TeleGeography 2001 Global Telecommunications Traffic Statistics and Commentary This work is based upon sources believed to be reliable, but the publisher does not warrant the accuracy or completeness of any information for any purpose and is not responsible for any errors or omissions. This work is for the contidential use of subscribers. Neither the whole nor any part of this publication may be reproduced or transmitted in any form or by any means, electronic, mechanical, photocopied, recorded or otherwise, with- out prior written consent from TeleGeography, Inc. All rights reserved. Copyright © TeleGeography, Inc. 2000 First Printing--October 2000 Second Printing--January 2001 This is book number TG01-110630 Do not reproduce. ISBN 1-886142-26-2 Printed in the United States of Amedca TeleGeography, Inc. ¯ 1730 Rhode Island Avenue, NW ¯ Suite 400 Washington, DC 20036 USA Tel. +1 202 467 0017 Fax +1 202 467 0851 E-mail: info@telegeography.com http://www.telegeography.com TeleGeol~raphy 2001 © TeleOeogrophg, Inc. 2000 TeleGeography, Inc. This report was prepared by TeleGeography, Inc., the Washington, D.C.-based research and publishing group of Band-X Ltd. The TeleGeography 2001 research and production team included: Jason Kowal .................... President & Editor-in-Chief Timothy Stronge .................. EVP & Director of Research Stephan Beckert .................. Director of Traffic Research Gregory C. Staple .............. Founder & Editorial Consultant Brain Dov Abramson ................. Senior Research Analyst Markus Krisetya ............. Research Analyst & Chief Designer Jenifer Neidhart de Ortiz ................... Research Analyst Janice Gentile .......................... Research Analyst Alan Mauldin ........................... Research Analyst Joseph Vigil ............................ Research Analyst Paul Melton ............................ Research Analyst Patrick Christian ......................... Research Analyst Trista Schroeder ......................... Research Analyst Graham Finnie ........................ Contributing Writer Jessica Marantz ....................... Contributing Walter The following publications also are available from TeleGeography, Inc.: International Bandwidth 2000 (ISBN 886142-22-2) Hubs and Spokes A TeleGeography Internet Reader (ISBN 886142-23-8) Global Communications Cable & Satellite Map (ISBN 886142-24-6) Global Communications Traffic Map (ISBN 886142-25-4) Direction of Traffic 1999 (ISBN 92-61-08041-3) New International Carriers Americas Edition (ISBN 886142-19-X) New International Carriers Asia/Pacific Edition (ISBN 886142-17-3) New International Carriers Europe Edition (ISBN 886142-18-1 ) For all inquiries, please contact: TeleGeography, Inc. 1730 Rhode Island Avenue, NW Suite 400 Washington, DC 20036 USA Tel. 1 202 467 0017 Fax +1 202 467 0851 E-maih info@telegeography.com For more information, please visit our web site: http://www.telegeography.com © TeleGeography, Inc. 2000 TeleGeography 2001 Contents List of Figures ............................................................................ z~ List of Country Traffic Tables ................................................................. 6 Preface ................................................................................. 7 Executive Summary ........................................................................ 9 Carriers The Growth of International Services Competition ............................................... 12 Market Shares of International Carriers ...................................................... 14 The Top 40 International Carriers .......................................................... 18 The Top U.S. International Carriers .......................................................... 19 A Primer on Bandwidth Exchanges ......................................................... 20 Pricing ’ Overview of International Pricing Trends ...................................................... 31 Elements of an International Call ........................................................... 35 International Carrier Call Costs from the U.S ............................................... 36 International Carrier Call Costs to the U.S ................................................ 38 International Private Line Pdces ....................................................... 40 International Settlement Rates ....................................................... 41 FCC and ITU Settlement Benchmarks ................................................... 43 National Interconnection Rates ....................................................... 45 The New Calculus: A Primer on Interconnection Accounting ........................................ 46 Retail Pricing ,, Retail Prices for a Three Minute Call .................................................... 50 Retail Pricing Trends, 1997-1999 ..................................................... 52 Retail and Wholesale Rates: PS-FN versus VolP ............................................ 53 Follow the Money: Network-to-Network Payments for lnternet Telephony and Other IP Traffic Streams ........... 54 Facilities MANs: The Golden Mile ................................................................ 71 Submarine Cable Systems ............................................................... 79 International Circuit Usage by U.S. Carriers .................................................... 89 International Communications Satellites ...................................................... 93 lnternet Backbones International Internet Backbones .......................................................... 102 Traffic Analysis Overview of International Traffic Trends ..................................................... 121 Bypass and Refile Traffic ................................................................ 124 VolP Routes ~. Traffic .................................................................. 128 International Traffic from Mobile Phones ..................................................... 132 International Call Quality Metrics ......................................................... 134 Traffic Summary Global Traffic Review .................................................................. 139 International Traffic by Region ............................................................ 141 International Traffic by Country ........................................................... 145 International Traffic by Route ............................................................ 148 Country Traffic Statistics International Traffic Tables (see page 6 for a list of countries) ................................................... 152 Survey Methodolo~oy .................................................................. 285 Global Reference: Blue Pages National Telecommunications Indicators ..................................................... 288 International Dialing Codes .............................................................. 292 North American Dialing Codes ........................................................... 297 A Primer on Bits ..................................................................... 302 About TeleGeography, Inc ................................................................... 304 TeleGeo~raphy 2001 © TeleGeography, Inc. 2000 List of Figures Carriers The Grewth of International Services Competition Figure 1. The International Carder Boom ............. 12 Figure 2. Countries with International Telephone Services Competition ........................... 13 Market Shares of International Carriers ....................... 14 The Top 40 International Carders ......................... 18 The Top U.S, Internabonal Carders ...................... 19 A Pnmer on Bandwidth Exchanges Rgure 1. Figure 2. Rgure 3 Rgure 4. Rgure 5. Figure 6. Selected Bandwidth Exchanges ............... 21 Common Bandwidth Trading Parameters .......... 22 Case Study: Bandwidth corn ................ 23 Case Study. Band-X ...................... 25 Pooling Points and Physical Matchmakers ......... 26 Case Study: Arbinet-thexchange ............... 27 Pricing Overview of International Pdcing Trends Figure 1. U.S. Carder Revenues and Settlement Outpayments, 1980-1999 .................. 31 Figure 2. U.S. Carrier Revenues for Intemabonal Voice Service, 1998-1999 ...................... 32 Figure 3. AT&T Settlement Payments per Minute to Select Destinations, 1999 ....................... 32 Figure 4. Settlement Payments per Minute for Outgoing Calls, 1998-1999 ...................... 33 Figure 5. U.S. Camer International Call Revenue by Destination, 1999 ....................... 34 Elements of an International Call Figure 1. International Call Components .............. 35 International Carder Call Costs fTom the U.S ..................... 36 International Carrier Call Costs to the U.S ..................... 38 International Private Line Prices Figure 1. International Private Line Lease Prices fi-om U,S., 1995-1999 ............................. 40 Rgure 2. Band-X Bit Index, 1998-2000 ................. 40 International Settlement Rates ............................... FCC and ITU Settlement Benchmarks ....................... 43 National Interconnection Rates .............................. 45 The New Calculus: A Primer on lnterconnect~on Accounting Figure 1. Call Costs from U.S. to U.K. and U.S. to Japan ....... 46 Figure 2. A Trade Dispute for the 21st Century .............. 47 Figure 3. EU Best Practices ......................... 48 Figure 4. Regional Termination Rates, 1998-2000 ......... 49 Retail Prices for a Three Minute Call .......................... 50 Retail Pricing Trends, 1997-2000 ............................. 52 Retail and Wholesale Rates: PSTN versus VoIP .................... 53 Follow the Money: Network-to-Network Payments for Intemet Telephony and Other IP Traffic Streams Rgure 1. The Death of Distance: International Cartier Bankruptcies and Distress Sales ................ 55 Figure 2. Figure 3. F~gure 4. Figure 5. Rgure 6. Rgure 7 Figure 8. Figure 9 Figure 10. How the Accounting Rate System Works for International Telephony ................... 56 Market Shares of Internet Backbone Prowders ...... 57 Time to Pay Up? ITU-T Recommendation D.(iH) International Internet Connections .............. 58 VolP Clearinghouse Payment Flows ............. 60 Clearinghouse Terms for VolP Traffic Termmabon .61 Band-X Routed IP Pricing ................... 82 WorldCom Costs for Transporbng U.S. Long Distance Voice and Data Traffic ............. 63 The Missing bnk: IP Billing Systems ............. 64 Network Choke Points: Telcos versus IP Backbone Providers .............................. 66 Facilities MAiNs: The Golden M~le Figure 1. No MAN ~s an Island ....................... 71 Rgure 2 Metromedia Fiber Network--London Area Map ..... 73 Figure 3. Pan-European MAN Deployment .............. 74 Figure 4. Where are the First--tier Cities? ................. 75 Figure 5. The Urban Jungle .......................... 76 Submarine Cable Systems Figure 1 Transoceanic Submanne Cable Capacity ........... 79 Figure 2. Submanne Cable Deployment Trends ............. 80 Figure 3. Major Submarine Cable Growth ................. 81 Rgure 4 Major Submarine Cable Systems by Region ......... 82 Figure 5 Map of Major Submanne Cables in the Umted States {Pacific Coast} ........................... 83 Figure 6. Map of Major Submarine Cables in the United States (Atlanbc Coast} & the Caribbean .............. 84 Figure 7 Map of Major Submarine Cables in Europe ....... 85 Figure 8. Map of Major Submanne Cables in Latin America and the Caribbean .......................... 86 Figure 9. Map of Major Submarine Cables in East Asia ....... 87 Figure I0. Map of Major Submarine Cables in Central Asia ..... 88 International C~rcuit Usage by U.S. Carners Figure 1. Internabonal Circuit Usage Summary, 1995-1999 ..... 89 Figure 2. International Circuit Usage for Selected Routes, 1997-1999 ............................... 90 Figure 3. International Circuit Usage by Region, 1997~1999 .... 91 Rgure 4. Idle Orcuits of U.S, Carders by Region, 1997-1999 .92 International Communication Satellites Figure 1. The Big Picture ............................ 93 Figure 2. Interregional Satellite Capacity, 1989-2001 ......... 94 Figure 3. Major Intemabonal Telecommunications Satellites in Geostationary Orbit ........................ 95 Figure 4. Major International Telecommunications Satellites in Geostationary Orbit ......................... 96 Figure 5. Map of Major Commumcations Satellites in the Atlantic Ocean Region ...................... 98 © TeleGeography, Inc. 2000 TeleGeol~raphy 2001 Figure 6. Rgure 7. Map of Major Communications Satelhtes ~n the Indian Ocean Reoon ...................... 98 Map of Major Communications Satellites ~n the Pacific Ocean Region ....................... 99 Inter~et Backbones International Internet Backbones Figure 1. Inten~egional Internet Bandwidth, 2000 .......... 103 Rgure 2. A Question of Scale ....................... 105 Figure 3. The Top 50 international lnternet Routes, 2000 ..... 106 Figure 4. The Top 50 International Internet Hub C~t~es, 2000 .107 Figure 5. Map of Major AsiaJPacific International Backbone Routes, 2000 .......................... 108 Figure 6. The Top 50 International Backbone Routes in Asia/Pacific, 2000 .......................... 109 Figure 7 Map of Major European International Backbone Routes, 2000 ........................... 110 Figure 8, The Top 50 International Backbone Routes in Europe, 2000 .......................... 111 Figure 9. Map of Major South American International ’Backbone Routes, 2000 ................. 112 Figure 10 The Top 50 International Backbone Routes in Latin Amedca and the Caribbean, 2000 ......... 113 Figure 11. Map of Major African International Backbone Routes, 2000 ......................... 114 Figure 12. The Top 50 International Backbone Routes in Africa, 2000 ............................. 115 Figure 13. Map of Major International Backbone Routes in the U.S and Canada, 2000 .................. 116 Figure 14. The Top 50 International Backbone Routes in the U.S. and Canada, 2000 ................... 117 Traffic A~alysis Overview of International Traffic Trends Figure 1. Incumbent Market Shares Go in One Direction ..... 121 Figure 2. Falling Pdces, Falling Revenues ................ 122 Figure 3. Charge of the Challengers .................... 122 Rgure 4 Bypass at Work? .......................... 123 Bypass and Refile Traffic Figure 1. Top ISR Routes with the U.S,, 1997-1999 ........ 124 Figure 2. Call Delivery Methods ...................... 125 Figure 3. Bypass Targets .......................... 126 Figure 4. File Refile Shell Game ...................... 126 Vo]P Routes & Traffic Figure 1. International VolP and PSTN Traffic Summary ....... 128 Figure 2. The Top 20 U.S.-Odginated VolP Routes, 1999-2000 ............................ 129 Rgure 3. Major VolP Carders & Traffic ................. 130 Figure 4. Major VolP Clearin!]houses .................... 131 International Traffic from Mobile Phones Figure 1. Mobile versus Fixed International Traffic and Subscribers by World Region, 1999 ............ 132 Rgure 2 Percent of International Traffic from Mobiles, 1998-1999 .......................... 133 Rgure 5. Mobile Subscribers and Internationa]Traffic for Selected Countries, 1999 ................... 133 International Call Quality Metncs Rgure 1. Call Quality from Germany to Mobile and F~xed Line Telephones ....................... 134 Rgure 2. Quality Metrics on Calls from Germany and the U.K., June-August 2000 .................. 135 Traffic Summary Global Traffic Rewew Figure 1. International Traffic and Main Line Growth ......... 139 Figure 2. International Traffic, Revenue, and Subscriber Growth .......... ................... 159 Figure 3. Intercontinental Traffic Flows, 1997 & 1999 ....... 140 International Traffic by Region Figure 1. Interregional Traffic Flows, 1999 .............. 141 Figure 2 International Traffic by Origin, 1999 ......... 141 Figure 3. Traffic Growth by Region, 1998-1999 .......... 141 Figure 4. European Telecommunications Traffic Flows, 1999 ... 142 Figure 5 Latin Amedcan Telecommunications Traffic Flows, 1999 .............................. 143 Figure 6 As~an Telecommunications Traffic Flows, 1999 .... 144 International Traffic by Country Figure 1. Outgoing International Telephone Traffic Growth for Selected Countries, 1998-1999 ........... 145 Figure 2. Telephone Traffic Balance for Selected Countnes, 1999 ......................... 145 Figure 3. International Traffic Indicators, 1999 ........... 146 International Traffic by Route F~gure 1 Figure 2. Figure 3 Figure 4. The Top 50 International Routes, 1999 .......... 148 Traffic Imbalances on Selected U.S. Routes, 1999 .... 149 Traffic Imbalances on Selected Non-U.S. Routes, 1999 ............................... 149 International Outbound Routes with Rapidly Growing Traffic, 1998-1999 .................. 50 Country Traffic Statistics (see page 6 for a hst of countnes] Global Reference: Blue Pages National Telecommunications Indicators .................. 288 International Dialing Codes, by Country ................ 292 Wodd Diahng Codes Map ........................... 294 International Diahng Codes, by Number ................. 296 North Amencan Area Codes, by Number ................ 297 North Amencan Area Codes, by Jurisdiction ............... 298 North American D=aling Codes Map ................ 300 TeleGeography 2001 © TeleGeography, Inc. 2000 List of Country Traffic Tables Albania ................... 152 Algeria .................... 153 Andorra .................... 1 Angola .................... 155 Argentina ................... 156 Armenia ................... 157 Australia ................... 158 Austria .................... 159 Azerbaijan .................. 160 Bahamas ................... 161 Bahrmn .................... 162 Bangladesh ................. 163 Belarus .................... 164 Belgium .................... 165 Benin ..................... 166 Bolivia .................... 167 Bosnia-Herzegovina ........... 168 Brazil ..................... 169 Brunei ..................... 170 Bulgaria .......... ’ ......... 171 Burkina Faso ................ 172 Burundi .................... 173 Canada .................... 174 Central African Republic ......... 175 Chad ...................... 176 Chile ...................... 177 China ..................... 178 Costa Rica .................. 179 C6te d’lvoire ................. 180 Cuba ...................... 181 Cyprus .................... 182 Czech Republic ............... 183 Denmark ................... 184 Dominican Republic ........... 185 Egypt ..................... 186 El Salvador ................. 187 Eritrea ..................... 188 Estonia .................... 189 Ethiopia ................... 190 Finland .................... 191 France ..................... 192 French Polynesia .............. 193 Georgia .................... 194 Germany ................... 195 Ghana ..................... 196 Gibraltar ................... 197 Greece .................... 198 Guatemala .................. 199 Guyana ................... 200 Hong Kong .................. 201 Hungary ................... 202 india ...................... 203 Indonesia .................. 204 Iran ...................... 205 Ireland .................... 206 Israel ..................... 207 Italy ...................... 208 Jamaica .................... 209 Japan ..................... 210 Jordan .................... 211 Kazakhstan ................. 2 ! 2 Korea, Republic of ............ 213 Kuwait .................... 214 Kyrgyzstan .................. 215 Latvia ..................... 216 Luxembourg ................. 217 Macau .................... 218 Macedonia .................. 219 Madagascar ................. 220 M alaysia ................... 221 Malta ..................... 222 Mauritania .................. 223 Mauritius ................... 224 Mexico .................... 225 Moldova ................... 226 Mongolia ................... 227 Morocco ................... 228 Mozambique ................ 229 Myanmar ................... 230 Namibia ................... 231 Netherlands ................. 232 New Caledonia ............... 233 New Zealand ................ 234 Nicaragua .................. 235 Niger ..................... 236 Norway .................... 237 Oman ..................... 238 Pakistan ................... 239 Palestinian Authority ........... 240 Panama .................... 241 Paraguay ................... 242 Peru ...................... 245 Philippines .................. 244 Poland .................... 245 Portugal ................... 246 Qatar ..................... 247 Russia ..................... 248 Rwanda .................... 249 Saudi Arabia ................ 250 Senegal .................... 251 S~ngapore .................. 252 Slovak Republic .............. 253 Slovenia ................... 254 South Africa ................. 255 Spain ..................... 256 Sri Lanka ................... 257 Sudan ..................... 258 Sudname ................... 259 Swaziland .................. 260 Sweden .................... 261 Switzerland ................. 262 Syria ...................... 263 Taiwan ..................... 264 Tajikistan ................... 265 Thailand ................... 266 Togo ...................... 267 Trinidad & Tobago ............. 268 Turkey ..................... 269 Turkmenistan ................ 270 Ukraine .................... 27 United Arab Emirates .......... 272 United Kingdom-Outgoing ....... 273 United Kingdom-Incoming ....... 274 U.K. Top 100 Correspondents ..... 275 United States-Outgoing ......... 276 United States-Incoming ......... 277 U.S. Top 100 Correspondents ..... 278 Uruguay ................... 279 Uzbekistan .................. 280 Venezuela .................. 28! Yugoslavia .................. 282 Zambia .................... 283 Zimbabwe .................. 284 © TeleGeography, Inc. 2000 TeleGeography 2001 Preface Global communications--to paraphrase a familiar axiom--is a journey, not a destina- tion. There is no finish line in our industry. Technology continues to evolve rapidly and relentlessly, the players are in a constant state of transformation, and even the rules of competitive engagement are in flux. All of this makes for an extremely exciting trip. But trying to keep abreast of this change and understand where we’re headed is a tremen- dous challenge. Fortunately, TeleGeography’s annual report helps us make some sense out of the indus- try swirl surrounding us. For 1 I years now, this report has been one of the most reli- able and comprehensive overviews of the global communications industry--an essential "roadmap" for understanding the journey we’re on. It provides accurate profiles of the industry’s major carriers and their market shares; Internet backbone maps and rank- ings; international traffic statistics and analyses; and detailed pricing information. It also helps to clarify and define important industry trends such as deregulation and pri- vatization. In this edition, for the first time, you will find a summary of the minutes-of- use for Voice over IP (VolP) traffic, a useful addition given the recent growth of this sec- tor and the attention it’s getting from end-users, ISPs, data vendors, and existing tele- corn operators. As we all know, next generation services like VolP are a small part of a much larger phe- nomenon-the explosive growth of the Internet worldwide. Here in the United States, this phenomenon has already firmly taken hold. The U.S. alone accounts for more than 40 percent of current global Internet usage, and our e-commerce spending is expected to reach over $35 billion this year. According to a number of recent reports, however, the biggest growth in Internet usage and spending during the next several years is like- ly to take place in Europe and Asia-Pacific. For example, Europe is expected to add more new Internet users than any other world region, and Asia-Pacific will experience the most rapid annual growth rate. With those kinds of expectations, it’s extremely important to have a solid understanding of the global marketplace, the key players who shape it, and the relationships and alliances that unfold and shift almost daily. Clearly, a report like TeleGeography’s will continue to be invaluable. On behalf of the entire WorldCom family, I am pleased to commend this TeleGeography 2001 report to your attention. We’re proud to sponsor this report and to continue our long-standing support of the annual TeleGeography series--one of the few tools avail- able today that actually makes the often unpredictable global telecommunications jour- ney just a little easier to navigate. Robert K. Lacy Vice President WorldCom International Expansion Support TeleGeography 2001 © TeleCeograph~l, Inc. 2000 Acknowledgements Wf e wish to thank the numerous carriers, government departments, regulators, and international organizations from around the world who responded to our requests or information. This report would not exist without the help of the dedicated people at these organizations who took the time to ensure that the data reported here are as current and accurate as possible. We would also like to thank the people who helped review early and final drafts of this book, including: Heidi Harris Bersin, Teddy Chu, Richard Elliott, Kevin Heneghan, Cathy Hsu, Cynthia Hswe, Doug Johnson, Tim Kelly, Robert Luczak, Craig McTaggart, Michael Minges, Nigel Panter, Chris Reid, Don Weightman, Brian Walsh, and Brent Wilkins. TeleGeography 2001 was supported, in part, by a publication grant from WorldCom, Inc. As in the past, however, this grant was made without any precondition; TeleGeography, Inc. is sole- ly responsible for the report’s editorial contents. The Editors © TeleGeographg, Inc. 2000 TeleGeography 2001 Executive Summary The web of supply and demand in the international telecom industry becomes more intricately woven each year. The purpose of the TeleGeography annual report is to help read- ers understand how that web is evolving through careful observation and rigorous reporting. The highlights of this year’s edition follow. Competition As of July 2000, more than 2,800 companies worldwide were authorized to build international telephone networks. Three years before, there were less than 600. Although most of these companies are too small to be noticed, their gross impact on global traffic flows is hard to miss. In total, the facilities-based carriers which started business since 1989 now carry almost a quarter of the world’s interna- tional call minutes. In places like Hong Kong and Germany, for example--where the incumbents had only lost their monopolies in’ ]998--new entrants gained more than a third of the international minutes market in just one year. Pricing Prices for international calls are falling fast. Cutthroat com- petition in the international services industry is providing the incentive---and falling settlement and bandwidth costs the means--for carriers to slash prices. Call prices from parts of recently liberalized Western Europe (e.g., Germany) to many international destinations have fallen 90 percent in just two years. Ironically, the only thing holding up interna- tional call prices on many competitive routes is the cost of local interconnection at either end of the call. In the long run, as the settlements regime disintegrates in favor of an interconnection model, we can expect local, domestic long distance, and international consumer prices to converge. Just as we were going to press, the German regulator helped prove the point by permitting Deutsche Telekom to charge just 9 pfennigs (4¢) per minute for calls to the United States--only one pfennig more than a call to the apartment next door. Facilities The undersea bandwidth boom reached an unprecedented single-year growth rate in 2000. Submarine cables installed in 2000 increased aggregated trans-Atlantic bandwidth by a factor of 2 in just one year, to over two terabits per sec- ond. And while huge growth rates in long-haul capacity have been standard fare for the latter half of this decade, bandwidth at the edges--in the metropolitan area network (MAN)--has been in short supply until recently. This short- Figure 1. Falling Prices. Falling Revenue Lowest Available Retail Price for Calls from Germany US "~ Turkey Poland Austria UK ~Year End 1997 mJanuary2000 0.00 0.25 0.50 0.75 1.00 1.25 1 50 Deutsche Mark per Minute International Service Revenues and Call Volumes, 1998-1999 40% ~Annual Call Volume Change 30% ¯ Annual Revenue Change 10% E° 0% -20% -30% -40% Source: Regierungsbehbrde for Telekommunikation und Post (RegTP), company reports, TeleGeography research © TeleGeography, Inc, 2000 TeleGeography 2001 © TeleGeographyo Inc. 2000 Figure 2. Top Carriers and Competition 120 International Traffic by Carrier Type, 1989-1999 New Carriers In Competitive Markets ~: Existing Carriers zn Competztive Markets ¯ Monopohes 20 0 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 The Top 10 International Carriers Outgoing Traffic (millions of minutes) Rank Company (Country) 1998 1999 1. AT&T (U.S.) ’ 10,798.5 10,816.5 2. WorldCom (U.S.) 7,195.0 8,294.9 3. France T~l~com (France) 3,911.0 4,390.0 4. BT (U.K.) 4,249.3 4,029.1 5. Deutsche Telekom (Germany) 4,711.0 3,860.0 8. Sprint (U.S.) 2,916.0 3,714.4 7. C&W Com. (U.K.) 2,646.2 3,177.0 8. Telecom Italia (Italy) 2,339.4 2,390.6 9. Swisscom (Switzerland) 2,258.0 2,259.0 10. China Telecom (China) 1,711.5 1,950.0 Top 10 Total 42,735.9 44,881.5 Global Total 93,360.0 107,800.0 Source; TeleGeogra phy researc h © TeleGeography, Inc. 2000 age sparked a MAN "building boom--first in the U.S. and then in Europe. In most international business cities, at least three (and often more) networks are being construct- ed, creating a unprecedented capacity infrastructure filled with many hundreds of fiber pairs. Internet Backbones So what is filling up all this new capacity? Although people do make more phone calls each year, much new bandwidth is being devoured by hungry lnternet service providers (ISPs). In 2000, ISPs began to take advantage of the fiber explosion, and some upgraded their international backbone connections from 155 Mbps to 2.5 Cbps (or 2,500 Mbpsl. This led to a tripling or quadrupling of bandwidth on many routes, especially those connecting North America to Europe (13 to 56 Gbps) and to Asia (6 to 20 Gbps). Traffic Flows International telephone traffic grew by over 15 percent in 1999, to 107.8 billion minutes, fueled by falling prices and the mobile phone boom in Europe and Asia. Call volume grew especially rapidly in Western Europe, where new carri- ers piled into recently liberalized markets, and where mobile operators added 75 million customers. International traffic from countries such as the Netherlands and Germany, which had been growing at five percent or less in 1996 and 1997, increased by 14 to 18 percent in 1999. Nevertheless, demand did not grow fast enough to compensate for the steep drop in prices, as many carriers, including Telstra, Sprint, and Telmex, reported increased call volumes, but lower revenues from international calls. Mobile phones played an increasingty important role in international telephone traffic in 1999--approximately 11.5 percent of international calls were placed from mobile phones. Almost two-thirds of this traffic was generated in Europe, where cross-border roaming contributed substan- tially to international call volumes. Swisscom, for example, reported that mobiles originated one-third of outbound international calls in 1999. Voice over Internet Protocol (VolP) traffic began to have an appreciable impact on international call volumes in 1999. Total international VolP traffic grew more than tenfold, to approximately 1.7 billion minutes. Although VolP account- ed for only approximately 1.6 percent of total international traffic in 1999, it had a disproportionate impact on some routes, particularly from the U.S. to developing countries. The largest route for international VolP, by far, is from the U.S. to Mexico. In 1999, calls between these two countries accounted for nearly 30 percent of all international VolP minutes. Conclusion This year’s edition of TeleGeography--the most compre- hensive yet---expands on the points above with a collection of 15 topical essays and over 250 statistical tables and charts. Like the markets we cover, however, the form and function of TeleGeography are evolving. We welcome your questions, comments, and criticisms to help improve future editions. Please send your correspondence to the coordi- nates listed on the title page of this book. ~i~ Carriers TeleGeography 2001 © TeleGeograph~l, Inc. 2000 The Growth of International Services Competition Sixty Percent Growth in Competition As of July 2000, more than 2,800 companies worldwide were authorized to build facilities to offer international tele- phone service. Three years before, there were less than 600 (see Figure 1. The International Carrier Boom). In the fastest growing markets (the U.S. and Western Europe), the pace is not likely to slow down, even in the face of the recent stock market slide. One might assume that closing the door to international capital would impede new entrants. But many new carriers do not own extensive submarine cable capacity and switching assets, so their start-up costs can be minimal. In the U.S. especially, the hundreds of small com- panies that are authorized to own networks may never build them. Nonetheless, a handful of carriers have built new networks, and they have collectively chipped away at incumbent mar- ket shares. In total, t~he facilities-based carriers which start- ed business since 1989 now carry almost a quarter of the world’s international telephone traffic (see the "Overview of International Traffic Trends" in the Traffic Analysis section below). The relationship between the network builders and the swarm of "virtual" carriers--which repackage the facilities and services of network builders--is one of symbiosis. New market entrants, while they represent a competitive threat, can also be the incumbent’s best customers. And, in some cases, new specialist wholesale carriers are serving up their facilities in the other direction--to established carriers that are encumbered by marketing expenses and bureaucratic p~ocesses. The New Breed of Virtual Carrier Both facilities-based and virtual carriers alike are always on the hunt for new ways to cut prices without shrinking profit margins. The latest development in alternative traffic rout- ing is creating a new kind of packet-switched symbiosis. Once the network builders determine how to send commer- cial grade traffic on IP networks reliably and to devise a way to settle accounts properly, the ranks of international carri- ers will swell even more rapidly. Indeed, IP connectivity may lead to unregulated international carriers on virtually every street corner, in every corner of the world. ~i~..~ Figure 1. The International Carrier Boom Global Growth of International Carriers, July 1995 July 2000 3000 2805 2500 2000 1500 1000 500- 367 1995 586 1760 1042 1996 1997 1998 1999 2000 Regional Growth of Authorized International Carriers, 1995-2000 Americas Europe Asia-Pacific Africa ¯ J uly 2000 [] July 1999 July 1998 July 1997 7 July 1996 [] July 1995 200 400 500 800 1000 1200 International Carriers Note: Figures include all carriers authorized to provide international facilities-based service or international simple resale (ISR). Source: TeleGeogra phy researc h © TeleGeography, Inc, 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Figure 2. Countries with International Telephone Service Competition Number of Authorized International Carriers July 1999 July 1998 July 1997 July 1996 115 Rank Country July 2000 July 1995 1 United States 1,100 679 393 175 65 2 United Kingdom 306 215 144 100 65 35 3 Hong Kong 150 80 4 1 1 1 4 Japan 115 50 13 3 3 3 5 Germany 90 40 32 1 1 1 6 France 89 50 29 1 1 1 7 Canada 75 49 21 21 19 18 8 Netherlands 60 30 23 3 1 1 9 Italy 52 15 9 1 1 1 10 Switzerland 50 40 21 1 1 1 11 Denmark 45 18 11 9 7 1 12 Australia 40 28 14 10 8 8 13 Austria 40 17 13 1 1 1 14 Ireland 40 25 5 3 3 1 15 Korea, Rep. 40 24 3 2 2 2 16 Singapore 40 1 1 1 1 1 17 Norway 35 14 7 1 1 18 Russia* 30 30 1 1 1 1 19 Spain 30 16 9 1 1 1 20 Sweden 26 16 13 11 9 7 21 ,Belgium 21 18 11 1 1 1 22 I~ew Zealand 21 19 11 9 9 2 23 Finland 20 8 8 8 8 5 24 Peru 19 18 1 1 1 1 25 Mexico 16 16 15 9 1 1 26 Portugal 15 1 1 1 1 1 27 Philippines 12 12 12 9 9 9 28 Chile 10 10 9 9 9 9 29 El Salvador 10 10 10 1 1 1 30 Luxembourg 10 4 1 1 1 1 31 Taiwan 10 1 1 1 1 1 32 Iceland 8 3 1 1 1 1 33 Malaysia 5 5 5 5 5 4 34 Argentina 4 2 1 1 1 1 35 Colombia 3 3 3 1 1 1 36 Dominican Rep. 3 3 3 3 3 3 37 Israel 3 3 3 3 1 1 38 Kazakhstan 3 3 3 1 1 1 39 Ecuador 3 3 3 3 1 1 40 Bermuda 2 2 2 2 2 1 41 Brazil 2 2 1 1 1 1 42 Brunei 2 2 2 2 2 2 43 China 2 2 2 2 2 2 44 Dominica 2 2 1 1 1 45 Georgia 2 2 1 1 1 1 46 Guatemala 2 2 1 1 1 1 47 Indonesia 2 2 2 2 2 2 48 Nepal 2 1 1 1 1 1 49 Ukraine 2 2 2 2 2 2 Note: Figures include all carriers authorized to provide faciliues-based international service or international simple resale as of July 1 for each year. Estimates include Russian carriers authorized to provide service only in certain municipalities. Source: TeleGeography research © TeleGeegrsphy, inc, 2000 TeleGeography 2001 © TeleGeography, Inc. 2000 Market Shams of International Carriers Percentage of Outgoing Minutes Country/Carrier 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 Australia Telstra 100.0 98.0 87.0 76.3 73.4 62.0 55.0 49.0 49.5 C&W Optus 2.0 13.0 21.9 23.4 27.0 26.0 22.0 21.9 AAPT 11.0 13.4 13.6 Prirnus 3.0 4.0 5.0 Teleglobe 4.4 4.4 Others 1.8 3.2 11.0 5.0 7.2 5.6 Austria Telekorn Austria 100.0 95.0 80.0 UTA Telekom 1.5 6.0 Tele2 5.0 tele.ring 3.0 Others 3.5 6.0 Belgium Belgacom 100.0 87.0 81.0 Others 13.0 19.0 Canada Stentor 71.0 70.0 70.0 69.0 66.0 54.0 44.0 44.0 41.0 40.0 n.a. Bell Canada ’, 23.0 Teleglobe 29.0 30.0 30.0 31.0 29.0 33.0 30.0 23.0 26.0 24.0 20.0 Sprint Canada 15.0 21.0 17.0 18.0 20.0 AT&T Canada 1.0 5.0 8.0 9.0 10.0 14.0 9.0 Primus 9.0 Telus 6.0 Others 4.0 8.0 3.0 3.0 5.0 4.0 13.0 Chile CTC Mundo/Globus I.0 17.5 31.2 31.0 31.5 33.0 35.0 33.0 ENTEL Chile 100.0 80.0 57.5 40.0 40.6 37.3 33.0 31.0 31.0 Chile Sat ’ 20.0 25.0 19.7 19.4 15.2 17.0 13.0 15.0 BellSouth Chile 6.6 6.8 10.0 10.0 10.0 10.0 AT&T (F=rstCom) 1.2 I.0 2.8 3.0 5.0 3.0 TransAm I.0 1.0 2.8 3.0 3.0 3.0 Others 1.0 1.0 1.0 1.0 3.0 5.0 Denmark Tele Danrnark 100.0 92.5 84.4 67.5 55.3 Tele2 4.0 6.6 12.4 13.2 Telia Danrnark 3.5 6.3 9.9 10.7 Global One 6.6 RSL Corn 3.7 Interoute 3.7 Teleglobe 3.1 Others 2.7 10.3 3.7 Dominican Republic Codetel 100.0 >90.0 85.8 83.0 77.0 73.8 72.2 78.1 Tricom n.a. 6.7 7.5 12.8 12.9 15.5 14.2 AACR n.a. 7.5 9.5 10.2 13.3 12.3 7.7 Notes: Data based on outgoing international traffic for the pubhc switched neWvork and Internatzonal S~mple Resale (ISR) covering the full calendar or fiscal year. Some data aggre- gated in "Others" rows znclude market shares for carriers shown ~ndividually m later years. Market shares may not total to 100 percent due to rounding. Canada: The Stentor alliance, wh=ch was dissolved in 1999, included Bell Canada, Teluso MTS, SaskTel, and Aliant. BCE, the parent company of Bell Canada, announced the purchase of Te[eglobe in Februa~/2Q00. Untd October 1998, Teleglobe held a monopoly, on a~l non-U.S, routes. Sprint Canada market shares include traffic carried by Fonorola, whzch merged wzth Sprint Canada in 1998. AT&T Canada market shares include ACC traffic prior to 1999 merger. Primus acquired the consumer divzsion of AT&T Canada in May 1999. Chile: CTC Mundo/Globus market shares prior to 1998 merger aggregate CTC Mundo and Globus (formerly V]’R) traffic. Source: TeleGeography research © TeleGeography, Inc. 2000 © TeleCeogrczphg, Inc. 2000 TeleGeography 2001 Percentage of Outgoing Minutes Country/Carrier 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 El Salvador CTE Antel Telecom 100.0 91.5 85.0 Telef6nica de El Salvador 7,0 Teleglobe 6.0 Others 8.5 2.0 Finland Sonera 100.0 90.0 72.8 66.0 58,9 54.7 53.0 Rnnet International 5.0 19.1 24.2 28.2 28.0 25.7 Telia Finland 3.0 7.7 8.8 9.3 12.0 8.6 RSL Corn 5.6 Others 2.0 0.4 0.9 3.5 5.2 7.1 France France T~l~com 100.0 93.0 88.7 Teleglobe 3.6 4.6 Cegetel 1.9 4.0 Others 1.0 2.7 Germany Deutsche Telekom 100.0 80.3 55.4 Mannesmann 7.2 13.1 WoddCom 1.8 4.9 Via9 Interkom 1.4 3.9 Teleglobe 1.8 2.4 RSL Corn 2.3 Others 7.5 18.0 Hong Kong C&W Hong Kong Telecom 100.0 90.0 61.3 New World Telephone 2.0 14.3 New T&T Hong Kong 2.0 12.0 Teleglobe 5.1 Others 6.0 7.3 Indonesia PT Indosat 100.0 99.5 95.4 88.5 84.8 88.3 86.5 PT Satelindo 0.5 4.6 11,5 15.2 11.7 13.5 Ireland Eircom 100,0 91.0 78.0 74,0 Esat Telecommunications 5.0 8.0 11.0 WorldCom 3.0 3.0 3.0 Others 1.0 11.0 12.0 Israel Bezeq 100.0 72.5 51.4 45.9 Barak ITC 15.0 24.8 30.0 Golden Lines 12.5 23.7 24.1 Italy Telecom Italia 100.0 88.6 77.1 Infostrada 4.5 8.4 Teleglobe 3.8 4.2 Albacore 1.0 1.6 Wind Telecomunicazioni 1.6 Others 2.0 7.1 Notes’ Germany: Mannesmann market shares include international traffic earned by mobile operator Mannesmann Mobilfunk and the long distance carriers Mannesmann ARCOR and o.tel o Hong Kong" C&W Hong Kong Teleoom was acquired by Pacific Century CyberWorks in August 2000. Source: TeleGeography research © TeleGeography, Inc. 2000 TeleGeography 2001 © TeleGeogrnphy, Inc. 2000 Market Shares of International Carriers (continued) Percentage of Outgoing Minutes Country/Carrier 1989 1990 1991 1992 1993 1994 1999 1996 1997 1998 1999 Japan KDD 93.3 88.0 73.3 69.7 66.9 66.3 66,2 63,9 62.7 58.0 51.1 C&W IDC 3.7 6.5 13.3 15.3 16.9 17.3 17.3 18.7 18.4 18.2 17.5 Japan Telecom 3.0 5.5 13.4 15.0 16.2 16.4 16.5 17.5 19.0 18.3 17.4 Teleglobe 3,2 N’l-r 1.2 Others 5.5 9.6 Korea, Rep. Korea Telecom 100.0 79.9 74.5 68.7 72.6 73.5 69.0 66.6 59.5 DACOM 20.1 25.5 31.3 27.4 26.5 27.0 21,9 24.7 Onse 4.0 11.5 15.8 Malaysia Telekom Malaysia 100.0 90.0 80.0 77.0 58.5 Celcom 8.0 11.0 10.0 14.5 Maxis 7.6 11,2 TIME Telekom 5.0 8.7 Digi 7.2 Others 2.0 9.0 1.0 n.a. Mexico Telmex Alestra Avantel Teleglobe Protel lusacell Others Netherlands P’l-r Telecom (KPN) RSL EnerTel Teleglobe Others New Zealand Telecom N. Zealand CLEAR Teleglobe Others Norway Te~enor Tele2 Norge World Access Telia Others 100.0 83.0 78.0 68.0 8.5 10.5 16.0 7.5 8.5 10.0 2.0 1.0 1.0 1.0 1.0 2.0 2.0 100.0 92.0 82.0 80.0 78.4 74.8 8.0 18.0 20.0 21.6 25.2 100.0 95.0 84.9 75.6 3.0 2.0 2.0 2.5 1.6 3.0 13.3 17.3 78.0 78.2 74.6 77.5 72.5 22.0 19.8 20.2 12.3 17.9 6.9 6.2 2.0 5.2 3.3 3.4 100.0 93.5 73.0 7.0 6.0 5.0 5.0 1.5 9.0 Notes; Japan: Japan Telecom market shares ~nclude ITJ prior to 1997 merger. Malaysia: Binariang changed its name to Maxis Communications in 1999. Source: TeleGeography research © TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Percentage of Outgoing Minutes Country/Carrier 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 Philippines PLDT 100.0 91.6 84.2 69.0 68.0 79.0 73.0 69.0 59.2 Globe Telecom 2.0 7.0 8.6 17.6 Digitel 2.0 3.0 4.3 5.8 Bayan Tel 1.0 4.0 5.0 5.7 5.5 Capitol Wireless 1.0 1.0 1.0 1.0 3.5 4.6 Eastern 7.0 6.0 5.0 7.0 6.4 4.0 Philippine Global Corn, 8.4 15.8 23.0 23.0 6.0 3.0 1.1 1.8 Islacom 1.0 1.0 1.0 1.4 Others 1.0 1.0 Spain Telef6nica 100.0 90.5 86.0 Retevisi~)n 4.5 7.0 Teleglobe 2.5 Lince 2.5 Others 5.0 2.0 Sweden Telia 100,0 92.0 87.0 76.0 69.0 66.0 62.0 53.0 Tele2 8.0 13.0 21.0 22.0 22.0 24.0 18.0 RSL Corn, 8.0 Telenordia 7.0 World Access 4.0 WorldCorn 4.0 Teleglobe 2.0 Others 3.0 9.0 12.0 14.0 4.0 Switzerland Swisscom 100.0 93.5 82.7 diAx 0.7 6.7 Sunrise 3.0 5.1 Others 2.8 5.5 United Kingdom BT 91.0 86.0 81.0 76.8 74.2 68.6 67.7 60.0 54.9 51.6 39.7 C&W Com. 9.0 14.0 19.0 23.2 24.0 28.1 25.8 26.8 30.3 32.2 31.3 WorldCom 6.6 5.1 5.1 10.0 Teleglobe 4.2 4.8 RSL Corn 3.0 WorldxChange 3.0 3.6 3.0 2.0 Global One 3.1 1.5 1.5 2.0 Others 1.8 3.3 6.5 1.0 4.6 2.2 7.2 United States AT&T 83.3 78.4 74.8 70.3 62.2 60.1 54.3 50.2 44.7 39.6 36.5 WorldCom 10.2 14.6 17.8 21.2 25.4 28.6 32.0 32.9 31.2 28.8 28.0 Sprint 5.8 6.4 6.3 7.3 10.3 11.1 11.3 13.2 12.0 11.7 12.5 Teleglobe USA 1.3 3.3 5,7 World Access 1.0 2.7 3.8 Viatel 0.3 0.8 3.0 Prirnus 0,3 0,5 2.9 STAR Telecorn. 0.5 1.8 2.7 Others 0.7 0.7 1.1 1.2 2.1 0.2 2.4 3,7 8.7 10.8 4.9 Notes: Phihppines: PLDT market shares include Smart Commumcations traffic prior to 1999 acquisition. Umted Kingdom" The figures for Cable & Wireless Communications reflect data for Mercury prior to its April 1997 merger with Bell Cablemed=a, V~deotron, and NYNEX CableComms. WorldxChange market shares include ACC Long Distance U.K. traffic pnor to 1999 acquisition. United States’ Market shares for U.S. carriers prior to 1993 exclude traffic to Canada and Mexico. WorldCom market shares prior to 1998 merger aggregate MCI and WorldCom traffic. WorLd Access market shares include Facd=Com traffic prior to 1999 merger. In February 2000, World Access announced agreements to acquire STAR and WorldxChange. Viatel traff=c includes Dest~a, which was acquired by V~atel =n November 1999. Source: TeleGeography research @ TeleGeography, Inc 2000 TeleGeography 2001 © TeleGeogrophy, Inc. 2000 The Top 40 International Carriers Outgoing Traffic (millions of minutes) Rank Company Origin Country 1999 1998 Change 98-99 Total 1. AT&T (a) U.S. 10,816.5 10,798.5 0.2% $62.6 2. WorldCom (a) U.S. 8,294.9 7,195.0 15.3% $37.1 3. France T~l~com France 4,390.0 3,911.0 12.2% $29.0 4. BT (b) U.K. 4,029.1 4,249.3 -5.2% $35.3 5. Deutsche Telekom ~a) Germany 3,860.0 4,711.0 -18.1% $37.8 6. Sprint (a) U.S. 3,714.4 2,916.0 27.4% $19.9 7. C&W Com. (a, b) U.K. 3,177.0 2,646.2 20.1% $3.0 8. Telecom Italia (a) Italy 2,390.6 2,339.4 2.2% $33.9 9. Swisscom Switzerland 2,259.0 2,258.0 0.0% $7.4 10. China Telecom China 1,950.0 1,711.5 13.9% $33.7 11, Teleglobe USA (a) U.S. 1,679.7 830.3 102.3% $2.9 12, C&W Hang Kong (a, b) Hang Kong 1,668.3 1,681.6 -0.8% $3.6 13, Telef~nica (a) Spain 1,665.0 1,518.0 9.7% $24.5 14. PTT Telecom (KPN)(a) Netherlands 1,625.0 1,600.0 1.6% $8.6 15. Singapore Telecom (a, b) Singapore 1,350.0 1,235,0 9.3% $2.9 16. Bell Canada Canada 1,305.0 1,350.0 -3.3% $12.6 17. Belgacom (a) Belgium 1,288.0 1,271.0 1.3% $4.9 18. Teleglobe (a) Canada 1,130.0 1,145.0 -1.3% $2.9 19. Sprint Canada Canada 1,130.0 865.0 30.6% $19.9 20. World Access(a) U.S. 1,129.5 678.8 66.4% $0.5 21, Telekom Austria (a) Austria 1,080.0 1,100.0 -1.8% $4.0 22. Telmex (a) Mexico 1,063.1 1,022.8 3.9% $10.1 23. Saudi Telecom Saudi Arabia 1,060.0 932.6 13.7% $4.0 (est.) 24. Telstra (b) Australia 1,046.0 836.0 25.1% $12.5 25. WorldCom U.K. (a, b) U.K. 1,015.0 425,0 138,8% $37.1 26. KDD (a) Japan 1,000.0 1,100.0 -9.1% $5.4 27. Etisalat U.A.E. 963.0 874.8 10.1% $1.7 28. Chunghwa Telecom Taiwan 949.3 862.0 10.1% $6.4 29. Rostelecom (a) Russia 928.2 1,038.3 -10.6% $0.9 30. Mannesmann Germany 915.0 425.0 115.3% $9.7 31. Viatel (a) U.S. 901.6 202.3 345.6% $0.2 32. Primus (a) U.S. 868.5 124.9 595.4% $0.8 33. STAR Telecom. (a) U.S. 785.8 457.4 71.8% $1.1 34. Eircom (a, b) Ireland 749.1 613.0 22.2% $2.0 35. aTE Greece 725.7 681.3 6.5% $2.0 36. Telia Sweden 725.0 750.0 -3.3% $8.3 37. Turk Telekomunikasyon Turkey 698.4 644.1 8.4% $3.7 38. Telekomunikacja Polska Poland 624.0 602.4 3.6% $3.3 39. Telecom NewZealand (b) NewZealand 590.6 473.3 24.8% $2.2 40. Embratel (a) Brazil 574,8 545.8 5.3% $3.1 1999 Revenue (US$ billions) Iot’l Service $4.9 $3.5 $1.3 $2.0 (est.) $1.5 $0.8 $1.1 (est.) $1.4 $0.5 $2.0 (est.) $0.1 $1.4 $0.8 $1.7 $1.0 $0.3 (est.) $0.6 $0.4 $0.4 (est.) $0.1 $0.2 (est.) $1.2 $0.2 (est.) $0.6 $0.4 (est.) $1.5 $0,2 (est.) $0.6 (est.) $0.6 $0.3 (est.) $0.1 $0.1 $0.1 $0.1 $0.6 $0.3 $0.2 (est.) $0.2 (est.) $0,2 $1 .o Note: Traffic figures are for public swztched telephone network (PSTN) c~rcuits only (service resale is excluded). Data for U.S. and U.K carriers include International S=mple Resale (ISR). Carrier rankmgs based on originating country minutes only; when based on the aggregated traffic of all subsidiaries, the top multinational carriers include: AT&T/BT (Cencert), WorldCem, and Te[egiobe. lnternat=onal service revenues generally reflect net of PSTN sen/ice revenues after adding or subtracting for settlement payments, but may also include some private line revenue All revenue figures converted from anginal currency at conversion rate current to year end reposed Some revenues figures have been estzmated (est.). a Data based on bilhng point of call, not originating point. b Data are for the fiscal year ending 31 March. Telstra and Telecom New Zealand FY ends 30 June. Source TeleGeography research, FCC, and company reports © TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeieGeo{~raphy 2001 The Top U.S. International Carriers Rank 1 2 3 4 5 6 7 8 9 10 11 12 13 15 Outgoing Facilities-Based Traffic (millions of minutes) Company AT&T WorldCom Sprint Teleglobe USA World Access Viatel Primus Telecommunications STAR Telecommunications RSL Communications Pacific Gateway Exchange Startec Global Communications IDT Corporation GTE Corporation TelefGnica Larga Distancia (Puerto Rico) Tricom USA 1999 10,816.5 8,294.9 3,714.4 1,679.7 1,129.5 901.6 868.5 785.8 389.5 284.1 207.2 151.3 60.9 42.5 41.6 19911 10,798.5 7,195.0 2,916.0 830.3 678.8 202.3 124.9 457.4 214.3 641.4 20.0 9.4 52.7 45.5 66.1 Growth 0% 15% 27% 102% 66% 346% 595% 72% 82% -56% 936% 1510% 16% -7% -37% Carrier Share of Outgoing Traffic, 1990-1999 30.0 25.0 ¯ Others Sprint ,, WorldCom =AT&T 0.0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 Note: Traffic figures are for public switched networt circuits based on billing point of call, not origina~ng point. International Simple Resale ~lSR) is included in facilities- based totals, Source: TeleGeography research and FCC carrier filings © TeleGeography, Inc. 2000 TeleGeography 2001 © TeleGeographg, Inc. 2000 A Primer on Bandwidth Exchanges The term "bandwidth exchange" has been pulled and stretched to cover the canvas of diverse business models and operational approaches. But what exactly is a band- width exchange? Part of the definitional problem originates from the complexity inherent in the telecom industry’s web of supply and demand--whether it be for cross-border tele- phone calls, intracity fiber optic connections, or access to the Internet’s cloud. A second, more obvious, obstacle to a clear definition is the newness of the bandwidth exchange business--many companies have redirected and redefined their strategic focus (and marketing programs) more than once in just a few short years of existence. Nonetheless, these bandwidth bazaars all share a common purpose: to facilitate transactions between buyers and sell- ers. Their challenge~and the source of their diversity--is to develop the best methodolo~ for facilitating each type of transaction. This ’.primer differentiates the services exchanged, as well as the role played by the matchmaker in each deal. We will begin with a brief description of the com- panies involved--including Band-X, now TeleGeography’s corporate parent. Background The traditional process of buying and selling communica- tions bandwidth--for carriers, ISPs, and multinational cor- porations alike--can be time-consuming and labor-inten- sive. The process typically requires direct negotiation over price, quality, and delivery. An exchange, however, can extend, complement, or replace all or part of a buyer’s or seller’s sales force at various stages of the process. Furthermore, bandwidth buyers can use exchanges to find quickly the best price/quality ratio on offer, and bandwidth sellers with excess capacity can earn incremental revenues with minimal effort. A bandwidth exchange may consist of a bullpen of brokers, perhaps part of a larger team of traders, who spend their days scanning price listings and phoning potentially inter- ested parties. We classify these brokerages, along with bul- letin-board operators, as "virtual matchmakers." Alternatively, an exchange may be based upon a switch con- nected to a computerized system where anonymous buyers and sellers swap traffic. Such exchanges, which have facili- ties where members interconnect their networks for physical delivery, fall under the category of "physical matchmakers." Both operational models, virtual and physical, assist buyers and sellers of bandwidth in finding counterparties and corn- pleting transactions. Of the more than 35 companies with actual or stated plans to trade bandwidth, at least twelve have facilities which route capacity between buyers and sell- ers. The remaining two-thirds of existing exchanges, accord- ingly, are virtual matchmakers. Bandwidth exchanges also differ by their own degree of involvement in the bandwidth transaction. Some exchanges are neutral, favoring neither buyer nor seller; others are cre- ated by a party to the trades. The latter category includes "market maker" exchanges--exchanges whose founders are in the carrier or capacity building business. The oft-cited market maker example is Enron, the energy company which is investing heavily in both a nationwide network build-out and the development of a bandwidth trading exchange (see Figure 1. Selected Bandwidth Exchanges). But why would an energy company enter the world of telecommunications bandwidth? The power industry’s experience with energy trading may prove applicable to the emerging bandwidth marketplace. Enron’s vision of the communications market relies on the commoditization of bandwidth, where a liquid market allows the trading of for- ward contracts and financial derivatives by bandwidth users, as well as speculators and arbitrageurs. This vision may bear fruit. Our discussion here, however, focuses on the current state of the bandwidth exchange industry and emphasizes the delivery market for bandwidth--the buying and selling of bandwidth for actual use--rather than the still mostly theoretical trading world of bandwidth financial instruments. Services To date, most bandwidth trading has focused on interna- tional telephone calls, or "minutes," and raw network con- nectivity, or "bandwidth circuits." More recently, bandwidth exchange Band-X has also begun trading lnternet transit, or "routed IP" services. Some exchanges may also broker related services to their customers, such as colocation space in carrier hotels, or empty ducts between points in a city. Minutes. Telephone carriers typically meter service to their customers in minutes or a portion thereof. Thus, because carriers around the world have agreed on standard defini- tions of a "conversation minute," they are able to route minutes easily from one network to another through their switches. © TeleGeogrophy, Inc. 2000 TeleGeography 2001 Figure 1. Selected Bandwidth Exchanges Exchange Asia Capamty Exchange (ACE) www.ace-asia.com AIG Telecom vvvwv.aigtelecom.com Arbinet-thexchange vvww.thexchange.com Band-X vvww.band-x.com Bandwidth.corn vvww.bandwidth.com Chapel Hill Broadband www.chbroadband.com Enron Broadband Services vvvwv.enron.net Global TeleXchange (TheGTX) vvww.thegtx.com RateXchange vvww.ratexchange.com Type of Matchmaker Hybrid Physical Physical Hybrid Products Currently Traded Minutes Bandwidth circuits Minutes Minutes Minutes Bandwidth circuits Internet access Colocation Switch, Router, or Hub Locations Hong Kong Los Angeles Jersey City, NJ New York London London Hong Kong New York Paris Amsterdam Dublin Notes ¯ Formed e-bandwidthtraders.org (eBTO) with TheGTX and RateXchange to promote interests of online exchanges ¯ Plans to broker options on bandwidth within next six months ¯ AIG is the buyer to every seller, and seller to every buyer ¯ Handles billing and settlement, takes on counterparty risk ¯ Automated integration of web site, switch, and OSS ¯ Plans to expand into new cities and products (e.g., bandwidth circuits) ¯ Minutes and Internet access services are routed through Band-X facilities ¯ Acts as virtual matchmaker for bandwidth circuittransactions and co- location ¯ Is deploying facilities in ten countries this year, including India, Brazil, and S. Africa, where offices are already open Virtual Bandwidth ¯ Partnered with Chapel Hill Broadband, c=rcuits which brokers wholesale capacity leads Virtual Bandwidth n.a. circuits Colocation New York Los Angeles New York London Miami Los Angeles Frankfurt Amsterdam 8 delivery hubs in North America London Frankfurt Bandwidth circuits Minutes Bandwidth circuits Physical Hybrid Hybrid Bandwidth circuits ¯ Provides carrier-neutral brokering of wholesale capacity, including dark fiber and wavelengths ¯ May also represent buyer or seller in negotiating specific contracts ¯ Intending to create a commodities market for bandwidth circuits ¯ Developing nationwide network of pooling points ¯ Brokers bandwidth circuits as virtual matchmaker ¯ Building "mesh network" of interconnected hubs to trade minutes ¯ Also intends to trade application services ¯ Operates an electronic trading system for commoditized bandwidth trading ¯ Formed alliance with brokerage Amerex Bandwidth Note: The list of bandwidth exchanges presented here is not exhaustive. City Ioeat~orts are listed in approximate order of deployment and magnitude as of Oct. 2000. Source: TeleGeography research and companyreports ©TeleGeography, Inc. 2000 TeleGeography 2OO1 © TeleGeographyo Inc. 2000 Figure 2. Common Bandwidth Trading Parameters Minutes Bandwidth Circuits ¯ Origination location (switch/city) ¯ Termination location (city or country) ¯ Quality metric (Answer Seizure Ratio, Post Dial Delay, or Call Quality Index) ¯ Rate (price) ¯ Sming/Availability (start and end date of service) ¯ Origination location (city) ¯ Termination location (city or country) ¯ Rate (price) ¯ Speed (e.g., T3/45 Mbps) ¯ Transport (e.g., fiber-optic cable) ¯ Commitment (in months) ¯ Availability (start date) Note’. Adapted from the Asia Capacity Exchange (ACE) website, www.ace-asia.com © TeleGeography, Inc. 2000 Many exchanges, including Band-X, Arbinet, and TheGTX, currently trade minutes. The total volume of minutes trad- ed through exchanges is difficult to ascertain. But our research indicates that international traffic flows through exchanges could reach 300 to 500 million minutes in 2000--about 0.5 percent of the world’s traffic. The minutes bought and sold through an exchange may include either traditional, circuit-switched minutes, or Voice- over-IP (VolP) minutes. Some exchanges specialize in IP minutes (e.g., Pulver.com’s Min-X); others allow members to specify if VolP minutes are desired (e.g., Arbinet). Those exchanges with their own switching facilities may trade cir- cuit-switched and IP minutes without differentiation if their switches route both SS7 (circuit-switched) signaling and IP protocols. Although hard numbers are difficult to come by, it is generally agreed that VoiP minutes make up a small, though growing, share of the total minutes traded. Arbinet, for example, reports that nearly ten percent of its traded traffic is Voice-over-IR Minutes offers are typically listed on exchanges on a coun- try-to-country or city-to-city basis: for example, "U.S. to India at $0.13 per minute." And although each minute is, by definition, the same 60 seconds of connect time, a minute’s quality is not standard and can vary considerably. For this reason, minutes offers frequently contain a quality metric as well (see Figure 2. Common Bandwidth Trading Parameters). Bandwidth Circuits. International carriers of voice and fax services may purchase wholesale minutes, as discussed above, to route calls to their desired destinations. Alternatively, buyers may lease or purchase bandwidth cir- cuits between points and provision switches at the ends. Thus, minutes and bandwidth circuits can be, in some instances, substitutes for each other. But there the com- parison ends. Minutes are a service which ride on a phys- ical circuit; bandwidth circuits are the capacity which may be provisioned to carry any application. A buyer purchases minutes by quantity, and bandwidth circuits by capacity, for a given time period. Bandwidth circuits are typically listed in city pairs and by potential carrying capacity: for example, New York to London at 2 Mbps. In addition to the geographic and capacity parameters, bandwidth circuit exchangers must specify their commitment period. Commitm,ents can range from one year leases to Indefeasible Rights of’Use (IRUs) for the lifetime of the facility. Bandwidth circuits take various forms. The circuit may be a satellite link, a segment of a terrestrial network, or an undersea submarine cable connection between world regions. As the medium used to carry the traffic differs, so too does the circuit’s level of provisioning. A company may purchase dark fiber, which is optical fiber not connected to transmission equipment. Alternatively, a company may pre- fer a circuit which is already "lit" to handle its application of choice. A new class of bandwidth products--wavelengths--- has recently emerged for buyers and sellers of optical fiber. A wavelength, or a single channel on an optical fiber system, is typically sold at 2.5 Gbps or 10 Gbps increments (see TeleGeography’s International Bandwidth 2000 report for a detailed explanation of the various options available). Given the wide range of bandwidth circuit increments and technologies, determining universal standards and contracts is far from a simple task. Band-X has developed a stan- dardized Service Level Agreement (SLA), which includes both provisioning time and SLA post-provisioning, for band- width circuit trades on its exchange. A few organizations, © TeleGeogra~hy, Inc. 2000 TeleGeography 2001 Blessed with a highly prized domain:name, Bandwidth.corn began : buyers with their pricinglquetes, and the two parties negotiate the operations as a bulletin-b0ard virtual matchmaker for bandwidth terms of the.contract on their own. Provisioning of the bandwidth circuit sales. The company generates enterprise;: or retail, cUs- Circuit and finandal settlement of the transaction also occur "off- tomer sales leads through its website and passes them along to line." For its role as a matchmaker, Bandwidth.corn receives a carders. Bandwidth.corn’s partnership with brokerage Chapel :Hill referral fee from the sellin~ cartier of either ten percent 0fthe sale Broadband allows:it to provide wholesale bandwidth matchmaking up-front, or two to five percent of the sale’s monthly revenue. : services as well. Bandwidth.corn a/so serves carriers andlSPs lookingto buy whole- Atypical userofthe Bandwidth.corn website may:be, for example, sale bandwidth through a partnership with brokerage firm Chapel a corporate network manager searching for a dedicated circuit Hill Broadband. Bandwidth,corn sends along these wholesale between two of its locations. The potential bandwidth buyer fills leads from :its website to Chapel Hill. which then contacts, the out a form on Bandwidth.corn’s home page where he indicates his potentialsellers ithas on hand.:When a probable matchis foun~, bandwidth needs, contact information, and any additional details Chapel Hill may continue to work with the counterparties t0. for- of interest to sellers. Bandwidth.c0m users may request informa~ mulate 1~he bandwidtli contract. Whereas Bandwidth.corn exem- tion on point-t0,point bandwidth circuits of speeds ranging from plifies:the passive, lead-generating bulletin,board mode]~ Chapel fractionalT-~ sto OC-48s. When the potential.brayer submits his Hill activeJy matches buyers and sellers through the:efforts of its request, Bandwidth,corn forwards it to the sales representatives .of brokers. Together, the two companies illustrate the wide :range of more than 50 carriers. At:~his point, the web site’s participation available virtual matchmaker services. inth~ bandwidth deal ends. :Carriers directly contact the potential : .... Virtual Matchmaker Models: Bulletin Boards and Brokers Bulletine Board i Bids "’"’"",., Physical Interconnection ....................... Information Exchange In one virtual matchmaker model, buyers and sellers post bids and offers on an electronic bulletin board. When the counterparties come to terms, they interconnect with each other, outside of the exchange. Brokers, also virtual matchmakers, may reference an electTonic bulletin board in help- ing to put buyers and sellers together. TeleGeography 2001 © TeleGeograph~l, Inc. 2000 including one organized by carrier assoi:iation CompTel, have convened to resolve the issue; not surprisingly, one of the thorniest issues has been the liquidation of damages if one party does not abide by the terms of a bandwidth deal. As with creating quality of service measurements for band- width circuits, standardizing bandwidth contracts across multiple exchanges and carriers will take time. How active is bandwidth circuit trading? As the pooling points necessary for facilities-based bandwidth circuit trad- ing are still in their early stages of development, the num- ber of deals facilitated by virtual matchmakers far outnum- ber those by physical matchmakers. From July to September 2000, brokerage Chapel Hill Broadband report- ed two to three hundred requests for wholesale capacity through the Bandwidth.corn lead-generating website. Physical matchmaker Enron, in contrast, says it will facilitate about 100 bandwidth deals this year. lnternet transit (IP routed). Exchanges may also help match buyers and sellers of upstream Internet access, or transit. To date, Band-X remains the only bandwidth exchange to provide physical matchmaker services for Internet transit. Like bandwidth circuit buyers, Internet transit buyers purchase dedicated capacity for a specific time period. However, unlike bandwidth circuits, lntemet transit capacity is not restricted to a point-to-point route. Instead, a buyer of Internet transit receives a guaranteed connection to an Internet backbone, which will carry the buyer’s traffic over various, unspecified paths to reach its intended recipient. Recipient locations are IP addresses, not physical places. The exchange itself does not provide the IP transit; rather, the exchange helps the buyer to find and connect to an "upstream" ISP. For example, a large company may require high-speed lnternet access for the next few months. Instead of negotiating with a number of ISPs, determining the best price and quality match, and signing a contract with the chosen provider for a fixed length of time, the company may enlist the services of a bandwidth exchange. Buyers can choose from various, anonymous ISPs based on price and performance, and, in some cases, can switch providers as often as once a month (for a detailed description of Band- X’s routed IP service, see Figure 4. Case Study: Band-X). Operational models The business practices of bandwidth exchanges are in con- stant flux, and many include elements of multiple business models. Keeping this in mind, we consider two operational categories: virtual and physical matchmakers. The exam- ples highlighted in the accompanying figures illustrate the individual approaches of certain exchanges. Virtual matchmaker. Virtual matchmakers help to put bandwidth buyers and sellers together without physically interconnecting the counterparties. There are two non- exclusive categories of virtual matchmakers: electronic bul- letin-board operators and over-the-counter (OTC) brokers. Electronic bulletin-board operators post sellers’ offers and .buyers’ bids on a website, speeding the information-gather- ing process for the parties involved. The posting process may be automated, with website members directly listing their prices; or, employees of the bulletin-board operator may enter the information into the site after communicating with the buyers and sellers. Generally, a bulletin-board offers a passive means of generating leads, as it is left to the interested parties to act on the bids and offers posted (see Figure 3. Case study: Bandwidth.corn). Band-X, a virtual matchmaker for bandwidth circuits, has developed a more interactive approach by holding reverse auctions on its web- site, where sellers place competitive offers on specified routes in real time. Frequently referred to as "dating services," OTC brokers search for bandwidth or minutes terms which match buyers’ or sellers’ requirements; the broker often works out the details of the deal over the phone. An OTC broker may rely on electronic bulletin-board services to find prices, and then call clients to alert them to attractive deals. Or a broker may have his own "inventory" of available bandwidth that he has collected, as an independent operator, from contact- ing sellers of capacity. In contrast to bulletin-board opera- tors, OTC brokers provide active matching services to buy- ers and sellers, often adding value to their Clients through their personal connections and effort. OTC brokers are generally carder-neutral--their concern is to complete bandwidth deals, not to favor one buyer or sell- er over another At times, however, a broker may represent one party from the bid or offer stage through to contract negotiation. For example, Chapel Hill Broadband, which most often acts as a neutral broker, may also be hired to negotiate a specific transaction on behalf of the buyer or seller. Virtual matchmakers, in addition, may facilitate the delivery of the minutes or bandwidth trade. For example, although by definition a virtual matchmaker owns no interconnect facilities, it may hold information on where its clients are colocated, and suggest locations for interconnection. As with other deals made through virtual matchmakers, these transactions are not anonymous, as both parties must work together to provision their networks for delivery. Physical matchmaker. Physical matchmakers do more than match buyers to sellers; they actively facilitate the delivery of the bandwidth deal through their own facilities. Such © TeleGeography, Inc. 2000 TeleGeography 2001 Figure 4., Case Study: Band-X Band-X Ltd. operates trading floors for four products: switched minutes, routed IP transit, telehouse cotocation, and network bandwidth (including clear channel circuits, dark fiber, ducts, and wavelengths). The first two trading floorsmswitched minutes and routed ]P--are facilities-based, meaning that buyers and sellers plug in to Band-X’s switching or routing equipment in London to complete the trade. Bandwidth and colocation services are agency-based; for these services, Band-X helps the buyer and sell- er find each other and assists with the delivery of the transaction. For bandwidth and colocation, therefore, Band-X acts as a virtual matchmaker. To understand how Band-X’s physical matchmaking works, we examine the muted IP service in more detail. "Routed IP" refers to public Internet access at specific levels of capacity. For exam- ple, a buyer of routed tP may be a smaller lnternet Service Provider (1SP) who chooses to connect to Band-X in lieu of direct- ly connecting to a larger ISR Band-X, in turn, has interconnected with participating sellers of IP capacity--including Cable & Wireless, Level 3, and Telia--and negotiated transit and peering arrangements with them. Connecting to Band-X allows the buyer to comparison shop between multiple providers by price and qual- ity metdcs, and to change providers on request as new offer infor- mation, updated daily, becomes available. Buyers gain flexibility in their choice of ISP in addition to a s~eady stream of quality mon- itoring information. Also, the buyer benefits from price declines as the market moves: if a seller reduces its price on Band-X, the buyer’s price is automatically adjusted downward to the new level Price increases, however, only take effect after a I Z~-day notice, so that buyers have the opportunity to switch to a new provider, if desired. The routed IP buyer finds his supplier of choice through Band-X’s web site, where he can examine anonymous sellers’ transit prices and associated quality indexes. Band-X derives its quality index from measurements made at over 200 representative web sites across several regions. The quality metdcs include traceroutes (the number of hops needed for a packet to reach its destination and back}, latency (packet round-trip time), pings (packet loss due to network congestion or other problems), and throughput (the rate at which traffic transits the network). Band-X updates its quality index metdcs approximately every half hour and allows routed IP buyers to view a provider’s quality measure for different geo- graphical zones and over vadous pedods of time. Thus, IP buyers may use region-specific quality measurements to choose targeted areas of coverage. Once an IP buyer has selected a provider based on his preference for price and quality, he inputs his required port type and connec- tion speed--from one to 155 Mbps---into Band-X’s web site. The website returns a pdce quote, which the buyer may accept or decline. Acceptance of the quote notifies Band-X, which confirms the transaction by phone or email. The routed IP buyer, once he has accepted the quote, works with Band-X to interconnect with Band-X’s IP exchange. Other ser- vices, including Domain Name Service [DNS) and mail back-up, may also be provided by the exchange. Band-X configures its routers for the buyer, allocating any needed IP addresses, and noti- fies the buyer when his connection is in operation. Once turned up, the buyer may view his traffic through the connection by log- ging in to Band-X’s web site. Band-X bills the buyer and settles all accounts. Band-X Model: Internet Access Website ISP Network IPExchange Physical Interconnecbon ....................... Inforrnat~on Exchange In the model illustrated, buyers and sellers of Internet access connect to a physical matchmaker’s touters. Buyers choose anonymous sellers based on offers posted on the exchange’s webszte. Source: TeleGeography research and company reports © TeleGeography, Inc. 2000 TeleGeography 2001 © TeleGeograph~l, Inc. 2000 Figure 5. Trading bandwidth on a physical level, whether it be wavelengths or STM ls, presents some unique technological difficulties. In order to switch circuits between anonymous counterparties in real time, not only must carriers resolve standardization and contractual issues~they must have the infrastructure in place that wilt allow the physical exchange and subsequent monitoring oftraded cap~c- ity. Traditional colocation or telehouse facilities may provide phys- ical locations of interconnection, but most are not equipped to per- mit trading circuits of varying increments, nor to measure the delivered quality of service. In addition, masking the identity of the bandwidth traders at traditional interconnect fadlities is gen- erally impossible. The creation of "pooling points" aims to resolve these intercon- nection and monitedng issues. Although pooling points may be virtual, which involves utilizing intermediary carriers for intercon- nection, our discussion focuses on the physical locations where car- riers plug in to each other for the purposes of exchanging band- width circuits (see "MANs: The Golden Mile" in the Facilities sec- tion of this report). Central to the pooling point is a "bandwidth manager," a piece of equipment that allows capacity to be switched remotely and automatically between buyers and sellers-- while measuring the quality of the delivered bandwidth and ensur- ing anonymity of the counterparties. Energy company Enron, a pioneer in developing bandwidth trading facilities, has built pooling points in New York and Los Angeles, and Pooling Points and Physical Matchmakers plans to have more than ten others in the United States alone this year. The company also intends to deploy pooling points overseas, including Tokyo, Amsterdam, Frankfurt, and Paris. Enron claims that its investments in pooling point infrastructure, as well as its role in convening bandwidth contract standardization meetings, are intended to speed the development of a commoditized band- width circuit trading market. Recent market entrant LighTrade Inc. focuses exclusively on deploying pooling points throughout the United States and other countries. The company started in 2000 with plans to install equipment in Atlanta, Chicago, Denver, Dallas, Miami, San Francisco, San Jose, Seattle, and Washington, D.C. The heart of LighTrade’s pooling points consists of a Lucent WaveStar Bandwidth Manager, which allows the switching of capacity at specified increments, as well as remote quality-of-service monitor- ing. While most observers agree that pooling points are a prerequisite for a commoditized bandwidth circuit trading market, the immedi- acy of their implementation is less certain. Developing the infra~ structure to support an array of circuit technologies remains a sig- nificant challenge, as does convincing carders of the benefits of interconnection. © TeteGeography, Inc. 2000 exchanges deploy switches or hubs where their members may interconnect. The decision of which switch or hub equipment to use depends upon the kind of services the exchange provides; facilities that carry minutes differ from facilities that allow bandwidth circuit trades. For minutes, switches of various shapes and sizes route calls from seller to buyer and monitor their flow. Bandwidth circuits, in con- trast, are not so easily switched between carriers. A num- ber of companies are currently developing an infrastructure for trading bandwidth circuits which provides the necessary measurement and routing capabilities. At this writing, how- ever, physical trading of bandwidth circuits remains in its infancy (see Figure 5. Pooling Points and Physical Matchmakers). A physical matchmaker may deploy one or many switches or hubs and may or may not interconnect them. Clearly, more hubs extend an exchange’s geographic reach, allowing more buyers and sellers to interconnect at a lesser cost. Some exchanges have multiple, separate facilities (e.g., Band-X); others are building a network of hubs that are linked togeth- er (e.g., TheGTX). Again, an exchange’s network architec- ture reflects the orientation of the service provided by the exchange, as well as the exchange’s preferred technological blueprint. Even if an exchange owns its own hubs, it may broker deals that are "off-hub," if it is more cost-efficient for the parties to interconnect through their own’ .facilities. Owning facilities may allow a physical matchmaker to offer anonymous trading services to its interconnected members and to measure actively the traded minutes or circuits. Monitoring of service quality, in such instances, becomes critical for buyer confidence. Physical matchmakers may post quality metrics on their website or even guarantee quality levels to buyers. Band-X has pioneered its own algo- rithm for determining the quality of IP network access that is available through its facilities (again, see Figure 4). A few physical matchmakers have developed facilities that allow transactions to happen in "real time"--instantly, with- out human intervention. Once connected to the exchange’s hub, a member of such an automated exchange can find a counterparty and provision the service delivery entirely through a Web-based interface (see Figure 6. Case Study: Arbinet). Payment and Risk As exchanges differ by service and operational model, so too do they vary by fee structure. Buyers and sellers com- pensate bandwidth exchanges through commissions on © TeleGeography, Inc. 2000 TeleGeography 2001 Figure 6. Case Study: Arbinet-thexchange Arbinet-thexchange Inc. has linked its website to its New York switching facilities and operating support system (OSS), allowing traders to buy and sell minutes automatically. Call originators and terminators who wish to use Arbinet’s auto- mated trading floor first pay $5,000 in application fees to under- go a background and credit check. Then they connect to Arbinet either directly or through an agreement with local network opera- tor FiberNet, which offers access to Arbinet’s New York switch from major area telehouses. Each month thereafter the customer pays a flat-rate connection fee ranging from about $750 to $1,100, depending on the size of their connection into the exchange’s switch. Arbinet’s fee structure also varies by the type of service level desired by the customer. But unlike many other brokers or exchanges, Arbinet’s fees now include no usage component; the company does not take a commission on any of the trades going through its switch. A buyer or seller wishing to place abid or an offer enters Arbinet’s trading floor by logging in to Arbinet’s website. The buyer/selle~ then chooses from over 20 different variables to list the bid/offer. These variables specify the desired mute, quality, and price. Geographically, Arbinet’s customers may identif~ countr~ or city endpoints for the calls (e.g., New York to Brazil). Quality metrics include ~,SR (answer seizure ratio), PDD (post dial delay), and VQoS (a qualitative ranking}. Pricing is listed at a cents per minute rate. The buyer/seller then prioritizes the variables by importance, allowing the software to search for the best match based on the customer’s specifications and rankings. Arbinet’s system searches for the customer’s "price or better" at "quality or better." In other words, if a buyer lists 4¢ pe~ minute from New York to Brazil in his bid, and a 3¢ per minute offer is available, Arbinet will match and charge the buyer at the :5¢ per minute rate. Once a match is found, Arbinet’s operating system notifies its switch, and minutes are routed automatically from one customer to another This process is anonymous; the buyer and seller do not know the identity ofthe counterparty. As the minutes flow through its switch, Arbinet monitors the quality of the calls and the con- stantly changing pdces listed on the floor. The company uses this monitoring to offer dynamic routing, where Arbinet’s server re- mutes a buyer’s calls as new options become available. A buyer’s minutes, for example, may start out being routed to carders X, Y, and Z. If carrier X drops its price considerably and its quality remains the same, the remaining buyer’s minutes would all then be automatically routed to carrier X’s network. The buyer would be aware of the decrease in pdce but would have no knowledge of the underlying call muting changes. Similarly, if the quality on a route changes, Arbinet’s switch re-mutes the calls accordingly. With the delivery underway, the web site and switch automatical- ly interface with the billing systems to settle accounts. Arbinet invoices the buyer and pays the seller, generally before receiving payment from the buyer. In this way, the exchange is taking on full "counterparty risk." The seller is guaranteed its payment, even if Arbinet never collects from the buyer Physical Matchmaker Model: Minutes Sw~tch Phys=cal Interconnect~on ....................... Informat=on Exchange A phys=cal matchmaker specializing =n m~nutes connects buyers and sellers to its sw=tch. Pnce matching may be done through the exchange’s webs~te, Source: TeleGeogrsphy research and company reports © TeleGeogrsphy, tnc. 2000 27 TeleGeography 2001 © TeleGeography, Inc. 2000 deals they facilitate, fees for membership in the exchange, or revenue from advertising or other ancillary services. Consistent with practices in the enersy industry, brokerage houses and lead generators may generate commissions only from bandwidth sellers, not buyers. But not always. Chapel Hill Broadband, for instance, has a variable fee structure that depends on the type and volume of the transaction it enables. Chapel Hill’s commission can range to as high as seven percent, if assessed on both buyer (three percent) and seller (four percent). On its switched minutes trading floor, Band-X takes a five percent total commission--two percent from buyers and three percent from sellers. For "networks," or bandwidth circuit deals, Band-X charges the successful bid/offer poster, whether buyer or seller, a two and a half percent c6mmis- sion on the first $200,000 and one percent thereafter. Some exchanges have moved away from charging commis- sions altogether. Arbinet, for example, has chosen to replace commissions on minutes trades with flat-rate capac- ity-based fees. Similarly, RateXchange has forsaken com- missions on bandwidth circuit deals in favor of increasing its market presence. Bandwidth exchanges can also take on counterparty risk in the financial transaction. In such cases, the exchange pays the seller directly, regardless of whether it is able to collect payment from the buyer. The seller, therefore, is relieved of any worries of bad debt or risk management. Exchanges which offer these clearing services "take title" to the minutes or bandwidth sold, becoming the buyer to every seller, and the seller to every buyer. For example, AIG Telecom buys blocks of minutes from the selling carrier, pays the seller, sells the minutes to the purchaser, and later invoices and collects from the buyer. To protect themselves, bandwidth exchanges often require participating members to undergo extensive background credit checks. Summary Bandwidth exchanges have a common goal: to facilitate the buying and selling of communications capacity. Most notably, exchanges help companies deal in minutes, band- width circuits, and Intemet access. Exchanges provide virtual or physical matchmaking ser- vices-or a combination of the two. Virtual matchmakers facilitate bandwidth trades without physically linking buyers and sellers. Of the virtual matchmakers, bulletin-board operators serve as passive lead generators, while OTC bro- kers play an active role in bringing counterparties together. Physical matchmakers, which have facilities where their members interconnect, may offer anonymous trading and monitoring capabilities at varying degrees of automation. Payment structures of exchanges include differing commis- sion schedules and membership fees. Bandwidth exchanges may assume counterparty risk by guaranteeing payment to sellers. Difficult as it may be to categorize bandwidth exchanges, the industry’s ongoing evolution is sure to bring greater standardization as well as significant change. ~i~ TeleGeography, Inc. is a wholly-owned subsidiary of the bandwidth exchange Band-X Ltd. The, editors of TeleGeography 2001 are solely responsible for the accura- cy and completeness of this article. Pricing TeleGeography 2001 © TeleGeograph~l, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Overview of International Pricing Trends There are not many industries in which revenues fall despite double digit volume growth. Unfortunately, developments in 1999 suggest you can count the international telecom service market among them. In the U.S., plunging prices overcame a 17 percent increase in outgoing call minutes to force down gross international service revenues for the sec- ond year in a row (see Figure 1. U.S. Carrier Revenues and Settlement Outpayments, 1980-1999). Still, the news was not all bad. Although billed revenues for outgoing interna- tional calls fell by an average of $0.06 per minute, retained revenues actually increased (see Figure 2. U.S. Carrier Revenues for International Voice Service, 1998-1999). U.S. carriers can thank rapidly falling settlement rates--the fees they must pay to foreign telcos to terminate international calls--for the dramatic cost savings. U.S. operators were not the only ones to benefit. Carriers in other markets also experienced substantial decreases in settlement costs (see Figure 4. Settlement Payments per Minute for Outgoing Calls 1998-1999). A Welcome Demise? Three significant factors accounted for the erosion of settle- ment rates. First, unilateral regulatory action by the U.S. Federal Communications Commission (FCC) may be taking effect. In 1997, the FCC adopted an order requiring U.S. carriers to negotiate settlement rates at or below prescribed "benchmark" rates. These rate levels, and their required implementation dates, varied according to the per capita income of correspondent countries. Benchmarks ranged from $0.15 per minute by January 1999 for the wealthiest countries to $0.23 per minute by 2003 for some develop- ing countries. Although settlement rates between the U.S. and numerous countries remain high, a number of routes have seen dramatic decreases. For example, the settlement rate for traffic between the U.S. and Kuwait has fallen from $0.78 to $0.15 per minute in just the last year (see the sec- tion on "International Settlements Rates" below). Even though benchmarks directly affect only U.S. carriers and their foreign correspondents, the FCC’s efforts may well be Figure 1. U.S. Carrier Revenues and Settlement Outpayments, 1980-1999 $2.00 Total Billed Revenues $12 $1.50 $1.00 $0.50 Revenue per Minute 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 Note: Excludes callsto Canada and Mexico. Source: FCC ©TeleGeography, Inc. 2000 TeleGeography 2001 © TeleGeography, Inc. 2000 Figure 2, U.S. Carrier Revenues for International Voice Service, 1998-1999 Total Receipts (US$ millions) Average Revenue per Minute (US$]minute) Billed Settlement Retained Settlement Net Billed Settlement Retained Settlement Revenue Outpayment Revenue Inpayment Revenue Revenue Outpayment Revenue Inpayment 0.31 0.40 0.23 AT&T (1998) 7,700.3 3,373.6 4,326.6 1,072.6 5,399.3 0.71 AT&T (1999) 6,677.0 2,544.4 4,132.6 730.5 4,863.1 0.62 0.24 0.38 0.15 WorldCom (1998) 4,298.1 2,308.9 1,989.2 755.5 2,744.7 0.60 0.32 0.28 0.25 WorldCom (1999) 5,051.1 2,116.9 2,934.2 548.7 3,482.9 0.61 0.26 0.35 0.18 Sprint (1998) 1,421.4 796.8 624.6 297.9 922.5 0.49 0.27 0.21 0.16 Sprint (1999) 1,379.0 776.0 803.0 221.7 824.8 0.37 0.21 0.18 0.12 Top 3 Total (1998) 13,419.8 6,479.4 6,940.5 2,126.0 9,066.4 0.64 0.31 0.33 0.22 Top 3 Total (1999) 13,107.1 5,437.3 7,669.8 1,500.9 9,170.7 0.57 0.24 0.34 0.15 Note: This table breaks down international vo=ce service revenue for the three largest U,S, international carriers. In t999, for example, AT&T collected $6.7 billion from customers for U.S. international outgoing calls, and paid foreign carriers $2.5 billion to terminate those calls. Thus, the company gained $4.1 billion by carrying U.S. outgoing calls. Because FCC regulations generally entitled each U.S. carrier to terminate incoming calls based on the percentage of U.S, outgoing traffic r~ originates, AT&T also collected a significant sum ($731 million) on foreign settlement inpayments, netting $4.9 billion on international voice service. Source: FCC carrier fitings © TeleGeogrsphy, Inc. 2000 having a wider impact. Carriers in other countries can lever- age these publicly-available rates--together with the threat of refile through the U.S.--to negotiate their own, lower rates. Second, the threat of illegal bypass may be pushing down rates. A number of countries have implemented an uneven Figure 3. AT&T Settlement Payments per Minute to Select Destinations, 1999 Canada United Kingdom Germany Italy Japan $0.00 $0.25 $0.50 $0.75 Source: FCC carrier filings ©TeleGeogrsphy, Inc. 2000 schedule for market liberalization, permitting competition for Value Added Network (VAN) and Internet services while keeping traditional Public Switched Telecommunications Network (PSTN) services in the domain of the national monopoly. In these markets, some foreign operators have sneaked incoming international traffic into the local PSTN via these VANs and the Internet, avoiding settlement pay- ments altogether. With international call ’volumes from Voice-over-IP (VolP) alone accounting for 1.7 billion minutes in 1999 and nearly four billion minutes in 2000, the mere threat of bypass may be forcing some countries to realize that high settlement rates are increasingly difficult to sustain (see the "VolP Routes and Traffic" article in this report for more information on VolP traffic flows). Finally, the spread of competition for public switched ser- vices has helped push down settlement rates. In many com- petitive markets, foreign carders seeking to land their traffic now have the option of eschewing traditional settlement rates in favor of direct interconnect fees. These charges~ usually the same fees that domestic long distance operators pay to send traffic onto local networks~often are far lower than traditional settlement rates. The difference between settlement rates in competitive and non-competitive mar- kets is dramatic (see Figure 3. AT&T Settlement Payments per Minute to Select Destinations, 1999). With two-thirds of the world’s international telephone traffic flows among countries with direct interconnect options in place, it should come as little surprise that carrier termination costs are falling. © TeleGeography, Inc. 2000 TeleGeography 2001 Figure 4. Settlement Payments per Minute for Outgoing Calls, 1998-1999 $0.70 S0.60 $0.50 $o.~o $0.20 $O.lO m1998 1999 $0.00 AT&T Japan Telecom KoreaTelecom OTE Sprint Swisscom WofldCom Source: FCC carrier filings, company reports, and TeleGeography research © TeleGeography, Inc. 2000 The Other Side of the Coin Combined, these developments mean that many carriers have been able to pass on some savings to consumers through lower international call prices, while protecting their own bottom line. Of course, not all companies have bene- fited from the changing economics of international call ser- vice provision. Some carriers in developing countries, where net traffic inflows are common, have depended on foreign settlement inpayments for a substantial portion of their total revenue. For these carriers, falling settlement rates are bad news indeed. PLDT, the dominant carrier in the Philippines, receives four times as many calls as it sends. Settlement payments from just one foreign carrier (AT&T] accounted for nearly one-fifth of all PLDT international ser- vice revenues in 1999. In January 2000, PLDT lowered its settlement rate with U.S. carriers from 33¢ per minute to the FCC benchmark rate of 19¢. With the tightening of set- tlement cash inflow and price pressure at home, interna- tional service revenue represents an ever shrinking portion of PLDT’s total revenue: 56 percent in 1997, 36 percent in 1998, and just 32 percent in 1999. Falling international settlement rates are forcing PLDT, like so many other carri- ers in developing markets, to seek revenue from other sources (e.g., residential line charges, local calling, domestic interconnect fees, and wireless services) to shore up its bot- tom line. What Lies Ahead The following pages trace the lines of this complex story in greater detail, providing sample cost data at each stage of an international call (bandwidth, settlement/interconnec- lion, wholesale rates, and retai~ tariffs). The analysis includes: ¯ International Call Costs to and from the U.S. These tables summarize the various cost elements required to complete an international call. They illustrate how inter- connection fees, compared to capacity prices, account for a relatively large and growing portion of carriage costs, and how carriers often can save money by circum- venting the traditional settlement rate mechanism where possible. ¯ IPL Lease Rates. These charts illustrate why band- width costs are a shrinking portion of call costs by track- ing how fast--and on what routes~lease rates for International Private Line (IPL) circuits have fallen. ¯ Settlement Rates. These charts compare U.S. and U.K. settlement rates over time and by destination. The FCC and ITU benchmark settlement rates are also pro- vided, showing which countries have met the U.S. crite- ria and which have not. ¯ "The New Calculus: A Primer on Interconnection Accounting." Even on competitive routes where regula- tions permit bypass of the traditional settlement rate regime, local access fees account for a large percentage of international carriage costs. This section summarizes the regulatory methodologies used to establish intercon- nect fees, and provides termination rates to 30 destina- tions in 1998, 1999, and 2000. TeleGeography 2001 © TeleGeographU, Inc. 2000 Figure 5. U.S.Carrier International Call Revenue by Destination, 1999 Average U.S. Outgoing Call Revenue by Region $1.50 $1.25 $1,00 $0.75 so.50 ¯ Settlement Payments Retained Revenue $0.25 $0.00 = E ~= = =o ,,=, ,,=, Average U.S. Outgoing Call Revenue by Country $1.50 $1.25 ¯ Settlement Payment~ Retained Revenue $1.00 $0.75 $0.50 $0.25 $0.00 Note: Cha~s show average revenue on U,S. international outgoing calls for the three largest U.S. intemationa| carriers. Total column height shows the average price for calls on a given route. RetaZned revenue equals average price minus settlement payment, and includes such components as access fees for origination, network costs, and profit. Source: FCC carrierfiiings © TeleGeography, lnc. 2000 ¯ Wholesale Rates and Retail Tariffs. So how do shifts in call component costs affect prices for the end-user? This section includes sample wholesale rates using circuit switched and Voice-over-IP (VolP) transmission and a matrix of international rates from 24 countries. ¯ "Follow the Money: Network-to-Network Payments for Internet Telephony and Other IP Traffic Streams." Settlement rate controversies are not limited to the realm of circuit switched traffic. Inter-network payments are an increasingly urgent topic in the Internet world as well. The Pdcing chapter concludes with this essay, which explores options for IP network settlements. © TeleGeography, Inc. 2000 TeleGeography 2001 Elements of an International Call An international service provider has a number of options to send its customers’ calls abroad. This section compares the cost elements of those options. Referring to the table on pages 36 to 37, let’s use a call from Washington, DC to Berlin as an example. Not including call-back, refile, and other forms of non-traditional traffic switching, a U.S. carder has five basic methods of transporting a customer’s call to its destination in Germany: I. Ownership/Settlement. To switch the call from the cus- tomer’s telephone to its own long distance network, the international carrier pays the local exchange carder (LEC] in Washington an origination fee, and then uses its own capac- ity to bdng the call to New York, where the international cable to Germany begins. Costs for the domestic portion of the call equal approximately 1¢ per minute. The carrier shifts the call onto the international "half circuit" it owns, then pays the German carder a settlement fee to transfer the call onto its matching half circuit and to the final destination. The U.S. carrier’s marginal cost of using its own backhaul and international circuit is insignificant: 0.1¢ per minute. The settlement rate, at 10.0¢ per minute, is far more expen- sive. Total cost: 11.1 ¢ per minute. 2. Lease/Settlement. A carrier is not required to own the circuits that it uses. Instead, it can lease both the domestic capacity between cities, and the half circuit to Germany. Total cost, including origination, backhaul, private line lease, and settlement payment: 1 1.8¢ per minute. 3. Ownership/Interconnect. Competition rules in Germany permit foreign carriers to interconnect directly with the domestic telephone network. Rather than financing a half circuit and paying a settlement fee, a U.S. carrier can pur- chase a whole circuit all the way to an international gateway in Germany, then pay the German carder a 2.2¢ per minute fee to switch the call to Berlin. Total cost, including origina- tion and backhauh 3.3¢ per minute. 4. Lease/Interconnect. Also known as International Simple Resale (ISR), a carrier can lease capacity to carry the call over a whole circuit from Washington to Berlin. Total cost, including origination, backhaul, pdvate line lease, and inter- connection in Germany: 4¢ per minute. 5. Service Resale. A telephone service provider may wish to avoid carrying its own traffic to Germany altogether by purchasing the minutes transported over another carrier’s network in bulk and marketing those minutes as its own. The charge required for end-to-end service resale is a "wholesale rate" covedng origination, U.S. domestic long distance, and the underlying carder’s international transport and termination charges. Total cost: 8.2¢ per minute. The following pages examine the component costs of transmit- ting an international call on selected routes, both to and from the United States. The calculations exclude Selling, General, & Administrative (SG6~) costs, which can form a significant por- tion of actual carrier expenses. ~i~.~. Figure 1. International Call Components Country A Co-~ntry O Natlon~= Network ., -. Half Circuit Half C~rcuit Local National Backhaul International Exchange Transpor~ Transpor[ Network Elements Covered by Settlement Rate Network E.ements Covered by Se~leme~.~ Rate Source: TeleGeography research © TeleGeography, Inc. 2000 35 TeleGeography 2001 © TeleCeographg, Inc. 2000 International Carrier Call Costs from the U.S. Per Minute Cost (U.S. cents) Origination Int’l Circuit Int’l Circuit Settlement Interconnect Wholesale Total Cost Ownership Lease Rate Rate Rate Cost Retail Price/ Profit (Loss) Americas U.S.-Canada (Toronto) 7.0 Own Settlement 1.0 0.04 -- 10.0 -- -- 11.1 (4.1) Own Interconnect 1.0 0.04 -- -- 0.5 -- 1.6 5.4 Lease Settlement 1.1 -- 0.1 10.0 -- -- 11.2 (4.2) Lease Interconnect 1.1 -- 0.1 -- 0.5 -- 1.7 5.3 Wholesale for resellers .... 3.0 3.0 4.0 U.S.-Mexico 39.0 Own Settlement 1.0 0.8 -- 19.0 -- -- 20.8 18.2 Own Interconnect .... n.a. n.a. Lease Settlement 1.1 -- 3.7 19.0 -- -- 23.8 15.2 Lease Interconnect ..... n.a. n.a. Wholesale for resellers 19.4 19.4 19.6 U.S.-Chile 45.0 Own Settlement 1.0 0.3 -- 35.0 -- -- 36.4 8.6 Own Interconnect 1.0 0.3 -- -- 1.8 -- 3.2 41.8 Lease Settlement 1.1 -- 1.4 35.0 -- -- 37.4 7.6 Lease Interconnect 1.1 -- 1.4 -- 1.8 -- 4.2 40.8 Wholesale for resellers .... 13.2 132 31.8 Europe U.S.-German¥ 17.0 Own Settlement 1.0 0.1 -- 10.0 -- -- 11.1 5.9 Own Interconnect 1.0 0.1 -- -- 2.2 -- 3.3 13.7 Lease Settlement 1,1 -- 0.7 10.0 -- -- 11.8 5.2 Lease Interconnect 1.1 -- 0.7 -- 2.2 -- 4.0 13.0 Wholesale for resellers .... 8.2 82. 8.8 U.S.-U.K. 10.0 Own Settlement 1.0 0.1 -- 10.0 -- -- 11.1 (1.1) Own Interconnect 1.0 0.1 -- -- 1.7 -- 2.8 7.2 Lease Settlement 1.1 -- 0.3 10.0 -- -- 11.4 (1.4) Lease Interconnect 1.1 -- 0.3 -- 1.7 -- 3.1 6.9 Wholesale for resellers .... 6.8 6.8 3.2 Notes See following page Source" TeleGeography research © TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Origination Cost Per Minute Cost (U.S. cents) Int’l Circuit Int’l Circuit Settlement Interconnect Wholesale Ownership Lease Rate Rate Rate Total Retail PriceJ Cost Profit (Loss) Asia U.S.-Australia 17.0 Own Settlement 1.0 0.6 -- 14.5 -- -- 16.1 0.9 Own Interconnect 1.0 0.6 -- -- 1.7 -- 3.3 13.7 Lease Settlement 1.1 -- 2.1 14.5 -- -- 17.7 (0.7) Lease Interconnect 1.1 -- 2.1 -- 1.7 -- 4.9 12.1 Wholesale for resellers .... 8.9 8.9 8.1 U.S.-Hong Kong Own Settlement 1.0 0.2 -- 6.5 -- -- Own Interconnect 1.0 0.2 -- -- 1.7 -- Lease Settlement 1.1 -- 2.7 6.5 -- -- Lease Interconnect 1.1 -- 2.7 -- 1.7 -- Wholesale for resellers .... 7.8 34.0 7.7 26.3 2.9 31.1 10.2 23.8 5.4 28.6 7.8 26.2 U.S.-India Own Settlement 1.0 1.9 -- 54.0 Own Interconnect ... Lease Settlement 1.1 -- 12.7 54.0 Lease Interconnect ... Wholesale for resellers 66.0 -- 56.9 9.1 -- n.a. n.a. -- 67.8 (1.8) -- n.a. n.a. 53.3 53.3 12.7 U.S.-Japan 26.0 Own Settlement 1.0 0.2 -- 14.0 -- -- 15.3 10.7 Own Interconnect 1.0 0.2 -- -- 2.4 -- 3.6 22.4 Lease Settlement 1.1 -- 3.2 14.0 -- -- 18.2 7.8 Lease Interconnect 1.1 -- 3.2 -- 2.4 -- 6.6 19.4 Wholesale for resellers .... 9.5 9.5 16,6 Notes: Prices are ind=catlve of carriers’ cost per call, but may not reflect actual costs. Selhng, General, & Admimstrat~ve (SG&A) costs are excluded. All costs expressed in U.S. cents and are exclusive of taxes. Component costs may not appear to sum to total cost due to rounding. Rates are baaed on ~nternatmnal calla originating from Washington, D C at peak hours. All rates are current as of August 2000. 0rig~natmn cost includes access charges paid to Local Exchange Carrier (Verizon! and U.S. domestic network costs for transm=ttlng calls to internat*onal gateway. Circuit ownership costs reflect half circ u~t ownership for India. All other mrcult own- ersh~p costs are for whole c~rcuits. C~rcuit ownership costs include price of backhau( Source: TeleGeography research Calculatzons converting circuit ownership prices to per minute costs assume that each 64 Kbps circuit is used for ten years and that each voice path is used four hours (240 minutes) per day. Interconnectmn rates show price for natmnal termination, except for Canada and Japan, where the regional rate is used. 0irect interconnection by foreign carriers to the domestic public switched telephone network is not permitted in India or Mexico Settlement rates are for peak rate traff=c termin ated by the largest fore=gn carrier U.S.-Mex~co settlement rates vary according to originating and terminating ]ocatmns in both countries. The s~mp~e average for all U.S.-Mexlco rates is presented here. Retail rates are based on the WorldCom One calhng plan. © TeleGeography, Inc, 2000 TeleGeography 2001 © TeleGeographg, Inc. 2000 International Carrier Call Costs to the U.S. Per Minute Cost (U.S. cents) Origination Int’l Circuit Int’l Circuit Settlement Interconnect Wholesale Cost Ownership Lease ’ Rate Rate Rate Total Retail/ Profit (Loss) Americas Canada-U.S. 13.0 Own Settlement 0.5 0.04 -- 10.0 -- -- 10.6 2.4 Own Interconnect 0.5 0.04 -- -- 1.0 -- 1.5 11.5 Lease Settlement 0.5 -- 0.1 10.0 -- -- 10.7 2.3 Lease Interconnect 0.5 -- 0.1 -- 1.0 -- 1.6 11.4 Mexico-U.S. Own Settlement 2.6 0.8 Own Interconnect -- -- Lease Settlement 2.6 -- Lease Interconnect -- -- Chile-U.S. Own Settlement 1.8 0.3 Own Interconnect 1.8 0.3 Lease Settlement 1.8 -- Lease Interconnect 1.8 -- Europe Bermany-U.S. Own Settlement 2.2 Own Interconnect 2.2 Lease Settlement 2.2 Lease Interconnect 2.2 47.9 -- 19.0 -- -- 22.4 25.5 .... n.a. n.a. 3.7 19.0 -- -- 25.3 22.6 ... n.a. n.a. 38.0 -- 35.0 -- -- 37.1 0.9 -- -- 1.0 -- 3.1 34.9 1.4 35.0 -- -- 38.2 (0.2) 1.4 -- 1.0 -- 4,2 33.8 0.1 0.1 9.2 -- 10.0 -- -- 1213 (3.1) -- -- 1.0 -- 3.3 5.9 0.7 10.0 -- -- 12.9 (3.7) 0.7 -- 1.0 -- 3.9 5.3 0.1 0.1 32.0 -- 10.0 -- -- 11.8 20.2 -- -- 1.0 -- 2.8 29.2 0.3 10.0 -- -- 12.0 20.0 0.3 -- 1.0 -- 3.0 29.0 U.K.-U.S. Own Settlement 1.7 Own Interconnect 1.7 Lease Settlement 1.7 Lease Interconnect 1.7 Notes: See following page. Source. TeleGeography research © TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Origination Cost Per Minute Cost (U.S. cents) Int’l Circuit Int~l Circuit Settlement Interconnect Wholesale Ownership Lease Rate Rate Rate Asia Australia-U.S. Own Settlement 1.7 0.6 -- 14.5 -- Own Interconnect 1,7 0.6 -- -- 1.0 Lease Settlement 1.7 -- 2.1 14.5 -- Lease Interconnect 1.7 -- 2.1 -- 1.0 Hong Kong-U.S. Own Settlement 1.7 0.2 -- 6.5 -- Own Interconnect 1.7 0.2 -- -- 1.0 Lease Settlement 1.7 -- 2.7 6.5 -- Lease Interconnect 1.7 -- 2.7 -- 1.0 India-U.S. Own Settlement 2.1 1.9 -- 54,0 Own Interconnect ... Lease Settlement 2.1 -- 12.7 54.0 Lease Interconnect ... RetaiU Total Profit(Loss) 22.0 16.8 5.2 3.3 18.7 18.3 3.7 4.8 17.2 38.0 8.4 29.6 2,9 35.1 10.9 27.1 5.4 32.6 132.0 58.0 74.0 ¯ tl,a, n.a, 68.8 63.2 g.a. n.a. Japan-U.S. 37.0 Own Settlement 2.4 0.2 -- 14.0 -- -- 16.6 20.4 Own Interconnect 2.4 0.2 -- -- 1.0 -- ", 3.6 33.4 Lease Settlement 2.4 -- 3.2 14.0 -- -- 19.6 17.5 Lease Interconnect 2.4 -- 3.2 -- 1.0 -- 6.5 30.5 Notes: Pnces are indzcat~ve of carriers’ cost per call, but may net reflect actual costs Selling, General, &. Administrative (SG&A) costs are excluded. All costs expressed m U,S cents and are excluszve of taxes. Component costs may not appear to sum to total cost due to roundzng. Retail rates are based on residential discount call plans of the largest forezgn carrier. All rates reflect tnternatlonal calls terminating in Washington, DC at peak hours and are current to August 2000. Non-U.S. carriers may own significant portions of home country local networks, in which case origination costs are counted as intra-corporate transfers. Circmt ownership costs reflect haft circuit ownership for india. All other circuit own- ersh=p costs are for whole circuits. Source: TeleGeography research C=rcult ownership costs include price of backhaul. Origination charges for India are estimated. Calculations conver~mg c[rcu=t ownership prices to per minute costs assume that each 64 Kbps c~rcmt zs used for ten years and that each voice path is used four hours (240 minutesJ per day. Direct interconnection to the U.S domestic public switched telephone network is not permitted for carriers from India or Mexico. Settlement rates are for peak rate traffic terminated by the largest foreign carrier. U S.-Mexico settlement rates vary according to originating and termmatrng/ocations in both countries. The szmple average for all U.S.-Mexico rates zs presented here. ©TeleGeography, lnc 2000 TeleGeography 2001 © TeleGeography, Inc. 2000 International Private Line Prices Figure 1. International Private Line Lease Prices from U.S., 1995-1999 $30,000 $25,000 $20,000 $15,000 $10,000 $5,ooo O E Europe O E Europe O, Africa & M East O, As a O Africa & M East (~ S America O S America O Asia W Europe World Avg Europe O W Ill World Avg O N & C America O N & C America 1996 O E Europe Africa & M East )S America Asia W Europe World Avg O N & C America Europe Africa & M East ,O E Europe ~} ~frica and M East S America S America Asia Asia ~WEuroDe WEurope World Avg World Avg N & C Amenca © N & C America 1995 1997 1998 1999 Note: Data reflect averages of annual revenue collected by U.S. international carriers for 64 Kbps circuit leases to countries within each region. Source; FCC carrier filings and TeleGeography research © TeleGeography, Inc, 2000 Figure 2. Band-X Bit Index, 1998-2000 120.0 100.0 Oct Nov Dec I Jan Feb Mar Apr May Jun Jul Aug Sap Oct Nov Dec IJan, Feb Mar Apr May Jun Jul Aug 1998 1999 2000 Note: The Band-X Bit Index measures relative price movement for one-year E-I or T-1 circuit leases (depending upon the geographic area) on major routes. This chart summar}zes index values into regional indices, based on simple averages for the following groupings: Eurepe-Asia (London to Sydney, Hong Kong); Trans-Pacific (Los Angeles to Beijing, Hong Keng, To~o);Trens~Atlarcdc {New Yorkto Frankfurt, London, Moscow); |ntra-Eurepe {London to Amsterdam, Brussels, Frankfurt, Madrid, Milan, Paris); Composite (all tracked routes). Source: Band-X Ltd. (vwwv.band.x.com) © TeleGeography, ln~, 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 International Settlement Rates United States Destination 1998 1999 2000 Andorra 0.29 0.29 0.28 Argentina 0.35 0.33 0.19 Australia (Teistra) 0.15 0.15 0.15 Austria 0.13 0.14 0.14 Bahamas 0.30/0.15 0.30/0.15 0.15 Bahrain 0.65 0.55 0.55 Bangladesh 0.80 0.80 0.67 Belarus 0,43 0.43 0.35 Belgium 0.14 0,14 0.14 Bolivia 0.46 0.43 0.29 Brazil 0,33 0,30 0.19 Canada 0.10/0.06 0.10/0.06 0.10/0.06 Chile 0.35 0.35 0,35 China 0,70 0.58 0.50 Colombia 0.40 0.38 0.33 Costa Rica 0.35 0.29 0.28 Croatia 0.28 0.25 0.22 Cyprus 0.38 0.37 0.37 Czech Republic 0.28 0.18 0.18 Denmark 0.11 0.11 0.12 Dominican Republic 0.30 0.26 0.19 El Salvador 0.39 0.31 0.27 Finland 0.16 0.14 0.14 France 0.11 0.10 0.10 French Polynesia 0,70 0.70 0.70 Germany 0.11 0.10 0.10 Ghana 0.50 0.50 0.38 Greece 0.28 0.17 0.13 Guyana 0.85 0.85 0.85 Hong Kong 0.36 0.07 0.07 Hungary 0.28 0.19 0.16 Iceland 0.24 0.14 0.14 india 0.64 0.64 0.54 Indonesia 0.53 0.48 0,25 Iran 1.05 0.90 0.78 Ireland 0.11 0.10 0,10 Israel 0.30 0.15 0.15 Italy 0,11 0,11 0,11 Japan 0.15 0.14 0.14 Jordan 0.68 0.50 0.50 United Kingdom 1998 1999 0.22 0.13 0.86 0.56 0.28/0.17 0.24/0.08 0.20 0.19 0.38 0.36 0.82 0.64 0.99 0.97 0.35 0.34 0.10 0.10 0.90 0.89 0.49 0.36 0.I0/0.08 0.10/0.04 0.90 0.89 1.08 0.89 0.90 0.56 0.69 0.47 0.33 0.33 0.25 0.20 0.21 0.20 0.10 0.07 0.67 0.56 1.54 1.18 0.15 0.13 0.11 0.10 1.64 ’ 1.27 0.10/0.08 0.10/0.04 0.66 0.52 0.29 0.24 0.90 0.89 0.45 0.42 0.18 0.18 0.23 0.23 0.95 0.87 1.21 0.64 1.21 1.18 0.16 0.16 0.25 0.24 0.16 0,13 0.59 0.48 1.21 0.97 Notes: 1. All rates expressed ~n US$. Equivalent dollar values are presented for accounting rates that are estabhshed in Specml Drawing R=ghts (SDRs), gold francs, or pounds sterling 2. The average U.S. settlement rates for 1998 are wmghted by the total minutes bet~veen the U.S. and each location in that year. Rates ~n subsequent years are for August 1999 and July 2000. 3. Where two rates are shown, there are peak]off-peak rates or growth-based rates (traffic above a benchmark level is ehgible for a lower rate). 4 Rates are for the largest carrier serving the route. D~fferent settlement rates may apply to competing carriers © TeleGeography, Inc. 2000 TeleGeography 2OO1 © TeleGeographg, Inc. 2000 United States Destination 1998 1999 Kazakhstan 0.69 0.51 Korea, Rep. 0.43 0.36 Kuwait 0.80 0.78 Luxembourg 0.14 0.14 Macau 0.50 0.50 Malaysia 0.40 0.40 Mexico 0.37 0.19 Moldova 1.04 1.04 Netherlands 0.10 0.07 New Zealand 0.13 0.14 Norway 0.09 0.08 Oman 0.75 0.68 Pa kista n 0.60 0.60 Panama 0.48 0.45 Paraguay 0.50 0.40 Peru 0.43 0.33 Philippines 0.36 0.29 Poland 0.28 0.21 Portugal 0.22 0.15 Russia 0.40 0.35 Saudi Arabia 0.87 0.68 Singapore 0.26 0.15 SIovak Republic 0.29 0.29 SIovenia 0.35 0.34 South Africa 0.40 0.35 Spain 0.13 0.14 Sri Lanka 0.80 0.60 Sweden 0.06 0.06 Switzerland 0.14 0.14 Taiwan 0.23 0.23 Thailand 0.45 0.35 Turkey 0.38 0.33 Ukraine 0.50 0.50 United Arab Emirates 1.00/0.65 1.00/0.65 United Kingdom 0.11/0.07 0.11/0.07 Uruguay 0.43 0.33 U.S. n.a. n.a. Uzbekistan 0.70 0.60 Venezuela 0.40 0.32 Vietnam 1.65/1.00/0.93/0.85 0.78 Yugoslavia 0.38 0.34 Note: All rates are expressed in US$. Source: FCC and 0FTEL 2000 0,34 0.26 0.15 0.14 0.15 0.19 0.19 1.04 O.07 0.14 0.08 0.60 0.51 0.28 0.40 0.25 0.19 0.19 0.15 0.30 0.67 0.15 0.20 0.34 0.30 0.14 0.60 0.06 0.14 0.15 0.30 0.27 0.22 0.14 0.10/0.06 0.19 n,a, 0.45 0.23 0.67 0.30 United Kingdom 1998 1999 0.82 0.64 0.64 0.50 0.82 0.80 0.23 0.24 0.53 0.51 0.51 0.50 0.45 0.44 0.30 0.24 O.O7 O.O6 0.20 0.19 0.07 0.07 0.82 0.80 0.66 0.64 0.77 0,64 0.90 0.80 0.74 0.72 0.49 0.48 0.28 0.25 0.23 0.18/0.14 0.37 0.27 1.27 0.89 0.59 0.32 0.19/0.10 0.19/0.10 0.18/0.16 0.16/,0.09 0.66 0.48 0.16 0.16 0.90 0.89 0.12 0.12 0.08/0.07 0.08/0.04 0.57 0.44 0.82 0.80 0.32 0.30 0.31 0.29 0.49 0.32 n.a. n.a. 0.97 0.95 0,12/0,08 0.12/0.04 0.99 0.80 0.82 0.80 1.31 1.29 0.28 0.27 © TeleGeography, Inc. 2000 © TeleGeograph~t, Inc. 2000 TeleGeography 2OO1 FCC and ITU Settlement Benchmarks Motivated by the annual multi-billion dollar settlements out- flow of U.S. carriers, the Federal Communications Commission (FCC) proposed in 1996 a set of "benchmark" or model set- tlement rates. Beginning in 1999, these benchmarks capped the amount U.S. carriers could pay their foreign correspon- dents for traffic exchange at rates ranging from $0.15 to $0.23 per minute. The FCC calculated benchmarks based on the price for the three network elements used to provide inter- national phone services, including international transmission facilities, international switching facilities, and national exten- sion facilities (domestic transport and termination). The FCC adopted the Benchmarks Order in August 1997, with implementation staggered over several years, based on national incomes. Settlement rates to high and upper-middle income countries have already been affected, following the 1999 deadline. As the table below demonstrates, most coun- tries in the upper income bracket have adopted settlement rates at or below benchmarks. Settlement rates for countries that have already met FCC benchmarks are shown in bold. Separate from the FCC’s efforts, a Focus Group of the International Telecommunication Union (ITU) issued a recom- mended set of "indicative target" settlement rates in November 1998. The Focus Group established seven bench- mark brackets based on country teledensity, with separate categories established for small island states and least devel- oped countries (LDCs). Adopted in June 1999, the ITU set- tlement targets were calculated using the average of the low- est 20 percent of published settlement rates for each brack- et. Initially, the ITU’s proposed rates ranged well outside the FCC’s prescribed band--from $0.06 to $0.45 per minute compared to the FCC’s $0.15 to $0.23 However, as the average of the lowest 20 percent is recalculated annually, the current targets ($0.05 to $0.30) are now much lower than when first established, particularly for countries in the low teledensity brackets. The settlement rate targets take effect December 31, 2001, with an extension to 2004 for LDCs. FCC Benchmarks and ITU Target Recommendations (U.S. cents) ITU Target Rate ITU Target Bate FCC Settlement August 2000 Country 1998 2000 Benchmarks Settlement Rate with U.S. Upper Income Bracket for FCC Benchmarks: Effective January 1999 Australia 6.0 4.9 TS.0 I5.0 Austria 12.0 10.7 15.0 14.0 Bahamas 12.0 15.5 15.0 15.0" Belgium 6.0 4.9 15.0 14.0 Denmark 6.0 4.9 15.0 12.0 France 6.0 4.9 15.0 10.0 Germany 6.0 4.9 15.0 10.0 Hong Kong 6.0 4.9 15.0 7.0 ireland 12.0 10.7 15.0 10.0 Israel 12.0 10.7 15.0 15.0 Italy 12.0 10.7 15.0 11.0 Japan 12.0 10.7 15.0 14.0 Kuwait 16.0 14.9 15.0 15.0" Netherlands 6.0 4.9 15.0 7.0 New Zealand 12.0 10.7 15.0 14.0 Norway 6.0 4.9 15.0 8.0 Portugal 12.0 10.7 15.0 15.0 Singapore 6.0 4.9 15.0 15.0 Spain 12.0 10.7 15.0 14.0 Sweden 6.0 4.9 15.0 6.0 Switzerland 6.0 4.9 15.0 14.0 Taiwa n 6.0 4.9 15.0 15.0* United Arab Emirates 12.0 10.7 15.0 14.0" United Kingdom 6.0 4.9 15.0 10.0/6.0 TeleGeo~raphy 2001 © TeleGeography, Inc. 2000 FCC Benchmarks and ITU Target Recommendations (continued) ITU Target Rate ITU Target Rate FCC Settlement August 2000 Country 1998 2000 Benchmarks Settlement Rate with U.S. Upper Middle Income Bracket for FCC Benchmarks: Effective January 2000 Argentina 16.0 14.9 19.0 19,0" Barbados 16.0 15.5 19.0 50.0 Brazil 19.0 15.3 19.0 19.0" Chile 16.0 14.9 19.0 35.0 Czech Republic 12.0 10.7 19.0 18.0 Greece 6.0 4.9 19.0 13.0 Hungary 12.0 10.7 19.0 16.0 Korea, Rep. 12.0 10.7 19.0 26.0 Malaysia 16.0 14.9 19.0 19.0" Mexico 19.0 15.3 19.0 19.0 South Africa 19.0 15.3 19.0 30.0 Trinidad 16.0 14.9 19.0 41.5 Uruguay 16.0 14.9 19.0 19.0" Lower Middle Income Bracket for FCC Benchmarks: Effective January 2001 Colombia 19.0 15.3 19.0 33.0 Costa Rica 16.0 14.9 19.0 28.0 Dominican Republic 29.0 19.1 19.0 19.0" Ecuador 29.0 19.1 19.0 34.0 El Salvador 29.0 19.1 19.0 27.0 Guatemala 29.0 19.1 19.0 29.0 Indonesia 35.0 21.9 19.0 25.0 Jamaica 19.0 15.3 19.0 ", 35.0 Jordan 29.0 19.1 19.0 50.0 Panama 19.0 15.3 19.0 28.0 Peru 29.0 19,1 19.0 25.0 Philippines 35.0 21.9 19.0 19.0" Poland 16.0 14.9 19.0 19.0" Russia 19.0 15.3 19.0 30.0 Thailand 29.0 19.1 19.0 30.0 Turkey 16.0 14.9 19.0 27.0 Venezuela 19.0 15.3 19.0 23.0 Lower Income Bracket for FCC Benchmarks: Effective January 2002 China 29.0 19.1 23.0 50.0 Egypt 29.0 19.1 23.0 45.0 Guyana 29.0 19.1 23.0 85.0 Halt= 45.0 29.6 23.0 50.0 Honduras 35.0 21.9 23.0 39.0 India 35.0 21.9 23.0 54.0 Kenya 45.0 29.6 23.0 55.0 Nicaragua 45.0 29.6 23.0 36.0 Pakistan 35.0 21.9 23.0 51.0 Vietnam 35.0 21.9 23.0 67.0 Notes: Rates became comphantw~th FCC benchmarks in 2000 are noted w~th an asterisk (*). ITU target rates are established in Spemal Drawing R~ghts (SDRs). Equivalent U.S. dollar values are subject to exchange rate adjustments. Source: FCC and ITU © TeleGeograph,/, inc. 2000 © TeleGeographyo Inc. 2000 TeleGeo~raphy 2001 National interconnection Rates Argentina Australia Austria Belgium Canada Chile China Colombia Denmark Finland France 6ermany Hong Kong Hungary Ireland Israel Italy Japan Luxembourg Mexico Netherlands New Zealand Norway Peru Portugal Spain Sweden Switzerla nd U.K. U.S. (Verizon) Fixed to Mobile Local Termination Regional Termination National Termination Termination (U.S. cents) (U.S. cents) (U.S. cents) (U.S. cents) 1998 1999 2000 1998 1999 2000 1998 1999 2000 2000 n.a. 2.35 1.10 n.a. 2.35 1,10 n.a. 2.35 1.10 n.a. 1.62 2.15 0.82 5.30 2.18 1,42 9.90 4.00 1.65 n.a. 2,00 1.90 0.97 ZOO 1.90 1.46 2.63 2.50 2.15 22.48 1.23 1.11 0.78 2.33 1.87 122 3.26 2.67 1.58 18.00 n.a. n.a, n.a, 5.33 0.78 0.51 n.a. n.a. n.a. n.a. n.a. n.a. 129 n.a. n.a. 1,79 n.a. n.a. 1.79 n.a. n.a. n.a, 1.50 n,a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 2.82 n.a. n.a. 2.82 n.a. n.a. 2.82 n.a. 1.09 1.03 0.81 2,02 1.92 1.52 2.46 2.33 1.83 17.00 1.56 1.67 1.36 1.58 1.67 n.a. 3.12 4.12 1.44 n.a. 0.78 0.63 0.56 1.90 1,56 1,13 2.80 2.32 1.69 20.00 1.10 1.05 0.83 1.88 2.26 1,80 2.86 2.74 2.18 24.00 n.a. 1.65 1.65 n.a. 1.65 1,65 n.a. 1.65 1.65 0.13 n.a. n.a. 6.61 n.a. n.a. 6,61 n.a. n.a. 6.61 3.79 2.44 1.08 0.98 4.61. 1.67 1,41 8.75 2.36 1.93 n.a. n.a. 0.80 0.80 n.a. 1.30 1,30 n.a. 2.50 2.50 n.a. 1.68 1.03 0.98 2.74 1.86 1,55 n,a, 2,69 2.19 23.00 1.81 1.74 1.54 3.73 3.31 2.38 n,a. n.a. n.a. 29.99 2.23 2.34 1.43 2.23 2.34 1,43 2,23 2,34 1.43 n.a. n.a. n.a. n.a. 5.32 2.61 2,61 n,a. n.a. n.a. 18.00 1.30 1.16 0.91 1.78 1.74 1,30 2.29 2.11 1,39 18.00 n.a. 1.43 1.38 n.a. n.a. n.a. n.a. 6.66 n.a. n.a. n.a. 1.00 0.82 n.a. 1.38 1,17 n.a. 1.63 1.75 15.60 n.a. 2.90 1.68 n.a. 2.90 1,68 n.a. 2.90 1.68 15.62 1.33 2.87 0.63 2.63 5.74 1,24 19.98 11.48 2.15 3.00 1.65 1.03 0.86 1.65 1.66 1,44 4.63 3.20 2.55 20.00 1.27 0.77 0.62 1.96 1.07 0,82 2.68 1.52 1.00 2.33 n.a. n.a. n.a. 2.72 1.50 0,32 3.73 2.08 0.59 29.54 0.60 0.62 0.56 0.96 0.82 0,82 1.73 1.76 1.68 20.42 n.a. n.a. n.a. 0.79 1.20 0,97 n.a. n.a. n.a. 2.64 Notes: All ~nterconnection charges are for peak period. All rates are established in national currenmes. Equivalent U.S. dollar values are sublect to exchange rate fluctuation. Local termination ts the lowest level of interconnect~on, typically giwng a carrier access to a srngle town or part of a city Regional and national terrninat~on are also known as s~ngle tandem and double tandern terrn=nat~on, Regional terrn~nat~on generally g~ves a carrier access to all subscribers within a metropohtan area or a North American area code. L/.S. termination fees vary according to Local Exchange Carrier (LECI. U,S. average for regional terrninat~on was 1.05¢ as of July 2000. Source: National regulatory agencies, OECD, and ITU © TeleGeography, Inc 2000 TeleGeography 2001 © TeleGeographg, Inc. 2000 The New Calculus A Primer on lnterconnection Accounting The Local Money Pit Transmitting international phone calls was once an expen- sive proposition. But technology and deregulation have pushed down long-haul transmission costs to the point that it is now the beginning and end of a call--were calls are handed to and from a local operator--that incur the great- est expense (at least on competitive routes that are well- served by modern telecom infrastructure}. After Selling, General, and Administrative (SG&A} costs, therefore, the majodty of call costs on these routes fall outside the direct control of the service provider (see Figure 1. Call Costs from U.S. to U.K. and U.S. to Japan}. How, then, can interna- tional service providers anticipate their termination and origination costs? In most markets, the local exchange carrier (LEC) ultimate- ly determines its own network access rates; however, it is often the regulator that prescribes the framework for calcu- lation. In order to make this process efficient, transparent, and responsive to changing market conditions, national reg- ulatory agencies (NRAs} have sought to establish a clear methodology for setting interconnect rates. Most NRAs have worked, in principle, to ensure that fees reflect actual local carrier costs. That may seem fair enough. But the tougher question is: Which "actual" costs should be count- ed? Should such costs reflect only per minute transmission costs incurred when sending calls over a network? Or should interconnect fees compensate LECs for a wider range of expenses, such as network upgrades and expansion? Many NRAs are wrestling with these questions, and some have proposed diverging methodologies to deal with them. This article clarifies the differences among interconnection regimes and draws out the assumptions of the underlying methodologies. The Old and New Regimes Interconnect rates compensate local carriers for use of their networks by long-distance operators. Thus, these access charges perform a similar function to the traditional settle- ment rate system employed by international carriers. There are, however, some important differences. First, under the settlement rate regime, an international carrier theoretically hands off a call at the mid-point of an international circuit, or between "half circuits." The originating carrier then relies upon its terminating counterpart to switch the call onto its domestic long-distance network to the final destination. In contrast, direct interconnection requires tb.e originating operator to carry a call all the way to the national gateway in the country of termination, through a "whole circuit." In the long term, the originating operator may wish to build Figure 1. Call Costs from U.S. to U.K. and U.S. to Japan U.S. to Japan U.S. to U.K. Settlement System Direct Interconnect Origination 12% Transpon 0rlglnatlol~ 1% 31% Settlement System 0ng~na~on 8% Transpo~ 2% Direct Interconnect Transpo~ 3% Termination Terminabon Termlnalion 87% 66% 90% 65% Total cost: 8.6€ Total cost: 2.8€ Total cost: 12,8€ Total cost: 2.6¢ Source: TeleGeography research © TeleGeography, Inc, 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 out its own long-haul network within the country, pushing the point of interconnection closer to the final call destina- tion. Interconnection regimes exist in most countries with competitive markets, where a single operator no longer monopolizes both long distance and local service. Second, settlement rates are generally one size fits all--no matter where inside a country a carrier sends the call, the rate is the same. On the other hand, interconnect fees tend to be distance sensitive, often divided into three bands. Although definitions vary by country, typical rate classifica- tions are local (calls handed off at the local exchange), sin- gle transit or single tandem (calls terminated within the metropolitan area or within a set regional distance), and double transit or double tandem (calls handed off at the national level). For example, a carrier sending traffic to France would pay France T~l~com 1.7¢ per minute to deliv- er a call from an international gateway in Paris to a business in Marseille. If that same operator carried the call on a leased line all the way to the local switching office in Marseille, it would only pay only a 0.6¢ per minute fee. Finally, settlement rates tend to be far more expensive than interconnect rates. The settlement rate between Canadian operator Teleglobe and U.S. carriers is $0.I0 per minute, whereas the regional interconnect rate in Canada is half of that. Nevertheless, when using direct interconnection, orig- ination and termination fees still account for a huge portion of per minute call costs. In the setUement rates model for originating carriers, termination charges can comprise near- ly 90 percent of call costs, edging down to 65 percent when paying direct interconnect fees (again, see Figure I). Methodologies Most governments have, in some form, decided that inter- connect fees should reflect the actual costs of network usage and development. But what is the best way to do that? Economic theory suggests that interconnection rates should be set to achieve the most efficient level of financing for net- work maintenance and expansion. If the rate is too high, competitors will have a strong incentive to build their own networks, creating excessive infrastructure. If too low, com- petitors will have little incentive to build and maintain net- works. To strike an efficient balance, regulators have devel- oped methods to establish prices systematically and trans- parently, and then to update those prices continuously. Although NRAs employ dozens of different methodologies, most can be categorized into three broad groupings: short run marginal cost (SRMC), fully distributed cost (FDC), and long run incremental cost (LRIC). The latter of the three, LRIC, has been the most widely adopted cost calculation method. Short Run Marginal Cost (SRMC) Short Run Marginal Cost (SRMC), also known as short run incremental or variable cost, measures the cost of resources required for the production of one more unit of output. In this case, SRMC reflects one additional minute of call time on a local network. SRMC takes into account only "direct" and "current" costs. Direct costs are exclusively related to the particular service (e.g., switches specifically designated to interconnect with long-distance networks). Common or joint costs, which are shared among ,different activities and services, are not considered. Curreht costs exclude the Figure 2. A Trade Dispute for the 21st Century Although the rules for compensa~ng re/cos for the few miles of cop- per wire that runs between a local switch and their customers may seem a parochial concern, these rules have emerged as a hot global trade issue, One long-standing dispute between the U.$. and Japan over Japanese local incumbent Nippon Telephone and Telegraph’s (NT~) high interconnection rates was resolved in July 2000. According to the U.S. government negotiators, Japan was protecting state-owned NTT through high switching termination charges; NT/applied the revenue from the high fees toward overseas expansion. In 1999, NTT’s-local switch termination Charges were over 60 percent higher than the world average, and regional switch termination charges were more than 70 percent above the average, In the July agreement, N’[-r agreed to a schedule of reductions that lowered its regional switch rates 55 percent and local rates 19 percent by 2002. A similar disa~eement--this time between the U.S. and Mexico-- heated up just as the U.S.-Japan dispute was cooling off, Telmex, the incumbent local and long-distance operator in Mexico, had made some downward adjustments to its domestic interconnection rates (Mexico does not allow direct international carrier interconnect) since the market was opened to long distance competition in 1998. But the new termination rates (one cent above the world average 1.6¢) still did not align with some U,S.-owned carriers’ interpretation of ,Mexico’s market openin{~ commitments made in the 1997 World Trade Organization (WTO] agreement on basic telecom services. As a result, the U~S. government began the first steps to filing a formal complaint with the WTO in July 2000, although the process remained in the "consultation" phase at this writing. If an a~reement can not be reached, the U.S. may bring the matter before a WTO panel for judgement. If the padel rules al~ainst Mexico, the U.S. may invoke trade sanctions on Mexico. © TeleGeography, Inc, 2000 TeleGeography 2001 © TeleGeograph~l, Inc. 2000 Figure 3. EU Best Practices The European Union [EU} decided in 1997 to transfer policy mak- ing for network interconnection and telecom competition from the domain of Member States to the Community level. As par~ of this effort, the European Parliament and Council issued a directive on intemonnection in telecommunications to promote fair competition, following the liberalization of the European telecommunications market in 1998. Although the directive concluded in favor of a rate system based on long-run incremental cost analysis, it also acknowledged that the establishment of methodological parame- ters and the research necessary to complete the model could take several years. In the intedm, the EU is using "best practice" rates as a check on excessively high interconnect fees. The process for establishing these rates is simple: the EU Commission collects local, single tran- sit, and double transit interconnect rates from the fifteen member states; fees from the three lowest cost member states in each inter- connection band form the range for best current practices. For the year 2000, these rates were drawn from the following countries: France, Sweden, and the U.K. {local interconnection}; and, the Netherlands, Sweden, and the U.K. {both single and double transit interconnection}. These best practice rates are recommended but not enforced, intended to shape the ongoing debate over intercom nection. For the March 2000 amendment to the EU’s interconnec- tion recommendations, see: www.ispo.cec.belinfosoc/te[ecompoli- cy/enirec2OcOen.pdf, © TeleGeography, Inc. 2000 operator’s cost of setting up the network (fixed costs) or for future maintenance, upgrades, and expansion. Cdtics of this method argue that this interconnection rate would be too low to maintain any level of growth, because the LEC would not recover enough of its very substantial fixed and network costs. Competitors would not have suffi- cient incentive to build their own networks, either, because of the low cost at which they can access the incumbent’s network. Not surprisingly, SRMC is often trumpeted by long-distance operators in newly-opened markets as the methodoloBy of choice. Despite this support, SRMC has not been applied by NRAs in any markets as of this writing. Fully Distributed Cost (FDC) Fully Distributed Cost (FDC), also known as fully allocated cost, is much more inclusive than SRMC. Beyond account- ing for direct costs, FDC also covers common costs. In addi- tion, FDC often employs historical accounting methods, cal- culating costs over past expenditures rather than present or future ones. Historical accounting tends to benefit incum- bent carriers because rates calculated using past costs are usually higher than those based on current or future costs, as costs overall have continuously declined. One of the major criticisms of FDC is that it encourages operation and investment inefficiency by local carriers, because fees are linked tightly to spending. When the U.K. introduced a telecom duopoly in 1984 and the system of rate accounting was examined for the first time, the Office of Telecommunications (OFTEL) determined that local calling prices were being subsidized by overpriced long distance rates. Instead of allowing British Telecommunications (BT) to increase its local line rate to better reflect the balance of costs, OF[EL implemented an FDC methodology that allowed BT to recover some of its lost profit through interconnection fees. Then, after local and long distance tariffs had been rebalanced by the late 1990s, OFTEL again shifted its accounting system to what they deemed a more appropriate long term solution: LRIC. This third methodology is described below. Long Run Incremental Cost (LRIC) Several NRAs have attempted to find a middle ground between SRMC’s lack of incumbent cost recovery and FDC’s strong compensation for historical costs in the Long Run Incremental Cost (LRIC) methodology. LRIC is similar to SRMC inasmuch as sunk costs--unrecoverable, past, fixed costs to build the network are not considered. The differ- ence between the two methodologies hinges on the time frame for cost inclusion. While SRMC considers the margin- al cost of a minute of traffic based on the network in the short run, LRIC considers the long run possibility of techno- logical improvements and capacity increases. In fact, LRIC defines the "long run" as the period of time in which alter- ations or expansions to the network can be implemented. Therefore, LRIC includes not only the marginal cost of today’s network, but also the marginal cost of developing a network. The methodology, therefore, tends to be friendlier to rate-payers than FDC, but less harsh to incumbents than SRMC. The basic structure of LRIC allows room for precise adapta- tion to each market. Models may differ over actual classifi- cation of specific direct, current, and potential future costs, as well as definitions of the "long run." Distinct LRIC vari- ations exist as a result--LRAIC (average), TSLRIC (total ser- vice), TELRIC (total element)--while all share fundamental principles of long run incremental cost. This shift from fully distributed cost accounting to LRIC methodologies may explain some of the recent decreases in termination rates (see Figure 4. Regional Termination Rates, 1998-2000). For example, Danish interconnect fees have dropped over 20 percent since the introduction of LRIC in 1999, compared to a six percent drop the previous year. Canadian interconnect rates have also declined over 90 per- cent since the implementation of an incremental cost methodology in 1998. Across the board, origination and © TeleGeography, Inc. 2000 TeleGeography 2001 Figure 4. !Regional Termination Rates, 1998-2000 6¢ 0¢ m1998 1~1999 1:32000 Source: TeleGeography research © TeleGeography, Inc. 2000 termination rates continue to decrease at a faster pace with LRIC implementation than with FDC. Conclusion As the price of international bandwidth continues to decline, the costs at either end of a call are becoming a more impor- tant portion of international service providers’ costs. Regulators have only recently begun to scrutinize these costs and develop long term rate-setting policies. Though carriers and regulators agree that rates should be tied to local carriers’ actual costs, the precise methodolo~ by which to calculate those costs remains a matter of debate. Today, most countries implement an LRIC strate~ to allow for adequate compensation of costs for incumbent local car- riers while offering a fair and competitive rate for interna- tional service providers. TeleGeography 2OO1 © TeleCeographg, Inc. 2000 Retail Prices for a Three Minute Call From/l’o Australia Belgium Canada Czech Rep. Finland France Germany Greece Hong Kong Ireland Italy Australia n.a, 1.97 1.42 2.85 1.83 1.83 1.68 1.55 1.26 1.40 1.36 Austria peak 0.86 0.81 0.86 0,65 0.81 0.81 0.65 0.86 0.86 0.81 0.65 Austria off-peak 0.77 0.68 0.77 0.54 0.68 0.68 0.54 0,77 0.77 0.68 0.54 Belgium peak 2.66 n.a. 1.33 1.99 1.33 1.00 1.00 1,33 2.66 1.33 1.33 Belgium off-peak 2,66 n.a. 1.33 1.99 1.33 1.00 1.00 1.33 2.66 1.33 1.33 Czech Rep. peak 1.77 1.14 1.70 n.a. 1.20 1.14 1.07 1,20 3.35 1.20 1.14 Czech Rep. off-peak 1.14 0.69 1.07 n.a. 0.76 0.69 0.63 0.76 2.65 0.76 0.69 Denmark peak 1.82 1.05 1.03 1.17 0.59 1.04 0.71 1.32 3.20 1.05 1.03 Denmark off-peak 1.40 0.81 0.71 0.94 0.48 0.68 0.46 0.95 2.91 0.85 0.65 Finland 1.90 1.27 1.27 1.27 n.a. 1.27 1.12 1.27 3.39 1.27 1.27 France peak 1.25 0.43 0.49 0.90 0.43 n.a. 0.43 0.68 1.25 0.43 0.43 France off-peak 1.12 0.42 0.51 0.86 0.42 n.a. 0.42 0.68 1.12 0.42 0,42 Germany 0.62 0.29 0.29 0.57 0.29 0.29 n.a. 0.57 "1.85 0.29 0.29 Greece 0.86 0.86 0.86 1.22 0.86 0.86 0.86 n.a. 1.22 0.86 0.86 Ireland peak 1.91 0.85 0.65 1.08 1.08 0.85 0.85 1,08 1,91 n.a. 1.08 Ireland off-peak 1.53 0.75 0.57 0.94 0.94 0.75 0.75 0.94 1.53 n.a. 0.94 Italy peak 2.38 0.92 0.92 1.29 0.92 0.92 0.92 0.92 2.96 0,92 n.a. Italy off-peak 2.38 0.92 0.92 1.29 0.92 0,92 0.92 0.92 2.96 0.92 n.a. Japan peak 6.43 8.14 4.07 8.14 8.14 5.49 5.49 8.14 8.14 8.14 8.14 Japan off-peak 5.11 6.81 3.41 6.81 6.81 5.30 5.30 6.81 6.81 6.81 6.81 Korea, Rep. 3.00 3.56 3,92 3,56 3.56 3.53 3.53 3.56 2.62 3.56 3.53 Mexico peak 5.49 4.85 3.49 4.85 4.85 4.85 4,85 4.85 5.49 4.85 4.85 Mexico off-peak 3.66 3.23 2.62 3.23 3.23 3.23 3,23 3.23 3.66 ’ 3.23 3.23 Netherlands peak 0,56 0,31 0.31 0.76 0.61 0.34 0.27 0.87 1.03 0,66 0.44 Netherlands off-peak 0.56 0.31 0.31 0.76 0.61 0.34 0,27 0.87 1.03 0.66 0.44 Poland peak 2.55 1.14 2.55 1.02 1.25 1.25 1,14 1.25 4.61 1.39 1.25 Poland off-peak 2.55 1.14 2.55 1.02 1.25 1.25 1.14 1.25 4.61 1.39 1.25 Portugal 2.15 0.88 0.88 1.62 0.88 0.85 0.85 0.88 2.84 0.88 0.88 Singapore 1.04 1.74 0.68 3.30 1.74 1.74 1.74 2.43 1.22 2.43 1.74 Spain 2.90 0.92 1.65 1.31 0.92 0.92 0.92 0.92 2.90 0.92 0.92 Sweden 0.94 0.35 0.29 0.94 0.28 0.35 0.29 0.54 1.96 0,54 0.54 Switzerland 0.43 0.43 0.22 1.08 0.43 0.22 0.22 0.43 1.08 0.43 0.22 Turkey 3.57 1.39 2.08 1.39 1.39 1.39 1,39 1.39 3.57 1.39 1.39 U.K. peak 1,88 1.09 0.91 1.48 1.48 1.09 1.09 1.36 1,88 0.88 1.36 U.K. off-peak 1.61 1.02 0.86 1.33 1.33 1.02 1.02 1.14 1.61 0.76 1.14 U.S. (WorldCom One) 0.51 0.51 0.21 1.89 0.51 0.51 0.51 0.51 1.02 0.51 0.51 U.S. (WorldCom Basic) 6.21 6.27 2.19 7.92 5.91 5,4 5.07 8.55 7.47 5.49 6.21 U.S. (AT&T One Rate) 0.51 0.87 0.21 1.89 0.87 0.51 0.51 0.51 0.45 0.51 0.51 U.S. {AT&’I Basic) 6.66 6.81 2.37 8.25 6.48 5,88 5.46 8.55 8.07 5.94 6.63 Notes: 1. 2. All rates are =n US$ and exclusive of taxes and were current on August 31, 2000. Peak hours are be’e~veen 9:00-19:30, Monday-Friday. Rates have been calculated ~n real t~me usrng meter step (rounded up to next meter step for a 3 minute call). 3. Fees are $2 with domestic long d=stance per month for AT&T One Rate and $3 per month for WorldCom One 4. Rates for calls from the U.S. to Canada and Mexico are from Washington, D.C to Montreal and Mexico City, © TeleGeography, Inc. 2000 TeleGeography 2001 Japan Korea, Rep. Mexico Neth’lands Poland Singapore Spain Sweden Turkey U.K. U.S. To]From 1.93 1.76 285 1.78 2.57 1.33 2.25 1.83 2.16 1.40 1.42 Austraha 0.86 0.88 1,28 0.81 0.92 0.86 0.81 0.81 0.92 0.81 0.86 Austria peak 0.77 0.77 1.14 0.68 0.92 0.77 0.68 0.68 0.92 0.68 0.77 Austria off-peak 2.66 3 65 4,65 1.00 1.99 2 68 1.33 1.33 1.99 1 00 1.33 Belgzum peak 2.66 3.85 4,85 1.00 1.99 2.66 1.33 1.33 1.99 1.00 1.33 Belgium off-peak 1.77 4.29 3,53 1.14 1.07 3.53 1.20 1.14 1.84 1.14 1.70 Czech Rep. peak 1.14 3,28 2,65 0.69 0.63 2.65 0.76 0.69 1.26 0.69 1.07 Czech Rep off-peak 2.55 4.21 4,21 0 91 0.85 2.91 1.32 0.48 1.60 0.71 1.03 Denmark peak 2.04 3.34 3.34 0.71 0.74 2.33 0.95 0.37 1 11 0.53 0.78 Denmark off-peak 3.13 4 14 4 14 1.27 1.27 3.39 1.27 0 41 1.27 1.27 1.27 Finland 1.25 1.25 1.66 0.43 0.90 1.66 0.43 0.43 0.90 0.43 0.43 France peak 1.12 1.12 1,52 0 42 0.86 1.52 0.42 0.42 0,86 0.42 0.42 France off-peak 1.85 1.85 2 42 0.29 0.57 2.57 0.29 0.29 0.57 0.29 0.29 Germany 1.22 1.22 2.01 0 86 1.22 1.22 0.86 0.86 1.05 0.86 0.86 Greece 1.91 2.74 1.71 0.85 1.08 1.91 1.08 1.08 2.06 0.34 0.65 Ireland peak 1.53 2.74 1.54 0.75 0.94 1.53 0.94 0.94 1,78 0.32 0,57 Ireland off-peak 2.38 2.98 3.86 0.92 1.29 2.96 0.92 0.92 1.88 0.92 0.92 Italy peak 2.38 2.96 3.86 0.92 1.29 2.96 0.92 0 92 1.88 0.92 0.92 Italy off-peak n.a. 3.50 8.72 8.14 8.14 5.20 8.14 8.14 8.14 4.07 1.70 Japan peak n.a. 2.84 5.58 6.81 6.81 4.54 6.81 6.81 6.81 3.97 1.42 Japan off-peak 2.22 n.a. 4.07 3.53 3.56 2.62 3.53 3.53 3.56 2.98 1.90 Korea, Rep. 5.49 5.49 n.a. 4.85 4.85 5.49 4.85 4.85 4.85 4.85 3.09 Mex=co peak 3.66 3.66 n.a. 3.23 3.23 3.66 3.23 3.23 3 23 3.23 2.32 Mexico off-peak 1.03 1 86 2 48 n.a. 0.87 1.28 0.44 0.36 0.92 0.25 0.21 Netherlands peak 1.03 1.86 2.48 n.a. 0.87 1.28 0.44 0.36 0.92 0.25 0.21 Netherlands off-peak 4.61 4 61 4.61 1.14 n.a. 4.61 1.38 1.14 1.38 1.25 2.55 Poland peak 4.61 4.81 4.61 1.14 n.a. 4.61 1.38 1.14 1.38 1.25 2.55 Poland off-peak 2.84 2.84 2.98 0.88 1.82 3.80 0.74 0.88 1.62 0.85 0.85 Portugal 1.56 2 17 3.47 1.74 3.30 n.a. 2.43 1.74 3.30 1.02 0.68 Singapore 2.90 2.90 2.20 0.92 1.31 2.90 n.a. 0.92 1.54 1.00 1.07 Spa~n 0.94 2.60 1.96 0.35 0.54 1.31 0.54 n.a. 0.94 0.29 0.29 Sweden 1.08 2.05 2 59 0.43 1.08 1.08 0.43 0.43 1.08 0.43 0 22 Sw=tzerland 3.57 3.57 3.57 1.39 1.39 3.57 1.39 1.39 n a, 1.39 2.08 Turkey 2.59 4.15 4.15 1.09 1.48 2.24 1.36 1 09 2.59 n.a. 0.91 U.K. peak 2.20 3.94 3,94 1.02 1.33 2.13 1.14 1.02 2.20 n.a. 0.86 U.K. off-peak 0.78 0.81 1.17 0.51 1.02 1.05 0.51 0.51 1.53 0.30 n.a. U.S. (WoddCom One) 6.00 7.47 4.98 5.34 6.63 6.72 6.63 5.34 7.56 4.50 n.a. U.S. (WorldCom Basic) 0.48 0.45 1.05 0.75 0.84 0.84 0.51 0.60 1,35 0.30 n.a, U,S, (AT&T One Rate) 6.39 7.98 8.24 5.79 6.93 7.26 7.14 5.79 8.07 4.77 n.a. U.S (AT&T Baszc) Source: Tar~fic a a urnt of The Phdhps Group, 3rd Floor, 19 Thomas More St., London E1 9YW, U K Tel +44 20 7423 4500 Fax +44 20 7423 4501 Ernml: consult@tanflca.com ¯ www.tarlfica corn Eve~ effort has been made to ensure that thzs przcing data zs up-to-date and accurate However, Tarff~ca can not be held responsible for any losses arzszng from use of the data Source for U S. rates TeleGeography research © The Phzlhps Group and TeleGeography, Inc 2000 © TeleGeographyo Inc. 2000 TeleGeography 2001 Retail Pricing Trends, 1997-2000 Figure 1. Retail Tariffs for Selected Countries $6 00 $600 $4OO $200 na $600 Aust~aha Germany Japan Mexzco UK. $6 00 $600 $400 $200 $6OO Aust~aha Germany Japan Mexzco ~om Ausb’alia from Belgium U S U.S from France from Germany U.K. U S Aust~’aha Germany Japan Mexico U K. $6O0 $6 00 $4OO $200 na $600 -- Ausb’aha Germany Japan Mexico Ul~ US $6O0 $600 $400 $2.OO $000 $6,00 $600 $4OO SZ00 $000 from Czech Rap. Australia Germany Japan Mexico UK. US from Italy Australia Germany Japan Mexico U K. US $6oo $6oo $4.00 $2oo $6oo Aust~aha Germany Japan Mexico fromJapen U I~ U.S from U.K. Aust~aha Germany Japan Mexzco U K. US Key ¯ [] [] [] 1997 1998 1999 2000 Ratefora 3 mmutecal[ to desbnabon $6 00 from U.S. ~00 $2.O0 $600 ~. ’~ ~’~ AustTaha Germany Japan Mexico UK. U S Note: All rates are for peak calling and are expressed in US dollars. Source: Philips Tadfica and TeleGeography research © TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2OO1 Retail and Wholesale Rates: PSTN versus VolP PSTN Rates VolP Rates Country Retail Wholesale Retail Wholesale Argentina 0,52 0.23 0.25 0.14 Australia 0.17 0.09 0.08 0.03 Austria 0.17 0.09 0.08 0.04 Bahamas 0.37 0.27 0.21 0.09 Belarus 0.63 0.32 0.30 0.20 Belgium 0.17 0.08 0.08 0.03 Brazil 0.38 0.19 0.21 0.14 Canada 0.07 n.a. 0.04 0.03 Chile 0.45 0.13 0.11 0.05 China 0.59 0.22 0.21 0.07 Colombia 0.52 0.19 0.22 0.13 Cyprus 0.62 0.24 0.26 0.13 Czech Republic 0.63 0.19 n.a. 0.18 Dominican Republic 0.44 0.15 0.23 0.09 Finland 0.17 0.11 0.08 0.05 France 0.17 0.08 0.08 0.03 French Polynesia 1.54 0.38 0.66 0.19 Germany 0.17 0.08 0.08 0.03 Ghana 0.68 0.29 0.49 0.21 Guyana 0.17 n.a. 0.20 0.11 Hong Kong 0.34 0,08 0.08 0.03 India 0.66 0.53 0.55 0.49 Indonesia 0.53 0.28 0.36 0.19 Iran 0.85 0.58 0.95 0.48 Israel 0.28 0.12 0.08 0.06 Japan 0.26 0.09 0.08 0.04 Malaysia 0.44 0.15 0.19 ,0.06 Mexico 0.39 0.19 0.19 0.12 Netherlands 0.17 0.08 0.08 0.03 New Zealand 0.17 0.09 0.08 0.03 Pakistan 0.75 0.58 0.78 0.46 Peru 0.52 0.33 0.31 0.17 Poland 0.34 0.21 0.29 0.12 Russia 0.47 0.28 0.21 0.12 Saudi Arabia 0.70 0.52 0.82 0.39 Singapore 0.35 0.13 0.15 0.04 South Africa 0.53 0.31 0.30 0.20 Spain 0,17 0.12 0.08 0.03 Sri Lanka 1.13 0.63 0.81 0.26 Switzerland 0.17 0.09 0.08 0.03 Taiwan 0.33 0.12 0.11 0,06 Turkey 0.51 0.24 0.17 0.17 Ukraine 0.48 0.31 0,20 0.13 United Arab Emirates 0.55 0.30 0.40 0.18 United Kingdom 0.10 0.07 0,08 0.02 Venezuela 0.35 0.26 0.26 0.18 Vietnam 0.89 0.78 0,97 0.65 Yugoslavia 0.84 0.28 0.42 0.20 Note: Rates are for calls from the United States Source’ PSTN retad rates: WorldCom One; VolP retail rates: deltathree com PC-to-Phone. Wholesale rates reflect wvo rna]or ~nternat~onal carriers’ rates as of September 2000. © TeleGeography, Inc. 2000 TeleGeography 2001 © TeleGeogrophg, Inc. 2000 Follow the Money Network-to-Network Payments for Internet Telephony and Other IP Traffic Streams As the Internet diverts more international telephone traf- fic from the existing carrier-to-carrier sgstem for settling accounts, there is a growing interest in new wags to share the network costs of global Internet communica- tions. This article profiles compensation arrange- ments~learinghouses and paid transit--used todag bg networks providing cross-border Internet telephong. It also looks at several factors (e.g., Internet billing soft- ware) that are likelg to affect the adoption of new inter- network compensation schemes. The article suggests that the current finandal pressure on major Internet ser- vice providers and long distance car~ers alike may accelerate implementation of new compensation models. Introduction The bloom is off the rose. For decades, international phone calls provided a large source of high-margin revenue for tele- phone operators worldwide. No more. In a few short years, open markets, abundant new transmission capacity, and the rise of the Internet have driven long distance call prices closer to their cost, especially for wholesale providers. This has led to hard times for well-established companies; many new card- ers have fared much worse (see Figure 1. The Death of Distance). Even AT&T, the largest U.S. telco, has not been immune from the stock market’s revaluation of the long distance business. By June 2000, The Financial Times reported that AT&T’s share price actually reflected a negative value for its long distance unit after taking account of the company’s cable TV, wireless, and other assets. Similarly, in August 2000, CS First Boston concluded that if WorldCom were broken up, its consumer long distance business--still the bulk of WorldCom’s revenue~ would only be worth $5 billion out of a $60 billion total valu- ation. In the 1990s, the pressure on the margins for international calls came chiefly from innovative least-cost routing arrange- ments that worked within the existing settlement system (call- back and traffic refile). [1] These practices sharply eroded the wholesale termination fees or settlements paid by one telephone operator to another for completing international calls. Today, one of the greatest pressures on international settlements comes from Internet telephone and fax services; they threaten to bypass the settlement syster~ altogether. International telephone companies appear to be caught between two worlds, neither of which is viable. On the one hand, despite the internet’s allure, the bulk of most operators’ phone traffic (at least outside Western Europe and North America) is still subject to the accounting rate system, and the payments the operators receive from foreign correspondents remain a significant revenue source. For example, according to the ITU’s study, Challenges to the Network, in 1998, net settlement revenue from international telephone services totaled over $125 million for the Dominican Republic, Poland, Egypt, Lebanon, Vietnam, and Pakistan; net revenues were approximately $500 million or more for China, the Philippines, India, and Mexico. For carriers serving these countries, Internet telephony is a clear economic threat. On the other hand, even in developing countries, most opera- tors acknowledge the rough consensus that has been forged among the world’s telecom engineers: over the next decade, the lnternet’s packet switched protocols will provide a com- mon global platform for almost all communication services. The migration of voice traffic from dedicated, long-haul trans- mission facilities for switched telephony to multi-purpose packet-based data networks is no longer a question of "if" but "when." [2] Hence, while the Internet may be a near-term revenue threat to long distance services, in theory, it holds out the promise of new revenues from other products. For a start, Internet telephony may offer telcos an entr6e to various unified-messaging services by offering a common means for sending real-time and stored messages (e.g., voice- mail, email, and fax). And that is just the beginning. Over the next few years, the Net also may provide operators an oppor- tunity to move up the value chain by sharing the revenue from multi-media and e-commerce services (e.g., music, films, games, and shopping). At least that is the theory. The closer technology gets to bringing this vision to the mar- ket, the greater the interest in new network-to-network pay- ment schemes. As with telephone service, the scale necessary to provision Internet services is expensive, and it requires the cooperation of numerous other networks. In the telephone © TeleGeography, Inc. 2000 TeleGeography 2001 Figure 1. The Death of Distance: International Carrier Bankruptcies and Distress Sales Who: Cherry Communications What: Bankruptcy When: 199? Why: In 1996, Cherry was a top U,$, international reseller with I0 percent of the market (over 650 million minutes). But poor cost con- trois, "shady" selling j~ractices, and a large number of uncollectible accounts put it into bankruptcy, it wasthen bought by World Access, Inc., a company partly owned by WorldCom, which has since bought several other distressed carriers. Who: Telegroup What: Bankruptcy When: February 1999 Why: As a pioneering call-back operator, Telegroup had few peers. Revenues rose from $270,000 in 1990 to S337 million in 1997, allowing a $40 million IPO. But Asia’s economic cdsis dampened the appetite for calls billed in U.S. dollars, just as the company was spending heavily to build out its global network. In June 1999, Pdmus Telecommunications bought Telegroup’s U.S. business out of bankruptcy for $72 million, Who: STAR Telecommunications What: Distress Sate When: February 2000 Why: Once a stock market favorite (in the spring of 1998, STAR’s shares were up over 400 percent from the I997 tPO), by mid-1999 STAR was long on capacity and short on cash. And, despite a boom- ing traffic base (over two billion minutes in 1999}, half from its PT-I pre-paid card affiliate, the pressure on call margins was unrelenting. STAR sold receivables to raise cash and then later agreed to be bought by WorJd Access, with PT-1 sold off to Canadian-owned Counsel Communications. Who: GSTTetecommunications What: Bankruptcy when: May 2000 Why: GST focused on the western United States and called itself an "Integrated Communications Provider," offedng lnternet and local access services. But most of its revenues (and investments) were linked to its long distance and international calling business. And pricing pressure never allowed the company to earn enough to fund its ambitious network construction plans. Who: Pacific Gateway Exchange What: Asset Sales when: Spdng 2000 why: With backing from Japan’s KDD, Pacific Gateway Exchange (PGE) was often touted as a sure bet to capitalize on "emerging glob- al telecom opportunities." It specialized in trans-Pacific routes and, buoyed by its rapid growth, went "long" on numerous undersea cable systems. Yet as per-minute margins fell and growth slowed, PGE found its cable portfolio more of a liability than an asset, especially as cheaper capacity came online. The market began to lose faith and, in March 2000, the company restated its 1999 earnings, A flur- ry of stockholder actions followed, ddving PGE’s share price below $5 a share (it had reached $62 in 1998). To keep afloat, PGE sold much of its cable capacity to Metromedia Fiber Network and is currently refocusing on international data and VolP services. One-Year Index of International Carrier Stock Prices 160% 140% 120% 100% 80% 60% 40% 20% 0% AT&T RSLCOM Source: Company reports and U.S. Securities and Exchange Commission (SEC) © TeleGeography, Inc. 2000 TeleGeography 2001 © TeleGeogrophy, Inc. 2000 world, however, most operators receive significant payments from other networks in exchange for providing end-to-end ser- vice. Not so for the Internet. As explained more fully in Part II, neither of the prevailing inter-network compensation mod- els for the lnternet--peering (where no money changes hands) and transit (where payments are one-way)--provides an obvi- ous replacement for the bilateral revenue sharing arrange- ments which now prevail in the telephone industry. [3] In these circumstances, the greater an operator’s reliance on network-to-network compensation payments, the greater the concern about the transition to Internet-based services. This concern has grown during the last year as the stock markets have begun to write down the capital value of almost every company which provides long-haul telecommunication services (again, see Figure 1). Put bluntly: How will the costs of handling two-way traffic on IP networks be recouped where the great majority of service providers are downstream ISPs which make net outpayments to upstream networks? From end-user charges alone? Or advertising? Or non-network services? Similarly, in tomorrow’s lnternet-based world, who will bear the cost of implementing much needed cross-network Quality of Service (QoS) standards for telephony and other applica- tions? And, where network traffic flows are asymmetrical--the sine qua non of the public Internet--what is the incentive for smaller networks to make new investments when most of the money flows to their larger correspondents? The remainder of this article is organized into three main sec- tions. Part I1 briefly reviews the existing settlement systems for international telephone operators and contrasts that with the prevailing inter-network compensation schemes among global ISPs. Part III looks in more detail at emerging inter-network payment arrangements for lnternet telephony and, to a lesser extent, for other IP-based services. The article concludes with Figure 2. How the Accounting Rate System Works for International Telephony Country A Half Circuzt H-~lf Circuit Local National Backhaul International Exchange Transport Transport Network Elements Covered by Settlement Rate Cot~ntry B ’, ! Nmwork Elements Covered by Settlement Rate 1. What is an accounting rate? The accounting rate is a bilaterally agreed per minute rate for carrying public switched telephone network traffic between two carriers (A and B] in different coun- tries. This is the wholesale rate of the call from beginning to end. 2. Carders use the accounting rate to settle accounts with one anoth- er for handling traffic in both directions. The carrier originating the greater number of minutes in an agreed period (typically, one month) pays out funds to the other carder in an amount equal to the number of surplus minutes multiplied by one-half the accounting rate (known as the settlement rate). Source: Adapted from Direction of Traffic t994. TeleGeographyllTU. For example, if in a Oven period carrier A sends 100 minutes to carder B, and carrier B sends 150 minutes to carrier A, then car- dora has a surplus of 50 minutes. 4. Assume now that the accounting rate is $0.30 per minute. Multiplying this 50-minute surplus by one-half the accounting rate, carrier B pays to carrier A an amount equal to 50 x $0.15 or $7.50. © TeleGeography, Inc. 2000 © TeleGeography, /nc. 2000 TeleGeography 2001 a look at some of the factors which are likely to advance or retard new network compensation schemes for IP services. II. From Settlements to Transit A. Settlements The classic paradigm for sharing the cost of international telecommunications is based on joint ventures for switched telephony. Two Public Telephone Operators (~PTOs], one from country A and one from country B, agree to exchange traffic and each PTO covers the cost of provisioning and maintaining a circuit to a half-way point that links their two networks. To cover the provisioning costs, the two operators pay each other one-half of an agreed wholesale charge, known as an account- ing rate, for each minute of traffic the other terminates. At the same time, each operator retains the freedom to set the end- to-end retail charge for traffic originating on its own network. Inter-operator settlements are based on net traffic balances. Thus, on any given route, the operator originating more traffic than it terminates pays out funds to its correspondent (see Figure 2. How the Accounting Rate System Works for International Telephony). Despite the fact that long-haul transmission costs have fallen rapidly since the mid-1990s, a PTO that receives more traffic than it sends has an obvious incentive to keep its accounting rate high so as to maintain incoming settlement payments. High, yes, but not so high that it encourages too much traffic bypass; that would be self-defeating. Thus, in the face of alternative call routing techniques (third country refile, ISR, "leaky" corporate branch exchanges, and mobile networks), most operators have been gradually forced to reduce their set- tlement rates or face a growing loss of traffic and revenues. In richer countries, settlement rates have come under even greater pressure because the majority of international traffic is now data, not voice, and a large portion of that data stream is Internet traffic. Consequently, even though most of the rev- enues that carriers receive for use of their international net- works still come from retail charges and settlements related to voice, more and more of the bandwidth and switching costs are attributable to data. And, unless these data/Internet costs are recovered from the retail charges assessed on Internet cus- tomers or from network connection charges, voice subscribers will be subsidizing them. So let us now take an initial look at how international opera- tors are typically compensated for hauling Internet traffic. We will concentrate here only on the network-to-network pay- ments involved, leaving end-user charges aside. At the net- work or wholesale level, two types of payment models cur- rently predominate: peering and transit. To put these models in context, a short digression is useful on the Internet’s global structure. From a service provider’s perspective, the Internet looks something like a river system. On the Internet, the headwa- ters are formed by thousands of small, often local, networks. They route their customer’s traffic to the rest of the Net via hundreds of tributary networks which, in turn, connect to the rest of the world via a top tier of Amazon-sized backbone net- works (see Figure 3. Market Shares of Internet Backbone Providers). In June 2000, the U.S. Department of Justice (DO J) provided a useful overview of this architecture when it sought to block two of the top tier ,backbone networks-- WoridCom and Sprint--from merging and thus gaining undue market power vis-a-vis other backbone networks. Figure 3. Market Shares of Internet Backbone Providers Rank 1 2 3 4 5 Total Boardwatch Provider WorldCom Sprint Cable & Wireless AT&T Verio Market Share 21% 13% 8% 6% 5% 53% DataCommunications Provider WorldCom Cable & Wireless AT&T Genuity Market Share 37% 13% 12% 12% 74% U.S. Department of Justice Study Provider Market WorldCom Sprint B.a. n.a. Share 37% 16% B.a. 53% Notes: Boardwatch rankings are based on number of t~ansit relationships with other ISPs as of December 1,999, Data Communications rankings are baaed on the routing tables of the 500 mosttrafficked web sites as of June 1999. U.S. Department of Justice study is based on March 2000 data. Source: Boardwatch Magazine, Data Communications Magazine, and the U.S. Department of Justice © TeleGeography, Inc, 2000 TeleGeegraphy 2001 © TeleGeography, Inc. 2000 Figure 4. lime to Pay Up? ITU-T Recommendation’D.(iii) International lnternet Connections ITU-T Recommendation D.50--perhaps better known as draft D.(iii}--was adopted in Apdt 2000 by representatives of the world’s telecom ministers in Geneva. The text appeared to be a model of economic diplomacy: "It is recommended that administrations negotiate and agree to bilateral commercial arrangements applging to direct inter- national lnternet connection wherebg each administra~on wilt be compensated for the costs that it incurs in carrying traffic that is generated by the other administration." l But this rather anodyne text--that the costs of Intemet transmission capacity used by two parties for bilateral service be shared among them--soon triggered a fierce debate. On the one side was the U.S. government and major U.S, international carders, most of which are also U.S. Intemet Backbone Providers (IBPs}. They saw the draft International Telecommunication Union (ITU) Recommendation D.(iii) as an attempt to impose a traffic-based settlement system on the Intemet akin to the "discredited" accounting rate regime used for international telephony. That is the worst kind of regulatory med- dling, said the Americans, and it would stifle the ability of tnternet Service Providers {ISPs) to negotiate their own interconnect terms~ terms that have thus far led to a rapid build-out of global bandwidth, fostered bread lntemet connectivity, and stimulated the necessary investment to scale the core backbone networks in the U.S. to accommodate lnternet transit traffic from around the world. Not so, said D.(iii}’s proponents, led by Australia and other Asia- Pacific nations. [2] We are not "asking that the telephony model be adopted" for the tntemet but are only "proposing that principles such as ’non-discrimination’, ’transparency’ and ’cost-based’ be applied to all services...." First, its proponents argued, D.(iii) eschews government intervention; it seeks to rely on commercial negotiations. Second, it is pro-com- petitive and would encourage new ISPs worldwide because non-U.S. ISPs must now subsidize the top tier of largely U.S.-based IBPs. How is that? Off-shore tSPs typically pay the full cost of the international transmission capacity to the U.S. even though, once provisioned, the capacity is made available--without charge--to U.S.-based tSPs to send lntemet traffic in the opposite direction to off-shore lSPs and their customers. One result is that non-U.S. ISPs face disproportionately high interna- tional "transit" or interconnect costs which must be recouped from their customers. That raises the costs of local tnternet access, said D,(iii)’s advocates, and also deters new entrants in poorer and mid- dle income countries. At best this is half true, replies the U.S. Yes, U.S. IBPs require off- shore [SPs to acquire their own transmission facilities if they wish to connect in the U.S., but many IBPs have off-shore points of presence (PoPs),. And non-U.S. ]SPs are free to build or buy their own U.S. backbones if they wish; the market for lnternet backbone services is wide open. Witness, for example, NTT’s acquisition of Vedo, a major U.S. ISP, though this deal apparently was subject to unprecedented terms by U.S. law enforcement officials. In addition, said the U.S., draft Recommendation D.[iii) is insuffi- ciently inclusive. It singles out international leased line costs and fails to consider cost-sharing for Internet facilities generally (e.g., for domestic links, hubs, web sites, maintenance). The fight over cost-sharing for global lnternet access continued in May 2000 at the CancSn, Mexico, meeting of telecom ministers from the Asia-Pacific Economic Cooperation (APEC) group. After extensive debate, the following terms were appended to a proposedaction pro- gram for review by APEC’s economic ministers at their November 2000 meeting in Brunei: Governments need not intervene in private business agreements on International Charging Agreements for lntemet Services achieved in a competitive environment, but where there are dom- inant players or de facto monopolies, governments must play a role in promoting fair competition. ¯ Internet Charging Agreements between providers of network services should be commercially negotiated and, among other issues, reflect: The contribution of each network to the communication; The use by each party of the interconnected network resources; and, The end-to-end costs of international transport link capacity. In Cancan, as ]n Geneva, U.S. telecom officials dedded the need for government oversight of the Internet’s privately owned global infra- structure, absent the existence of any dominant players. However, the very next month, Amedcan antitrust officials contended that the private sector could not be trusted to safeguard the Internet’s glob- al architecture. The U.S. Department of Justice (DO J) went to court to block the merger of America’s two largest IBPs, WorldCom and Sprint, because a combined company would control 53 percent of U.S. Intemet traffic--five times that of the next largest tBP. According to the DO J, that would have given the merged entity the incentive and ability to impair rival ISPs by raising their costs and/or degrad- ing the quality of ]nterconnections. [3] Back to the 1TU. In October 2000, after a 10-day meeting in Montreal, the ITU’s World Telecommunication Standardization Assembly (v~q’SA) adopted a compromise version of Recommendation D.(]ii} based on text brokered by Canada and the Netherlands. [4] Adopted by consensus, with the U,S. and Greece taking reserva.tions, the 1TU now recommends that: continued on next page © TeleGeographg, Inc. 2000 TeleGeography 200~ Figure 4. (continued) "administrations involved in the provision of internaiional Internet connections nego~ate and agree to bilateral commer- cial arrangements enabling direct international Internet con- nections that take into account the possible need for compen- sa~on between them for the value of elements such as traffic flow, number of mutes, geographical coverage and cost of international transmission amo.ng others," Notes: I]] In ITU parlance, administration means a telecommunications operator or R~o~ni~ed Opera~in~ A~ency (ROA). Accordin~ to Australia, a pdm~ advocate of Recommendation D.(iii), the term ROA "includes any operator of facilities that provides public access and connection to the communication infrastructure and address space of the global Internet." See ’Australia Recommendation D.(iii)," Document 84-E, September 2000, WTSA-2000, available at www.itu.intlitudocr/itu-tlwtsa/docs/index.html (resistration required). [2] For backsround, see ’~Asia-Pacific Telecommunity Study Group 3 Proposed Recommendation Dill", Document 81-E, September 2000, WTSA-2000. For the U.S.. position, see Documents 49~E [Substantive Issues] and 52-E [Procedural Issues]. All documents available at the ITU URL referenced in Note I above. [3] See www.usdoj.gov/atr/cases/fS000/5051 .htm. [4] For details, see "Draft Report of Committee 6," Document 149- E, WtSA-2000, and "Report of Committee 6," Document 160-E, WTSA-2000. © TeleGeography, Inc. 2000 As the DOJ explained, "Because the Internet comprises thou- sands of separate networks, direct interconnections [among these networks] would be impractical. Instead, the Internet has developed a... structure in which smaller networks are interconnected with one of the few large Internet ’backbone’ networks, which operate high-capacity long-haul transmission facilities and are interconnected with each other." Thus, in a typical Intemet session, "an ISP sends data from one of its customers to the large network that the ISP uses for backbone services, which in turn sends the data to another backbone network, which then delivers it to the ISP serving the end user to whom the data is addressed." The small number of lnternet backbone providers (IBPs) at the top of this food chain all sell transit service to substantial num- bers of ISPs and sell dedicated Intemet access directly to cor- porate customers or other enterprises. Tier One IBPs have large, high capacity national and international networks, and typically maintain pdvate peering relationships with all other Tier One IBPs on a settlement-free basis. Smaller IBPs, often referred to as Tier Two or Tier Three IBPs, also may sell transit to ISPs (or IBPs) and sell dedicated Intemet access to end-users. However, as the DOJ explains, these Tier Two or Tier Three IBPs typically purchase transit from, rather than peer with, one or more Tier One IBPs and/or rely substantially upon exchanging traffic at "inferior" (and those are the DOJ’s words) public interconnection facilities. Any lower tier IBP that must purchase significant connectivity thus operates "at substantial cost disadvantages compared to Tier One IBPs" which rely upon peering. Let us now return to the economics of peering and transit, fol- lowing which we shall consider how these schemes may affect the flow of funds for VolP and other specific IP services. Peering As noted above, peering exists mostly between ISPs with like- sized networks (e.g., between the handful of global IBPs and among smaller Tier Three networks). Among peers, no money is exchanged. Payment comes in the form of reciprocal net- work use: VII terminate your traffic if you terminate mine. These reciprocal obligations are not service-specific. That is, peers terminate VolP and other traffic indifferently. There are some caveats. Peers only agree to accept traffic from each other if that traffic is destined for their network cus- tomers (i.e., end-users) or for a network buying transit ser- vices from them. Peers will not hand off or "transit" traffic for each other. And each peer typically pays for one-half of the cost of the leased lines necessary to connect its network to the other. In the mid-1990s, as differences among’ISPs grew--for exam- ple, in terms of customer base and network infrastructure~ universal peering fell out of favor. ISPs with more extensive transmission networks began charging smaller ISPs and some hosting facilities (e.g., web server farms) for the right to mute traffic over their networks to reach end-users. This arrange- ment is generally known as transit. [5] Transit Transit agreements typically involve a standard monthly inter- connection fee based upon the line speed a customer requires plus a traffic-sensitive fee for use of additional capacity. For these transit fees, the user gets full Internet connectivity (see Figure 7. Band-X Routed IP Pricing). F, ew question the economic rationale for the current system of transit charges--larger networks provide more value. But like the accounting rate system for switched telephone traffic, it tends to be self-reinforcing. The larger an IBP’s network, the more it may be able to dictate access terms for downstream networks. And the greater a network’s ability to acquire TeleGeography 2001 © TeleGeographg, Inc. 2000 potential competitors, the greater the barrier to entry for new- comers. In these circumstances, it may not be surprising that U.S. and European Union regulators have had to step in repeatedly in the 1990’s to keep the top tier of Internet service providers competitive: network effects make larger horizontal combina- tions economically compelling. In addition, the market power of the largest IBPs may allow them to shift significant facility costs for cross-border access onto downstream networks--a matter which is of special concern to non-U.S. ISPs (see Figure 4. Time to Pay Up? ITU-T Recommendation D.(iii) International Intemet Connections). There is no mistaking the privileged position Tier One IBPs have within the sector. In mid-2000, for example, when GTE spun off its Tier One IBP, Genuity, the company explained the importance of maintaining Genuity’s position this way in its stock prospectus: ’7~ny significant loss to market share.., could cause the loss of our status as a Tier One Internet backbone provider, which would make our services significantly less attractive to existing and potential customers and would likely result in a significant loss of revenues. In addition, the loss of... Tier One [sta- tus].., would adversely affect our ability to maintain our free private peering relationships with other Tier One [IBPs]. Currently these relationships allow us to have direct, cost-free exchange of traffic with other Tier One Internet backbone Figure 5. VolP Clearinghouse Payment Flows Originating ITSP PSTN :Clearinghouse’ IP Networkjlnternet Key ......... Signaling Transport sssssssssss Billing Terminating ITSP PSTN 1. Both originators and terminators of traffic pay an initiation fee to the clearinghouse that usually includes a metering software-licensing fee and a charge for facilitating the initial connection to the clearinghouse, which includes training of personnel. 2. Both originators and terminators provision their own connections to e=ther an IP network designated bythe clearinghouse orto the public Internet. The choice =s based on quality and cost. 3. Originators pay the advertised rate for termination to the location of their choice. See Figure 6. Clearinghouse Terms for VolP Traffic Termination for sample rates. 4. The clearinghouse takes a portion of the rate from an originator. Industry clearing rates differ, and rates between certain customers of the same clearinghouse differ as well. The range is anywhere from five to 17.5 percent. The primary element affecting the clearing rate is volume; the more traffic an originator sends to a clearinghouse, the lower the clearing rate. 5. The clearinghouse then arranges for payment to the terminator for its services. Using an average clearing rate of 11 percent for a three minute call to Russia, here is how the money flows: a) The originator pays $0.36 for a three minute call to Russia placed through the clearinghouse. b) The clearinghouse gets the clearing fee of 11 percent or approximately $0.04. c) The terminator receives $0.32. Source: TeteGeograph¥ research © TeteGeography, inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 providers and allow us to avoid the congestion of public peer- ing points when directing traffic to users connected to those other Internet backbones. If we are unable to maintain these free peering relationships our operating costs will increase and our results of operations will suffer." B. Settlements vs. Transit There is actually less difference than one might think between the settlement systems now used for compensating interna- tional telephone carriers and the compensation systems that characterize the lnternet. They are both generally based on the amount of the traffic handed off from one network to another--measured for the telephone network in minutes; for the Internet in megabytes. And, on both systems, where the volume of traffic flowing in each direction is more or less in hal- ance, the effective rate of exchange--whether from peering or netting settlement payments--is zero. The difficulties begin when traffic flows are not balanced or where autonomous systems are unequal in size; consequently, a clamor for compensation results. Here, the two regimes diverge significantly: ¯ On the telephone network where traffic is unbalanced, the net flow of payments tends to be in the same direction as the net traffic flow. In other words, on the internation- al telephone network, the carriers of most calls pay a land- ing fee to the terminating service provider to deliver their messages. ¯ On the lnternet, however, especially the World Wide Web, the net flow of payments, notably for transit, is in Figure 6. Clearinghouse Terms for VolP Traffic Termination Sample Terms and Conditions Quality of Service (OoS) All members connected to clearing- house are ranked by quality for each termination mute. Ranldngs are based on such cdteria as packet loss, network availability, 24x7 network operations center support, availability of backup muting and hardware (e,g., second gateway up and running). Seller’s Obligations All members must license clearinghouse metering software and pay one-time joining fee. Terminating partners to a clearinghouse must submit to PoP-by-PoP network audits to certify and regularly authenticate quality rating. All rankin~s are published on clearinghouse member web site. In addition, total bandwidth capacity information will be published on member web site. Buyer’s Obligations Must license clearinghouse software and pay one-time joining fee, and submit to credit check (usually a presentation of audited financials and letter of credit or deposit]. A deposit oftwo or throe times highest-planned usage is usually required. Comply with monthly settlement cycle. Clearinghouse’s Obligations Handles all related settlements, billing and administration. Guarantees payment to terminating partners ’for traffic muted via clearinghouse. No guarantees either express or implied as to the quality of any given route. 5. lnterconnectibn Options Members should connect directly to the clearinghouse’s backbone for optimum quality ranking (although high-quality indirect connections may also merit top ranking]. 6. Facilities Needed All members must provide connection to dear- inghouse,network and provision their own IP ~ateway. Data col- lection and metering software also must be installed in members’ network. Liabilities/Damages Members are liable to any physical damages they inflict on clearinghouse members’ equipment. Some clear- inghouses also limit their liability by requiring members to state that they have legal authority to operate in their respective terri- tories. Additional warranties may apply based on specific con- tracts between members and the clearinghouse, Dispute Resolution With metering software deployed on the orig- inator’s network, the clearinghouse network and the terminator’s network, there are three points at which traffic is metered, allow- ing members to double check invoices. Any billing disputes are handled through the clearinghouse. Duration Negotiable. Also, since muting specifications are left in the control of member networks, originators can re-mute traf- fic flexibly and for any reason. Sample Termination Rates by Route (Sept. 2000) Cost of transit via clearinghouse includes a one-time joining fee and cost of connection to the clearinghouse backbone. Call ter- mination costs are based on 3-m~nute calls for different service grades. Prices expressed in US$ and includes a clearing fee: Australia $0.031 China $0.116 Czech Republic $0.125 India $0.501 Indonesia $0.183 Italy $0.039 Korea, Rep. $0.046 Philippines $0.136 Russia $0.118 Source: Adapted from Concert Global Clearmghouse © TeleGeography, Inc. 2000 TeleGeo~’aphy 2001 © TeleGeographg, Inc. 2000 Figure 7. Band-X Routed IP Pricing Seller A B C D E F G H A B C D E F G H 1 Mbps $1,105 $939 $289 $361 $361 $578 $723 $1,105 $939 rl.a. $289 $361 $361 $578 $723 2 Mbps $1,301 $1,879 $1,662 $578 $668 $723 $723 $1,301 $65O $939 $831 $289 $334 $361 $361 $65O 4 Mbps $3,396 $3,757 $3,400 $1,156 $1,734 $1,445 $3,179 $649 $939 $650 $289 $434 $361 $795 Price per Month 8 Mbps $6,792 $7,514 $6,800 $2,312 $3,468 $2,89o $6,214 Price per Megabit $849 $939 $85O $289 $434 $361 $777 n.a. 10 Mbps $8,316 $8,671 $8,3O9 $2,89O $4,335 $3,613 $7,514 $832 $867 $831 $289 $434 $361 $751 45 Mbps $33,945 $33,815 $31,160 $9,754 $15,022 $16,257 $33,960 n.a. $754 $751 $692 $217 $334 $361 $755 155 Mbps $71,676 $65,862 $33,598 n.a. $55,997 n.a. r].a. $554 $217 n.a. ~61 Note: Se~lers offering transit via Band-X post their rates anonymously and are identified only by letters te.g., A-H). Rates are fortransit originating at Band-X’s London PoP and have been converted from UK£to US$, Source: Band-X Ltd., September 200D © Tele6eography, Inc. 2000 the opposite direction to the net tr&ffic flow. On the lnternet, smaller ISPs with less traffic (customers) pay larger ISPs with more traffic (customers) for transit. And though they have a little upstream traffic, most users pay the originating ISP a fee to cover the cost of sending large volumes of web traffic downstream to them. The international telephone regime tends to promote the flow of resources from more developed (large traffic) to less devel- oped (small traffic) networks, in large part because richer countries make more calis to poorer ones than vice-versa. Yet, it also tends to subsidize inefficient national monopolies that send less traffic than they receive. The Internet tends to sup- port bottom-up, user-driven growth. But current Internet compensation schemes provide incentives for routing more and more traffic and resources to the largest and most efficient service providers, encouraging increasing horizontal concen- tration of control. ill. Compensation Schemes for NextGen Carriers With the foregoing review of peering and transit relations in mind, let us turn now to the specific network-to-network com- pensation schemes which apply when phone calls are routed via the lnternet. First, some definitions. Internet telephony commonly refers to a variety of services based on the termi- nals used (computer or telephone), the form of long-haul transmission (IP or other packet protocol), and the location of the interface or gateway between the ]nternet and the public switched telephone network (PSTN) (whether on the phone company’s premises or the ISP’s). Here, however, we will use Voice-over-lnternet Protocol (VolP)~ and Internet telephony to refer interchangeably to any voice communications routed over the Internet or an IP-compatible network for some por- tion of its transmission, without regard to the end-users’ ter- minals. In other words, as used here, VolP is the provision of voice telephony in packet form over an IP network. This broad definition seems appropriate because inter-network compensation and payment flows concern almost all types of VolP players--small ISPs and big ones, legacy telcos as well as private network operators--and it is with this issue that we are primarily concerned. A. Traffic Clearinghouses At this writing, VolP clearinghouses provide the only widely availab}e option for most ISPs that wish to recoup a portion of their network costs from other networks for delivering VolP traffic. [6] Though the details vary, all are based on a similar model, namely, the provision of a centralized billing mecha- © TeleGeography, Inc. 2000 TeleGeography 2001 Figure 8. WorldCom Costs for Transporting U.S, Long Distance Voice and Data Traffic Transmission Costs for Voice and Data Traffic 5O% 45% 40% 35% 30% 25% 20% 15% 10% l~i I Minute of Voice Traffi~ ¯ I Megabit of Data Traff_ic 10% 6% 4% Switching Transpor~ 35% 9% 5% 43% SG&A 0SS Access 4O% Source: WorldCom, Inc. © TeleGeography, Inc. 2000 nism for originating VolP providers and secured payment for terminating ISPs (see Figure 5. VolP Clearinghouse Payment Flows). In so doing, the clearinghouse provides the bridge-- often physical as well as financial--between one VolP provider and another. By working through a clearinghouse, an ISP can obtain compensation for terminating VolP calls to its cus- tomers without negotiating an interconnection agreement with the hundreds of widely dispersed ISPs in other countries whose customers may originate such calls. As a facilitator, clearing- houses generally are provider-independent. This can be par- ticularly important for those houses which provide multiple termination options by route. Independence also is desirable to ensure impartial monitoring of the quality of each partici- pating ISP; members must be afforded the ability to switch ter- minating pai-tners on a non-discriminatory basis if they are not satisfied with their current termination options. What about the finances? Clearinghouses typically charge a per-call "clearing fee" or commission of five to 15 percent per call, which is usually paid by the originating provider based on the termination price of the call. Clearinghouse participants set their own termination rates. These rates typically include the cost of a PSTN interconnection (i.e., the last-mile cost), but typically exclude an international PSTN settlement charge because the traffic is routed across borders via IP networks. To boost margins, however, rates posted at a clearinghouse may only be 20 to 30 percent below the prevailing international settlement rate. A VolP provider’s net revenue will depend on the total number of minutes it handles by route multiplied by the termination rate, minus its costs and minus the termina- tion payments it makes to other clearinghouse members. B. Transit Services In contrast to clearinghouses where network compensation typically flows in both directions, that is, to the sending and receiving ISPs and carriers, transit paym.ents are by definition one-way--from the customer to the vendor. And while VolP clearinghouses offer originating and terminating networks a defined split of the revenue associated with each call or ses- sion, the actual per call cost of paid transit is embedded in the overall charge. Few, if any, ISPs buy or sell transit to support VolP services alone. Transit is a generic, service-independent offering. Yet, that does not make a comparison of clearinghouse and transit economics "apples" and "oranges." On the contrary, it is pre- cisely because ISP transit charges cover "apples" (VolP) and oranges, grapefruits, kiwis, etc. (XolP) traffic that it is worth- while trying to break out that portion of the transit charge which might be reasonably assigned to VolP traffic. Transit rates for VolP, however, reflect end-to-end delivery on an IP network, excluding the last mile. In contrast, as noted above, clearinghouse rates typically reflect the last-mile access costs of PSTN termination. Transit charges are typically billed monthly based upon the size of the daily traffic stream which is regularly delivered for transit. The price may vary significantly based upon location of the network connection point(s), length of the contract, the transit capacity desired, and the agreed level of network per- TeleGeography 2001 © TeleGeogrophg, Inc. 2000 Figure 9. The ,Missing Link: IP Billing Systems At an operational level, the challenge of inter-network compensation is less about money than software--installing the right applications on network devices so that any user’s traffic, whether a corporate customer or another network operator, can accurately be metered, profiled, and billed in real time. Without this type of IP business software, inter-network payments may never progress beyond the rough justice which now exists with peedng and transit. Likewise, absent new software, ISPs will have lit- tle ability to move from fiat-rate to service-specific pdces for VolP and other value-added products. The core of any 1P traffic billing system is a mechanism for gathering customer usage data. The second cdtical element involves mediation software that processes the data and converts it in’to a billable record. Today’s IP bilking systems generally can be distinguished by their approach to these tasks. Most of the current billing vendors--leaders include Amdocs Ltd., Belle Systems NS, Convergys Corporation, Kenan (a Lucent sub- sidiary), Portal Software, Savera Systems, and XACC]-Technologies-- typically create billing records based upon log files generated by net- work routers, gateways, and application servers, or by using software probes (for example, generated by RMON (ReMOte Network) diag~ nostic tools) to profile IP traffic streams. Cisco’s NetFIow software can provide similar information based upon an analysis of incoming packets received by a router but cannot identiflj the particular appli- cations involved. These data collection techniques are used to com- pile session-specific information for a user. The resulting billing records are similar to the call detail records (CDRs) generated by telco switches--records which also can provide the basis for carrier- to-carder settlements in the telephone world. An even more sophisticated set-of Intemet monitoring and billing tools has been developed by NARUS, Inc. (www.narus.com). The NARUS technology, known as Semantic Traffic Analysis (S’I’A), enables a service provider to detect and analyze all of its IP data streams in real time without mining log files or using RMON probes. STA relies upon network appliances with proprietary software which are con- nected at key locations in the service provider’s network. The soft- ware is able to extract packet header and payload information from each packet, and thus can reconstruct any user’s sessions in real time. Needless to say, the granularity of this data provides the basis for extraordinarily detailed traffic analysis and bilhng options. NARUS markets these applications separately through its IP Billing Mediation and l~telligence product lines. The application of new IP billing tedhnologies is likely to be enhanced in the next few years by the work of the Intemet Protocol Detail Records [IPDR) group, a cross-industry forum, founded by Dr. Jerry Lucas. It aims to standardize the format for exchanging usage infor- mation on IP networks. In June 2000, the group published its first technical specification. It proposes rules for ordering the relationship between the IP network devices which generate usage data and the business support systems (BSS), such as billing and fraud detection software, which must work with this data. The new specification also relies upon XML [Xtensible Markup Language) to define several sam- ple records for billing VolP, streaming video, and e-commerce ser- vices. The specification is available from www.ipdr.org. For Lucas, standardized IPDRs have become a quest. The big busi- ness event of the 21 st century, says Lucas, will come when "two car- ders use IPDR for inter-carder clearing [and] financial settlements." That may take time, "The toughest nut to crack in the billing indus- try is moving billing to center stage in carrier CEO thinking," adds Lucas. However, every ISP’s future is at stake: "The corporate marching order has to be ’We can’t beat the competition without IPDR.’" © TeleGeography, Inc. 2000 formance, which is typically written into a Service Level Agreement (SLA). For transit originating in the U.S., typical fiat rate charges in mid-2000 for a 45 Mbps connection were $13,000 to $23,000 per month, depending upon the average throughput. Prices offered by non-U.S. 1BPs vary significantly. One public benchmark for transit pricing is Telstra Big Pond Direct Internet Access (see telstra.com.au/bigpond/direct/pricinga2.htm). As of October 2000, Australian transit pricing for a 512 Kbps connection was A$8,400 for less than 40 percent utilization and A$0.29 per megabyte received thereafter. The great majority of lSPs (and large corporate users) still pur- chase upstream transit services directly from Tier One or Two backbone providers. However, comparison shopping can be quite difficult as IBPs typically require users to keep the terms confidential; after all, IBPs do not consider themselves "com- mon carders" and have no obligation to publicly tariff their services. They also may price discriminate, subject only to general competition laws. Also, because the access points for competing IBPs may be geographically disparate, any cus- tomer wishing to shift its business to a new provider may face significant transaction costs. For these and other reasons, London-based bandwidth exchange Band-X launched an Internet transit exchange in 1999. [7] The Band-X routed service consists of a PoP through which downstream networks can gain access to tran- sit routes offered by various sellers. Similar to the clearing- house for voice minutes, Band-X routed allows buyers to select between multiple anonymous sellers, allowing quality and price to be their guide. (For a detailed discussion of Band-X’s © TeleGeography, Inc. 2000 TeleGeography 2001 IP muted service, please see the case study in the "Primer on Bandwidth Exchanges" in this report.) Band-X believes that its exchange-based transit contracts enable ISPs to provision incrementally, thus permitting shorter lead time for handling new customers or services, and also reducing the inventory cost of holding bandwidth on reserve. Band-X also maintains that the build-out of transit exchange globally will give many regional ISPs the ability to provide global connectivity without necessarily transiting U.S. back- bones. With this overview of transit in mind, let us once again try to "follow the money" for VolP service. As noted, many ISPs carry VolP traffic outside the clearinghouse system by simply bundling it with other IP traffic streams for onward delivery via their standard transit contracts. The cost per call can be astonishingly low. For instance, according to Philip Mutooni, a product manager with iBasis, a reasonable working assumption is that a I Mbps transit service purchased from a Tier One IBP could handle an average of 100 to 150 Internet telephone calls per minute, assuming roughly 10 Kbps is allocated per call and standard encoding schemes are used. From the previous discussion, we know that a mid-range estimate for I Mbps transit is $500 per month. Thus, taking the low end of Mutooni’s assumption, for $500 a month an ISP could theoretically transit 100 VolP calls a minute 24 hours a day, making the average cost per call less than $0.00001 per minute (although actual costs would be higher because the network would not be fully utilized all month long). [8] It is numbers like these that may help to explain why the stock market has recently marked down the value of legacy long distance carriers so sharply (recall Figure I). Two final points about transit and network-to-network com- pensation. Though the cost of transiting a few thousand min- utes of VolP traffic may be financially trivial, overall transit costs can still be significant for downstream ISPs, as are the in- payments, of course, for the upstream IBPs. Further, it may be misleading to classify all downstream (or upstream) ISPs alike. Mid-tier ISPs may buy upstream global connectivity from a Tier One player but, in turn, offer value-added services (e.g., web hosting and caching) to smaller regional players, employ- ing their upstream capacity to provide connectivity to their downstream customers: Many off-shore ISPs find themselves in this position, for example, offering in-region transit, termi- nation and colocation services to "foreign" ISPs or web con- tent providers, while simultaneously buying transit from Tier One backbones. On closer look, therefore, the flow of money associated with transit services may not be as black and white as is commonly supposed. C. Does Inter-Network Compensation Really Matter? Let’s step back and take stock. We began by pointing to the economic threat the Internet poses to large international oper- ators, as switched traffic bypasses the settlement system, and then looked at the limited opportunities for alternative net- work-to-network payments for international VolP operators. The result: as with other Internet services, most global VolP players must rely primarily on end-user charges to cover their costs. To which many readers may say: "We know that. But so what?" Have we got the premise wrong? That is, do ISPs really need a new network-to-network compensation model to accommo- date large volumes of VolP and new "value-added" Internet services? Or will current arrangements and, in particular, end- user charges plus paid transit suffice? There are several outstanding issues to resolve before we can answer these questions. To be equitable, network settlements should be cost-based. But costs are a moving target. Core transmission and switching costs for many ISPs are falling by perhaps 30 to 50 percent a year (as they are for long-haul telephone operators) and are now only a small percentage of end-to-end costs. [9] Selling, General, & Administrative (SGOA) costs plus last-mile (local access) expenses are now the key to competitive success for long-haul data and voice networks alike, and ISPs should not expect settlements to cover most of that (see Figure 8. WorldCom Costs for Transporting U.S. Long Distance Voice and Data Traffic). Settlements are impractical, say the skeptics. Why? There is no agreed way to allocate most ISPs"’incremental (or even fixed) costs to VolP and other specific services. Contrast the extensive costing models developed for telephone services (see "The New Calculus: A Primer on Interconnection Accounting" in this report). Moreover, Internet services are too fluid, as is the quality--another measurement problem. In any case, the transaction costs for implementing a defensible cost model would likely outweigh the benefits. If ISPs think they have a "free rider" problem with other networks, they should find an engineering fix and, if they can’t, they should get their end-users to pick up the tab or limit the access of any "resource hogs." This argument has a second variant: settlement schemes are premature because we don’t have the right tools yet. There are but a few software packages available for metering and billing service-specific IP streams. ISPs are testing different vendors but, as yet, there is no agreement on the key compo- nents which would make industry-wide settlements (e.g., call data records for VolP) viable (see Figure 9. The Missing Link: IP Billing Systems). And, even assuming common billing records are agreed, the industry will also need a standard protocol for an ISP in one TeleGeography 2001 © TeleGeographg, Inc. 2000 Figure 10. Network Choke Points: Telcos vs. IP Backbone Providers Tier One Internet Backbone Providers Source Transmission Famht~es Incumbent International Operations incumbent typically owns only submanne cable landing statmns and available backhaul to domestic~network; it also may own cable capacity on preferential terms. Unrivaled customer base makes top 4 or 5 Internet Backbone Providers (IBPs) essential facility for downstream service providers. Position of IBPs may be enhanced through bundled offering of international private line (IPL) facilities for access by "off-shore" customers. Interconnection Mandated by laws to ensure non-discriminatory terms Unregulated. Favorable terms for "peer" networks; and conditions for competitors, others must pay transit charges. Private peering arrangements typically provide better Quahty of Service (QoS) than public exchange points. Traffic Management Enhanced services require access to intelhgent network Top I BPs may use network reach to set de facto QoS (e.g., network databases, service control points) and and billing standards. Traffic filtering/metering for may not be resold to competitors. QoS/billing as well as caching technologies could be used to favor affihated media sites. Numbering Control of telephone numbering plans (e.g., for mobile Scarcity of I P addresses and lira=ted portability of phone numbers) and historical lack of local number addresses may "lock in" major customers and reduce portability has enhanced incumbent’s position, competition. Source: TeleGeography research @ TeleGeography, Inc. 2000 part of the world to swap the records with its counterpart in another region to ensure timely payment. That is why clear- inghouses all require their users to sign onto a standard billing package; large equipment vendors that are developing OoS software for installation on corresponding networks do the same. Once these products prove themselves, the argument goes, new payment models may be feasible. The market is working, the argument continues. IP phone traf- fic is growing rapidly. To date, ISPs have afforded the circuits and routers for VolP and other much more bandwidth-inten- sive services (e.g., Napster for swapping MP3 audio files). VolP players have raised large sums in the equity markets, with players such as iBasis, ITXC and Net2Phone going public in 1999. In addition, with large cash infusions from legacy players, such as AT&T, it appears the VoIP market will contin- ue to grow, whether a network-to-network compensation scheme is hammered out in the immediate future or not. So, do the skeptics have the better of this argument? Or will network-to-network compensation schemes be an integral part of tomorrow’s lnternet? Let’s start with the "falling costs" argument. The cost of band- width and routers is declining, to be sure, but ISP expenses may still be rising, largely because the Net is getting more bandwidth-intensive per customer. [10] The pace of equip- ment and transmission cost declines, in other words, may not be keeping pace with increases in end-user demand. If this is the case, the cost issue is real and one felt most keenly by overseas ISPs in Asia where the bandwidth market is tess com- petitive. Moreover, non-U.S. ISPs typically must pay the full cost of the links that connect them to the U.S. backbone even though, once provisioned, the links are used in both directions (recall Figure 4). And cost concems are not limited to "off-shore" ISPs. it is no secret that profits in the ISP business are scarce: very few ISPs (apart from AOL) operate in the "black" and all the major VolP companies operate at significant losses. So too, appar- ently, do the Net’s main backbones. In June 2000, John Sidgrnore, WorldCom’s CEO, told the SuperComm trade show in Atlanta that despite 1999 revenues of $34 billion (exclud- ing the company’s Brazilian carrier, Embratel) with $11 billion of that from its data and Internet divisions, WorldCom’s data networking business was still not profitable on a stand-alone basis. 11 Genuity’s 2000 stock prospectus, which offered perhaps the most detailed financial information yet on the backbone indus- try, told a similar story, lntemet access and transport provi- sions accounted for more than $600 million, or close to 90 percent of Genuity’s total revenue in 1999. However, operat- ing expenses exceeded $1.55 billion. Hence, if Tier One IBPs, such as WorldCom and Genuity, cannot break even under the current scheme--and these players receive the lion’s share of transit payments--no wonder the downstream players are hurting. On to the last argument: billing tools. This is something of a chicken-and-egg situation. Without much wider agreement on the importance of inter-network compensation, metering and billing may seem like a wasted effort. But without these tools, no network-to-network compensation scheme is practical. The tools are coming if only because networks need to better man- age their own customers’ demands (e.g., to deter bandwidth "hogs"). And, once ISPs deploy the software needed to more accurately profile and bill for the network load that each cus- tomer imposes, doing the same for other networks--in effect, large customers~is likely to follow. © TeleGeography, Inc. 2000 TeleGeography 2001 Some fear that the tools necessary for real time monitoring of network workloads could lead to new choke points on the Net (see Figure 10. Network Choke Points: Telcos vs. IP Backbone Providers). It is said, for example, that larger networks could use new metering and traffic management software to enhance rather than to reduce their market power. On the other hand, if such tools are made widely available (e.g., from independent third-party suppliers), they may well allow any ISP to make a much better assessment of the costs and benefits of intercon- nection. Indeed, traffic monitoring and billing tools could be the Internet’s best hope for building a sustainable provisioning model when telephone bit streams are but a smaller and smaller part of the overall data flow. One final point: market failures can be hard to detect, like errors of omission, especially if the main impact is far away (e.g., in poor countries where backbone costs are still a barri- er to entry). Likewise, when capital is plentiful and someone else is paying for the corporate overheads, the "it isn’t broke" argument only goes so far; wait for the next market down draft. The major IBPs are all losing money, and the market cap of most long distance telecom providers is at a three-year low (again see Figure 1). But the real problem with the "it isn’t broke" argument is that it tends to ignore what has made the Internet so valuable to date--its ubiquity and open, end-to-end access for new ser- vices. Absent some new approaches to network cost-sharing, there are likely to be at least three consequences: (1) a more limited and slower regional build-out of the Net; (2) less com- petitive markets; and (3) the increasing rise of private intranets--gated communities--with higher interconnect fees and QoS for mission-critical corporate, e-commerce and, yes, entertainment services, while the public lnternet--the global commons--becomes poorer, more congested and unreliable. [12] Indeed, a two-tier lnternet already exists in many respects. That may not be all bad. Not everyone wants or wishes to pay for a telephone system QoS, and private intranets (corporate firewalls) can co-exist with the public Internet. But we should be under no illusions regarding the direction in which our legacy compensation schemes for IP traffic are taking us, or the consequences for the public Internet. And if you are not persuaded by principle alone, consider the bottom line. The investors who continue to sell off Internet backbone and long distance telco stocks alike may just have a message for the industry: show us a sustainable business model for hauling long distance IP traffic--whether based on inter-network payments or otherwise--or we will continue to put our money elsewhere. This essay was written by Cregory C. Staple, the founder of TeleGeography, Inc. Mr. Staple is currently an editorial advi- sor to TeleGeography and a partner in the Washington, D.C., law firm of Vinson & Elkins L.L.R The author serves as U.S. regulatory counsel to Telstra Corporation Ltd. and certain other international telecommunication service providers; how- ever, the views expressed here are his own. The author grate- fully acknowledges the assistance of Trista Schroeder in preparing this work. Mr. Staple may be reached by email at gstaple@velaw, com. Notes: [1] See, e.g., Gregory C. Staple, "A Primer on International Simple Resale," TeleGeography 1996/67. [2] Some incumbent operators--notably AT&T Jens in Asia--have sought to steal a match on the competition by providing VolP service first. ~his is happening even in developing countries. For example, in June 2000, the Telephone Organization of Thailand [TO’F] won regula- tory approval for a new VolP service. [3] The absence of analogous payment schemes is due, in part, to the absence of d~rect connection or privity between most of the originating and terminating networks which comprise the public lnternet. It also stems from the Internet’s U.S.-centdc origins (which generally gave U.S. ISPs the upper hand in setting connection terms), and the asymmetric nature of much Internet traffic (e.g., to and from web sites). For a full history of the N~t’s financial architecture, see the Internet primer Hubs and Spokes: A TeleGeography Internet Reader [TeleGeography, Inc., 2000). [4] Some entrepreneurs have reportedly taken this to extraordinary lengths by, for example, installing tens of rack-mounted mobile phones to provide domestic PSTN access for international telephone services. Even m developing countries, the mobile service market usually is com- petitive, and mobile operators may be only too happy to provide addi- tional phones, provided the user pays the connection charge. That allows a business customer to mix in its rebound international traffic with the mobile operator’s other traffic stream for PSTN ]nterconnec- tion, thus reducing its visibility to the incumbent operator. Also, the mobile operator generally does not have an international service and, hence, has no interest in protecting existing settlement revenue. [5] The evolution of peering and transit relationships is reviewed at greater lenBth in M. Kende, "The Digital Handshake: Connecting Internet Backbones," OPP Working Paper No. 32, Office of Plans and Policy, FCC (September 2000) available at www,fcc.govlopplwork- ingp.html. See also the discussion of peering terms in The Internet’s continued on next page TeleGeography 2001 © TeleGeography, Inc. 2000 Notes (continued): Coming of Age, National Research Council IOctober 2000) available at books.nap.edu/catalog/9823.html. [6] Though clearinghouses are important so far as internetwork com- pensation is concerned, they play a much smaller role in terms of the international VolP market as a whole, accounting for approximately 15 percent of global VolP traffic. The bulk of such traffic is now handled by wholesale VoIP carriers such as rrxc, iBasis, and deltathree.com, which may also share a portion of their revenues with ISP affiliates that terminate traffic in countries where the wholesale carrier lacks a pres- ence. In 1998, when AT&T and BT first announced the Concert joint venture, which would pool their international networks and migrate traffic onto a new IP-based global backbone, it was speculated that this novel venture would pioneer new correspondent agreements and set- tlement schemes for the carriers which have international operating agreements with BT and AT&T. For example, it was suggested that Concert might offer existing BT correspondents a common U.K.-U.S. termination rate or a set of regional rates based on the cost of trans- mitting traffic via Concert’s IP cloud. [7] Full disclosure: TeleGeography, inc. became an independent sub- sidiary of Band-X Ltd. in September 2000. [8] Calculated as follows: $500/(60 minutes x 24 hours/day x 30 dayslmonth x I00 callslminute). [9] An instructive and up-to-date review of the price-performance improvements in network switching and transmission facilities can be found in a September 2000 industry study co-authored by McKinsey & Company and J.R Morgan: "Backbone! How Changes in Technology and the Rise of IP Threaten to Disrupt the Long-Haul Telecom Services Industry." [10] See K.C. Coffman and A.M. Odlyzko, "lnternet Growth: Is There a ’Moore’s Law’ for Data Traffic?" Preliminary version, July 11, 2000, available at www.research.att.com/-amo/doc/networks.html. [I I WorldCom’s public reports do not break out profitability by mar- ket segments, however. For example, the company’s Securities and Exchange Commission (SEC) Form I0C~ for April-June, 2000, states: "Communication services are generally provided utilizing the Company’s network facilities, which do not make a distinction between the types of services. As a result, the Company does not allocate line costs or assets by segment." [12] See, e.g., Jonathan Angel, "Toll Lanes on the Information Superhighway," Network Magazine, February 2000, available at www.networkmagazine.com. Facilities TeleGeol~raphy :2001 © TeleGeographg, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 MANs: The Golden Mile What is a MAN? Fundamentally, Metropolitan Area Network (MAN) is just a fancy term for an intracity communications system. A MAN connects the major communications nodes of a city, includ- ing Central Office switches, telehouses, Internet exchange points, and corporate business centers. In the overall net- work hierarchy, MANs lie between the long-haul and access networks (see Figure I. No MAN is an Island). In the past decade, a shortage of MAN bandwidth emerged as network build-out near the edge did not keep up with the supply boom on long-haul networks. To make things worse, demand from end-users--both for international and intracity connections~--has increased at an astounding rate. This shortage has sparked a MAN building boom--first in the U.S. and then in Europe. Asia has not been too far behind, with new networks laid in Hong Kong, Tokyo, Sydney, and other places. In most international business cities, at least three [and often more) networks are being constructed, creating a unprecedented capacity infrastruc- ture filled with many hundreds of fiber pairs. To the uninitiated, MAN deployment may seem simple because the distances involved are insignificant in compari- son to crossing an ocean or sending radio signals into outer space. As many have discovered, however, building a MAN is very complex. In this article we present the MAN as both an element of an international communications system and a network of its own. Who Needs a MAN? Even though MANs have never been isolated networks, it makes even less sense today to see them as discrete centers of demand. On the one hand, major bandwidth consumers need seamless links between their city nodal points in the Figure 1. No MAN is an Island Core PoP Customer Premises Customer Premises Leaf PoP SAN Site Colocat~on S~te © TeleGeography, Inc. 2000 From a network topology perspective, MANs are networks that connect access networks and long-haul networks. Most MANs are designed as a ring, although larger MANs may be shaped as a series of rings, within rings. From a geographical perspective, what constitutes a metropolitan area is less clear, tn the U.S., some sprawling metropolitan areas would be considered regions elsewhere. At the same time, there are regions such as the Randstad in the Netherlands, which are Source: TeleGe0graphy research characterized by some builders (such as MFN) as a single con- tiguous market, if not exactly a metropolis. The Randstad region, one of Europe’s most important concentrations of commercial activity, includes Amsterdam, the burgeoning business parks around Schiphol Airport, Hilversum, Utrecht, The Hague, and Rotterdam. Other equally important conurbations include D0sseldorf, Dortmund, Essen, and Cologne in Germany, and the area inside the M25 dng read around London. © TeleGe0graphy, Inc. 2000 TeleGeography 2001 © TeleGeography, Inc. 2000 metropolitan area and the global network. On the other, the MAN is increasingly an extension of many corporate net- works as those companies store and retrieve information from sites located on metropolitan area networks, rather than private ones. MANs have been built and operated by incumbent opera- tots, municipal authorities, local entrepreneurs, and global super-carriers alike. For many international carriers, the main justification for building MANs has been that they could not get the high capacity they needed any other way. Until last year, incumbent telcos in some cities did not offer capacity above 2 Mbps, and many telcos still do not offer any capacity above 155 Mbps. When long distance net- works are running at 10 Gbps or more, that simply isn’t enough. Moreover, some incumbents still take up to six months to deliver very high-speed lines in the metro area. Even for lithe new carriers, getting bandwidth to buyers can take a long time. As a result, the desire to achieve point-to-point "bandwidth on demand" capabilities has driven the creation of "virtual pooling points" to speed up provisioning. A virtual pooling point is not a single telehouse or colocation site, but rather all the important, yet disparate, nodes in a metro area strung together in a single addressable system. The collec- tion of nodes represents all the major sources and destina- tions of traffic within a city; for many international players, the construction of this type of MAN is the main objective. Optical Technology and the MAN In long-haul networks, "fiber exhaust" was solved using dense wavelength division multiplexing (DWDM) equip- ment. Before DWDM, a fiber pair could carry a maximum of four channels of traffic on four different light frequencies at up to 2.5 Gbps each--a total of 10 Gbps. DWDM enables carefully tuned lasers to deliver 40 or more closely spaced light signals (often called "wavelengths" or "lamb- das") per fiber pair. Typical current DWDM configurations include 40 x 2.5 Gbps; 80 x 2.5 Gbps; 32 x 10 Gbps; and, 96 x I 0 Gbps. The commercial state of the art is 960 Gbps per fiber pair though much higher figures have been achieved in laboratories. Could DWDM be used in metro networks as well? Although almost all metro networks are based entirely on optical fiber, important differences between metro and long-haul net- works make metro DWDM solutions problematic. First, a high degree of uniformity exists in the long-haul net- work. This is not the case in the metro space. Because metro networks are much closer to the ultimate end user, they must carry a wide mix of lower speed channels, includ- ing voice channels at every node. These include dedicated private lines (64 Kbps to 622 Mbps), voice circuits (64 Kbps), ISDN circuits (64 Kbps to 2 Mbps), packetized frame relay and cell-based ATM traffic, transparent LAN (Ethernet) service, and private SONET rings. In long-haul networks, in contrast, traffic has already been converted to icientical high speed channels. Second, DWDM equipment is very expensive and inefficient for short distance links. To start, an optical network using DWDM requires lasers that cost four to five times as much as those on non-DWDM systems. DWDM equipment also consumes a large amount of space and power--two com- modities that are becoming a critical constraint in many cities. One recent estimate suggested that it would cost a long-haul carrier $100,000 per annum to house DWDM equipment in a single metro site. And finally, unlike long- haul networks, metro networks tend to have many nodes because they must add and drop off traffic at many points along the network. Until optical multiplexing equipment matures, each add-drop node in the metro network needs to also provision space for optoelectronic conversion equip- ment that compensates for signal losses created by present- ly deployed technology--an expensive proposition indeed. Optical add/drop multiplexers and wave filters may help address these problems in time. One intermediate solution is to use dark fiber with fewer channels of WDM traffic on it (four instead of forty channels of DWDM) and lease more dark fiber when the need arises. With suppliers like Metromedia Fiber Networks (MFN) deploying on average 864 fibers per cable in U.S. cities and 432 in European ones, this approach is a, viable option. However, metro network design will continue to be difficult in cities where fiber is abundant in some areas but scarce in others. Ultimately, the choice between WDM and DWDM options is determined in part by the scale of demand. If sig- nificant demand is expected, the benefits of investing in expensive DWDM equipment may outweigh the costs of incrementally adding more fibers, terminals, and regenera- tors in a less expensive WDM option. Designing the MAN Further complicating MAN deployment is the issue of archi- tectural standard. Traditionally, most MANs have been designed using either SONET ISynchronous Optical NETwork) or its close cousin SDH (Synchronous Digital Hierarchy). SONET and SDH are self-healing network archi- tectures that prevent interruption in service by rerouting traffic almost instantaneously in the event of a fiber cut. A ring topology, however, requires provisioning for the maxi- mum bandwidth required in the network on every segment irrespective of the actual load on the segment. Furthermore, half of the network remains idle for restoration purposes. © TeleGeography, Inc. 2000 TeleGeography 2001 Figure 2. ~Metromedia Fiber Network--London Area Map Metromedia Network Telecom Tower Trafalgar Square Buckingham Palace Bank of England ¯ Houses of Parliament I mile Tower of London Telehouse London Canada Tower Map by TeleGeography, Inc. Note: Map depicts approxemate year-end 2000 route, which is subject to change. Metro rings are not shown in this map. © TeleGeography, Inc, 2000 The ring topology was originally developed for fixed-band- width, circuit-switched traffic--that is, voice, rather than data traffic. Given the "hub and spoke" nature of data traf- fic flows, however, many have written off SONET rings in favor of "mesh" networks, which are better suited for point- to-multipoint data traffic flows. But SONEr and SDH are far from dead; the next generation of Ethernet standards has incorporated many of their features. (The 1 0 GigE alliance, which is developing 1 0 Gbps Ethernet, has created two ver- sions of the standard--one for LANs and one for WANs. The first commercial products to use these standards should be available by the end of 2000.) The wide acceptance that SONET currently enjoys suggests a network evolution with a gradual transition from SONET rings to meshed networks. Different solutions also exist in the type of fiber most suited for the metro space. This year Corning launched a new fiber called MetroCor, which uses less expensive transmission lasers and has the mind-boggling characteristic of dispersion that actually improves with distance~that is, the pulses of light become narrower. Hence, regenerators are unnecessary in 400 kin-long network segments--a key design objective for MAN builders. Lucent, on the other hand, has pioneered AccuRibbon fiber, which packs up to 864 fibers in a single narrow cable that can also be mass-spliced at node points~ a significant cost savings for metro builders squeezing large fiber counts in narrow underground ducts. Underground Regulation and the MAN In addition to technology choices, market regulation is another complex issue facing MAN deployment. In principle, markets are open in Western Europe, the U.S., and many Asian countries. MAN entrepreneurs ought to be able to power up the bulldozers and move right in. In practice, though, it’s far from simple. The regulatory environment for constructing MANs varies widely from one city to another, especially rules for digging the streets. According to companies such as MFN, building in cities is "an order of magnitude" more complicated than building intercity networks, largely because of rights of way issues. For example, MFN had to negotiate with over 50 rights-of-way (ROW) owners to build their MAN in New York. In some cities, municipal monopolies have been created to avoid the continuous disruption of torn-up streets. Service TeleGeography 2001 © TeleGeographu, Inc. 2000 Figure 3. Pan,European MAN Deployment Number of MANs 5 or r here ¯ 3-4 ¯ 1-2 Dublin ¯ Bit~irlgham 0slo¯ ¯ ¯ Budapest Madrid ¯ Rome ,¯ Source: TeleGeography research, company reports Note: MAN count to December 2000. In 1995, only two Metropolitan Area Network builders existed in Western Europe. Today, as many as six different builders compete for the same customers in a number of cities across this re~ion. Many of these players have also built or bought capacity on intemity tings to interconnect their MANs. The figure to the left illustrates the target cities of nine pan-European metro builders (metro builders which are building MANs in more than one country in Western Europe), These builders are: Carder 1, COLT Telecom, Globat Crossing, Global Metro Networks, GTS, Level 3, Metromedia Fiber Network, and Viatel. The figure does not include those metro networks being constructed by local muhicipalities or city- carriers, such as NetCologne (Cologne), MetroWeb (Milan}, and CompleTel (Paris). © TeleGeography, Inc. 2000 providers seeking to build MANs in these cities must then lease dark fiber or ducts built by the monopolies. These municipal bodies include StokAB in Stockholm and Sippera, which holds a franchise in the Parisian business district of La Defense. Elsewhere, regulations restrict when and how a company may dig. Tokyo utilities, for example, may only dig the streets for a few weeks in each year. In other cases, the local authorities may require providers to sham ducts, or at least to harmonize digs, as is general practice in several Italian cities. In some French cities, close relationships exist between the local city authority and ducts owners, making it difficult for third parties to get permission to dig. At the other extreme almost no restrictions exists in cities such as London. Licensed telecommunications carders have the right under "Code Powers" to dig up the streets wher- ever and as often as they like. Recently, the European Commission has stepped in to bring order to the inconsistent street digging laws in Europe. In its 1999 Policy Review on telecommunications rules, the European Commission referred specifically to rights-of-way and said that changes might be made if builders were unfairly obstructed. "There is some evidence to suggest that the currant rules am not providing new entrants with ade- quate and timely access to rights of way," it..said. A recent ruling suggested that incumbent telcos must make spare ducts available to third parties. Later, in August 2000, the European Commission announced that it had begun pro- ceedings against Luxembourg because the local city author- ity had refused Coditel, a Belgian telecommunications provider, the right to lay fiber in the city. The case is still pending at this writing. Where are the MANs? As in the long-haul market, MAN builders follow the money. The first cities to attract competitive build-out in the early 1990s were international financial centers, especially New York, London, Chicago, Paris, and Frankfurt, where as many as nine separately owned fiber cable infrastructures are now in the ground. Other global business and financial cities soon followed (see Figure 3. Pan-European MAN Deployment). So invariable is this approach to selecting MAN deploy- ments that market entrants have divided the world’s cities informally into a series of tiers, and strategists tend to talk about "first-tier, .... second-tier," and "third-tier" cities. Most business strategists identify about 30 to 40 cities among © TeleGeography, Inc. 2000 TeleGeography 2001 Figure 4. i Where are the First-’13er Cities? One of the most systematic attempts to classify and rank major dries has been made at Lougborough University, home of the Globalization and World Cities Study Group and Network [GaWC-- see www.lboro.ac.uk/departments/gy/research/gawc/rb/rb5.html). Researchers at GaWC developed a methodology that ranks major world cities into four groups, each of which contains several sub- groups. The first three groups appear heroin. The GaWC ranking is based not on population but on the city’s per- ceived significance in the global economic system. The theory argues that world cities are most easily identified by the relative availability of "advanced producer services," four categories of which are used in the GaWC schema: accounting, advertising, bank- ing, and law, GaWC aggregated.and scored public information on the concentration of firms in these areas to create its ranking, Like any other theoretical scheme, the GaWC top cities ranking does not correspond precisely to the MAN builder target cities list. Although a shared industry perception that a city is ’first’ or ’sec- ond’ tier may factor into a decision to build a MAN, other factors are important. These include its geographic location relative to the rest of the network and to other telecommunications traffic sources, the local regulatory situation, a~d a whole range of market data, such as household Internet penetration, Hence, some European cities that don’t appear on the GaWC listing are, nevertheless, extensively cabled or connected because they happen to sit between other major cities, or because the local regulatory envi- ronment is favorable, or both. Examples include Stuttgart in Germany and Strasbourg in France, Others such as Warsaw, which is on the GaWC ranking, remain relatively untouched by MAN fever. The World According to the GaWC Alpha World Cities: Beta World Cities: Gamma World Cities London San Francisco (Europe only): Pans Sydney Amsterdam New York Toronto Barcelona Tokyo ZOr~ch Berlin Chicago Brussels Budapest Frankfurt Madnd Copenhagen Hong Kong Mexico City Diisseldorf Los Angeles S~o Paolo Geneva Milan Moscow Hamburg S~ngapore Munich Prague Rome Stockholm Warsaw Source: Globahzation and World Cities StuOes Group © TeleGeography, Inc. 2000 the first-tier, with up to several hundred worldwide in the second-tier, and many hundreds of towns and cities in the third (see Figure 4. Where are the First-Tier Cities?). First-tier cities are the biggest sources of and destinations for traffic. Some estimates suggest, for example, that well over 60 percent of pan-European traffic is routed via four major cities: London, Frankfurt, Paris, and Amsterdam. Similarly in the U.S., estimates suggest that over 80 percent of data traffic terminates in 25 first-tier cities. Powerful con- centrations around one city also exist. Though statistics are sparse, some network analysts believe that well over one- quarter of corporate network traffic in France originates in Paris and as much as a half of Greek traffic comes from Athens. In other countries, especially Germany, and to a lesser degree Italy, traffic is much more widely distributed because commercial activity is more widely dispersed. Moreover, these first-tier cities tend to attract and generate even more traffic over time. In a liberalized market where cities compete for traffic, those that already have strong competitive infrastructures tend to win a significant share of new traffic. This, in turn, attracts communications-intensive companies to locate near to these major nodes, resulting in even more entrepreneurial MAN build-out. Small wonder, therefore, that competitive MAN activity is largely concen- trated in these first-tier cities. MAN deployment is beginning to disperse, however. In the U.S., for example, Brooks Fiber (acquired by WorldCom in 1998) built networks in over 90 second-tier commercial cities in the U.S. and earned a very high valuation because it was the only player in town other than the incumbent telco. In Europe, no company has emulated that approach precisely, but several players are now building in second tier cities such as Lyon, Hannover, and Birmingham. As always, however, economically less-developed regions and countries may have a long wait before MAN build-out reaches their cities. Who is Building the MANs? MANs come in many shapes and sizes and usually reflect characteristics of their builders (see Figure 5. The Urban Jungle). For example, full-service global telcos often have the most intricate MANs because they need to connect cor- porate customers directly. Wholesale carriers’ carriers tend TeleGeography 2001 © TeleGeography, Inc. 2000 Figure 5. TheiUrban Jungle Type of Company Market Orientation Incumbent telco Full-service global telco Municipal monopoly City carrier Metro carriers’ carriers Long-distance carriers’ carrier ’New wave’ corporate MAN builder Legally obligated to connect any metropolitan company or household; MANs serve all traffic and customer needs Offers complex services for multinationals and other large organizations Leases fiber or ducts to licensed carriers Acts mainly as wholesaler to other carriers Wholesales fiber; mainly to other carriers, but also to enterprises Mostly builds MANs for own needs; may bundle metro capacity in some contracts Offers very low cost bandwidth for metro corporations and to connect ’tall shiny buildings’ (sometimes known as a B-LECs) Network and Technology Very extensive, reaching almost all buildings in the city; often based on ’legacy’ technologies Extensive; mostly based on own fiber and ducts (though this may change) Fairly extensive; based on ownership of rights-of-way, ducts and fiber Scope is variable; almost all build complex networks in commercial areas; tend to own multiple ducts; fiber of varying age Focuses on key nodes and commercial districts; deploys multiple ducts and state of the art cable and fiber Simple MANs focusing on major nodes such as high- level switching nodes, telehouses and Internet NAPs; may lease dark fiber and deploy DWDM; may lay cables Major commercial district initially; leased dark fiber; DWDM; gigabit Ethernet routers Examples BT, Verizon, Deutsche Telekom COLT, WorldCom StokAB (Stockholm), Sippera (Paris La Defense) NetCologne (Cologne) MetroWeb (Milan) CompleTel (Paris) MFN, Global Metro Networks GTS Cogent Main products Retail and value-added products; may choose not to sell dark fiber, but in some cases legally obligated to offer both dark fiber and ducts Retail and value-added products; will not sell dark fiber Dark fiber Dark fiber, bandwidth, as well as enhanced services Dark fiber Bandwidth, IP transit, wholesale minutes; some will sell dark tibet; some offer ducts 10 Mbps/1 Gbps Source: TeleGeography research © TeleGeography, Inc. 2000 to have quite extensive networks and to lay a lot of fiber and multiple ducts because their customers tend to lease dark fiber. Retail and wholesale long distance carriers often build skeletal MANs that simply link five to ten major telecommunications nodes or server farms on one or two leased dark fibers. Not all carriers can be neatly buttonholed: Level 3, for example, is primarily a long distance player but is building fully cabled MANs, rather than leasing dark fiber locally. Level 3 competes not only with long distance carriers, but also with metro network operators such as MFN. Carrier1 has gone further, leasing ducts in its Amsterdam network to other carders to pay for the construction of its own network. What is clear is that different types of MANs will continue to exist. Buyers of capacity need to weigh the relative advan- © TeleGeography, Inc. 2000 TeleGeography 2001 tages and disadvantages of each type of MAN design and the business models of its builders. Choosing Your MAN There are currently several different product options of vary- ing complexity: Dark fiber. Now becoming widely available in first-tier cities, dark fiber is generally sold on long leases of 0 or 15 years, which has its caveats. First, the buyer needs to be realistic about the likely commercial lifetime of the fiber, given that most of the components in the system have a lifetime of less than five years. Second, while most dark fiber leases are priced per meter per annum, they also include an annual operations and maintenance charge. Depending on specific lease provisioning details--whether charges are fixed or variable, for exam- pie--such charges can become a significant cost (for more on bandwidth leases, see International Bandwidth 2000). Wavelengths. Recently available in major markets, wavelengths are usually sold at 2.5 Gbps increments. ]P over WDM is widely seen as a desirable solution for cus- tomers who are data-centric and carry mostly IP traffic. However, the management of wavelengths in metropoli- tan networks continues to be an area of controversy and ongoing experimentation. Dedicated Channels or Private Lines. Dedicated cir- cuits are a widely available and mature service, usually provided on short leases at speeds from 64 Kbps up to 622 Mbps. In cities where the only provider is the incum- bent, it can still be very difficult to get any service above 2 Mbps, but as competitors build out, this situation will change quickly. To some degree, the type of buyer determines the product choice. Large telcos and ISPs increasingly want to buy dark fiber, because they have enough bandwidth demand to jus- tify the up-front cost and the expertise to equip and man- age the network. Moreover, it is somewhat easier to control costs in a dark fiber network, since the cost of upgrading the network will almost certainly be lower than the cost of addi- tional leased bandwidth. Wavelengths are also attractive for this kind of buyer. Smaller, inexperienced carriers and corporate end-users, on the other hand, tend to prefer to lease fixed SONET or SDH bandwidth. However, the larger and more sophisticated multinational end-users are starting to lease dark fiber. MFN says that while the number of end-user customers (includ- ing the New York Public Library and Chase Manhattan Bank) is small, it is also growing rapidly. In addition to the various product options above, other dif- ferentiating factors include the following: Proximity. If a MAN doesn’t connect directly to the buyer’s building or the service provider’s colocation facil- ity, it had better come pretty close. The further away from the MAN, the more money and time it will require to get a leased line from the incumbent telco. Service Level Agreements. Metro networks are particu- larly vulnerable to breaks. Buyers need to know about the level of redundancy employed on routes and critical components, especially if the network does not use stan- dard SONET/SDH ring technology, Provisioning times. Difficulties in forecasting demand are driving on-demand bandwidth provisioning. Clearly if it is necessary to dig to the buyer’s building, provisioning will take weeks or months. But even where a buyer is already connected into the fiber, a request that entails complex routing may still take weeks to fulfill because engineers may need to visit many sites to manually fit the necessary equipment. Understanding the network config- uration and what is required to add bandwidth, allow buyers to anticipate provisioning lead-time. Bundled telehouse space. Bundled offers can be very attractive to users in cities where colocation space is a premium. Paying for Your MAN Anecdotal evidence from buyers suggests that prices for metropolitan capacity are still not falling nearly as fast as long-haul pricing. In general, metro prices are falling by around 20 percent a year in Europe, against 50 percent a year in the more competitive parts of the long-haul market. As a result, the effective gap between long-haul and metro pricing continues to widen. Broad variations in pricing exist between cities. Some com- panies claim that the price of dark fiber varies from a low of one dollar per meter per year in some German cities to as much as four dollars per meter per year in Paris, while some non-utility providers may be charging up to eight dollars per fiber pair. in some cases, providers add an Operation and Maintenance (O6dVI) charge, which will typically be two to three percent of the lease price per year. As in the long-haul market, new entrants are creating inno- vative pricing schemes to attract customers. For example, MFN allows customers with requirements in several cities to sign a general frame contract that entitles them to draw from a "bank" of fiber when they need it in the cities where demand emerges first. Like other providers, MFN also dis- counts quite steeply for volume. TeleGeography 2001 © TeleGeographg, Inc. 2000 Pricing schemes also vary by type of service. Wavelen~hs and SDH channels tend to be priced by distance, by the size of the channel, and the number of connected nodes. Fixed SDH channels, available at 155 Mbps (and in some cases 622 Mbps) cost around $50,000 per year in European cities. However, the variation in pricing levels and method- ology is so wide and so heavily customized that price points are difficult to determine. Some more extreme models, don’t charge by distance, except for the final "drop" to the customer site(s) in the form of a one-off connection charge, rather than an annual leasing charge. Conclusion Metropolitan area networks, like the cities they serve, are complex, expensive to build and are often difficult to man- age. The problems they present for network planners and designers are more formidable than those of long-haul net- works. Yet it is generally agreed that MANs are going to be one of the most dramatic areas of development in telecommunica- tions over the next two to three years. MANs perform a vital role in collecting, aggregating and grooming traffic for transmission onto international networks and in delivering services to end-users. Effectively implemented, they will be important catalysts and generators of international band- width demand. ~i~..~ © TeleGeography, Inc. 2000 TeleGeography 2001 Submarine Cable Systems Figure 1. Transoceanic Submarine Cable Capacity Interregional Capacity Trends Gbps 7,000 6,000 5,000 ,000 3,000 2,000 1,000 ~~ 200,2 U.S.-Europe~~~~/ 2001 U.S.-Latin America ~~ 2000 U.S.-Asia 1998 Europe-Africa-Asia 1996 1997 1995 Aggregate Capacity (Gbps) 1995 1996 1997 1998 1999 2000 2001 2002 U.S.-Europe 23.0 23.0 23.0 83.0 178.0 2,157.4 4,917.4 6,822.4 U.S-Latin America 4.3 10.8 10.6 10.6 15.6 275.6 1,595.6 4,165.6 U.S.-AsJa 4.1 4.1 14.1 14.1 104.1 344.1 384.1 1,224.1 Europe-Asia 1.1 1.1 11.1 11.1 31.1 31.1 71.1 171.1 Total 32.5 38.8 58.8 118.8 328.8 2,808.2 6,988.2 12,383.2 Source: TeleGeography r~search © TeleGeography, inc. 2000 Note: Capacity figures denote lit ca pac~ atl~e end of a particular year. Projected capaci~ is based on reported initial capaclty of systems during the yea~they are scheduled to be ready for service and assumes that cable systems with announced DWDM upgrade schedules will upgrade to one-half potentJal capacity by year- end 2002, TeleGeography 2001 © TeleGeogr~phg, Inc. 2000 Figure 2. Submarine Cable Deployment Trends 3,000 2,500 2,000 1,500 1,000 5OO 1996 ¯ U S.-Europe © U.S -Asia ¯ U.S-Latm America Europe-Afr=ca-Asia ¯ Intra-Asia FLAG Europe-Asza TPC 5 TAT 12/13 APCN ¯ __~ 1997 Submarine Cable Deployment History North Asmn Cable Tycom (Level 3) Trans-Atlant~c ARCOS 1 ¯ ¯ ~\~ ¯ ¯ EAC Mercus-1 FLAG Atlantic-1 ¯ APCN-2 360atlant=c SAm-1 Gemini 1998 AC-1 ¯ Chlna-U.S. Atlantis-2 © PC-1 Columbus-Ill SeaMeWe-3 G-P © [] Yellow/AC-2 ~SAC ¯ 360americas Austraha-Japan Japan-U.S. TAT 14 ©¯ © PAC Americas-2 Maya-1 Southern SAT-3/WAS C/SAFE 1999 2000 2001 2002 Ready for Service Date Route Kilometers and Landing Points 0o) 6 9 2 6 4 6 9 9 24 U.S.-Latin America Europe-Africa-Asia 40.000 km Landings Note: Included cables have a capacib/of 10 Gbps or more and will be ready for service byyear-end 2001. The C2C Cable Network is omitted from the top figure because its announced maximum capacity of 7,680 Gbps, mere than twice of any previously announced cable, would exceed the scale of thefigure. Source: TeteGeography research © TeleGe, ogra phy, tnc, 2000 TeleGeo{~raphy 2001 © TeleGeographg, Inc. 2000 Figure 3, Major Submarine Cable Growth Submarine Cable Systems Above 10 Gbps (1999) Submarine Cable Systems Above 10 Gbps (2002) System Capacity (Gbps) >2,560 2,560 1,280 640 10 Note: Fjgure includes systems with a capacity of t0 Gbps or more and are in-service atyear-end 2002, Cables have been scaled to their maximum upgradeable capacity, Source: Te~eGeography research © TeleGeography, Inc. 2000 TeleGeo~raphy 2001 © TeleGeographg, Inc. 2000 Ready for Service Fiber Bit Capacity Length Cost (RFS) Pairs Rate Channels (Gbps) (km) (US$ Millions) Europe-Africa-Asia Africa ONE 40 2002 4 10 16 640 32,000 1,900 FLAG Europe-Asia November 1997 2 2.5 2 10 28,000 1,500 SAT-3/VVASC/SAFE October 2001 2 2.5 16 80 18,900 796 SeaMeWe-3 September 1999 2 2.5 8 40 38,000 1,173 Intra-Asia APCN February 1997 1 5 4 20 5,234 550 APCN-2 September 2001 4 10 64 2,560 17,000 1,000 C2C Cable Network December 2001 8 10 96 7,680 18,000 2,000 East Asia Crossing (EAC) December 2000 4 10 64 2,560 17,700 1,280 North Asian Cable (Level 3) June 2001 4 10 64 2,560 6,000 n.a. U. S,-Latin America 360americas 2Q 2001 4 10 32 1,280 22,500 940 Americaso2 May 2000 4 2.5 8 80 7,350 365 ARCOS 1 September 2000 4 10 64 2,560 8,600 400 Atlantis-2 February 2000 1 2.5 8 20 8,500 230 Columbus-Ill December 1999 2 2.5 8 40 10,000 236 Maya-1 May 2000 3 2.5 2 15 4,962 152 Mercus-1 December 2001 4 10 64 2,560 10,000 950 Mid-Atlantic Crossing (MAC) February 2000 2 10 16 320 6,700 415 Pan-American Crossing (PAC) March 2000 2 2.5 16 80 9,000 280 SAm-1 March 2001 4 10 48 1,920 23,000 900 South American Crossing (SAC) September 2000 4 10 32 1,280 15,000 800 ,. U.S.-Europe 360atlantic March 2001 4 10 48 1,920 12,200 630 Atlantic Crossing-1 (AC-1) May 1998 4 2.5 16 160 14,521 750 FLAG Atlantic-1 (FA-1) March 2001 6 10 40 2,400 12,250 1,200 Gemini Cable System February 1998 2 2.5 6 30 12,115 600 TAT 12/13 September 1996 2 5 3 30 12,766 750 TAT 14 December 2000 4 10 16 640 15,000 1,500 TyCom Trans-Atlantic July 2001 4 10 64 2,560 13,000 650 Yellow/AC-2 September 2000 4 10 32 1,280 6,000 800 U.S.-Asia Australia-Japan July 2001 2 10 32 640 7,000 450 China-U.S. Cable System January 2000 4 2.5 8 80 30,800 1,400 Guam-Philippines (G-P) March 1999 2 2.5 4 20 3,600 100 Japan-U.S. Cable NeWvork (JUS) October 2000 4 2.5 64 640 21,000 1,000 Pacific Crossing-1 (PC-l) December 1999 4 10 16 640 21,000 1,200 Southern Cross Cable Network November 2000 3 2.5 16 120 30,000 900 TPC 5 January 1997 2 5 2 20 25,000 1,240 TyCom Trans-Pacific & Northern Asia July 2002 8 10 64 5,120 32,000 n.a. Notes: This table includes submarine cable systems that are currently in operation as of October 2000 or will be ready for service by year-end 2002. Reg=onal submarine cable systems and systems that have a capacity of less than t0 Gbps have been emitted. Technical cerffigurat~on denotes maximum upgr~deabte c~p~ity. Source: TeleGeography research © TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Figure 5. Map of Major Submarine Cables in the United States (Pacific Coast) Note: The following maps include submarine cable systems in operation as of October 2000 and those that will be ready for service by year-end 2002. Regional sub- marine cable systems and systems that have a capacity of less than 10 Gbps are omitted. Source: TeleGeography research, FCC filings, and company reports © ToleGeography, Inc. 2000 TeleGeo~aphy 2001 © TeleGeographg, Inc. 2000 Figure 6. Map of Major Submarine Cables in the United States (Atlantic Coast) & the Caribbean Now York Stations Mastic Beach o-- TAT 12/13 Shirley o-- Yellow/AC2 Brookhaven o~ i AC-1 MAC Crab Meadow Beach Long Beach FA-I New Jersey Stations Wall Township TyCom Trans-Atiantic Gemini Manasquan ~r’~"’----"----" TAT 14 Tuckerton -- : 360americas Mercus-1 Lynn Greenhill New York]Now Jersey Stations Ha~ax Vem Beach Boca I]aton Hollywood CancSn o, I.de CozumeiO Puerto Puerto Trujillo "~,Lempira Cortes ~’ Bluefields o~ MasSe~ Catls~nd CrookedlslandO--~! Providencia~s -- ~unta Cana o Punta F #~CCa Barranq=illa Puerto Viejo\ Camuri Gorda Uzqmo Cayenne Note: Please refer to the note on Figure 5 for an explanation of this map. © TeleGeography, Inc. 2000 © TeleGeographtj, Inc. 2000 TeleGeography 2001 Figure 7. Map of Major Submarine Cables in Europe Cornwall Stations TAT 14 Yellow/AC-2 "~ Bude Bay AC-1 FA-1 -~b~ Whitesands Bay TAT 12]13 ~o,~ Land’s End FA-1 FLAG Europe-Asia "~i Porthcurno Gemini f~o. Pentewan TAT 14 SeaMeWe-3 o Goonhilly Downs arreNebel Oxwich Bay Pl~rin Penmarch Norden ~Beverwijk Kat~viik aan Zee Oostende St. Valery Lisbon ’ Seaimbra Casablanca Valencia Algiers Pra~a Nouakchott Note: Please refer to the note on Figure 5 for an explanation of this map. © TeleGeography, Inc. 2000 TeleGeography 2001 © TeleGeography, Inc. 2000 Figure 8. Map of Major Submarine Cables in Latin America and the Caribbean Puerto Trujillo ~’~LLemp~ra o Puerto Cabezas Bluefields o e Town Puerto Plata ~_ Punta Cana Miramar Barran~l~lilla ~ Cartagena Puerto Viejo\ Puma Gorda aventura Cayenne Fortalez Lurin Natal Jo~o Pessoa Rec~e Macei6 Aracaju Salvador V~6ria Rio deJaneiro Santos Paranagu~ Florien6polis Porto Alegre Valparaiso Maldonado Las Toninas, Note: Please refer to the note on Figure 5 for an explanation of this map. © TeleGeogrephy, inc. 2000 © TeleGeograpl~y, Inc. 2000 TeleGeography 2001 Figure 9. Map of Major Submarine Cables in East Asia Tokyo Stations Kitaibaraki ~~ dapan-U.S. Ajigaura EAC PCol Maruyama Australia-Japa n Chikura o China-U.S. ,~ APCN-2 Tokyo North Asian Cable C2C Miura o -- FLAG Europe-Asia Ninomiya TPC 5 Pgsan Tokyo Stations Chon Shan PC-1 China-U,S. Hong Kong Stations Macau Tai ShantouT°U 0kinawa Japan-U.S, Oanang latangas Guam-Philippines Tumon B Ancol Note: Please refer to the note on Figure 5 for an explanation of this mep. © TeleGeography, Inc. 2000 TeleGeog~aphy 2001 © Tele(~eographyo Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 International Circuit Usage by U.S. Carriers Each year the U.S. Federal Communications Commission (FCC) releases aggregate circuit usage statistics based on reports filed by the three largest U.S. facilities-based card- ers (AT&T, WorldCom, and Sprint). Although the rapid entry of new carders reduces the relative representation of the top three carriers each year, the statistics are still useful for baseline comparisons along two axes. First, the data illumi- nate year-to-year growth trends in overall cable and satel- lite connectivity. Second, the statistics break down how much capacity is used for public switched telephone network (PSTN) traffic, international private lines (IPLs), as well as how much capacity is reported "idle" each year. Although private lines can carry voice traffic, the circuit usage statistics provide a rough proxy to determine the bal- ance of voice and data traffic on international networks con- necting to the U.S. Assuming that increased IPL circuit deployment represents increased data traffic flows, the voice/data "crossover"--the point at which data traffic exceeded voice---occurred sometime in 1998. Over the past five years, the PSTN’s share of used capacity dropped from 83 to 36 percent. If the trend continues, public tele- phone lines may contribute only 15 percent of used capac- ity by 2005. Figure 1. International Circuit Usage Summary, 1995-1999 Absolute Circuit Usage Relative Circuit Usage 64 Kbps Circuits 960,000 800,000 100% 90% 700,000 800,000 500,000 300,000 200,000 I0[],000 1995 199G 1997 1998 1999 1999 1996 1997 1998 1999 N ~S-rN ¯ I~ [] ~ale 64 Kbps Circuit Usage IPL PSTN Idle Total Available 1995 28,497 128,150 118,343 270,990 1996 91,362 140,518 7~762 306,642 1991 147,408 170,717 123,751 441,876 1998 198,369 177,049 241,052 616,470 1999 375,503 212,243 252,232 839,978 Source: FCC © TeleGeography, Inc. 2000 TeleGeography 2001 © TeleGeographg, Inc. 2000 Figure 2. International Circuit Usage for Selected Routes, 1997-1999 U.S. Carrier 64 Kbps Circuit Usage For Private For Public Switched Total Circuits Idle Total Lines Network In Use Circuits Available Can ada 1997 37,383 50,343 87,726 32,178 119,904 1998 53,302 54,719 108,021 120,961 228,982 1999 103,546 72,999 176,545 108,901 285,446 M exi c o 1997 19,155 38,935 58,090 1,148 57,238 1998 24,463 38,301 62,764 4,080 66,844 1999 41,523 50,259 91,782 7,414 99,196 Hong Kong 1997 3,058 1,221 4,279 1,825 6,104 1998 4,685 1,027 5,712 3,623 9,335 1999 6,218 924 7,142 4,199 11,341 Japan 1997 10,087 6,149 16,236 17,178 33,414 1998 11,907 6,098 18,005 26,042 44,047 1999 21,998 6,401 28,399 28,120 56,519 Singapore 1997 1,617 570 2,187 571 2,758 1998 1,959 608 2,567 1,999 4,566 1999 3,412 638 4,050 2,473 6,523 U.K. 1997 23,008 14,662 37,670 20,118 57,788 1998 47,410 11,818 59,228 27,671 86,899 1999 101,298 13,695 114,993 34,100 149,093 Note: Data based on year-end FCC circ uit status reports filed by AT&T, WorldCom, and Sprint, for circuits originating in the continental U.S. as well as Puerto Rico, Guam, and other U.S. territories. "Idle" circuits are owned by a carrier at year end but not in use. The FCC estimates that 25-30 per- cent of total submarine cable capa city landed in the U.S. is controlled by foreign carriers and thus not reported here. Also, up to 100 percent of used capacity goes unrepo, rted because it is reserved for restoration purposes only. Canada Mexico Hong Kong Japan Singapore 1997 1998 1999 1997 1998 I 1999 1997 1998 1999 1997 1998 1999 1997 1998 1999 1897 1898 1999 I 0% lO% Circuit Usage and Idle Capacity, 1997-1999 40% 50% 60% 70% 100% 20% 30% 80% 90% Source: FCC © TeteGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleOeography 2001 Figure 3. International Circuit Usage by Region, 1997-1999 N. and C. America U.S. Carrier 64 Kbps Circuit Usage For Private For Public Switched Total Circuits Idle Total Lines Network In Use Circuits Available 1997 64,230 88,989 153,219 34,504 187,723 1998 78,601 94,952 173,553 126,197 299,750 1999 144,257 125,328 269,585 116,683 386,268 South America 1997 5,928 6,900 12,828 3,986 16,814 1998 7,958 7,716 15,674 5,536 21,210 1999 10,823 7,953 18,776 6,752 25,528 Caribbean 1997 1,034 6,478 7,512 4,006 11,518 1998 1,439 7,028 8,465 1,977 10,442 1999 1,622 7,063 8,685 4,223 12,908 W. Europe 1997 43,784 34,476 78,260 46,245 124,505 1998 69,051 34,133 103,184 52,937 156,121 1999 153,806 3~777 192,583 6~386 256,969 E. Europe 1997 1,326 3,742 5,068 1,719 6,787 1998 1,004 4,418 5,422 1,231 6,653 1999 1,161 4,791 5,952 1,298 7,250 Middle East 1997 1,432 3,096 4,528 479 5,007 1998 1,920 2,807 4,727 844 5,571 1999 2,808 2,934 5,742 1,085 6,827 Africa 1997 699 2,608 ~307 292 3,599 1998 1,080 2,712 3,792 320 4,112 1999 1,036 2,630 3,666 917 4,583 Asia 1997 23,545 19,567 43,112 30,830 73,942 1998 30,563 19,262 49,825 45,915 95,740 1999 48,513 19,932 68,445 50,170 118,615 Oceania Totals 1997 5,430 ~861 1~291 1,690 11,981 1998 6,753 4,023 10,776 6,095 16,871 1999 11,477 2,835 14,312 6,718 21,030 1997 147,408 170,717 318,125 123,751 441,876 1998 19~369 177,049 375,418 241,052 616,470 1999 375,503 212,243 587,746 252,232 839,978 Note: Data based on year-end FCC circuit status reports filed by AT&T, WorldCom, and Sprint, for circuits originating in the continental U.S. as well as Puerto Rico, Guam, and other U,S.territories. "Idle" circuits are owned by a carrier atyear end but notin use. The FCC estimates thst 25-30 per- cent of total submarine cable capacity landed in the U.S, is controlled by foreign carriers and thus not reported here. Also, up to 100 percent of used capacity goes unreported because it is reserved for restoration purposes only. Source: FCC © TeleGeography, Inc. 2000 TeleGeography 2001 © TeleGeography, Inc. 2000 Figure 4. Idle Circuits of U.S. Carriers by Region, 1997-1999 Circuit Usage and Idle Capacity, 1997-1999 1997 N.&C America 1998 1999 1997 South America 1998 1999 1997 Car=bbean 1998 1999 1997 W. Europe 1998 1999 1997 E. Europe 19~8 1999 1997 Middle East 1998 1999 1997 Africa 1998 1999 1997 As=a 1998 1999 1997 0ceama 1998 1999 0% ¯ IPLs [] PSTN E3ldle 100% 10% 20% 30% 40% 50% 60% 70% 80% 90% Note: Data based on year-end FCC circuit status reports filed by AT&T, WoddCom, and Sprint, for circuits originating in the continental U.S. as well as Puerto Rico, Guam, and other U.S. territories. "idle" circuits are owned by a carrier atyear end but not in use. The FCC estimates that 2530 per- cent oftotal submarine cable capacity landed in the U.S. is controlled by foreign carders and thus not reported here. Also, up to 100 percent of used capac~ goes unreported because it is reserved for restoration purposes only. Source: FCC © TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 International Communications Satellites Early Bird entered service in 1965 as the first geostationary satellite launched to carry intercontinental telephone traffic. By the following decade satellites became the dominant means of providing all kinds of international telecommunications. That dominance only began to fade--and slowly at first--when the first of the new generation of fiber optic submarine cables was deployed in the late 1980s. By 1994, trans-Atlantic fiber was carrying more voice traffic than trans-Atlantic satellites. It is clear that, on very busy point-to-point routes, fiber-optic submarine cables are superior to satellites. Although the qual- ity of voice conversations is appreciably better on fiber, the pri- mary reason for the decline of satellites’ central role is price. The cost of submarine cable bandwidth on some trans-Atlantic routes is one-tenth that of comparable satellite capacity. And the gulf between cable and satellite bandwidth pricing is bound to widen in the coming years. Whereas most fiber bandwidth suppliers forec~lst price declines of 25 to 50 percent per year for the next three years, satellite bandwidth suppliers antici- pate declines of no more than five to ten percent per year. Recently, the explosion of fiber optic capacity has even eroded satellites’ traditional role in submarine cable restoration. In previous years, carriers could lease capacity on satellites as an emerBency backup when cable faults occurred. Yet submarine cable bandwidth now far exceeds total satellite capacity (see Figure I. The Big Picture). Indeed, a single cable system, such as the 2.4 Tbps FLAG Atlantic- I, will soon dwarf all the world’s satellite capacity combined--by at least one order of ma~i- tude. Suppliers of these state of the art cable systems must rely on network ring architectures or other cables for restora- tion; satellites no longer present a viable option, The technological and competitive revolution transforming long- haul networks is also leading a strategic business realignment among satellite owners. For example, when the world’s largest satellite operator, ]ntelsat, spun off five of its satellites in a 1998 pdvate venture, their primary tarBet was the video ser- vices market. Most of New Skies’ capacity is dedicated to video transmission for broadcasters. Likewise, Intelsat’s first privately-backed competition, PanAmSat, collects more than three quarters of its revenue from video. Although Intelsat’s business is still dominated by traditional telephone trunldng services, much of its new commercial activity focuses on video services and lntemet services. Intercontinental Submarine and Satellite Capacity Submarine Cable Capacity Satellite Capac~ 100 1989 1991 1993 1995 1997 1999 2001 leGeo iphy:researCh and Euroconsult : :: Countries Connected by Multi-Gigabit Submarine Cable Systems 100% 90% 80% 70% 30% 0% Pre- 199~ 1994 1995 1996 1997 1998 1999 2000 2001 2002 TeleGeo{~raphy 200! © TeleGeography, Inc. 2000 Figure 2. Interregional Satellite Capacity, 1989-2001 Interregional Satellite Capacity Growth 180 160 Africa & Middle East U.S. & Canada 0 1980 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 140 120 100 8O 6O 40 20 Note; Satellite capacity used for tnterregional voice and data services accounts for 12 percent of capacity shown above. Capacity figures assume that a pair of 36 MHz equivalent transponders will yield approximately 40 Mbps of transmission c apacity. Source: Euroconsult Design © TeleGeography, Inc. 2000 Although the prevalent model in the geostationary satellite business has shifted considerably over the last decade, satel- lites still matter in the international bandwidth regime. First, satellites are crucial on routes under-served by fiber optic infra- structure. This is important to telecom companies, ISPs, and corporate networkers seeking connectivity to this surprisingly large segment of the world’s population (see Figure 1). Second, unlike fiber, satellite capacity is readily available in asymmetric configurations. This application is ideal for ISPs connecting to lnternet hubs, where traffic is largely going in one direction--from content-rich web sites to downstream end- users. Finally, due to this same asymmetry, satellites are actu- ally better suited for some applications. While this is clear in the case of one-way television broadcast services, the same logic applies to certain lnternet applications, including the "rnulticast" of newsfeeds and caches of recently accessed web content. ~i~ Adapted from "International Bandwidth 2000, TeleGeography’s guide to supply and demand of interna- tional telecommunications capacity. © TeleGeograpl~yo Inc. 2000 TeleGeography 2001 ¯ O ~1 O 0 O O TeleGeography 2001 © TeleGeography, Inc, 2000 Orbital Slot (°E) Launch Date C-Band Ku-Band Bus Atlantic Ocean GE Americom G E-I E 5.0° November 1997 Eutelsat W3 7.0° May 1999 Eutelsat W1 10.0° 3Q 2000 Eutelsat HOT BIRD 1 13.0° March 1995 Eutelsat HOT BIRD 2 13.0° November 1996 Eutelsat HOT BIRD 3 13.0° September 1997 Eutelsat HOT BIRD 4 13.0° February 1998 Eutelsat HOT BIRD 5 13.0° October 1998 Eutelsat W2 16.0° November 1998 Eutelsat II-F4 21,5° July 1999 Eutelsat I-F4 25.5° September 1987 PanAmSat PAS-5 302.0° August 1997 PanAmSat PAS-9 302.0° July 2000 Intelsat IS-805 304.5° June 1998 Intelsat IS-708 307.0° May 1995 Intelsat IS-709 310.0° June 1996 BE Americom BE-li 313.0° 2003 GE Americom TDRSo6 313.0° January 1993 PanAmSat PAS-1 315.0° June 1988 PanAmSat PAS-1R 315.0° November 2000 PanAmSat PAS-3 317.0° January 1996 PanAmSat PAS-6 317.0° August 1997 PanAmSat PAS-6B 317.0° December 1998 New Skies Satellites NSS 806 319.5° February 1998 GE Americom 515 322.3° January 1989 Intelsat IS-601 325.50 October 1991 I ntelsat IS-904 325.5° 3Q 2001 Intelsat IS-801 328.5° February 1997 Intelsat IS-907 328.5° TBD Intelsat IS-511 330.5° June 1985 Intelsat IS-605 332.5° August 1991 Intelsat IS-905 332.5° 4Q 2001 Intelsat IS-603 335.5° March 1990 Intelsat 1S-903 335.5° 2Q 2001 GE Americom GE-2E 336.0° 4Q 2002 New Skies Satellites NSS 7 338.5° 4Q 2001 New Skies Satellites NSS 803 338.5° September 1997 New Skies Satellites NSS K 338,5° June 1992 Intelsat IS-705 342.0° March 1995 Intelsat IS-906 342.0° TBD Loral Skynet Telstar 12 345.0° October 1999 Eutelsat Atlantic Bird 1 347.5° 2Q 2001 Eutelsat II-F2 347,5° January 1991 Intelsat IS-707 359.0° March 1996 24,0 24,0 360 42,0 42.0 72.0 12,0 24.0 36.0 25.1 36.0 27,0 64,0 76.0 64.0 n.a. 42.0 64.0 76,0 64,0 76.0 58.0 61.0 42.0 42.0 24.0 n.a. 48.0 Spacebus 480 Spacebus 16.0 Spacebus 18.3 Eurostar 18.8 Eurostar 18.8 Eurostar 23.0 Eurostar 48.0 Spacebus 23.0 Spacebus 200 B. Aero 24.O HS-601 24.0 HS-601 6.0 AS-7000 28.0 FS-1300 2O,O FS-1300 TRW 12.0 GE 3000 36.0 HS-601 25,1 HS-601 36O FSq300 32.0 H S-601 6.O LM 7000 Ford Aero 24.0 HS-393 20.0 Loral SS 12.0 AS-7000 n.a n.a. 12.0 E Aero 24.0 H S-393 20.0 n.a. 24.0 HS-393 20.0 Loral SS n.a 47.0 Loral SS 12.0 LM 7000 21,0 LMA2100 20.0 FS-1300 n.a. n.a. 45.0 Eurostar 36.0 n.a. 23.0 Spacebus 28.0 FS-1300 Note: Included satellites provide international coverage and belong to a constellation with at least 15 transponders dedicated to international voice and data services. Source: TeleGeography research and company reports © TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Figure 4. Major International Telecommunications Satellites in Geostationary Orbit (continued) Transponders (36 Mhz) Orbital Slot(°E) Launch Date C-Band Ku-Band Bus Indian Ocean Eutelsat II-F3 36.0° December 1991 Eutelsat SESAT 36.0° April 2000 Eutelsat W4 36.0° May 2000 Loral Skynet Telstar 11 37.5° November 1994 Eutelsat II-F1 48.0° August 1990 New Skies Satellites NSS 703 57.0° October 1994 40 Intelsat IS-604 60.0° June 1990 64 Intelsat IS-902 60.00 2Q 2001 76 Intelsat IS-602 62.00 October 1989 64 Intelsat IS-901 62.0° 4Q 2000 76 Intelsat IS-804 64.00 December 1997 64 Intelsat IS-704 66.0° January 1995 42 PanAmSat PAS-10 68.5° 1Q 200; 12 PanAmSat PAS-4 68.5° August 1995 25.1 PanAmSat PAS-7 68,5° September 1998 14 Loral Skynet Telstar 10/APSTAR-IIR 76.5° October 1997 27.8 Intelsat APR-1 83.0° June 1999 11 GE Americom 6E-1A 108.0° 3Q 2000 Pacific Ocean APSTAR APSTAR-IA 134,0° 1994 28 APSTAR APSTAR-I 138,0° July 1996 28 PanAmSat PAS-8 166.0° November 1998 24 PanAmSat PAS-2 169.0° July 1994 251 G E Americom 6 E-2i 172.0° 2003 60 Intelsat IS-802 174.00 June 1997 64 Intelsat IS-702 177.0° June 1994 42 Intelsat IS-701 180.0° October 1993 42 New Skies Satellites NSS 513 183.0° May 1988 42 6E Americom TDRS-5 185.7° August 1991 12 Loral Skynet Telstar 13 239.0° June 2002 n.a. Satmex Satmex 5 243.2° December 1998 24 Satmex Solidaridad 2 247.0° October 1994 24 GE Americom GE-4 259,0° November 1999 24 PanAmSat Galaxy Villi 265.0° December 1997 {3E Americom GE-Spacenet 4 279.0° April 1991 24 G E Americom G E-6 288.0° 4Q 2000 24 23 Spacebus 18 n.a. 29.3 Spacebus 48 Eurostar 23 Spacebus 20 FS-1300 24 H S-393 20 Loral SS 24 HS-393 20 Loral SS 12 AS-7000 20 FS-1300 12 HS 601 24.6 HS-601 30 FS-1300 23.5 FS-1300 n.a. 28 n.a, HS-376 HS-376 24 FSo1300 25.1 HS-601 12 AS °7000 20 FS-1300 20 FS-1300 16 Loral SS TRW n,a. Echostar 24 HS 601 24 HS 601 32 A2100 21.3 HS 601 32 A2100 Note: Satellites included provide international coverage and belong to a constellation with at least 15 transponders dedicated to international voice and data services, Source: TeleGeography research and company reports © TeleGeography, Inc. 2000 TeleOeo~aphy 2001 © TeleCeography, Inc. 2000 Figure 5. Map of Major Communications Satellites in the Atlantic Ocean Region ¯ Six36-MHztransponder equivalents for the C-band (4/6 GHz) ¯ Six36-MHztransponder equivalents for the Ku-band (11-12/14 GHz) WI~’ Outline denotes satellites under construction as of September 2000 Note: Satellites included provide international coverage and belong to a constellation with at least 15 transponders dedicated to international voice and data services, Figure 6. Map of Major Communications Satellites in the indian Ocean Region Six 36-M Hz transponder equivalents for the C-band (4/6 G Hz) ¯ Six36-MHztransponder equivalents for the Ku-band (11-12/14 GHz) [] ~’ Outline denotes satellites under construction as of September 2000 Note: Satellites included provide international coverage and belong to a constellation with at least 15transponders dedicated to international voice and data services, © TeleGeograpl~lo Inc. 2000 TeleGeography 2001 Figure 7. Map of Major Communications Satellites in the Pacific Ocean Region (~ TeleGeog~i~lnc. 2000 ¯ Six36-MHz transponder equivalents for the C-band (4/6 GHz) ¯ Six 36-MHz transponder equivalents for the Ku-band (11-12/14 GHz) [] V Outline denotes satellites under construction as of September 2000 Note: Satellites included provide international coverage and belong to a constellation with at least 15 transponders dedicated to international voice and data services. TeleGeography 200t © TeleGeography, Inc. 2000 lO0 lnternet Backbones TeleOeography 2001 © TeleGeography, Inc. 2000 International lnternet Backbones Internet Backbones What is an Internet backbone? And when is it internation- al? The questions are not as straightforward as they might seem. International lnternet backbones are private data links which cross international political borders, run the Internet Protocol (IP), are reachable from other parts of the Internet, and carry general Internet traffic: e-mail, Web pages, and most of the other popular services which have come to define today’s Internet. That means that international IP links devoted to just one type of traffic--notably, Voice-over-IP (VolP)--are excluded from our definition of backbones on the public Internet. If VolP is excluded, though, then why publish international lnternet backbone data in a book on international telepho- ny? The answer: because it just might be important. Despite a history stretching back more than 30 years, today’s Internet really began its push toward ubiquity dur- ing the 1990s in a rapid transition from academic network to commercial networks. What evolved was a decentralized infrastructure whose end-to-end design made it possible for users to create new network applications without asking too many people’s permission. The resulting infrastructure took media services based on text and simple graphics and turned them into the most widespread media platform since television. That ubiquity only fueled its popularity, however, and soon people were stuffing two-way voice telephony, streaming video, and other bandwidth-intensive applications into the public Intemet. They did this not because the lnternet’s then- infrastructure was particularly well-suited to such services, but because running them over the Internet meant bringing together multiple services on a single platform. On net- works, the whole is always more than the sum of its parts. The so-called "public Internet" is at a crossroads. How will it accommodate very different types of traffic inside the same networks? Some want to solve the problem by bestowing Quality of Service (QoS) provisions upon IP so that networks can distinguish between what needs to be delivered immediately and what needs to be delivered with care. Some, pressed for time, prefer to forego fancy traffic engineering by throwing more bandwidth at the problem, hoping to give every packet enough room to get to its des- tination in style. And some are abandoning the public Internet altogether: distinct backbones are emerging for 102 self-similar traffic generators, like VolP or the Usenet’s text- and photo-oriented discussion groups. Those highly spe- cialized, single-service backbones are not included here. That sharpens the scope of what we mean by "Internet" backbones. But it doesn’t close off their possibilities. There is increasing excitement over a "new public network" infra- structure which meshes PSTN (public switched telephone network) and IP infrastructure into the backbone of tomor- row’s public communications facilities. If the feverish activ- ity taking place around the world can successfully achieve the economies of scale and creative possibilities that inter- operable communications services represent, these back- bones will have to come together to look like the lnternet as many engineers have always drawn it--a cloud. Bandwidth, Not Traffic The maps and statistics on the following pages show inter- national Internet backbone capacity, or bandwidth--not traffic. There are several reasons to keep track of interna- tional Internet bandwidth. One is to provide a rough met- ric for matching supply and demand. But there is another reason: bandwidth take-up may provide a clue to Internet traffic statistics, which are still in very short supply. International Internet bandwidth is growing faster than international Internet traffic, however. In the past few years, tremendous physical infrastructure builds began to come on-line. Because raw bandwidth does not translate imme- diately into Internet capacity, however--it must first be lit, sold, deployed, and integrated into data network opera- tions--the numbers showed what, to casual observers, appeared to be a mismatch between physical capacity and Internet capacity. In 2000, however, bandwidth started dif- fusing up the network layers, moving from physical installa- tion to actually existing network services. As Internet capacity bega5 to take advantage of the fiber explosion, two-and-a-half gigabit per second OC-48 (STM-16) links running Intemet Protocol became, if not common, at least widespread. All this new Internet capacity makes network bandwidth less useful as a proxy for traffic. But it does provide important insights as to how traffic is routed. Historically, steep intraregional bandwidth costs, a comparative lack of local content, and limited regional coordination had caused the U.S. to become the Internet’s central switching office, even for data flows within a region. Last year, we found that the © TeleGeography, Inc. 2000 TeleGeography 2001 Figure 1. Interregional internet Bandwidth,2000 357.1 Mbps U .S ./ Canada Asia/ Pacific Europe ,/ International Region I Region 2 Intarnet Bandwidth Africa Africa 0ntraregional) 9.0 Mbps Asia/Pacific 3.8 Mbps Europe 171.5 Mbps Latin America/Caribbean n.a. U.S./Canada 467.6 Mbps Region total 661.9 Mbps Asi~z/Pac~fic Africa 3.8 Mbps Asia/Pacific (intraregional) 3,124.0 Mbps Europe 357.1 Mbps Latin America/Caribbean n.a. U.S./Canada 19,716.5 Mbps Region total 23,201,4 Mbps Europe Africa 171,5 Mbps Asia/Pacific 357,1 Mbps Europe (~ntrareglonal) 176,594,9 Mbps Latin America/Caribbean 127.0 Mbps U.S./Canada 56,241.2 Mbps Region total 233,491.7 Mbps Latin America/Caribbean Africa n.a. Asia/Pacific n,a. Europe 127.0 Mbps Latin America/Caribbean (~ntraregional) 71.0 Mbps U.S./Canada 2,637.6 Mbpa Region total 2,835.6 Mbps U.S./Canada Africa 467.6 Mbps Asia/Pacific 19,716.5 Mbps Europe ’ 56,241.2 Mbps Latin America/Caribbean 2,837.6 Mbps U.S./Canada (intraregional) 33,973.5 Mbps Region total 113,036.3 Mbps Note: Figures represent estimated |ntemet bandwidth between Consolidated Metropolitan Statistical Areas or equivalents. Domestic backbone routes are omitted. Data as of mid-200O, Source: TeleGeegraphy research © TeleGeography, Inc. 2000 103 TeleGeography 2001 © TeleGeographyo Inc. 2000 Internet was still U.S.-centric but that places like Western Europe and, to a lesser extent, eastern Asia were beginning to develop as secondary hubs (see Figure 1. lnterregional Internet Bandwidth, 2000). In 2000, this trend continued. In Europe, international capacity between European countries again grew at a noticeably faster clip than did outgoing international band- width, nudging the continent’s in-region connectivity to beyond 75 percent of its total international lnternet band- width. Asia’s intraregional connectivity, too, grew more than twice as fast as to anywhere else, doubling the proportion remaining within the region to one-sixth of total internation- al Internet bandwidth. Enormous differences still remain from country to country. The U.S.-centric pattern wanes only with substantial and sustained infrastructure builds of the sort that has swept Europe, rolled into Asia, announced itself in Latin America, and stalled in most of Africa. The connectivity divide is reflected in the larger pattern of global net bandwidth (see Figure 2. A Question of Scale). Although fiber builds in Latin America and the Caribbean increased substantially in the last year, the Internet had not yet caught up as of mid- 2000, and the contin’ued reliance on slower, more expensive satellite links for international connectivity translated into another year of the World Wide Wait. In continental Africa, similarly, the number of countries connected at above 10 Mbps went from three to six--but Africa-U.S. Internet band- width remained the continent’s fastest-growing route. The Players Approximately 300 International Internet Service Providers (liSPs) own, lease, or otherwise get hold of transborder net- work capacity; place routing computers at either end; and use these segments to cobble together logical networks that, together, form the Internet’s international backbone. Three hundred may seem like a lot. Not all backbones are equal, however: in mid-2000 the ten largest liSPs controlled three-quarters of international lnternet bandwidth. Some observers try to make sense of the lnternet’s snarl of networks by dividing them into three or four tiers. Under that framework, "Tier Ones" are the handful of global back- bone operators who have rich interconnection relationships with all other significant providers; "Tier Twos" are the not- quite-Tier-One backbones who end up having to pay for some of their direct backbone connectivity; and "Tier Threes" and "Fours" are the national/regional and local ISPs, depending on the context and topology, in question. Those definitions are somewhat fuzzy--and for good reason. As a whole, the Internet service provider world is not seg- mented into hierarchical divisions, so hard-and-fast typolo- gies just aren’t possible on a global scale. The same is true, 104 perhaps more so, of the liSP segment. Instead, we have identified four groupings around which liSPs cluster. However imperfect, these markings on the lISP spectrum help understand which way they lean: Global liSP. Two kinds of liSPs engage in activities which place them in the "global" range. One is a set of very large players who have strong historical roots in the U.S. lnternet, either in origin (AT&T/Concert, Genuity, WorldCom, PSINet, Sprint) or by osmosis; Cable & Wireless acquired much of MCI’s Internet backbone as a result of the MCI-WorldCom merger. The.other group of liSPs are providers who feature managed IP bandwidth over bent-pipe (point-to-point) satellite as important parts of their services portfolios, typically Intelsat signa- tories like Telecom Italia, whose Seabone offering con- nects many countries around the world, or service providers like Interpacket. The effort to move from the second set of global liSPs into the first--Teleglobe attempted this during the late 1990s--is a key dynamic within this grouping. Regional IISR A regional liSP specializes in operating backbone connectivity between different countries in a single region, like GTS E-Bone in Europe, Pacific Century CyberWorks or Telstra in Asia, and Africa Online. Because of the impressive build-out in Europe during the past two years, Western Europe is probably the best example of the impact that regional liSPs can have on reconfiguring a sin’gle region’s topology map. At the other end of the spectrum, a number of liSPs continue to vault into the regional area by purchasing existing small- er, nationally-based lISP networks; once, these were dominated by former incumbents, but an increasing number of new entrants have borrowed this strategy as well, fuelled by the international spread of venture capi- tal and Initial Public Offerings. National lISP. Typically, this is an Internet provider which has acquired international connectivity as part of a national or local service; which acts increasingly as an upstream provider for other providers who have little or no international connectivity; and which moves to expand into neighboring countries. In 2000, this sector actually shrunk, as existing players federated or were bought up to form regional liSPs. Academic liSP. Research networks, including those operated by academic institutions, often act as interna- tional connectivity providers alongside commercial liSPs. In many environments, they operate high-capacity, lead- ing-edge systems, catalyzing Internet development-- examples are Europe’s DANTE TEN-155 and G~dkrr pro- jects--but they are increasingly specializing in exclusive- ly academic and research traffic as part of the interna- © TeleGeography, Inc. 2000 TeleGeography 2001 Figure 2. A Question of Scale Top Routes Compared Europe London to New York U.S.]Canada New York to London Asia/Pacific Tokyo to San Franczsco Latin America/Caribbean S~o Paulo to New York 444.1 Mbps Africa 124.0 M bps o Johannesburg to New York One Route Scaled Europe 310 Mbps (Scale 1 : I) U.S./Canada 310 Mbps (Scale 1 : 0.58) Asia/Pacific 310 Mbps (Scale 1 : 0 23) Latin America/Caribbean Africa How much does lnternet backbone capacity vary from region to region? A IOt. As the figure on the left illustrates, the largest single international lnternet route connected to Africa is but a fraction of its European counterpart. That meant that our regional lnternet maps couldn’t be drawn without varying the scale we used to represent bandwidth from map to map. To get an idea of just how much the scale varied, look to the right-hand figure above. The line that is so thin on top and so thick on the bottom represents ,510 Mbps, the equivalent of two OC-5 links, as it is scaled on each of the cantinental maps. Of course, some routes were too small to be shown on the maps altogether---even 310 Mbps stretches thin in the bandwidth-rich Source: TeleGeography research regions of Europe and U.S. & Canada. Other routes had to be omitted, not because they did not meet the inclusion criteria, but in order to improve map visibility or because the routes were located outside of the base map projection. For each region, routes are ordered by backbone capacity and then ]isled alphabetically, where both cities are in the region, or with the local city first, when they aren’t. That doesn’t imply any hierarchy. But it does mean some routes dropped off because they were too far down in the alphabet, even if they had the same capacity as the fiftieth-largest route listed.. Only so much bandwidth can fit on the pdnted page. © TeleGeography, Inc. 2000 tional coordination of Internet2 and advanced research applications. Methodology The data depicted on the following pages--using different scales for different regions--result from a TeleGeography, Inc. study completed in October 2000. The research focused on the network topologies of over 300 liSPs oper- ating international Internet links--routers or switches direct- ly connected across an international border over an internal network. These links and their capacities were then tracked through over 300 cities in more than 180 countries. Each IISP’s network routes and capacities were derived from a combination of public documents, confidential interviews, and computer-based network analysis tools. The study grouped specific switch and router locations according to Consolidated Metropolitan Statistical Area, Census Metropolitan Area, or the equivalent. Only the IP network was mapped, instead of the physical network infra- structure which runs beneath it. In cases where liSPs had provisioned relatively new dedicated IP capacity, the study did not include the capacity unless it was believed to be operational and available for public Internet traffic as of mid-2000 (i.e., bandwidth kept in reserve was excluded). A final note: due to the complex and ever-changing nature of network architectures, omissions may have occurred. 105 TeleGeography 2OO1 © TeleGeographg, Inc. 2000 Figure 3. The Top 50 International Internet Routes, 2000 Rank City, Country City, Country 1 London, U.K. New York, U.S. 26,680.5 Mb 2 London, U.K. Paris, France 24,340,0 Mb 3 Frankfurt, Germany Paris, France 14,148.0 Mb 4 Amsterdam, Netherlands London, U.K. 12,347.0 Mb 5 Amsterdam, Netherlands Frankfurt, Germany 10,510.0 Mb 6 Amsterdam, Netherlands NewYork, U.S. 9,958.0 7 Montreal, Canada New York, U.S. 8,749.0 8 Amsterdam, Netherlands Brussels, Belgium 8,368.0 9 Amsterdam, Netherlands DLisseldorf, Germany 7,999.0 10 Geneva, Switzerland Paris, France 7,778.5 11 San Francisco, U.S. To,o, Japan 7,550.0 12 Chicago, U.S. Toronto, Canada 6,575.0 13 Brussels, Belgium London, U.K. 6,204.0 14 Frankfurt, Germany London, U.K. 4,975.0 15 NewYork, U.S. Toronto, Canada 4,862.0 16 Seattle, U.S. Vancouver, Canada 4,707.0 17 Frankfurt, Germany Milan, Italy 4,305.0 18 Frankfurt, Germany Stockholm, Sweden 3,897.0 19 Copenhagen, Denmark Stockholm, Sweden 3,848.0 20 Amsterdam, Netherlands Paris, France 3,820.0 21 Frankfurt, Germany Washington, U.S. 3,663.0 22 Brussels, Belgium Paris, France 3,657.0 23 Amsterdam, Netherlands Stockholm, Sweden 3,414.0 24 Madrid, Spain Paris, France 3,212.0 25 London, U.K. Madrid, Spain 2,880.0 26 Frankfurt, Germany Geneva, Switzerland 2,877.0 27 Amsterdam, Netherlands Geneva, Switzerland 2,833.5 28 Frankfurt, Germany New York, U.S. 2,729.0 29 Barcelona, Spain Lyon, France 2,688.0 30 Amsterdam, Netherlands Madrid, Spain 2,612.0 31 Geneva, Switzerland Lyon, France 2,567.0 32 Amsterdam, Netherlands Vienna, Austria 2,533.0 33 Moscow, Russia Stockholm, Sweden 2,522.0 34 Amsterdam, Netherlands Oslo, Norway 2,488.0 Chicago, U.S. Montreal, Canada 2,488.0 Copenhagen, Denmark Diisseldorf, Germany 2,488.0 D[Jsseldorf, Germany London, U.K. 2,488.0 DLisseldorf, Germany Prague, Czech Republic 2,488.0 Paris, France Washington, U.S. 2,488.0 Prague, Czech Republic Vienna, Austria 2,488.0" Seattle, U.S. Toronto, Canada 2,488.0 42 London, U.K. Washington, U.S. 2,378.0 43 Toronto, Canada Washington, U.S. 2,176.0 44 Milan, Italy NewYork, U.S. 1,734.0 45 Frankfurt, Germany Vienna, Austria 1,676.0 46 NewYork, U.S. Stockholm, Sweden 1,553.0 47 Los Angeles, U.S. Tokyo, Japan 1,520.5 48 San Francisco, U.S. Seoul, Korea, Rep. 1,336.6 49 Milan, Italy Paris, France 1,242.0 50 Hong Kong, China San Francisco, U.S. 1,209.5 Internet Bandwidth Mb~s Mb ~s Mb~s Mb~s Mbps Mb~s Mb~s Mb~s Mb~s Mb~s Mb~s Mb~s Mb ~s Mb~s Mb ~s Mb ~s Mb ~s Mb~s Mb~s Mb ~s Mb~s Mb~s Mb~s Mb~s Mb ~s Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Note: Rgures represent estimated Internet bandwidth between Consolidated Metropolitan Statistical Areas or equivalents. Domestic backbone routes are omitted, Data as of mid*2000. Source: TeleGeography research © TeleGeography, Inc, 2000 106 © TeleGeographuo Inc. 2000 TeleGeography 2001 Figure 4. The Top 50 International Internet Hub Cities, 2000 Rank City Country 1 London U.K. 66,5893 Mb 2 Amsterdam Netherlands 68,301.6 Mb 3 Paris France 62,196.7 Mb 4 New York U.S. 61,070.6 Mb 5 Frankfurt Germany 52,332,0 Mb 6 Stockholm Sweden 18,652.5 7 Brussels Belgium 18,630.8 8 G eneva Switzeda nd 17,848.8 9 Toronto Canada 16,398.6 10 DLisseldorf Germany 15,863.4 11 San Francisco U.S. 14,712.5 12 Washington U.S. 12,029.7 13 Montreal Canada 11,671.8 14 Chicago U.S. 10,935.5 15 Tokyo Japan 10,835.1 16 Milan Italy 9,262.6 17 Madrid Spain 9,246.0 18 Vienna Austria 8,273.3 19 Seattle U.S. 7,658.3 20 Copenhagen Denmark 7,520.0 21 Prague Czech Republic 6,186.0 22 Vancouver Canada 5,547.0 23 Lyon France 5,410.0 24 Oslo Norway 3,928.0 25 Los Angeles U.S. 3,529.6 26 Z0rich Switzerland 3,198.3 27 Moscow Russia 2,956.4 28 Hong Kong China 2,949.4 29 Barcelona Spain 2,873.0 30 Seoul Korea, Rep. 2,317.4 31 Sydney Australia 2,162.4 32 Helsinki Finland 1,841.0 33 Singapore Singapore 1,793.8 34 Taipei Taiwan 1,310.9 35 Budapest Hungary 1,019.5 36 Dublin Ireland 917.0 37 Munich Germany 827.0 38 Osaka Japan 820.0 39 Palermo Italy 818.5 40 Calgary Canada 777.0 41 Warsaw Poland 787.0 42 Mexico City Mexico 749.5 43 Miami U.S. 632.7 44 Shanghai China 626.0 45 Auckland New Zealand 592.0 46 $8o Paulo Brazil 566.6 47 Athens Greece 560.1 48 Nova Scotia Canada 499.9 49 Lisbon Portugal 499.0 50 Stuttgart Germany 481.9 International Interact Bandwidth )S )s )s )s )$ Mb )s Mbps Mbps Mb)s Mb )s Mb )s Mb)s Mb)s Mb)s Mb ~s Mb~s Mb)s Mb)s Mb~s Mb )s Mbps Mbps Mb)s Mb)s Mb )s Mb )s Mb)s Mb)s Mb)s Mb)s Mb )s Mb)s Mb )s Mb)s Mb )s Mbps Mbps Mbps Mbps Mbps Mb )s Mb)s Mb)s Mb)s Mb )s Mb)s Mb)s Mb )s Mb)s Mb)s Note: Rgures represent estimated tnternet bandwidth between Conso|idated Metropolitan Statistical Ames or equivalents. Domestic backbone routes are omitted. Data as of mid-2000 Source: TeleGeography research © TeleGeography, Inc. 2000 107 TeleGeography 2001 © TeleGeography, Inc. 2000 Figure 5. Map of Major Asia/Pacific International Backbone Routes. 2000 Sacramento Seattle Chicago Aggregate lnternet backbone bandwidth 2 Gbps I Gbps 500 Mbps !00 Mbps Intraregional Interregional Note: Map includes international backbone routes with at least 100 Mbps of aggregate capacity. Figures represent estimated Internet bandwidth beWveen Consolidated Metropolitan Statistical Areas or equivalents. Domestic backbone routes are omitted. Data as of mid-2000, Source: TeleGeography research © TeleGeograph% Inc. 2000 108 © TeleGeographyo Inc. 2000 TeleGeography 2001 Figure 6. The Top 50 International Backbone Routes in Asia/Pacific, 2000 Rank City, Country 1 Tokyo, Japan 2 Tokyo, Japan 3 SeouI, Korea, Rep. 4 Hang Kong, China San Francisco, U.S. 5 Sydney, Australia San Francisco, U.S. 6 Singapore, Singapore San Francisco, U.S. 7 Taipei, Taiwan San Francisco, U.S. 8 Sydney, Australia Los Angeles, U.S. 9 0saka, Japan San Francisco, U.S. 10 Hang Kong, China Tokyo, Japan 11 Tokyo, Japan Seattle, U.S. 12 Seoul, South Korea, Rep. Los Angeles, U.S. 13 Hang Kong, China Shanghai, China 14 Auckland, NewZealand San Francisco, U.S. 15 Hang Kong, China Los Angeles, U.S. 16 Auckland, NewZealand Los Angeles, U.S. 17 Perth, Australia San Francisco, U.S. 18 Seoul, Korea, Rep. Sacramento, U.S. 19 Shanghai, China San Francisco, U.S. 20 Taipei, Taiwan Los Angeles, U.S. 21 Seoul, Korea, Rep. Tokyo, Japan 22 6uangzhou, China Hang Kong, China 23 Abu Dhabi, U.A.E. NewYork, U.S. Osaka, Japan New York, U.S. Osaka, Japan Sacramento, U.S. 26 Tel Aviv, Israel New York, U.S. 27 Sydney, Australia Tokyo, Japan 29 Auckland, New Zealand Sydney, Australia 29 Hang Kong, China Singapore, Singapore Shanghai, China Sydney, Australia 31 Hang Kong, China Taipei, Taiwan 32 Tokyo, Japan Chicago, U.S. Tokyo, Japan New York, U.S. 34 Hang Kong, China Seoul, Korea, Rep. 35 Manila, Philippines Singapore, Singapore 36 Taipei, Taiwan Tokyo, Japan 37 Manila, Philippines San Francisco, U.S. 36 Bangkol~Thailand San Francisco, U.S. 39 Perth, Australia Los Angeles, U.S. 40 Mumbai, lndia Singapore, Singapore 41 Mumbai, India New York, U.S. 42 Singapore, Singapore Sydney, Australia Singapore, Singapore Tokyo, Japan 44 Singapore, Singapore Los Angeles, U.S. 45 Haifa, Israel New York, U.S. 46 Manila, Philippines Tokyo, Japan 47 Kuala Lumpur, Malaysia Singapore, Singapore 48 Kuala Lumpur, Malaysia Tokyo, Japan 49 Beijing, China Tokyo, Japan 50 Manila, Philippines Los Angeles, U.S. City, Country Intemet Bandwidth San Francisco, U.S. 7,550.0 Mbps Los Angeles, U.S. 1,520.5 Mbps San Francisco, U.S. 1,336.6 Mbps 1,209.5 Mbps 1,030.0 Mbps 932.0 Mbps 741.5 Mb )s 496.0 Mb )s 465.0 Mb )s 403.5 Mb ~s 400.0 Mb )s 310.0 Mb )s 296.0 Mb )s 255.0 Mb ~s 245.0 Mb )s 200.0 Mb ~s 196.1 Mb )s 189.0 Mb )s 165.0 Mb )s 161.5 Mbps 157.3 Mb )s 156.0 Mb ~s 155.0 Mb ~s 155.0 Mb )s 155.0 Mb ~s 151.0 Mb ~s 145.0 Mb ~s 135.0 Mb ~s 118.0 Mb~s 118.0 Mb )s 103.0 Mb ~s 100.0 Mb )s 100.0 Mb )s 96.5 Mb ~s 90.4 Mb zs 87.1 Mb )s 79.0 Mb )s 77.0 Mb )s 75.0 Mb )s 70.0 Mb ~s 68.0 Mb ~s 61.0 Mb ~s 61.0 Mb ~s 59.0 Mb ~s 58.5 Mb ~s 57.0 Mb ~s 56.0 Mb ~s 55.1 Mb ~s 53.1 Mb ~s 53.0 Mb ~s Note: Rgures represent estimated Internet bandwidth between Consolidated Metropolitan Statistical Areas or equivalents. Domestic backbone routes are omitted, Data as of mid-2000. Source: TeleGeog ra phy resea rch © TeleGeography, Inc. 2000 109 TeleGeography 2001 © TeleGeography, Inc. 2000 Figure 7. Map of Major European International Backbone Routes, 2000 Aggregate Internet backbone bandwidth 20 Gbps 10 Gbps 5 Gbps 2 Gbps Interregiona! Note: Map includes international backbone routes with at feast 2 Gbps of aggregate capacity. Figures represent eCtimated Internet bandwidth betwee~ Consolidated Metropolitan Statistical Areas or equivalents. Domestic backbone routes are or~itt~d. Data as of mid-2000. Source: TeleGeography research © TeleGeography, Inc. 2000 110 © TeleGeograp#~to Inc. 2000 TeleGeography 2001 Figure 8. The Top 50 International Backbone Routes in Europe, 2000 City, Country Internet Bandwidth NewYork, U.S. 26,880.5 Mbps Rank City, Country 1 London, U.K. 2 London, U.K. Paris, France 24,340.0 3 Frankfurt, Germany Paris, France 14,148.0 4 Amsterdam, Netherlands London, U.K. 12,347.0 5 Amsterdam, Netherlands Frankfurt, Germany 10,510.0 6 Amsterdam, Netherlands New York, U.S. 9,958.0 7 Amsterdam, Netherlands Brussels, Belgium 8,368.0 8 Amsterdam, Netherlands DLisseldorf, Germany 7,999.0 9 Geneva, Switzerland Paris, France 7,778.5 10 Brussels, Belgium London, U.K. 8,204.0 11 Frankfurt, Germany London, IJ.K. 4,975.0 12 Frankfurt, Germany Milan, Italy 4,305.0 13 Frankfurt, Germany Stockholm, Sweden 3,897.0 14 Copenhagen, Denmark Stockholm, Sweden 3,848.0 15 Amsterdam, Netherlands Paris, France 3,820.0 16 Frankfurt, Germany Washington, U.S. 3,663.0 17 Brussels, Belgium Paris, France 3,657.0 18 Amsterdam, Netherlands Stockholm, Sweden 3,414.0 19 Madrid, Spain Paris, France 3,212.0 20 London, U.K. Madrid, Spain 2,880.0 21 Frankfurt, Germany Geneva, Switzerland 2,877.0 22 Amsterdam, Netherlands Geneva, Switzerland 2,833.5 23 Frankfurt, Germany NewYork, U.S. 2,729.0 24 Barcelona, Spain Lyon, France 2,688.0 25 Amsterdam, Netherlands Madrid, Spain 2,612.0 26 Geneva, Switzerland Lyon, France 2,567.0 27 Amsterdam, Netherlands Vienna, Austria 2,533.0 28 Moscow, Russia Stockholm, Sweden 2,522.0 29 Amsterdam, Netherlands Oslo, Norway 2,488.0 Copenhagen, Denmark DOsseldorf, Germany 2,488.0 D0sseldorf, Germany London, U,K. 2,488.0 DLisseldorf, Germany Prague, Czech Republic 2,488.0 Paris, France Washington, U.S. 2,488.0 Prague, Czech Republic Vienna, Austria 2,488.0 35 London, U.K. Washington, U.S. 2,378.0 36 Milan, Italy New York, U.S. 1,734.0 37 Frankfurt, Germany Vienna, Austria 1,676.0 38 Stockholm, Sweden NewYork, U.S. 1,553.0 39 Milan, Italy Paris, France 1,242.0 40 Helsinki, Finland Stockholm, Sweden 999.0 41 Dublin, Ireland London, U.K. 872.0 42 Oslo, Norway Stockholm, Sweden 862.0 43 Frankfurt, Germany ZiJrich, Switzerland 834.0 44 Geneva, Switzerland Milan, Italy 735.0 45 London, U.K. Stockholm, Sweden 701.0 46 Frankfurt, Germany Prague, Czech Republic 593.0 47 Vienna, Austria ZOrich, Switzerland 567.0 48 Amsterdam, Netherlands Washington, U.S. 555.0 49 Paris, France NewYork, U.S. 469.0 50 Amsterdam, Netherlands NewYork, U.S. 465.0 Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mbps Mb ~s M~ ~s Mb is Mbls Mb ~s Mb ~s Mb~s Mb is Mbps Mbps Note: Figures represent estimated Internet bandwidth be~veen Consolidated Metropolitan Statistical Areas or equivalent~ Domestic backbone routes ere omitted. Data as of mid-2000. Source: Te|eGeogrephy research © TeleGeogrephy, lnc. 2000 111 TeleGeography 2001 © TeleCeography, Inc. 2000 Figure 9. Map of Major South American International Backbone Routes, 2000 ^ ^ Aggregate Internet backbone bandwidth 250Mbps 125Mbps 50Mbps 20Mbps Intraregional Interregional Note: Map includes international backbone routes with at least :~0 Mbps of aggregate capacity. Figures represent estimated Internet bandwidth between Consolidated Metropolitan Statistical Areas or equivalents. Domestic backbone routes are omitted, Data aS of mid-2000. Source: TeleGeogrephy research © TeleGeography, Inc. 2000 112 © TeleGeography, Inc. 2000 TeleGeography 2001 Figure 10. The Top 50 International Backbone Routes in Latin America and the Caribbean, 2000 City, Country Cit~, Country Internet Bandwidth $8o Paulo, Brazil New York, U.S. 444.1 Mbps Rank 1 2 Mexico City, Mexico Dallas, U.S. 165.0 3 Mexico City, Mexico Los Angeles, U.S. 159.0 4 Mexico City, Mexico NewYork, U.S. 156.5 5 Mexico City, Mexico Houston, U.S. 155.0 6 Buenos Aires, Argentina New York, U.S. 148.2 7 Rio de Janeiro, Brazil Montreal, Canada 136.0 8 Buenos Aires, Argentina Palermo, Italy 102.0 9 Santiago, Chile Washington, U.S. 90.3 10 Buenos Aires, Argentina Nova Scotia, Canada 79.0 11 S~o Paulo, Brazil Montreal, Canada 68.0 12 Monterrey, Mexico Houston, U.S. 52.5 13 Hamilton, Bermuda NewYork, U.S. 49.0 14 Caracas, Venezuela Miami, U.S. 47.0 15 Rio de Janeiro, Brazil Miami, U.S. 46.0 16 Caracas, Venezuela Atlanta, U.S. 45.0 Monterrey, Mexico Dallas, U.S. 45.0 Mexico City, Mexico Orlando, U.S. 45.0 Mexico City, Mexico Tampa, U.S. 45.0 Santiago, Chile Chicago, U.S. 45.0 21 Buenos Aires, Argentina Washington, U.S. 42.5 22 S~o Paulo, Brazil Washington, U.S. 33.5 23 Lima, Peru NewYork, U.S. 29.0 24 Buenos Aires, Argentina Montevideo, Uruguay 28.0 25 Quito, Ecuador Miami, U.S. 27.7 26 Lima, Peru Miami, U.S. 25.0 27 Bogota, Colombia Miami, U.S. 23.0 28 Lima, Peru Washington, U.S. 20.0 29 Caracas, Venezuela Washington, U.S. 18.8 30 Rio de Janeiro, Brazil Washington, U.S. 17.5 31 Bogota, Colombia New York, U.S. 16.0 32 Caracas, Venezuela Los Angeles, U.S. 15.0 33 Santiago, Chile Miami, U.S. 14.0 34 Rio de Janeiro, Brazil NewYork, U.S. 13.5 35 Buenos Aires, Argentina Mexico City, Mexico 12.0 Buenos Aires, Argentina Miami, U.S. 12.0 Buenos Aires, Argentina San Francisco, U.S. 12.0 Rio de Janeiro, Brazil Dallas, U.S. 12.0 Santiago, Chile Los Angeles, U.S. 12.0 S~o Paulo, Brazil Los Angeles, U.S. 12.0 Quito, Ecuador Orlando, U.S. 12.0 42 Bogota, Colombia Montreal, Canada 9.5 43 Caracas, Venezuela NewYork, U.S. 8.8 44 Santiago, Chile NewYork, U.S. 8.4 45 Bogota, Colombia Toronto, Canada 8.0 Buenos Aires, Argentina Lima, Peru 8.0 Buenos Aires, Argentina Santiago, Chile 8.0 Guatemala City, Guatemala Mexico City, Mexico 8.0 Lima, Peru Boston, U.S. 8.0 Lima, Peru Seattle, U.S. 8.0 Mbps Mbps Mb)s Mb )s Mb )s Mb )s Mb)s Mb)s Mb )s Mb)s Mb)s Mb )s Mb )s Mb )s Mbps Mb )s Mb)s Mb)s Mb )s Mb )s Mb )s Mb ~s Mb )s Mbps Mb)s Mb)s Mb )s Mb )s Mb )s Mb )s Mb~s Mb~s Mb ~s Mb )s Mb )s Mb~s Mb~s Mb~s Mb =s Mb =s Mb~s Mb~s Mb~s Mbps Mbps Mbps Mbps Mbps Mbps Note: Figures represent estimated In~ernet bandwldl~ between Consolidated Metropolitan Statistical Areas or equivalents. Domestic backbone routes are omitted. Data as of mid-2000. Source: TeleGeography research © TeleGeography, Inc, 2000 TeleGeography 2001 © TeleGeograph~/, Inc. 2000 Figure 11. Montreal Map of Major African International Backbone Routes, A Tunis Aggregate [nternet backbone bandwidth Note: Map includes international backbone routes with at least 5; Mbps of aggregate capacity, Figures represent estimated In~ernet bandwidth between Conso|idated Metropolitan Statistical Areas or equivalents. Domestic backbone routes are omitted. Data as of mid-20~O. Source: TeleGeography research © TeleGeography, Inc. 2000 114 © TeleGeography, Inc. 2000 TeleGeography 2001 Figure 12. The Top 50 International Backbone Routes in Africa, 2000 Rank 1 2 ¯ CiW, Country City, Coun’m/ Internet Bandwidth Johannesburg, South Africa New York, U.S. 117.0 Mbps Johannesburg, South Africa Washington, U.S. 90.0 Mbps Capetown, South Africa New York, U.S. 79.0 Mbps Johannesburg, South Africa Boston, U.S. 79.0 Mbps 5 Rabat, Morocco Paris, France 68,0 Mbps 6 Capetown, South Africa London, U.K. 34.0 Mbps 7 Capetown, South Africa Los Angeles, U.S. 12.0 Mbps Rabat, Morocco Palermo, Italy 12.0 Mbps Tunis, Tunisia Montreal, Canada 12.0 Mbps 10 Johannesburg, South Africa London, U.l~ 10,0 Mbps Tunis, Tunisia New York, U.S. 10.0 Mbps 12 Cairo, Egypt Amsterdam, Netherlands 8.0 Mbps Nairobi, Kenya Boston, U.S. 8.0 Mbps Tunis, Tunisia Palermo, Italy 8.0 Mbps 15 Cairo, Egypt NewYork, U.S. 6.5 Mbps 16 Gaborone, Botswana Montreal, Canada 6.0 Mbps Port Louis, Mauritius Paris, France 6.0 Mbps 18 Algiers, Algeria London, U.K. 4.0 Mbps Cairo, Egypt Atlanta, U.S. 4.0 Mbps Cairo, Egypt London, U.K, 4.0 Mbps Cairo, Egypt Montreal, Canada 4.0 Mbps Johannesburg, South Africa Windhoek, Namibia 4.0 Mbps Port Louis, Mauritius New York, U.S. 4.0 Mbps Rabat, Morocco New York, U.S. 4.0 Mbps 25 Cairo, Egypt Los Angeles, U.S. 2,0 Mbps Algiers, Algeria Luxembourg 2,0 Mbps Algiers, Algeria Washington, U.S. 2.0 Mbps Antananarivo, Madagascar Paris, France 2.0 Mbps Cairo, Egypt Toronto, Canada 2.0 Mbps Dakar, Senegal Montreal, Canada 2,0 Mbps Nairobi, Kenya Montreal, Canada 2.0 Mbps Tripoli, Libya Montreal, Canada 2,0 Mbps Victoria, Seychelles Hong Kong, China 2.0 Mbps 35 Accra, Ghana Montreal, Canada 2.0 Mbps Kinshasa, Congo, D.R. of Brussels, Belgium 2.0 Mbps 37 Lagos, Nigeria Montreal, Canada 2.0 Mbps 38 Abidjan, Cote-d’tvoire NewYork, U.S. 2.0 Mbps Accra, Ghana Washington, U.S. 2,0 Mbps Annaba, Algeria Paris, France 2.0 Mbps Bamako, Mall Boston, U.S. 2,0 Mbps Dar es Salaam, Tanzania Boston, U.S, 2.0 Mbps Dar es Salaam, Tanzania Oslo, Norway 2.0 Mbps Harare, Zimbabwe Montreal, Canada 2.0 Mbps Harare, Zimbabwe Oslo, Norway 2.0 Mbps Harare, Zimbabwe Washington, U.S. 2.0 Mbps Kampala, Uganda Boston, U.S. 2.0 Mbps Nairobi, Kenya NewYork, U.S. 2.0 Mbps Nairobi, Kenya Paris, France 2.0 Mbps Nairobi, Kenya Singapore, Singapore 2.0 Mbps Ouagadougou, Burkina-Faso Montreal, Canada 2.0 Mbps Note: Rgures represent estimated Internet bandwidth between Consolidated Metropolitan Statistical Areas or equivalents. Domestic backbone routes are omi~ed. Data current to October 2000 Source: TeleGeography research © TeleGeography, Inc. 2000 115 TeleGeo~raplqy 2001 © TeleGeography, Inc. 2000 116 © TeleGeography, Inc. 2000 TeleGeography 2001 Figure 14. The Top 50 International Backbone Routes in the U.S. and Canada. 2000 City, Country Internet Bandwidth London, U.K. 26,680.5 Mbps Rank City, Country 1 NewYork, U.S. 2 NewYork, U.S. Amsterdam, Netherlands 9,958.0 Mhps 3 Montreal, Canada NewYork, U.S. 8,749.0 Mbps 4 San Francisco, U.S. Tokyo, Japan 7,550.0 Mbps 5 Chicago, U.S. Toronto, Canada 6,575.0 Mbps 6 NewYork, U.S. Toronto, Canada 4,862.0 Mbps 7 Seattle, U.S. Vancouver, Canada 4,707.0 Mbps 8 Washington, U.S. Frankfurt, Germany 3,663.0 Mbps 9 NewYork, U.S. Frankfurt, Germany 2,729.0 Mbps 10 Chicago, U.S. Montreal, Canada 2,488.0 Mbps Seattle, U.S. Toronto, Canada 2,488.0 Mbps Washington, U.S. Paris, France 2,488.0 Mbps 13 Washington, U.S. London, U.K. 2,378.0 Mbps 14 Toronto, Canada Washington, U.S. 2,176.0 Mbps 15 NewYork, U.S. Milan, Italy 1,734.0 Mbps 16 New York, U.S. Stockholm, Sweden 1,553.0 Mhps 17 Los Angeles, U.S. Tokyo, Japan 1,520.5 Mbps 18 San Francisco, U.S. Seoul, Korea, Rep. 1,336.6 Mbps 19 San Francisco, U.S. Hong Kong, China 1,209.5 Mbps 20 San Francisco, U.S. Sydney, Australia 1,030.0 Mbps 21 San Francisco, U.S. Singapore, Singapore 932.0 Mbps 22 Calgary, Canada Chicago, U.S. 777.0 Mbps 23 San Francisco, U.S. Taipei, Taiwan 741.5 Mbps 24 Chicago, U.S. Vancouver, Canada 622.0 Mbps 25 Washington, U.S. Amsterdam, Netherlands 555.0 Mbps 26 Los Angeles, U.S. Sydney, Australia 496.0 Mbps 27 New York, U.S. Paris, France 469.0 Mbps 28 NewYork, U.S. Amsterdam, Netherlands 465.0 Mbps New York, U.S. Palermo, Italy 465.0 Mbps San Francisco, U.S. Osaka, Japan 465.0 Mbps 31 NewYork, U.S. S~o Paulo, Brazil 444.1 Mbps 32 Seattle, U.S. Tokyo, Japan 400.0 Mbps 33 San Francisco, U.S. London, U.K. 355.0 Mbps 34 NewYork, U.S. Ziirich, Switzerland 324.0 Mbps 35 Los Angeles, U.S. Seoul, Korea, Rep. 310.0 Mbps 36 San Francisco, U.S. Auckland, NewZealand 255.0 Mbps 37 NewYork, U.S. Copenhagen, Denmark 245.0 Mbps Los Angeles, U.S. Hong Kong, China 245.0 Mbps 39 NewYork, U.S. Moscow, Russia 242.0 Mbps 40 Los Angeles, U.S. Auckland, New Zealand 200.0 Mbps San Francisco, U.S. Frankfurt, Germany 200.0 Mbps 42 San Francisco, U.S. Perth, Australia 196.1 Mbps 43 Sacramento, U.S. Seoul, Korea, Rep. 189.0 Mbps 44 Dallas, U.S. Mexico City, Mexico 165.0 Mbps Sen Francisco, U.S. Shanghai, China 165.0 Mbps 46 Los Angeles, U.S. Taipei, Taiwan 161.5 Mbps 47 Los Angeles, U.S. Mexico City, Mexico 159.0 Mbps 48 NewYork, U.S. Mexico City, Mexico 156.5 Mbps 49 Chicago, U.S. Calgary, Canada 155.0 Mbps Houston, U.S. Mexico City, Mexico 155.0 Mbps Note: Rgures represent estzmated Internet bandwidth between Consolidated Metropolitan Statistical Areas or equivalents. Domestic backbone routes are omitted, Data as of mid-2000. Source: TeleGeography research © TeleG, eography, Inc. 2000 117 TeleGeo{~raphy :2001 © TeleGeography, Inc. 2000 118 Traffic Analysis Tele~3eography 2001 © Tele~eog~’aphy, Inc. 2000 120 © TeleGeography, Inc. 2000 TeleGeography 2001 Overview of International Traffic Trends The effects of telecom market liberalization began to take hold in 1999, having been initiated in many countries only a year earlier Falling retail prices, combined with the sustained rapid growth of mobile phone usage, caused traffic growth to surge from approximately 13 percent in 1998 to over 15 percent in 1999. Worldwide international telephone traffic volume reached 107.8 billion minutes in 1999. The strong growth of international public switched minutes appears all the more remarkable in light of slumping fax traffic volumes and the increasing number of minutes moving off the Public Switched Telephone Network (PSTN) thanks to bypass mechanisms such as Voice-over-lnternet Protocol (VolP). Such trends dudng 1999 suggest that the international telecom industry had one foot on the accelerator, but the other on the brake. Pedal to the Metal Mobile Teleph6ny. Tens of millions of new mobile subscribers boosted international call volumes in 1999. In Europe alone, cellular operators added 75 million customers. The surge in cross-border roaming calls, which regularly are counted as international traffic, also contributed to heavier call volumes. Together, mobile roaming and subscribership are key factors in explaining why total international traffic from countries such as Germany and the Netherlands, which had been growing at rates of five percent or less in 1996 and 1997, showed double digit growth in 1999. Intensifying competition. Competition has intensified far more quickly in countries which liberalized their international long distance (ILD) markets in the late 1990s, compared to the more gradual increases observed in those markets that liberal- ized ten to 15 years ago. In the U.S., competition in ILD ser- vices began in 1982-1983; 12 years later, in 1995, AT&T’s market share was still more than 50 percent. In contrast, com- petition was introduced in Germany in 1998, and Deutsche Telekom’s market share had fallen to 55 percent by 1999. While Deutsche Telekom’s decline in dominance has been par- ticularly swift, its experience is certainly not unique (see Figure 1. Incumbent Market Shares Go in One Direction). New Carriers. The rise of the multinational carrier (MNC) model is one reason why incumbent market shares are shrink- ing more quickly in the newly liberalized markets than in coun- tries that introduced competition in the 1980s and early 1990s. Prior to the 1998 "Big Bang" telecom liberalization in Europe and elsewhere, only a handful of countries permitted competition. Usually, only a few significant competitors Figure 1. Incumbent Market Shares Go in One Direction 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 Source: TeleGeography research © TeleGeography, Inc. 2000 TeleGeography 2001 © TeleGeography, Inc. 2000 Figure 2. Falling Prices, Falling Revenues Lowest Available Retail Price for Calls from Germany US Turkey Poland Austria UK ~Year End 1997 ¯ January 2000 0.00 0.25 0.50 0.75 1.00 1 25 1.50 Deutsche Mark per Minute International Service Revenues and Call Volumes, 1998-1999 4o% Annual Call Volume Change 30% ¯ Annual Revenue Change Note: In January 2000, one U.S. dollar equaled 1.94 Deutsche Marks. Source: RegierungsbehiJrde ~rTelekommunikation und Post (RegTP), company reports, and TeleGeography research © TeleGeography, Inc. 2000 emerged to challenge the incumbent. These carriers often focused exclusively on one market. The proliferation of compe- tition since 1998 has created a critical mass of deregulated markets to support a multinational strategy, in which a carder provides facilities-based international service from many differ- Figure 3. Charge of the Challengers Traffic Base of Carriers Beginning Operations after 1989 25 0 20.0 15.0 10.0 5.0 0.0 Global Market Share of New Careers 24% ^^o~ 0.31’ 0.03% U.UU ~o 2% 3% 7% 5% 9% 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 Note: New carriertraff=c ihcludes only carriers that began facilities-based operation after 1989, Source: TeleGeography research ©TeleGeography, Inc. 2000 ent countries (see "The Growth of International Services Competition" in the Carders section of this report). MNCs hold a number of advantages over challengers pursuing a single market st-rategy. First, they enjoy economies of scale. They can apply the technical expertise, marketing skills, and company funding developed through their initial efforts in one country to new markets. Second, MNC affiliates have access to parent company international networks, which permit end-to-end rout- ing in many cases, thus facilitating the evasion of costly settle- ment charges. Finally, because many of their clients are them- selves multinationals, MNCs have a ready pool of customers when they move into new markets. For all these reasons, multi- national carders can steal away incumbent market shares with astonishing alacrity. In 1999, the number of minutes carded by new competitors grew over 70 percent, and accounted for 24 percent of world PSTN traffic (see Figure 3. Charge of the Challengers). Taken together, the subsidiaries of many multinational carders now carry more traffic than the incumbent operators of most countries. Some of these operators have emerged very quick- ly. In just the last year, U.S.-based World Access has pur- chased, among others, multinationals FaciliCom and WoddxChange as well as German operator TelDaFax. If it com- pletes its proposed merger with STAR Telecommunications, World Access will be the fourth largest international carder in the U.S. Multinational carriers need not be new operators. Teleglobe, for example, has transformed itself from the overseas telecom 122 © TeleGeography, Inc. 2000 TeleGeography 2001 service monopoly of Canada into a major global player. If the five billion minutes of international traffic from all of its country affiliates were to be aggregated, Teleglobe would be the third largest carrier in the world, ahead of all but AT&T and WorldCom. Price cuts. Fieme price competition has spurred call volume growth in many countries. However TeleGeography’s research indicates that competitive pressure has been so intense in some countries that pdces have fallen much faster than demand can possibly increase. For example, in Germany, prices on some routes have tumbled by as much as 90 percent in only two years, while call volumes Brew by only 45 percent (see Figure 2. Falling Prices, Falling Revenues). The aggressive price-cutting strategies of new carriers have placed incumbent carriers in a quandary~ tradition of monopoly and heavy mgnlation have rendered many of them uniquely ill-equipped to engage in pdce competition. If they cut prices to compete with their rivals, they will undermine the steep ILD profit margins on which they have based their busi- ness. However, if they do not cut their prices, they will lose cus- tomers to their new competitors. In Germany, prices fell so rapidly that Deutsche Telekom had no choice but to follow suit. However their c’hallengers cut pdces far morn aggressively, and Deutsche Telekom lost not just market sham, but also sub- scfibers. In the past two years, Deutsche Telekom’s interna- tional voice traffic has fallen by 19 percent, and revenues have plunged by 44 percent. In this, too, Deutsche Telekom’s expe- rience has not been unique (again, see Figure 2). Foot on the Brake Fax Traffic. The public switched traffic tracked by TeleGeography includes both voice and fax minutes. Since the 1980s, fax has been a critical component in the traffic mix. Fax traffic is particularly important on routes dominated by busi- ness calling and between countries that do not sham a common language--that is, in cases when written language is morn reli- able than oral communication. Fax traffic has also been signif- icant between countries separated by multiple time zones, where fax and email am morn reliable than real-time applica- tions (e.g., voice). With fax, a person can send a message even if the other party has left the office for the day. Estimates for the mid- 1990s put fax minutes at 30 percent of total international traffic. Yet by 1999, fax appeared to account for only half that sham. Evidence points to email as the culprit. In many situations, email has proven quicker, cheaper and easier to use than facsimiles; however fax traffic does remain significant on certain routes. Both email and fax volumes am likely high from the United States, where 55 per- cent of outgoing international traffic travels to countries morn than four time zones away. Fax as a sham of total traffic is highest from Asia, where multiple character sets render email a less reliable technology than the trusty fax machine. Figure 4. Bypass at Work? Reported U,S. Billed Traffic 40% 30% 20% 10% 0% -20% -30% ¯ 1996 to 1997 -40% 1997 to 1998 1998 to 1999 -50% Source: TeleGeography research ©TeleGeography, Inc. 2000 Bypass Traffic. In the case of bypass, the description of inter- national traffic as "flows" is apt because these calls, like water, seem to route around obstacles--such as high settlement rates. The traffic tables in TeleGeography 2001 reflect the global total of 107.8 billion minutes of public switched calls. We estimate that, in 1999, an additional five to eight billion minutes skirted the settlement rate mechanism, unaccounted (see Figure 4. Bypass at Work?). Growth of Voice-over-IP (VolP), a form of bypass, is particular- ]y rapid. While VolP accounted for only about 1.6 percent of total international voice traffic in 1999, it represents a much larger proportion of traffic on certain key routes. For example, TeleGeography estimates that VolP equaled at least 20 percent of traffic between the U.S. and China in 1999. In 2000, we expect global VolP volumes to morn than double. The Road Ahead The following pages explore in detail the directions in which international telecommunications am heading. First, an analy- sis of bypass and mille traffic provides a look at the drivers of alternative traffic arrangements. Results from a throe-month survey complement this analysis with an in-depth study of international VolR Information on mobile traffic volumes adds another perspective, showing how the wireless industry shapes international telecommunications. Finally, statistics on interna- tional call quality suggest how metrics besides call prices and volumes am important for understanding international telecom trends. 123 TeleGeography 2001 © TeleGeographyo Inc. 2000 Bypass and Refile Traffic Until just a few years ago, international telecommunication companies shared the cost and revenue for nearly every cross-border public switched call in accordance with the decades-old accounting rate regime. To send a call abroad, a carrier would route the signal onto its own international "half circuit," then transfer the call onto the matching net- work of its foreign counterpart for final termination. For this service, the originating carrier would pay the foreign telco a hefty settlement fee, usually equal to one-half the account- ing rate negotiated by the two carriers. However, as com- petition shrinks profit margins, carriers are finding new ways to protect their bottom lines; one approach is to retain more of their existing revenue streams by sending traffic around this settlement system. Legal Bypass Legal bypass, which’eschews traditional international settle- ment in favor of direct interconnection with foreign local exchange carriers (LECs), accounts for the largest portion of alternatively routed traffic. Regulators often refer to the practice as International Simple Resale (ISR), reflecting the practice of leasing and re-using private lines from foreign carriers--for many years, the only way to acquire a line to international destinations. This practice is gradually giving way to new options, such as the outright ownership of band- width between and within multiple countries. In 1999, legal bypass traffic accounted for approximately 15 percent of international call volumes worldwide. This percentage is surprisingly large considering the relatively small number of countries (at present, only a few dozen) that permit these alternative termination arrangements. However, legal bypass represents a growing portion of inter- national traffic because, although few in number, these countries tend to have very high traffic volumes. Of the 20 largest countries ranked by international minutes, 16 offer some form of direct interconnection. Although the U.S. reg- ulator permitted ISR with only 23 countries in 1999, these routes accounted for one-quarter of all U.S. outgoing traffic (see Figure 1. Top ISR Routes with the U.S., 1997-1999).. Yet those sums are just the beginning. Not all U.S. carriers used ISR to connect their calls to these 23 countries. If Figure 1. Top ISR Routes with the U.S., 1997-1999 80% O% ISR Traffic from U.S, 1997 1998 1999 684 336 7,153 2,954 914 World Germany United Canada Average Kingdom 2,216 ISR Traffic to U.S. 80% ~,, 1997 1998 1999 50% 149 40% 3,959 30% 20% 1,652 59O lO% 3Ol o% World Germany United Canada Average Kingdom 2,156 Note; Numbers above columns indicate millions of minutes sent via International Simple Resale (ISR). ISR with Germany was not legal m 1997, Source: FCC carrier filings © TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 200] Figure 2. Call Delivery Methods Standard Public Switched Call 1. Customer dials internat{onal number. 2. Call is sent over the Public Switched Telephone Network (PSTN) by originating carrier, wh=ch pays settlement charge to terminating career ~n desti- nation country. 3. Call is delivered to its final des- tination by terminating carrier. Refile 1. Customer dials international number. 2. Originating carrier sends call to hub country v~a PSTN or over international private ~ine, known as International Simple Resale (ISR). 3. Refile carrier re-originates call over PSTN. 4. Call is delivered to final desti- nation wa refile carrier, which pays settlement charge to ter- m=nat~ng carrier. Switched Bypass 1. Customer dials ~nternational number. 2. Call is routed over international private line to switch in desti- nation country but outside net- work incumbent telco (e.g., to closed user group or mobile operator). 3. Call is re-rerouted to incum- bent telco’s nebNork and com- pleted as a local call on PSTN. No international settlements are paid bythe originating car- tier. VolP Bypass 1. Customer dials international number. Call ~s routed over PSTN to gateway computer. 2. Call is converted from analog voice to Internet Protocol (IP) format and sent over the Internet to a gateway in termio hating country. 3. Call is converted back to ana- log format. 4. Call ~s completed as a local call on PSTN. No international set- tlements are paid by the origi- nating carrier. © TeleGeography, Inc. 1999 every carrier did, nearly 50 percent of U.S. traffic would take advantage of legal bypass. Illicit Bypass Although calls to countries where ISR is legal constitute almost 50 percent of U.S. outgoing traffic, they account for a mere 17 percent of total U.S. settlement outpayments. The remaining 80 percent of settlement revenues flow to countries where accounting rate bypass is illegal and per minute termination rates are, on average, four times higher. These countries present the greatest cost savings opportu- nities for bypass of the settlement rate and are, therefore, the most attractive targets for carriers seeking to evade set- tlement payments (see Figure 3. Bypass Targets). Some countries, (appearing in Figure 3 as gray circles hugging the "x" axis) have very high settlement rates but low volumes of incoming traffic. Other countries, Istacked along the "y" axis in Figure 3) receive substantial amounts of incoming calls but have low settlement rates. It is the combination of relatively high settlement rates and heavy traffic volumes that has historically created large volumes of illicit bypass traffic (e.g., to China, Jamaica, Philippines, Brazil, India, and Mexico). 125 TeleGeography 2001 © TeleGeographu, Inc. 2000 Figure 3. Bypass Targets Traffic and Settlement Rates from U.S. Carriers, 1999 1000 I~ Mexico India Brazil "~ Phd~pp~nes 200 ,~ China $0.00 $0.20 $0.40 $0.60 $0.80 $1.00 Settlement Rate wKh U.S. per M~nute Note: Data are for t~ethree targest U.S. carriers (AT&T, WorldCom, and Sprint). Traffic to Mexico is offthis scale at 3.6 billion minuteswith a settle- ment payment of $0.26 per minute. Source: TeleGeography research © TeleGeography, Inc. 2000 Carriers looking to smuggle phone calls into foreign coun- tries have a range of options. Internet Protocol (IP) tele- phony is one, relatively new, avenue for bypass. The trans- mission of voice over packet-switched data networks may occur primarily over the public Internet, or through private managed lines and terminating on traditional handsets. Companies can also accomplish bypass through older, switch-based technologies. Although few countries permit direct interconnection with the domestic PSTN necessary for ISR, many are more lenient in licensing intra-cc~mpany ser- vices for multinational corporations, provided that calls are contained within a "closed" user group. A common ploy for illegal bypass is to lease international circuits from an incumbent apparently for such closed use, then "leaking" non-group international minutes onto the local network. So how much international traffic illegally bypasses the accounting rate system? By virtue of their illicit character, traffic volumes in this "gray market" are extraordinarily dif- ficult to track. A smuggling arrangement is often transient: carriers lease a private line, aggressively ramp-up interna- tional call volumes, and then terminate the operation just as quickly, before local authorities even raise an eyebrow. TeleGeography estimates that illicit bypass volume was somewhere in the range of four to seven percent of global international traffic in 1999. Of these calls, roughly one- third traveled as VolP, the remainder as switched bypass over leased lines. Moreover, VolP volumes are set to climb dramatically in the next few years. (See "VolP Routes & Traffic" below.) Although illicit bypass accounts for only a small percentage of total world traffic, the practice wreaks havoc in some countries, siphoning up to 35 percent of incoming international traffic. In 1999, an estimated $200- 250 million dollars in settlement payments were lost due to illicit bypass from the U.S. alone. Figure 4. The Refile Shell Game U.S. & O.K. Settlement Rates to Selected Countries $1.00 $0.80 $0.20 Cheaper from U.S. Kuwait o o o Peru o°~ °oO ° °°° o oo oO o o Cheaper from U.K. $0.00 $0.00 $0.20 $0.40 $0.60 $0.80 $1 00 U.S. Settlement Rate (S/rain) Source: FCC, Office of Teleeommunica’~ons (OFTEL), and TeleGeography research Hypothetical Refile Transactions Settlement Refile Settlement Gain w/o Refile Fee w/Refile (Loss) Origin. U.K. ($0.66) ($0.25) $0.41 Hub U.S. $0.25 ($0.15) $0.10 Destin. Kuwait $0.66 $0.15 ($0.51) Origin. U.K. ($0.59) ($0.35) $0.24 Hub U.S. $0.35 ($0.25) $0.10 Destin. Peru $0.59 $0.25 ($0.34) Origin. U.S. ($0.34) ($0.23) $0.11 Hub U.K. $0.23 ($0.13) $0.10 Destin. Cyprus $0.34 $0.13 ($0.21) © TeleGeography, Inc. 126 © TeleGeograph~!, Inc. 2000 TeleGeography 2001 Refile Refile represents a third form of alternatively muted traffic. Instead of avoiding accounting rates altogether, those carri- ers employing refile bend the rules of the international set- tlement regime to their advantage. Refiling occurs when a carrier secretly re-mutes an outgoing international call through a third country, taking advantage of the intermedi- ate country’s lower settlement rate with the final destination country. Although the legal status of refile is more debat- able than that of many other forms of bypass, the practice is certainly illicit. With the intent of disguising the true ori- gin of traffic, the refile carrier in the intermediate country strips the numbering code, which identifies the originating country, replacing it with its own country code. This ruse makes economic sense in cases where sett}ement rate dis- parity exists between originating countries. For example, in mid-year 2000, the official settlement rate for traffic to Kuwait was $0.15 per minute from the UoS. and $0.66 per minute from the U.K. (see Figure 4. The Refile Shell Game). By charging British carriers a fee somewhere between the U.S. and U.K. rates--say, $0.25 for a one-minute call--a U.S.-based refiler could turn a $0.10 profit. Another winner would be the British carrier, saving $0.41 (minus the negli- gible transmission costs of re-muting the call through the U.S.). In contrast, the Kuwaiti telco would lose $0.51 in potential settlement income from the transaction. Refile traffic accounted for approximately 9.5 percent of world international call volumes in 1999. This percentage is much higher on routes to some developing countries, which tend to have higher settlement rates with most of their cor- respondents. For example, up to one-third of all calls to India, Pakistan, and Malaysia may have been refiled in ! 999, resulting in a significant loss of settlement income in those countries. Conclusion Countries wishing to stamp out refile and illegal bypass face two realistic options. The first option is to invest in the sophisticated technology necessary to detect the many sources of illicit flows, with an eye to quashing bypass oper- ations in the long term. The other option is to lower the offi- cial settlement rate, which would reduce the incentive for bypass and refile, forcing the would-be illicit operators above ground by drying up their margins. ~i~..~. ~27 TeleGeography 2001 © TeleGeography, Inc. 2000 VolP Routes & Traffic Entering Adolescence Since TeleGeography began tracking international phone calls more than a decade ago, market forces and technolog- ical innovation have driven down prices and increased traf- fic flows across the globe. The Internet has no doubt played a significant role in accelerating this process in the last few years, but quantifying the effect on actual traffic flows has been largely a speculative practice. Just two years ago, the combined traffic of all the companies routing international calls over Internet Protocol (IP) networks accounted for less than one-half of one percent of the world’s international minutes. Although Voice-over-lP [VolP) has only recently left its infancy as an alternative to traditional circuit- switched calling, the core infrastructure and support sys- tems necessary for making VolP a serious choice have now come online. .. In 1999, cross-border VolP call volumes reached approxi- mately 1.7 billion minutes--about 1.6 percent of total inter- national voice traffic. Based on TeleGeography’s survey results for the first half of 2000, the total international VolP market should reach 3.7 billion minutes for calendar year 2000 and, if the growth continues, 6.2 billion minutes in 2001 (see Figure I. International VolP and PSTN Traffic Summary). But the VolP industry is still young and unpredictable. While new and incumbent carriers alike are laying plans for IP networks that will carry all of their voice traffic in coming years, they have a way to go yet. The industry is still dri- ven by specialist VolP providers acting as carriers’ carriers to established phone companies, taking advantage of the arbitrage opportunities between official PSTN settlement fees and the de facto terminati~)n rates they can achieve (see Figure 3. Major VolP Carriers & Traffic). VolP clear- inghouses that link together the raft of new entrants which need a means to exchange traffic and settle accounts are also important (see Figure 4. Major VolP Clearinghouses). The myriad new retail VolP businesses that offer free PC-to- phone calling by way of advertising on Web-based commu- Figure 1. International YelP and PSTN Traffic Summary Total YelP and PSTN Traffic, 1997-2001 6’221m1 160,000 140,000 International VolP Traffic 120,000 1,699~ 100,000 8m 80,000 60,000 40,000 20,000 0 1997 1998 1999 2000 2001 YelP Traffic by Regional Destination, 1999 Africa U.S. & Canada 3% 7% Europe 16% Asza-Pac~fic 35% Latin America 39% Note: In these figures, Voice-over-tP (VolP) traffic includes all cross-border voic~ calls carried on IP networks buttermmated on public switched telephone net- works; PC-tmPC communications and private network:raffle ~re excluded. PSTN traffic includes circuit-sw’rtehed voice and fax traffic carried on traditional interna- tional faeilities as well as international simple resale (ISR) facilities. Source: TeleGeography survey results © TeleGeography, |~c. 2000 128 © TeleGeogra~hy, Inc. 2000 TeleGeography 2001 Largest U.S.-Originated VolP Routes, 1999 2. U.S. to Israel ................ 6% 3. U.S. toChina ................ 5% 4. U.S. to Korea, Rep ............ 4% 5. U.S. tolndia ................. 4% 6. U.S. to U.K ................... 3% 7. U.S. to Singapore ............ 3% 8. U.S. toBrazil ................ 3% 9. U.S. to Canada ............... 2% 10. U.S. to Ecuador .............. 2% 11. U.S. to Greece .............. 2% 12. U.S. to France ............... 2% 13. U.S. to Peru ................. [] 2% 14. U.S. to Philippines ............ [] 2% 15. U.S. to Germany ............. [] 1% 18. U.S. to Malaysia ............. [] 1% 17. U.S. to Russia ............... [] 1% 18. U.S. to Australia ............. [] 1% 19. U.S. to Taiwan ............... []1% 20. U.S. toHongKong ............ [] 1% Other routes ................ ..... .... 7 ¯ 26% Routes expressed as a percentage of reported ~nternatmnal VolP traffzc Largest U.S.-Originatod VolP Routes, 2000 1. U.S. to Mexico ............... ........ # 35% 2. U.S. to China ................ ~7% 3. U.S. to U.K ................... 4% 4. U.S. to Brazil ................ 3% 5. U.S. tolsrael ................ 3% 6. U.S. to Russia ............... 2% 7. U.S. to Canada ............... [] 2% 8. U.S. to India ................. [] 2% 9. U.S. to Bulgaria .............. I~ 2% 10. U.S. to Poland ............... [] 2% 11. U.S. to S. Africa .............. [] 2% 12. U.S. to Romania .............. [] 2% 13. U.S. to Greece ............... [] 2% 14. U.S. to Lebanon .............. [] 2% 15. U.S. to Philippines ............ [] 1% 16. U.S. to Peru ................. [] 1% 17. U.S. to Colombia ............. [] 1% 18. U.S. toTaiwan ............... [] 1% 19. U.S. to Bangladesh ........... [] 1% 20. U.S. to Korea, Rep ............ [] 1% Other routes ................ 25% Routes expressed as a percentage of reported ~nternatmnal YelP traffic Note; Route rankings are based on actualtraffic reports by major wholesale and retail VolP carriers as well as VolP clearinghouses¯ Figures do not include all VolP car~ers and routes, however, so some omissions may have occurred. Year 2000 rankings are based on statistics supplied for the first s~x months of 2000. Source: TeleGeegraphy survey results © 129 TeleGeography 2001 © TeleCeogmphu, Inc. 2000 1999 (Jan.~Dec.) 2000 (Jan.-June) Market Cap. Major Traffic (min.) Revenues Traffic (min.) Revenues (Oct. 2,2000) Shareholders 120 rn $12.4 m 130 m $15.3 m $104.5 m RS/Corn (95.5%) n.a. $3.0 rn 30 m $20.0 rn $1,255.0 m Verizon (10%) 157 m $19.4 m 205 rn $23.3 m $541.1 m n.a. ’150 m $28.0 rn 366 m $33.7 rn $563.6 m n.a. 259 m $41.0 rn 230 m $50.0 m $1,238.8 m 38 m n.a. 46 m $2.6 m $61.0 m Company deltathree.com NASD: DDOC) Genuity NASD: GENU) iBasis NASD: IBAS) ITXC NASD: ITXC) Net2Phone NASD: NTOP) Wherever.net NASD: WNET) 100% 90% 80% 70% 60% 5O% 40% 30% 20% lo% Market Share of International VolP Minutes, 1998-2001 o% 1998 1999 2000 2001 AT&T (32%) IDT (17%) Hontung Venture Capital (12%) nications portals constitute a third driver of traffic, although focus on what we do know. The statistics and analysis pre- most of these flows do not cross international borders as of sented on these pages are based on TeleGeography’s first yet. annual VolP routes survey, concluded in September 2000. Our Survey Given the still nascent stage of the VoIP industry, the installed base of circuit-switched transmission equipment, and the difficulty of tracking calls terminated in places where you may not want to advertise your success, making pre- dictions is hazardous business. Instead, in this essay we The goal of our survey was twofold: first, to measure how much VolP traffic transits international networks; and sec- ond, to establish where it is going. The data presented here include international phone calls that transit public or pri- vate IP networks at some point but are ultimately terminat- ed on traditional fixed or mobile networks. PC~to-PC corn- ,130 © TeleGeography, Inc. 2000 TeleGeography 2001 Figure 4. MajorVolP Clearinghouses 1999 (Jan.-Dec.) 2000 (Jan.-June) Clearinghouse]URL Members Traffic Revenue Members Traffic Revenue Concert Global Clearinghouse 84 90 m n.a. 100 120 m n,a. (www.concert.com) GRIC Comm., Inc. 250 98 m $8.0 m 350 118 rn $11.7 m (www.gric.com) ipx, inc. 3 15 m $2.3 m 31 19 rn $2.8 m (www.ipx-inc.com) Telia Clearinghouse Services 31 7 m $0.7 m 34 17 rn $1.8 m (clearinghouse.telia.corn) Note. Concert, ipx, and Telia traffic and revenue figures are for international VolP minutes only; GRIC traffic and revenue figures include per-minute tracking of dial- up lnternet service roaming. Clearinghouse membership figures for 1999 are current to December and 2000 figures are currentte June. Source: Te~eGeographysurvey research and company reports © Te/eGeography, inc. 2000 munications and private corporate network traffic are excluded because neither are directly comparable to PSTN traffic flows. And, because our survey is based on the reports of most--but not all--companies carrying VolP traf- fic, some routes may be under-reported. As a result, the route tables in Figure 2 present traffic flows in relative pro- portions rather than absolute minutes as we do in our PSTN tables. The Results Overall, our findings prove what we already suspected--that VolP is a new means to an old end. Because U.S.-based companies have had a head start in setting up their busi- nesses, most of the world’s VolP traffic currently originates in the U.S.--each of the world’s top 20 VolP routes for 1999 and 2000 originate there. Furthermore, because the Internet-remains U.S,-centric, U.S.-based VolP carriers have access to the most international IP bandwidth at the lowest prices. And, just as the U.S. continues to act as the prima- ry hub for Internet traffic, the U.S. may retain its position as a refile hub for VolP traffic even as the ranks of VolP card- ers proliferate into Western Europe and Asia. Although VolP calling patterns run roughly parallel to estab- lished PSTN demand, the largest share of VolP traffic has terminated in the countries where existing PSTN settlement rates are highest relative to the actual cost of getting the call there. Also, because quality expectations may be lower on many popular arbitrage routes, VolP calls compare favor- ably to the equally mediocre quality of many circuit- switched calls. The obvious example of the initial trend is traffic on the U.S.-Mexico route, which accounted for about one-third of global VolP traffic in 1999 and 2000. In the near future, we also expect that traffic into China, Brazil, and India will increase dramatically as VolP termination arrangements expand and IP infrastructure matures, providing a viable alternative to prohibitively high PSTN settlement rates (see Figure 2. The Top 20 U.S.-Originated VolP Routes, 1999- 2ooo). But VolP is not all about arbitrage. Although established carriers have yet to migrate much of their traffic onto VolP networks, many have begun trials or made announcements of their transition to more efficient all-IP infrastructure in coming years. Moreover, as new IP communications ser- vices and devices become available, they may stimulate new demand and increase VolP traffic flows beyond the growth rates characteristic of the traditional voice telephony mar- ket. We will be watching--and reporting--these develop- ments as they occur ~i~ 131 TeleGeography 2001 © TeleGeographg, Inc. 2000 International Traffic from Mobile Phones "By 2001, one in ten international calls will orig- inate on mobile terminals." -TeleGeography 2000 That bold statement, penned just one year ago in the first- ever study of mobile-based international call volumes, had the right idea but the wrong date. Mobile-originated inter- national calling already has exceeded the envisioned one in ten ratio, increasing from eight percent of international calls in 1998 to 11.5 percent in 1999 (see Figure 1. Mobile vs. Fixed International Traffic and Subscribers by World Region, 1999). At current growth rates, mobile calls in 2001 will account for one out of every four international minutes. Two factors in particular help explain the growing importance of mobile networks in international calling: wireless subscriber- ship growth and roam, ing. Subscribership Sometimes, the simplest explanations are the best: mobile- originated international minutes are increasing because more people are using cellular phones. While fixed line sub- scribership growth remains in the single digits in most regions of the world, mobile subscribership growth has not slowed from its torrid pace in the 1990s. In Africa, the number of new mobile phone lines doubled during 1999. In Europe, cellular subscribership grew by 60 percent--a fact all the more remarkable when one considers that there were already 100 million mobile phone users in 1998. Mobile users now account for one-third of all phone subscribers worldwide, and appear set to exceed fixed lines around the middle of this decade. The growth of mobile-originated international traffic stem- ming from subscribership changes may be as much a case of traffic substitution as it is of traffic creation. Many of the international calls placed from wireless phones would other- wise have been dialed from fixed line sets. Furthermore, although most people still use cellular phones as a supple- ment to their fixed line phones, growing numbers of sub- scribers are switching off their fixed lines altogether. Between 1998 and 1999, the total number of active fixed lines actually declined in fourteen countries, including Austria, Japan, and Israel. The growing disparity in wireless versus fixed subscribership patterns helps explain why inter- Figure 1. Mobile vs. Fixed International Traffic and Subscribers by World Region, 1999 From Mobiles From Fixed International Traffic L America & Africa Caribbean ~~ --U S & Canada ~As=a & Oceama Europe" Tot~l = 12,400 m minutes Africa L America & .Caribbean Total = 95,600 m minutes Subscribers Mobiles A~ca L America & Caribbean Fixed Africa L America & Caribbean Total=900 m Source: TeleGeog’raphy research and ITU © TeleGeography, tnc. 2000 132 © TeleGeography, Inc. 2000 TeleGeography 2001 national call minutes from mobile phones increased by a robust 66 percent in 1999 compared to a relatively paltry 11 percent growth from fixed line phones. Roaming Calls from cell phone users when traveling abroad represent a second major impetus for traffic growth. The reason has to do with the roundabout circuit that many roaming calls take. If a German businessman traveling jn Z(~dch were to call someone in Switzerland, for example, the call likely would first be switched through his operator’s facilities in Germany before traveling back across the Swiss border. Each roaming call--even calls to local destinations in Z0rich--would count as international traffic from Germany to Switzerland. In contrast to the subscribership effect, which shifts some international traffic from fixed lines onto mobile phones, roaming largely creates new international traffic that other- wise would not exist. In 1999, for example, the number of international call minutes from Germany to Switzerland increased from 400 to 650 million minutes, a fact that is hard to explain except for roaming. Roaming is particularly significant in Europe, Afdca, and Asia, where th~ common standard Global System for Mobile Communications (GSM) permits interconnectivity. The wide- spread use of roaming over GSM networks helps explain why international mobile calls appear so high in these regions (see Figure 2. Percent of International Traffic from Mobiles, 1998-1999). Roaming is even beginning to have a notice- ably upward impact on total international traffic volumes. Mobile-originated international traffic is one reason why Figure 2. Percent of International Traffic from Mobiles, 1998-1999 20.0% ~1998 m1999 10.0% 2.9 8,2 5.0% 0.0% Africa U.S.& L. America Asia & Canada &Carib. Oceania Europe World Total Note: Numbers show total mobile-originated international minutes in billions. Source: TeleGeography research and ITU © TeleGeography, Inc, 2000 Figure 3. Mobile Subscribers and International Traffic for Selected Countries, 1999 Italy 806.0 27.2 1,399.50 0.2 13.1 30.3 UK. Z=mbabwe Sw=tzerland 2,9 955.5 Germany 1,184.1 23.5 Indones=a 0% 47.0 86.1 ¯ Subscribers (m) 2.2 Int’l Traffic (m mins) 10% 20% 30% 40% 50% 60% Percent from Mobdes Note: Numbers show absolute levels of subscribership and call minutes from mobiles. Lengths of bars show relative levels from mobiles. Source: TeleGeography research and [TU © TeleGeography, inc. 2000 total international traffic from Europe in 1999 grew at its fastest pace in over a decade. The roaming phenomenon is also a factor in the low level of international calls placed from cellular phones in the U.S., where GSM networks are less common (see Figure 3. Mobile Subscribers and International Traffic for Selected Countries, 1999). Limits to Growth Despite the spectacular increases in mobile-originated inter- national calling, limits to its growth do remain. Worldwide, individuals place 105 minutes of international calls per year on their fixed line sets but only 25 minutes per year on their mobile phones. In Italy, over half of all telephones are mobiles, and yet 74 percent of international calls come from wireline phones. The reason is that, despite their ddve toward ubiquity, mobile terminals are not perfect substitutes for wireline telephones. In many countries, call prices from mobile phones remain more expensive than from wireline sets. Also, many people prefer to place international calls from their offices or homes, where they have more reliable call quality and less background noise. As we concluded in our TeleGeography 2000 analysis, it seems safe to say that "the fixed line telephone will remain the dominant medium for originating and terminating international calls" for the foreseeable future. But then again, the astonishing growth of cellular telephony has a history of leaving even the best predictions in its wake. ~iL~ 133 TeleGeography 2001 © TeleGeography, Inc. 2000 International Call Quality Metrics Minutes, revenues, bandwidth--all are vital statistics for track- ing changes in the telecom industry. In fact, much of what we know about international telecommunications traffic reflects such volumetric data. Yet statistics that describe call quantity paint only a partial picture.all quality is also a critical com- ponent. However, while collecting volumetric data is relatively straightforward; quality, on the other hand, is subjective. So how can quality be quantified? Measuring the Subjective Monnet UK Ltd., an independent Quality of Service (OoS) arbiter, is implementing one approach. In addition to monitor- ing call quality on its clients’ networks, Monnet also constructs industry benchmarks, pooled from data provided by participat- ing carders. Figure 2 shows survey results for 45 destination countries, based on a sample of 30 million international calls from German and u.K., carriers between June 1 and August 31, 2000. Monnet employs three indicators to measure call quality: ¯ Answer Seizure Ratio. ASR measures the percentage of successful call attempts between a switch and a given desti- nation. A 50 percent ASR means that only one-half of all call attempts were answered by a person or device; an unan- swered call or busy signal counts as an unsuccessful call. That means ASR is affected not only by performance fac- tors--availability of dial tone and the network’s ability to establish a transmission path or switch a call--but also by phenomena ranBing from a changed dialing code to a holi- day season, leading to more unanswered calls due to wrong numbers or busy signals. ASR standards vary significantly by reBion. For example, the range of acceptable ASR for calls.to developed countries generally is 60 to 75 percent. ¯ Post Dial Delay. PDD measures the time it takes a net- work to establish a connection once the caller has finished dialing. Hence, a PDD of 8.2 means that an average of 8.2 seconds elapse between the dial and the ring at the other end. ¯ Call Quality Index. CQI, expressed on a scale of 0 to 100, consists of a basket of five qualitative factors: signal level, noise, echo path loss, echo path delay, and speech activity. All five factors are based on a technical model pro- vided in ITU-T Recommendation G.107 (vwwv.itu.int/~doc/~- t!redg/g100-699/s_g107.htm). To earn a "best" ranking, a call must post a COl score between 80-100; on the other end of the scale, a COl of less than 60 is characterized as "poor" iiiiii:ii!il 0.0 2.0 4.0 6.0 "~ ::i:i:iiii:ii 0.0 20.0 40.0 60.0 80.~ 1~0.0 !:ii:iii!i!iiiiill !!iii:::i Calls to Axed Line Phones :i~i~i!ii~i!I Ca s to Fixed L ne Phones i:i~ii:~ili~ili:i 134 © TeleGeography, Inc. 2000 TeleGeography 2001 Answer Seizure Ratio Post Dial Delay (seconds) from Call Quality Index Destination from Germany from U.K. from Germany from U.K. Germany from U.K. Australia 42% 37% 8.2 9.0 61.4 53.8 Austria 67% 29% 5.4 4.5 78.1 68.5 Belgium 54% 30% 2.7 6.1 67.7 60.6 Brazil 34% 37% 5.5 4.5 62.5 64.9 Canada 63% 72% 4.9 1.7 83.8 83.6 Chile 35% 13% 3.6 5.1 60.6 67.2 China 33% 28% 7.1 10.3 49.1 52.5 Colombia 34% 22% 5.6 4.9 55.5 58,2 Denmark 53% 56% 6.0 3.6 44.6 64.6 Ecuador 16% 24% 7.1 4.1 52.5 53.6 Finland 47% 60% 5.3 2.9 66.4 65.2 France 34% 57% 4.5 3.4 58.5 54.0 Germany 44% 61% 7.6 2.3 68,0 77.2 Ghana 14% 19% 6.1 6.4 42.9 38.8 Greece 46% 45% 4.9 2.3 58.8 64.1 Hong Kong 60% 56% 6.0 4.0 75.8 47.7 India 19% 24% 7.3 7.4 47.4 55.9 Ireland 52% 43% 2.9 4.7 81.7 62.3 Israel 20% 36% 8.7 6.9 61.1 63.5 Italy ¯ 31% 24% 4.6 9.4 53.7 58.1 Japan 42% 20% 7.0 6.3 79.8 76.4 Korea, Rep. 27% 32% 8.1 6.0 69.3 59.4 Kuwait 46% 36% 4.2 4,9 56.1 58.3 Macedonia 27% 36% 6.7 4.9 48.2 55.1 Malaysia 44% 46% 6.9 5.9 77.4 63.4 M exic o 41% 39% 5,7 5.6 68.0 70.4 Netherlands 47% 44% 2,9 5.9 67.9 58.8 Norway 14% 52% 4.0 4.2 51.7 67.6 Pakistan 14% 19% 6,6 5.7 46.0 44.1 Peru 31% 42% 3.9 5.1 51.9 61.5 Philippines 34% 34% 7.2 5.3 52.8 56.6 Pola nd 47% 46% 4.2 4.3 66.9 65.2 Romania 47% 25% 2.0 7.1 60.0 63.8 Russia 34% 31% 6.0 5.2 64.9 58.1 Saudi Arabia 36% 40% 4.3 4.2 54.0 59.2 Singapore 60% 61% 7.3 3.2 60.2 55,2 South Africa 59% 45% 7.5 5.3 55,0 59.0 Spain 30% 20% 6.1 6.2 58.7 59.6 Sweden 49% 61% 2.4 6.2 74.0 60.7 Switzerland 34% 49% 4.8 3.7 65,6 69.4 Taiwan 37% 51% 7.7 5.8 58.0 57.6 Turkey 26% 28% 4.2 7.6 53.7 56.6 UAE 34% 38% 9.2 6.4 64.7 63.0 United Kingdom 48% 73% 2.8 2.4 71.2 68.4 United States 66% 73% 3.2 2.8 81.5 68.8 Average (Summer 2000) 40,0% 41.0% 5.0 5.4 65.2 61.4 Average (Summer 1999) 58,0% 42,9% 3.4 4.7 75.6 60.0 Source: Monnet UK Ltd., 2 Honey Lone, Cheapside, London EC2v 8BT, U.K. Tet +44 20 7367 5350 Fax +44 20 7387 5360 Ema~l: info@monnef=uk.com ¯ http~/wW~N.monnet.uk.cem D es=gn © TeleGeog raphy, In c. 2000 TeleGeography 2001 © TeleGeography, Inc. 2000 Many factors affect CO1 scorns, including basic infrastructure problems, packet loss in IP networks, the excessive use of compression, and switching calls between many service providers. Window on Industry Change For carriers that subscribe to quality testing services such as Monnet’s, industry-wide benchmarks provide an essential tool for pinpointing those network links that are not up to par with the competition. Beyond the immediate commercial benefit to subscribers, benchmarks also identify wider industry trends, such as the predictable gap between call quality to developing and developed countries due largely to weaker telecom infra- structure. Quality data vary not only by destination, but by the country of origination and time period studied, as well. For example, calls measured by Monnet during the June-August 1999 peri- od from Germany scored markedly higher than those calls from the U.K. In the summer of 2000, however, averages of call quality statistics from the two countries were nearly identical. Possible explanations for this convergence point to wider impli- cations for the industry: ¯ Mobile Traffic.’.When a call transits a mobile network, a number of characteristics appear that tend to drive down Call Ouality Index scores--noise, echo, and delay {see Figure 1. Call Ouality from Germany to Mobile and Fixed Line Telephones). The economics of sending calls to mobiles further complicate the metric; high intercon- nect fees to mobile networks induce terminating carriers in some countries to answer those incoming calls des- tined for mobile phones with a busy signal. This practice may partially explain the siidinB German call quality dis- cussed earlier, considednB the hiBh Browth rate of traffic to mobile terminals from Germany. ¯ Rapidly expanding call volumes. Especially in newly opened markets such as Germany, eme~inB carriers sometimes attract more traffic than originally anticipated by network planners. Some network links simply cannot handle these unexpectedly heavy traffic loads, and the network upgrades necessary to accommodate such traf- fic volumes require investment over a long time period. In order to continue offerinB service while networks are overloaded, some carriers have resorted to "call gap- ping." Using this practice, a carrier accepts only a limit- ed portion of total placed calls at any one time; individ- uals whose calls are blocked generally hear-a recorded message statinB that "all circuits are busy." ¯ Price/Quality Tradeoff. In Germany, call prices on some international routes have plummeted 90 percent in just two years, squeezinB profit marBins. In response, more service providers are willinB to purchase minutes from wholesale carriers at mediocre quality--as IonB as they deliver at rock bottom pdces. Many of these whole- sale carriers operate in the Bray market of international telecommunications, using alternative routing technolo- gies such as Voice-over-lnternet Protocol (VolP) to evade costly PSTN settlement charges. While these mecha- nisms enable cost-cutting by carders, they can also frus- trate call quality guarantees. ¯ Least Cost RoutinB. Overall, the correlations between Germany- and U.K.-oriBinated call quality metrics were markedly hiBher in 2000 than in 1999. For example, COl ratinBs from Germany and the U.K. tended to be more hiBhly correlated in 2000 than in 1999 to various destinations (e.B., India and Kuwait). This converBence of U.K. and German call quality data suBBests a Breater reliance on least cost routinB. FuelinB this trend are the many fast-BrowinB multinational carriers in Germany with a substantial presence in other major markets, such as the U.K. Often, these carriers switch calls throuBh affili- ate networks before sendinB them on to their final desti- nations. If, as seems likely, calls from Germany and the U.K. are carried increasinBly over the same international links, it stands to reason that call quality levels for these two countries will converBe. Call quality metrics are a critical part of the movement toward a more robust standard of international service. First and foremost, specific call quality metrics enable carri- ers to monitor flow and to diaBnose their networks for main- tenance and upBrades. However, industry benchmarks also illuminate technoloBical and reBional trends that impact wider business development decisions. ~.~ 136 Traffic Summary TeleGeography 2001 © TeleCeogrophg, Inc. 2000 138 © TeleGeography, Inc. 2000 TeleGeo{~raphy 2001 Global Traffic Review Figure 1. International Traffic and Main Line Growth 2,500 2,250 2,000 1,750 1,500 1,250 1,000 750 5OO 25O 0 Fixed main lines ¯ Cellular subscribers --o-International call minutes 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 25O 225 2OO 175 150 125 100 75 5O 25 0 2003 Note: Data include outbound intemadunaJ traffic on public networks only. Prejec~ons assume 15% t~affic grow~, 5% main line growth, and 30% mobile subscriber growth annually. Source; ITU and TeleGeography research ©TeteGeography, Inc.2000 Figure 2. International Traffic, Revenue, and Subscriber Growth Historica/ Trend S/ow Growth Same Growth CAGR CAGR CAGR Indicator 1995 1999 1995-99 2003 1999-2003 2003 1999-2003 Ca~ls (bn) 17.1 34.7 19.3% 60.4 14.9% 67.1 18.0% Minutes (bn) 61.6 107.5 14.9% 169.2 12.0% 188.0 15.0% per main line subscriber 89.0 119.4 7.6% 151.7 6.2% 163.9 8.2% per main line plus mobile 78.9 77.6 -0.4% 73.6 -1.3% 74.3 -1.1% Revenue (US$ bn) 55.0 66.7 4.9% 71.9 1.9% 69.9 1.2% Assumptions Call length (rains) 3.6 3.1 -3.7% 2.8 -2.5% 2.8 -2.5% Price per minute (US$) 0.90 0.62 -8.9% 0.43 -9.0% 0.37 -12.0% Main lines (m) 691 900 6.8% 1,115 5.5% 1,147 6.3% Mobile subscribers (m) 89 485 52.8% 1,184 25.0% 1,385 30.0% Total subscribers (m) 781 1,385 15.4% 2,299 13.5% 2,532 16.3% Fast Growth CAGR 2003 1999-2003 71.9 20.0% 201.4 17.0% 170.8 9.3% 72.2 -1.8% 65.4 -O.6% 2.8 -2.5% 0.32 -15.0% 1,180 7.0% 1,611 35.0% 2,791 19.1% Note: 1995-99 based on repo~ed data. 2000-2003 based on ITU and Te/eGeography forecasts. Scenarios are as follows: 1. Slow Growth: Traffic growth slows as minutes move off the pubhc switched neWvork (PSTN) and large markets mature. 2. Same Growth: Traffic growth continues at szmilar rate of last five years, assuming that faster rates of price-cutting keep traffic on the PSTN. 3 Fast Growth: Traffic growth increases, assuming a faster growth rate of network subscribers and faster rates of pr~ce-cutting, plus a significant component of new demand created by internatmnal traffic generated from mobdes Source: ITU World Te/ecommumcation/ndicators Database, ITU estimates, and TeleGeogrephyresearch © TeleGeography, Inc. 2000 139 TeleGeo{~raphy 2001 © TeleGeogrophg, Inc. 2000 Figure 3. Intercontinental Traffic Flows, 1997 & 1999 Key Traffic Flows ~,5~ 1,[~Q 500 10O Mdhon Minutes Each band is proportional to the total annual traffic on the public network in both directions between each pair of countries. These maps show at! intercontinen- tal routes with an annual volume of more than 100 million minutes. The total volume of these routes in 1999 was 26.6 billion minutes, approxi- mately 25 percent of global interna- tional traffic. © TaleGeography, Inc. 2000" 140 © TeleGeographtj, Inc. 2000 TeleGeo{~raphy 2001 International Traffic by Region Figure 1. Interregional Traffic Flows, 1999 Key All figures are given in btlhons of mi- nutes of telecommunications traff=c forthe public telephone network. 5 Million Minutes Each band ~s proportional to the total annual traffic from one region to another. Arrows indicate d=rection of traffic between regions Note: These interregiona! traffic flows total 30.6 billion minutes. That sum does not equal global total of 107.8 billion minutes because (1) data set based on top 20 out- going routes for 130 largest ~euntries on~y; |2) traffic w~tbin regions account for a further 52.2 billion minutes based on data set; (3) interregional routes below 100 mil- lion minutes not shown. © TeleGeography, Inc. 2000 Figure 2. International Traffic by Origin, 1999 Latin Amer=ca & Caribbean 4.2% 0ceama Africa 1.9% Note: Global traffic was 107.8 bn mznutes in 1999. © TeleGeography, Inc. 2000 Figure 3. Traffic Growth by Region, 1998-1999 30% 25% 20% Global Average 15% 10% 5% 0% Latin Asia America & Caribbean Europe Afrzca U S & 0ceama Canada Note: Global traffic was 107.8 bn minutes in 1999. @TeleGeography, Inc. 2000 TeleGeography 2001 © TeleGeographu, Inc. 2000 Figure 4, European Telecommunications Traffic Flows, 1999 SWEDEN FINLAND o CZECH HUNGARY raphy, inc. 2000 Key All figures are given m millions of minutes of telecommunications traffic for the public telephone network~ The map shows all intra-Eurepean routes with a combined t999 volume of more than 135 million minutes Traffic Flows 600 400 200 Million Minutes Each band is proportional to the total annual traffic on the public telephone network in both directions between each pair of CeU~ltl~es. (~)Total Outgoing Traffic 50 ,000 5,000 illio~ Minutes The area of each circle is proportional to the volume of the total annual outgoing traffic from each country, Balance of Traffic On routes where traffic in one direction accounts for at least 60% of the total, an arrow shows the direction of most of ~he traffic flows, © TeleGeography, Inc. 2000 © TeleGeographyo Inc. 2000 TeleGeography 2001 MEXICO 64%-~;~ COLOMBIA BOLIVIA URUGUAY ARGENTINA © TeieGeography. Inc. 2000 Key All figures are given in millions of minutes of telecommunications traffic far the p~b~ic telephone network. The map shows all routes within Latin America with e combined 1999 volume of more than 10 million minutes, Traffic Flows 100 50 Million Minutes Each band is proportional to the total annual traffic on the public telephone network in both directions between each pair of countffes. (~)Total Outgoing Traffic Million Minutes The area of each circle is propo~onal to the volume of the total annual outgoing traffic from each country. Balance of Traffic On routes where traffic in one direction accounts for at least 60% of the total, an arrow shows the direction of most of the traffic flows. © TefeGeograptiy, Inc, 2000 143 TeleGeography 2001 © Te/eGeograph~/, Inc. 2000 Figure 6. Asian Telecommunications Traffic Flows, 1999 AUSTRALIA © TeleGeography, inc. 2000 Key Traffic Flows All figures ere given in millions of minutes of telecommunications ¯ ~’affic for the public telephone 500 200 50 network, Million Minutes The map shows all intra-Asian routes with a combined 1999 volume of more than 50 minutes. Each band is proportional to the total annual traffic on the public telephone network in both directions between each pair of countries. (~)Total Outgoing Traffic ( 0o 500 1,500 Milt~on Minutes The area of each circle is propor~onaf to the volume of the total annual outgoing traffic from each country. Balance of Traffic On routes where traffic in one direction accounts for at least 60% of the total, an arrow shows the direction of most of the t~affic flowsL © TeteGeography, [nc, 2000 144 © TeleGeographg, Inc. 2000 TeleGeography 2001 International Traffic by Country Figure 1. Outgoing International Telephone Traffic Growth for Selected Countries, 1998-1999 45% 4O% 35% 3O% 25% 20% 15% 10% 5% O% -5% -10% -15% Global Average © TeleBeography, Inc. 2000 Figure Z. Tolephone Traffic Balaoces for Selected Countries, 19~ Countries with a traffic surplus (incoming exceeded outgoing minutes) Chart to Scale U.S. Deficit Dominican Republic ~~ Turkey ~, Egypt ~: Jamaica Brazil Portugal Cuba Venezuela Syna Guatemala Greece Japan Kuwait Ta~wan Norway Hong Kong ~~~~~J~l~l~ United K,ngdom Unrted States Countries with a traffic deficit (outgoing exceeded incoming minutes) -3,500 -2,500 -1,500 -500 500 1,500 2,500 3,500 International Telephone Traffic Balance (Millions of Minutes) © TeteGeography, Inc. 145 TeleGeo{~raphy 2OO1 © TeleGeographg, Inc. 2000 Figure 3. International Traffic Indicators, 1999 International Traffic Outgoing Incoming Balance Population Minutes (Out) Main Lines Minutes (Out) (m minutes) (m minutes) (m minutes) (m) per Capita (thous.) per Main Line Albania (a) 74.6 121.7 47.1 3.9 19.4 140 531.4 Algeria (d) 121.3 n.a. n.a. 30.1 4.0 1,477 82.1 Andorra 53.2 n.a. n.a. 0.1 708.3 n.a. 1,516.9 Angola 35.0 33.1 -1.9 12.5 2.8 96 362.9 Argentina 377.6 n.a. n.a. 36.6 10.3 7,357 51.3 Armema 33.7 89.8 56.0 3.5 9.6 547 61.6 Australia (b) 2,115.0 n.a. n.a. 18.9 111.8 9,857 214.6 Austria (c) 1,350.0 n.a. n.a. 8.2 165.1 3,939 342.7 Azerbaijan (a) 32.2 68.6 36.4 7.7 4.2 730 44.1 Bah a mas (d) 63.5 90.0 26.5 0.3 214.6 106 600.1 Bahrain (a) 134.1 106.5 -27.5 0.7 201.6 165 810.8 Bangladesh (d) 41.8 196.2 154.4 124.8 0.3 378 110.5 Belarus (a, c) 161.2 195.6 34.4 10.3 15.7 2,683 60.1 Belgium (a) 1,590.0 n.a. n.a. 10.2 156.6 5,100 311.8 B olwvi a 29.7 82.2 52.5 8.1 3.6 502 59.1 Bosnia-Herzegovina (c) 97.2 200.5 103.3 3.8 25.3 368 264.2 Brazil (a) 574.8 838.5 263.7 168.0 3.4 24,985 23.0 Brunei (c) 18.8 21.7 2.9 0.3 58.3 n.a. 227.7 Bulgaria 98.9 n.a. n.a. 8.3 11.9 2,833 34.9 Canada (a) 5,680.0 n.a. n.a. 30.5 186.3 19,957 284.6 Chile 270.0 n,a, n,a. 15.0 18.0 3,109 86.9 China 1,950.0 n.a. n.a. 1,266.8 1.5 108,716 17.9 Costa Rica (a) 94.1 109.0 14.9 3.9 23.9 803 117.3 CSte d’lvoire (d) ’, 57.3 46.6 -10.7 14.3 4.0 170 337.1 Cuba 32.6 225.3 192.7 11.2 2.9 434 75.1 Cyprus (a, c) 168.2 134.1 -34.0 0.8 216.0 424 396.6 Czech Republic (a) 364.0 452.2 88.2 10.3 35.5 3,806 95.6 Denmark (a) 800.0 n.a. n.a. 5.3 150.6 3,638 219.9 Dominican Republic 185.7 920.0 734.3 8.4 22.2 n.a. 229.3 Egypt 171.0 554.6 383.6 67.2 2.5 4,686 36.5 El Salvador (c) 47.0 n.a. n.a. 6.2 7.6 466 100.4 Estonia (a) 74.6 84.8 10.2 1.4 51.6 510 146.3 Finland (c) 423.9 n.a. n.a. 5.2 82.1 2,850 148.7 France 4,950.0 n.a. n.a. 58.9 84.1 34,100 145.2 Georgia (c) 46.7 65.7 19.0 5.5 8.6 672 69.5 Germany (a) 6,965.0 n.a. n.a. 82.2 84.8 48,300 144.2 Ghana 30.1 118.4 88.2 19.7 1.5 159 189.9 Greece (c) 725.7 794.2 68.5 10.6 68.3 5,611 129.3 G u ate mala 83.3 208.6 125.3 11.1 7.5 605 137.7 Guyana 16.1 101.0 84.9 0.9 18.8 64 250.8 Hong Kong (a, b) 2,720.3 1,747.2 -973.1 6.9 395.2 3,869 703.1 Hungary (a) 343.9 n.a. n.a. 10.2 33.7 4,109 83.7 India (a, b, c) 473.3 1,772.5 1,299.2 996.1 0.5 26,511 17.9 Indonesia (a, c) 269.6 n.a. n.a. 209.3 1.3 6,080 44.3 Iran 200.4 216.3 15.9 66.8 3.0 8,371 23.9 Ireland (b) 1,015.0 n.a. n.a. 3.7 273.9 1,770 573.4 israel (a) 804.0 n.a. n.a. 6.1 131.8 2,800 287.1 Italy 3,100.0 n.a. n.a. 57.3 54.1 26,502 117.0 Jamaica (a) 66.4 335.9 269.5 2.6 25.9 510 130.3 Japan (b) 1,956.6 1,929.6 -27.0 126.5 15.5 62,490 31.3 Jordan (a) 132.5 n.a. n.a. 6.5 20.4 n.a. 244.7 Kazakhstan (c) 104.5 149.8 45.3 16.3 6.4 1,760 59.4 Korea, Rep, (c) 898.0 n.a. n.a. 46.5 19.3 20,518 43.8 Kuwait 170.0 120.0 -50.0 1.9 89.6 456 373.2 Kyrgyzstan (c) 23.5 n.a. n.a. 4.7 5.0 356 66.0 Notes: Data are in millions of minutes of publzc sw=tched traffic a. International minutes based on billing point of traffic. b. Internatmnal traffic for year end=ng 31 March, 2000. Australia, Mauritius, Pak=stan ends 30 June. c. Traffic data exclude some carriers or routes. (See count~, table for details.) d. Data are for 1998. © TeleGeography, Inc. 2000 146 © TeleGeography, Inc. 2000 TeleGeography 2001 Figure 3. International Traffic Indicators, 1999 (continued) International Traffic Outgoing Incoming Balance Population Minutes (Out) (m minutes) (m minutes) (m minutes) (m) per Capita Latvia (a) 55.6 90.0 34.4 2.4 22.8 Luxernbourg 319.1 277.5 -41.5 0.4 743.4 Macau (a) 132,8 97.7 -35.1 0.4 303.3 Macedonia (c) 82.3 152.5 70.3 2.0 40.9 Malaysia (a, b) 690.0 n.a. n.a. 21.8 31.6 Malta 39.0 50.2 11.2 0.4 101.1 Mauritius (b) 31.4 43.3 11.9 1.1 27.3 Mexico (a) 1,563.0 4,007.5 2,444.5 97.4 16.1 Mold ova (a) 49.0 101.1 52.1 4.4 11.2 Morocco 219.5 n.a. n.a. 27.9 7.9 Mozambique 20.3 38.8 18.5 19.3 1.1 Myanmar 17.4 29.8 12.4 45.1 0.4 Narnibia 61.2 51.2 -10.0 1.7 36.1 Netherlands 2,150.0 n.a. n.a. 15.8 135.7 New Zealand (b) 815.0 n.a. n.a. 3.8 212.9 Nicaragua (a) 52.0 72.7 20.7 4,9 10.5 Norway (a) 567.0 386.9 -180.1 4.4 127.5 Ornan (a, c) 101.3 83.4 -17.9 2.5 41.2 Pakistan (b, c) 75.1 644.9 569.8 134.5 0.6 Palestinian Authority (c) 34.9 n.a. n.a. 3.1 11.3 Panama (a) 53.6 95,8 42.2 2.8 19.1 Paraguay 34.7 54.8 20.1 5.4 6.5 Peru (a) 88.9 302.6 213.7 25,2 3.5 PhilippineS,(a, b) 218.0 n.a. n.a. 74,5 2.9 Poland (a) 624.0 n.a. n.a. 38,7 16.1 Portugal (c) 532.8 753.3 220.5 10,0 53.4 Qatar 128.5 84.0 -44.5 0,6 218.1 Russia (c) 928.2 929.3 1.1 147,2 6.3 Saudi Arabia 1,060.0 n.a. n.a. 20,9 50.7 Senegal 36.5 111.1 74.7 9,2 3.9 Singapore (a, b) 1,350.0 n.a. n.a. 3.2 418.6 Slovak Republic (a) 162.8 208.7 45.9 5.4 30.3 Slovenia (d) 129.6 137.0 7.4 2.0 65.0 South Africa 461.1 n.a. n.a. 39.9 11,6 _Spain (a) 1,935.0 n.a. n.a. 39.4 49.1 Sfi Lanka 45.5 n.a. n.a. 18.6 2.4 Sudan (c) 21.9 105.3 83.3 28.9 0.8 Swaziland (b) 29.3 n.a. n.a. 1.0 29.9 Sweden (a) 1,365.0 n.a. n,a. 8.9 154.0 Switzerland 2,730.0 n.a. n.a. 7.1 382.1 Syria (a) 125.6 256.7 131.1 15.7 8.0 Taiwan (a) 949.3 882.0 -67.3 22.1 42.9 Thailand (a, c) 298.7 327.8 29.1 60.9 4.9 Trinidad & Tobago (a, b) 67,2 158.8 91.6 1.3 52.2 Turkey (c) 698.4 1,122.7 424.3 68.2 10.2 Turkmenistan (c) 16.5 n.a. n.a. 4.4 3.8 Ukraine 359.2 n.a. n.a. 50.7 7.1 United Arab Emirates 963.0 n.a. n.a. 2.4 401.6 United Kingdom (a, b) 10,141.0 6,853.4 -3,287.6 58.7 172.6 United States (a) 29,608.8 10,640.8 -18,968.0 276.2 107.2 Uruguay 80.1 98.3 18.2 3.3 24.2 Uzbekistan (a) 68.5 75.0 6.6 23.9 2.9 Venezuela (a) 160.2 315.3 155.2 23.7 6.8 Yugoslavia (a) 227,0 498.8 271.7 10.6 21.3 Zirnbabwe (b) 65.6 59.0 -6.6 11.5 5.7 Main Lines Minutes (Out) (thous.) per Main Line 732 76.0 311 1,026.3 178 744.3 471 174.7 4,431 155.7 198 197.5 257 122.2 10,927 143.0 555 88.2 1,467 149.7 78 260.4 249 69.9 108 565.7 9,610 223.7 1,877 434.2 147 354.3 3,165 179.1 220 459.5 2,986 25.2 n.a. n.a. 462 115.9 297 116.8 1,689 52.7 2,940 74.1 10,068 62.0 4,230 126.0 155 829.5 30,388 30.5 n.a. 347.4 ’166 219.9 1,861 725.6 1,655 98.4 757 171.2 5,493 83.9 16,480 117.4 679 66.9 251 87.3 31 959.8 5,889 231.8 4,992 546.9 1,600 78.5 12,044 78.8 5,216 57.3 276 243.9 18,054 38.7 359 46.0 10,074 35.7 975 987.5 33,750 300.5 188,331 157.2 897 89.3 1,599 42.8 2,586 61.9 2,281 99.5 239 274.5 Notes: Data are m miJhens of minutes of pubhc switched traffic. a International minutes based on billing point of traffic. b. International traff=c for year ending 31 March, 2000. Australia, Mauritius, and Pakistan ends 30 June c. Traffic data exclude some careers or routes. (See country table for details.) d. Data are for 1998. © TeleGeography, Inc. 2000 ~47 TeleGeography 2001 © TeleGeography, Inc. 2000 International Traffic by Route Figure 1. The Top 50 International Routes, 1999 Countries Minutes Each Way Total Minutes 1. U.S.- Canada 4,491.3 -- 3,925.0 8,416.3 2. U.S. Mexico 4,084.3 -- 1,364.6 5,449.0 3. U.S.- U.K. 1,909.6 -- 1,610.9 3,520.5 4. Hong Kong- China 1,263.7 -- 1,020.0 2,283,7 5. U.S.- Germany 1,525.3 -- 455.0 1,980.3 6. U.K.- Germany 913.3 -- 605.0 1,518.3 7. U.K.- Ireland 809.3 -- 700.0 1,509.3 8. U.S.- Japan 874.2 -- 406.4 1,280,6 9. Germany- Switzerland 650.0 -- 620.0 1,270.0 10. U.K.- France 655.8 -- 555.0 1,210.8 11. Austria- Germany 625.0 -- 555.0 1,180.0 12. France- Germany 565.0 -- 540.0 1,105,0 13. U.S.- India 996.4 -- 59.9 1,056.4 14. Germany- Italy 590.0 -- 465.0 1,055.0 15. U.S.- Italy 772.0 -- 210.0 982.0 16. U.S.- France 626.3 -- 350.0 976.3 17. Netherlands- Germany 450.0 -- 440.0 890.0 18. U.S.- Philippines 773.4 -- 55.0 828.4 19. Singapore- Malaysia 465.0 -- 350.0 815.0 20. U.S.- Brazil 623.5 -- 183.6 807.1 21. Fran’ce- Italy 400.0 -- 370.0 770.0 22. U.S.- Dominican Republic 626.8 -- 137.1 763.9 23. Switzerland- France 420.0 -- 330.0 750.0 24. France- Belgium 405.0 -- 345.0 750.0 25. U.K.- Spain 435.2 -- 300.0 735.2 26. U.S.- Australia 420.5 -- 300.0 720.6 27. Germany- Poland 445.0 -- 240.0 685.0 28. Netherlands- Belgium 340.0 -- 325.0 665.0 29. U.K.- Italy 396.6 -- 260.0 656.6 30. U.K.- Australia 325.6 -- 325.0 650.6 31. New Zealand Australia 365.0 -- 280.0 645.0 32. Switzerland- Italy 370.0 -- 270.0 640.0 33. Germany-Turkey 420.0 -- 204.9 624.9 34. France- Spain 315.0 -- 300.0 615.0 35. Spain- Germany 320.0 -- 280.0 600.0 36. U.K.- Netherlands 348.1 -- 250.0 598.1 37. U.S.- Israel 362.6 -- 200.0 562.6 38. U.S.- Korea, Rep. 351.4 -- 210.0 561.4 39. Canada U.K. 295.0 -- 247.9 542.9 40. U.S.-Taiwan 333.4 -- 162.0 495.5 41. Taiwan- China 286.9 -- 205.0 491.9 42. U.S.- Hong Kong 242.1 -- 240.3 482.4 43. Russia- Ukraine 264.1 -- 215.4 479.5 44. U.S.- China 416.0 -- 50.0 486.0 45. U.S.- Netherlands 284.3 -- 135.0 419.3 46. U.S.- Spain 325.2 -- 90.0 415.2 47. Japan- China 285.1 -- 130.0 415.1 48. U.S.- Jamaica 364.3 -- 44.5 408.8 49. U.K.- Belgium 234.6 -- 155.0 389.6 50. Sweden- U.K. 200.0 -- 186.3 386.3 Note: All data m milhons of minutes of telecommumcations traffic The country whzch generates more traffic on each route ~s listed f~rst The routes hsted above total 55.5 bilhon minutes, equal to 51 percent of all international trafftc. Data for Austra ha, Hong Kong, Ireland, Japan, Malaysia, New Zealand, Singapore, and the U.K. are for f=scal year 1999/2000 148 © TeleGeography, inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Figure 2. Traffic Imbalances on Selected U.S. Routes, 1999 U.S. Canada U.S. Umted Kingdom U.S. Mexico U S Brazil U.S. Italy U.S. Dominican Republic U.S Jordan U.S. Morocco U.S. Russia U.S. Poland U.S. Suriname U.S. Sri Lanka U.S. Cuba U.S. Ecuador U.S. Ph~hppmes U.S. Indza U.S Ukraine U S. Ha~= U S. Pakistan 1:1 5:1 10:1 15’1 20:1 25:1 30 1 35:1 Ratio of Outgoing to Incoming Traffic Note: Court ~tr~ywJth traffic deficit on route listed first. A ra~o of 1:1 would indicate a perfect balanee on a route, U.S. data baaed on bitling point of call, and may not reflect actual call ratios due to refi|e and call-bac}~ © Te]eGeography, Inc. 2000 Figure 3. Traffic Imbalances on Selected Non-U.S. Routes, 1999 Germany-Turkey Japan China France Morocco Taiwan Thailand Austral=a Indonesia Mexico Cuba Taiwan Phihppines Jordan Egypt Japan Phihppmes Malaysia Bangladesh India Pak=stan Saud= Arabia o Egypt Spain Cuba Canada India Israel- Ukraine Italy Ch=le U.K. Pakistan U.A.E. Pakistan Saudi Arabia Pakistan 1:1 31 6:1 9:1 12:1 15:1 Ratio of Outgoing to Incoming Traffic Note: Country with traffic deficit on route I~sted f~rst: A ratio of 1:1 would indicate a perfect balance on a route. Data for some countries based on bilhng point of cal~, and may not reflect actual eaJl ret~os due to refile and cal,-back. © Tele6eography, Inc. 2000 149 TeleGeography 2001 © TeleGeogrophUo Inc. 2000 140% ::1 120% t-,-. 100% 80% 60% 150 International Traffic Statistics TeleGeography 2001 © TeleGeography, Inc. 2000 Albania Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. Italy ............................ 26.8 2. Greece .......................... 24.9 3. Germany ......................... 4.4 4. United KinGdom ................... 2.5 5. United States ..................... 2.4 6. Switzerland ....................... 1.8 7. YuGoslavia ........................ 1.6 8. Turkey ........................... 1.2 9. France ........................... 1.2 10. Macedonia ....................... 1.1 11. Belgium .......................... 1.1 12. Austria ........................... 6.6 13. Netherlands ...................... 0.5 14. Canada .......................... 6.4 15. Spain ............................ 6.3 16. Croatia ........................... 6.3 17. Bulgaria .......................... 6.3 18. Romania ......................... 6.2 19. Hungary .......................... 0.2 26. Denmark ......................... 6.2 Other ............................ 1.3 Percentage of Outgoing Traffic ~® 3.4% 3.2% 2.5% 2.2% 1.6% 1.6% 1.5% 1.5% 0.8% 0.7% 0.5% 0.4% 0.4% 0.4% i 0.3% 0.3% ! 0.3% 1.7% Total ............................ 74.6 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 in..c0ming .................... ...................... n.a.~ .................. 93.~ ................. !2...!_:7 ....... 0 u~.g..°..!..n g ............................................. ~0.~ ................ ~....9.:.2. ................. 74,6 .~.[P I U,§..(.O.efi c i~) ........................ n:a: .............. 4~:..4. ..................... Total Volume n.a. 142.8 196.3 Note: Data are in millions of minutes of outgoing public switched telecommunicattons traffic. Data based on billing point of traffic. 152 ©’feleGeograph~, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Algeria Largest Telecommunications Routes, 1998 Destination Minutes (millions) 1. France .......................... 62.0 2. United Kingdom ................... 7.9 3. italy ............................. 5.5 4. Morocco ......................... 5.1 5. Spain ............................ 4.7 6. Germam~, ......................... 4.0 7. Canada .......................... 3.9 8. United States ..................... 3.4 9. Belgium .......................... 3.3 10. Switzerland ....................... 2.5 11. United Arab Emirates ............... 1.9 12. Saudi Arabia ...................... 1.7 13. Syr!a ............................ 1.3 14. Egypt ............................ 1.2 15. Lebanon ......................... 1.1 16. Netherlands ...................... 1.1 17. Sweden .......................... 6.4 18. Greece ........................... 0.2 19. Denmark ......................... 6.2 20. Australia ......................... 0.1 Other ....................... ’ ..... 9.8 Percentage of Outgoing Traffic 4.5% ~i 3.9% 3.3% 3.2% 2,8% 2.1% 1.6% 1.4% 1.1% ~, 1.0% o.9% ~i o.9% I o.2% Total ........................... 121.3 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 !ncomi,Bg ...................................... n,:ff. ¯ ................ n.a. ....... ...... n.~.,, ,. ...... O ut..g...0...!.n g ............................................... n.a.. ................ !2....I..:.3.. ........ ~:a.. ...... S u~rp lo~..s... (D efi..€.!t) ........................ n,:~: ................... n.a.’ ................ n.._a: .... Total Volume n.a. n.a. n.a. Note; Data are in rn~ll~ons of minutes of outgoing public switched telecommunicatmns traffic, 1999 traffic data are not available. Data exclude cross-border traffic to Tunzs=a. © TeleGeography, Inc. 2000 155 TeleGeography 2OO1 © TeleGeogrophg, Inc. 2000 Andorra Largest Telecommunications Routes, 1999 Destination 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Minutes (millions) Spain ........................... 37.0 France ........................... 9.0 Portugal .......................... 2.5 United Kingdom ................... 1.1 Belgium .......................... 0.3 6ermany ......................... 0.3 Denmark ......................... 0.2 Italy ............................. 0.2 Netherlands ...................... 0.2 Switzerland ....................... 0.2 United States ..................... 0.2 Other ............................ 2.0 Total ............................ 53.2 Percentage of Outgoing Traffic ~~~ii~~~i 69.5% 16.9% 4.7% ~2.1% ! 0.6% i 0.6% ! 0.4% 0.4% 0.4% 0.4% 0.4% 3.8% © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ..... O ut~g~!~.g ......................................... 42.~ .................. 47.4 ........... ~..~2 ..... ¯ S.urp~!us(Defi~it~! ............................. (12.,.!) .......... (,!,,.5,,:,2,),, ............. n:a-. ...... Total Volume 72.3 79.6 n.a. Note: Data are m mdlions of minutes of outgoing pubhc switched telecommunications traffic. 154 © TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Angola Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic 2. South Africa .5.1 ~~~ 14.5% 3. France ........................... 1.6 4.6~/o 4. United Kingdom ................... 1.4 ~i 4.0% 5. United States ..................... 1.3 ~3.6% 6. Namibia .......................... 1.0 ~3,0% 7. Brazil ............................ 1.0 2.9% 8. Germany ......................... 0.7 2.0% 9. Belgium .......................... 0.6 ~1.7% 10, Gambia .......................... 0.6 ~;1.6% 11. Netherlands ...................... 0.5 1.3% 12. Spain ............................ 0.4 1.2% 13. Zimbabwe ........................ 0.4 ~i 1,1% 14. Switzerland ....................... 0.4 1,0% 15. Guinea ........................... 0.3 1.0% 16. Mali ............................. 0.3 ~0.9% 17.~ Italy ............................. 0.3 0.9% 18. Congo, Dem. Rep ................... 0.3 ~0.8% 19. Nigeria ........................... 0.2 ~0.6% 20. Russia ........................... 0.2 ~! 0.6% Other ............................ 4.0 ~~~i 11.6% Total ............................ 35.0 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 Outgoing ....................... 21.9 u.r...P. !..US (Deficit) Total Volume 40.7 49.6 68.0 Note: Data are in m~ll~ons of minutes of outgoing pubJic switched telecommunications traffic. © TeleGeography, Inc. 2000 155 TeleGeography 2OO1 © TeleGeography, Inc. 2000 Argentina Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. United States .................... 70.8 2. Uruguay ......................... 46.0 3. Brazil ........................... 43.6 4. Chile ............................ 32.2 5. Paraguay ........................ 28.9 8. Bolivia .......................... 25.8 7. Spain ........................... 25.3 8. Italy ............................ 17.2 9. Peru ............................ 14.7 19. Mexico .......................... 8.9 11. France ........................... 7.9 12. Germany ......................... 8.3 13. United Kingdom ................... 5.8 14. Colombia ......................... 4.2 15. Venezuela ........................ 4.9 16. Canada .......................... 3.9 17. Switzerland ....................... 2.3 18. China ............................ 1.9 19. Dominican Republic ................ 1.2 20. Cuba ............................ 1.1 Other ........................... 28.7 Percentage of Outgoing Traffic 12.2% ~~%~ 8.5% ~~~ 7.7% ~~~ ~.7~o 2.1% ~.~% 1.5% ~.~% 1.1% 1.0% 0.8% 0.5% 0.3% 0.3% Total ........................... 377.6 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ...... .S..u_~P !..U..S._ ! D....e.f..i Cit) ...................................... 220.8 ............ n:~.. Total Volume 667.6 n.a. n.a. Note: Data are in mdhons of minutes of outgoing public switched telecommumcat=ons traffic. 156 © TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeo~raphy 2001 Armenia Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. Russia .......................... 24.1 2. Ukraine .......................... 2.3 3. Georgia .......................... 1.5 4. United States ..................... 1.1 5. Belarus .......................... 0.4 6. Germany ......................... 0.4 7. 6reece ........................... 0.4 8. France ........................... 0.3 9. I{azakhstan ....................... 0.3 10. Iran ............................. 0.2 11. United Kingdom ................... 0.2 12. Uzbekistan ....................... 0.2 13. Tu~ey ........................... 0.2 14. United Arab Emirates ............... 0.2 15. Turkmenistan ..................... 0.1 16. Netherlands ...................... 0.1 17. Bulgaria .......................... 0.1 18. Belgium .......................... 0.1 19. Italy ............................. 0.1 20. Poland ........................... 0.1 Other ............................ 1.5 Percentage of Outgoing Traffic 4.4% 3.3% 1.1% 1.o% 1.o% 0.9% i 0.6% 0.5% i o.5% 1 0.5% ! 0.5% I 0.4% i 0.4% i 0.3% 10.3% 0.3% ! 0.3% 4.5% Total ............................ 33.7 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ...... I n c 0_,m,,,i n,.,g" ...................................... £.a., ................ 94.0, .............. 8~.8 ¯ O,u,tg 0,i,# g ............................................... 48,;~ ............ 56:6 ................... 3,3...~ ,_, ~urp,[#S (.D, ofi£!,t,)~ ................................ n.,:a,,., ..................... 3,7-4 ............. Total Volume n.a. 150.7 123.5 Note: Data are m millions of mznutes of outgoing public switched telecommunications traffic. TeleGeography, Inc. 2000 157 TeleGeo{~raphy 2001 © TeleGeography, Inc. 2000 Australia Largest Telecommunications Routes, FY 1999/00 Destination Minutes (millions) Percentage of Outgoing Traffic 1. United Kingdom .325.0 ~~~~ 15.4% 2. United States ................... 300.0 ~~~ 14.2% 3. New Zealand .280.0 ~~~ 13.2% 4. Japan ........................... 90.0 ~,~ 4.3% 5. Singapore ....................... 90.0 4.3% 6. Indonesia ....................... 85.0 ~4.0~ 7. Hong Kong ...................... 75.0 3.5% 8. China ........................... 55.0 2.6% 9. Canada ......................... 45.0 2.1% 10. Philippines ....................... 45.0 2.1% 11. Taiwan .......................... 45.0 2..1% 12. Germany ........................ 40.0 1.9% 13. Malaysia ......................... 40.0 1.9% 14. Argentina ....................... 40.0 1.9% Other .......................... 560.0 ~~~~t ~~ 26.5 % Total ......................... 2,115.0 © TeleGeography, Inc. 2000 National Traffic Balance Minutes FY 1997/98 FY 1998/99 FY 1999/00 ........ !,nc~in.g ................................. !,~50.~ ..................... n.a:~ ................. no.,.a. ~ogtg0~!~g .......................... ~,.~10:0 ......... !~.0, 2,115.0 S.~,~plus, (,D ~!Fit) ........ (2,~0:~!, ,,n~:.~.~ ............. Total Volume 2,760.0 n.a. n.a. Note: Data are in millions of minutes of outgoing public switched telecommunications traffic. Fiscal year ends 30 June. 158 © TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Austria Largest Telecommunications Routes. 1999 Destination Minutes (millions) 1. Germany ....................... 625.0 2. Italy ............................ 80.0 3. Yugoslavia ....................... 60.0 4. Hungary ......................... 53.0 5. United Kingdom .................. 40.0 6. Poland .......................... 30.0 7. Netherlands ..................... 34.0 8. Turkey .......................... 34.0 9. Croatia .......................... 32.0 10. United States .................... 31.0 11. France .......................... 30.0 12. Czech Republic ................... 30.0 13. Sl~vak Republic .................. 20.0 14. Romania ........................ 19.0 15. Slovenia ......................... 19.0 Other .......................... 207.0 Percentage of Outgoing Traffic ~44~ 3.0% 2.7% 2.5% 2.5% 2.4% 2.3% 2.2% 2.2% 1.5% 1.4% 1.4% Total ......................... 1,350.0 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 Total Volume 1,953.2 n.a. n.a. Note: Data are zn millions of minutes of outgoing public sw=tched telecommumcatmns traffic Traffic figures exclude most cross-bor- der traffzc to Swrtzefland © TeleGeography, Inc. 2000 159 TeleGeography 2001 Aze rb aij an Largest Telecommunications Routes, 1999 Destination © TeleGeographg, Inc. 2000 Minutes (millions) 1. Russia .......................... 17.3 2. Turkey ........................... 3.1 3. Ukraine .......................... 1,8 4. Georgia .......................... 1.3 5. Iran ............................. 0.9 6. United Kingdom ................... 0.8 7. Kazakhstan ....................... 0.8 8. United States ..................... 0.6 9. Germany ......................... 0.5 10. United Arab Emirates ............... 0.5 11. Belarus .......................... 0.4 12. Uzbekistan ....................... 0.4 13. TurkmeniStan ..................... 0.3 14. France ........................... 0.2 15. Switzerland ....................... 0.1 16. Italy ............................. 0.1 17. Moldova ......................... 0.1 18. Netherlands ...................... 0.1 19. Norway .......................... 0.1 20. Poland ........................... 0.1 Other ............................ 2.7 Percentage of Outgoing Traffic 5.6~ 4.0~ 2.8% 2.5% 1.8% ~.9% :0.3% 0.3% 0.3% 0.3% ,0.3% 0.3% 8.~% Total ............................ 32.2 © Tele6eography, In c. 2000 National Traffic Balance Note: Data are zn milhons of minutes of outgmng pubhc switched te]ecommumcat~ons traff=c. Data based on billing point of traffic. 160 ©TeleGeograph~, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Bahamas Largest Telecommunications Routes, 1998 Destination Percentage of Outgoing Traffic ~~~~,~,~ 78.3% 2. 5.5% 3. 2.3% 4. ii~ 2.2% 6. 7. 8. 9. 10. 12. 13. 15. 17. 18. 20. Minutes (millions) United States .................... 49.8 Canada .......................... 3.5 Jamaica ......................... 1.5 United Kingdom ................... 1.4 Switzerland ....................... 0.6 Haiti ............................. 0.4 Turks & Caicos Islands ............. 0.4 Mexico .......................... 0.4 France ........................... 0.4 Germany ......................... 0.4 Trinidad & Tobago ................. 0.2 Cayman Islands ................... 0.2 Ita’ly ............................. 0.2 Brazil ............................ 0.2 Spain ............................ 0.2 Bermuda ......................... 0.2 Dominican Republic ................ 0.1 China ............................ 0.1 Colombia ......................... 0.1 Guyana .......................... 0.1 Other ............................ 3.3 0.9% 0.6% 0.6% o.6~o 0.6% 0.6% 0.3% 0.3% 0.3% 0.3% 0.3% 0.2% 0.2% O.2% 0.2% 0.2% 5.2% Total ............................ 63.5 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ...... !nco~i~g ............................... .Q..a., ........ ~,9~g£ing .................... .. .. 62.7 .................... 63.~ ..... S u,rp!,u,.s, (D,e,,f!c!t) ............................. n,,,:,a,.~ ........... ,.2,&5 Total Volume n.a. 153.5 n.a. Note: Data are ~n millions of minutes of outgoing public switched tetecommunications traffic. 1999 traffic data are not available. ©TeleGeography, Inc. 2000 161 TeleOeography 2001 © TeleGeography, Inc. 2000 Bahrain Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic 1. India .27.4 ~~~~~,~%~i~ 20.4% ........................... 2. Saudi Arabia .17.6 ~~~~ 13.1% 3. United Arab Emirates .............. 14,9 11.1% 4. United Kingdom ................... 8.~ ~~,~s.4% Pakistan .8.3 ........................ Kuwait .5.8 .......................... 4.~% United States ..................... 5.0 ~3,7% E~ ............................ 4.~ ~.~ Qatar ............................ 4.5 ~3.3% Philippines ....................... 3.5 2.6% Oman ............................ 2.2 Bangladesh ....................... 2.1 1.5% Jordan .’ .......................... 2.0 1.5% Morocco ......................... 1.8 ~.4% Sri Lanka ......................... 1.7 1.2% Syria ............................ 1.2 0.9% Lebanon ......................... 1.0 0.8% France ........................... 1.0 O.8% Yemen ............................ 0.9 0.7% Germany ......................... 0.8 0.fi% Other ........................... 21.2 ~~~~ 15.8% 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. Total ........................... 134.1 © TeleGeography, n’c.,200D National Traffic Balance Minutes 1997 1998 1999 Out_go!~B. ................................................ ~!,0~:6 ........ !~4.4 ......... ~4.~!~ ..... Surp!us (.D~fic i~) ........................ (2!.2). ............. (22.~3_) ..................... Total Volume 192.0 226.5 240.6 Note: Data are in millions of minutes of outgoing public switched telecommunications traffic. Data based on billing point of traffic. © TeleGeograph~, Inc. 2000 © TeleCeography, Inc. 2000 TeleGeography 2OO1 Bangladesh Largest Telecommunications Routes, 1998 Destination Minutes (millions) 1. india ............................. 7.1 2. United Kingdom ................... 5.1 3. United States ..................... 5.0 4, Saudi Arabia ...................... 3.5 5, Singapore ........................ 2.8 8. United Arab Emirates ............... 2.4 7. Hong Kong ....................... 1.8 8. Pakistan ......................... 1.6 9. Malaysia ......................... 1,3 10. Korea, Rep ........................ 1.2 11. China ............................ 1.0 12. Italy ............................. 0.9 13. J~pan ............................ 0.9 14. Canada ......................... ,.0.9 15. Germany ......................... 0.8 16. Thailand ......................... 0.7 17. France ........................... 0.7 18. Australia ......................... 0.6 19. Taiwan ........................... 0.6 20. Indonesia ........................ 0.3 Other ............................ 2.7 Total ............................ 41.8 © TeleGeographv, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ... O..utg.o...i n g .......................................... 46.:..9 .................. 41.:.8. ...... S u..r.pl ~§ (D .ef!..c it) ........................ 140.~" ................. ~4.4 .................. n.:a.. Total Volume 233.9 237.9 n.a. Note: Data are in rnilhons of mznutes of outgoing public switched telecommunications traffic. 1999 traffic data are not available. © Tel~Geography, Inc. 2000 TeleGeography 2001 © TeleGeogr~zph~l, Inc. 2000 ( Belarus Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. Russia .......................... 91,8 2. Ukraine ......................... 21.9 3. IVloldova ......................... 1.9 4. Kazakhstan ....................... 1.7 5. Azerbaijan ........................ 0.9 6, Armeni~ .......................... 0.9 7. Uzbekist~n ....................... 0.7 8. Georgia .......................... 0.6 9. Kyrgyzstan ....................... 0.2 10. T~jikistan ......................... 0.2 11. Turkrnenistan ..................... 0.2 Other ........................... 40.2 Total ........................... 161.2 Percentage of Outgoing Traffic ~~~~ 13.6% 12% 1.1% 0,6% 0.6% 0.4% 0.4% 0.1% 0.1% 0.1% ©TeleBeography, inc. 2000 National Traffic Balance Minutes 1997 1998 1999 _!~€.~oming ....... 1.85:~ ......... 193.~ ......... !95.~6 Total Volume 333.8 369.6 356.8 Note: Data are in m~llions of minutes of outgoing pubhc switched telecommumcat~ons traffic. Data based on bdhng point of traffic. The "Other" category may include routes to non-members of the Commonwealth of Independent States that rank among the top rlesti- nations for outgoing traffic. © TeleGeograph~, Inc. 2000 164 © TeleGeography, Inc. 2000 TeleGeography 2001 Belgium Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1, France ......................... 345.0 2. Netherlands .................... 325.0 3. Germany ....................... 190.0 4. United Kingdom ................. 155.0 5. Italy ............................ 80.0 6. Luxembourg ..................... 55.0 7. United States .................... 55.0 8. Spain ........................... 52.0 9. Switzerland ...................... 34.0 10. Sweden ......................... 20.0 Other .......................... 279.0 Percentage of Outgoing Traffic ~! ~..-~ ~..- 9.7°/o ~i~ ~.OO/o 3.5°/o 1.3% Total ......................... 1,590.0 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ..... !nco~!n~g ............... 1,420.0 n.a. ¯,,~.a., ....... ...... u rP lU,,S~,(D ef!,9i~), ..i ....... .................. 80"0. .................. n.a,.,. ................ n.a.~, Total Volume 2,760.0 n.a. n.a. Note: Data are m mflhons of mznutes of outgmng pubhc switched telecommunications traffic. Data based on bilhng pmnt of traffic © TeleGeography, Inc. 2000 165 TeleGeography :2001 © TeleGeographg, Inc. 2000 ( Benin Largest Telecommunications Routes, 1999 Destination Minutes (thousands) 1. France ........................ 2,840.0 2. "]’ogo ......................... 1,296.0 3. CSte d’lvoire ................... 1,102.0 4. Nigeria ......................... 776.0 5. Niger .......................... 515.0 6. United States ................... 513.0 7. Gabon ......................... 424.0 8. Senegal ........................ 397.0 9. Burkina Faso .................... 338.0 10. Germany ....................... 269.0 11. United Kingdom ................. 205.0 12. Belgium ........................ 200,0 13. Cameroo~ ...................... 196.0 14. Ghana ......................... 180.0 15. Mall ........................... 138.0 16. Italy ........................... 137.0 17. Switzerland ..................... 132.0 18, Canada ........................ 111.0 19. Netherlands ..................... 83.0 20. Lebanon ......................... 64.0 Other .......................... 695.0 Percentage of Outgoing Traffic ~ill~~~ 27.1% ~~~ 1o.5% ~~ 7.4% 2.6% 1.9% 1.9% 1.7% 1.3% 1.3% 1.3% 1.1% 0.8% 0.6~ ~~J~ 6.6% Total ........................ 10,495.0 ©TeleGeography, Inc. 2000 National Traffic Balance Note: Data are in mdhons of m~nutes of outgoing public switched telecommumcations traffic. 166 © TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2OO1 Bolivia Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. United States ..................... 6.6 2. Argentina ........................ 5,3 3. Peru ............................. 3.5 4. Rr8zil ............................ 3.4 5. Chile ............................. 3.0 8. Italy ............................. 0.6 7. Colombia ......................... 0.6 8. Cuba ............................ 0.5 9. Ecuador .......................... 0.5 10. Spain ............................ 0.5 11. Germany ......................... 0.5 12. Paraguay ......................... 0.4 13. Ca’nada .......................... 0.4 14. Mexico .......................... 0.4 15. Venezuela ........................ 0.4 16. United Kingdom ................... 0.4 17. Japan ............................ 0.3 18. China ............................ 0.2 19. Uruguay .......................... 0.2 20. France ........................... 0.2 Other ............................ 1.8 Percentage of Outgoing Traffic ~!~ 17.7°/o ~~~ 1o.2% 2.1% 2.1% 1.7% 1.6% 1.5% 1.4% 1.4% 1.3% 1.2% 0.9% 0.8% 0.7% 0.7% Total ............................ 29.7 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ~ln~om~ng ...................................... ~.~3 ............... 76.~4 ............... ~.2 Q~_~ o.!.n..9 ............................................ 22:7_ ...................... ~1 ..~ ............. 29:~ Total Volume 92.0 108.0 111.9 Note: Data are ~n millions of minutes of outgoing pubhc switched telecommunications traffic TeleGeography, Inc. 2000 167 TeleGeography 2001 © TeleGeographg, Inc. 2000 Bosnia-Herzegovina Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic ......................... i~- ~i~ ~. o 1. Croatia .27.8 ~% 28.6~ 2. Germany .17.9 ! 18.5% ....................... ~_ 3. Yugoslavia ....................... 11.3 ~~~ 11.6% 4. Slovenia ......................... 7.5 ~i@~ 7.7% 5. United States ..................... 6.2 ~~ 6.3% 6. Austria ........................... 5.7 5.9% 7. Switzerland ....................... 3.5 ~3.6% 8. Italy ............................. 2.8 ~2.9% 9. Hungary .......................... 2.4 2.4% 10. Sweden .......................... 2.3 2.4% 11. United Kingdom ................... 1.8 1.9% 12. Netherlands ...................... 1.8 1.8% 13. France .’ ........................... 1.7 ~1.8% 14. Turkey ........................... 1.4 1.4% 15. Macedonia ....................... 1.1 1.2% 16. Norway .......................... 1.0 1.0% 17. Czech Republic .................... 0.3 0.3% 18. SlovakRepublic ................... 0.2 i0.2% Other ..0.6 0.6% Total ............................ 97.2 © TeleGeography, Inc, 2000 National Traffic Balance Minutes 1997 1998 1999 ..... !nF0,~i~g .............................. ~,n,.. a: ............... 159:~2_ ............ 2~,0:5 ~, ,,Q,~t,g~i~,,g,. ............................ .............. 6§o.~ ................ 94.9 .............. 97~.~ .......... ..... ,..S. U rp.l,~.F (D~it) .............................. p.~a: ........... ~;3 .............. ~q~.~3 ..... Total Volume n.a. 254.1 297.7 Note: Data are in millions of minutes of outgoing public switched telecommunications traffic. Data exclude cross-border traffic to Albania. 168 © TeleGeographY, Inc.2000 © TeleCeographg, Inc. 2000 TeleGeography 2001 Largest Telecommunications Routes, 1999 Brazil Destination 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. Minutes (millions) United States ................... 183.6 Argentina ....................... 43.1 Portugal ......................... 26.3 Italy ............................ 21.5 Spain ........................... 20.9 United Kingdom .................. 20.7 Germany ........................ 18.9 France .......................... 14.3 Japan ........................... 14.1 Uruguay ......................... 11.4 Chile ............................ 10.6 Paraguay ........................ 10.6 Canada .......................... 9.6 Switzerland ....................... 8.6 Mexico .......................... 6.7 Netherlands ...................... 5.0 Bolivia ........................... 5.0 Peru ............................. 3.6 Israel ............................ 3.3 Venezuela ........................ 2.8 Other .......................... 135.0 Percentage of Outgoing Traffic ~~ 7.5% 4.6% 2.5% 2.0% 1.8% 1.8% 1.6% 1.5% 1.2% 0.9% 0.9% 0.6% 0.6% o.5% Total ........................... 574.8 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 Inc0m!ng ............................ 7~!,.~ .............. ~06~.9 0.utg,,0,.,!n g ,. 459.1 Total Volume 1,220.4 1,352.7 1,413.3 Note= Data are in millions of minutes of outgoing public switched telecommunications traffic. Data based on bilhng point of traffic. © TeleBeography, Inc. 2000 169 TeleGeography 2001 © TeleGeographgo Inc. 2000 ( Brunei Largest Telecommunications Routes, 1999 Destination Minutes (thousands) 1. SinGapore ..................... 3,352.0 2. Malaysia ...................... 3,261.0 3. Philippines .................... 2,690.0 4. Indonesia ..................... 1,997.0 5. United Kingdom ................ 1,885.0 6. Australia ....................... 975.0 7. Thailand ........................ 948,0 8. India ........................... 824.0 9. United States ................... 326.0 10. HonG KonG ...................... 251.0 11. NewZealand ................... 200.0 12. Nepal .......................... 169.0 13. ,Japan ..’ ........................ 164.0 14. Canada ........................ 118.0 15. Taiwan .......................... 87.0 16. France .......................... 83.0 17. Germany ........................ 83.0 18, China ........................... 76.0 19. Saudi Arabia ..................... 71.0 20. United Arab Emirates .............. 61.0 Other ......................... 1,141.0 Percentage of Outgoing Traffic 17.9% ~%~’~ 10.6% 5.1% 1.7% 1.1% 0.9% 0.9% 0.6% 0.5% 0.4% 0.4% 0.4% 0.4% 0.3% ~~7.1% Total ........................ 18,763.0 © TeleGeegraphy, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ....... !~o,.€,o~!ng ............................................. .~2..o ................ 25.5, ............. Qu~,go,!ng ............................................. 3#:6, ................. ,2,3.4 ,I,.,,,....S. U [£!..u.s,~..(O e....f.,i c !t) ............................ (]3.6) ....................... 2,,1, ................ ,,2,..9 Total Volume 57.6 48.9 40.4 Note: Data are in millions of minutes of outgoing public switched telecommumcat~ons traffic. Data exclude traffic originated on mobile telephones. © TeleGeographv, Inc. 2000 © TeleGeograph~l, Inc. 2000 TeleGeography 2001 Bulgaria Largest Telecommunications Routes, 1999 Destination 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. Minutes (millions) Greece .......................... 15.0 Germany ........................ 13.0 Turkey .......................... 13.0 Yugoslavia ........................ 9.0 United States ..................... 7.0 Italy ............................. 5.0 France ........................... 4.0 Romania ......................... 4.0 Russia ........................... 4.0 United Kingdom ................... 4.0 Macedonia ....................... 4.0 Austria ........................... 3.0 Ut~raine .......................... 3.0 Netherlands ...................... 2.0 Czech Republic .................... 2.0 Other ............................ 7.0 Total ............................ 98.9 © TeleGeography, Inc~ 2000 National Traffic Balance Note: Data are in milhons of minutes of outgmng pubhc sw=tched telecommumcations traffic. © TeleGeography, Inc. 2000 171 TeleGeography 2001 Burkina Faso © TeleCeograph~l, Inc. 2000 Largest Telecommunications Routes, 1999 Destination Minutes (thousands) Percentage of Outgoing Traffic ....................... ~. ~,~. ~,~ 1. France .2,754.7 ~: 25.2% .................. ,~ 2. CGte d’lvoire .2,459.6 22.5~o .......................... ~, 4. United States .596.7 ~~ 5.5% 5. Australia ....................... 539.9 4.9% 6. Senegal ........................ 460.1 4.2% 7. Mall ........................... 404.9 3.7% Niger .......................... 310.1 2.8% Benin .......................... 297.5 2.7% United Kingdom ................. 287.4 2.6% Belgium ........................ 263.5 2.4% Ghana ......................... 201.0 1.8% Italy ...’ ......................... 163.1 ~1.5% ~ermany ....................... 155.7 1.4% Swi~erland ..................... 127.0 1.2% Canada ........................ 110.3 1.0% Nigeria .......................... 90.7 0.8% Gabon .......................... 82.4 0.8% Netherlands ..................... 78.4 0.7% Papua New Guinea ............... 78.4 0.7% ~~~ 7 ~% Other .......................... 800.0 ( .... 9. 10. 12. 13. 15. 16. 17. 18. 20. Total ........................ 10,942.0 © TeleGeography, Inc. 2000 National Traffic Balance Note: Data are in millions of minutes of outgoing public switched telecommunications traff=c. 172 ©TeleGeography, Inc. 2000 © TeleGeograph~j, Inc. 2000 TeleGeography 2001 Burundi Largest Telecommunications Routes, 1999 Destination 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. Minutes (thousands) Belgium ........................ 571.0 France ......................... 252.0 Kenya ........................... 222.0 Italy ........................... 112.0 Canada ........................ 110.0 U nited States ................... 108.0 United Kingdom ................. 101.0 Switzerland ...................... 88.0 Tanzania ........................ 73.0 South Africa ..................... 72.0 Germany ........................ 59.0 Netherlands ..................... 58.0 Greece .......................... 47.0 Uganda ......................... 36.0 Rwanda ......................... 32.0 Senegal ......................... 28.0 COte d’lvoire ..................... 24.0 Ethiopia ......................... 22.0 China ........................... 21.0 Cameroon ....................... 17.0 Other .......................... 420.0 Percentage of Outgoing Traffic ¯ % ~’~’~" ~-, o 9.0% ~i~ 4.6% ~~ 4.5% ~~ 4.4 4.~% ~.0% 2,4% 1.5% 1.3% 1.1% 1.0% 0.9% 0.9% 0.7% ’%~ .... ~, ,,~,. Total ......................... 2,458.0 © TeleBeography, Inc, 2000 National Traffic Balance Minutes 1997 1998 1999 !ncq,.m, ing ................................ n.¯.a,,,~. .................. 3:,6, ............... ,,,,.O.,,.u t g,,o.,,!,£,g ............................... n.a.~ ................ ,,,,2.4 .......................... 2,,.:5 S,u,,,,,r,p! us (O,#,fiQ,t) .................. n.~,: .............. 1:,! .............. Total Volume n.a. 6.0 5.9 Note: Data are in millions of minutes of outgoing pubhc switched telecommunications traffic. Data based on bllhng point of traffic. ©~JeGeograph~ Inc.2000 173 TeleGeography 2001 Canada Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic 1. United States .................. 3,925.0 ~~~~ 69.1% 2. United Kingdom ................. 295.0 5.2% 3. Hung Kon9 ...................... 100.0 1.8% 4. Germany ........................ 85.0 1.5% 5. France .......................... 80.0 ~1.4% 6. India ............................ 70.0 1.2% 7. Italy ............................ 55.0 ~1.0% 8. Australia ........................ 45.0 0.8% 9. Philippines ....................... 45.0 0.8% 10. Japan ........................... 42.0 ~0.7% 11. Netherlands ..................... 33.0 0.6% 12. Taiwan .......................... 33.0 0.6% 13. Vietnam i ......................... 33.0 ~0.6% 14. Jamaica ......................... 27.0 ! 0.5% 15. Korea, Rep ....................... 27.0 i 0.5% 16. Mexico ......................... 27.0 i0.5% 17. Switzerland ...................... 27.0 i 0.5% 18. Sri Lanka ........................ 27.0 ~0.5% Other .......................... 704.0 12.4% © TeleC, eogrophy, Inc. 2000 Total ......................... 5,680.0 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ...... Su r~!~o.s. (~fici~) .................... 3~.8 ................... n.~ ................... n~:~ Total Volume 8,921.4 n.a. n.a. Note= Data are in millions of minutes of outgoing public switched telecommunications traffic. Data based on bdhng point of traffic. 174 © TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Central African Rep. Largest Telecommunications Routes, 1999 Destination Minutes (thousands) 1. France ........................ 1,901.0 2. Cameroon ...................... 446.0 3. Chad ........................... 248.0 4. C~te d’h/oire .................... 214.0 5. Senegal ........................ 179.0 6. United States ................... 143.0 7. Lebanon ........................ 114.0 8. Togo ............................ 92.0 9. Belgium ......................... 87.0 10. Gabon .......................... 74.0 11. Egypt ........................... 52.0 12. Benin ........................... 49.0 13. Mall ............................ 46.0 14. Burkina Faso ..................... 43.0 15. Italy ............................ 43.0 16. Canada ......................... 36.0 17. Germany ........................ 34.0 18. Nigeria .......................... 33.0 19. Switzerland ...................... 33.0 20. Morocco ........................ 26.0 Other .......................... 205.0 Percentage of Outgoing Traffic ~~ ~0.~% 4.9% ~4.1% 2.6% 2.1% 2.0% ~.7% 1.1% 1.1% 1.0% 1.0% 0.8% 0.8% 0.8% 0.8% 0.6% Total ......................... 4,330.0 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 I nco,,m_,ing" ..................................... ,,n.a.,. .............. 3;,.6~ ................... n.a; ....... ,,~£,.U,,tgo, ing ............................................. 3,:6 .................... ,3:5 .............. ,,.4:,3 Surp!,£S,’ (D,,,,efic,,!~) ........................... n.a: ................... 0.~! .............. n.a: Total Volume n.a. 7.1 n.a. Note: Data are in millions of mznutes of outgoing pubhc switched telecommunications traffic. © TeleGeography, Inc, 2000 175 TeleGeography 2001 © TeleGeography, Inc. 2000 Chad Largest Telecommunications Routes, 1999 Destination Minutes (thousands) Percentage of Outgoing Traffic 1. France .1,095.0 o ~’,~ ,,~®~ ~, 38.6% ....................... 2. Cameroon ...................... 441.0 ~~ 15.5% 3. Saudi Arabia .................... 274.0 ~~ 9.7% 4. United States ................... 145.0 5.1% 5. CentralAfrican Rep ............... 109.0 3.8% 6. C~te d’lvoire ..................... 97.0 ~,i~ 3.4% 7. Ni0eria .......................... 87.0 3.1% 8. Sudan .......................... 70.0 2.5% 9. Egypt ........................... 67.0 2.4% 10. Canada ......................... 60.0 2.1% 11. Senegal ......................... 56.0 ~2.0% 12. Belgium ......................... 55.0 1.9% 13. Benin .o’ .......................... 49.0 1.7% 14. Germany ........................ 46.0 1.6% 15. Gabon .......................... 39.0 1.4% 16. Switzerland ...................... 34.0 1.2% 17. Burkina Faso ..................... 34.0 ~! 1.2% 18. Italy ............................ 32.0 1.1% 19. Togo ............................ 28.0 1.0% 20. Libya ........................... 19.0 0.7% Other .......................... 137.0 4.8’/o Total ......................... 2,837.0 © TeleGeography, Inc, 2000 National Traffic Balance Minutes 1997 1998 1999 ....... Lncom~..ng .......................... ..q.~... .......... .n:~. ..................... ~n:a.~ ........ ¯ O~tgoing, 2.8 3.2 2.8 ,,,Su~plus (Deficiti .............. n.a. n.a. n.a. Total Volume n.a. n.a. n.a. Note: Data are ~n millions of m~nutes of outgoing public swztched telecommunicatmns traffzc. Data based on billing point of traffic. 176 © TeleGeograph~, Inc. 2000 © TeleGeographg, Inc. 2000 TeleGeo~raphy 2001 Largest Telecommunications Routes, 1999 Chile Destination 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. Minutes (millions) United States ................... 100.0 Argentina ....................... 31.0 Spain ........................... 14.3 Peru ............................ 10.7 Brazil ..................... ..... 10.0 Germany ......................... 6.5 Canada .......................... 6.2 Japan ............................ 6.0 Bolivia ........................... 5.4 France ........................... 4.4 Mexico .......................... 4.3 United Kingdom ................... 3.7 E~uador .......................... 3.5 Venezuela ........................ 3.5 Italy ............................. 3.4 Sweden .......................... 3.1 Colombia ......................... 2.2 Australia ......................... 2.0 Uruguay .......................... 1.9 Paraguay ......................... 1.4 Other ........................... 46.5 Percentage of Outgoing Traffic ~~o~ .......... ~,~ 37.0% ~~ 11.5% ~ii 5.3% ~i~ 4.0% 2.4% 2.3% 2.2% 2.0% 1.6% 1.6% 1.3% 1.3% 1.3% 1.1% 0.8% 0.7% 0.7% 0.5% ~7.2% Total ........................... 270.0 © TeleSeography, Inc, 2000 National Traffic Balance Minutes 1997 1998 1999 ¯ ,,!,ncom~i~g ..................................... n.~a. ,,. ~..u...t....g~o, in,g_ .................................. 2.42.0 ......... ~59.4. ..Surplus. (Oe.f.!€,,it) .......................... n..a: Total Volume n.a. n.a. n.a. Note: Data are =n millions of m~nutes of outgoing pubhc switched telecommumcations traffic. ©TeleGeography, Inc. 2000 177 TeleGeography 2OO1 © TeleGeograph~l, Inc. 2000 ( China Largest Telecommunications Routes, 1999 Destination 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. Minutes (millions) Hon9 Kon9 .................... 1,020.0 Taiwan ......................... 205.0 Japan .......................... 130.0 United States .................... 50.0 Korea, Rep ....................... 45.0 Macau .......................... 40.0 Singapore ....................... 35.0 Australia ........................ 18.0 Canada ......................... 18.0 Germany ........................ 18.0 United Kingdom .................. 16.0 France .......................... 12.0 Italy ...’ .......................... 10.0 Malaysia ......................... 9.0 Russia ........................... 9.0 Other .......................... 315.0 Percentage of Outgoing Traffic 10.5% 6.~% 2.6% 2.3% 2.1% 1.8% 0.9% 0.9% 0.9% 0.8% 0.6% 0.5% 0.5% 0.5% ~~ 16.2% Total ......................... 1,95Q.0 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 Total Volume n.a. n.a. n.a. Note: Data are =n milhons of minutes of outgozng pubhc switched telecommumcations traffic. 178 ©.TeleGeograph~,, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Costa Rica Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. United States .................... 34.7 2. Nicaragua ....................... 19.0 3. Panama .......................... 5.9 4. Mexico .......................... 5.6 5. Guatemala ........................ 4.6 6. El Salvador ....................... 3.8 7. Honduras ......................... 2.8 8. Colombia ......................... 2.6 9. Canada .......................... 1.6 10. Spain ............................ 1.3 11. Italy ............................. 1.1 12. Cuba ............................ 1.0 13. 6e’rmany ......................... 1.0 14. Venezuela ........................ 0.9 15. Dominican Republic ................ 0.9 16. Puerto Rico ....................... 0.8 17. Peru ............................. 0.8 18. Argentina ........................ 0.7 19. Chile ............................. 0.6 20. Brazil ............................ 0.6 Other ............................ 3.9 Percentage of Outgoing Traffic ~~~~ 36 8% 6.3% ~.~% ~i 4.~% ~4.1% 2.8% 1.7% 1.4% 1.2% 1.0% 1.0% 1.0% 0.8% 0.~% 0.7% o.~% Total ............................ 94.1 © TeleGeography, Inc, 2000 National Traffic Balance Minutes 1997 1998 1999 Inco..m.,!ng .................................... !.!1:6 ............ !.!2~9 ........... !09.:.0., %,t,g°iqg ............................................. 6,,6,,9 ................ 82.7. ................ 94.! ...... ..... Surp,!,.u.,s (p,,ef!cit) ................ ,44.? ........... 3,0,.2 ............ !.4.9 ,,, Total Volume 178.5 195.6 203.1 Note: Data are in millions of minutes of outgmng publzc switched telecommumcations traffic. Data based on bilhng point of trafhc. © TeleGeography, Inc. 2000 ¯ ......... TeleGeography 2001 © TeleGeographM, Inc. 2000 ( C6te d’Ivoire Largest Telecommunications Routes, 1998 Destination Minutes (millions) 1. France .......................... 17.3 2. Senegal .......................... 4.4 3. United States ..................... 3.6 4. Italy ............................. 3.1 5. Burkina Faso ...................... 2.3 6. Mali ............................. 2.2 7. Nigeria ........................... 2,1 8. Lebanon ......................... 1.8 9. United Kingdom ................... 1.6 10. Benin ............................ 1.4 11. Togo ............................. 1.2 12. Gabon ........................... 1.1 13. Belgium ’. .......................... 1.0 14. Germany ......................... 1.0 15. Congo, Rep ........................ 0.9 16. Cameroon ........................ 0.9 17. Switzerland ....................... 0.9 18, Ghana ........................... 0.8 19. Morocco ......................... 0.8 20. Canada .......................... 0.8 Other ............................ 8.1 Percentage of Outgoing Traffic ~~~ 30.2% ~1~ 7"6°/° ~~ 6.2% 2.1% 2.0% 1.8% 1.7% 1.6% 1.6% 1.6% 1.4% 1.3% 1.3% ~~~ 14.2% Total ............................ 57.3 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1’999 ¯ I_~,c. q,,m,, ing .............................................. 5,~,.6 46.6 ...... n ~.a~.~ ......... 0,, ,U.,tg o!,,,n,.,g ............................... ~O.! ............ 5,,7,:3 ...................... .... Su~p!us (D,~f!cit~) .................................... ,1~.5 ................... (10.7,! ........................ n,,.,,a,. ,,, Total Volume 90.7 103.9 n.a. Note: Data are in rndhons of rnmnutes of outgoing public switched telecommumcat~ons traffic. 1999 route data are not available. 180 © TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Largest Telecommunications Routes, 1999 Cuba Destination 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. Minutes (millions) United States ..................... 6.7 Spain ............................ 6.2 Canada .......................... 4.6 Italy ............................. 4.3 Mexico .......................... 2.7 France ........................... 1.0 Brazil ............................ 0.7 Colombia ......................... 0.7 Argentina ........................ 0.6 Panama .......................... 0.6 Germany ......................... 0.6 Venezuela ........................ 0.6 united Kingdom ................... 0.5 Chile ............................. 0.5 Dominican Republic ................ 0.5 Puerto Rico ....................... 0.4 Switzerland ....................... 0.3 Costa Rica ........................ 0.2 Barbados ......................... 0.1 dapan ............................ 0.1 Other ............................ 0.5 Percentage of Outgoing Traffic ~~ ~4.~% @~~i ~.40/o 2.3% 2.O0/o 1.9% 1.80/o 1.7% 1.6% 1.6% ~.6% 1.3% 0.9% 0.7% o,4% 0.3% Total ............................ 32.6 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 .... !,~com,!ng ................................... 16~:2 ......... 2,0,3.0 ...... ,,.2,,2~.3 0U,,,tg0in~_ ..................................... 27.,8. ............... 29:,,0. ................... 32,.,6 ,, ,S,urp~u,~ (0.e, f.!..€,it) ...................... 13.3.,,4 ........... !.7,,4..0,.. ............ Total Volume 189.0 232.1 257.8 Note: Data are =n mdlions of minutes of outgoing pubhc switched telecommumcatzons traffzc. © TeleGeography, Inc. 2000 181 TeleGeography 2001 © TeleGeography, Inc. 2000 Cyprus Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1, Greece .......................... 46.8 2. United Kingdom .................. 36.5 3. Egypt ........................... 11.8 4. Lebanon ......................... 10.4 5. Russia ........................... 7.3 6. Germany ......................... 4.8 7. United States ..................... 4.0 8. Bulgaria .......................... 4.0 9. Romania ......................... 3.5 10. Ukraine .......................... 3,1 11. Yugoslavia ............. ........... 2.9 12. Italy ............................. 2.4 13. France .’ .......................... 2.1 14. Syria ............................ 2.1 15. Sweden .......................... 1.8 16. Switzerland ....................... 1.8 17. Netherlands ...................... 1.8 18. israel ............................ 1.6 19. Belgium .......................... 1.1 20. Australia ......................... 1.0 Other ........................... 17.5 Percentage of Outgoing Traffic 2.8% 2.1% 1.8% 1.7% 1.4% 1.3% 1.3% 1.1% 1.0% 1.0% 1.0% 0.7% 0.6% 10.4% Total ..... ..................... 168.2 © TeleBeography, inc. 2000 National Traffic Balance Minutes 1997 1998 1999 0,...U..t.g oin,g, .......................................... ,!,.5.4:..4, ........ 182:,.0, ............. ...... S u r,p.,[..U.S (D..efic!,t..! .......................... (39.,2). ................... (..6...I.,,.4) ................ Total Volume 269.6 302.7 302.3 Note: Data are in millions of minutes of outgoing public switched telecommunications traffic. Data based on billing point of traffzc. 182 © TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeo~raphy 2001 Czech Republic Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic 1. Germany ........................ 90.0 ~~~~~~ 24.7% 2. Slovak Republic .75.0 ~~~~il 20 6% ................ :~1~i %. ¯ ~, 3. Austria .26.0 ~’.....~:~ 7.1% 4. United Kingdom .................. 20.0 ~~ 5.5% 5. Poland .......................... 15.0 ~4.1% 6. Italy ............................ 13.0 3.6% 7. France .......................... 12.0 3.3~/o 8. United States .................... 11.0 3.0% 9. Netherlands ..................... 10.0 2.7% 10. Ukraine ......................... I0.0 2.7% Other ........................... 82.0 22.5% Total ........................... 364.0 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 In,~P~m~ ! ~ng ......................... 3~.0 .............. ~06:~ ....... 45~:2 ..... S~[~p~lus. (Deficit) ................. 48~ ........... 89.5 ...... 8~;2 ..... Total Volume 661.1 724.4 816.2 Note: Data are in millions of m~nutes of outgoing pubhc switched telecommumcations traffic Data based on billing po=nt of traffic. © TeleGeography, Inc. 2000 185 TeleGeography 2001 © TeleGeographu, Inc. 2000 ( Denmark Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. Germany ....................... 135.0 2. Sweden ........................ 124.0 3. United Kingdom ................. 100.0 4. Norway ......................... 76.0 5. United States .................... 37.0 6. Netherlands ..................... 34.0 7. France .......................... 30.0 8. Italy ............................ 24.0 9, Spain ........................... 17.0 10. Switzerland ...................... 17.0 11. Belgium ......................... 16.0 12. Finland .......................... 15.0 13. Poland .’ .......................... 14.0 14. Faroe Islands .................... 12.0 15. Turkey ........................... 9.0 Other .......................... 140.0 Percentage of Outgoing Traffic ~~i 3.8% 3.0% 2.1% ~.z~, 2.1% 2.0% o 1.9~ 1.1% Total ........................... 800.0 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 In£,0, min, g ..................... 68.2.0 ............... ~.~a; ................. ,n, .a,~ .......... ¯ _,0,, ut, g,0.,!n gv ................................ 60,.7.5 .......... 7!,0.0 ............... #OO:0, S,u, rP!Us (Deficit,,) ........ ...................... 7,4,:,5 ......................... #:.a,,: ........ .,n..a; ....... Total Volume 1,289.5 n.a. n.a. Note: Data are in millions of minutes of outgoing public switched telecommunications traffic Data based on billing point of traffic. 184 ©TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Dominican Republic Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic 1. United States .137.1 ~~~~ 73,8% 2. Spain ............................ 7.1 ~3.8% 3. Italy ............................. 4.4 2.3~o 4. Canada .......................... 2.7 1,4% Germany ......................... 2.6 1,4% Cuba ............................ 2.1 ~.~% Venezuela ........................ 2.1 1.1% Mexico .......................... 1.8 1.0% France ........................... 1.5 0.8% Swi~erland ....................... 1.4 0.8% Netherlands Antilles ............... 1.2 0.7% Haiti ............................. 1.2 O.6% Argentina ........................ 1.2 0.6% Colombia ......................... 1.2 0.6% Panama .......................... 1.0 O.5% United Kingdom ................... 0.9 ~0.5% Netherlands ...................... 0.7 0.4% Costa Rica ........................ 0.7 0.4% Chile ............................. 0.4 0.2~o Brazil ............................ 0.3 0.2% Other ........................... 14.3 7.7% 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. t6. 17. 18. 19. 20. Total ........................... 185.7 © Tele6eography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 l.n..€ o. m!n..g ..................................... ,.4..-/. 6.9 .............. .7.~.0.5 ...... 920.:0. Total Volume 618.9 888.0 1,105.7 Note: Data are ~n millions of minutes of outgmng public sw=tched telecommunications traffic. © Tele~3eegraphy, Inc. 2000 185 TeleGeography 2001 © TeleGeograpbgo Inc. 2000 ( Egypt Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. Saudi Arabia ..................... 28.9 2. Italy ............................ 13.3 3. United Kingdom .................. 12.6 4. United States .................... 12.0 5. United Arab Ernirates .............. 11.6 6. Germany ........................ 10.0 7. France ........................... 9,4 8. Kuwait ........................... 8.4 9. Lebanon ......................... 4.8 10. Yemen ........................... 3.7 11. Jordan ........................... 3.6 12. Canada .......................... 3.5 13. Switzerland ....................... 3.0 14. Netherlands ...................... 2.7 15. Syria ............................ 2.5 16. Greece ........................... 2.3 17. Spain ............................ 2.3 18, Libya ............................ 2.1 19. Qatar ............................ 2.1 20. Belgium .......................... 1.9 Other ........................... 30.2 Percentage of Outgoing Traffic ~~ ~.~% 1.4% 1.3% ~)~ 1.3% 1.2% 1.15 Total ........................... 171.0 © TeleGeegraPhy, inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ¯ ! ~nc~.~!ng ..................................... ~51.2 ......... 475.3 ......... 5~4.6 Total Volume 570.5 BO2.B 725.B Note: Data are in milhons of m~nutes of outgoing pubhc switched telecommunicabons traffic. 186 © TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 El Salvador Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. United States .................... 24.0 2. Guatemala ....................... 6.8 3. Honduras ........................ 2.9 4. Mexico .......................... 2.9 5. Costa Rica ....................... 2.8 6. Nicaragua ....................... 1.6 7. Canada .......................... 0.8 8. Panama ......................... 0.8 9. Colombia ........................ 0.4 10. Spain ........................... 0.4 11. Argentina ........................ 0.2 12. Brazil ........................... 0.2 13. Ch’ite ............................ 0.2 14. France .......................... 0.2 15. Germany ......................... 0.2 16. Peru ............................ 0.2 17. Venezuela ....................... 0.2 18. Italy ............................. 0.2 19. Dominican Republic ............... 0.1 20. United Kingdom ................... 0.1 Others ........................... 1.8 Percentage of Outgoing Traffic ~~ 14.5 6.0% 3.4% 1.7% 1.7% 0.9% 0.9% 0.4% 0.4% o.~% 0.4% 0.4% o.4% 0.4% 0.4% ~o.2% 0.2% 3.8% Total ............................ 47.0 © TeleGeography, Inc. 2000 National Traffic Balance Note: Data are =n mdlions of minutes of outgmng public switched telecommunicat=ons traffic. 1998 data are for ANTEL only. © TeleGeography, Inc. 2000 187 TeleOeography 2OO1 © TeleGeograph~l, Inc. 2000 Eritrea Largest Telecommunications Routes, 1999 Destination Minutes (thousands) Percentage of Outgoing Traffic 1. Italy ,408.1 ~~i~ 16.3% 2. SaudiArabia .................... 301.0 ~~~ 12.0% 3. United States ,289,6 ~~~~ 11.6% ................. ,~]~ ............... ,~ 4. United Kingdom ..158.6 6.3% United Arab Emirates ............. 148.0 ~~ 5.9% Germany ....................... 135.1 ~~ 5.4~ Egypt .87.4 ~~ 3.5% Korea, Rep ....................... 85.8 ~~ 3.4~o Kenya .80.1 3.2% Libya ........................... 61.6 2.5% Sudan ..53.4 2.1% Sweden ......................... 46.3 1.8% Swi~erland ...................... 41.0 1.6% Netherlands ..................... 39.9 1.6% Canada ......................... 34.4 1.4% China ........................... 32.0 1.3% Yemen .......................... 32.0 1.3% France .......................... 31.0 1.2% India ............................ 28.0 1.1% Denmark ........................ 26.3 1.0% Other .387.9 15.5% ......................... 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 17. 18. 19. 20. Total ......................... 2,507.6 © TeleGeography, In c. 2000 National Traffic Balance Minutes 1997 1998 1999 Total Volume o n.a. 15.7 16.3 Note: Data are zn millions of m~nutes of outgoing pubhc switched telecommumcat~ons traffic 188 © TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Estonia Largest Telecommunications Routes, 1999 Destination 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. Minutes (millions) Finland ........................... 21.0 Russia .......................... 14.6 Sweden .......................... 5.9 Latvia ............................ 5.3 Germany ......................... 4.6 Ukraine .......................... 3.1 Lithuania ......................... 2.9 United Kingdom ................... 2.2 Denmark ......................... 1.8 United States ..................... 1.4 Norway .......................... 1.2 Belarus .......................... 1.1 It~ty ............................. 0.9 Poland ........................... 0.9 Netherlands ...................... 0.9 France ........................... 0.8 Belgium .......................... 0.6 Spain ............................ 0.5 Switzerland ....................... 0.4 Austria ........................... 0.4 Other ............................ 4.1 Total ............................ 74.6 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ._!n, com.!,ng ....................................... ,67:p ........ 7,9:2. .............. 84.8~ 0 u...tg.~iq,g .................................. 66:3 ................. 75:1 .................. ..... ~9,,F,p I,..U.,§ £~efi~!t) .......................... Oo:~Z .......... 4-! .................. 10,2, Total Volume 133.3 154.3 159.4 Note: Data are in millions of m~nutes of outgoing pubhc switched telecommumcatzons traffic. Data based on bilhng pomt of traff=c. © Tel~Geography, Inc. 2000 189 TeleGeoi~raphy 2001 Ethiopia Largest Telecommunications Routes, 1999 Destination Minutes (thousands) Percentage of Outgoing Traffic 1. United States .................. 1,722.0 ~~i~i 13.6% 2. Djibouti .1,412.2 ...................... ¯ ~~~ 11.2% ................ 3. United Kingdom .931.4 ~~1 7.2% 4. Kenya .832.7 ~~ 6.4% Israel .723.1 ~~ 5.6% SaudiArabia .................... 703.8 ~~~ 5.6% 6ermany ....................... 506.3 4.0% Algeria ......................... 444.9 3.2% France ......................... 439.2 3.2% United Arab Emirates ............. 430.8 3.2% Switzerland .277.8 2.4% South Africa .................... 263.5 2.4% Sweden ’ .259.6 2.4% Netherlands .255.7 2.4% India ........................... 247.3 ~! 1.6% Yemen ......................... 189.0 1.6% Egypt .......................... 170.0 1.6% Canada ........................ 161.2 1,6% 6reece ......................... 124.8 ~.o~o Uganda ........................ 121.0 0.8% Other ......................... 2,237.3 19.2% 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18, 19. 20. © TeleGeographg, Inc. 2000 Total ........................ 12,453.4 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 0utgg!~ng ...................... .. ...... .. 10:~ ........... !3:7 .................... ~.5 Total Volume 50.3 n.a. 59.0 Note: Data are in mdhons of minutes of outgoing pubhc switched telecommumcat~ons traffic. 190 © TeleGeography, Inc. 2000 © TeleGeographg, Inc. 2000 TeleGeography 2001 Finland Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. Sweden ........................ 135.0 2. Germany ........................ 38.0 3. United Kingdom .................. 37.0 4. Estonia .......................... 31.0 5, Russia .......................... 28.0 6. Norway ......................... 18.0 7. United States .................... 16.0 8. Denmark ........................ 12.0 9. France .......................... 12.0 10. Netherlands ...................... 8.0 Other ........................... 89.0 Percentage of Outgoing Traffic 9.0Yo ~~ 7.3% 6.6% 4.2% 2.8% 1.9% ~’,~ ,,, ~, ~,,,,, 21.0% Total ........................... 423.9 © TeieGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 !£com!,,,ng ....................................... n.a; .............. ~.n,.a: ......................... n.~: O~u,,,~g,,o,. [,,n,~,.g, ......................................... 37,,,!:! ................... 4!.0:.8, ........ 4,,2,,3-9,,, S,# rp!,,u,, S, (D e[!cit) .......................... n.,a_: ............... n-a., .............. n.,~,: Total Volume n.a. n.a. n.a. Note: Data are in millions of m=nutes of outgmng pubhc switched telecommumcations traffic. 1999 data exclude an est=mated 25 mil- lion minutes outgoing Voice over IP traff=c. © Tel~Geography, Inc. 2000 TeleGeography 2001 f France © TeleGeography, Inc. 2000 Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic ...................... ~- .~ ."’ ¯ 1. Germany .5~5.0 11.4%o ................ ’~ ¯ 2. United Kingdom .555.0 ~~ 11.2% 3. Belgium ........................ 405.0 ~~~~~~l~’~~ 8.2°o 4. Italy ........................... 400.0 8.1% 5. United States ................... 350.0 ~~ 7.1% 6. Switzerland .330.0 6.7% ......................... ~-’,~ o 7. Spain .315.0 ~~ 6.4Yo "~%~’ ¯ o 8. Morocco ....................... 205.0 4.1~/o g. Portugal ........................ 185.0 ~~ 3.7% 10. Netherlands .................... 170.0 ~~ 3.4% 11. Algeria ......................... 180.0 ~3.2% 12. Canada ........................ 120.0 2.4% 13. Tunisia .’ ......................... 110.0 ~_~ 2.2% 14. Turkey .......................... 85.0 1.3% 15. Poland .......................... 59.0 ~1.2% 16. Luxembourg ..................... 55.0 1.1% 17. Sweden ......................... 40.0 ~i 0.8% 18.. Austria .......................... 34.0 0.7% 19. Greece .......................... 34.0 ~0.7% 20. Denmark ........................ 33.0 0.7% Other .......................... 700.0 15.4’/, Total ......................... 4,950,0 © TeleGeography, Inc, 2000 National Traffic Balance Minutes 1997 1998 1999 ....... Inc0,~g ................................... 3,609,:0 ..... n.a~: .............. ,,,,O,.,~tgoi~,g 3,545..0 ................... ~:~,!,,~:P .......... 4,,,,9~.0 Total Volume 7,154.0 n.a. n.a. Note: Data are in millions of minutes of outgoing public switched telecommumcations traff=c. © TeleGeography, Inc. 2000 192 © TeleGeography, Inc. 2000 TeleGeography 2001 French Polynesia Largest Telecommunications Routes, 1999 Destination Minutes (thousands) 1. France ........................ 9,380.0 2. United States .................. 1,507.0 3. New Caledonia .................. 806.0 4. New Zealand ................... 438.0 5. Australia ....................... 384.0 6. Italy ........................... 164.0 7. Cook islands .................... 156.0 8. Japan .......................... 133.0 9. United Kingdom .................. 99.0 10. R~union ......................... 95.0 11, Germany ........................ 90.0 12. China ........................... 79.0 13. Ph’ilippines ....................... 59.0 Other ......................... 1,240.0 Percentage of Outgoing Traffic lO.3% 5.5% 3.0% 2.6% 1.1% 0.9% 0.7% 0.6% o.6% 0.5% i 0.4% Total ........................ 14,630.0 © TeleBeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 !..n._.€...O..~i ng ....................... ~.a.. ............... n:a..: .............. n:.a. Outg0i,~,g ........... n:.~.: ........ 12.3 .................. 14._~ S....U.rP.!.Us (..DefiCit). ................ .~.a.. ............... n.a... ..................... Total Volume n.a. n.a. n.a. Note: Data are in mdllons of minutes of outgoing public switched telecommumcat~ons traffic © TeleGeography, Inc. 2000 195 TeleGeoi~raphy 2001 ( Georgia Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. Russia .......................... 28.6 2. Ukraine .......................... 3.6 3. Armenia .......................... 2.0 4. Azerbaijan ........................ 2.0 5. Kazakhstan ....................... 0.5 6. Belarus .......................... 0.5 7. Uzbekistan ....................... 0.3 8. Moldova ......................... 0.2 9. Turkmenistan ..................... 0.2 10. Kyrgyzstan ....................... 0.1 Other ............................ 8.8 © TeleGeographg, Inc. 2000 Percentage of Outgoing Traffic 4.3% 4.3% 1.1% 1.1% 0.6% 0.4% ! 0.4% Total ’ ....46.7 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 I ~C~O mj,,n, g, ........................................ ~5.9~ ............. 72:,0 ......... 6,5.7 Total Volume 104.4 n.a. 112.4 Note: Data are in millions of mznutes of outgoing public switched telecommumcations traffic The "Other" category may include routes to non-members of the Commonwealth of Independent States that rank among the top destinations for outgoing traffic. 194 @ TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Germany Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. SwitzerJand ..................... 650.0 2. United Kingdom ................. 605.0 3. Italy ........................... 590.0 4. Austria ......................... 555.0 5. France ......................... 540.0 6. United States ................... 455.0 7. Poland ......................... 445.0 8. Netherlands .................... 440.0 9. Turkey ......................... 420.0 10. Spain .......................... 280.0 11. Belgium ........................ 160.0 12. Denmark ....................... 140.0 13. Czech Republic .................. 140.0 14. Greece ......................... 135.0 15. Croatia ......................... 115.0 16. Yugoslavia ...................... 105.0 17. Hungary ......................... 86.0 18. Portugal ......................... 80.0 19. Romania ........................ 72.0 20. Sweden ......................... 67.0 Other .......................... 885.0 Percentage of Outgoing Traffic ~~~ ~.O’/o ~~~ 6 0% ~~~ ,~.o% ~2o% ~.o% 1.5% 1.o% ~.o% Total ......................... 6,965.0 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ,,, !ncq,~in.g ..................................... 5,6,18.0 .............. p.a: 0 y.tg .o.!.~g ................................. .4..,.8 !..3..0 ............... .5.,870.0 ...... .6..:96.5..0 ...... ..S u rP..! US.(D. efi €!.t..) ..................... 8...°5.0. ............ .n.a... ................ n:a.:... Total Volume 10,431.0 n.a. n.a. Note: Data are in mdlions of minutes of outgoing public switched telecommunicatmns traffic. Data based on bilhng poznt of traffic. © TeleGeography, Inc. 2000 195 TeleGeography 200| © TeleGeography, I~. 2000 Ghana Largest Telecommunications Routes, 1999 Destination Percentage of Outgoing Traffic 3. 6.3°/o 4. 4.0% 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18, 19. 20. Minutes (millions) United Kingdom .................. 12.2 United States ..................... 7.4 Germany ......................... 1.9 Italy ............................. 1.2 South Africa ...................... 1.0 Canada .......................... 1.0 France ........................... 1.0 Japan ............................ 0.7 Netherlands ...................... 0.7 Togo ............................. 0.5 Belgium .......................... 0.3 Benin ............................ 0.2 Egypt .’ ........................... 0.1 Burkina Faso ...................... 0.1 Liberia ........................... 0.1 Senegal .......................... 0.1 Zimbabwe ........................ 0.1 Niger ............................ 0.1 Kenya ............................ 0.1 Gambia .......................... 0.1 Other ............................ 1.5 3.5% 3.4% ~3.~% 2.4% 2.3% 1.5% 0.9% 0.5% 0.4% 0.4 ~/o 0.3% 0.3% O.3% 0.2% O.2% 0.2% Total ............................ 30.1 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 Iotal Volume 101 ,I 129.7 148.5 Note: Data are in millions of minutes of outgoing public switched telecommunications traffic 196 © TeleGeographi,, Inc. 2000 © TeleGeographg, Inc. 2000 TeleGeography 2001 Gibraltar Largest Telecommunications Routes, FY 1999/00 Destination Minutes (millions) Percentage of Outgoing Traffic 1. United Kingdom .7.0 429% 2. Spain ........................... .6.1 ~~~¯~ ~" ¯ ¯ ~’ ~¯ ¯~ 37.2% 3. Morocco ......................... 0.6 3.5% 4. United States ..................... 0.3 1.7% 5. Switzerland ....................... 0.3 1,6% 6. Germany ......................... 0.2 1.1%o 7. France ........................... 0.2 1.0% 8. Denmark ......................... 0.1 ~0.9% 9. Ireland ........................... 0.1 ~0.8% 10. Portugal .......................... 0.1 ~0.8% 11. Netherlands .0.1 ~0.7% 12. Russia ........................... 0.1 ~0.6% 13. Hong Kong ....................... 0.1 ~0.6% 14. Greece ........................... 0.1 ~0.5% 15. Italy ............................. 0.1 ~0.5% 16. India ............................. 0.1 ~0.4% 17. Belgium .......................... 0.1 ~0.4% 18. Canada .......................... 0.1 ~0.4% 19. Sweden .......................... 0.1 i0.4% 20. South Africa ...................... 0.1 0.3% Other ............................ 0.6 3.7% Total ............................ 16.4 © TeleGeography, Inc, 2000 National Traffic Balance Note: Data are in millions of m~nutes of outgmng public switched telecommunications traffic. © TeleGeography, Inc. 2000 ......... TeleGeography 2001 © TeleGeographg, Inc. 2000 Greece Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic 1. Germany .112.2 ~~~~ 15.5% 2. United Kingdom .106.5 14.7% ................ 3. Italy ............................ 59,9 4. United States .39.8 ~~ 5.5% 5. Albania .36,7 ~~ 5.1% 6. France .29.1 ~3~ no~ ......................... 7. Cyprus .......................... 27.8 ~3,8% 8. Bulgaria ......................... 24.6 3.4% 9. Romania ........................ 21.0 ~2.9% 10. Netherlands ..................... 17.9 2.5% 11. Belgium ......................... 14.3 ~2.0% 12. Sweden ......................... 12.8 1.8% 13. Swi~erland ...................... 12.8 1.8% 14. Ukraine ......................... 11.4 1.6% 15. Australia ........................ 10.1 1.4% 16. Yugoslavia ....................... 10.1 1.4% 17. Austria .......................... 10.0 1.4% 18. Russia .......................... 10.0 ~.4% 19. Turkey ........................... 9.6 1.3% 20. Canada .......................... 9.2 1.3% Other .......................... 139.9 19.3% Total ........................... 725.7 © TeleGeography, Inc. 2000 National T~affi¢ Balance Minutes 1997 1998 1999 ...... !nc~!n9 .............................. ~34.6 .......... 7~10.1 .......... 7~94=~ _ Total Volume 1,228.3 1,391.4 1,519.9 Note: Data are in millions of mznutes of outgoing pubhc switched telecommunications traffic. Data exclude cross-border traffic to Macedoma. 198 © TeleGeography, Inc. 2000 © TeleGeographu, Inc. 2000 TeleGeography 2001 Guatemala Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. United States .................... 38.8 2. Mexico ......................... 10.8 3. El Salvador ...................... 10.0 4. Costa Rica ........................ 4.8 5. Honduras ......................... 4.8 6. Nicaragua ........................ 2.7 7. Panama .......................... 1.6 8. Colombia ......................... 1.4 9. Canada .......................... 1.1 10. Spain ............................ 1.0 11. Korea, Rep ........................ 0.8 12. Italy ............................. 0.6 13. Germany ......................... 0.5 14. Venezuela ........................ 0.5 15. Argentina ........................ 0.5 16. Peru ............................. 0.4 17. Brazil ............................ 0.4 18. France ........................... 0.3 19. Belize ............................ 0.3 20. Dominican Republic ................ 0.3 Other ............................ 1.9 Percentage of Outgoing Traffic ~~~ 12.9% ~~~ 12.o% ~.8% 3.3% ~# 2.0% 1.7% 1.3% 1.2% 0.7% 0.7% 0.6% 0.6% 0.6% 0.5% 0.4% 0.4% 0.4% 0.4% 2.3% Total ............................ 83.3 © TeleGeography, Inc. 2000 National Traffic Balance Minutes FY 1997/98 FY 1998/99 1999 ..... !ncom!ong .......................... !65.0~, .............. ~....a: ........ 208.§ ~_u.t, going ................................................. ~8:.,5. .............. 60.0 .......... 8~.3 ¯ ,.~urpl,u,~ (D...efi~.!t) ..................... .1....!.§.5 ............. ~..a.. ............... ,!25.,3., Total Volume 213.5 n.a. 291.9 Note: Data are tn millions of minutes of outgoing public switched telecommumcat=ons traffic. © Tele~eography, Inc. 2000 199 TeleGeo~raphy 2001 © TeleGeography, I~¢. 2.000 ( Guyana Largest Telecommunications Routes, 1999 Destination Minutes (thousands) 1. United States .................. 7,590.1 2. Canada ....................... 2,087.9 3. Trinidad &Tobago .............. 1,558.5 4. Barbados ....................... 879.6 5. United Kingdom ................. 630.0 6. Suriname ....................... 279.0 7. ,Jamaica ........................ 241.7 8. Antigua & Barbuda .............. 206.1 9. Venezuela ...................... 100.5 10. Saint Kitts & Nevis ................ 87.4 11. Netherlands Antilles .............. 83.9 12. China ........................... 82.0 13. French Guiana ................... 75.4 14. Brazil ........................... 72.9 15. Grenada ......................... 70.2 16. Saint Vincent & the Grenadines ..... 68.2 17. Bahamas ........................ 46.9 18. Dominica ........................ 32.4 19. Turks & Caicos Islands ............ 28.6 20. Dominican Republic ............... 26.9 Other ......................... 1,812.8 Percentage of Outgoing Traffic ~,,,~ 13.0% N~ 9.7o/o ~i 5.5O/o 3.9% 1.7% 1.5% 1.3% 0.5% o.5"/. ~.5% o.5% o.5% 0.4% o.~% 0.3% 0.2% 0.2% 0.2% Total ........................ 16,061.0 © TeleGeography, tnc, 2000 National Traffic Balance Note: Data are in mflhons of minutes of outgmng public switched telecommumcatlons traffic. 200 © TeleGeography, Inc. 2000 © TeleGeographg, Inc. 2000 TeleGeography 2001 Hong Kong Largest Telecommunications Routes, FY 1999/00 Destination 1. 2. 3. 4. 5. 6. 7. 8. 9. Minutes (millions) China ......................... 1,263.7 United States ................... 240.3 Philippines ...................... 189.1 Canada ........................ 174.3 United KinGdom ................. 125.0 Taiwan ......................... 112.7 Australia ....................... 106.4 Japan ........................... 80.0 Singapore ....................... 79.5 Other .......................... 349.3 Percentage of Outgoing Traffic ~~ 8.8% ~1 3.9% 2.9% ~,’--"~ 2.9% Total ......................... 2,720.3 © Tele(~eography, Inc. 2000 National Traffic Balance ........ @in~,!,,e,s .................................... FY ,!997./98, ..... ,EY.199,8199 .... Inc.~minog. ...................................... 2,1.00.3 .......... !.;83.3.-~ ........ !,747;7 ..... .... ,,,O,,,,Ut90..~Ln, 9 .................... !,,.7!_8;~ ...... 1,,.879.8, ......... 2,~20-3 S,,u,,,r~,,!us (D.,~Ti~tii,+,,’~ ...................... 78,2.7_ ............ (4~,.8,1 ............... (973.~,!1 ....... Total Volume 3,818.3 3,712.8 4,407.5 Note: Data are in millions of m=nutes of outgoing pubhc switched telecommunications traff=c. Data based on billing point of traffic. Fiscal year ends 31 March. TeleGeography, Inc. 2000 201 TeleGeography 2001 © TeleGeography, Inc. 2000 t Hungary Largest Telecommunications Routes, 1999 Destination 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Minutes (millions) G e rrnany ........................ 88.5 Austria .......................... 39.2 Romania ........................ 30.0 Yugoslavia ....................... 19.5 Italy ............................ 16.8 United Kingdom .................. 16.1 United States .................... 14.1 France .......................... 11.0 Netherlands ...................... 8.0 Switzerland ....................... 7.6 Other ........................... 93.1 Percentage of Outgoing Traffic ~~~~~.~’~,~,~ ,~ ~.~ "~ ’~" ", 25.7% ~~® ~.7O/o ~@~~ 5.7% ~4,1% 3.~% 2.3% Total ’ .343.9 © TeleGeogrephy, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 J£c o,,m,,i,n,g .................................. ,32,4.~6 ........ 3~,4,.:5 ................ O,.u,.tgo[n,,g .......................................... 2,.,8,.7,:1 ................ 296.3 ............. #43,..#, ..... # u rp.,I,u.,s,.,(B e~!,c it) ........................... 37,..5 ........... 78:~ ............ ~£,.,a., ......... Total Volume 611.7 670.8 n.a. Note: Data are in milhons of minutes of outgoing public switched telecommunications traff=c Data based on billing point of traffic. © TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 India Largest Telecommunications Routes, F"Y 1999/00 Destination Minutes (millions) Percentage of Outgoing Traffic 1. SaudiArabia ..................... 91.2 ~~~~ 19.3% 2. United States .59,9 ~~~@~ 12.7% ................... 3. United Arab Emirates .............. 42.2 ~~ 8.9% 4. United Kingdom .................. 42,0 ~~ 8,9% ...................... ~,~i ~:’ "" 4 50 5. Singapore .21.2 i ¯ Vo 6. Kuwait .......................... 16.1 3.4% 7. Oman .15.4 3.2% 8. Germany ........................ 14.3 ~i" 3.0’/o 9. Canada .......................... 9.2 1.9% 10. Sri Lanka ......................... 9.2 1.9% 11. Australia ......................... 8.7 ~1.8% 12. Hong Kong ....................... 8.4 1.8% 13. Japan ............................ 8.2 1.7% 14. France ........................... 8.1 1.7% 15. Italy ............................. 7.6 1.6% 16. Malaysia ......................... 6.6 1.4% 17. Qatar ............................ 6.3 1.3% 18. Pakistan ......................... 5.0 1.1% 19. Bangladesh ....................... 4.5 t~ 1.0% 20. Philippines ....................... 4.5 ~i 0.9% Other .......................... .84.7 ’’ ’ ,~ ,~ ~~ ~’ ~’"- ~’~;~:~’~ 17.9~° Total ........................... 473.3 © TeleGeography, Inc, 2000 National Traffic Balance Minutes F’Y 1997/98 F’Y 1998/99 FY 1999/00 ........ !~co~!~g ............................. !,,2,5. 6.6 .............. 1,;~98,.~ ...... !,~772.~¯ O,ut, g,,°,~in~g, .......................... .... ~20,.5 ............. 43,6.:2 ............... 4,7.~3~.3 ..... Su[plu§ (D e~!,cit) .............................. 836,..1 ............. 1,~62:~ ........ !¯~,~99.2 .... Total Volume 1,677.1 1,9:35.0 2,245.8 Note: Data are =n millions of minutes of outgoing public switched telecommunications traffic. Data based on bdhng pmnt of traffzc. Rscal year ends 31 March. Data exclude some cross-border traffic w~th Bangladesh, Nepal, and Pakistan. © TeleGeography, Inc. 2000 ......... TeleGeography 2001 © TeleGeographu, Inc. 2000 ( Indonesia Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. Singapore ....................... 62.8 2. Malaysia ........................ 29.2 3. Australia ........................ 27.8 4. United States .................... 23.6 dapan ........................... 20.3 Taiwan .......................... 11.6 Hong Kong ....................... 9.2 Korea, Rep ........................ 9.0 United Kingdom ................... 7.8 India ............................. 6.4 China ............................ 5.8 Thailand ......................... 5.2 Germany’, ......................... 5.0 Philippines ....................... 4.6 Netherlands ...................... 4.0 France ........................... 3.3 Brunei ........................... 2.2 Italy ............................. 2.0 Canada .......................... 1.9 Switzerland ....................... 1.2 Other ........................... 26.6 8. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. Percentage of Outgoing Traffic ~~~ 7.5% 4.3% 3-4% 3.3% 2.9% ~L 2.4 ~o o ~.7 ~o 1.5% 1.2% 0.8% o.7~o 0.7% 0.5% Total ........................... 269.6 © TeleGeography, Inc. 2000 National Traffic Balance Note: Data are in millions of minutes of outgoing pubhc switched telecommumcations traffic. Data based on billing point of traffzc. 1999 outgoing traffic data exclude at least 50 million minutes of calls sent illegally over international private hnes. ©TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleOeography 2001 Iran ¯ Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic ............. ~. .~ ~.’. ~,..~ .~. 1. United Arab Emirates .23.1 11.5% 2. Germany ........................ 19.2 i~/~~ 9.6% ................... ’~ ~"~ ~.-’." 3. United States .19.1 ~i,~ii~~ 9.5% 4. Pakistan ......................... 16.0 ~~,~ 8.0’/o 6. United Kingdom .10.2 ~~ 5.1% 7. Kuwait ........................... 9.9 4.9’/o 8. Sweden .......................... 6.8 ~3.4% 9. Japan ............................ 4.4 ~2.2% 10. Azerbaijan ........................ 3.9 1.9% 11. Italy ............................. 3.8 1.9% 12. France ........................... 3.5 1.7% 13. Algeria ........................... 3.2 1.6% 14. Netherlands ...................... 2.8 1.3% 15. Austria ........................... 2.3 1.2% 18. Qatar ............................ 2.2 1.1% 17. Canada .......................... 2.2 1.1% 18. Switzerland ....................... 2.1 1.0% 19. Russia ........................... 2.0 ~1.0% 20. India ............................. 1.7 ~0.9% Other .51.1 ~~~’~,~,~~~ 25.5’/o Total ........................... 200.4 © Tele[~eography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ..... !nco~,i_ng ............................. ~,30.2 .................... 185:,~ ........ 276.~ O.....U..tg~oin g ........................... 1.6o0.:? I,.7..7.0 ................. 2~0~4 Total Volume 290.9 362.7 416.8 Note: Data are in millions of minutes of outgmng pubhc switched telecommunicatmns traffic © TeleGeography, Inc. 2000 205 Tele(3eography 2001 © TeleGeography, Inc. 2000 t lreland Largest Telecommunications Routes, FY 1999/00 Destination Minutes (millions) Percentage of Outgoing Traffic 1. United Kingdom ................. 700.0 ~~,~~ ~~!!~ 69.0% 2. United States .90.0 8.9% ................... 3. France .......................... 29.0 2.9% 4. Germany ........................ 29.0 ~2.9% 5. Netherlands ..................... 25.0 2.5% 6. Canada ......................... 14.0 ~1.4% 7. Italy ............................ 12.0 1.2% 8. Spain ........................... 10.0 ~1.0% 9. Belgium .......................... 8.0 ~0.8% 10. Australia ......................... 7.0 0.7% Other ........................... 91,0 9,0% Total ’ .1,015.0 © TeleGeography, Inc. 2000 National Traffic Balance ¯ M!~n.q..t~..s.. ................... ~...!.,9,..9,7!98 ............ !9~8/99 ....... ..FY..1,.9..,99/00 .... ]£,€,oming ............................................................ n,:,~:, ...................... ,.n,,: a,.,: .............. ,n.:,a,,~, ...... ~, Surp!u,,,s,,(,,D,,,efi£!,.t,) ........................... p,:,,,a.. ............... £,,..a. ................ £,.,,.a,,. ,, Total Volume n.a. n.a. n.a. Note: Data are in mdhons of minutes of outgoing public switched telecommumcat~ons traffic. Fiscal year ends 31 March. 206 ©TeleGeography, Inc. 2000 © TeleGeographg, Inc. 2000 TeleGeography 2001 Largest Telecommunications Routes, 1999 Israel Destination 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Minutes (millions) United States ................... 200.0 United KinGdom .................. 67,0 Canada ......................... 50.0 France .......................... 43.0 Germany ........................ 43.0 Italy ............................ 30.0 Russia .......................... 27.0 Jordan .......................... 26.0 Ukraine ......................... 25.0 Netherlands ..................... 16.0 Other .......................... 277.0 Percentage of Outgoing Traffic ¯ ,,~, ,~ 3.2% 2.0% Total ........................... 804.0 © TeleOeographyo Inc. 2000 National Traffic Balance Note: Data are ~n millions of minutes of outgoing public sw[tched telecommunications traffzc Data based on billing pmnt of traffic © TeleGeography, Inc. 2000 207 TeleGeography 200 ! Italy Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic 1, Germany ....................... 465.0 ~~~ 15.0~/o 2. France .370.0 ~~~~" 11.9% ....................... 3. Switzerland ..................... 270.0 ~~ 8.7% 4. United Kingdom .260.0 ~i~~ o ~o~ ............... ~~~%~,"®1~ o.,+ ~o United States ................... 210.0 ~iI~ 6.8% Spain .......................... 135.0 ~~4.4% Romania ........................ 90.0 2.9% Austria .......................... 90.0 2.9% Belgium ......................... 80.0 2.6% Netherlands ..................... 65.0 2.1% Poland .......................... 65.0 2.1% Greece .......................... 60.0 1.9% Morocco’ ......................... 58.0 1.9% Canada ......................... 55.0 1.8% Albania ......................... 43.0 ~.~ 1.4% Tunisia .......................... 41.0 1.3% Croatia .......................... 41.0 1.3% Chile ............................ 30.0 1.0% Sweden ......................... 27.0 0.9% Russia .......................... 25.0 0.8% Other ........................ .620.0 ~~~~~~~~’, ~®- ~;~’~"~’~l ’~ ~’~ ..... 20.0% 6. 7. 8. 9. 20. Total ......................... 3,100.0 © TeleGeogropho, Inc. 2000 © TeleGeographyo Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ,! n € o..~oi n g, .............................. 2.:.475:.! ............ n:oa:.. ......................... n. ~0~tg°!Fg .......................... 2.,,~!:9 ................. ~,640-0 ................ ,3.J 9.0,:0 Su[p..!us (Deficit)_ ............................. 12.3..:2 ................... .n.:~.. .................... n.a,...- Total Volume 4,827.0 n.a. n.a. Note: Data are in millions of minutes of outgoing pubhc switched telecommumcations traffic Data exclude some traffic to France, Poland, Slovema, and Switzerland © TeleGeography, Inc. 2000 © TeleGeographg, Inc. 2000 TeleGeography 2001 Jamaica Largest Telecommunications Routes, 1999 Destination 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. Minutes (millions) United States .................... 44.5 United Kingdom ................... 5.4 Canada .......................... 4.6 Cayman Islands ................... 1.8 Trinidad & Tobago ................. 1.4 Bahamas ......................... 1.1 Barbados ......................... 1.0 Germany ......................... 0.6 Turks & Caicos Islands ............. 0.3 Cuba ............................ 0.3 Saint Lucia ....................... 0.3 Antigua & Barbuda ................ 0.3 N~therlands ...................... 0.3 India ............................. 0.3 Italy ............................. 0.2 Panama .......................... 0.2 Switzerland ....................... 0.2 Mexico .......................... 0.2 Saint Vincent & the Grenadines ...... 0.2 Japan ............................ 0.2 Other ............................ 3.2 Percentage of Outgoing Traffic ~@ 8.1% 7.0% 2,7% 2.2% 1,6% ~.5% 0.8% ;0.5% 0.5% ~0.4% 0.4% ;0.4% 0.4% 0.3% 0.3% 0.3% 0.3% 0.3% 0.2% 4.8% Total ............................ 66.4 © TeloGeography, Inc, 2000 National Traffic Balance Note: Data are zn mitlions of minutes of outgoing public switched telecommumca~ons traffic. Data based on bilhng point of traffic © TeleGeography, Inc. 2000 2O9 TeleGeograpby 2001 © TeleGeography, Inc. 2000 ( Japan Largest Telecommunications Routes, Fit 1999/00 Destination Minutes (millions) Percentage of Outgoing Traffic 1. United States ................... 406.4 ~~~~ 20.8% 2. China .......................... 285.1 ~~~~ 14.6S 3. Philippines ...................... 174,8 ~~ 8.9% .................... ~,.,-’ ~, ,,~ 4. Korea, Rep, .167.0 8.5% 5. Taiwan .107.1 ~~%~ 5.5o/o 6. Brazil ........................... 79.8 41% 7. Thailand ......................... 79.4 4.1% 8. United Kingdom .................. 60.9 3.1% 9. Hong Kong ...................... 58.4 ~3.0% 10. Singapore ....................... 48.0 2.5% 11. Australia ........................ 42.9 ~2.2% 12. Indonesia ....................... 38.1 1.8% 13. Seychelle,s ....................... 35.8 1.8% 14. Germany ........................ 30.1 ~1.5% 15. Malaysia ........................ 28.2 1.4% 16. Canada ......................... 27.1 ~1.4% 17. France .......................... 27.0 ~1.4% 18. Russia .......................... 19.0 ~1.0% 19. Vietnam ......................... 15.7 0.8% 20. India ............................ 14.2 0.TYo Other .211.6 10.8% Total ......................... 1,956.6 © Tele6eography, Inc. 2000 National Traffic Balance Minutes ...... ,F~.~,997~8 ....... ,~.Fy~,?~998/~ ..... _F~..7999/00 S U[P !,US,,(,,D, efi €, it) .......................... !! 36.,7,) ......... (,320,:0) ............ (2,7.0) ..... Total Volume 3,406.7 3,470.0 3,886.2 Note: Data are m millions of minutes of outgomg public switched telecommunications traffic Fiscal year ends 31 March. © TeleGeography; Inc. 2000 210 © TeleGeogrephg, Inc. 2000 TeleGeography 2001 Jordan Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic I. Saudi Arabia ..................... 20.5 ~:~~~~~~~,;;~’,",~ ,’~-’i~ "/,~ ’~ 15.5° 2. Egypt ........................... 15.1 11.4% 3. United Arab Emirates .............. 11.4 ~~,~’ ~"~ 8.6°~o 4. Syria. .10.4 ~~~ 7.9% 5. Iraq ...9.9 ~~ 74~o .......................... %~ ¯ 6. Israel ............................ 9.1 ~~~ 6.9% 7. United States ..................... 8.5 ~~~ 6.4% 8. Kuwait ........................... 5.7 ~~~ ~-~ 4.3°~ 9. United Kingdom ................... 4.6 3.4% 10. Qatar ............................ 3.0 2.3% 11. Germany ......................... 2.6 1.9% 12. Italy ............................. 1.7 ~.3% 13. France ........................... 1.7 ~1.3% 14. Canada .......................... 1.6 ~1.2% 15. Oman ............................ 1.5 ~1.1% 16. Yemen ........................... 1.4 ~1.1% 17. Bahrain .......................... 1.3 1.0% 18. Turkey .1.1 0.8% 19. Swi~erland ....................... 0.7 0.6% 20. India ............................. 0.7 0.5% Other .......................... .20.0 ~~,~5~~~~,~ ;,~, ~- ,~ ,~ .... ,~,, ]5.1 Total ........................... 132.5 © TeleBeography, Inc, 2000 National Traffic Balance Minutes 1997 1998 1999 I n.€,9,,~ i,,n,.,g ................................. 1~:0,, ............ !,76.9 .............. n.,a: ..... Sur~,~lus (D ef,!cit~ .................................... .5,3.,!, ............... ~4"~ ............. ~,.a. _ ...... Total Volume 23B.9 299.5 n.a. Note: Data are in millions of minutes of outgoing public switched telecommumcat~ons traffic. Data based on billing point of traffic. ©TelelGeography, Inc. 2000 211 TeleGeography 2001 © TeleGeographg, Inc, 2000 Kazakhstan Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. Russia .......................... 66.4 2. Uzbekistan ....................... 7.2 3. Kyrgyzstan ....................... 5.4 4. Ukraine .......................... 3.5 5. Azerbaijan ........................ 1.6 6. Belarus .......................... 1.5 7. Tajikistan ......................... 1.0 8. Armenia .......................... 0.7 9. Turkmenistan ..................... 0.7 10. Georgia .......................... 0.6 11. Moldova ......................... 0.3 Other ........................... 15.5 Tetal ........................... 104.5 Percentage of Outgoing Traffic 3.3% 1.4% 1.0% 0.7% 0.7% ! 0.6% ! 0.3% ~~~ ~4.~% © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ...Outgg.!ng, ...................................... !!~.7, ............ ,,1,,,!8.9 ............. !0~,.,5~ Su~plu.,~ (De~!~!t) ................................ ~:a. ............ !.~;6 ............ ~5:3 Total Volume n.a. 256.4 254.3 Note: Data are in mdhons of minutes of outgoing public switched telecommunications traffic. The "Other" category may =nclude routes to non-members of the Commonwealth of Independent States that rank among the top dest~natzons for outgoing traffic. © TeleGeography~ Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Korea, Rep. Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. United States ................... 210.0 2. Japan .......................... 150.0 3. China .......................... 140.0 4. Philippines ....................... 26.0 5. Hon9 Kong ...................... 25.0 6. Canada ......................... 24.0 7. Australia ........................ 22,0 8. Indonesia ....................... 18.0 9. Germany ........................ 17.0 10. Vietnam ......................... 17.0 11. Taiwan .......................... 14.0 12. Pakistan ......................... 13.0 13. Sihgapere ....................... 13.0 14. Thailand ......................... 10.0 15. United Kingdom .................. 10.0 Other .......................... 189.0 Percentage of Outgoing Traffic 2.9% 2.8% 1.9% 1.6% 1.4% 1.4% 1.1% 1.1% Total ........................... 898.0 © TeieGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ..... !,~com!,~g .............. 782.0 ......... ~7~1~.~ n.a. S.uFplq~¯¯(D e~(.~!t) ................... (!03.q). ...... .(.!88.3! Total Volume 1,667.0 1,627.1 n.a. Note: Data are in millions of minutes of outgoing public switched telecommunications traffic. 1999 data are for Korea Telecom, DACOM and Onse only. © TeleGeography, Inc. 2000 213 TeleGeo{~raphy 2001 © TeleGeogmph~l, Inc. 2000 ( Kuwait Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic 1. Egypt .......................... .35.0 ~~~~~~ ~’~i ¯’~ 20.6% 2. SaudiArabia ..................... 18.6 10.9% 3. India ............................ 18.2 ~~ 10.7% 4. Syria ............................ 11.9 ~~~ 7.0% 5. United Arab Emirates .11.2 ~~ 6.6% ............. 6. United States .9.3 ~~ ®~ ~. 5.5% 7. Iran ............................. 7.0 ~4.1% 8. Pakistan ......................... 6.9 ~4.1% 9. United Kingdom ................... 6.5 3.8% 10. Jordan ........................... 5.8 ~.~ 3.4%0 11. Lebanon ......................... 5.2 3.1% 12. Bahrain .......................... 4.2 2.5% 13. Qatar ..; .......................... 1.8 1.1% 14. Philippines ....................... 1.7 1.0% 15. Canada .......................... 1.7 1.0% 16. Bangladesh ....................... 1.7 1.0% 17. Sri Lanka ......................... 1.3 ~0.6% 18. Oman ............................ 1.2 ~0.7% 19. France .......................... 1.2 ~0.7% 20. Germany ......................... 1.1 0.7% Other ........................... 18.2 10.9% Total ........................... 170.0 © TeleGeography, Inc. 2000 National Traffic Balance Note: Data are in milhons of minutes of outgoing pubhc switched telecommumcations traffic. 214 © TeleGeography, Inc. 2000 © TeleGeography, In~. 2000 TeleGeography 2001 Kyr zstan Largest Telecommunications Routes, 1999 Destination 1. 2. 3. 4. 5. 6. 7. 8. Minutes (millions) Russia .......................... 10.3 Kazakhstan ....................... 5.9 Uzbekistan ....................... 2.7 Tajikistan ......................... 0.5 Ukraine .......................... 0.5 Belarus .......................... 0.2 Turkmenistan ..................... 0.2 Azerbaijan ........................ 0.1 Other ............................ 2.9 Percentage of Outgoing Traffic .... ~~ 11,5% Hi 2.1% 2.1% I~ 0.9% 0.9% i 0.4% ~~ 12.3% Total ............................ 23.5 © TeleGeography, Inc, 2000 National Traffic Balance Minutes 1997 1998 1999 I~co,~!~0g .................................. 5.] ............. 3~0.1 ......... n.a. ~rplus (D~ficit~)~ ......................... (~2.~9)~ .............. (P:,3) .................. n.a.~ ...... Total Volume 34.3 60.5 n.a. Note: Data are ~n milhons of minutes of outgoing public switched telecommumcat=ons traffic. The "Other" categow may include routes to non-members of the Commonwealth of Independent States that rank among the top destinations for outgoing traff=c, © TeleGeography, Inc. 2000 215 TeleOeography 2OO1 © TeleGeography, Inc. 2000 ( Latvia Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic 1. Russia ,13,3 23.9% ........................ ~.~ 700 2. Lithuania .6.5 11. 3. Estonia ........................... 5.3 ~~ 9.4~ ~. ~er~a~y ......................... 3.~ ~~~ 6.~% 5. Belarus .3.8 ~~ 6.7% ......................... 6. Sweden .......................... 2.9 5.2% 7. Ukraine .......................... 2.6 ~4.8% 8. United Kingdom ................... 2.1 ~3.7% 9. Finland ........................... 1.8 3.2% 10. Denmark ......................... 1.4 2.6% 11. Poland ........................... 1.4 ~2.4% 12. Noway .......................... 0.8 1.4% 13. Netherlands ...................... 0.7 ~, 1.3% 14. France ........................... 0.7 1.3% 15. Italy ............................. 0.7 1.2% 16. United States ..................... 0.6 ~.1% 17. Swi~erland ....................... 0.6 1.1% 18. Belgium .......................... 0.5 0.9% 19. Austria ........................... 0.4 0.7% 20. Israel ............................ 0.3 ~0.6% Other ............................ 5.5 ~~~ 10.0% Total ............................ 55.6 © TeteBeography, tnc. 2000 National Traffic Balance Minutes 1997 1998 1999 .!n c.o....m...!on g .................................................. 81.8 ..................... .8..~:2 ............. ..9.0.0.- Total Volume 125.8 142.5 145.6 Note: Data are in millions of minutes of outgoing public switched telecommumcat~ons traffic. Data based on billing point of traffic. 216 © TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Luxembourg Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic 1. France .......................... 76.0 2. Germany .75.0 ~~~~% ....................... ’,~ .... -’"~! ,~’: ~’~" ¯ ~-"-’~" 23 5 ~/o 3. Belgium .73.0 ~~~~~ 22.9% 4. Portugal ......................... 20.0 ~~~~@ l~ 6 3°~/o 5. United Kingdom .................. 18.0 ~@~;~ 5.6% 6. Italy ............................ 15.0 ~47~ 7. Netherlands ..................... 11.0 3.4% 8. Switzerland ....................... 9.0 2.8% 9. United States ..................... 7.0 ~i 2.2% 10. Spain ............................ 5.0 1.6% Other ........................... 10.1 3.2% Total ........................... 319.1 © TeleGeography, Inc. 2000 National Traffic Balance Note: Data are in millions of minutes of outgoing pubhc switched telecommunications traffic. © TeleGeography, Inc. 2000 ......... TeleGeography 2001 © TeleGeographg. Inc. 2000 ( Macau Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic 1. Hong Kong .,53,0 ....... ...... ~Y ....... .................... 39.9 Yo 2. China ........................... 50.7 38.1% 3. Taiwan ........................... 7.5 ~5.7% 4. United States ..................... 4.1 3.1% 5. Portugal .......................... 3.9 2.9% 6. Canada ........................... 2.6 ~2.0% 7. Philippines ....................... 1.9 1.5% 8. United Kingdom ................... 1.9 1.4% 9. Australia ......................... 1.8 1.3% 10. Thailand ......................... 1.1 0.8% 11. Singapore ........................ 0.6 ~0.5% 12. Japan ............................ 0.4 ~0.3% 13. Malaysia’, ......................... 0.4 i 0.3% 14. Korea, Rep ........................ 0.3 i0.2% 15. France ........................... 0.3 i0.2% 16. Vietnam .......................... 0.2 i 0.2% 17. Indonesia ........................ 0.2 1 0.1% 18. NewZealand ..................... 0.2 i 0.1% 19. Cambodia ........................ 0.1 I0.1% 20. Germany ......................... 0.1 i0.1% Other ............................ 1.5 1.2% Total ........................... 132.8 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 Total Volume 211.2 220.3 230.5 Note: Data are ~n milhons of minutes of outgoing pubhc switched telecommumcat~ons traffic. Data based on bdiing pmnt of traffic. 218 © TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Macedonia Largest Telecommunications Routes, 1999 Destination 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. Minutes (millions) Yugoslavia ....................... 24.6 Germany ........................ 11.9 Switzerland ....................... 4.8 Greece ........................... 4.4 Bulgaria .......................... 4.3 Italy ............................. 3.9 Slovenia ......................... 3.3 United States ..................... 3.2 Turkey ........................... 3.0 United Kingdom ................... 3.6 Croatia ........................... 2.4 France ........................... 2.3 Austria ........................... 2.6 Albania .......................... 1.4 Australia ......................... 1.0 Bosnia-Herzegovina ............... 0.9 Hungary .......................... 0.8 Belgium .......................... 0.8 Netherlands ...................... 6.8 Denmark ......................... 0.5 Other ............................ 2.8 Total ............................ 82.3 © TeleGeography, Inc. 2000 National Traffic Balance Mi,e,utes ................................. 1997 ............ 199,8. ..... !.999 ..... .... !,,,nco.,,m.!.,ng ....................................... ,.85;0 ....... 9.!:7 ........... 15,2.5 ,,, O..u...t,.g,o,! n g .......................................... 5~!.,;7 ............... 37:] .............. 82.:3 ¯ .,.S......U~lp!u§ (.D#,..fic.i~,t,) ............................. 33.3 54.6 70.3 Total Volume 136.7 128.9 234.8 Note: Data are in millions of minutes of outgmng pubhc switched telecomrnumcations traffzc. Data for 1998 exclude an estzmated 20 milhon minutes of traffic to Yugoslawa. © TeleGeography, Inc. 2000 219 TeleGeography 2001 © TeleGeography, Inc. 2000 Madagascar Largest Telecommunications Routes, 1999 Destination Minutes (thousands) Percentage of Outgoing Traffic 1. France ....................... .4,768.7 ~~~~.~"~~-"- "~.-..-’% .~+ ’il~ -. ~, "~ ~. ’~ ~":’:~--’: 49.5% 2. R~union .1,246.4 ~~ 12.9% 3. Mauritius ....................... 606.8 6.3% 4. Italy ........................... 250.9 2.6% 5. China .......................... 211.2 2,2% 6. South Africa .................... 197.4 2.0% 7. India ........................... 172.8 ~1.8% 8. Comoros ....................... 144.0 1.5% 9. United States ................... 142.9 1.5% 10. United Kingdom ................. 140.2 1.5% 11. Belgium ........................ 140.2 1.5% 12. Switzerland ..................... 126.8 1.3% 13. Mayotte’, ....................... 120.6 1.3% 14. Germany ....................... 118.4 1.2% 15. Canada ......................... 85.0 0.9% 16. Thailand ......................... 66.9 0.7% 17. Kenya ........................... 66.1 ~0.7% 18. Hong Kong ...................... 65.8 ~0.7% 19. Japan ........................... 53.7 ~0.6% 20. Singapore ....................... 44.4 0.5% ~..’..’ ....-,..~, Other .......................... 871.4 9.0% Total ......................... 9,640.5 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ,,,O,,,.U,.,,.t g, o,!,£,g ....................... 7:,.6, .............. ,,9..# ........... 9:§ ........ .... S u,r P,[,u,,s,,(Def,.!€,it,) ......................... 5.0~ ................... 6;#_ ............... ,0.0 .... Tota I Vo I u m e 20.2 25.3 19.3 Note: Data are ~n mdhons of minutes of outgoing publzc switched telecommunications traffic. 220 © TeleGeography, Inc. 2000 © TeleGeogKzphy, Inc. 2000 TeleGeography 2001 Malaysia Largest Telecommunications Routes, FY 1999/00 Destination 1. 2. 3. 4. 5. 6. 7. 8. 9, 10. 11. 12. 13. 14. 15. Minutes (millions) Singapore ...................... 350.0 Indonesia ....................... 45.0 Thailand ......................... 40.0 Japan ........................... 29.0 Australia ........................ 25.0 United States .................... 25.0 United Kingdom .................. 24.0 Hong Kong ...................... 15.0 Philippines ....................... 12.0 India ............................ 11.0 Taiwan .......................... 11.0 China ............................ 7.0 6~rmany ......................... 6.0 Bangladesh ....................... Korea, Rep ........................ 4.0 Other ........................... 80.0 Percentage of Outgoing Traffic 4.2% ~!~ 3.6% 3.6% 3.5% 2.2% 1.7% 1.6% 1.6% I.O% 0.9% 0.9% 0.6% ~®~ 11.6% Total ........................... 690.0 © TeleGeography, Inc, 2000 National Traffic Balance ...... Mi,gu~es~. ........................... .Fy._799,7~98 ..... .~F~__7998/~ ..... ,F~’_~ 1999/00 !,n,€~om!ng ......... 592.0 n.a. n.a. .... S,~,~p I,y S,,,(D e~c it) ............ ................. 3:5 ................. ~.a~:.............. n.a. Total Volume 1,180.5 n.a. n.a. Note: Data are in mdlions of mznutes of outgmng public switched telecommunications traffic. Data based on billing point of traffic. Fiscal year ends 31 March. Totals for FY 1997/98 were for Telkom Malaysia only. ©TeleGeography, Inc. 2000 221 TeleGeography 2001 © TeleGeographg, Inc. 2000 Malta Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. United Kingdom .................. 12,3 2. Italy ............................. 5.6 3. Germany ......................... 3.5 4. Libya ............................ 1.7 France ........................... 1.6 United States ..................... 1.4 Netherlands ...................... 1.4 Australia ......................... 0.9 Switzerland ....................... 0.8 Belgium .......................... 0.7 Russia ............................ 0.7 Austria ........................... 0.6 Turkey ’. .......................... 0.5 Sweden .......................... 0.5 Canada .......................... 0.4 Spain ............................ 0.4 Ireland ........................... 0.4 Greece ........................... 0.4 Tunisia ........................... 0.3 Norway .......................... 0.3 Other ............................ 4.7 6. 7. 8. 9. 10. 12. 13. 14. 15. 16. 17. 18. 19. 20. Percentage of Outgoing Traffic 14.3% 4.3%0 3.6% 3-5°/o ~1 2.4% 2.0% 1.7% 1.7% 1.4% 1.4% 1.3% 1.1% ~, 1.1% 1.1% 1.1% 0.9% 0.9% ~~~ 11.9% Total ............................ 39.0 © Te~eGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ,!,,n ,c o,,,rn~ n g ..................................... ,3,,7,.,,:0 ......... 43.~ ............. 50:2~ .... O,u,,,,tg o,!,n g,~ 34A ................ 37.3 ...... 39.0 Total Volume 71.4 80.7 89.2 Note: Data are ~n millions of minutes of outgoing public switched telecommunications traffic. 222 © TeleGeograph~, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Mauritania Largest Telecommunications Routes, 1999 Destination Minutes (thousands) 1. France ........................ 3,464.4 2. Senegal ....................... 1,378.7 3. United States .................. 1,216.2 4. Spain .......................... 989.2 5. Morocco ....................... 405.7 6. COte d’lvoire .................... 21Z6 7. United Arab Emirates ............. 204.2 6. 6abon .......................... 71.1 9. Canada ......................... 33.7 10. Togo ............................ 23.2 11. Tunisia .......................... 23.0 12. Burkina Faso ..................... 22.3 13. Be’nin ........................... 19.5 14. Egypt ........................... 17.4 Other ........................... 23.3 Percentage of Outgoing Traffic ~~ 16.6% ~~ 14.6% ~~ 11.9% 4.9% 2.6% 2.5% 0.9% 0.4% o,3% 0.3% 0.3% 0.2% 0.2% 2.7% Total ......................... 8,304.6 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 Total Volume n.a. 15.7 n.a. Note: Data are in millions of m=nutes of outgmng public switched telecommunications traffic. © TeleBeography, Inc. 2000 ¯ ......... TeleGeography 2001 © TeleGeograph~t, Inc. 2000 Mauritius Largest Telecommunications Routes, F’Y 1999/00 Destination Minutes (millions) Percentage of Outgoing Traffic .......................... 2. United Kingdom ................... 4.8 ~~~~ 15.4% 3. R~union .......................... 3.2 ~~~~ 10.2% ..................... 4. South Africa .3.0 ~~i 9.5% 5. India ............................. 2.1 ~~6.6% 6. Italy ............................. 1.2 3.8% 7. Australia ......................... 1.0 3.1% 8. Madagascar ...................... 0.9 9. Germany ......................... 0.9 2.9% 10. Switzerland ....................... 0.6 2.0% 11. China ............................ 0.6 1.9% 12. Singapore ........................ 0.6 1.9% 13. United States ..................... 0.6 1.8% 14. Hong Kong ....................... 0.5 1.7% 15. Belgium ........................... 0.5 1.5% 16. Seychelles ....................... 0.4 ~.4% 17. Canada .......................... 0.3 ~i1.1% 18. Taiwan ........................... 0.2 0.7% 19. Malaysia ......................... 0.2 0.6% 20. Kenya ............................ 0.2 ~0.5% ........................... ¯ ...... 108% Other °3.4 ~~~~! Total ............................ 31.4 © TeleBeegraphy, Inc. 2000 National Traffic Balance Minutes FY 1997/98 FY 1998/99 FY 1999/00 .... In..€...oming ................... .......... ... 35.0 ................ .3.9.5 ............... ~.3.,..~_ ..... ._...0. ut 9,~.!.~g ............................................... 2,.4.:,,,6. .................. ..7, ................... 3.1.4 ¯ S...U...rP I u s ,(D efi € !.t) ........................... 1.°-4 ........ 9:8 ............... !.~,,~ Total Volume 59.6 69.2 74.7 Note: Data are in millions of minutes of outgoing pubhc switched telecommunications traffic. Rscal year ends 30 June. 224 © TeleGeography’, Inc. 2000 © TeleGeographg, Inc. 2000 TeleGeography 2001 Mexico Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. United States .................. 1,364.6 2. Canada ......................... 21.4 3. Spain ........................... 15.9 4. France .......................... 10.5 5. Guatemala ....................... 10.2 6. Cuba ............................ 9.9 7. Colombia ......................... 9.8 8. Germany ......................... 9.0 9. Argentina ........................ 8.5 10. United Kingdom ................... 7.7 Other ........................... 95.5 Percentage of Outgoing Traffic 1.4% 1.0% 0.7% 0.7% 0.6% 0.6% 0.6% I 0.5% i 0.5% ~6.1% Total ......................... 1,563.0 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 .0 ~t~g~9~n g ................................... 1~.,~.1~3.6 .................. 1,~!0:~ ........... 1,563.~ ........ ~q.rplusJ~De~9it) ......................... o! ,~6o~.~7 ......... ~,750:0 .......... 2,44 ~4.~5 Total Volume 4,032.9 4,370.0 5,570.5 Note: Data are in millions of minutes of outgoing pubhc switched telecommunications traffzc~. Data based on billing point of traffic. © TeleGeography, Inc. 2000 TeleGeography 2001 © TeleGeography, Inc. 2000 Moldova Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. Russia .......................... 15.4 2. Ukraine ......................... 13.8 3. Romania ......................... 5.9 4. Greece ........................... 1.8 5. Germany ......................... 1.6 6. Italy ............................. 1.5 7. Belarus .......................... 1.5 8. Turkey ........................... 1.4 9. United States ..................... 0.6 10. Israel ............................ 0.5 11. Bulgaria .......................... 0.5 12. France ........................... 0.5 13. China ..’., ......................... 0.3 14. Poland ........................... 0.3 15. Portugal .......................... 0.3 16. United Kingdom ................... 0.3 17. Hungary .......................... 0.3 18. Czech Republic .................... 0.3 19. Austria ........................... 0.2 20. Spain ............................ 0.2 Other ............................ 2.0 Percentage of Outgoing Traffic ~~~1~~~~~’~ 28.1% 11.9% ~...j~ 3.1% 2.8% 1.1% 1.o% 1.0% 0.9% 0.7% 0.6% # 0.6% 0.6% 0.6% 0.5% 0.4% 0.4% Total ............................ 49.0 © TeleGeography, Inc, 2000 National Traffic Balance Min..ute..s ........... ..................................... !9~.,7.. ....................... ,1~.9,,,8. ............... .... Ln,€ o,,,,m, i n g ....................................... #0.2 ................ 90.3 ..................... .... 0 u.tg £!,,n g ................ 5 ~,.,,6, ........... .#~5,:8 ............... ,.49.0 Total Volume 135.8 146.1 150.1 Note: Data are in millions of minutes of outgo=ng public switched telecommunications traffic. Data based on billing point of traffic, 226 © TeleGeography, Inc. 2000 © Tele~eography, I~. 2000 TeleGeography 2001 Mongolia Largest Telecommunications Routes, 1999 Destination 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. Minutes (thousands) Russia ......................... 834.5 China .......................... 772.7 United States ................... 432.6 Japan .......................... 402.6 Korea, Rep ...................... 367.1 Germany ....................... 293.4 United Kingdom ................. 135.1 Hang Kong ...................... 86.4 Singapore ....................... 58.7 Australia ........................ 52.0 France .......................... 45.5 India ............................ 28.1 Pfii~ippines ....................... 27.1 Taiwan .......................... 26.4 Thailand ......................... 16.1 Malaysia ........................ 10.5 Vietnam .......................... 9.2 Indonesia ........................ 9.1 Pakistan ......................... 7.9 New Zealand ..................... 5.7 Other .......................... 368.5 Percentage of Outgoing Traffic ~~~i~ 10.8% ~~i~’~"~"" ~"~ : 9.2% ~~ 7.4% 3.4% 2.2%° 1.5% 1.3% 0.7% 0.7% 0.7% 0.4% 0.2% 0.2% 0.2% Total ......................... 3,989.2 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 l.nc2~ing ................................. !.~:2 .............. !...I...4. ................ !8.! .... ..0.. ~tg.o.!.n...g .................................................... 2.~ ................... .3:9 ................... ~:0 Surplus (,D efi_~,!~) ................................. 7:3 .................. 7:~ ........... 1.4.:.I Total Volume 13.1 15.3 22.1 Note: Data are in mdlions of minutes of outgoing public swztched telecommunications traffic, TeleGeograph~/, Inc. 2000 227 TeleGeography 2001 © TeleGeographg, Inc. 2000 Morocco Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic 1. France ......................... ,90,0 ~~~I~~~’ "~’~ ~" ~’’~’’’~’ ~"~;~’~ 41,0% 2. Spain ........................... 20.0 ~~ 9.1~ 3. United Kingdom .................. 17.0 7.7% 4. Italy .16.0 ~. ........................... 7.3% 5. Germany ......................... 9.0 ~4.1% 6. United States ..................... 9.0 ~4.1% 7. Belgium .......................... 9.0 ~4.1% 8. Netherlands ...................... 8.0 ~3.6% 9. SaudiArabia ...................... 8.0 3.6% 10. Canada .......................... 4.0 ~1.8% Other .29.5 "~ .......................... 13.4°/o Total ... : ........................ 219.5 © Tele6eography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 .... ~nc~o.~ing ........................................... 36~:~ ............ ,460..~.0 .............. ~.a.~ ,,.0 utgo,!n_,g., ......................................... 1,49.9,,, ................. 1.81.0_ ............... 2!.9,,:5 ,,, S,,,,u,[,P,].us (D,,oefic,],t,,) ........................ ..2..!4:1 ............ ,2,79.0 .................. n.a:, ...... Total Volume 513.9 641.0 n.a. Note: Data are in millions of minutes of outgoing pubhc switched telecommunmatmns traffic. © TeleGeographyl Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Mozambique Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic ..................... ~" 1. South Africa .9.7 2. Portugal .2.9 14~% ......................... 3. Zimbabwe ........................ 0.7 ~3.7% 4. United Kingdom ................... 0.4 2.0% 5. Swaziland ........................ 0.4 1.8% 6. Italy ............................. 0.3 1.3% 7. United States ..................... 0.2 1.2% 8. Malawi .......................... 0.2 1.0% 9. France ........................... 0.2 1.0% 10. Tanzania ......................... 0.2 0.9% 11. Germany ......................... 0.2 ~0.8% 12. Spain ............................ 0.1 0.7% 13. Br’azil ............................ 0.1 0.6% 14. Zambia ........................... 0.1 ~0.6% 15. Pakistan ......................... 0.1 0.6% 16. Netherlands ...................... 0.1 0.5% 17. India ............................. 0.1 ~0.4% 18. Angola ........................... 0.1 I0.4% 19. Sweden .......................... 0.1 I0.3% 20. Botswana ........................ 0.1 !0.3% Other ........................... .4.0 Total ............................ 20.3 © TeleGeography, Inc, 2000 National Traffic Balance Minutes 1997 1998 1999 !~pmi~,g ............................... ~9.~!~ .......... q.a.~ ........ 3~8.8 Total Volume 45.5 n.a. 59.1 Note: Data are zn millions of minutes of outgoing pubhc sw=tched telecomrnumcatzons traff=c. ©~/eGeograph~lnc. 2000 229 TeleGeography 2001 f Myanmar Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic 1. Thailand ........................ .3.4, "~ ~" ~, ¯ ; ~; ! ~" ’ ~’ 19.7% 2. Singapore ........................ 3.2 ¯ ~.~,~-~,~-~ o 3. Japan ............................ 2.2 ~~~~12.5~ 4. Malaysia .1.4 ~~8.0% 5. United States .1.2 ~~ ~qo/, ........................... ’" o 8. China .1.1 ~6.0~ 7. Taiwan ........................... 1.0 ~5.7% ............................ ~:~ o 8. India .0.6 3.3~ 9. AustraLia ......................... 0.4 2.2% 10. Hong Kong ....................... 0.4 ~2.1% 11. United Kingdom ................... 0.3 1.9% 12. Korea, Rep ........................ 0.3 ~1.8% 13. Indonesia ......................... 0.2 0.9% 14. France ........................... 0.1 0.7% 15. Philippines ....................... 0.1 0.7% 16. Pakistan ......................... 0.1 0.7% 17. Bangladesh ....................... 0.1 0.6% 18. Germany ......................... 0.1 ~0.5% 19. iVlacau ........................... 0.1 0.4% 20. Vietnam .......................... 0.1 0.4% Other ............................ 1.1 ~~ 6.4% © TeleGeography, Inc. 2000 Total ............................ 17.4 © TeleGeography, inc, 2000 National Traffic Balance Minutes 1997 1998 1999 ....... I n o.m...!~.g ...................... 3...5.0 ................... 36,:3 ........................ 29:8 ..~ Outg_oing ............ ,1~:3 ............... ~19..! ........... ~]:4~ ......... ¯ _S~r.P..lu...S (Defi~iiOi .... ’ .. ". ..................... !.8.]~ .............. !7:2 ..................... ’12....4.. ...... Total Volume 51.3 55.4 47.2 Note: Data are zn millions of minutes of outgoing public switched telecommunications traffic. 230 © TeleGeographyl Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Namibia Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic .................... ~’~ .~ ~.-~’~ ~@~’ ~’-’,’.-..~.~ o 1. South Africa .51.2 83.6’~o 2. Germany ......................... 2.3 ~3.8% 3. United Kingdom ................... 1.0 1.7% 4. United States ..................... 0.7 1.1% 5. Zimbabwe ........................ 0.7 1.1% 6. Angola ........................... 0.4 i0.7% 7. Spain ............................ 0.4 ~0.7% 8. Zambia ........................... 0.4 i0.7% 9. France ........................... 0.3 ~0.4% 10. Netherlands ...................... 0.2 i 0.4% 11. China ............................ 0.2 10.3% 12. Portugal .......................... 0.2 i 0.3% 13. switzerland ....................... 0.2 i0.3% 14. Italy ............................. 0.2 i0.3% 15. Russia ........................... 0.2 10.3% 16. Australia ......................... 0.1 10.2% 17. Austria ........................... 0.1 i0.2% 18. Belgium .......................... 0.1 i0.2% Other ............................ 2.3 3.7% Total ............................ 61.2 © Telel3eography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 !,n, com,,!.ng ............................... ..4.2:3.., ............ 45.3 ................ 51.~ O,,u..tgoin.g ..................................... 49.......7, ......... 6!...9 ........... 6,!,;2 ..... .S.,u rplu .S,,(D elicit) .......... (7:4) ............ (!..6,.6} ........... Total Volume 92.0 107.2 112.4 Note: Data are in mzlhons of minutes of outgmng public switched tetecornmunications traffic. © TeleGeooraphy, Inc. 2000 251 TeleGeography 2001 © TeleGeographg, Inc. 2000 Netherlands Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. Germany ....................... 450.0 2. RelGium ........................ 340.0 3. United Kingdom ................. 250,0 4. France ......................... 165.0 5. United States ................... 135.0 6. Italy ............................ 70.0 7. Spain ........................... 60.0 8. Switzerland ...................... 50.0 9. Turkey .......................... 40.0 10. Canada ......................... 40.0 11. Sweden ......................... 32.0 12. Denmark ........................ 27.0 13. Austria .’ .......................... 25.0 14. Poland .......................... 24.0 15. Morocco ........................ 17.0 Other .......................... 425.0 Percentage of Outgoing Traffic ~~ ~.~O/o 3.3% ~@ 2.~% 1,9% 1.5% 1.3% 1.2% 1.1% 0.8% Total ......................... 2,150.0 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ........ !nc0~ming ................................................ n,,a:.. ........ n,,,a,~: .............. .,n:~a,~ ..... ...O....u.tg !..n..g_ ................................. ~,6!~:o ................. !.:~5.....o ................ .~.,..!.5~0.o S~[plus,, (.D.efici~). ............................. .n:,.,a.., ....... n,:~.: ................ ~:a,,..,. Total Volume n.a. n.a. n.a. Note: Data are zn millions of minutes of outgoing public swztched telecommumcations traffic. 232 © TeleGeography; In c. 2000 © TeleGeographU, Inc. 2000 TeleGeo{~raphy 2001 New Caledonia Largest Telecommunications Routes, 1999 Destination 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. Minutes (thousands) France ........................ 9,235.3 AustraLia ...................... 1,829.1 French Polynesia ............... 1,009.8 New Zealand ................... 451.9 Vanuatu ........................ 428.8 Japan .......................... 179.3 United States ................... 143.8 Indonesia ....................... 129.5 Vietnam ........................ 120.1 Fiji .............................. 99.4 Germany ........................ 53,8 Singapore ....................... 51.7 Uriited Kingdom .................. 50.0 Thailand ......................... 32.6 Hong Kong ...................... 29.5 Philippines ....................... 23.9 China ........................... 21.4 Solomon Islands .................. 14.7 Papua New Guinea ............... 11.4 Malaysia ........................ 11.2 Other ......................... 1,978.9 Percentage of Outgoing Traffic ~~,~~ 58.1% 11.5°/o ~.3% 2.8% 2.7% 1.1% 0.9% o.~% 0.8% o.6% o.3% 0.3% 0.3% 0.2% ~o.2% 0.2% }o.1% 0.1% 0.1% 0.1% Total ........................ 15,904.0 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 Ou,,.,,tg, o !,.n,, g ......... n.,a,,. ............... !,3.1 ............ ,1,,5.9 ....... ,S,u,rp,,!,,~s ,!D eficiti,,,,,;i,,~iii "",, ~,ii~iii i ~;i .............. n.a.. ................. n,.a,. ...... n.a,. Total Volume n.a. n.a. n.a. Note: Data are =n milhons of minutes of outgozng public switched telecommumcat~ons traffic. © TeJeGeography, Inc. 2000 TeleGeography 2001 © TeleGeographg, Inc. 2000 ( New Zealand Largest Telecommunications Routes, F’Y 1999100 Destination Minutes (millions) Percentage of Outgoing Traffic 1. Australia ....................... 365.0 2. United States ...100.0 ~~~ 12.3% ................ 3. United Kingdom .................. 95.0 ~~ 11.7% 4. Canada ......................... 40.0 5. Philippines ....................... 21.0 2.6% 6. Hung Kong ...................... 14.0 1.7% 7. Japan ........................... 14.0 ~1.7% 8. Fiji .............................. 13.0 1.6% 9. Malaysia ........................ 10.0 1.2% 10. Singapore ....................... 10.0 1.2% ......................... Other .133.0 ~~i 16.3% Total ... ’. ........................ 815.0 © TeleBeography, Inc. 2000 National Traffic Balance Mi,,,n,,ute.s. ................................... FY,.o!, 99,,7~8, ..... !998/9.9_ ........... ~1.999/00. ! n ,c_o..,,m~n,g .................... 4,30.0 ........................ ,,,O,,,..utg ~i ~..g. .................................... ~.,0 ............ 6!~:0, ............... ~.~.5.:~, ..... S u,,ro.P..! u S, (~ e~!o.€. !.t) ....................... ......... 23:.,0, ................. ..n,a.- Total Volume 837.0 n.a. n.a. Note: Data are ~n millions of mznutes of outgoing public switched telecommunications traffic. F~scal year ends 31 March. 234 © TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Nicaragua Largest Telecommunications Routes, 1999 Destination 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. Minutes (millions) United States .................... 23.8 Costa Rica ....................... 12.1 Guatemala ........................ 2.9 Honduras ......................... 2.2 El Salvador ....................... 2.0 Mexico .......................... 1.3 Panama .......................... 1.2 Spain ............................ 1.0 Canada .......................... 0.9 Dominican Republic ................ 0.3 Colombia ......................... 0.3 Cuba ............................ 0.2 G~many ......................... 0.2 Italy ............................. 0.2 Brazil ............................ 0.2 Netherlands ...................... 0.1 Peru ............................. 0.1 Chile ............................. 0.1 France ........................... 0.1 Venezuela ........................ 0.1 Other ............................ 2.8 Percentage of Outgoing Traffic 4.3% 2.6% 2.4% ~!~ 2.0% 1.7% 0.6% o.5% 0.4% 0.4% 0.3% i 0.3% 0.3% i 0.2% ! 0.2% 10.2% io.2% ~:~ 5.0% Total ............................ 52.0 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ..!ncom!n~g~ .................................. ~52.~ ......... 5.9.:~ .......... 72.7. Total Volume 92.9 106.2 124.7 Note: Data are in m~llions of minutes of outgoing public switched telecommumcatzons traffic. Data based on billing point of traffzc. © TeleGeography, Inc. 2000 255 TeleGeography 2001 © TeleGeograph~l, Inc. 2000 Niger Largest Telecommunications Routes, 1999 Destination Minutes (thousands) Percentage of Outgoing Traffic ....................... 1. France .2,955.2 ~~~,. ,,~,,~ 48.4% 2. C6te d’lvoire .................... 539.1 8.8% 3. Benin .473.5 ~7.8% 4. Burkina Faso .................... 432.2 7.1% 5. Senegal ........................ 184.5 ~3.0% 6. United States ................... 173.3 ~i 2.8% 7. Nigeria ......................... 169.6 2.8% 8. Belgium ........................ 132.9 ~2.2% 9. SaudiArabia ..................... 82.7 1.4% 10. Germany ........................ 80.5 ~i 1.3% Other .......................... 876.5 ~~ 14.4% Total .. ;,. ..................... 6,100.0 © TeleGeography,,Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ..... !,n~ i n g ................................ ~:,~" ........................ ~o:~" ......... ~:,.a..: ........ S~[!?lus (~.efi~it) .................... n.a: ........................... ~:a. ........... ~:a., Total Volume n.a. n.a. n.a. Note: Data are zn millions of m~nutes of outgoing public switched telecommunications traffic. © TeleGeograph,~, Inc. 2000 © TeleGeogrophg, Inc. ~000 TeleGeography 2001 Norway Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. Sweden ........................ 153.2 2. Denmark ........................ 81.5 3. United Kingdom .................. 72.7 4. United States .................... 44.7 5. Germany ........................ 38.5 6. Netherlands ..................... 17.6 7. France .......................... 16.2 8. Spain ........................... 14.4 9. Finland .......................... 12.7 10. Italy ............................. 8.1 Other .......................... 107.2 Percentage of Outgoing Traffic ~~~ 7 9°/, ~~ ~.~. 2.9% 2.5% 2.2% 1.4% Total..~ ......................... 567.0 © TeleGeography, inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ¯ ~!~gcom!ng .......................................... 5~15.~ ............... n’a.’~ .......... 386.:~ .... O.~go!og ................................ 4.~.’1.0 ................ 5~4~:0 .... 567.0 Total Volume 996.0 n.a. 953.9 Note: Data are in mdlions of minutes of outgoing public switched telecommumcations traffzc. Data based on billing point of traff=c, ©~e~eography, lnc. 2000 237 TeleOeography 2001 © TeleGeographg, Inc. 2000 ( Oman Largest Telecommunications Routes, 1999 Destination Minutes (millions) I. India ............................ 30,3 2. United Arab Ernirates .............. 27.4 3. Pakistan ......................... 6.5 4. United Kingdorn ................... 5.8 EGypt ............................ 4.1 Saudi Arabia ...................... 2.8 BanGladesh ....................... 2.7 Bahrain .......................... 2.4 United States ..................... 2.3 Jordan ........................... 1.7 Sri Lanka ......................... 1.5 Kuwait ........................... 1.5 Philippines ....................... 1.4 Qatar ............................ 1.4 Tanzania ......................... 1.1 South Africa ...................... 1.0 Sudan ........................... I).9 6ermany ......................... 0.8 France ........................... 1).8 Netherlands ...................... 0.6 Other ............................ 4.3 6. 7. 8. 9. 10. 11. ,12. 13. 14. 15. 16. 17. 18. 19. 20. Percentage of Outgoing Traffic ~~ 6.4% 2.8% ~I~ 2,6% 2.4% 2.2% 1.6% 1.5% t~ 1.5% 1.4% ~i 1.3% 1.1% 1.0% 0.9% ~i. 0.8% 0.8% 0.6% Total ........................... 101.3 © Te)ol~eography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 .~ ~tg o.}...n..g ........................................... ...7..4:3. ................ 90.0 ................. 1.....01.3 ,,, SUr.plus~.D..efici.t) .................... (~:9) ................ (.!.8.3.o! ........... Total Volume 144.7 161.8 184.7 Note: Data are in millions of minutes of outgoing public switched telecommunications traffzc. Data based on billing po=nt of traffic. Data exclude some cross-border traffic to the United Arab Emzrates © TeleGeography, Ino. 2000 258 © TeleGeography, Inc. 2000 TeleGeography 2001 Pakistan Largest Telecommunications Routes, F’Y 1999/00 Destination Minutes (millions) Percentage of Outgoing Traffic 1. United Kingdom ..15.2 ~il/~i~~l~ 20.2% 2. United States ..................... 9.7 i........... 12.9% 3. United Arab Emirates ..9.5 ~~ ~2.6% 4. Saudi Arahia .7.6 ~~~ 10.2% Canada .......................... 6.1 ~’~ -~ 8.1% Italy ............................. 3.2 ~%~ 4.3% Germany ......................... 2.2 2.9% Iran ............................. 2.1 2.8% Japan ............................ 1.9 2.6% France ........................... 1.6 2.1% Qatar ............................ 1.4 1.9% Singapore ........................ 1.3 1.7% China ............................ 1.1 ~’,,~ 1.5% Bangladesh ....................... 1.1 1.5% Hung Kong ....................... 1.0 1.3% India ............. ................ 1.0 1.3% Oman ............................ 0.8 1.1% Netherlands ...................... 0.8 1.0% Kuwait ........................... 0.8 1.0% Australia ......................... 0.7 1.0% 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 17. 18. 19. 20. Total ............................ 75.1 © TeleGeography, Inc. 2000 National Traffic Balance ¯ _Minutes ............ ,~1~.~!997/98 ........ ,!~" 7998/,99 I:y!999/,~0 ....... ..... ~utgq~n,g ............................................. 84:~, ............... ~7~.5 ............. 75:.~ .......... ~p~lu~ (Deficit)" ........................ 47~,:7 .......... 55~:9~ ....... 5~9.~8,, Total Volume 641.9 727.9 720.0 Note: Data are ~n mtll~ons of minutes of outgoing pubhc switched telecommunicatmns traffic. Rscal year ends 30 June. Data exclude some cross-border traff=c to India. © TeleGeography, Inc. 2000 TeleGeography 2001 © TeleGeography. Inc. 2000 ( Palestinian Authority Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic ........................ 1 Jordan ...14,9 ~~~!~~,~,~42.7O/o 2. United States ..................... 4.7 ~~ 13.5% 3. Egypt ............................ 2.2 ~6.2% 4. Saudi Arabia ...................... 1.7 5.0% 5. United Arab Emirates ............... 1.3 3.6% 6. Germany ......................... 0.9 ~i 2.6% 7. United Kingdom ................... 0.8 2.2% 8. Italy ............................. 0.6 ~1.6% 9. France ........................... 0.5 ~1.4% 10. Turkey ........................... 0.4 1.2% 11. Syria ............................ 0.4 1.1% 12. Canada .......................... 0.4 1.1% 13. Morocco’, ......................... 0.3 0.8% 14. Kuwait ........................... 0.3 15. Qatar ............................ 0.2 ~0.7% 16. Sweden .......................... 0.2 ~0.5% 17. Tunisia ........................... 0.2 0.5% 18. Iraq ............................. 0.2 ~0.5% 19. Switzerland ....................... 0.2 0.5% 20. China ............................ 0.2 ~0.4% Other ............................ 4.5 i~~ 12,9% Total ............................ 34.9 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 0ut, g~!~g ............................................... n ~.~a:~ .... 27.6 ........... ~4:~ _Su[plu.s...(Def.!..€.it) ................................ n.a.. ..... ~..ii.~(..!.i.~)ii ................. ~.a:o ..... Total Volume n.a. 44.3 n.a. Note: Data are in millions of m~nutes of outgoing public switched telecommumcations traff=c. Data exclude traffic with Israel © TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 ~leGeography 2001 Panama Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. United States .................... 19.0 2. Colombia ......................... 5.6 3. Costa Rica ........................ 4.2 4. Mexico .......................... 2.3 Guatemala ........................ 1.4 Dominican Republic ................ 1.3 Venezuela ........................ 1.3 Nicaragua ........................ 1.0 Spain ............................ 1.0 El Salvador ....................... 1.0 Ecuador .......................... 0.9 Cuba ............................ 0.9 H(~nduras ......................... 0.7 Peru ............................. 0.7 Brazil ............................ 0.7 Canada .......................... 0.7 Argentina ........................ 0.5 Chile ............................. 0.5 United Kingdom ................... 0.5 China ............................ 0.4 Other ............................ 8.8 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. Percentage of Outgoing Traffic ~~@~ lO.5O/o ~j~ 7.8% ~.~% 2.6% 2.4% 2.4% 1.9% 1.9% 1.9% 1.6% 1.6% 1.3% 1.3% 1.3% ~.o% 1.o% 1.o% 0.8% Total ............................ 53.6 © TeleBeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ,,SurPlUs (Dpt)ci~) .......................... 5~3.7 .......... 45.5 ........... 4~:2 Total Volume 136.5 145.5 149.4 Note: Data are =n m~llions of minutes of outgomg public switched telecomrnun~cations traff(c Data based on billing point of traffic. © TeleGeography, Inc. 2000 241 TeleGeography 2001 © TeleGeography, Inc. 2000 Paraguay Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. Argentina ....................... 11.8 2. Brazil ............................ 8.5 3. United States ..................... 3.6 4. Uruguay .......................... 1.5 Lebanon ......................... 1.3 Chile ............................. 1.3 Germany ......................... 0.6 Bolivia ...... -. .................... 0.5 Spain ............................ 0.5 Peru ............................. 0.5 Taiwan ........................... 0.4 China ............................ 0.3 Italy ...’ .......................... 0.3 Korea, Rep ........................ 0).3 France ........................... 0.3 Mexico .......................... 0.3 Japan ............................ 0.2 Canada .......................... 0.2 Switzerland ....................... 0.2 United Kingdom ................... 0.2 Other ............................ 2.0) 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. Percentage of Outgoing Traffic ~~ ~0.2O/o 3-8°/o ~3.~ 1,7% 1.4% 1.4% 1.3% 1.2% 0.9% 0.8% 0.8% 0.8% 0.7% 0.7% 0.6% ~o.5~ 0.4~o 5.9% Total ............................ 34.7 © TeleGeography, Inc. 2000 National Traffic Balance Note: Data are in mdhons of m~nutes of outgoing pubhc switched telecommunications traffic. © TeleGeography, Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. United States .................... 33.5 2. Chile ............................ 8.2 3. Argentina ........................ 6.5 4. Spain ........................... 6.0 5. Colombia ........................ 3.4 6. Venezuela ....................... 3.2 7. Bolivia ........................... 3.0 8. Mexico ........................... 3.0 9. Brazil ........................... 2.8 10. Ecuador ......................... 2.1 11. ,Japan ........................... 2.1 12. Canada .......................... 1.8 13. Germany ......................... 1.5 14. United Kingdom ................... 1.2 15. France .......................... 0.9 16. Switzerland ...................... 0.8 17. Panama ......................... 0.7 18. Costa Rica ....................... 0.5 19. China ........................... 0.5 20. Uruguay" ......................... 0.5 Other ............................ 6.9 Peru Percentage of Outgoing Traffic ~~ 9.2% ~~ 7.3% 3.6% 3.4% 3.4~o 3.~o 2.4% 2.4~o 2.0% ~.6~o 1.o% 0.9% 0.8% 0.6% 0.5% 0.5% Total ............................ 88.9 © TeleGeography0 Inc. 2000 National Traffic Balance ~in..[es .................... ,~.,!99~/98 ....... 1~98/99 .......... !~ ..... incoming ........... 2,.56.,.9, ........... 27°2:,# ..... 302.6 ¯ ,0,u,,,tg o,,!n, g ~,, ........................... 79.4 .................... 90.~ ................... ,, S,u rplus,(D ef!,,cit)" .............. !,77.5, ......... 18,2.3 ......... 213.Z Total Volume 336.3 363.0 391.5 Note: Data are in milhons of minutes of outgoing public switched telecommunicatzons traffic. Data based on billing point of traffic © TeleGeography, Inc. 2000~ ¯ ......... TeleGeography 2001 Philippines Largest Telecommunications Routes, FY 1999/00 Destination Minutes (millions) Percentage of Outgoing Traffic 1. United States .55.0 ........... .... ................... 2. Japan .40.0 ~~~ 18.3% 3. Hong Kong ...................... 20.0 ~~~92% 4. Singapore .15.0 ~~ 6.9% 5. Taiwan .......................... 15.0 6.9% 6. SaudiArabia .15.0 ~~ 6.9% 7. Canada ......................... 15.0 ~6.9% 8. Australia ........................ I0.0 4.6% 9. Korea, ~ep ........................ 5.0 2.3% 10. Malaysia ......................... 5.0 2.3% Other ........................... 23.0 ~~ 10.6% © TeleGeograph~, Inc. 2000 Total ...’ ........................ 218.0 TeleGeography, tnc. 2000 National Traffic Balance Minutes FY 1997/98 FY 1998/99 FY 1999/00 !~...C.o.~.i n.g ............................ .9.30.0. ............ .n..a.. ....................... .n.:.a... ...... 0 u~90!.n..~ ..................................... 27..4.:0 .............. 26.?:0 ................ 2!.8:.0. ...... S...~ rp!.~.s.. (D .e...f~.€.it) ....................................... 656:0 ................... n.a.: .................. ~.a: ..... Total Volume 1,204.0 n.a. n.a. Note: Data are in millions of minutes of outgoing public switched telecommumcat~ons traffic. Data based on bilhng point of traffic. Rscal year ends 31 March. © TeleGeography, Inc. 2000 244 © TeleGeography, Inc. 2000 TeleGeography 2001 Poland Largest Telecommunications Routes, 1999 Destination 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Minutes (millions) Germany ....................... 240.0 United Kingdom .................. 45.0 Italy ............................ 42.0 France .......................... 34.0 United States .................... 27.0 Austria .......................... 21.0 Netherlands ..................... 21.0 Ukraine ......................... 20.0 Sweden ......................... 17.0 Czech Republic ................... 15.0 Other .......................... 142.0 To~al ........................... 624.0 Percentage of Outgoing Traffic 7,2% @~@ 6.7% 3.4-°/. ~i 3.2% 2.4% © TeleGeography, Inc~ 2000 National Traffic Balance Minutes 1997 1998 1999 ....... !.nco+m+ing .................................. .800.2 ............. ~o,i~.4.2 ........... n..a.. ¯ o+.u.t.g o!n.g ..................................... 529.0 .................... 6~2~ .Surp!.~.s (~.e.ficit! .......................... 27~¯._2 ............ 547:8 ................ Total Volume 1,329.2 1,746.6 n.a. Note: Data are in millions of minutes of outgoing public swztched telecommunicatmns traffzc. Data based on billing point of traffic. ©TeJeGeograpby, Inc+ 2000 245 TeleGeography 2001 © TeleGeography, Inc. 2000 ( Portugal Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic 1. France ......................... 102.9 ~~~~~~° ’~ ~, ,~ ’" ~i~,, !~ 19.3 2. Spain .......................... .91.1 % .... ~i ~’~" 17.1% 3. United Kingdom .................. 54.9 ~~ 10.3% 4. Germany ........................ 51.3 ~~,~ 9.6% 5. Brazil .28.5 ~l~ 5.3% 6. Switzerland ...27.3 ~~i 5.1°/o o 7. United States .................... 23.9 ~43 8. Italy ............................ 18.1 ~13-4% 9. Netherlands .16.4 ~,~l/" 3.1% 10. Be[gium ......................... 13.0 2.4% 11. Angola .......................... 12.6 2.4% 12. Canada .......................... 8.8 1.7% 13. Cape Ver~e ....................... 8.4 1.6% 14. Luxembourg ...................... 5.9 1.1% 15. Sweden .......................... 4.9 ~0.9% 16. Venezuela ........................ 4.2 ~0.6% 17. South Africa ...................... 4.1 0.6% 18. Mozambique ...................... 4.0 ~0.8% 19. Guinea-Bissau .................... 4.0 0.7% 20. Ireland ........................... 2.9 ~0.6% Other ........................... 45.7 ~~ 8.6 Total ........................... 532.8 © Telel3eography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 0u~g°ing ................................ 393.3 .................... 467"8 ................. ~3o2.:8 ~. Total Volume 1,022. I I ,I 78.8 1,286.0 Note: Data are ~n milhons of minutes of outgoing public switched telecommumcations traffic. Data are for Portugal Telecom only and may exclude some cross-border traffic to Spain. © TeleGeography, Inc. 2000 246 © TeleC, eogrQphy, Inc. 2000 TeleGeoi~raphy 2001 Largest Telecommunications Routes, 1999 Qatar Destination 1. 2. 3. 4. 5. 6. 7. 8. 9. 10, 11. 12, 13. 14. 15. 16. 17. 18. 19. 20. Minutes (millions) India ............................ 24.0 United Arab Emirates .............. 21.2 Saudi Arabia ..................... 10.0 Egypt ............................ 9.7 Bahrain .......................... 7.5 Pakistan ......................... 6.4 United Kingdom ................... 5.9 United States ..................... 3.8 Bangladesh ....................... 3.7 Jordan ........................... 3.7 Kuwait ........................... 3.1 Sudan ........................... 3.1 Sri,Lanka ......................... 2.7 Philippines ....................... 2.6 Iran ............................. 2.1 Oman ............................ 2.0 Lebanon ......................... 2.0 Syria ............................ 1.8 France ........................... 1.3 Yemen ........................... 1.1 Other ........................... 10.7 Percentage of Outgoing Traffic ~~ 7.6% ~~ 5.8% ,,, ,~ ~I 3.0% 2.9% 2,4% 2.o% 1.6% 1.6% 1.0% 0.8% Total ........................... 128.5 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 S ,u.~[P!US, (Qef!.~!,t) ........................ (40,.,,4) ........... (42;5) .... (~4. ,5,) ..... Total Volume 159.4 182.5 212.5 Note: Data are in millions of m=nutes of outgoing public swztched telecommumcations traffic. © TeleGeography, Inc. 2000 2z~7 TeleGeography 2001 Russia © TeleGeographg, Inc. 2000 Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. Ukraine ........................ 264.1 2. Belarus ........................ 100.9 3. Kazakhstan ...................... 59.0 4. Germany ........................ 47.2 5. Azerbaijan ....................... 42.1 6. Georgia ......................... 32.2 7. Uzbekistan ....................... 30.8 8. Armenia ......................... 24.7 9. United States .................... 20.9 10. Moldova ........................ 18.1 11. Latvia ........................... 16.1 12. United Kingdom .................. 13.9 13. Turkey .’ ......................... 13.1 14. Lithuania ........................ 13.0 15. Kyrgyzstan ....................... 11.7 16. China ........................... 11.4 17. Italy ............................ 10.9 18. France .......................... 10.7 19. Finland .......................... 10.5 20. Estonia ........................... 8.7 Other .......................... 168.2 Percentage of Outgoing Traffic ~~~ 10.1% 7.4% 5.0% 3.9% ~3.1% 2.0% 1.4% 1.4% 1.2% 1.2% 1.2% 1.2% 1.1% 1.o% 1.o% 0.8% Total ........................... 928.2 © TeleGeography, inc. 200g National Traffic Balance Note: Data are in millions of minutes of outgoing public switched telecommunications traffic. Data are for Rostelecom only. Other operators carned an estimated 200 million minutes of addition traffic ~n 1998 and 1999. 248 © TeleGeography, Inc. 2000 © TeleCeogrophy, Inc. 2000 TeleGeography 2001 Rwanda Largest Telecommunications Routes, 1999 Destination Minutes (thousands) 1. Belgium ........................ 753.8 2. United States ................... 565.4 3, Kenya .......................... 368.5 4. France ......................... 319.4 5. South Africa .................... 314.1 6. Uganda ........................ 313.3 7, Burundi ........................ 193.5 8. United Kingdom ................. 172.7 9. Germany ....................... 139.2 10. Netherlands .................... 113.4 11. Italy ........................... 109.4 12. india ............................ 68.1 13. Canada ......................... 41.4 14. United Arab Emirates .............. 39.3 15. COte d’lvoire ..................... 36.5 16. Senegal ......................... 34.9 17. Ethiopia ......................... 33.7 18. Cameroon ....................... 30.8 19. Egypt ........................... 28.2 20. China ........................... 22.0 Other ......................... 1,242.8 Total ......................... 4,742.8 Percentage of Outgoing Traffic ~~ 11.9°/o ~i~ 6.7% ~~ 6.6% 2.~% 2.3% ~.4% 0,9% 0.8% 0.8% 0.7% o,7% o.6% 0.6% 0.5% © TeleGeography, Inc, 2000 National Traffic Balance Minutes 1997 1998 1999 I~£0,,..ming ......................................... n-a: .......... ~.:a.. .... n.a:,, Outg£i,,,n,,g ....................................... n,...~., ............ 4:6 ............. 4:7 Su[plus,,...(,D, e..f..ic!t), ....................................... .n,,.a., ............. n:a. ............ n.,~.: Total Volume n.a. n.a. n.a. Note: Data are =n milhons of minutes of outgo=ng pubhc sw~tehed telecommunications traffzc. © TeleGeography, Inc. 2000 249 TeleGeography 2001 ( Saudi Arabia Largest Telecommunications Routes, 1999 © TeleGeographg, Inc. 2000 Desl~ination Minutes (millions) 1. EGypt .......................... 164.9 2. India ........................... 123.0 3. Pakistan ......................... 92.0 4. United Arab Emirates .............. 49.0 5. Syria ............................ 48.8 6. Philippines ....................... 48.4 7. Sudan .......................... 46.0 8. Yemen .......................... 41.0 9. Bahrain ......................... 38.5 10. United States .................... 33.0 11. Kuwait .......................... 27.0 12. Jordan .......................... 24.8 13. Lebanon :,.. ...................... 20.0 14. Morocco ........................ 18.8 15. United Kingdom .................. 17.0 16. Bangladesh ...................... 15.0 17. Canada ......................... 14,0 18. Qatar ........................... 10.8 19. Iran ............................ 10.9 20. France .......................... 10.0 Other .......................... 207.4 Percentage of Outgoing Traffic ~~~ 15.6% 11.6% ~~ 4.6% ~~ ,~.6% ~3.~% 2.3% 1.9% 1.6% 1.4% 1.3% 1.0% 0.9% Total ......................... 1,060.0 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ,In_~o~!,~,g ............................... n.a. n.a. n.a. Ou~9oing ..................................... 801.3 ......... 9~2.6 ............ ~,o~q,;q ,,S,.u,,,rP,!~,S, ( ,D,,.,,e,,f i £!,,t. ,),. ............................. q.a: ................. n:,~: ............................ n.a:, ..... Total Volume n.a. n.a. n.a. Note: Data are in millions of minutes of outgoing public switched telecommunications traffic. © TeleGeography, Inc. 2000 © TeleGeographg, Inc. 2000 TeleGeography 2OO1 Senegal Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. France .......................... 11.0 2. COte d’lvoire ...................... 2.4 3. United States ..................... 2.3 4. Italy ............................. 1.5 5. Marl ............................. 1.4 0. Gambia .......................... 1.2 7. Guinea ........................... 0.9 8. Mauritania ....................... 0.9 9. Gabon ........................... 0.7 10. Spain ............................ 0.6 11. Morocco ......................... 0.6 12. Canada .......................... 0.5 13. B~rkina Faso ...................... 0.4 14. Germany ......................... 0.4 15. Switzerland ....................... 0.4 16. United Kingdom ................... 0.4 17. Cameroon ........................ 0.3 18. Benin ............................ 0.3 19. Portugal .......................... 0.3 20. Togo ............................. 0.3 Other ............................ 9.7 Percentage of Outgoing Traffic ~i~ 6.6% ~4.1% 2.4% 2.0% 1.6% 1.5% 1.4% 1.2% 1.1% 1.1% 1.0% 0.9% ~j~ 0.9 ~o 0.9% 0.9% Total ............................ 36.5 © TeleGeography, Inc. 2000 National Traffic Balance Minutes FY 1997/98 FY 1998/99 1999 .... Inc,,0,,~ing ................................. 68.7 ..................... ,9,.~.8 ............. !11.,! ...... ..... ~.~,tg°,i~g ................................... 27~.7 .............. 29:6 ................. ,3,~;5 ....... SurpI,us !,O,e, fiq!~) ............................ ~1.0, .............. 6~.2 ..... 74.7 Total Volume 96.4 123.4 ........ i 47~~ Note: Data are m millions of m=nutes of outgo=ng pubhc switched telecommunications traffic. © TeleGeography, Inc. 2000 251 TeleGeography Singapore © TeleGeography, Inc. 2000 Largest TelecommunicationsRoutes, F’Y 1999/00 Destination Minutes (millions) Percentage of Outgoing Traffic 1. Malaysia ....................... 465.0 ~~. ~!~ 2. Hong Kong .100.0 .... o ..................... 7.4 yo 3. indonesia .70.0 5.2% 4. United States .70.0 5.2% 5. Australia ........................ 65.0 ~4.8% 6. China ........................... 65.0 ~4.8% 7. Japan ........................... 55.0 ~4.1% 8. Thailand ......................... 45.0 3.3% 9. Philippines ....................... 38.0 2.8% 10. India ............................ 33.0 2.4% 11. United Kingdom .................. 33.0 2.4% Other .......................... 311.0 ~~~~1~ 23.0 Yo Total ......................... 1,350.0 © TeleGeography, Inc. 2000 National Traffic Balance Minutes FY 1997/98 FY 1998199 FY 1999/00 ...... I n,,c,,.o, m!,q.,g, ...................................... n,:,~., ............ p:a,.,, .............. n:a,; ,,,Q u~!,,0,,,in g .............................................. .!,,~!,6 !,,,.,~ .............. ,.1,,2...35.,0 ................ !,,;,,3. 5,~0,.,,0 .... S,,,U r p! U,, S, (D,,,,efi c,,!,t,) ...................................... ~.a; ................... ,n,,..,,a., ................. p;,a,,,: .... Total Volume n.a. n.a. n.a. Note: Data are tn millions of minutes of outgoing public switched telecommunications traffic Data based on bilhng point of traffic. Fiscal year ends 31 March © TeleGeography, Inc. 2000 252 © TeleGeography, Inc. 2000 TeleGeography 2001 Slovak Republic Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. Czech RepubJJc ................... 64.6 2. Germany ........................ 19.7 3. Austria .......................... 13.8 4. Hungary .......................... 8.3 5. United Kingdom ................... 6.9 6. Italy ............................. 6.7 7. Poland ........................... 5.5 8. Ukraine .......................... 4.6 9. United States ..................... 3.6 10. France ........................... 3.4 11. Switzerland ....................... 3.1 12. Netherlands ...................... 2.2 13. Russia ........................... 2.1 14. Belgium .......................... 1.5 15. Spain ............................ 1.5 16. Croatia ........................... 1.3 17. Yugoslavia ........................ 1.3 18. Canada .......................... 0.9 19. Greece ........................... 0.9 20. Sweden .......................... 0.8 Other ........................... 10.1 Percentage of Outgoing Traffic 8.5% 5.~% 3.4% 2.8% 2.2% ~2.1% 1.9% 1.4% 1.3% 0.9% 0.9% 0.8% 0.8% 0.6% 0.50/o 0.5% Total ........................... 162.8 © TeleGeography, Inc. 2000 National Traffic Balance Note: Data are m millions of minutes of outgoing pubhc switched telecommumcations traffic. Data based on billing point of traff=c © TeleGeography, Inc. 2000 TeleGeography 2001 ( Slovenia Largest Telecommunications Routes, 1998 Destination Minutes (millions) Percentage of Outgoing Traffic 1. Croatia .......................... 31.3 ~~~~i~ 24.1% 2. Germany ........................ 15.7 ~~~~ 12.1% 3. Austria .......................... 14.6 ~~ 11.3% 4. Italy .14.4 11.1% 5. Yugoslavia .13.2 ~~ 10.2% ...................... 6. Bosnia-Herzegovina ............... 6.1 4.7% 7. Macedonia ....................... 3.4 2.6% Switzerland ....................... 3.0 2.3% United Kingdom ................... 2.9 2.2% France ........................... 2.8 2.1% Hungary .......................... 2.3 1.8% United States ..................... 2.1 1.6% Russia .’. .......................... 1.7 1.3% Netherlands ...................... 1.6 1.2% Czech Republic .................... 1.4 1.1% Sweden .......................... 1.2 0.9% Belgium .......................... 1.1 0.9% Poland ........................... 0.8 0.6% Canada .......................... 0.8 0.6% Chile ............................. 0.7 0.5~° Other ............................ 8.5 ~~~ 6.6% 9. 11. 12. 13. 14. 15. 17. 18. 20. Total ........................... 129.6 © TeleGeographg, Inc. 2000 © Tele6eography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ~. I n,,~€ °m !...n.~9 .................................. .!,!,,8:9 ........... 137:,0 ............... ,, O.u....t...g..o i ng~ ................................................. 113...5 ................. .!..2.9..6 ............... n.a.~ ............ ...Sur~us (D...ef!~..i.t) .................................. 5:4 .................. ~.4., ....... .n....a.,.. Total Volume 232.4 266.6 n.a. Note: Data are m millions of minutes of outgoing public sw=tched telecommumcatmns traffic. 1999 traffic data are not available. © TeleGeography, Inc. 2000 © TeleGeograph~l, Inc. 2000 TeleGeography 2001 South Africa Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. United Kingdom .................. 96.9 2. United States .................... 43.3 3. Namibia ......................... 41.0 4. Zimbabwe ....................... 36.5 5. Botswana ....................... 18.8 6. Mozambique ..................... 18.8 7. Germany ........................ 17.1 8. Swaziland ....................... 14.2 9. Australia ........................ 12.5 10. Lesotho ......................... 11.4 11. Canada .......................... 9.1 12. France ........................... 9.1 13. Netherlands ...................... 9.1 14. italy ............................. 8.0 15. Zambia ........................... 6.8 O~her .......................... 108.3 Percentage of Outgoing Traffic ~~ 9.~% ~4.1% ~;~ 3.7% ~.7% 2.0% 2.o% 1.7~o Total ........................... 461.1 © TeleGeography, Inc, 2000 National Traffic Balance Minutes 1997 1998 1999 .!nc..o.ming ................................. 3..4.3.2 ............ n..a. ......... .n..a. 0 utg...O...!~ g .......................... 3.68:8 .............. 40.5-0¯ ............ ~61.~ ..... .... .Su..r~ lu S. (D.ef!ci:t) .................... (25.~) .................. n:.~.. .......... ~:a.. ..... Total Volume 712.0 n.a. n.a. Note: Data are in millions of minutes of outgoing public switched telecommunications traffic. © Tel~Geography, Inc. 2000 255 TeleGeoEaphy 2001 Spain Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic 1. Germany ....................... 320.0 ~J~~J~~~~ 16.5% ....................... ¯ ¯ 2. France ...300.0 15.5% 3. United Kingdom .300.0 15.5% ............... .......................... i~ ¯ ’~ 4. Italy .130.0 i 6.7% 5. Portugal ......................... 90.0 ~~4.7% 6. United States .................... 90.0 ~~ 4.7% 7. Switzerland ...................... 60.0 3.1% 8. Belgium ......................... 55.0 ~2.8% 9. Netherlands ..................... 55.0 2.8% 10. Cuba ............................ 40.0 ~2.1% 11. Morocco ........................ 35.0 1.8% 12. Argentina ....................... 28.0 1.4% 13. Andorra : ......................... 27.0 1.4% 14. Sweden ......................... 23.0 1.2% 15. Brazil ........................... 20.0 ~1.0% 16. Chile ........................ i...20.0 ~1.0% 17. Mexico ......................... 20.0 ~1.0% 18. Colombia ........................ 17.0 0.9% 19. Denmark ........................ 16.0 ~0.8% 20. Austria .......................... 13.0 0.7% Other .......................... 276.0 ~~~~~%~ 14.3% © TeleGeograph~/, Inc. 2000 Total ......................... 1,935.0 © TeteGeography, In c. 2000 National Traffic Balance Note: Data are m mdhons of minutes of outgoing public sw~’tched telecommunications traffic. Data based on bilhng point of traffic. L ©TeleGeographyl Inc. 2000 256 © TeleGeography, Inc. 2000 TeleGeography 2001 Sri Lanka Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. India ............................. 7.5 2. United Kingdom ................... 4.7 3. Singapore ........................ 2.9 4. Japan ............................ 2.8 5. United States ..................... 2.6 8. Australia ......................... 2.1 7. Germany ......................... 1.8 8. United Arab Emirates ............... 1.8 9. I-long Kong ....................... 1.7 10. Saudi Arabia ...................... 1.4 11. Maldives ......................... 1.2 12. Korea, Rep ........................ 1.2 13. Ita’ly ............................. 1.0 14. Canada .......................... 0.9 15. France ........................... 0.9 16. Pakistan ......................... 0.8 17. Malaysia ......................... 0.7 18. Thailand ......................... 0.7 19. Switzerland ....................... 0.6 20. Kuwait ........................... 0.6 Other ............................ 7.8 Percentage of Outgoing Traffic ~~ 10.4% 6.3~ ~~ 5.7/o ~@i~ 5.6% ~~ 4.o% ~’,~ 3.6% 2.3% 1.9% 1.~% ~3~.4% Total ............................ 45.5 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 In~m!,~g ............................................. !~4.3 ........... 146,8 n.a. S¯ur.¯.p.l.~ s(D.e~! € it) ........................ 91..!. .......... !07.5 Total Volume 157.5 186.1 n.a. Note: Data are in millions of minutes of outgmng public switched telecommunications traffic. © Te/eGeography, Inc. 2000 ¯ ......... TeleGeography 2001 Sudan Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. Saudi Arabia ...................... 6.9 2. United Arab Emirates ............... 2.5 3. Egypt ............................ 1.6 4. United Kingdom ................... 1.1 5. Qatar ............................ 0.6 6. United States ..................... 0.5 7. China ............................ 0.4 8. Jordan ........................... 0.4 9. Syria ............................ 0.4 10. 6ermany ......................... 0.3 11. India ............................. 0.3 12. France ............................ 0.2 13. Netherlands ...................... 0.2 14. Italy ............................. 0.2 15. Canada .......................... 0.2 16. Libya ............................ 0.2 17. Yemen ........................... 0.2 18. Oman ............................ 0.2 19. Switzerland ....................... 0.1 20. Kenya ............................ 0.1 Other ............................ 5.3 © TeleGeographg, Inc. 2000 Percentage of Outgoing Traffic ~~ 7.4O/o 6.2% 2,9% 2.4% 1.8% 1.7% 1.6% 1.3% 1.~% 1.1% 1.0% 0.9% 0.8% 0.8% 0.7% 0.6% Total ............................ 21.9 © TeleGeography, Inc. 200D National Traffic Balance Minutes 1997 1998 1999 I,E£o m i,,n,,,g,, .................. ............... ~3..2 ........... ~8.;0 ............... 1~05...3 0ut,gg.i, ng ........................... 1,4,,.,~ ............... !8.~ .............. 2!,.,.,9, ...... ,,~...u. rp I u S..(..D. e fi €!t) ................ ,.28.4 ................ 6..9.:6 ....... 83.4 Tota I Volu m e 58.0 106.4 127.2 Note: D.ata are in m~llmns of minutes of outgoing public switched telecommumca~ons traffic. Data exclude some cross-border trafftc to Chad. 258 © TeleGeograph¥, Inc. 2000 © TeleGeograph!j, Inc. 2000 TeleGeography 2001 Suriname Largest Telecommunications Routes, 1999 Destination 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. Minutes (thousands) Netherlands ................... 6,348.5 United States .................. 1,325.6 Rrazil .......................... 898.3 Taiwan ......................... 524.4 Guyana ........................ 341.6 Netherlands Antilles ............. 316.9 French 6uiana .................. 198.2 Trinidad & Tobago ............... 188.7 Canada ........................ 156.2 United Kingdom ................. 155.0 Dominican Republic .............. 138,5 Aruba .......................... 108.6 V~nezuela ....................... 75.7 France .......................... 59.7 Japan ........................... 37.3 Jamaica ......................... 32.6 Germany ........................ 29.0 Colombia ........................ 23.1 Argentina ....................... 21.5 Panama ......................... 17.8 Oth er ......................... 1,380.2 Percentage of Outgoing Traffic ~~ 51.3% 2.8% 2.6% 1.6% 1,5% ~i~ 1.3% 1.3% 1.1% 0.9% 0.6% 0.5% i 0.3% i 0.3% i 0.2% !o.2% ! 0.2% i 0.1% Total ........................ 12,377.0 © TeleGeography, Inc, 2000 National Traffic Balance Minutes 1997 1998 1999 ¯ .!.n €.o...m,..i n g, ........................... 3,!.:0.. ............... 3~.:4, ......... .,2.7.9 .... 0.....utg o!..n., g ........................................ ,4,:8, ............... 5...£ .................. 12:~ ~urp, l#s (De,f!ci....t) .............................. 2,6.4 ........... .2.,9.4. ............... !5.5. Total Volume 35.6 39.4 40.3 Note: Data are in millions of minutes of outgmng public swztched telecomrnunicatmns traffic. © TeleBeography, Inc. 2000 259 TeleGeography 2001 © TeleGeographg, Inc. 2000 ( Swaziland Largest Telecommunications Routes, F’Y 1999/00 Destination Minutes (thousands) Percentage of Outgoing Traffic ................. ; "’~L~ ~" ~. "~ ¯ 1. South Africa .25,693.6 2. Mozambique .................... 865.6 3.0% 3. United Eingdom ................. 421.2 1.4% 4. United States ................... 169.3 0.6% Lesotho ........................ 162.2 0.6% Zimbabwe ...................... 158.5 0.5% Zambia .......................... 96.7 0.3% Eenya ........................... 57.2 0.2% Germany ........................ 56.1 0.2% Namibia ......................... 54.4 0.2% Netherlands ..................... 50.7 0.2% Tanzania ........................ 47.3 0.2% Maiawi :,. ....................... 43.7 ~0.1% Uganda ......................... 41.7 0.1% Italy ............................ 40.9 0.1% India ............................ 37.9 0.1% C~n~da ......................... 34.6 0.1% ~h~na .......................... 33.0 0.1% ~ustralia ........................ 32.4 0.1% Belgium ......................... 30.9 0.1% Other ......................... 1,200.0 4.1% 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. Total ........................ 29,340.2 © TeleBeography, inc. 2000 National Traffic Balance Note: Data are in mdhons of minutes of outgoing public switched telecommumcations traffzc Fiscal year ends 31 March. 260 © TeleGeographyi Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Sweden Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic ................ .~ ~- ¯ ,~ "~ 1. United Kingdom .200.0 14.7% 2. Finland .180.0 13.2’,’o 3. Norway .165.0 ~~ 12.1% 4. United States ................... 145.0 10.6%o 5. Denmark ....................... 135.0 ~~~ 9.9% 6. Germany ....................... 120.0 8.8% 7. Poland .......................... 70.0 ~~ 5.1% 8. France .......................... 60.0 ~~~ 4.4% .................... ¯ o 9. Netherlands .45.0 3.3~/o 10. Switzerland .35.0 2.6% Other .......................... 210.0 15.4% Total ......................... 1,365.0 © TeleGeography, Inc. 2000 National Traffic Balance Note: Data are m millions of m~nutes of outgoing pubhc switched telecomrnumcatmns traffic. Data based on bilhng point of traffzc. © TeleGeography, Inc. 2000 261 TeleGeography 2001 © TeleGeographu, Inc. 2000 Switzerland Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. Germany ....................... 620.0 2. France ......................... 420.0 3. Italy ........................... 370.0 4. United Kingdom ................. 145.0 5. Austria ......................... 135.0 6. United States ................... 115.0 7. Portugal ......................... 80.0 8. Spain ........................... 80.0 9. Netherlands ..................... 65.0 10. Yugoslavia ....................... 50.0 Other .......................... 650.0 Percentage of Outgoing Traffic ~~ ~.~°/o 2.9% Total ....: ..................... 2,730.0 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 .... ~ut90ing ....................................... 2,16~:0 ....... 2~;~ ........ 2,7~0:0 ....... Total Volume 3,887.0 n.a. n.a. Note: Data are in millions of minutes of outgo;ng public switched telecommunications traffic. 262 © TeleGeography, Inc. 2000 © T~leGeograph~/, Inc. 2000 TeleGeography 2001 Syria Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic 1. Lebanon,. ...................... .36.0 ~~~~~’~" ~’ "~ ~% =~4~i~/ ~"~ 28.7% 2. Saudi Arabia ..................... 20.8 ~~~~~ %~ 16.5% 3. United ArabEmirates ............... 8.1 ~,~,~6.~~~ 4° 4. Jordan ........................... 7.6 ~~6.1~o 5. Kuwait .5.4 ~o~ .......................... ~,~/o .................... ~,~ 6. United States .5.2 4.1% 7. Egypt ............................ 3.0 2.4% 8. France ........................... 2.6 ~2.1% 9. United Kingdom ................... 2.2 1.8% 10. Turkey ........................... 1.9 1.5% 11. Germany ......................... 1.8 ~1.4% 12. Iraq ............................. 1.7 1.4% 13. canada .......................... 1.4 1.1% 14. Italy ............................. 1.4 1.1% 15. Russia ........................... 1.0 ~0.8% 16. Greece ........................... 0.9 0.7% 17. Qatar ............................ 0.7 ~0,6% 18. Sudan ........................... 0.7 ~0.5% 19. Sweden .......................... 0.6 ~0.5% 20. Yemen ........................... 0.6 ~0.4% Other ........................... 22.2 ~~~ Total ........................... 125.6 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ..... !~ncom!~g ................................... !~3.2, .............. ~a~ :~ .............. 256.7~ Total Volume 262.5 n.a. 382.3 Note: Data are in mzllions of minutes of outgoing public switched telecommunications traffic, Data based on bdling point of traff=c. Te~eGeography, Inc. 2000 265 TeleGeography 2001 © TeleGeographg, Inc. 2000 ( Taiwan Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. China .......................... 286.9 2. United States ................... 162.0 3. Japan ........................... 80.3 4. Hong Kong ...................... 73.7 5. Philippines ....................... 55.5 6. Thailand ......................... 45.0 7. Canada ......................... 29.6 8. Singapore ....................... 26.8 9. Indonesia ....................... 23.5 10. Vietnam ......................... 20.8 11. Australia ........................ 18.5 12. Malaysia ........................ 15.8 13. Korea, Re’p ....................... 13.0 14. Russia .......................... 11.8 15. 6ermany ........................ 10.4 16. United Kingdom .................. 10.0 17. New Zealand ..................... 6.8 18. France ........................... 6.7 19. Macau ........................... 4.4 20. Netherlands ...................... 4.3 Other ........................... 44.1 Percentage of Outgoing Traffic ~t~~~ 30.2% 7.8% ~3.1% 2.2% 2.0% 1.6% 1.2% 0.7% 0.7% 0.5% .6% Total ........................... 949.3 © Tele6eography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ~...u....t_9 O! ~.g. ......................................... ~8_9...0 .......... 862:~ .................. 9~.9:~ .... Total Volume ............... i ~6~i"i’~ 1,643.9 1,831.3 Note: Data are in millions of m=nutes of outgoing public switched telecommunications traffic. Data based on billing point of traffic. 264 ©TeleGeography; Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Tajikistan Largest Telecommunications Routes, 1999 Destination 1. 2. 3. 4. 5. 6. 7. Minutes (millions) Russia ........................... 5.4 Uzbekistan ....................... 1.6 Kazakhstan ...................... i(3.6 Ukraine .......................... 0.4 Kyrgyzstan ....................... 0.3 Turkmenistan ..................... 0.2 Belarus .......................... 0.1 Other ............................ 0.4 Percentage of Outgoing Traffic ~~~,,,,~,~,~~~ ~o.o% 4.4% 3.3~o 22% 1.1% 4.4% Total ............................. 9.0 © Telol3eography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 .... Incoming .................................... n:a,, .................. n.a.~ ...... n;a. O..~tg oi, n,g ..................................................... ~3;6 ................. ~:9 ............... ~.,0.,. ~,urplu§ (~,,~ficit), ............................ n,,,:a. .................... n.a., ......... ~.a., Total Volume n.a. n.a. n.a. Note: Data are in millions of minutes of outgoing pubhc switched telecommunications traffic. The =Other" category may include routes to non-members of the Commonwealth of Independent States that rank among the top destinations for outgoing traffic. @TeleGeography, Inc. 2000 :265 TeleGeography 2001 Thailand Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic 1. ,Japan ........................... 45,0 15.1% ................... 2. United States .40.0 ~~ 13.4% 3. Singapore .34.0 11.4°/, ..................... 4. Hong Kong .22.0 ~~ 7.4% ................. 5. United Kingdom .17.0 ~~~ 5.7% 6. Taiwan .......................... 16.0 ~~~ 5.4% 7. Australia ........................ 15.0 5.0% S. China ........................... 13.0 ~~~4.4% 9. Germany ........................ 13.0 ~~ 4.4% 10. France ........................... 7.0 2.3% 11. India ............................. 7.0 ~2.3% 12. Korea, Rep ........................ 6.0 2.0% 13. Philippines ....................... 6.0 2.0% 14. Indonesia ........................ 5.0 1.7% 15. Switzerland ....................... 4.0 1.3% Other .49.0 .......................... ~~~ 16.4% © TeleGeograph~, Inc. 2000 Total ........................... 298.7 © TeleBeography, tnc, 2000 National Traffic Balance Minutes 1997 1998 1999 .... ! nF~¯9~ming ........................................... ~.~8-5 .................... ̄358.6 ................ 3~7.8 ...... Sq~plus ,(Defici~)~ .............................. ~30:,!~ ................. 62.~ ......... 2~.1 .......... Total Volume 686.9 655.0 626.5 Note: Data are in millions of rninutes of outgoing public switched telecornrnumca’oons traffic. Data based on billing point of traffic. Data exclude some cross-border traffic with Laos, Malaysm, and Myanrnar. © TeleGeography, Inc. 2000 266 @ TeleGeographg, Inc. 2000 TeleGeo~raphy 2001 Largest Telecommunications Routes, 1999 Destination Minutes (thousands) 1. France ........................ 2,336.2 2. Benin ......................... 1,009.4 3. COte d’lvoire .................... 923.2 4. Burkina Faso ..... ; .............. 496.4 5. Ghana ......................... 417.1 6. United States ................... 400.8 7. Germany ....................... 375.5 8. Senegal ........................ 260.6 9. Niger .......................... 256.5 10, Nigeria ......................... 224.6 11, Belgium ........................ 155.7 12, Mall ........................... 154.7 13. Gabon ......................... 151.8 14. United Kingdom ............ .... 149.6 15. Switzerland .......... ., .......... 107.6 16. Canada ......................... 85.8 17, Italy ............................ 75.6 18. Cameroon ....................... 72.9 19. Netherlands ..................... 68.5 20. China ........................... 53.3 Other .......................... 705.0 Percentage of Outgoing Traffic ~~~l~: 10.9% ~, 4.7% 4.4% 3,0% 1.8% 1.8% 1.8% 1.8% 1.3% 1.0% o.~% 0.8% 0,8% 0.6% ~~ ~.3% Total ......................... 8,478.6 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 !~€om,,!ng ...................... n:a.. .............. .1.7...! ........... 2!.6, ,O~,t,,gging ................................. 7:9 ............... .8...4 ........................ ~.5 Su,rp!,,U,s (D,,eficit) ............................ n~.a..._. 8.7 ................ 13.1. .... Total Volume n.a. ~5.5 30.1 Note: Data are ~n millions of m=nutes of outgo=ng public switched telecommunications traff=c. © TeleGeography, Inc, 2000 267 TeleGeography 2001 f Trinidad & Tobago © TeleGeographg, Inc. 2000 Largest Telecommunications Routes, F’Y 1999/00 Destination Minutes (millions) Percentage of Outgoing Traffic 2. Canada .......................... 6.0 ~8.9% 3. United Kingdom .4.7 7.0% 4, Barbados ......................... 3.7 ~5.6~ 5. Grenada ......................... 2.9 4.4% 6. Guyana .......................... 2.2 3.3% 7. Jamaica ......................... 2.1 ~3.1% 8. Saint Vincent &the Grenadines ...... 2.0 2,9% 9. Venezuela ........................ 1.9 2.8% 10. Saint Lucia ....................... 1.7 2.5% 11. Antigua & Barbuda ................ 1.0 1.5% 12. Dominica ......................... 0.5 0.7% 13. Germany’, ......................... 0.5 0.7% 14. Netherlands Antilles ............... 0.5 0.7% 15. Saint Kitts & Nevis ................. 0.5 0.7% 16. British Virgin Islands ............... 0.4 0.6% 17. Bahamas ......................... 0.3 !0.4% 18. Cayman Islands ................... 0.3 //°.4% Other ............................ 5.5 8.2% Total ............................ 67.2 © TeleGeography, Inc. 2000 National Traffic Balance Mi~p~F_S ........................ 1997/,98 .... .F3f 1998/99 ................ 1~999/OP Total Volume 192.0 208.0 226.0 Note: Data are m milhons of minutes of outgoing public switched telecommunications traffic. Data based on bilhng point of traffic. Fiscal year ends 31 March. 268 © TeleGeography, Inc. 2000 © TeleGeographg, Inc. 2000 TeleGeography 2001 Turkey Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic 1. Germany ...................... .204.9 ,,",’" "" ~,,"~ "’~’-~’,’.~ ,~ ~4~’~- ~2~ ~" "~’ "~’, ’"::"~ -~", -~ 29.3% 2. United Kingdom .48.6 ~-~-~,i~ 7 noz ................. .u/o 3. France .......................... 35.4 5.1°/o 4. Netherlands ,30.4 l~ 44% .................... 5. United States .26.5 3.8~ 6. Bulgaria ......................... 20.0 ~t~ii~ 2.9% 7. Romania ........................ 18.4 ~2.6% 8. Switzerland ...................... 17.0 2.4% 9. Austria .......................... 16.6 2.4% 10. Italy ............................ 15.3 2.2% 11. Belgium ......................... 13.5 1.9% 12. Russia .......................... 13.3 ~1.9% 13. A~erbaJjan ....................... 11.3 1.6% 14. Greece .......................... 11.1 1.6% 15. Ukraine ......................... 10.3 1.5% 16. Iran ............................. 7.5 1.1% 17. Moldova ......................... 7.1 1.0% 18. Israel .6.6 ~/0.9% ........................... ~.~ 19. Saudi Arabia ...................... 6.2 0.9% 20. Sweden .......................... 6.0 0.9% Other .172.4 ~~~~ 24.7% Total ........................... 698.4 © TeleGeographv, lnc, 2000 National Traffic Balance Minutes 1997 1998 1999 i.,n, co~.!ng ............................................. 83~o ......... 9..5_5.9 ........... !,..,.,.122.:~ .... ,,, o~,,u,..t...g0ing ...................................... .5.57....~ ............... ~44:! ............. 69,8.4 ¯ ..S..urplus, (De, fici~,t), ........................ 2.7.8:5 ............. 3!,1.7 ............ ~2.4.,~ .,. Total Volume 1,393.5 1,600.0 1,821.1 Note: Data are =n millions of m~nutes of outgoing public switched tetecommunications traffic. Data exclude some cross-border traffic to Iran © TeleGeography, Inc. 2000 269 TeleGeography 2001 © TeleGeographg, Inc. 2000 ( Turkmenistan Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic Russia ........................ ...5.5 ~~~~~~~ ~. "! ~,’~’° ’~ ~® 33.3~/o~ Uzbekistan ....................... 1.1 6.7% Ukraine .......................... 1.0 6.1O/o Azerbaijan .. ..................... 0.9 5.6% Kazakhstan ....................... 0.7 ~!~ 4.2o/o Armenia .......................... 0.4 ~’~ 2.4% Belarus .......................... 0.3 1.6% Tajikistan ......................... 0.3 1.8% Georgia .......................... 0.2 1.2% Kyrgyzstan ....................... 0.2 1.2% Moldova ......................... 0.2 1.2% Other ...5.8 ......................... 35.2% Total ............................ 16.5 2. 3. 4. 6. 7. 8. 9. 10. 11. © TeleBeography, In c. 2000 National Traffic Balance Minutes 1997 1998 1999 In,~0m!,n~g. ,, ..................... n-a:~ ........... n~.~a~ .................... ~.a.~ ¯ ¯ Total Volume n.a. n.a. n.a. Note: Data are ~n milhons of minutes of outgoing public switched telecommumcations traffic. The "Other" category may include routes to non-members of the Commonwealth of Independent States that rank among the top destinatmns for outgoing traffic. © Tele6eography~" Inc. 2000 © TeleGeography, Inc. 2000 TeleGeography 2OO1 Ukraine Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic 2. Belarus ......................... 19,8 ~5.5% 3, rvloldova ........................ 15.0 4.2% 4. Gerrnany ........................ 13.9 3.9% 5. Pofand .......................... 10.3 2,9% 6. Czech Republic .................... 5.2 1.4% 7. Armenia .......................... 5.0 1.4% 8. Italy ............................. 4.3 1.2% 9. United States ..................... 4.1 1.~% 10. Kazakhstan ....................... 4.0 1.1% 11. United Kingdom ................... 3.0 1.0% 12. 6eorgia .......................... 3.5 1.0% 13. H~ngary .......................... 3.4 0.9% 14. Azerbaijan ........................ 3.3 0.9% 15. Turkey ........................... 3.2 0.9% 16. Latvia ............................ 3.2 0.9% 17. Israel ............................. 3.1 ~0.9% 18. Uzbakistan ....................... 3.1 0.9% 19. Lithuania ......................... 2.0 0.7% 20. Bulgaria .......................... 2.3 0.6% Other ........................... 30.9 9.6% Total ........................... 359.2 @ TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ~O.,.u..,,t,,g o i n,.,g .................................... ~,4,86:,,8 ................. 465,:9 ......... 359,2 ....... .~u rp!.,u,s (.D,,,.efi~,!t) ............................ ,n,, a..,: ............. ~,,a,, ........... n,,.a.: .... Total Volume n.a. n.a. n.a. Note: Data are in milhons of minutes of outgoing public switched telecommunications traffic. © TeleGeography, Inc. 2000 271 TeleGeo~raphy 2001 ( United Arab Emirates Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic 1. India ........................... 238.2 ~~!~~ ~. 24.7% 2. Pakistan .94.7 ~~ 9.8% 3, Egypt ........................... 56.0 ~~ 5.8% 4. United Kingdom .52.0 ~~ 5.4% 5. SaudiArabia ..................... 50.4 5.2% 6. Syria ............................ 37.6 3.9% 7. Oman ........................... 36,0 ~3.7% 8. United States .34.0 o ................... ~,~,~ 35~/o. 9. Iran ’ .27.2 ~2.8% 10. Jordan .......................... 25.2 ~2.6% 11. Philippines ....................... 22.4 ~2.3% 12. Lebanon ......................... 22.3 2.3% 13. Qatar ..’. ......................... 21.6 ~2.2% 14. Bahrain ......................... 19.5 2.0% 15. Sudan .......................... 18.7 1.9% 16. Bangladesh ...................... 18.5 1.9% 17. Kuwait .......................... 16.7 1.7% 18. Sri Lanka ........................ 12.7 1.3% 19. Canada ......................... 11.4 1.2% 20. Germany ........................ 11.1 ~iI~ 1.2% Other .......................... 136.8 14.2% © TeleGeography, Inc. 2000 Total ........................... 963.0 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 _~,ln~pm,!n~ ............................................. n~.a~: ................ n,.,.~. ........ Total Volume n.a. n.a. n.a. Note: Data are in mdhons of m~nutes of outgoing public switched telecommumcatzons traffic. © TeleGeography, Inc. 2000 272 © TeleCeography, Inc. 2000 TeleGeography 2001 United Kingdom--Outgoing Largest Telecommunications Routes, FY 1999/00 Destination Minutes (millions) Percentage of Outgoing Traffic 1, United States .1,610.9 ~~~ 15.9% 2. Germany ....................... 913.3 ~~ 9.0% 3. Ireland ......................... 809.3 ~~ 8.0% ........................ "%~" ,~’~ 65% 4. France .655.8 ......................... ~..-’~ 5. Spain .435.2 4.3% 6. Italy ........................... 396.6 3.9% 7. Netherlands .................... 348.1 3.4%° 8. Australia .325.6 ~o~ ...................... o.~ ~o 9. Canada ........................ 247.9 2.4% 10. Belgium ........................ 234.6 2.3% 11. Switzerland ..................... 201,1 2.0% 12. Greece ......................... 196.5 1.9% 13. Sweden ........................ 186.3 1.8% 14. India ........................... 181.3 1.8% 15. Turkey ......................... 153.2 1.5% 16. South Africa .................... 150.3 1.5% 17. Pakistan ........................ 136.0 1.3% 18. Portugal ........................ 116.9 1.2% 19. Poland ......................... 116.6 1.1% 20. Austria ......................... 115.7 1.1% Other ......................... 2,609.8 25,7% Total ........................ 10,141.0 © TeleGeography, Inc. 2000 National Traffic Balance Minutes ¯ !~ ~omi,nog~ ........... ..... Putg,~9!ng .......... ~, S~P!U,S (.Def!Fit) Total Volume .............. ,800;0. ........... ................. ........ n.a. 14,625.0 16,994.4 Note: Data are in milhons of minutes of outgoing pubhc switched telecommunicat=ons traffic. Data based on bilhng point of traff=c. Fiscal year ends 31 March. Data =nclude znclude approximately two bilhon mznutes of traffic refiled via the U.K, thus overstating U.K.- originated volumes. ©TeleGeography, Inc. 2000 273 TeleGeography 2001 © TeleGeography, Inc. 2000 f United Kingdom--lncoming Largest Telecommunications Routes, FY 1999/00 Destination Minutes (millions) Percentage of Outgoing Traffic ................. ~’~ 1. U~ited States 1,479.8 21.6% 2. Ireland .629.2 ~~ 9.2% ....................... ~~ 3. Germany ....................... 603.0 8.8% 4. France 552.2 ~~~ 81% ........................ ¯ 5. Spain .......................... 299.1 ~~ 4.4% 6. Canada ........................ 294.8 ~4.3% 7. Australia 289.4 ~~o~ ....................... -.~ 8. Netherlands .................... 237.7 3.5% 9. Italy ........................... 228.0 3.3% 10. Sweden ....................... 155.0 2.3% 11. Swi~erland .................... 139.3 2.0% 12. Belgium ....................... 117.2 1.7% 13. Greece .’ ....................... 106.2 ~1.5% 14. South Africa ..................... 91.5 1.3% 15. NewZealand .................... 82.5 1.2% 16. Singapore ....................... 78.6 1.1% 17. Hong Kong ...................... 76.2 1.1% 18. Norway ......................... 70,2 1,0% 19. Austria ......................... 63.2 0.9% 20. Denmark ........................ 63.0 0.9% Other ........................ 1,197.3 ~~ 17.5% Total ......................... 6,853.4 © TeleGeography, Inc. 2000 274 © TeleGeogra~hy, Inc. 2000 TeleGeography 2001 U.K. Top 100 Correspondents Outgoing Incoming Outgoing Incoming Country Q3 98/99 Q3 99/00 Q3 98/99 Q3 99/00 Country Q3 98/99 Q3 99/00 Q3 98/99 Q3 99/00 Albania 1.7 2.3 0.2 0.7 Kuwait 4.4 3.0 2.1 1.0 Algeria 3.5 3.2 2.1 2.7 Latvia 4.0 4.0 0.5 0.6 Argentina 2.7 3.3 1.1 1.3 Lebanon 4.4 8.5 1.6 4.7 Australia 65.3 78.1 67.6 80.0 Lithuania 1.8 5.0 0.3 0.3 Austria 25.7 25.3 7.2 15.1 Luxembourg 6.5 7.0 3.9 4.5 Bahrain 2.2 2.3 1.6 2.2 Macedonia 1.6 3.4 0.4 1.1 Bangladesh 5.8 7.7 2.3 2.7 Malaysia 5.8 7.4 3.7 7.4 Barbados 2.3 2.7 1.0 0,9 Mali 2.6 2.2 0.7 0.1 Belg=um 37.2 63.2 33.3 29.0 Malta 4.0 3.5 3.1 2.9 Brazil 6.7 11.2 4.1 4.7 Mauritius 3.2 3.4 2.1 3.5 Bulgaria 3.8 4.1 0.9 0.9 Mexico 2.8 3.7 1.3 1.7 Cameroon 1.5 3.1 0.3 0.3 Morocco 7.3 9.0 2.7 3.8 Canada 48.8 63.9 53.8 72.4 Netherlands 63.2 86.3 55.7 57.4 Chile 3.6 4.8 0.9 1.6 New Zealand 15.0 13.9 14.2 24.6 China 6.7 11.7 3.6 5,0 Nigeria 16.4 14.3 4.9 5.6 Colombia 7.0 5.5 1.2 2.2 Norway 21.6 20.1 21.1 17.9 Croatia 3.3 4.8 1.7 1.7 Oman 1.9 2.2 1.5 1.5 Cuba 1.7 2.0 1.5 0.4 Pakistan 23.5 33.2 4.0 4.4 Cyprus 10.9 10.3 7.6 6.5 Peru 1.5 5.2 0.4 0.4 Czech Republi~ 11.3 8.9 4.5 4.5 Philippines 5.3 8.9 1.3 3.8 Denmark 23.5 22.3 16.6 14.9 Poland 28.0 37.1 9.4 11.5 Dominican Rep. n.a. 1.9 n,a. 0.4 Portugal 19.8 26.6 13.1 15.0 Ecuador 1.7 4.8 0.2 0.8 Romania 5.8 6.0 1.8 1.9 Egypt 9.1 14.0 2.9 7.0 Russia 14.7 16.6 3.5 5.3 Finland 11.1 13.6 5.4 7.3 Saudi Arabia 7.7 9.8 6.9 7.5 France 135.0 141.2 106.0 130.0 Senegal 1.3 3.9 0.2 0.4 Gambia 1.7 4.3 0.3 0.9 Singapore 8.8 10.3 5.2 19.7 Georgia n.a. 2.6 n.a. 0.1 SIovak Republic 4.9 5.1 2.0 1.5 Germany 177.9 232.9 81,6 156.8 Slovenia 1.6 8.1 0.9 1.1 Ghana 9.2 12.1 2.7 3.4 South Africa 25.5 43.7 21.0 22.3 Gibraltar 2.4 2.8 1.7 1.9 Spain 83.1 104.4 59.0 69.3 Greece 31.0 48.0 25.0 28.2 Sri Lanka 6.5 5.0 1.2 1.9 Guyana n.a. 2.7 n.a. 0.6 Sweden 37.2 49.2 38.3 37.9 Honduras n.a. 2.2 n.a. 0.1 Switzerland 35.7 46.8 23.7 21,6 Hong Kong 22.4 21.4 7.6 19.6 Syria 4.3 4.9 0.6 1.4 Hungary 13.2 16,2 4.0 3.9 Taiwan 3.7 4.1 2.3 2.9 Iceland 1.7 3.2 1.4 1.5 Tanzania 2.2 4.5 0.6 0.7 India 33.0 39.7 9.4 15.2 Thailand 5.2 6.4 3.6 5,0 Indonesia 2.0 3.4 2.0 2.2 Trinidad & Tobago 2.0 2.4 1.2 1,3 Iran 5.2 5.4 3.1 3.0 Tunisia 3.9 6.4 0.9 1,1 Iraq n.a. 2.1 n.a. 0.2 Turkey 30.0 44.5 12,6 13,5 Ireland 197.4 213.5 149.2 149.9 U.A.E. 11.1 14.0 9.4 13,3 Israel 10,7 14.2 6.4 7.2 Uganda 1.8 2.1 0.6 0.7 Italy 90.7 94.0 43.8 53.5 Ukraine 8.6 3.4 1.3 1,3 Jamaica 6.5 10.8 1.3 2.5 United States 323.0 390.8 344.0 387.3 Japan 20.4 27.1 11.5 14.5 Venezuela n.a. 2.1 n.a. 0.7 Jordan 3.9 3.6 1.5 2.2 Vietnam 5.0 6.7 4.3 3.7 Kazakhstan 1.8 3.2 0.6 0.4 Yugoslavia 4.9 10,6 1.3 2.6 Kenya 4.4 4.1 1.8 0.4 Zaire n.a. 2.1 n.a. 1.2 Korea, Rep. 2.5 4.4 2.7 2,8 Zimbabwe 1.9 2.4 2.6 3.3 Note: Data are mdlions of m~nutes of pubhc switched traffic for the largest U.K international carriers during the third quarters (October-December) of FY 1998/1999 and FY 1999/2000. Route data may include some calls ref~/ed wa the U.K., thus overstating actual U,K.-originated traffic. Source: Offzce of Telecommunicatzons (OFTEL) ©TeleGeography, Inc. 2000 275 TeleGeography 2001 © TeleGeography, Inc, 2000 United States---Outgoing Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic 1. Canada ....................... 4,491.3 ~~i~~~ ’~ 15.2% 2. Mexico ....................... 4,084.3 13.8% 3. United Kingdom ................ 1,909,6 ~,,~,~,-~ ~6~,’/~ 4. Germany ...................... 1,525.3 5.2% 5. India ........................... 996.4 ~3.4% 6. Japan .......................... 874.2 3.0% 7. Philippines ...................... 773.4 2.6% 8. Italy ........................... 772.0 2.6% 9. Dominican Republic .............. 626.8 2.1% 10. France ......................... 626.3 2.1% 11. Brazil .......................... 623.5 ~2.1% 12. Australia ....................... 420.5 1.4% 13. China ’ ...416.0 1.4% 14. Jamaica ........................ 364.3 1.2% 15. Israel .......................... 362.6 1.2% 16. Korea, Rep ...................... 351.4 1.2% 17. Taiwan ......................... 333.4 1.1% 18. Spain .......................... 325.2 ~1.1% 19. Co/ombia ....................... 318.8 1.1% 20. Pakistan ..300.7 ...................... 1~ 1.0% ....................... ¯ "~ ~, ~. ~" i~ =~ o Other .9,112.7 3 30.8 Total ........................ 29,608.8 © TeleGeography, Inc. 2000 National Traffic Balance Note: Data are in mdhons of m~nutes of outgoing public switched telecommumcations traffic. Data based on billing point of traffzc. Data include znclude at least Wvo bilhon m~nutes of traffic refded vza the U.S., thus overstating traffic originating from the U.S. Carriers and traffic from points beyond the United States, Puerto Rico, and the U.S. %firgin Islands are excluded. © TeleGeography, Inc. 2000 276 © TeleGeographg, Inc. 2000 TeleGeography 2001 United States--Incoming Largest Telecommunications Routes, 1999 Destination Minutes (millions) 1. Canada ...................... 3,920.9 2. United Kingdom ............... 1,126.4 3. Mexico ....................... 1,124.1 4. Australia ....................... 437.7 5. Oermany ....................... 327.2 6. Japan ......................... 305,7 7. Israel .......................... 237.4 8. France ........................ 224.7 9. Italy ........................... 201.1 10. Korea, Rep ...................... 199.5 11. Brazil .......................... 185.9 12. Sweden ....................... 144.0 13. N~therlands .................... 136.2 14. Dominican Republic ............. 122.0 15. Taiwan ........................ 121.2 16. Hong Kong ..................... 103.6 17. Switzerland ..................... 81.7 18. Spain ........................... 74.0 19. Colombia ....................... 57.1 20. Jamaica ........................ 54.2 Other ........................ 1,456.2 Percentage of Outgoing Traffic ~~,~ 10.6% o.6% 2.9% 2.2% 2.1% 1.9% 1.9% 1.7% 1.4% 1.3% 1.1% 1.1% 1.0% 0.8% 0.7% 0.5% ’ 0.5% Total ..................... 10,640.8 © TeleGeography, Inc. 2000 277 TeleGeography 2000 © TeleGeographg, Inc. 1999 Top 100 Correspondents Country Antigua & Barbuda 66.7 62.3 Argentina 234.0 114.2 Aruba 25.8 26.2 Australia 459.2 420.5 Austria 76.0 108.0 Bahamas 82.7 94.4 Bangladesh 52.7 65.9 Barbados 47.7 52.6 Belarus 12.6 15.1 Belgium 116.7 135.1 Belize 15.1 17.8 Bermuda 42.8 49.3 Bolivia 52.8 42.9 Brazil 609.5 623.5 Bulgaria 14.6 29.4 Canada 4,024.6 4,491.3 Cayman Islands 25.8 25.7 Chile 128.2 93.0 China 446.6 416.0 Colombia ’224,3 318.8 Costa Rica 74,1 82.4 Croatia 24.0 43.6 Cuba 141,8 77.5 Cyprus 11,1 17.6 Czech Republic 30.3 54.7 Denmark 75.8 72.2 Dominica 29,3 20.6 Dominican Rep. 453.2 626.8 Ecuador 113.5 192.8 Egypt 113.8 168.0 El Salvador 198.9 235.3 Outgoing Minutes Incoming Minutes Outgoing Minutes Incoming Minutes 1998 1999 1998 1999 Country 1998 1999 1998 1999 7.1 6.3 Jamaica 291.5 364.3 51.9 54.2 46.1 32.4 Japan 839.5 874.2 336.4 305.7 5.7 6.6 Jordan 51.6 46.8 6.8 6.9 261.3 437.7 Kenya 26.5 35.4 4.1 3.8 26.1 22.6 Korea, Rep. 399.9 351.4 219.7 199.5 49.7 42.2 Kuwait 50.7 52.7 8.8 5.5 4.7 3.7 Lebanon 57.7 74.1 11.2 6.1 12.6 9.4 Lithuania 7.0 16.4 1.1 0.9 1.1 0.8 Luxembourg 16.0 17.4 6.9 6.7 39.7 34.6 Malaysia 103.7 72.8 25.2 19.6 5.0 3.9 Mauritius 14.0 15.5 0.6 0.2 30.0 19.6 Mexico 3,121.1 4,084.3 1,086.3 1,124.1 5.5 6.5 Morocco 74.9 63.4 6.4 6.0 180.2 185.9 Netherlands 280.8 284.3 122.2 136.2 1.6 3.3 New Zealand 75.6 90.3 54.7 42.4 3,359.0 3,920.9 Nicaragua 62.3 112.3 8.2 13.7 14.2 10.0 Nigeria 163.9 132.4 17.4 9,2 81J 52.7 Norway 90.6 80.0 51.8 41.0 79.4 48.9 Pakistan 213.5 300.7 15.6 8.8 51.2 57.1 Panama 69,6 68,7 18.9 16.9 26.6 24.0 Paraguay 16.2 17.8 3.7 2.9 5.1 4.3 Peru 185.5 232.4 31.4 27.5 1.2 7,4 Philippines 589.5 773.4 50.7 37.9 4.6 5.4 Poland 189.5 254.8 27.9 24.8 9.9 9.1 Portugal 72.0 93.1 23.1 13.2 26,5 19.7 Romania 76.1 61.6 31.5 7.5 2.1 2.0 Russia 176.3 17~3.1 22.0 16.6 111.4 122.0 Saint Lucia 20.1 21.4 3.5 4.2 13.6 14.3 Saudi Arabia 120.7 111.0 35.8 32.0 11.7 11.2 Senegal 19.9 32.5 2.0 1.5 16.0 24.7 Singapore 175.7 116.1 65.6 26.3 23.2 1.8 1.9 Slovak Republic 11.2 29.5 3.3 2.5 35.6 14.1 13.3 South Africa 130.8 163.4 36.5 38.9 626.3 270.8 224.7 Spain 254.0 325.2 69.5 74.0 1,525.3 365.9 327.2 Sri Lanka 21.2 28.6 3.1 7.9 47.6 7.8 5.8 Sweden 151.0 173.9 113.8 144.0 182.4 42.8 38.6 Switzerland 199,9 238.8 102.1 81.7 67.4 2.3 3.0 Syria 16,2 24.3 4.1 2.6 178.5 17.9 18.7 Taiwan 334.4 333.4 153.0 121.2 134.2 7.2 6.8 Thailand 130.3 118.5 32.2 23.5 167.6 15.1 14.6 Trinidad & Tobago 105,9 132.1 31.5 27.9 242.1 214.7 103.6 Turkey 113.7 231.5 28.4 27.0 83.0 12.7 11.9 Ukraine 46.1 70.8 6.3 3.8 20.0 8.9 8.3 U.A.E. 60.4 90.7 32.3 32.0 996.4 58.5 44.1 United Kingdom 1,524.7 1,909’.6 1,193.5 1,126.4 98.0 28.6 20,5 Uruguay 36.5 35.5 8.2 7.0 111.9 17.7 20.0 Venezuela 233.3 230.4 60.2 49.2 193.5 69.0 50.1 Vietnam 201.9 246.8 4.5 7.4 362.6 171.4 237.4 Yemen 29.8 20.7 1.3 1.2 772.0 134.0 201.1 Yugoslavia 33.9 27.7 8.4 3.2 Ethiopia 22.0 Finland 32.1 France 593.8 Germany 1,175.8 Ghana 44.8 Greece 116.3 Grenada 63.8 Guatemala 145.6 Haiti 100.0 Honduras 150.0 Hong Kong 623.0 Hungary 39.4 Iceland 14.1 India 776.9 Indonesia 103.4 Iran 61.1 Ireland 161.0 Israel 221.9 Italy 546.9 Note: All data are millions of m~nutes of pubhc switched and international Simple Resale (ISR) traffic Because data are based on the bdling point of the traff=c, route data maynotexactlycorrespondwithtraff~cvo{umesasmeasured bythe originating pointoftraffic~see Methodology on page285) Carners and traff~c from outside the U.S. states, Puerto Rico, and the U.S. Virgin Islands (e.g., Guam) are excluded © TeleGeography, Inc. 2000 278 © TeleGeographg, Inc. 2000 TeleGeography 2001 Uruguay Larges~ Telecommunications Routes, 1999 Destination Minutes (millions) 1, Argentina ....................... 41.3 2. Brazil ........................... 10.7 3. United States ..................... 9.1 4. Spain ............................ 3.3 Chile ............................. 2.0 Paraguay ......................... 1.8 Italy ............................. 1.4 Mexico .......................... 1.0 France ........................... 0.9 Peru ............................. 0.8 Canada .......................... 0.6 Germany ......................... 0.6 V~ezuela ........................ 0.5 Switzerland ....................... 0.5 Australia ......................... 0.4 Colombia ......................... 0.4 United Kingdom ................... 0.4 Cuba ............................ 0.3 Ecuador .......................... 0.2 Bolivia ........................... 0.2 Other ............................ 3.9 6. 7. 8. 9. 10. 11. 12. 13. 14. 17. 18. 19. 20. Percentage of Outgoing Traffic 13.4% 4.1% ~!i~ 2.5% 2.2% 1.8% 1.2% 1,1% 0.9% 0.8% 0.7% 0.6% 0.6% 0.5% 0.4% ! 0.4% i 0.3% ~0.3% 10.2% 4.9% Total ............................ 80.1 © Te]eGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ,,,~,ut~ging ................................... ~8.4,. .................. 7~.~ ................ 80.1 Total Volume 162.1 175.3 178.4 Note: Data are in mzllions of rnznutes of outgoing public swztched telecommunicatmns traffzc. © Tel~Geography, Inc. 2000 279 TeleGeosraphy 2001 ( Uzbekistan Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic 1. Russia ......................... .29.0 ~,,, ,~ "~@~ ,~o ~i~i~;~,,~!~ 42.3% 2. Kazakhstan ...................... 10.3 ~~ 15.0% 3. Ukraine .......................... 3.6 5.3% 4. Kyrgyzstan ....................... 3.3 ~4.8%o Tajikistan ......................... 2.7 3.9% Turkmenistan ..................... 1.9 2.7% Turkey ........................... 1.9 2.7% Germany ......................... 1.6 2.4% United States ..................... 1.5 2.2% United Kingdom ................... 1.0 1.5% Belarus .......................... 0.9 1.4% Korea, Rep ........................ 0.9 1.3% China ...’ ......................... 0.8 ~1.1% Azerbaijan ........................ 0.7 1.0% Pakistan ......................... 0.6 0.9% France ........................... 0.4 O.6% India ............................. 0.4 0.5% Japan ............................ 0.4 o.5% Italy ............................. 0.4 0.5%. Greece ........................... 0.2 ! 0.3% Other ............................ 6.0 ~~ 8.8% © TeleCeogrophg, Inc. 2000 6. 7. 8. 9. 10. 11. 13. 15. 16. 17. 18. 20. Total ............................ 68.5 © TeleGeogrophy, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 I n~.~ i~ g ....................................... Su[plus (Deficit) Total Volume n.a. 166.5 143.5 Note: Data are in mdhons of m~nutes of outgoing public switched telecommumcations traffic. Data based on bilhng point of traffic. © TeleGeography, inc. 2000 © TeleGeograph~J, Inc. 2000 TeleGeography 2001 Venezuela Largest Telecommunications Routes, 1999 Destination 1, 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. Minutes (millions) LJnited States .................... 65.6 Colombia ........................ 24.9 italy ............................ 11.3 Spain ........................... 10,0 Canada .......................... 9.8 United Kingdom ................... 4.2 Peru ............................. 3.9 Mexico .......................... 3.0 Portugal .......................... 2,5 Argentina ........................ 2.5 Brazil ............................ 2.1 Dominican Republic ................ 2.0 Fiance ........................... 2.0 Chile ............................. 1.9 Ecuador .......................... 1.3 Trinidad & Tobago ................. 1.3 Germany ......................... 1.2 Panama .......................... 1.0 Honduras ......................... 0.9 Netherlands Antilles ............... 0.8 Other ............................ 8.6 Percentage of Outgoing Traffic ~~ ~6.6O/o 6.3°/o 2.6% 2.4% 1.8% 1.5% 1.5% 1.3% ~.2% 0,8% 0,8% 0,8% 0,6% 0.6% 0.5% 5.3% Total ........................... 160.2 © Tele~eography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 ,,!n co_~,!ng ....................... 28~,.9 ................ 2~,~.! .......... 3!5:~ 0~goi~,g ..... ~_ ~,~,~~ ........................ !59:,,2,. ............. !,64..5~ ................. 1~,0.2 S~U rpl~s (Q,.ef,!~cit) ............... ~127.,7 .......... 1~.6 ................ ~155:,2 Total Volume 446.1 462.6 475.5 Note: Data are in millions of minutes of outgoing public swztched telecommunications traffic, Data based on billing point of traffic. © TeleGeography, Inc. 2000 281 TeleGeography 2OO1 © TeleGeography, Inc. 2000 Yugoslavia Largest Telecommunications Routes, 1999 Destination Minutes (millions) Percentage of Outgoing Traffic 1, Germany ........................ 38.0 16.8% 2. Austria .......................... 20.6 ~~~ 9.1 ~/o 3. Croatia .......................... 17,9 ~~~ 7.9%° 4. Switzerland ...................... 16.8 ~~ 7.4% 5. Macedonia ...................... 14.4 ~~ 6.3’/~ 6. Italy .14.2 ~:~l~@~’~ 6.3% ........................... o 7. Hungary ......................... 11.7 ~5,2’/~ 8. Bosnia-Herzegovina. .11.2 ~%~4.9% ............. 9. Slovenia ......................... 8.9 ~~3.9% 10. United States ..................... 8.8 3.9% 11. France ........................... 8.7 ~~3.8% 12. Greece ........................... 7.9 ~3.5% 13. United Kingdom ................... 7.8 3.4’/~ 14. Bulgaria .4.4 ......................... 1.9% 15. Sweden .......................... 4.4 1.9% 16. Russia ........................... 4.1 1.8% 17. Netherlands ...................... 3.3 1.4% 18. Romania ......................... 2.8 1.2% 19. Canada .......................... 2.4 ~" 1.1% 20. Czech Republic .................... 1.7 0.8% Other ........................... 17.0 ~~ 7.5% Total ........................... 227.0 © TeleGeography, Inc. 2000 National Traffic Balance Minutes 1997 1998 1999 !n~0~i,ng ........................................... 332.0~ ................... ~423:~ ............... 49~.~ Total Volume 549.0 642.9 725.8 Note; Data are in mdhons of minutes of outgoing public switched telecornmumcations traffic. Data based on bilhng point of traffic. 282 © TeleGeography, Inc. 2000 © TeleGeograph~l, Inc. 2000 TeleGeography 2001 Zambia Largest Telecommunications Routes, FY 1999/00 Destination Minutes (thousands) Percentage of Outgoing Traffic 1. South Africa ................... 4,554.1 ~~~~ 30.0% 2. United Kingdom ................ 2,244.2 14.8% 3. United States .................. 1,734.6 ~~ 11.4% .................... ~%~ ¯ ~r~’.~’ 4. Zimbabwe ,1,469.8 i 9.7% 5. Japan ........................ 1,221.4 ~~ 8.1% ~. India ........................... 514.7 ~i 3.4% 7. Botswana ...................... 314.9 ~2.1% 8. Kenya .......................... 314.1 2.1% 9. Italy ........................... 308.8 2.0% 10. Tanzania ....................... 277.7 1.8% 11. Germany ....................... 272.2 1.8% 12. Netherlands .................... 257.4 1.7% 13. M’alawi ......................... 254.1 1.7% 14. France ......................... 208.9 1.4% 15. Canada ........................ 166.2 1.1% 16. Belgium ........................ 107.9 t~ 0.7% 17. Namibia ......................... 96.7 0.6% 18. Denmark ........................ 84.9 ~0.6% 19. Sweden ......................... 77.7 ~0.5% 20. Norway ......................... 76.0 !~ 0.5% Other .......................... 800.0 4.0% Total ........................ 15,874.2 © TeleGeography, Inc. 2000 National Traffic Balance o...u...t.gg..!n g ..................................... n:a.: ................. !,3:o5 .......... !,5:9... ......... .....S...urp..!us (p,.ef!.€.!..t) ................................. n.:a.. .................... 6.~ ............... 0....0 .... Total Volume n.a. 33.8 31.8 Note: Data are m millions of m~nutes of outgoing pubhc switched telecommunicat=ons traff=c. © TeleGeography, Inc. 2000 28~ TeleGeography 2001 Zimbabwe © TeleGeography, Inc. 2000 Largest Telecommunications Routes, FY 1999/00 Destination Minutes (millions) Percentage of Outgoing Traffic 1, South Africa .29.7 ~~ 45.3% 2. United Kingdom .................. 11.8 ~~ 18.0% 3. United States ,,.3.7 5.6% 4. Botswana ........................ 3.0 ~4.6% 5. Zambia ........................... 2,5 ~3.8% 6. Ualawi .......................... 1.3 ~i~ 2.1% 7. Mozambique ...................... 1.1 1.7% 8. Germany ......................... 0.8 ~1.3% 9. Kenya ............................ 0.8 i~ 1.2% 10. France ........................... 0.6 1.0% 11. Netherlands ...................... 0.6 0.9% 12. India ...................... ., ...... 0.5 ~0.8% 13. Canada .’ ......................... 0.5 ~0.7% 14. Switzerland ....................... 0.5 0.7% 15. Namibia .......................... 0.4 ~0.7% 16. China ............................ 0.4 ~0.6% 17. Japan ............................ 0.4 ~0,5% 18. Belgium .......................... 0.4 0.5% 19. Angola ........................... 0.3 0,5%0 20. Australia ......................... 0.3 ~0.5% Other ............................ 5.9 ~~ 9.0% Total ............................ 65.6 © TeleGeography, Inc. 2000 National Traffic Balance Note: Data are =n milhons of minutes of outgoing public switched telecommumcations traffzc. 284 © TeleGeogra phy’i In c. 2000 © TeleGeography, Inc. 2000 TeleGeography 2001 Methodolo The traffic statistics in TeleGeography 2001 were compiled pri- marily from an independent survey of telecommunications ser- vice providers. For some countries and carriers, traffic data have been estimated based upon annual reports, government publications, and industry interviews. To enable comparisons of countries’ international traffic statis- tics, TeleGeography has endeavored to apply a consistent methodology. When reviewing the traffic statistics in TeleGeogruphy 2001, however, readers should keep in mind the following issues. Public Switched Network vs. Private Line Traffic Traffic volumes in TeleGeography 2001 are generally reported in minutes. In most cases, the statistics refer to paid minutes on public switched circuits and thus include voice as well as fax traffic. Traffic volumes include traffic carried by wholesale carders that is resold by "pure" resellers. These resellers do not own or lease their owh international transmission facilities. Instead, they resell the traffic of other carriers; thus, pure resale traffic is counted as part of the minutes for the facilities-based carrier whose services are resold. Traffic carried by International Simple Resale (ISR) carriers is also included. ISR carders resell the capacity of international private lines (IPLs) for switched services by interconnecting their IPLs to the public switched network at one or both ends. lllicit Bypass While traffic volumes include ISR, they generally do not include illicit bypass traffic that bypasses the international settlement rate regime. One form of illicit bypass is Voice-over-lnternet- Protocol (VolP). For an overview of Voice-over-IP traffic vol- umes, see "VolP Routes and Traffic." Cross-Border Traffic Neighboring countries may not classify local cross-border traf- fic in the same way. That is, one country may treat some cross- border traffic as domestic, while its neighbor counts all such traffic as international. Billing Point vs. Originating Point of Traffic Unless otherwise stated in the notes to a table, the outbound minutes reported for countries in TeleCeography 2001 refer to outbound traffic originated in the reporting country even if it is billed in another country. In the past, most international calls were billed at the point of origination. The number of billed minutes thus coincided with the volume of outgoing traffic. Billed minutes also included col- lect or reverse charge calls because the calls were set up by an operator in the originating country. However, the recent use of calling card and call-back services has shifted the billing point for many international calls. For example, calls from Italy to the United States (or a third country, such as Argentina) may now be set up and billed in the U.S. Some countries, including the U.S., report international traffic data based solely on the location where the traffic is billed. Consequently, "outbound" traffic data for these countries can include traffic actually originating in another Thus, incoming minutes reported for one country may not match the outgoing traffic on the same route by the correspondent country. Some double counting may also occur For example, a call from Thailand to the U.S. which is billed to a U.S. calling card is reported by the U.S. carrier as outbound U.S. traffic; the same call also may be reported as outbound minutes by Thailand. Accordingly, in countries where calling card and call-back ser- vices are widely used, a year-to-year comparison of national traffic also requires examining the statistics of countries, such as the U.S. and the U.K., where the calls are hubbed. Transit Traffic Unless otherwise stated, TeleGeograph9 2001 excludes refile traffic from the totals of countries acting as transit hubs. Notable exceptions include the U.K. and U.S. statistics, which do include some traffic reoriginated from other countries. Fixed vs. Mobile Traffic Traffic volumes include international calls originated and termi- nated on both fixed and mobile networks. Rounding Rounding may cause the figures on total national incoming and outgoing tra~c to appear inconsistent with other national data. Revised Data Some differences exist between the historical statistics (I 998 or earlier) reported in TeleGeography 2001 and data published in prior TeleGeography reports or Direction of Traffic. The vari- ations reflect corrections and/or revised data subsequently pro- vided to TeleGeography. 285 TeleOeography 2001 © TeleGeography, Inc. 2000 Global Reference: Blue Pages TeleGeography 2001 © TeleGeographg, Inc. 2000 National Telecommunications Indicators (A-K) GDP 1999 Population Main Lines Lines Per Cellular Users International Internet Hosts Countries (US$ billions) 1999 (millions) 1999 (thous.) 100 people 1999 (thous.) Carriers 1999 1999 (thous.) Albania (a) 3.1 3.9 140 3.6 11 1 n.a. Algeria 47.0 30.8 1,600 5.2 72 1 n.a. Andorra n.a. 0.1 35 46.7 18 1 1 Angola 5.9 12.5 96 0.8 17 1 n.a. Arg entin a 281.9 36.6 7,357 20.1 4,434 2 142 Armenia 1.9 3.5 547 15.5 9 1 2 Australia (b) 389.7 18.9 9,857 52.1 6,501 28 1,090 Austria (c) 208.9 8.2 3,939 48.2 4,242 17 263 Azerbaijan (a) 4.5 7.7 730 9.5 180 1 1 Bahamas n.a. 0.3 111 36.9 16 1 n.a. Bahrain (a) n.a. 0.7 165 24.9 134 1 1 Bangladesh 45.8 126.9 433 0.3 149 1 n.a. Belarus (a, c) n.a. 10.3 2,683 26.1 22 1 1 Belgium (a) 245.7 10.2 5,100 50.2 3,193 18 339 Bolivia 8.5 8.1 502 6.2 420 1 1 Bosnia-Herzegovina (c, d) n.a. 3.8 368 9.6 53 3 2 Brazil (a) 760,3 168.0 24,985 14.9 15,033 2 446 Brunei (c) n.a. 0.3 82 25.6 64 2 1 Bulgaria 12.1 8.3 2,833 34.2 350 1 17 Canada (a) 612.0 30.5 19,957 65.5 7,000 49 1,670 Chile 71.1 15.0 3,109 20.7 2,261 10 40 China 9912 1266.8 108,716 8.6 43,296 2 72 Costa Rica (a) 11.1 3.9 803 20.4 143 1 7 COte d’lvoire 11.2 14.5 219 1.5 257 1 1 Cuba n.a. 11.2 434 3.9 5 1 n.a. Cyprus (a, c) 9.0 0.8 424 54.5 148 1 6 Czech Republic (a) n.a. 10.3 3,806 37.1 1,945 1 122 Denma rk (a) 174.4 5.3 3,638 68.5 2,629 18 338 Dominican Republic 17.1 8.4 810 9.7 256 3 7 Egypt 92.4 67.2 4,686 7.0 481 1 2 El Salvador (c) 12.2 6.2 468 7.6 383 10 1 Estonia (a) 5.1 1.4 510 35.3 387 1 30 Finland (c) 126.1 5.2 2,850 55.2 3,364 8 462 France 1,410.3 58.9 34,100 57.9 21,434 50 1,233 Georgia (c) 4.2 5.5 672 12.3 78 2 1 Germany (a) 2,081.2 82.2 48,300 58.8 23,470 40 1,635 Ghana 7.6 19.7 159 0.8 70 I n.a. Greece (c) 123.9 10.6 5,611 52.8 3,300 1 75 Guatemala 18.0 11.1 605 5.5 351 2 2 Guyana n.a. 0.9 64 7.5 2 1 n.a. Hong Kong (a, b) 156.6 6.9 3,869 56.2 3,973 80 115 Hungan/(a) 48.4 10.2 4,109 46.2 1,628 1 120 India (a, b, c) 459.8 998.1 26,511 2.7 1,195 1 23 Indonesia (a, c) 141.0 209.3 6,080 2,9 2,221 2 21 Iran 101.1 66.8 8,371 12.5 491 I 1 Ireland (b) 84.9 3.7 1,770 47.8 1,460 25 64 Israel (a) 99.1 6.1 2,800 45.9 2,800 3 149 Italy 1,150.0 57.3 26,502 46.2 30,296 15 302 Jamaica (a) 6.1 2.6 510 19.9 144 2 n.a. Japan (b) 4,395.1 126.5 62,490 49.4 56,849 50 2,637 Jordan (a) 7.6 6.5 542 8.4 92 1 1 Kazakhstan (c) 15.6 16.3 1,760 10.8 50 3 4 Korea, Rep. (c, d) 406.9 46.5 20,518 44.1 23,443 24 283 Kuwait 29.6 1.9 456 24.0 300 1 4 Kyrgyzstan (c) n.a. 4.7 356 7.6 3 1 4 Source: TeleGeography research, ITU, and Wor/d Development Report2000/2001, World Bank, September 2000 288 © TeleGeographg, Inc. 2000 TeleGeogFaphy 2001 International Telephone Traffic (A-K) 1998 49.2 121.3 47.4 27.3 358.7 56.6 1,690.0 1,160.0 42.9 63.5 124.4 41.8 176.1 1,460.0 31.6 94.9 545.8 23.4 96.0 4,805.0 259.4 1,711.5 82.7 57.3 29.0 182.0 317.4 710.0 157.5 127.3 43.1 75.1 410.8 4,115.0 5,870.0 28.9 681.3 60.0 14.3 1,879.8 296.3 436.2 324.5 177.0 885.0 661.0 2,640.0 60.1 1,895.0 Outgoing MiTr (millions) Incoming MiTT (millions) Traffic Balance 1999 % Change 1998 19~J % Change 1998 1999 74.6 51.7% 83.6 121.7 30.0% 44.4 47.1 n.a. n.a. n.a. n.a. n.a. n.a. n.a. 53.2 12.2% 32.2 n.a. n.a. -15,2 n.a. 35.0 28.0% 22.3 33.1 48.4% -5.0 -1.9 377.6 5.3% n.a. n.a. n.a. n.a. n.a. 33.7 -40.4% 94.0 89.8 -4.5% 37.4 56.0 2,115.0 25.1% n.a. n.a. n,a. n.a. n.a. 1,350.0 16.4% n.a. n.a. n.a, n.a. n.a. 32.2 -24.9% 46.0 68.6 49.0% 3.2 36.4 n.a. n.a. 90.0 n.a. n.a. 26.5 n.a. 134.1 7.8% 102.1 106.5 4.3% -22.3 -27.5 n.a. n.a. 196.2 n.a. n.a. 154.4 n.a. 161.2 -8.5% 193.5 195.6 1.1% 17.3 34.4 1,590.0 8.9% n.a. n.a. n.a. n.a. n.a. 29.7 -6.1% 76.4 82.2 7.6% 44.8 52.5 97.2 n.a. t59.2 200.5 n.a. 64.3 10~3 574.8 5.3% 806.9 838.5 3.9% 261.1 263.7 18.8 -19.7% 25.5 21.7 -15.1% 2.1 2.9 98.9 3.0% 201.0 n.a. n.a. 105.0 n.a. 5,680.0 18.2% n.a. n.a. n.a. n.a. n.a. 270.0 4.1% n.a. n.a. n.a. n.a. n.a. 1,950.0 13.9% n.a. n.a. n.a. n.a. n.a. 94.1 13.8% 112.9 109.0 -3.4% 30.2 14.9 n.a. n.a. 46.6 n.a. n.a. -10.7 n.a. 32.6 12.2% 203.0 225.3 10.9% 174.0 192.7 168.2 -7.6% 120.6 134.1 11.2% -61.4 -34.0 364.0 14.7% 406.9 452.2 11.1% 89.5 88.2 800.0 12.7% n.a. n.a. n.a. n.a. n.a. 185.7 17.9% 730.5 920.0 25.9% 573.0 734.3 171.0 34.3% 475.3 554.6 16.7% 346.0 383.6 47.0 9.1% 149.2 n.a. n.a. 106.1 n.a. 74.6 -0.7% 79.2 84.8 7.1% 4.1 10.2 423.9 3.2% n.a. n.a. n.a. n.a. n.a. 4,950.0 20.3% n.a. n.a. n.a. n.a. n.a. 46.7 n.a. n.a. 65.7 n.a. n.a. 19.0 6,965.0 18.7% n.a. n.a. n,a. n.a. n.a. 30.1 4.3% 100.8 118.4 17.4% 72.0 88.2 725.7 6.5% 710.1 794.2 11.8% 28.8 68.5 83.3 38.9% n.a. 208.6 n.a, n.a. 125.3 16.1 t2~3% 93.8 101.0 7.7% 79.5 94.9 2,721L3 44.7% 1,t133.0 1,747.2 -4~7% -46.8 -973.1 343.9 16.1% 374.5 n.a. n.a. 78.2 n.a. 473.3 8.5% 1,498.8 1,772.5 18.3% 1,062.6 1,299.2 269.6 -16.9% 434.2 n.a. n.a. 109.7 n.a. 200.4 13.2% 185.7 216.3 16.5% 8.8 15.9 1,015.0 14.7% n.a. n.a. n.a. n.a. n.a. 804.0 21.6% 424.0 n.a. n.a. -237.0 n.a. 3,100.0 17.4% n.a. n.a. n.a. n.a. n.a. 66.4 10.5% 349.8 335.9 -4.0% 289.7 269.5 1,956.6 3.3% 1,575.0 1,929.6 22.5% -320,0 -27.0 122.6 132.5 8.1% 176.9 n.a, n.a. 54.4 n.a. 118.9 104.5 -12.1% 137.5 149.8 9.0% 18.6 45.3 907.7 898.0 n.a. 719.4 n.a. n.a. -188.3 n.a. 173.1 170.0 -1.8% 135.0 120.0 -11.1% -38.1 -50.0 30.4 23.5 -22.8% 30.1 n.a. n.a. -0.3 n.a. Notes: Data are in millions of minutes of public switched t~affic. a. International minutes based on billing point of t~affic. b. International t~affic for year ending 31 March. Australia, Mauritius, and Pakistan ends 30 June. c. Traffic data exclude some carriers or routes. (See countz7 table for details) d. 1998 and 1999 traffic data not directly comparable. (See country table for details) Countries Albania (a) Algeria Andorra Angola Argentina Armenia Australia (b) Austria (c) Azerbaijan (a) Bahamas Bahrain (a) Bangladesh Belarus (a, c) Belgium (a) Bolivia Bosnia-Herzegovina (c, d) Brazil (a) Brunei (c) Bulgaria Canada (a) Chile China Costa Rica (a) C~te d’lvoire Cuba Cyprus (a, c) Czech Republic (a) Denmark (a) Dominican Republic Egypt El Salvador (c) Estonia (a) Finland (c) France Georgia (c) Germany (a) Ghana Greece (c) Guatemala G~ana Hong Kong (a, b) Hungary (a) India (a, b, c) Indonesia (~z, c) Iran Ireland (b) Israel (a) Italy Jamaica (a) Japan (b) Jordan (a) Kazakhstan (c) Korea, Rep. (c, d) Kuwait Kyrgyzstan (c) © TeleGeography, Inc. 2000 289 TeleGeography 2001 © TeleGeographgo Inc. 2000 National Telecommunications Indicators (L-Z) GDP 1999 Population Main Lines Lines Per Cellular Users International Internet Hosts Countries (US$ billions) 1999 (millions) 1989 (thous.) 100 people 1989 (thous.) Carriers 1999 1999 (thous.) Latvia (a) 6.7 2.4 732 30.0 274 1 19 Luxembourg n.a. 0.4 311 72.4 209 3 10 Macau (a) n.a. 0.4 178 40.8 89 1 n.a. Macedonia (c, d) 3.4 2.0 471 23.4 50 1 1 Malaysia (a, b) 74.6 21.8 4,431 20.3 2,860 5 59 Malta n.a. 0.4 198 51.2 38 1 6 Mauritius (b) n.a. 1.1 257 22.4 102 1 1 Mexico (a) 475.0 97.4 10,927 11.2 7,622 16 405 Moldova (a) 1.1 4.4 555 12.7 17 1 1 Morocco 35.2 27.9 1,467 5.3 374 1 2 Mozambique 4.2 19.3 78 0.4 12 1 n.a. Myanmar n.a. 45.1 249 0.6 11 1 n.a. Namibia 3.1 1.7 108 6.4 30 1 2 Netherlands 384.8 15.8 9,610 60.6 6,900 30 959 New Zealand (b) 53.6 3.8 1,877 49.0 881 19 271 Nicaragua (a) 2.3 4.9 147 3.0 69 1 1 Norway (a) 145.4 4.4 3,165 71.2 2,745 14 439 Oman (a, c) n.a. 2.5 220 9.0 121 1 1 Pakistan (b, c) 59.9 134.5 2,986 2.2 266 1 5 Palestinian Authority (c) 3.6 3.1 n.a. n.a. n.a. 1 n.a. Panama (a) 9.6 2.8 462 16.4 242 1 1 Paraguay 8.1 5.4 297 5.5 436 1 2 Peru (a) 57.3 25.2 1,689 6.7 1,013 18 9 Philippines (a, b) 75.4 74.5 2,940 3.9 1,734 12 12 Poland (a) 154.1 38.7 10,068 26.0 3,956 1 171 Por~ug a l (c) 107.7 10.0 4,230 42.4 4,672 1 78 Qatar n.a. 0.6 155 26.3 84 1 n.a. Russia (c) 375.3 147.2 30,388 20.6 1,360 35 91 Saudi Arabia n.a. 20.9 3,051 14.6 627 1 3 Senegal 4.8 9.2 166 1.8 74 1 n.a. Singapore (a, b) 84.9 3.2 1,861 57.7 1,532 1 148 SIovak Republic (a) 19.3 5.4 1,655 30.8 918 1 28 Slovenia 20.7 2.0 802 40.3 210 1 24 South Africa 131.1 39.9 5,493 13.8 5,269 1 168 Spain (a) 562.2 39.4 16,480 41.8 12,300 16 470 Sri Lan ka 0.0 18.6 679 3.6 227 1 1 Sudan (c) 0.0 28.9 251 0.9 13 1 n.a. Swaziland (b) 0.0 1.0 31 3.1 11 1 1 Sweden (a) 226.4 8.9 5,889 66.5 5,125 16 523 Switze rl a n d 260.3 7.1 4,992 69.9 2,935 40 270 Syria (a) 19.4 15.7 1,600 10.2 4 1 n.a. Taiwan (a) 0.0 22.1 12,044 54.4 11,541 1 597 Thailand (a, c) 123.9 60.9 5,216 8.6 1,957 2 40 Trinidad & Tobago (a, b) 0.0 1.3 276 21.4 39 1 5 Turkey (c) 188.4 68.2 18,054 26.5 8,000 1 79 Turkmenistan (c) 0.0 4.4 359 8.2 4 1 n.a. Ukraine 0.0 50.7 10,074 19.9 116 1 29 United Arab Emirates 0.0 2.4 975 40.7 832 1 20 United Kingdom (a, b) 1,373.6 58.7 33,750 57.5 27,185 215 1,739 United States (a) 8,708.9 276.2 188,331 68.2 86,047 679 53,176 Uruguay n.a. 3.3 897 27.1 316 1 25 Uzbekistan (a) 0.0 23.9 1,599 6.7 40 1 n.a. Venezuela (a) 103.9 23.7 2,586 10.9 3,400 1 14 Yugoslavia (a) n.a. 10.6 2,281 21.4 606 2 11 Zimbabwe (b) 5.7 11.5 239 2.1 174 1 2 Source: ToleGeography research, ITU, and the World Bank © TeleGeography, Inc., 2000 290 © TeleGeo~raph~l, Inc. 2000 TeleGeography 2001 International Telephone Traffic (L-Z) 1998 55.4 293.8 125.2 37.1 685.0 37.3 29.7 1,310.0 55.8 181.0 17.6 19.1 61.9 1,885.0 610.0 46.5 540.0 90.0 87.5 Outgoing MiTr (millions) Incoming MiTr (millions) Traffic Balance 1999 % Change 1998 1999 % Change 1998 1999 55.6 0.4% 87.2 90.0 3.3% 31.8 34.4 319.1 8.6 % 242.6 277.5 14.4% -51.2 -41.5 132.8 6.1% 95.1 97.7 2.7% -30.2 -35.1 82.3 n.a. 91.7 152.5 n.a. 54.6 70.3 690.0 0.7% n.a. n.a. n.a. n.a. n.a. 39.0 4.7% 43.4 50.2 15.7% 6.1 11.2 31.4 5.7% 39.5 43.3 9.6% 9.8 11.9 1,563.0 19.3% 3,060.0 4,007.5 31.0% 1,750.0 2,444.5 49.0 -12.3% 90.3 101.1 11.9% 34.4 52.1 219.5 21.3% 460.0 n.a. n .a. 279.0 n.a. 20.3 15.8% n.a. 38.8 n.a. n.a. 18,5 17.4 -8.8% 36.3 29.8 -17.9% 17.2 12.4 61,2 -1.1 % 45.3 51.2 13.0% -16.6 -10.0 2,150.0 14.1% n.a. n.a. n.a. n.a. n.a. 815.0 33.6% n.a. n.a. n.a. n.a. n.a. 52.0 11.8% 59.7 72.7 21.8% 13.2 20,7 567.0 5,0% n.a. 386.9 n.a. n.a. -180.1 101.3 12.5% 71.7 83.4 16.3% -18.3 -17.9 75.1 -14.1% 640.4 644.9 0.7% 552.9 569.8 Countries Latvia (a) Luxembourg Macau (a) Macedonia (c, d) Malaysia (a, b) Malta Mauritius (b) Mexico (a) Moldova (a) Morocco Mozambique Myanmar Namibia Netherlands New Zealand (b) Nicaragua (a) Norway (a) Oman (a, c) Pakistan (b, c) 27.6 34.9 26.2% 16.6 n.a. n.a. -11.0 50.0 53.6 7.2% 95.5 95.8 0.3% 45.5 37.8 34.7 -8.2% 57.4 54.8 -4.5% 19.6 90.4 88.9 -1.6% 272.6 302.6 11.0% 182.3 262.0 218.0 -16.8% n.a. n.a. n.a. n.a. 602.4 624.0 3.6% 1,144.2 n.a. n.a. 541.8 462.8 532.8 15.1% 713.8 753.3 5.5% 250.9 112.5 128.5 14.2% 70.0 84.0 19.9% -42.5 1,038.3 928.2 -10.6% 1,029.8 929.3 -9.8% -8.5 932.6 1,060.0 13.7% n.a. n.a. n.a. n.a. 29.6 36.5 23.1% 93.8 111.1 18.5% 64.2 1,235.0 1,350.0 9.3% n.a. n.a. n.a. n.a. 151.8 162.8 7.3% 186.4 208.7 12.0% 34.6 129.6 n.a. n.a. 137.0 n.a. n.a. 7.4 405.0 461.1 13.9% n.a. n.a. n.a. n.a. n.a. Palestinian Authority (c) 42.2 Panama (a) 20.1 Paraguay 213.7 Peru (a) n.a. Philippines (a, b) n.a. Poland (a) 220.5 Portugal (c) -44.5 Qatar 1.1 Russia (c) n.a. Saudi Arabia 74.7 Senegal n.a. Singapore (a, b) 45.9 S]ovak Republic (a) n.a. Slovenia n.a. South Africa 1,675J) 1~35,0 15,5% n.a, n.a, n,a. n.a. 39,3 45.5 15.7% 146,8 n.a. n.a, 107.5 18.4 21.9 19.1% 88.0 105.3 19.7% 69.6 28.4 29.3 3.3% n.a. n.a. n.a. n.a. 1,230.0 1,365,0 11.0% n.a. n.a. n.a. n.a. 2,425.0 2,730.0 12.6% n.a. n.a. n.a. n.a. 103.0 125.6 21.9% n.a. 256.7 n.a. n.a. 862.0 949.3 10.1% 781.8 882.0 12.8% -80.2 296.4 298.7 0.8% 358.6 327.8 -8.6% 62.2 64.4 67.2 4.3% 141.5 158.8 12.2% 77.1 644.1 698.4 8.4% 955.8 1,1.__~.7 17.5% 311.7 15.3 16,5 8.0% n.a. n.a. n.a, n,a. 465.9 359.2 -22.9% n.a. n.a. n.a. n.a. 874.8 963.0 10.1% n.a. n.a. n.a. n.a. 8,225.0 10,141.0 23.3% 6,400.0 6,853.4 7.1% -1,825.0 n,a. ___ Spain ,(a) n.e. Sd Lanka 83.3 Sudan (c) n.a. Swaziland (b) n.a. Sweden (a) n.a. Switzerland 131.1 Syria (a) -67.3 Taiwan (a) 29.1 Thailand (a, c) 91.6 Trinidad & Tobago (a, b) 424.3 Turkey (c) n.a. Turkmenistan (c) n.a. Ukraine n.a. United Arab Emirates -3,287.6 United Kingdom (a, b) 25,163.8 29,608.8 17.7% 10,395.3 10,640.8 2.4% -14,798.5 78.3 80,1 2.3~- 97.0 98.3 1.3% 18.7 91.7 68.5 -25.4% 74.7 75.0 0.4% -17.0 164.5 160.2 -2.6% 298.1 315.3 5.8% 133.6 219.5 227.0 3.4% 423.3 498.8 17.8% 203.8 52.8 65.6 24.3% 53.2 59.0 10.9% 0.4 Notes: Date are in millions of minutes of public switched traffic. a. International minutes based on billing point of traffic. -18,968.0 United States (a) 18.2 Uruguay 6.6 Uzbekistan (a) 155.2 Venezuela (a) 271.7 Yugoslavia (a) -6.6 Zimbabwe (b) b. International tTaffic for year ending 31 March. Australia, Mauritius, and Pakistan ends 30 June. c. Traffic data exclude some carriers or routes. (See country table for details) d. 1998 and 1999 tTaffic data not directly comparable. (See country table for details) 291 TeleOeo~’aphy 2OO1 © TeleGeogrophu, Inc. 2000 International Dialing Codes, by Country Afghanistan ................. 93 Albania .................... 355 firana ................. 42 Algeria ................... 213 Algiers ................. 2 American Samoa ............ 664 Andorra ................... 376 Angola ................... .244 Luanda ................. 2 Anguilla ................. 1-264 Antigua & Barbuda ........ 1-268 Argentina ................... 54 Buenos Aires ............. 1 Armenia ................... 374 Yerevan ................. 1 Aruba .................... 297 Ascension Island .......... 247 Australia ................... 61 Melbourne ............... 3 Sydney .................. 2 Australian Territories ........ 672 Austria ..................... 43 Vienna .................. 1 Azerbaijan ................. 994 Baku .................. 12 Bahamas ................. 1-242 Bahrain .................... 973 Bangladesh ................ 680 Dhaka ................. 2 Barbados ................ 1-246 Rolarus .................... 375 Minsk ................ 172 Rolgium .................... 32 Brussels ................ 2 Bolizo .................... 501 Belmopan ............... 8 Benin .................... 3.2.9 Bermuda ................. 1-441 Bhutan .................... 975 Bolivia .................... 591 La Paz .................. 2 Bosnia-Herzegovina ........ .387 Sarajevo ................ 71 Botswana ................. 267 Brazil ...................... 55 Brasilia ................ 51 Rio de Janeiro ........... 21 S~io Paulo .............. 11 British Indian Ocean Tort.... 246 British Virgin Islands ...... 1-284 Brunei ..................... 673 Bandar Seri Begawan ....... 2 Bulgaria ................... 359 Sofia ................... 2 Burkina Fase ............... 226 Burundi ................... 25? Cambodia .................. 855 Cameroon ................. 2.37 Canada ...................... 1 Montreal ........... 514/450 Ottawa ................ 613 Toronto ............ 416/647 Vancouver .............. 604 Cape Verde ............... 238 Cayman Islands ........... 1-345 Central African Republic .... 236 Bangui ................. 61 Chad ..................... 235 Chile ....................... 56 Santiago ................ 2 China, People’s Republic of .... 86 Beijing ................. 10 6uangzhou .............. 20 Shanghai ............... 21 Colombia ................... 57 Bogota ................. 1 Cocos Islands; Norfolk & Christmas Islands .......... 672 Comoros .................. 269 Congo, Dem. Rep. of ......... 243 Kinshasa ............... 12 Congo, Republic of ......... 242 Brazzaville ......... 81/82/83 Cook Islands ............... 682 Costa Rica ................. 506 COte d’lvoire .............. 3.25 Croatia .................... 385 Zagreb .................. 1 Cuba ....................... 53 Havana ................. 7 Cyprus ..................... 357 Nicosia ................. 2 Czech Republic ............. 420 Prague .................. 2 Denmark .................... 45 Djibouti ................... 253 Dominica ................ 1-767 Dominican Republic ....... 1-809 East ~mor ................ 672-9 Ecuador ................... 593 Quito ................... 2 Egypt ...................... .20 Cairo ................... 2 El Salvador ................. 503 Equatorial Guinea ........... 240 Eritrea ..................... 291 Estonia .................... 372 Tallinn .................. 2 Ethiopia ................... 251 Addis Ababa ............. 1 Falkland Islands ............ 500 Faroe Islands ............... 298 Fiji ...................... 679 Finland .................... 350 Helsinki ................. 9 France ..................... 33 Paris ................... 1 Marseille .............. 491 French Antilles ............. 596 French Guiana .............. 564 French Polynesia ........... 689 Gabon ..................... 241 Gambia .................... 220 Georgia .................... 995 Tbilisi ................. 32 Germany .................... 49 Berlin .................. 30 Bonn ................. 228 Frankfurt ................ 69 Munich ................. 89 Ghana .................... .233 Accra .................. 21 Gibraltar ................... 350 Greece ..................... 30 Athens .................. 1 Greenland ................. 299 Grenada ................. 1-473 Guadeloupe ................ 590 Guam .................... 1-671 Guatemala ................. 502 Guinea ................... 2.24 Guinea-Bissau ............. .245 Guyana .................... 592 Georgetown ............. 2 Haiti ...................... 509 Honduras .................. 504 Hong Kong ................. 852 Hungary ................... .,36 Budapest ................ 1 Iceland .................... 354 India ....................... 91 Mumbai ............... 22 Calcutta ................ 33 New Delhi .............. 11 Indonesia .................. 62 Jakarta ................ 21 Inmarsat Special ................ 870 East Atlantic ............ 871 Pacific ................ 872 Indian ................. 873 West Atlantic .......... 874 International Freephone ...... 800 Iran ....................... 98 Tehran ................. 21 Iraq ...................... 964 Baghdad ................ 1 Ireland .................... 353 Dublin .................. 1 Israel ..................... 972 Jerusalem ............... 2 Tel Aviv ................. 3 Italy ....................... 39 Rome .................. 06 Milan .................. 02 Jamaica ................. 1-876 Japan ...................... 81 Osaka ................... 6 Tokyo ................... 3 Jordan .................... 962 Amman ................. 6 Kazakhstan .................. 7 Almaty ............... 3272 Kenya .................... 254 Nairobi ................. 2 Kiribati .................... 686 Korea, Dem. Rep. of ......... 850 Pyongyang ............... 2 Korea, Republic of ........... 82 Seoul ................... 2 Kuwait .................... 965 Kyrgyzatan ................. 996 Bishkek ............... 312 Laos ..................... .656 Latvia ..................... 371 Riga ................... 2 292 © TeleGeographg, Inc. 2000 TeleGeography 2001 Lebanon ................... 961 Beirut .................. 1 Lesotho ................... .266 Liberia ................... 231 Libya ..................... .218 Tripoli ................. 21 Liechtenstein ............... 423 Lithuania .................. 370 Vilnius .................. 2 Luxembourg ................ 352 Macau .................... 853 Macedonia ................. 389 Skopje ................ 91 Madagascar ............... .261 Malawi ................... .265 Malaysia ................... 60 Kuala Lumpur ............ 3 Maldives .................. 960 Mall ..................... 3.23 Malta ..................... 356 Marshall Islands ............ 692 Martinique ................. 596 Mauritania ................ 2.2.2 Mauritius ................. 230 Mayotte ................... 269 Mexico ..................... 52 Guadalajara .............. 3 Mexico Cih’ .............. 5 Monterrey .............. 8 Micronesia ................ 691 Moldova ................... 373 Chisinau ................ 41 Monaco .................. .377 Mongolia .................. 976 Ulaanbaatar .............. 1 Monteerrat ............... 1-564 Morocco .................. .212 Casablanca ............... 2 Rabat .................. 7 Mozambique .............. .258 Maputo ................. 1 Myanmar ................... 95 Namibia .................. .264 Windhoek .............. 61 Nauru ..................... 674 Nepal ..................... 977 Kathmandu .............. 1 Netherlands ................. 31 Amsterdam ............. 20 Netherlands Antilles ........ 599 New Caledonia ............. 687 New Zealand ................ 64 Auckland ................ 9 Wellington ............. 4 Nicaragua ................. 505 Managua ................ 2 Niger ..................... 2.27 Nigeria ................... 234 Lagos ................... 1 Niue ...................... 683 Northern Marianas ........ 1-670 Saipan ................ 322 Nonvay ..................... 47 0slo ................ 22/23 Oman ..................... 968 Pakistan .................... 92 Islamabad .............. 51 Palestinian Authority ........ 970 Palau ..................... 680 Panama ................... 507 Papua New Guinea .......... 675 Paraguay .................. 595 Asuncion ............... 21 Peru ....................... 51 Lima .................. 14 Philippines .................. 63 Manila .................. 2 Poland ..................... 48 Warsaw ............... 22 Portugal ................... 351 Lisbon ................. 21 Puerto Rico ............... 1-707 Qatar ...................... 974 R6union Island ............ .262 Romania .................... 40 Bucharest .............. 1 Russia ....................... 7 Moscow .............. 095 St. Petersburg ........... 812 Rwanda .................. .250 St. Helena ................. .250 St. Kitts & Nevis ........... 1-869 St. Lucia ................. 1-750 St. Pierre & Miquelon ........ 508 St. Vincent & the Grenadines .............. 1-784 San Marino ................ 378 Sag Tome and Principe ..... 239 Saudi Arabia ............... 966 Riyadh .................. 1 Senegal ................... 221 Seychelles ................. 248 Sierra Leone ............... 23~ Freetown ............... 22 Singapore ................... 65 Slovak Republic ............ 421 Bratislava .............. .7 Slovenia ................... 386 Ljubljana ............... 61 Solomon Islands ............ 677 Somalia ................... 252 Mogadishu .............. 1 South Africa ................ .27 Johannesburg ........... 11 Pretoria ................ 12 Spain ...................... 34 Madrid ................. 1 Barcelona ................ 3 Sri Lanka ................... 94 Colombo ................ 1 Sudan ..................... 248 Khartoum .............. 11 Suriname .................. 597 Swaziland ................. 268 Sweden .................... 48 Stockholm ............... 8 Switzerland ................. 41 Berne .................. 31 Zurich ................... 1 Syria ...................... 963 Damascus .............. 11 Tahiti ...................... 689 Taiwan .................... 886 Taipei .................. 2 Tajikistan .................. 992 Dushanbe .............. 37 Tanzania ................... 255 Dar Es Salaam ........... 22 Thailand .................... 66 Bangkok ." .............. 2 Togo ...................... 228 Tokelau .................... 690 Tonga ..................... 676 Trinidad & Tobago ......... 1-868 Tunisia .................... 216 Tunis ................... 1 Turkey ...................... 90 Ankara ................ 312 tstanbul ............... 212 Turkmenietan ............... 993 Ashkhabad ............. 12 Turks & Caicos ........... 1-649 Tuvalu ..................... 688 Uganda .................... 256 Kampala ............... 41 Ukraine .................... 380 Kiev ................... 44 United Arab Emirates ........ 971 Abu Dhabi ............... 2 Dubai ................... 4 United Kingdom .............. 44 Cardiff ............... 2920 Glasgow .............. 141 London ............ 207/208 Manchester ............ 161 United States ................. 1 Chicago ........ 312/773/872 Houston ........ 713/281/832 Los Angeles ......... 213/323 Miami ............. 305/786 New York ....... 212/646/917 Washington ............ 202 U.S, Virgin Islands ......... 1-340 Uruguay ................... 598 Montevideo .............. 2 Uzbekistan ................. 998 Tashkent ............... 71 Vanuata ................... 678 Vatican City ................ 379 Venezuela .................. 68 Caracas ................. 2 Vietnam .................... 84 Wallis & Futuna ............ 681 Western Samoa ............. 665 Yemen ..................... 967 Sanaa .................. 2 Yugoslavia ................. 381 Belgrade ................ 11 Zambia .................... 260 Lusaka .................. 1 Zanzibar (Tanzania) .......... 255 Zimbabwe ................. 263 Harare .................. 4 293 TeleGeography 2001 © TeleGeography, Inc. 2000 World Dialing Codes RUSSIA 7 CANADA 1 JAPAN 81 .... .~ ~u .. i ~: .... AUSTRALIA 61 NEW ZEALAND ~4 UNITED STATES 1 BRAZIL PERU 55 BOLIVIA 591 PARAGUAY 555 ARGENTINA: ,~, TeleGemGzre, p~ I r~,::, :Z.PGmO 294 © TeleGeographgo Inc. 2000 TeleGeogFaphy 2001 GREENLAND 299 & FRJ ICELAND SEE INSET 213 LI BYA ¯ 218 EGYPT ~HRA ~:~ OATAR 1~ ~,r,,,,, 20 SAUDI RUSSIA JKRAINE KAZAKHSTAN 7 TURKEY IRAN AFGHANISTAN PAKISTAN SOUTH AFRICA L~O’r~o 27 MONGOLIA 976 KOREA, DEM. NEPAL BHUTAN IT/ I N D IA BANB LADESH 91 ~OmUGAL 3~I SPAIN RUSSIA 7 i~BOUR6~ CZECH REP’4~SLOVAKREP ~1 MOLDOVA SWITZERLAND HUNGARY ROMANIA3T~ TeleGeography 2001 © TeleGeography, Inc. 2000 International Dialing Codes, by Number 1 Canada Guam Northern Marianas United States Caribbean 26 Egypt 212 Morocco 213 Algeria 216 Tunisia 218 Libya 220 Gambia 221 Senegal 2.2.2 Mauritania 2.2.3 Meli 224 Guinea 225 COte d’lvoire 226 Burkina Faso 227 Niger Togo 229 Benin 230 Mauritius 231 Liberia 232 Sierra Leone 233 Ghana 234 Nigeria 235 Chad 236 Central African Republic 237 Cameroon 238 Cape Verde 239 Sao Tome & Principe 240 Equatorial Guinea 241 Gabon 242 Congo, Republic of 243 Congo, Dem. Rep. of 244 Angola 245 Guinea-Bissau 246 British Indian OceanTerr. 2~,7 Ascension Island 246 Seychelles 249 Sudan 2_50 Rwanda 251 Ethiopia 252 Somalia 253 Djibouti 254 Kenya 2_55 Tanzania 2_56 Uganda 2_57 Burundi 258 Mozambique 260 Zambia 261 Madagascar 262 R~union 263 Zimbabwe 264 Namibia 265 Malawi 266 Lesotho 500 Falkland Islands 692 Marshall Islands 267 Botswana 501 Belize Kazakhstan 268 Swaziland 502 Guatemala Russia 269 Comoros & Mayotte 503 EISalvador 800 International Freephone 27 South Africa 504 Honduras 81 Japan 290 St. Helena 505 Nicaragua 82 Korea, Republic of 291 Eritrea 506 Costa Rica 84 Vietnam 257 Aruba 507 Panama 850 Korea, Dem. Rep. of 298 Faroe Islands 508 St. Pierre & Miquelon 852 Hong Kong 299 Greenland 509 Haiti 853 Macau 30 Greece 51 Peru 855 Cambodia 31 Netherlands 52 Mexico 856 Laos 32 Belgium 53 Cuba 86 China 33 France 54 Argentina 870 InmarsatSpecial 34 Spain 55 Brazil 871 Inmarsat East Atlantic 350 Gibraltar 56 Chile 872 Inmarsat Pacific 351 Portugal 57 Colombia 873 Inmarsat Indian 352 Luxembourg 58 Venezuela 874 Inmarsat West Atlantic 353 Ireland 590 Guadeloupe 880 Bangladesh 354 Iceland 591 Bolivia 886 Taiwan 355 Albania 592 Guyana ~0 Turkey 356 Malta 503 Ecuador 91 India 357 Cyprus 594 French Guiana 92 Pakistan 358 Finland 595 Paraguay 93 Afghanistan 359 Bulgaria 596 Martinique 94 Sri Lanka 36 Hungaw 597 Suriname 95 Myanmar 370 Lithuania 598 Uruguay 960 Maldives 371 Latvia 599 Netherlands Antilles 961 Lebanon 372 Estonia 60 Malaysia 962 Jordan 373 Moldova 61 Australia 963 Syria 37~, Armenia 62 Indonesia 964 Iraq 375 Belarus 63 Philippines 965 Kuwait 376 Andorra 64 NewZealand 966 SaudiArabia 377 Monaco 65 Singapore 967 Yemen 378 San Marino 66 Thailand 968 Oman 379 Vatican City 672 Australian Territories 970 Palestinian Authority 380 Ukraine 673 Brunei 971 United Arab Emirates 381 Yugoslavia 674 Nauru 972 Israel 385 Croatia 6?5 Papua New Guinea 973 Bahrain 386 Slovenia 676 Tonga 974 Qatar 387 Bosnia-Herzegovina 677 Solomon Islands 975 Bhutan 3~ Macedonia 678 Vanuatu 976 Mongolia 39 Italy 679 Fiji Islands 977 Nepal 40 Romania 680 Palau 98 Iran 41 Switzerland 501 Wallis & Futuna 992 Tajikistan 420 Czech Republic 502 Cooklslands 993 Turkmenistan 421 Slovak Republic 083 Niue 994 Azerbaijan 423 Liechtenstein 504 American Samoa 995 Georgia 43 Austria 685 Western Samoa 996 Kyrgyzstan ~1~, United Kingdom 686 Kiribati 998 Uzbekistan 45 Denmark 687 New Caledonia 46 Sweden 688 Tuvalu 47 Norway 689 French Polynesia 48 Poland 690 Tokelau 49 Germany 691 Micronesia 296 © TeleGeographg, Inc. 2000 TeteGeography 2001 North American Area Codes, by Number 201 New Jersey 33? Louisiana 519 Ontario ?08 Newfoundland 202 Dist. of Columbia 339 Massachusetts 520 Arizona 710 U.S. Government 203 Connecticut 340 U.S, Virgin Is. 525 Mexico City Emergency 204 Manitoba 341 California 530 California 712 Iowa 205 Alabama 345 Cayman Islands 540 Virginia 713 Texas 206 Washington 34? New York 541 Oregon 714 California 20? Maine 351 Massachusetts 958 St. Lucia 715 Wisconsin 208 Idaho 352 Florida 559 California 716 New York 209 California 360 Washington 561 Florida 717 Pennsylvania 210 Texas 361 Texas 562 California 718 New York 212 New York 369 California 564 Washington 719 Colorado 213 California 401 Rhode Island 570 Pennsylvania ?20 Colorado 214 Texas 402 Nebraska 571 Virginia 724 Pennsylvania 215 Pennsylvania 403 Alberta 573 Missouri 727 Florida 216 Ohio 404 Georgia 580 Oklahoma 732 New Jersey 217 Illinois 405 Oklahoma 586 Michigan 734 Michigan 218 Minnesota 406 Montana 601 Mississippi 737 Texas 219 indiana 407 Florida 602 Arizona 740 Ohio 224 Illinois 408 California 603 New Hampshire 747 California 225 Louisiana 409 Texas 604 British Columbia 752 California 228 Mississippi 410 Maryland 605 South Dakota 757 Virginia 229 Georgia 411 Directory Assist. 606 Kentucky 758 St. Lucia 231 Michigan 412 Pennsylvania 607 New York 760 California 234 Ohio 413 Massachusetts 608 Wisconsin 763 Minnesota 240 Maryland 414 Wisconsin 609 New Jersey 764 California 242 Bahamas 415 California 610 Pennsylvania 765 Indiana 246 Barbados 416 Ontario 611 Repair Service 767 Dominica 248 Michigan 417 Missouri 612 Minnesota 770 Georgia 250 British Columbia 418 Quebec 613 Ontario 773 Illinois 252 North Carolina 419 Ohio 614 Ohio 774 Massachusetts 253 Washington 420 Wisconsin 615 Tennessee 775 Nevada 254 Texas 423 Tennessee 616 Michigan 778 British Colombia 256 Alabama 424 California 617 Massachusetts 780 Alberta 262 Wisconsin 425 Washington 618 Illinois 781 Massachusetts 264 Anguilla 435 Utah 619 California 784 St. Vincent 267 Pennsylvania 440 Ohio 623 Arizona & Grenadines 268 Antigua 441 Bermuda 626 California 785 Kansas 270 Kentucky 442 California 627 California 786 Rorida 278 Michigan 443 Maryland 628 California 787 Puerto Rico 201 Texas 445 Pennsylvania 630 Illinois 800 Toll-free serv. 284 British Virgin Is, 450 Quebec 631 New York 801 Utah 289 Ontario 464 Illinois 636 Missouri 802 Vermont 301 Maryland 409 Texas 041 Iowa 803 South Carolina 302 Delaware 473 Grenada 646 New York 804 Virginia 303 Colorado 475 Connecticut 047 Ontario 805 California 304 West Virginia 478 Georgia 049 Turks & Caicos Is. 806 Texas 305 Florida 480 Arizona 650 California 807 Ontario 306 Saskatchewan 484 Pennsylvania 6~1 Minnesota 808 Hawaii 307 Wyoming 500 Pers. Comm. Serv. 557 California 809 Dominican Rep. 308 Nebraska (PCS) 660 Missouri 810 Michigan 309 Illinois 501 Arkansas 661 California 812 Indiana 310 California 502 Kentucky 662 Mississippi 813 Rorida 312 Illinois 503 Oregon 604 Montserrat 814 Pennsylvania 313 Michigan 504 Louisiana 669 California 815 Illinois 314 Missouri 555 New Mexico 670 Northern Marianas 916 Missouri 315 NewYork 506 New Brunswick 671 Guam 817 Texas 316 Kansas 507 Minnesota 678 Georgia 818 California 317 Indiana 508 Massachusetts 679 Michigan 819 Quebec 318 Louisiana 509 Washington 682 Texas 828 North Carolina 319 Iowa 510 California 701 North Dakota 630 Texas 320 Minnesota 512 Texas 702 Nevada 631 California 321 Rorida 513 Ohio 703 Virginia 832 Texas 323 California 514 Quebec 704 North Carolina 635 Pennsylvania 330 Ohio 515 Iowa 705 Ontario 843 South Carolina 331 Illinois 516 NewYork 706 Georgia 045 NewYork 334 Alabama 517 Michigan 707 California 847 Illinois 336 North Carolina 518 NewYork 708 Illinois 850 Florida 856 New Jersey 857 Massachusetts 858 California 859 Kentucky 860 Connecticut 863 Florida 864 South Carolina 865 Tennessee 867 NW Territories/Yukon 868 Trinidad &Tobago 869 St. Kitts & Nevis 870 Arkansas 872 Illinois 876 Jamaica 977 Toll-free serv. 878 Pennsylvania 8~0 Toll-free serv. 081 Toll-free serv. 082 Toll-free serv. 888 Toll-free serv. 900 Info, Servs. 901 Tennessee 802 Nova Scotia & Prince Edward Is. 903 Texas 804 Florida 905 Ontario 906 Michigan 907 Alaska 908 New Jersey 089 California 910 North Carolina 911 Emergency Servs. 912 Georgia 913 Kansas 914 New York 915 Texas 916 California 917 New York 918 Oklahoma 919 North Carolina 520 Wisconsin 925 California 931 Tennessee 935 California 836 Texas 937 Ohio 940 Texas 941 Florida ~47 Michigan 949 California 951 California 952 Minnesota ~54 Florida 955 Texas 959 Connecticut 970 Colorado 971 Oregon 972 Texas 973 New Jersey 978 Massachusetts 979 Texas 980 North Carolina 989 Michigan 297 TeleGeography 2001 © TeleGeography, Inc. 2000 North American Area Codes, by Jurisdiction Alabama Birmingham and northern Alabama ....... 205 Huntsville and northern Alabama ........ 256 Mobile, Montgomery, and southern Alabama ..................... 334 Alaska .................................... 907 Alberta Cal0ary, Banff, and southern Alberta ..... 403 Edmonton, ,Jasper, and northern Alberta .780 Anguilla ................................... 264 Antigua .................................... 268 Arizona Eastern Phoenix area .................. 480 Arizona except Phoenix area ............ 520 Central Phoenix area ................... 602 Western Phoenix area ................. 623 Arkansas Little Rock, Fayetteville, and northwestern Arkansas ................. 501 Jonesboro and southern Arkansas ....... 870 Bahamas .................................. 242 Barbados .................................. 246 Bermuda ................................... 441 British Columbia British Columbia except Vancouver area ,250 Vancouver area .................... 604/778 British Virgin Islands ........................ 284 California Stockton, Fresno, Modesto, and central California ...................... 209 Los Angeles ....................... 213/323 Malibu, Beverly Hills, and west Los Angeles suburbs ............... 310/424 Oakland, Berkeley, and Fremont areas ..................... 341/510 Solano County ........................ 369 San Jose, Sunnyvale, and Cupertino areas ................... 4081669 San Francisco ..................... 415/628 Encitas, San Marcus, and southeastern California ................. 442 Chico, Redding, and northeastern California ............................ 530 Fresno and central California ............ 559 Lon0 Beach .......................... 562 San Die0o ........................... 619 Pasadena ............................ 626 Santa Rose, Sonoma, Napa, and north central California ................. 627 San Mateo, Palo Alto, and south San Francisco suburbs ............. 650/764 Northern Orange County ............ 657/714 Bakersfield and south central California ...................... 661 Fort Bra0g, Eureka, Ukiah, and northwestern California ................ 707 Ontario, Pomona, and San Bernadino .................... 752/909 Barstow, EI Cen~’o, Palm Sprin0s, and southeast California ................... 760 Santa Barbara, Bakersfield, and central western California .............. 805 Burbank and Glendale areas ............ 818 Monterey, Santa Cruz, and central western California ..................... 831 Northern San Diego and Del Mar ........ 858 Sacramento .......................... 916 Concord, Livermore, Walnut Creek ....... 925 Eastern San Diego and southwestern California ................ 935 Anaheim, Irvine, and southern Orange County ........................ 949 Riverside, Hemet’ San Jacinto ........... 951 Cayman Islands ............................. 345 Colorado Denver area ....................... 303/720 Colorado Springs, Pueblo and southeastern Colorado ................. 719 Aspen, Durango, and northwestern Colorado ............................. 970 Connecticut Bridgeport, New Haven, and southwestern Connecticut ........... 203/475 Hartford, Bristol, and northeastern Connecticut ........... 860/959 Delaware .................................. 302 District of Columbia ......................... 202 Dominica .................................. 767 Dominican Republic ......................... 809 Florida Miami, Key West and southeastern Rorida ................ 305/786 Orlando and central eastern Rorida ........................... 321/407 Gainesville and central Florida ........... 352 West Palm Beach, Boca Raton, and east central Florida .................... 561 Tampa, St. Petersburg, and Clearwater area ................... 727/813 Pensacola, Tallahassee, and northwestern Florida ................... 850 South central Florida ................... 863 Jacksonville, Daytona, and northeastern Florida ................... 904 Bradenton, Sarasota, and southwestern Rorida ................... 941 Fort Lauderdale ....................... 954 6eorgia Albany, Valdosta, and south central Georgia .................. 229 Atlanta, Marietta, Norcross ...... 404/678/770 Macon, Swainsboro, and south central Georgia ....................... 478 Augusta and northern Georgia ........... 706 Savannah, Vidalia, and southeastern Georgia .............................. 912 6renada ................................... 473 6uam ..................................... 671 Hawaii .................................... 808 Idaho ..................................... 208 Illinois Champaign, Urbana, Springfield, and central Illinois ........................ 217 Northeastern Illinois and northwest Chicago suburbs ................... 224/647 Peoria, Rock island, and west central Illinois .................... 309 Chicago ....................... 312/’/73/872 Cen~’al Chicago suburbs ............ 331/630 Southern Chicago suburbs .......... 464/708 Alton, Mount Vernon, and southern Illinois ....................... 618 La Salle, Rockford, and northern Illinois ....................... 815 Indiana Gary, South Bend, and northern Indiana .o .219 Indianapolis .......................... 317 Central Indiana excluding Indianapolis .... 765 Evansville and southern Indiana ......... 812 Iowa Davenport, Dubuque, and eastern Iowa ...319 Des Moines, Ames, and central Iowa ..... 515 Mason City, Pella, and central Iowa ...... 641 Council Bluffs, Sioux City, and western Iowa ......................... 712 Jamaica ................................... 876 Kansas Dod0e City, Wich~a, and southern Kansas ...................... 316 Topeka, Lawrence, and northern Kansas .785 Kansas City and eastern Kansas ......... 913 Kentucky Paducah, Bowlin0 Green, and western Kentucky ..................... 270 Louisville, Shelbyville, and north central Kentucky ...................... 502 Eastern Kentucky ...................... 606 Richmond, Danville, and northeastern Kentucky ............................. 859 Louisiana Baton Rouge and central eastern Louisiana ............................ 225 Shreveport, Monroe, and northern Louisiana ............................ 318 Lake Charles, Lafayette, and southwestern Louisiana ................ 337 New Orleans and southeastern Louisiana ............................ 504 Maine ..................................... 207 Manitoba .................................. 204 Maryland Rockville, Hagerstown, and western Maryland ......................... 240/301 Baltimore, Annapolis, and eastern Maryland ......................... 410/443 Massachusetts Waltham, Lexington, and Boston suburbs .................... 339/781 Lowell, Salem, and northern Massachusetts .................... 351/978 Pittsfield, Springfield, and western Massachusetts ....................... 413 Framingham, Cape Cod, and southern Massachusetts ............ 568/774 Boston ........................... 617/857 Michigan Traverse City, Muskegon, and northwestern Michigan ................. 231 Pontiac, Southfield, and Oakland County .................... 248/947 Ann Arbor, Wayne, and Detroit suburbs .................... 278/734 Detroit ........................... 313/679 Lansing and central Michigan ........... 517 Flint, Flushing, and southeastern Michigan ......................... 586/810 Grand Rapids, Kalamazoo, and southwestern Michigan ................ 616 Marquette, Sault Ste. Marie, and northern Michigan ..................... 906 Alpena, Bay City, and central Michigan ...989 Note: Two or more codes separated by a slash (e.g., in Houston, Texas) indicate an overlay; multiple codes are used for the same geographic area. 298 © TeleGeography, Inc. 2000 TeleGeography 2001 Minnesota Duluth and northern Minnesota .......... 218 St. Cloud and central Minnesota ......... 320 Rochester and southern Minnesota ...... 507 Minneapolis .......................... 612 St. Paul .............................. 651 Fridley, Blaine, and northwest Minneapolis suburbs ................... 763 Bloomington, Minnetonka, and southwest Minneapolis suburbs ......... 952 Mississippi Gulfpo~ Biloxi, and southern Mississippi ........................... 228 Jackson and centTal Mississippi ......... 601 Greenville, Tupelo, and northern Mississippi ........................... 662 Missouri St. Louis ............................. 314 Joplin, Springfield, and southwestern Missouri ............................. 417 Jefferson City, Columbia, and eastern Missouri ............................. 573 Franklin and Jefferson counties .......... 636 Marshall and northern Missouri ......... 660 Kansas City ........................... 816 Montana .................................. 406 Montserrat ................................ 664 Nebraska North Platte and western Nebraska ...... 308 Omaha, Lincoln, and eastern Nebraska ...402 Nevada Las Vegas and southern Nevada ......... 702 Northern Nevada ...................... 775 New Brunswick ............................ 506 New Hampshire ............................ 603 New Jersey Hackensack, Jersey City, and northeastern New Jersey ............... 201 Atlantic City, Trenton, and southeastern New Jersey .............. 609 Middlesex, Monmouth, and Ocean counties ....................... 732 Camden, Millville, and southwestern New Jersey .......................... 856 Elizabeth, Warren, and northwestern New Jersey .......................... 908 Newark, Patterson, and Morristown ...... 973 New Mexico .............................. 505 New York Manhattan .................... 212/’646/917 Syracuse and northwestern New York ............................ 315 Brooklyn, State Island, Bronx, and Queens ....................... 347/718 Nassau County and western Long Island ........................... 516 Northeastern New York ................ 518 Binghamton and south centTa] New York ............................ 607 Lindenhurst, Islip, and eastern Long Island ........................... 631 Buffalo and western New York .......... 716 Albany, Poughkeepsie, and southeastern New York ................. 845 Westchester, White Plains, and southeastern New York ................. 914 Newfoundland ............................. 709 North Carolina Northeastern North Carolina ............ 252 Winston-Salem, Greensboro, and northwestern North Carolina ............ 336 Charlotte and south central North Carolina ..................... 704/980 Asheville and western North Carolina .... 828 Fayetteville and southeastern North Carolina ........................ 910 Raleigh and northeastern North Carolina ........................ 919 North Dakota .............................. 701 Northern Marianas ......................... 670 Northwest Territories/Yukon ................. 867 Nova Scotia and Prince Edward Island ........ 902 Dhio Cleveland ............................ 216 Youngstown, Akron, Canton, and northeastern Ohio .................. 234/330 Toledo and northwestern Ohio ........... 419 Northeastern Ohio excluding Cleveland ............................ 440 Cincinnati ............................ 513 Columbus ............................ 614 Southeastern Ohio excluding Columbus ...740 Southwestern Ohio excluding Cincinnati .937 Oklahoma Oklahoma City and central Oklahoma ..... 405 Southern and western Oklahoma ........ 580 Tulsa and northeastern Oklahoma ........ 918 Ontario Toronto ........................... 416/647 London and southwestern Ontario ....... 519 Ottawa and southeastern Ontario ........ 613 North Bay and northeastern Ontario ...... 705 Thunder Bay and western Ontario ........ 807 Hamilton and southeastern Ontario ...289/905 Oregon Portland, Salem, and northwestern Oregon ............... 503/971 Oregon except Portland and Salem areas .......................... 541 Pennsylvania Philadelphia ................... 215/267/445 Pittsburgh and western Pennsylvania .................. 412/724/878 Allentown, Reading, and southeastern Pennsylvania ....... 464/610/835 Scranton and northeastern Pennsylvania ......................... 570 Harrisburg and south central Pennsylvania ......................... 717 Erie and northwestern Pennsylvania ...... 814 Puerto Rico ................................ 787 Quebec Quebec City and eastern Quebec ........ 418 Southern Quebec excluding Montteal .... 450 Montreal ............................. 514 Western Quebec ...................... 819 Rhode Island .............................. 401 St. Kitts & Nevis ............................ 869 St. Lucia .................................. 558 St. Vincent & Grenadines .................... 764 Saskatchewan ............................. 306 Sooth Carolina Columbia and central South Carolina ..... 803 Charleston and eastern South Carolina ...843 Greenville and western South Carolina ...864 Sooth Dakota .............................. 605 Tennessee Chattanooga, Johnson City, and southeastern Tennessee ................ 423 Nashville ............................. 615 Knoxville, Jefferson City, and east central Tennessee ................. 865 Memphis and western Tennessee ........ 90! Central Tennessee excluding Nashville .. 331 Texas San Antonio .......................... 210 Dallas ........................ 214/469/972 Waco and centTal Texas ................ 254 Houston ....................... 281/713/832 Corpus Christi and southeastern Texas ................................ 361 Beaumont, Galveston, Port Arthur, and southeastern Texas ................ 409 Austin and San Marcus ............. 512/737 Fort Worth, Arlington, and Weatherford ...................... 682/817 Amarillo and northern Texas ............ 806 Uvalde, Kerrville, and southwest Texas ...830 Tyler and northeastern Texas ............ 903 El Paso, Odessa, Midland, and western Texas ........................ 915 Conroe, Nacogdoches, Lufkin, and southeastern Texas .................... 936 Wichita Falls, Denton, and northern Texas ........................ 940 Laredo, Brownsville, and southern Texas ........................ 956 Bwan, College Station, and southeastern Texas .................... 979 Trinidad & Tobago .......................... 868 Turks & Caicos Islands ...................... 649 U.S. Virgin Islands .......................... 340 Utah Utah excluding Salt Lake City ............ 435 Salt Lake City ......................... 801 Vermont ................................... 802 Virginia Roanoke and northwestern Virginia ...... 540 Alexandria, Arlington, and McLean ...571/703 Hampton, Norfolk, and southeastern Virginia .................. 757 Richmond and central Virginia ........... 804 Washington SeattJe ............................... 206 Tacoma and south Seattle suburbs ....... 253 Western Washington excluding Seattle .................. 360/564 North SeattJe suburbs .................. 425 Spokane and eastern Washington ........ 509 West Virginia .............................. 304 Wisconsin Racine and southeastern Wisconsin ...... 262 Milwaukee, Greenfield, and Oak Creek .... 414 Madison and southwestern Wisconsin .... 608 Eau Claire and northern Wisconsin ....... 715 Southeastern Wisconsin excludin9 Milwaukee .................. 920 Wyoming .................................. 307 299 TeleGeography 2001 © TeleGeogrophg, Inc. 2000 North American Area Codes Edmonton ¯ BRITISH COLUMBIA 780 250 SASKATCHEWAN ~’~’~ii~uver 306 i~r~ 71~ ALBERTA Seattle 206 WASHINGTON Porl:ia nd 360/564 509 OREGON 541 530 707 CALIFORNIA NEVADA 661 702 ~’lZr .,,~6. 760 Los AnoeIes ~l~Z,.~,. ~i! San 8iego ¯ 500 Regina ¯ MONTANA 406 IDAHO WYOMING ~07 801 UTAH COLORADO ARIZONA NEW MEXICO 5O§ 500 Personal Communication Services ~(]O/B’J’7~ Toll-free 8801881 Toll-free to U.S. from Canada and Caribbean 204 Winnipeg ¯ NORTH DAKOTA 701 SOUTH DAKOTA 605 NEBRASKA 308 402 785 KANSAS 316 OKLAHOMA 806 580 .... 940 682/817 915 TEXAS 254 512/737 830 956 361 218 MINNESOTA 763 320 9~2.6t; 507 515 616 913 417 918 5 903 Dallas 936 1 4O9 979 I-:c’uSt~,n © TeleGeographg, Inc. 2000 TeleGeography 2001 807 ONTARIO 705 flUEBEC 819 ~" NEW BRuNsw:,~::~: 506 450 ..... MAINE ,"514’, 207 902 71i~ !ll, 9r_-_-_~ 512 ...~’. 507 515 712 816 ~!3 417 918 501 9O3 715 WISCONSIN 920 231 519 Chica~ 989 ~m 608 ai2 ~-I ~l :~ 1 ;Z/"/"/At" ~1:~t 517 ~6 815 ~#~t~DIANf, 419 OHIO 755 7~ ¯ ~Pd,a~o~is~ 937 660 217 Z!7 T ¯ 5~3 CmDmne~i St. L~ 812 859 3~4~ 618 ~2 KE~rU~(Y M, SSQg~ 270 606 573 Nash~,l~e A~SAS ~1 ~!~a~SS~ 250 662 870 205 Dallas 318 ~-!.~4!~72 LOUISIANA MISSIS- ALABAMA SIPPI 601 334 936 F~!s’mn 337 504 Newgrlea~ 613 ~Z NH ¯ 315 603 .~’t~.~ |Torl3Tl~ 518 ./~-~’! t--617~7 ~n .. 11~7 N~ ’=O.~K .413 570 814 st~ .~7~1e .41~ ’~ a,] ~Ni~804 W. ~’rHG,~ IA 252 336 423 828 704.,"980 864 706 803 ¯ At!aftra ¯ 478 E EU~ El,6 229 912 850 FLORIDA 904 919 I",,’DRTF I 910 SUUTH.r, AROLIt’~, 843 3~2 321/4~7 727 813 863 561 954 ~I4T ¯ Miami 301 TeleGeography 2001 © TeleGeography, Inc. 2000 A Primer on Bits Measuring Bytes Bit by Bit Below are the standard metric prefixes used in the SI (Syst~me International) conventions for scientific measurement. With units of time (e.g., gigabits per second) or things that come in powers of 10, they retain their usual meanings of multiplication by powers of 1,000 = 103. When used with bytes (e.g., gigabytes of data storage) or other things that naturally come in powers of 2, they usually denote multiplication by pow- ers of 1,024 = 21°. Base 10 Kilobit/s = 1,0001 = 103 = 1,000 Megabit/s = 1,0002= 106 = 1,000,000 Gigabit/s = 1,0003 = 109 = 1,000,000,000 Terabit/s = 1,0004 = 1012 = 1,000,000,000,000 Petabit/s = 1,0005 = 1015 = 1,000,000,000,000,000 Exabit/s = 1,0006 = 10le = 1,000,000,000,000,000,000 Zettabit]s = 1,0007 = 1021 = 1,000,000,000,000,000,000,030 Yottabit/s = 1,0008 = 1034 = 1,000,000,000,000,000,000,000,000 Base 2 Kilobyte = 1,0241 = 21° = 1,024 Megabyte = 1,0242 = 220 = 1,048,576 Gigabyte = 1,0243 = 23o = 1,073,741,824 Terabyte = 1,0244 = 240 = 1,099,511,627,776 Petabyte = 1,0245 = 25o = 1,125,899,906,842,624 Exabyte = 1,0246 = 260 = 1,152,921,504,606,846,976 Zettabyte = 1,0247 = 27o = 1,180,591,620,717,411,303,424 Yottabyte = 1,024e = 28o = 1,208,925,819,614,629,174,706,176 © TeleGeography, Inc. 2000 Measuring Telecommunications Bandwidth--DS-0 to OC-192 Carder Technolo~/ Data Rate(Mbl~) Description ~4 Kbps Circuits* DS-0 0.064 Base rate in the Digital Signal (DS) level hierarchy 1 T-1 (DS-1) 1.544 Primary level of the American T-carrier multiplexing 24 system; capacity is the same as a DS 1 carrier T-2 (DS-2) 6.312 Four times the capacity of T-1 96 T-3 (DS-3) 44.736 28 times the capacity of T-1 672 T-4 (DS-4) 274.176 168 times the capacity of T-1 4,032 E-1 2.048 Primary level of the European E-carrier multiplexing system 30 E-2 8.448 Carries four multiplexed E-1 signals 120 E-3 34.368 Carries four E-2 signals 480 E-4 139.264 Carries four E-3 signals 1,920 E-5 565.148 Carries four E-4 signals 7,680 OC-1/STS-1 51.840 Basic signaling rate of SONET hierarchy 810 0C-3/STM-1 155.520 Exactly three times the capacity of 0C-1"* 2,430 0C-12/STM-4 622.080 12 times the capacity of 0C-1 9,720 0C-24 1,244.160 24 times the capacity of 0C-1 19,440 0C-48/STM-16 2,488.320 48 times the capacity of 0C-1 38,880 0C-192/STM-64 9,953.280 192 times the capacity of 0C-1 155,520 Key "T" "E" "OC" "STM" "STS" T-carrier system in U.S., Canada, and Japan with 1.544 Mbps as the primary level (24 voice channels x 64 Kbps per channel). Digital Signal that travels on the T-carrier or E-carrier. Used in countries other than U.S., Canada, and Japan. The hierarchy was established by the CEPT (Conference Europ~enne des Postes et T~l~communications) with 2.048 Mbps as the primary level ([30 voice channels 2 channels for overhead] x 64 Kbps per channel). Optical Carrier interface designed to work with STS-n (Synchronous Transport Signal) signaling rate in a SONET (Synchronous Optical Network). Synchronous Transport Module refers to a large carrier (base signal 155.52 Mbps) in a SONET. Synchronous Transport Signal is the elec~’ical counterpart to the Optical Carrier (0C). The number of 64 Kbps is presented for comparative purposes only. The actual number of simultaneous conversations possible over a given carrier may vary depending on the encoding scheme used. ~" In the "E° and "T" hierarchies, each higher level is set to be "almost b,,t not exactly’ a multiple of the bit rate for the previous order (plesiochronous). To elimi- nate problems associated with plesiochronous multiplexing, SONET, a synchronous hierarchy, was defined in the United States in 1986. As a result, the "OC" and "STM" carriers are exact bit-rate multiples of their primary levels, 0C-1 and STM-1, respectively. Source: TeleGeography research, Alcatel, Newton’s Telecom Dictionary © TeleGeograph¥, Inc. 2000 302 Put telegeography on your wall. Global Communica~ons Cable and Satellite Map 2000 _-.,~li’ill~ ~-- Map measures 0.9 m x 1.2 m (37" x 51") and is shipped folded or flat. © Tele6eography, Inc. 2000 TeleGeol~raphy, Inc. is the authoritative source for international telecom statistics and analysis. Contact us for details on our publications and wall maps: Tel.: +1 202 467 0017 Fax: +1 202 467 0851 Asian Telecommunications Traffic Flows Inset from Global Communications Traffic Map 2000 © TeleGeography, Inc. 2000 Email: info@telegeography.com http://www.telegeography.com TeleGeography, Inc. www.telel~eography.corn About TeleGeography, Inc. telegeography \tel "a-je-Sg’ra-fe n (1990) abbrv, of telecommunications geography [fr. Gk re/e, far off, at a distance and L. communicatus, pp. of communicare to impart fr. Gk geo (earth) graphein, (to write)] 1. a new branch of geography that maps the pattern of tele- phone traffic and other electronic communication flows; 2. places created by or perceived solely via telecommunications (e.g., a computer network address); 3. the telecommunications artifacts (radio antennae, terminals, signs) on a site; 4. the balance of telecommunications power in one country or region vis-b-vis another (cf. geopolitics, archaic). The old geography of countries and coast lines is giving way to a new geography marked by tele- phone codes, satellite footprints, and lnternet addresses. Electronic networks have made the world smaller. But they also have created countless new places, both virtual and physical. This expand- ing electronic terrain---call it telegeography--demands a new cartography. That is the purpose behind TeleGeography, Inc., the authoritative source for international telecom statistics and analysis. An independent subsidiary of Band-X Ltd., TeleGeography publishes reports and maps used by thousands of communications companies, consultancies, governments, and financial institutions in over 100 countries. The company’s flagship report on international traffic-- the self-titled TeleGeography series--has been published annually since 1989. TeleGeography also authors a sedes of related reports and maps on global telecom infrastructure and network topology, including: International Bandwidth 2000, a guide to supply and demand on long-haul networks; Hubs and Spokes: an Internet Reader, a pdmer on global Internet architecture; Direction of Traffic 1999, an historical traffic atlas compiled with the International Telecommunication Union (ITU); New International Carriers, a three-volume directory of competing international telephone companies; Global Communications Cable and Satellite Map, a poster-sized map of telecom infrastructure; and Global Communications Traffic Map, a wall map of internation- al traffic flows. To learn more, please visit us at http:llwww.telegeography.com. TELEGEOGRAPHY, INC. 1730 Rhode Island Avenue, NW ¯ Suite 400 Washington, DC 20036 USA Tel. 1 (202) 467-0017 ¯ Fax 1 (202) 467-0851 ¯ E-mail: info@telegeography.com http://www.telegeography.com