******************************************************** NOTICE ******************************************************** This document was converted from WordPerfect to ASCII Text format. Content from the original version of the document such as headers, footers, footnotes, endnotes, graphics, and page numbers will not show up in this text version. All text attributes such as bold, itallic, underlining, etc. from the original document will not show up in this text version. Features of the orginal document layout such as columns, tables, line and letter spacing, pagination, and margins will not be preserved in the text version. If you need the complete document, download the WordPerfect version or Adobe Acrobat version, if available. ***************************************************************** Before the Federal Communications Commission FCC 96-484 Washington, D.C. 20554 In the Matter of ) ) IB Docket No. 96-261 International Settlement Rates ) Notice of Proposed Rulemaking Adopted: December 19, 1996 Released: December 19, 1996 By the Commission: Comment Date: February 7, 1997 Reply Comment Date: March 10, 1997 TABLE OF CONTENTS Paragraph Numbers I. INTRODUCTION 1-4 II. BACKGROUND 5-25 III. DISCUSSION 26-91 A. Benchmark Methodology 31-38 B. Benchmarks 39-57 1. Benchmark Rates Based on Tariffed Components Prices 39-55 a. Ranges Based on Level of Economic Development 43-53 b. Country-Specific Benchmarks 54-55 2. Alternative Benchmark Methodologies 56-57 C. Implementation of Benchmarks 58-91 1. Transition to Benchmarks 58-68 2. Application of Benchmark Settlement Rates to Countries Committed to Competitive Reform 69-74 3. Applying Benchmarks to Address Anticompetitive Behavior 75-86 4. Enforcement Proposals 87-90 5. Effect of Settlement Rate Reductions on U.S. Collection Prices 91 IV. CONCLUSION 92-94 V. PROCEDURAL ISSUES AND ORDERING CLAUSES 95-107 I. Introduction 1. This Notice of Proposed Rulemaking presents proposals for revising our benchmark settlement rates for international message telephone service (IMTS) between the United States and other countries. This Notice represents the next step in an ongoing effort by the Commission, many foreign governments, and multilateral organizations such as the International Telecommunication Union (ITU) and the Organization for Economic Cooperation and Development (OECD) to reform the international accounting rate system. A multilateral consensus has emerged that the traditional accounting rate system must be reformed because it results in settlement rates that are substantially above costs and creates competitive distortions and inefficiencies in the global telecommunications market. We concur with the need for reform, and to that end, propose measures in this Notice to encourage settlement rates to move closer to cost. 2. We believe that revisions to our benchmark rates are necessary because the current benchmarks have been rendered obsolete by the significant changes occurring in the global telecommunications market in recent years. In particular, we believe our benchmark rates should be revised to reflect recent technological improvements, their associated cost reductions, and the market structure changes occurring in the global telecommunications market. We also believe these revisions are necessary to move settlement rates closer to the actual costs incurred by foreign carriers to terminate international traffic. As described in detail below, this Notice is part of a broad-based effort by the Commission to reform the international accounting rate system and thereby help ensure lower international calling prices for consumers. 3. We are committed to achieving settlement rates that more closely resemble the level that would be established in a competitive market for termination of international services. We believe that in such a competitive market settlement rates would be close to the incremental cost of providing international termination service. We recognize, however, that moving to cost- based settlement rates will require a significant adjustment for many countries. We therefore recommend a transition period for implementation of our new benchmark rates to ensure that this adjustment does not unduly disrupt foreign carriers' operations, particularly for foreign carriers from developing countries. 4. In this Notice, we solicit comment on four issues. First, how should benchmark settlement rates be calculated? Second, how long should the transition to benchmark rates last? In particular, should we provide a longer transition for developing countries and should we provide additional flexibility beyond any transition for countries committed to introducing competition? Third, what enforcement mechanisms are necessary to ensure carriers make progress in negotiating settlement rates within the benchmarks? Finally, can the benchmark rates be used to address to competitive problems in the U.S. IMTS market? II. Background 5. Our goals in this Notice are the same as the goals we stated in our Foreign Carrier Entry Order: (1) to promote effective competition in the global market for communications services; (2) to prevent anticompetitive conduct in the provision of international services or facilities; and (3) to encourage foreign governments to open their communications markets. Accomplishing these goals will help us to reach our objective under the Communications Act: ensuring that consumers and businesses in the United States are provided with reasonably priced, high quality, and technologically advanced international service. We believe the current above-cost accounting rate system is a significant impediment to achieving our goals in this proceeding because it restrains the development of competition in U.S. and foreign markets, creates the potential for distortions in the U.S. market for IMTS, and significantly increases prices for U.S. consumers. 6. The current accounting rate system was developed as part of a regulatory tradition that international telecommunications services were supplied through a bilateral correspondent relationship between national monopoly carriers. An accounting rate is the price a U.S. facility- based carrier negotiates with a foreign carrier for handling one minute of international telephone service. It was originally intended to allow each carrier to recover its costs for terminating an international call. Each carrier's portion of the accounting rate is referred to as the settlement rate. In almost all cases, the settlement rate is equal to one-half of the negotiated accounting rate. At settlement, each carrier nets the minutes of service it originated against the minutes the other carrier originated. The carrier that originated more minutes of service pays the other carrier a net settlement payment calculated by multiplying the settlement rate by the number of imbalanced traffic minutes. 7. The principal problem with the traditional accounting rate system is that in most cases, settlement rates greatly exceed the underlying costs of providing the service in question, i.e., terminating an international call. It is not unusual for settlement rates to be between five and ten times a reasonable estimate of the underlying cost of terminating an international call. Such significantly inflated settlement rates represent a major subsidy from U.S. consumers, carriers and their shareholders to foreign carriers and raise prices for international services to U.S. consumers many times above the costs of providing those services. They also distort IMTS market performance and restrict market growth. 8. The United States paid roughly $5 billion in settlements to the rest of the world in 1995, up from $2.8 billion in 1990. The U.S. outpayment results in part from the fact that U.S. consumers make more telephone calls to foreign countries than foreign consumers make to the United States. In fact, the size of the imbalance between U.S.-outbound and inbound minutes has accelerated in recent years, as the chart in Appendix C demonstrates. To the extent that these settlement payments exceed the actual costs foreign carriers incur in terminating U.S.- originated calls, they represent a significant subsidy to foreign carriers. Based on our estimate of the costs of international termination services, we estimate that at least three-quarters of the $5 billion in outpayments is such a subsidy from U.S. consumers, carriers and their shareholders to foreign carriers. 9. U.S. consumers pay on average 16› a minute for a domestic long distance call, but they pay 99› a minute for an international call. Yet, the difference in cost between providing domestic long distance and international service is no more than a few cents. As a result of recent technological advances, the underlying costs of providing telephony are becoming virtually distance insensitive. For example, because of new fiber optic technology, the cost of undersea cables on a per circuit basis is only one eighth of what it was seven years ago. We anticipate that increased competition in international satellite services will bring similar potential benefits to countries that are not now served by undersea cables and comparable land facilities. Differences in underlying costs therefore do not explain why international services are so much more expensive than domestic long distance services. The difference is attributable in part to limited competition in the IMTS market and in part to the inflated settlement rates paid by U.S. carriers to terminate traffic in foreign markets. 10. Inflated prices for international services restrict growth in the market for those services, both in the United States and in foreign countries. Past experience in the U.S. domestic market and in other countries has shown that calling patterns are highly price elastic. We anticipate, therefore, that reductions in the price of international telephone service would significantly stimulate traffic flows, thereby increasing revenues for U.S. and foreign carriers. A more robust market for international services will provide additional financing for network infrastructure and result in a more ubiquitous global telecommunications network. Policies that encourage such expansion of the global network are vital in light of the fact that some two thirds of the world's population have no access to a telephone. 11. The subsidies embedded in most current settlement rates also create significant competitive distortions in the rapidly growing international market for telecommunications services. These distortions impede the Commission's policy of creating greater competition in the IMTS market in order to lower international calling prices for U.S. consumers. An example of possible anticompetitive distortions in the U.S. market resulting from above-cost settlement rates is "one-way bypass," which could occur if a foreign carrier from a country that does not permit private line resale opportunities equivalent to those available under U.S. law were permitted to terminate its traffic inbound to the U.S. market over resold international private lines, which are not subject to settlement rates. The foreign carrier would receive an above-cost settlement rate to terminate U.S. outbound traffic in its market, while unilaterally cutting its own costs by bypassing the accounting rate system to terminate its switched traffic inbound to the U.S. market over resold private lines. Indeed, the potential for harm to U.S. carriers and consumers from such "one-way bypass" exists even if a foreign carrier does not enter the U.S. market, but instead terminates its traffic via private lines provided by an unaffiliated carrier in the U.S. market. In this case, the foreign carrier would gain a significant cost advantage over U.S. carriers in the global market because it still would be collecting above-cost settlement rates to terminate traffic in its market while U.S. carriers would not be collecting such rates to terminate traffic in the U.S. market. 12. Recent service innovations in the international market accentuate the market distortions caused by above-cost settlement rates. Call-back exemplifies the dilemmas and benefits posed by such service innovations. The traditional settlement rate system assumes that a customer's physical location determines the place of origin of an international call, with the carrier in the originating country paying a settlement rate to the carrier in the terminating country. By means of call-back, however, the customer in effect chooses which country he or she wishes to be the originating country and which will be the terminating country. In cases where the call is between a competitive market and a monopoly market, the competitive market (with competitive pressure on the retail prices charged for IMTS) will almost always be the less expensive point of origination. The caller would presumably choose to originate his or her call from the competitive market. As a result, so long as call-back is legally possible and technically comparable to conventional IMTS, competitive markets will see their balance of traffic with monopoly markets shift to a very heavy imbalance of outbound versus inbound minutes. 13. There are other technological developments that accentuate the market distortions caused by above-cost settlement rates. For example, the routing of bilateral traffic through third countries has become increasingly prevalent as a means to arbitrage settlement rate differences. Such re-routing can be helpful in undercutting the settlement rate system, but it can also lead to inefficient traffic routing patterns that are not aligned with underlying economic network costs. Use of the Internet also has emerged as an alternative to higher priced IMTS. Though internet traffic and switched voice traffic are carried over virtually identical facilities, the price for internet service is far cheaper because switched traffic is subject to international settlement rates, while internet traffic is exchanged outside of the traditional accounting rate system. 14. Already, the traditional settlement rate system is under significant pressure to reform as a result of the recent transformations in market structures occurring in many countries. In order to promote the development of their communications sectors, many countries are rapidly moving from the old model of monopoly providers to competitive markets for telecommunications services. The United States, which first introduced competition into long distance and international services in the early 1980's, has moved toward removal of barriers to competition in local markets this year with the enactment the Telecommunications Act of 1996. In Europe, the member states of the European Union have pledged to dismantle competitive barriers by January 1, 1998. Currently, the members of the World Trade Organization (WTO) are engaged in talks to liberalize trade in basic telecommunications services. If the WTO's Group on Basic Telecommunications (GBT) is successful in reaching an agreement by its deadline of February 15, 1997, the international market for telecommunications services will become far more open and competitive than at present. 15. Consistent with these market pressures, settlement rate reform has been the focus of much attention in recent years in international fora. For example, both the International Telecommunication Union (ITU) and the Organization for Economic Cooperation and Development (OECD) have been actively studying accounting rate reform and members of both organizations have reached a consensus on the need for such reform. In a speech last June, Dr. Pekka Tarjanne, Secretary-General of the ITU, reflected multilateral consensus on the need for accounting rate reform, arguing that "the situation is not sustainable." ITU Recommendation D.140 calls for countries to adopt nondiscriminatory, cost-oriented and transparent accounting rates. Work in the OECD similarly has focused on the principles of transparency, nondiscrimination and cost-based pricing. 16. The Australian delegation to the WTO's GBT has argued that the service of terminating international telecommunications traffic, "telecommunications termination services," should be provided on a cost-based, nondiscriminatory basis. Mexico and Sweden have explored accounting rate reform as well, and the United Kingdom has joined the United States in trying to reform the accounting rate system by improving its transparency. The United Kingdom now publishes its accounting rates with all OECD member nations and has stated that in December of this year it will publish its rates to all countries. Perhaps most far-reaching of all is the contemplated move in Europe after January 1, 1998 to reject the traditional accounting rate system for intra-European calls in favor of a system of domestic call termination charges. 17. We support the work done in international fora to reform the traditional settlement rate system and we propose to intensify our efforts to achieve multilateral consensus on measures to reform the settlement rate system to achieve cost-based, transparent and nondiscriminatory charges for terminating international calls. For example, we propose to work with U.S. carriers and other U.S. government agencies to suggest measures the ITU could undertake to promote more rapid progress in achieving cost-based rates. Recommendation D.140 has contributed to the progress in reducing settlement rates, but we believe additional steps by the ITU would encourage more rapid movement in this direction. We also will continue supporting the work of the OECD on settlement rate issues. Additionally, we will encourage regional organizations like APEC (Asian Pacific Economic Cooperation), Inter-American Telecommunications Commission (CITEL) and the European Union to consider ways to reform the international settlement rate system. We solicit comments on alternative approaches to reforming the international accounting rate system, including multilateral approaches through institutions like the ITU. 18. We believe, however, that in light of the significant technological and market changes occurring in the global telecommunications market we should go beyond these multilateral steps to encourage settlement rate reform. Given the rapidly increasing imbalance between U.S.-outbound and inbound traffic and the slow pace of change internationally, we believe we should act domestically to encourage lower settlement rates and ultimately, international calling prices to U.S. consumers. We have a statutory mandate to ensure that consumers pay reasonable charges for communications service. Current prices for international service are substantially inflated, in part as a result of above-cost accounting rates. In addition, we believe it is unfair for consumers and carriers from the United States and other competitive markets to continue to subsidize foreign carriers by paying international settlement rates that exceed the costs foreign carriers incur in terminating calls. 19. We emphasize that the measures we propose here are intended to fulfill our statutory mandate to ensure reasonable telephone rates. These measures are directed at U.S. carriers and the settlement rates they pay to foreign carriers. They are necessary because the level of international settlement rates has an effect on the price paid by U.S. consumers for IMTS. We believe the Commission has the authority under Sections 1, 4(i) 201-205 and 303 (r) of the Communications Act of 1934, as amended, relevant case law, and international regulations, to take the steps described in this Notice. We ask parties to comment on these steps, including our authority to undertake them. 20. We believe that the most effective way to ensure settlement rate reform that results in reasonable international calling prices is through the development of competitive markets for IMTS. Effective competition on both ends of an international call would ultimately drive international termination charges closer to costs and erode the subsidy embedded in current settlement rates. To the extent possible, therefore, we seek through our policies to reform the traditional accounting rate system and encourage the development of competitive markets for the termination of international services. 21. In our Accounting Rate Policy Statement, we announced a significant change in our approach to international settlement rates. As a way of encouraging a more competitive international market, we expanded beyond our private line resale policy to endorse new services that encourage alternatives to the traditional accounting rate system in the international market. We pledged also to tailor our settlement rate policies to reflect diverse national market structures. In particular, we announced our intention to develop settlement rate policies which differentiate among competitive markets, non-competitive markets, and developing countries with special infrastructure needs. 22. Our recent Flexibility Order took a further critical step in reforming our settlement rate policies by recognizing that we should allow for entirely new alternatives to the traditional correspondent accounting rate model where competitive markets exist in both the originating and terminating markets. We sought to encourage the development of competitive markets for the origination and termination of international calls. Accordingly, the Flexibility Order establishes a more flexible framework which permits carriers to take their international traffic off the traditional settlement rate system where competitive conditions permit and to negotiate alternatives for terminating international calls that more closely track underlying costs. 23. Despite the far-reaching changes occurring in the international services market, in the short term many countries are likely to continue to have monopoly carriers or limited competition. As long as settlement rates remain substantially above costs, these countries may have a disincentive to introduce competition in their telecommunications markets in order to protect the subsidy in above-cost accounting rates, notwithstanding the overall economic and development benefits that would accrue from a robust, competitive telecommunications sector. We therefore tentatively conclude that we cannot rely solely on market forces to achieve timely reform of accounting rates in markets with no or limited competition. 24. This Notice also recognizes, however, that many countries will need time to adjust to more cost-based settlement rates. Indeed, many countries already are grappling with the difficult transition from an inefficient, high-priced telephone system to a competitive and lower- priced structure that will better serve their consumers and economies. A number of countries, for example, are undergoing the politically difficult task of rebalancing rates. We note that current settlement rates are generally much higher in lower income countries than in high income countries and in some countries settlement payments represent a high proportion of total telecommunications revenue. Immediate enforcement of lower benchmark rates could impose a substantial burden on these countries as they seek to develop their telecommunications networks. We therefore propose to implement our benchmarks in a way that provides a longer transition for developing countries and permits flexibility for countries that are making the adjustments necessary to introduce competitive reforms and move toward a more cost-based system of international settlement rates. 25. We recognize that our efforts to reform the traditional accounting rate system and reduce settlement rates to a cost-based level will require substantial adjustments for many countries in the short term. In the long term, however, we are convinced that these reform efforts will benefit not only U.S. consumers and carriers, but also foreign consumers and carriers. In particular, these reforms will lower international calling prices, which in turn will stimulate additional service growth, expand market size, and increase revenue of U.S. and foreign carriers. This market expansion should lead to greater network buildout, so that the benefits of the global information infrastructure become more widespread. III. Discussion 26. The Commission first adopted benchmark settlement rates in 1992 in conjunction with other procedural reforms to remove U.S. regulatory impediments to cost-based, economically efficient, nondiscriminatory settlement rates. We adopted benchmark settlement rate ranges of 0.165 SDR to 0.275 SDR (23›-39›) to terminate U.S. telephone calls in Europe and 0.275 SDR to 0.42 SDR (39›-60›) to terminate U.S. telephone calls in Asia and other regions. Our goal in setting the benchmark ranges was to support U.S. carriers' efforts to reduce settlement rates to cost-based levels as expeditiously as circumstances permitted. U.S. carriers have made important strides in lowering settlement rates since the adoption of our current benchmarks. When we set our current benchmark ranges in 1992, the average U.S. settlement rate was 51.5› per minute. By November 1996 that rate had declined to 36.5› per minute. 27. Despite these recent decreases in rates negotiated under our current benchmarks, settlement rates remain significantly above the cost of providing international service. Moreover, technological change, increased power of digital technology, and greater reliance on market forces and competitive behavior have combined to reduce costs and render our existing benchmark ranges obsolete. For these reasons, we believe that our benchmarks should be revised. 28. We believe that adopting one of the methodologies described below to revise our benchmarks will ensure that they better reflect these changing cost, supply, and market conditions and continue to move settlement rates closer to actual costs. To avoid the problem in the future of our benchmarks not keeping pace with cost reductions, and to encourage further movement toward cost-based settlement rates, we propose to periodically revise our benchmarks. We will also consider approaches to benchmarks that more closely approximate incremental costs. We seek comment on how we can ensure that carriers that have committed to achieving settlement rates within the benchmarks are not adversely affected by any revisions to the benchmarks. 29. We invite interested parties to submit comments on our proposals for revising the benchmark settlement rates, including the methodology for calculating the rates and our proposal for periodic revisions to the rates. We also invite comments on our plan to implement the revised benchmark settlement rates in a manner that will promote our goal of achieving the cost- oriented, nondiscriminatory, transparent settlement rates necessary for the development of competition in the global telecommunications services market. 30. We note that the work of the WTO's Group on Basic Telecommunications (GBT) in negotiating an agreement to liberalize trade in basic telecommunications services is scheduled to conclude on February 15, 1997. The United States has been participating in these negotiations. Any commitments that the United States may ultimately make in these negotiations would have to be extended to all WTO members on a Most Favored Nation (MFN) basis. We invite comments on whether each of the proposals in this Notice is consistent with MFN and National Treatment obligations under a potential WTO agreement. A. Benchmark Methodology 31. Ideally, we would let competitive market forces determine the pricing for termination of international services. As we noted in our Flexibility Order, termination at nondiscriminatory interconnection charges in competitive markets would be an example of such an approach. However, such interconnection charges still do not exist in most foreign markets. We believe that a reasonable approximation of the interconnection charges that would exist in competitive markets is the total service long run incremental cost (TSLRIC) of terminating international service plus a reasonable contribution to common costs. 32. The TSLRIC of providing international termination services is the additional cost that a firm incurs as a result of providing that service. This cost includes a risk-adjusted return on capital. The term "total service," in the context of TSLRIC, indicates that the cost measured is that of providing an entire service, in this case, international termination service. The term "long run," in the context of TSLRIC, refers to a period long enough so that all of a firm's costs become variable or avoidable. Because long run incremental cost is the level to which prices would tend in a competitive market, we believe it should be the preferred cost standard for establishing benchmark settlement rates. Prices at an incremental cost level would maximize consumer welfare. In addition, we believe that it may be appropriate for international services to provide a reasonable contribution to the common costs of foreign carriers. Yet, we also believe that if termination services were competitively provided, the mark-up over incremental cost reflected in carriers' prices, i.e. their contribution to common costs, would be substantially less than the mark-up that is reflected in current settlement rates. We seek comment on this analysis, and on the appropriate costs that should be included or excluded from this approach. 33. Although we believe incremental cost ultimately may be the appropriate standard for estimating termination charges, we recognize that applying this approach to individual countries will not be easy. We do not have, and cannot obtain, the data necessary to calculate such costs for each foreign carrier. We do not have convincing and up-to-date studies of foreign carriers' costs, nor do we believe that current and reliable data on such costs exists in most countries. Until such data is available, we believe it is appropriate to use alternative methods to calculate benchmarks. Thus, we seek comment on alternative methods to establish benchmarks that more closely approximate costs than current settlement rates. 34. We recognize that foreign carriers should be able to charge a reasonable price for terminating U.S.-originated calls, but settlement rates appear in most instances to be well in excess of any estimate of reasonable termination costs. For example, as discussed below, paras. 50-51, we estimate that the incremental cost of terminating international traffic is no more than 6›-9› per minute. The average foreign settlement rate for U.S. traffic is approximately 36›, potentially as much as four to six times our estimate of the incremental cost of terminating international traffic. In our view, these rates are above reasonable levels. We accordingly propose two primary options in this Notice to calculate settlement rate benchmarks that treat foreign carriers in a consistent manner and place some limit on the subsidies in current settlement rates. 35. Each of the methodologies for calculating benchmarks that we propose in this Notice relies on the framework described in ITU-T Recommendation D.140, which calls for cost- oriented, nondiscriminatory settlement rates. That Recommendation contains cost guidelines that identify the three specific network elements that are used to provide IMTS and the cost components for those elements to be included in cost-oriented settlement rates. The specific network elements are: (1) international transmission facilities; (2) international switching facilities; and (3) national extension (domestic transport and termination). 36. The International Bureau conducted a study in which it calculated prices for these three elements. The price calculated for the international transmission and national extension network components is based on foreign carriers' tariff rates, and the price for the international gateway element is based on data published by the ITU. The alternatives we propose in this Notice use the sum of these tariffed prices for the international transmission and national network components and the price for the international gateway switching component, which we collectively refer to as the "tariffed components price," to calculate benchmarks. We are releasing a report prepared by the International Bureau that presents the results of its study. 37. Sixty-five countries are included in the Bureau's study, generally those having the largest traffic volumes with the United States. The Bureau used data collected during the fourth quarter of 1995 through mid-1996 to calculate tariffed components prices for these sixty-five countries. A description of the Bureau's methodology for calculating the tariffed price for each network element follows:  International facility component The international facility component consists of international transmission facilities, both cable and satellite, including the link to international switching facilities. This component includes only the half-circuit on the terminating end because originating carriers have traditionally been responsible for the half circuit on the originating end of a call. High capacity circuits, normally 1.544 mbps or 2.048 mbps circuits, are used for IMTS and most telephone administrations offer these circuits to customers on a dedicated basis. The cost element for this component, therefore, is based on foreign carriers' private line rates for dedicated circuits. Multiple 64 kbps circuits are derived from the high capacity channels and multiplexed into voice grade circuits based on standard U.S. operating practices. This information, along with average monthly traffic volume per circuit, is used to convert the private line rates to a charge per minute for each country.  International gateway component The international gateway component consists of international switching centers and associated transmission and signalling equipment. Foreign carriers do not generally offer a separate tariff rate for the international gateway component, so the study relies on information published by the ITU. The cost of this component varies with the level of digital facilities.  National extension component The national extension component consists of national exchanges, national transmission, and the local loop facilities used to distribute international service within a country. Foreign carriers' domestic rates and the distribution of U.S. billed service within a country are used to compute an average charge per minute for cost of this component. 38. Appendix E summarizes the tariffed components price and each network element tariff price for the sixty-five countries. Study procedures, data collection, and estimation methods are described in further detail in the Bureau Report. B. Benchmarks 1. Benchmarks Based on Tariffed Components Prices 39. We discuss below several options for calculating settlement rate benchmark ranges based on level of economic development and carriers' tariffed components prices. We present two primary options: (1) calculating benchmark ranges based on categories of economic development levels, in which the upper end of the range for each category is based on the average of the tariffed components prices in that category and the lower end of the range is based on an estimate of the incremental cost of terminating international traffic; and (2) setting a country-specific benchmark range for each country in which the upper end of each country's range is equal to that country's tariffed components price and, again, the lower end of each country's range is based on an estimate of the incremental cost of terminating international traffic. We believe the first option best serves the goals we seek to achieve in this proceeding. While both options will reduce the excess in current settlement rates in a fair manner, we believe that the first option is more equitable and will discourage inefficient pricing. We ask for comments on each option, and on any alternative methodology options using the tariffed components prices. 40. We believe that a carrier's tariffed components price, as calculated in the Bureau Report, may provide a sound basis for establishing benchmarks. The logic of basing benchmarks on a foreign carrier's tariffed prices is that those prices are the same tariff rates charged by a foreign carrier to its domestic customers. Nondiscriminatory treatment of U.S. carriers would require that foreign carriers assess U.S. carriers a comparable charge for terminating service from the United States to that they assess their own domestic customers. We see no justifiable economic basis for requiring a U.S. carrier to pay a foreign carrier more than that carrier charges its domestic customers for the same service. Moreover, tariff rates are publicly available, so benchmarks based on such rates can be revised as the tariff rates change. We also note that the tariffed components prices represent the prices U.S. carriers would pay if they were allowed to purchase the three elements of international termination service in an unbundled fashion. While these prices are still significantly above the cost of providing termination services, they represent a substantial reduction in the current price U.S. carriers pay for these elements on a bundled basis in the form of settlement rates. 41. We recognize that benchmark settlement rates based on foreign carriers' tariffed prices will not be based on incremental cost. As discussed below, para. 42, tariff prices are substantially greater than carriers' incremental cost of terminating international traffic. Nonetheless, calculating benchmarks based on carriers' tariffed components prices, as we propose in this Notice, will move international termination prices closer to costs relative to current settlement rates. Moreover, we believe that, in the absence of data on foreign carriers' costs to terminate international traffic, these options are a reasonable method for calculating benchmarks that treat foreign carriers in a consistent manner and move in the direction of cost- based settlement rates. We reiterate that we are committed to achieving settlement rates that are based on carriers' long run incremental cost of terminating international traffic, and we will continue to work toward that goal through implementation of the benchmarks we propose in this Notice and though future revisions, if necessary, of our benchmark methodology. 42. We also note that benchmark settlement rates based on tariffed components prices will permit foreign carriers to recover more than their incremental cost of terminating international service. This is because the tariff rates used in the calculations presumably reflect foreign carriers' incremental cost plus a significant contribution to common costs. In fact, the tariffed components prices will provide a generous margin over incremental cost. Because foreign carriers' retail rates include costs that are not incurred in the termination of international services, settlement rates based on retail rates will substantially exceed incremental cost. Marketing and commercial expenses, for example, are part of the cost of providing retail service but not of operating and maintaining the facilities used for international service. Similarly, allowances for overhead expenses and uncollectible revenues would be included in foreign carriers' retail rates but not included in cost-based settlement rates. Thus, benchmarks based on tariffed components prices will fully compensate foreign carriers for the costs they incur in terminating international traffic. a. Ranges Based on Level of Economic Development 43. We propose to categorize countries by level of economic development and to establish separate benchmark ranges for each category. For the reasons discussed below, paras. 45-46, we believe that the upper end of the benchmark ranges should be based on an average of the tariffed components prices. However, the tariffed components prices are generally significantly higher in lower income countries than in upper income countries. If the tariffed components prices of lower income countries were averaged with those of higher income countries to establish benchmarks, the differential between the new benchmark rate and current settlement rates would be much greater for lower income countries than for higher income countries. Indeed, for many higher income countries, there would be little difference between an average benchmark and current settlement rates. Establishing separate ranges based on level of economic development would mitigate this effect while still capturing some of the benefits of using an average discussed below. Moreover, economic development level is a logical way to cluster the tariffed components prices for purposes of averaging because there generally is an inverse correlation between the level of tariffed components prices and a country's level of economic development. We seek comment on this proposal. 44. A standard measure of economic development is income per capita. The World Bank classifies countries on the basis of four levels of economic development using gross national product (GNP) per capita. The ITU uses the same classification scheme. The four levels of economic development are: (1) low income, GNP per capita of $726 or less; (2) lower- middle income, $726-$2,895 per capita; (3) upper-middle income, $2,896-$8,955 per capita; and (4) high income, $8,956 or more. Appendix D lists countries by the level of economic development according to this classification scheme. Because our proposed method of calculating benchmark rates results in benchmarks that are almost identical for lower-middle income and upper-middle income countries, we propose to merge these two categories into one "middle income" group for purposes of calculating and implementing new benchmark settlement rates. Doing so simplifies our country classifications and the administration of our benchmarks, and has minimal effect on the countries in these two categories. We invite interested parties to comment on this approach. 45. As Appendix E shows, the tariffed components prices for different countries vary widely. Because the tariffed components prices rely on foreign carriers' widely divergent tariffs to set prices for two of the three network elements, any inefficiencies in a foreign carrier's tariffs are captured in its tariffed components price. For example, telephone service in many countries included in the Bureau Study is provided by monopoly carriers whose tariff rates may reflect protected market positions and an ability to charge prices not related to underlying costs. Moreover, many countries have rate structures that use high international or domestic long distance charges to offset below-cost local service fees. Using the average of the tariffed components prices for an economic development category as the basis for the upper end of a benchmark range mitigates the effect of carriers' inefficient pricing structures on our benchmark calculations by averaging the most inefficient rates with those that are less inefficient. At the same time, categorizing countries by level of economic development for purposes of calculating averages results in a less severe difference between current settlement rates and the benchmarks for lower income countries than calculating one average for all countries. We emphasize that we believe that almost all countries' tariffed components prices reflect inefficiencies that result in excessive tariff prices for terminating international service. Thus, using the average as the basis for the upper end of the benchmark ranges should permit all foreign carriers to recover far more than their incremental cost of terminating international traffic plus a substantial contribution to common costs. 46. We also note that an average figure is beyond the ability of any one carrier to alter significantly, so a carrier will have no incentive to change its tariff rates to affect the level of its benchmark. In addition, using the average as the basis for the upper end of the benchmark ranges removes any reward for carriers that have the most inefficient pricing structures and encourages carriers above the average to rebalance their rates more in line with underlying economic costs. Moreover, from a practical perspective, using the average of the tariffed components prices as the basis for the upper end of the range is a significant step toward reducing settlement rates. At the same time, however, it is a cautious approach that should permit all foreign carriers to recover substantially more than their incremental cost of terminating international service plus a substantial contribution to common costs. We seek comment on whether the average is an appropriate basis for setting the upper end of the benchmark ranges and on the advantages and disadvantages of this method. 47. We propose to use the simple average of the tariffed components prices in each economic development category as the upper end of the benchmark ranges. Doing so would result in upper ranges of 15.4› for carriers in upper income countries; 19.1› for carriers in middle income countries; and 23.4› for carriers in lower income countries. These proposed benchmark settlement rates would translate into accounting rates of 30.8›, 38.2›, and 46.8›, respectively. We seek comment on the advantages and disadvantages of using a simple average to calculate the upper end of the benchmark ranges. 48. We also seek comment on alternative approaches for setting the upper end of benchmark ranges using tariffed components prices in each economic development category. Another approach, for example, would be to establish ranges that reflect the variation within each development category of individual countries' tariffed components price around the category's average tariffed components price. This approach would reflect the fact that the tariffed components prices for some countries in a category are higher than the average value while others are lower, and that some are closer to the average value than others. We could do so, for example, by adding and subtracting one standard deviation from the average tariffed components prices for each economic development category. Using such an approach would result in an upper range for benchmark settlement rates of 9› to 22› for high income countries; 12› to 26› for middle income countries; and 13› to 33› for low income countries. 49. We note that a potential problem with these alternative approaches is that the measure of variation used to calculate the lower or upper bounds of the ranges, for example, one standard deviation or 1.5 standard deviations, is significant from a statistical perspective, but may not have any particular relevance for purposes of establishing appropriate benchmark ranges. Another potential problem is that these alternative approaches, while providing a statistical measure of the variation in tariffed components prices around the average, do not address and remedy the primary reason for that variation -- the fact that the tariffed components prices reflect carriers' pricing inefficiencies. We seek comment on the advantages and disadvantages of such alternative approaches, and on whether it is appropriate to use standard deviations around the average tariffed components prices to calculate ranges. 50. We propose to use an estimate of the incremental cost of terminating international traffic for the lower end of the benchmark because it is our goal ultimately to achieve settlement rates that are more closely cost-based than are current settlement rates. As stated above, para. 31, we believe that the appropriate cost standard for establishing benchmark settlement rates is the incremental cost of terminating international traffic because rates in competitive markets would tend towards that cost. However, we do not have sufficient data at this time to calculate a precise estimate of incremental cost. AT&T has provided an estimate of its "average network cost" for termination of inbound international calls. In the absence of data on carriers' incremental costs, we propose to use AT&T's estimate as a starting point to derive a preliminary estimate of incremental cost. We emphasize that our estimate is not a precise measure of incremental costs because we do not have the data to make such a calculation. We encourage both U.S. and foreign carriers to submit incremental cost data. 51. AT&T's estimate of 7.5› per minute includes the costs of the international half- channel, gateway switching, domestic interexchange carrier distribution, and local distribution. This estimate, except for local distribution, is derived from internal AT&T average network cost data and, as AT&T states, exceeds long run incremental cost. The local distribution component, moreover, reflects the average price AT&T pays to local exchange companies (LECs) to terminate traffic. AT&T's estimate therefore contains some contributions to common costs of its own network and local exchange carrier's (LEC) networks, including the access charges that AT&T pays to the LECs for use of the local loop. An estimate of incremental cost, however, would not include contributions to common costs. To get closer to an estimate of incremental cost, therefore, we remove the common costs that we can identify and quantify in AT&T's estimate of average network costs -- the common costs included in the access charges. Removing the access charges from AT&T's 7.5› estimate results in an estimate of 6›. We note this estimate is greater than incremental cost because it includes some contributions to the common costs of AT&T's network that we cannot identify on the basis of the data provided by AT&T. 52. We recognize that this estimate of the incremental cost of terminating international traffic is based on U.S. data, and that the incremental cost of terminating international traffic in a foreign country may differ from this estimate. Nonetheless, we believe it unlikely that the incremental cost in foreign countries will vary from our estimate by more than a few cents, especially given that our estimate most likely exceeds incremental cost. At most, we believe that incremental cost in foreign countries is not more than 9› per minute. We request comment on this assumption and this approach. In particular, we request comment on whether the incremental cost of terminating international traffic in the United States is different than the cost in foreign countries. We also request comment on whether individual foreign carriers' incremental cost is different than our 6›-9› estimate, or whether individual U.S. carriers' incremental cost is different than our estimate. We encourage foreign and U.S. carriers to submit data on their costs. If the costs of foreign carriers or individual U.S. carriers are different, to what extent are they different? What causes such cost differences? For example, are there differences in the cost of certain network elements, such as the international half circuit, depending on the geographic location of a country in relation to the United States? We also note that individual foreign carriers' incremental cost may be higher than our estimate of the incremental cost of terminating international traffic if they must pay access charges to local carriers that exceed the incremental cost of terminating traffic in the local exchange. We request comment on whether such access charges should be factored into our estimate of the incremental cost of terminating international traffic. 53. We also seek comment on any other alternatives for using the tariffed components prices to establish benchmark ranges. For example, we seek comment on whether, if we establish benchmark ranges based on the tariffed components prices for different economic development categories, we should also take into account variables other than economic development. Should we take into account a country's size or geographic location in relation to the United States in calculating benchmark ranges? We note that it might be reasonable to take such variables into account if they affect foreign carriers' costs of providing international services. We request comment on how we could take such factors into account, and on what factors might be relevant for purposes of calculating reasonable benchmarks. b. Country-Specific Benchmarks 54. We also seek comment on whether we should adopt benchmarks for each country based on that country's specific tariffed components price, instead of benchmark ranges for each level of economic development. We propose that if we adopt country-specific benchmarks, we set a range for each country in which the upper end of the range is equal to the country's tariffed components price and the lower end of the range is equal to an estimate of the incremental cost of terminating international service, as discussed above. Appendix E lists the tariffed components prices for each country in the Bureau's study. 55. We seek comment on the advantages and disadvantages of establishing country- specific benchmark rates based on tariffed components prices. One advantage of this approach is that, arguably, where we have tariffed components price information for a particular country, that information represents the best available information for setting a benchmark rate for that country. However, a potential problem of setting country-specific benchmarks based on the tariffed components prices of each country's carrier(s) is that it results in different benchmarks for similarly situated countries. For example, countries that are similar in terms of geographic location, size, and economic development level may have significantly different tariffed components prices, and therefore have different benchmarks. This potential problem arises in part from using foreign carriers' domestic tariffs, which vary substantially from carrier to carrier, to calculate the national extension component. We note that this problem is mitigated by an approach that sets benchmark ranges based on averages within economic development categories as described in the preceding section. The effect of setting such a range is to treat similarly all carriers located in countries of the same economic development category. We also note that unavoidable problems with the accuracy of the data used to calculate the tariffed components prices and the inherent uncertainty in any estimation approach may caution against relying on the accuracy of an individual carrier's tariffed components prices to set a benchmark. For example, the tariffed components prices are based on exchange rates that fluctuate daily and global usage factors rather than individual country estimates. 2. Alternative Benchmark Methodologies 56. Finally, we seek comment on alternative methodologies for calculating benchmarks other than our tariffed components price approach. In particular, we request comment on whether there are differences among countries and carriers that affect the cost of providing international service and that are not adequately reflected in our tariffed components price approach. For example, does the geographic location of the country in which a carrier is located in relation to the United States, the size of a country, or its population density affect the costs of providing international service to an extent that should be reflected in our benchmark methodology? Are there other country or carrier characteristics that affect costs? We also seek comment on whether we should, and indeed whether it is possible to, calculate benchmarks that take into account such cost-causative country and carrier characteristics. Additionally, we seek comment on steps we might take to stimulate the production of incremental cost data, particularly from recognized global institutions such as the World Bank, the ITU, and the OECD. 57. We emphasize that, regardless of the method we ultimately adopt for calculating benchmarks, any carrier may challenge the validity of its benchmark on the grounds that it is not indicative of the carriers' actual costs to receive, transmit, and terminate international service. To successfully challenge a particular benchmark on these grounds, we would require the challenging carrier to demonstrate that its costs are higher than the established benchmark. A carrier may also challenge its benchmark on the grounds that it has not been calculated correctly under whatever methodology we adopt in this proceeding. C. Implementation of Benchmarks 1. Transition To Benchmarks 58. Because the benchmarks we propose here likely overestimate carriers' costs of providing international termination service, we believe U.S. carriers' settlement rates with all foreign carriers should already be at or below the benchmarks. This is particularly true for the high income, developed countries. We note that ITU-T Recommendation D.140, when adopted in 1992, called for cost-oriented settlement rates within one to five years. While some settlement rates have been dramatically reduced, most have not. In fact, a majority of settlement rates are not even within our current benchmark ranges, despite the fact that costs have declined significantly since the ranges were adopted. 59. We acknowledge the argument that substantially above-cost settlement rates are justified because they are used to subsidize network development in lower income countries. Such network development benefits not only the economies of lower income countries, but also the economies of the United States and other countries by providing the telecommunications infrastructure necessary to support international commerce and trade. This is not the only way to look at settlement payments, however. Bringing settlement payments closer to cost will, in the long run, lead to lower calling prices. Lower calling prices will in turn stimulate traffic flows. In fact, as noted above in para. 10, experience in the United States and other countries that have introduced competition in their telecommunications markets shows that additional reductions in the price of international telephone service stimulate demand. We believe that the rapid growth in internet usage and call-back, which are in part a response to high IMTS prices, are indicators that the elasticity effects of price decreases should be dramatic. The increase in U.S. outgoing traffic that will result from decreased prices -- and the associated settlement payments -- should temper any decline in revenue per minute from settlement rates. 60. In addition, the technological means for bypassing the settlements regime are developing rapidly and the current highly inflated settlement rates provide a strong incentive for such bypass. These growing bypass capabilities and incentives mean that the traditional monopolists' revenue streams no longer provide secure financing for investment in telecommunications infrastructure. We believe that open and competitive markets that welcome private capital offer a more reliable and sustainable means to finance infrastructure development than the traditional monopoly-based accounting rate system. 61. We nonetheless realize that countries will need time to make the adjustments necessary to introduce competitive reforms. We also recognize additional time may be needed to enable U.S. carriers to negotiate for lower settlement charges with their foreign correspondents without forcing undue disruption of both parties' operations. For example, carriers in many developing countries have significantly distorted rate schedules involving cross-subsidies from users of international services to those using domestic services. These carriers also may have substandard telecommunications infrastructure, including low levels of network buildout and low levels of network reliability. An immediate shift to cost-based settlement rates thus could create adjustment problems for carriers in these countries while they are trying to rebalance rates and upgrade their network. Moreover, the difference between current settlement rates and the benchmarks we propose here generally is greater for lower income countries than for upper income countries. Implementation of the benchmarks therefore will require greater reductions in current settlement rates for developing countries than for upper income countries. 62. We therefore propose a transition schedule based on countries' level of economic development that will enable all carriers to achieve rates at or below the benchmarks in a four to five year period. We base the timeframe for full compliance with our benchmarks on a country's level of economic development because, for the reasons discussed in the preceding paragraph, we believe that a U.S. carrier's ability to negotiate a charge that complies with our benchmarks without undue disruption of its or its foreign correspondent's operations diminishes as the level of economic development decreases. We seek comment on this approach. 63. We propose to require that settlement rates for U.S. carriers with high income countries be at or below our benchmarks within one year of the effective date of our order in this proceeding. Alternatively, we request comment on whether the transition schedule for upper income countries should be two years. For upper middle income and lower middle income countries, we propose to require that settlement rates be at or below our benchmarks within two years. Alternatively, we request comment on whether we should adopt a three year transition schedule for these two middle income categories, or whether we should adopt a three year transition schedule for carriers in the lower middle category only. For U.S. carriers in low income countries, we propose to require that settlement rates be at or below our benchmarks within four years. Alternatively, we request comment on whether the transition schedule for low income countries should be five years. We seek comment on these proposed transition schedules. 64. Additionally, we request comment on whether U.S. carriers should be asked to make reasonable progress in achieving rates at or below the benchmarks throughout the transition period. For example, should we ask U.S. carriers to negotiate a certain percentage reduction annually of the spread between current settlement rates and benchmarks for carriers during the transition period? 65. We also seek comment on alternatives to our proposed transition approach based on level of economic development. For example, we might have U.S. international carriers negotiate settlement rates at or below our proposed benchmarks with all of their foreign correspondents by a date certain. This date could be a defined period, such as two to four years from the effective date of the rules we adopt in this proceeding. If we adopt this approach, should we require U.S. international carriers to demonstrate periodically that they are making progress in negotiating settlement rates at or below our proposed benchmarks? This method would have the advantage of simplicity and consistency for all international routes, but would lack the flexibility inherent in the phased-in approach discussed above. We seek comment on the advantages and disadvantages of this proposal. 66. Another possibility would be to adopt a single deadline for transition to benchmark settlement rates, but to allow U.S. carriers to request a waiver of this requirement for particular routes. Reasons supporting waivers of the benchmarks might include undue hardship of moving to the benchmark for some countries due to low levels of economic development or similar reasons. Moreover, to the extent governments are making a good faith effort to move to cost-based settlement rates by adopting pro-competitive policies, this factor might justify a temporary waiver of our benchmark rules. We seek comment on the reasonableness of this approach. 67. We also seek comment on whether we should provide an additional period of transition in negotiations with foreign carriers for which annual reductions in the spread between their current settlement rate and their benchmark will exceed a certain percentage, such as twenty-five percent. Alternatively, we seek comment on whether we should provide additional transition time for negotiations with foreign carriers for which transition to the relevant benchmark would entail a loss of greater than a certain percentage of their annual revenue. Such an additional transition period may be necessary to prevent carriers facing significant declines in their accounting rates as a result of our benchmarks from experiencing undue disruptions of their networks. On the other hand, a longer transition period for carriers whose settlement rates are substantially above their benchmark could be seen as rewarding carriers who had previously refused to negotiate accounting rate reductions. We seek comment on the advantages and disadvantages of this proposal. We also invite parties to comment on whether these transition periods would be consistent with the United States' MFN obligations in the event the GBT reaches an agreement on liberalizing trade in basic telecommunications service. 68. We reiterate that we believe U.S. carriers' settlement rates should currently be at or below the benchmarks proposed here, particularly with high income, economically developed countries. The transition rate goals and transition periods we propose are not intended to be a substitute for the benchmarks proposed in this Notice or to postpone achieving rates at or below the benchmarks. 2. Application of Benchmark Settlement Rates to Countries Committed to Competitive Reform 69. As discussed above, para. 20, we believe the best way to create an alternative to the traditional accounting rate system is to introduce effective competition. Indeed, we believe that in competitive markets our benchmark rates would not be necessary because international call termination rates in such markets will be below any benchmark rates that we adopt. It may, therefore, be appropriate for the Commission to forbear from applying its benchmark rates where there is effective competition for international services on a route and where substantial progress has been made toward achieving rates that represent the incremental cost of terminating international service. We seek comment on whether this would be appropriate. 70. We also seek comment on whether it would be appropriate to permit additional flexibility in the application of our benchmarks beyond the transition periods we propose here for U.S. carriers and their foreign correspondents in developing countries that have demonstrated an actual commitment to fostering entry and promoting competitive market environments. Such a policy would recognize the challenges to developing countries posed by the introduction of cost- based rates and the consequent reduction or elimination of the revenue stream generated by the current system. At the same time, however, it would also encourage the development of competitive markets necessary to achieve cost-based settlement rates. 71. We believe that both foreign and U.S. consumers will ultimately benefit from greater competition in foreign telecommunications markets. Experience in the global telecommunications market has shown that where competitive market structures have been promoted, prices have decreased, service quality has increased, and networks have been expanded. For example, as a result of procompetitive policies at the local loop, investment in networks and cable telephony in the United Kingdom has increased dramatically. In Chile, where the telecommunications market has been privatized and liberalized, the number of telephones more than trebled from 591,000 lines in 1988 to more than 1.8 million in mid- 1995. The digitalization of Chile's telecommunications network increased from 37.9 percent in 1988 to 100 percent in early 1995. Competition in the market for terminating international services will exert price pressure on settlement rates. We have already seen this in markets that have introduced competition, such as Sweden, where it costs U.S. carriers 9› a minute to terminate an international call. We therefore believe that both network development and settlement rate reductions are best achieved by a policy that promotes the development of competitive markets in developing countries. 72. We thus propose to consider additional flexibility in the application of our benchmarks for U.S. carriers serving countries in the low and middle income development categories. We propose as a threshold requirement that a country demonstrate an actual commitment to promote competition before we will consider additional flexibility. We further propose to allow flexibility based on a more limited commitment to competition for countries that are in the lower middle income and low income categories and whose level of network development is very low. We seek comment on this proposal. We also seek comment on the appropriate measure for determining the level of network development. For example, should we use teledensity or some other measure? We also seek comment on the level of economic and network development to use as thresholds for allowing flexibility in the application of our benchmarks. Additionally, we seek comment on whether we should require a link between any additional flexibility we grant and infrastructure development, and how this might be done. 73. We propose to leave initiatives on proposals for developing country flexibility to carriers. If a U.S. carrier and its foreign correspondent believe they need additional flexibility beyond any transition rate goals and transition periods we adopt, they may propose an alternative transition plan. These carriers may, for example, propose a plan allowing additional time to achieve settlement rates within the appropriate benchmark. These carriers may also propose an alternative schedule for achieving settlement rates within the appropriate benchmark that is of several years duration. However, we will seek assurances that these carriers will achieve settlement rates within the benchmarks by the end of the transition plan. By the end of the transition plan the settlement rates between the carriers must be in full compliance with the benchmark rates. We invite interested parties to comment on this approach for allowing additional flexibility in the application of our benchmarks in certain circumstances. We also invite parties to comment on whether this approach would be consistent with the United States' MFN obligations in the event the GBT reaches an agreement on liberalizing trade in basic telecommunications service. 74. We also note that it may be appropriate to consider additional alternatives to our international settlements policies for carriers in developing countries beyond the flexibility in application of our benchmarks that we propose here. We seek comment on whether we should permit such additional alternatives for developing countries. 3. Applying Benchmarks to Address Anticompetitive Behavior 75. While we are cognizant of the difficulties some U.S. carriers will face in negotiating settlement rates at or below our benchmarks, the subsidies contained in above-cost settlement rates can distort performance in the IMTS market. These distortions impede the Commission's policy of creating greater competition in the IMTS market in order to lower international calling prices for U.S. consumers. As discussed above, para. 11, there is a great potential for distortion flowing from above-cost settlement rates when a foreign carrier collecting those rates is able to send its switched service over resold international private lines into the United States, but U.S. carriers are unable to send their traffic over private lines in the reverse direction, and must continue to pay a relatively expensive settlement rate. It has also been argued that foreign carriers could use the subsidy embedded in above-cost settlement rates to cross-subsidize an affiliate providing international services in the U.S. market. According to this argument, a foreign carrier's U.S. affiliate could afford to price its services in the U.S. market below the costs of providing those services because its foreign parent would be earning substantial subsidies from terminating traffic at above-cost settlement rates. However, if a foreign carrier is collecting cost-based settlement rates, or if its ability to collect above-cost settlement rates is constrained by the existence of effective competition in its home market, concerns about anticompetitive behavior will be significantly diminished. Therefore, we propose to take into account the level of competitive risk posed by above-cost settlement rates in enforcing our settlement rate benchmarks. We seek comment on this approach. We also seek comment on whether there are safeguards applied by other countries to address these potential anticompetitive concerns that we should consider. 76. We propose to condition various types of authorizations to provide U.S. international services to address the potential market distortions created by above-cost settlement rates. First, when a carrier seeks authorization to provide international facilities-based service from the United States to an affiliated foreign market, whether to provide switched service or private line service, we propose to condition any authorization to serve that affiliated market on the foreign affiliate offering U.S. licensed international carriers a settlement rate within the benchmark range we are proposing in this Notice. Consistent with our existing International Settlements Policy (ISP), all U.S. carriers would receive the same settlement rate for traffic on that route. If, after the carrier has commenced service to the affiliated market, we learn that the carrier's service offering has distorted market performance on the route in question, we propose to order that settlement rates to that country be reduced to the bottom of the range (our estimate of cost-based termination) or to revoke the authorization of the carrier to serve the affiliated market. 77. We ask for comment on what mechanism or approach we should use to determine when there has been a distortion of competition in the IMTS market and the lower settlement rate should be applied. For example, if a foreign carrier seeks to distort traffic flows in order to increase net settlement payments to its foreign affiliate, to evade requirements of our ISP (including our proportionate return and no special concessions rules), or to use substantially above-cost settlement rates on the foreign end to price its services in the U.S. market in an anticompetitive fashion, would such actions constitute distortion of competition? We also invite comment on how the Commission's reporting system could be modified in order to make monitoring and enforcement more effective. 78. We seek comment on this proposal. We also seek comment on how this proposal would affect the effective competitive opportunities (ECO) test we adopted in our Foreign Carrier Entry Order. For example, should this proposal be used in conjunction with the ECO test, replace the ECO test, or should we modify the ECO test to ensure that it is compatible with this proposal? 79. We emphasize that the purpose of our proposal is to prevent U.S.-licensed carriers from distorting the IMTS market through service to affiliated markets with excessive settlement rates. The proposal does not serve as a barrier to market entry for foreign carriers. Under the proposal, we would not limit the ability of foreign carriers to enter the U.S. market. Rather, all U.S.-licensed carriers (U.S. or foreign-owned) would face similar conditions on service to affiliated foreign markets. 80. We also seek comment on whether it is necessary to apply this safeguard. The concern has been raised that foreign carriers would have the incentive to use the subsidy embedded in above-cost settlement rates to cross-subsidize an affiliate providing international services in the U.S. market. This argument, however, appears to ignore the opportunity costs to the foreign parent of offering service through an affiliate in competition with U.S. carriers that formerly purchased termination service from the parent. In serving its home market directly through its affiliate, the foreign parent would no longer receive the settlement payment it formerly received from U.S. carriers to terminate traffic in that market. We seek comment on this analysis. In particular, we request comment on whether a foreign carrier has the incentive or ability to use above-cost settlement rates to cross-subsidize a U.S. affiliate. We also request comment on whether, if a foreign carrier does not have this incentive or ability, competitive safeguards are nonetheless necessary to help ensure that irrational anticompetitive behavior does not occur. 81. Our second proposed condition affects the resale of international private lines to provide switched service, which we believe presents a significant danger of competitive distortion. Currently we address this concern by only permitting the resale of international private lines for switched service to countries which offer equivalent resale opportunities to U.S. carriers. We now propose to address this concern by imposing settlement rate conditions on authorizations to resell international private line services to provide switched services. 82. In the case of international private line resale, traffic is not settled because it is carried over private lines. Therefore, it could be meaningless to require the applicant to reduce its settlement rate to a level within the benchmark range because a pure reseller would not have established a settlement rate with any U.S. correspondent. We accordingly propose to grant carriers' applications for authority to resell international private lines to provide switched service to the United States on the condition that accounting rates on the route or routes in question are within the benchmark range. Under such a rule, if settlement rates are outside the benchmark range, a carrier would not be permitted to use its authorization to provide international private line resale service until such time as settlement rates on the route in question are brought within the benchmark range. When settlement rates are within the benchmark range, the authorization could be used. This condition would also apply to U.S. carriers seeking to provide international private resale line service because they, too, would have the ability to distort competition on the route to the extent they accepted "one-way bypass" traffic from foreign carriers. 83. If we learn that competition on the route has been distorted in fact, we propose to order all U.S. international carriers to pay settlement rates at the low end of the benchmark range, that is, at a level that we believe represents the actual cost of terminating international traffic in the United States. By ordering all carriers to pay settlement rates at the lower end of the benchmark range for switched traffic, the Commission would eliminate the financial distortion from above-cost settlement rates that makes competitive harm possible. We ask for comment on what mechanism or approach we should use to determine when competition has been distorted and the lower settlement rate should be applied. For example, if our monitoring determines that the inbound to outbound ratio of international switched minutes subject to settlements shifts after such resale is authorized to a country which does not allow inbound resale, should this be considered evidence of "one-way bypass" and qualify as a distortion of competition? 84. Alternatively, we seek comment on whether we should require that accounting rates on the route or routes in question be at the low end of the benchmark range as a condition of carriers' authorizations to resell international private lines to provide switched service to the United States. We also seek comment on how our proposed treatment of international resale would affect our equivalency test. For example, should this proposal be used in conjunction with our equivalency test, replace our equivalency test, or should we modify our equivalency test to ensure that it is compatible with this proposal? 85. Of course, we have long taken the view that effective competition will best ensure that settlement rates are set at cost-based levels and thereby eliminate the subsidy in above-cost rates. We therefore propose to presume that carriers from countries that have opened their markets to meaningful competition have fulfilled these conditions. We seek comment on this proposal and on whether we should evaluate the competitiveness of a market through an ECO test, or some other means. We also seek comment on whether we should impose these conditions on existing Section 214 certificate holders that serve affiliated markets. We believe that the same potential for distortions in the IMTS market from foreign carriers collecting above- cost accounting rates may also exist with respect to existing authorized carriers. Alternatively, we also seek comment on whether specific safeguards provided for in existing section 214 authorization conditions are sufficient to address these competitive concerns, without the conditions we propose here. 86. We also invite parties to comment on whether these safeguards would be consistent with the United States' MFN obligations in the event the GBT reaches an agreement on liberalizing trade in basic telecommunications service. 4. Enforcement Proposals 87. While we expect carriers to negotiate settlement rates at or below our benchmarks within the relevant transition periods, we believe additional steps may be needed to assure adequate progress. We propose to identify foreign carriers that are reluctant to engage in meaningful progress toward negotiating settlement rates at or below the relevant benchmark. Specific factors we would examine to identify such foreign carriers are: (1) lack of a commitment to competitive reforms in the near future; (2) settlement rates significantly above the relevant benchmark and transition rate goals; (3) substantial, rapidly growing net settlement payments, particularly to countries receiving large net settlement payments; and (4) an unwillingness to negotiate reasonable settlement rates, as evidenced by continuing high rates and no meaningful change in recent years. We will, however, consider these factors in the context of any unique or unusual problems faced by individual carriers or countries such as unforeseen natural disasters, countries impoverished by armed conflict, or routes that have exceedingly low volumes. 88. Once we have identified the foreign carriers that have failed to make meaningful progress toward complying with our benchmarks, we propose several steps to reduce settlement rates with those carriers consistent with our new benchmarks. First, we propose to convey to the responsible government authorities our concern about continued high settlement rates and the lack of meaningful progress, and to seek their support in lowering settlement rates. In our contacts with the responsible government authorities, we propose to emphasize the need for cooperation in achieving the goal of cost-based rates; enlist the active participation of government authorities in achieving that goal; cite relevant ITU recommendations such as Recommendation D.140; and suggest further discussions between responsible government authorities. We request comment on whether additional steps in multilateral organizations such as the ITU would be appropriate in such instances. 89. Second, we would consider stronger steps in those cases where foreign carriers fail to respond to U.S. carriers' initial efforts to achieve settlement rate progress. In each case our actions would apply to U.S. carriers within our jurisdiction, not to their foreign correspondents. We would consider:  Directing U.S. carriers to negotiate settlement rate agreements that provide for a fixed expiration date until a foreign carrier agrees to a reasonable schedule of reductions aimed at reaching the benchmark level.  Directing U.S. carriers to settle at a rate that is no higher than the transition rate goals until a foreign carrier agrees to a reasonable schedule of reductions aimed at reaching the benchmark level. Each U.S. carrier would then be required to settle with the correspondent at that rate and resume payment at a negotiated rate only after it is determined by the Commission that the lower rate represents adequate progress.  Directing U.S. carriers to settle at or below the benchmark rate and to continue paying at that rate until the Commission determines that meaningful progress in the form of accounting rate reductions is being made. Normal settlement practices would then resume.  Directing U.S. carriers to pay a settlement rate no higher than the benchmark rate. We invite comments on the advantages and disadvantages of each of our proposed approaches, or on any alternative approaches, to encourage foreign carriers to engage in meaningful settlement rate negotiations that result in settlement rates that are at or below the benchmarks. 90. Finally, we reiterate our commitment to vigorous enforcement of the nondiscrimination requirement of our ISP. Recent International Bureau actions make clear that discriminatory and retaliatory behavior by foreign carriers in violation of the ISP will not be tolerated, and we will continue in the future to take strong actions against foreign carriers in response to complaints of discrimination by U.S. carriers. 5. Effect of Settlement Rate Reductions on U.S. Collection Prices. 91. As noted above, we seek to reform the traditional settlement rate system in order to provide U.S. consumers with just and reasonable prices for IMTS service. Thus, we seek comment on how to encourage U.S. carriers to reflect the reductions they receive in their settlement rates. If U.S. carriers lowered their rates in response to settlement rate reductions, calling volume would be stimulated. This would not only likely benefit U.S. carriers by increasing their collection revenues, it would also benefit foreign carriers because they could offset lower settlement rate levels with an increase in the number of minutes terminated, thus moderating the effect of lowering settlement rate levels. It would also further our goal of ensuring that U.S. consumers enjoy international calling prices that are more closely related to underlying costs than are current prices. IV. Conclusion 92. In this Notice, we tentatively conclude that our benchmark ranges for international settlement rates adopted in 1992 no longer accurately reflect supply conditions in the industry and, therefore, need to be revised. We propose to replace our current benchmark ranges with ranges based on level of economic development and foreign carriers' tariffed rates. We request comment on other alternatives to setting benchmark ranges. 93. We propose to implement our new benchmarks in a way that promotes competition in the global telecommunications services market and prevents potential anticompetitive conduct in the U.S. market by foreign carriers. We propose a transition period that will enable all carriers to achieve settlement rates at or below the benchmarks within four years. We also propose to provide flexibility in the application of our benchmarks to developing countries that are committed to competitive reforms. Finally, we suggest potential steps that may be required by the Commission if reasonable progress towards cost-based settlement rates fails to materialize. 94. We seek comments from interested parties, including foreign governments and carriers, on the options we propose for revising our benchmark settlement rates and our suggestions to implement the benchmarks in a way that promotes competition and deters anticompetitive conduct. We also encourage interested parties to provide additional information and data that will assist us in establishing our benchmark rates. V. Procedural Issues A. Ex Parte Presentations 95. This is a non-restricted notice-and-comment rulemaking proceeding. Ex parte presentations are permitted, except during the Sunshine Agenda period, provided that they are disclosed as provided in the Commission's rules. See generally 47 C.F.R.  1.1202, 1.1203, and 1.1206. B. Initial Regulatory Flexibility Analysis 96. Pursuant to the Regulatory Flexibility Act of 1990, 5 U.S.C.  601-612, the Commission's Initial Regulatory Flexibility Analysis with respect to the Notice of Proposed Rulemaking is as follows: 97. Reason for Action: The Commission is issuing this Notice of Proposed Rulemaking seeking comment on possible changes in the benchmark ranges applied to settlement rates for international message telephone service between U.S. facilities-based carriers and foreign carriers and related issues. The Commission believes that its benchmark rates should be revised to reflect recent technological improvements, their associated cost reductions, and the market structure changes occurring in the global telecommunications market. We also believe these revisions are necessary to move settlement rates closer to the actual costs incurred by foreign carriers to terminate international traffic 98. Objectives: The objective of this proceeding is to attain reform in the international accounting rate system and thereby help ensure lower international calling prices for consumers. In particular, this proceeding seeks to remove the primary obstacle to accounting rate reform -- the anticompetitive effects of substantially above-cost settlement rates. The Commission will achieve this objective by revising its benchmark settlement rates so that they more closely resemble the underlying costs of providing international termination services. 99. Legal basis: The Notice of Proposed Rulemaking is adopted pursuant to Sections 1, 4(i), 201-205 and 303(r) of the Communications Act of 1934, as amended, 47 U.S.C. 151, 154(i), 201-205, and 303(r). 100. Description, potential impact, and number of small entities affected: The Commission has not developed a definition of small entities applicable to international facilities- based common carriers. Therefore, the applicable definition of small entity is the definition under the Small Business Administration (SBA) rules applicable to Communications Services, Not Elsewhere Classified. This definition provides that a small entity is expressed as one with $11.0 million or less in annual receipts. Based on preliminary 1995 data, at present there are 29 international facilities-based common carriers that qualify as small entities pursuant to the SBA's definition. The number of small international facilities-based common carriers has been growing significantly, and by the end of 1996 that number could increase to approximately 50. The revised benchmark rates will apply to all international facilities-based common carriers, including small entities, that enter into an operating agreement with a foreign carrier that provides for the payment of settlement rates. We note that the revised benchmark rates should result in lower settlement rates for carriers. After evaluating the comments in this proceeding, the Commission will further examine the impact of any rule changes on small entities and set forth findings in the Final Regulatory Flexibility Analysis. The Secretary shall send a copy of this Notice of Proposed Rulemaking to the Chief Counsel for Advocacy of the Small Business Administration in accordance with Section 603(a) of the Regulatory Flexibility Act, Pub. L. No. 96-354, 94 Stat. 1164, 5 U.S.C. 601, et seq. (1981). 101. Reporting, recordkeeping and other compliance requirements: None. 102. Federal rules which overlap, duplicate or conflict with the Commission's proposal: None. 103. Any significant alternatives minimizing impact on small entities and consistent with stated objectives: The Notice of Proposed Rulemaking solicits comments on a variety of alternative methodologies for calculating benchmark settlement rates, but these have no impact on small entities. The Notice of Proposed Rulemaking also solicits comments on enforcement mechanisms that may be necessary to support U.S. carriers, including small entities, in their negotiations with foreign carriers. We seek comment on the impact of these alternatives on small entities. 104. Comments are solicited: Written comments are requested on this Initial Regulatory Flexibility Analysis. These comments must be filed in accordance with the same filing deadlines set for comments on the other issues in this Notice of Proposed Rulemaking, but they must have a separate and distinct heading designating them as responses to the Regulatory Flexibility Analysis. The Secretary shall send a copy of the Notice to the Chief Counsel for Advocacy of the Small Business Administration in accordance with Section 603(a) of the Regulatory Flexibility Act, 5 U.S.C.  601, et seq. C. Comment Filing Procedures 105. Comments and reply comments should be captioned in IB Docket No. 90-337 only. Pursuant to applicable procedures in Sections 1.415 and 1.419 of the Commission's rules, 47 C.F.R.  1.415, 1.419, interested parties may file Initial Comments on or before February 7, 1997, and Reply Comments on or before March 10, 1997. To file formally in this proceeding, you must file an original and four copies of all comments, reply comments, and supporting comments. If you want each Commissioner to receive a personal copy of your comments, you must file an original and nine copies. Comments and reply comments should be sent to Office of the Secretary, Federal Communications Commission, 1919 M Street, N.W., Room 222, Washington, D.C. 20554, with a copy to Kathryn O'Brien of the International Bureau, 2000 M Street, Room 822, Washington, D.C. 20554. Parties should also file one copy of any documents filed in this docket with the Commission's copy contractor, International Transcription Services, Inc. 2100 M Street, N.W., Suite 140, Washington, D.C. 20037. Comments and reply comments will be available for public inspection during regular business hours in the FCC Reference Center, 1919 M Street, N.W., Room 239, Washington, D.C. 20554. D. Ordering Clauses 106. Accordingly, IT IS ORDERED that, pursuant to Sections 1, 4(i), 201-205, and 303(r) of the Communications Act of 1994, as amended, 47 U.S.C.  151, 154(i), 201-205, and 303(r) a NOTICE OF PROPOSED RULEMAKING is hereby ADOPTED. 107. IT IS FURTHER ORDERED that the Secretary shall send a copy of this NOTICE OF PROPOSED RULEMAKING, including the regulatory flexibility certification, to the Chief Counsel for Advocacy of the Small Business Administration, in accordance with paragraph 603(a) of the Regulatory Flexibility Act, 5 U.S.C.  601 et seq. (1981). Federal Communications Commission William F. Caton Acting Secretary The Appendices to the Order are Excel and Harvard Graphics Files identified below and are contained in the Zip Version. APPENDIX A -- 96484a.prs APPENDIX B -- 96484b.xls APPENDIX C -- 96484c.prs APPENDIX D -- 96484d.xls APPENDIX E -- 96484.xls ----------------------------------------------------------------------------------------------------- I. Introduction and Summary This report describes a study undertaken by the Commission's International Bureau to develop prices for each of the three network elements identified by the International Telecommunications Union (ITU-T) used to provide international telephone service. These prices are used by the Commission to calculate proposed international settlement rate benchmarks in the absence of current, reliable data on the costs foreign carriers incur to terminate international traffic. In this report we describe the structural framework developed to calculate these prices, and the data collection procedures and tariff rate information used as inputs in the model. We explain the estimation procedures used to compute the network element prices, including the underlying assumptions for each element, and present several examples to demonstrate the estimation procedures. We provide a summary of each country's prices and append to the report the underlying basic tariff rate information for each of the countries in the study. When the Commission originally developed and adopted its benchmark settlement rate ranges in 1992, it was unable to estimate foreign telephone administrations' costs to terminate calls from the United States because relevant, reliable information on those costs for international service did not exist in the public domain. Indeed, such information may not have been available to many of the telephone administrations themselves. Useful information needed to develop reliable cost estimates for terminating international service is still largely unavailable to the public. Ideally, the goal of this study would be to estimate foreign telephone administrations' total service long run incremental cost (TSLRIC) of receiving and terminating calls from the United States. In the absence of the cost data needed to perform such a study, this report seeks to develop alternative data to be used in establishing benchmarks that more closely approximate costs than current settlement rates. In particular, we calculate prices for the three network elements that are used to provide IMTS identified by the ITU in Recommendation D.140. These three elements are: (1) international transmission facilities; (2) international switching facilities; and (3) national extension (domestic transport and termination). This report uses foreign carriers' unbundled tariff rates to calculate prices for each of these three elements based on cost components identified by the ITU for each element. We refer to these prices as the "tariffed component price" of each element, and the sum of these prices for each foreign carrier as the "tariffed components price" (TCP). We note that the TCPs exceed the underlying costs borne by foreign carriers to handle international service because they include a potentially significant profit component, overhead costs that are inappropriate for international service, and various joint and common costs of domestic and international service. We also note that the tariffed prices used to calculate the TCPs are the same tariff rates that a foreign carrier charges to its domestic customers. TCPs for sixty five countries, primarily those with which the United States has its largest calling volume, were completed for the study. The countries represent all geographic regions of the world and various levels of economic development. These sixty five countries accounted for 69 percent of U.S. IMTS minutes in 1994. The TCPs for each country are summarized in Table 1. More detailed results, which include each of the component elements described in this report, are presented in Appendix B. Table 1. Tariffed Components Prices (per minute) Country TCP Country TCP Argentina 32.1› Jordan 23.0› Australia 18.7› Kenya 42.6› Austria 31.4› Korea 12.8› Bahamas 19.9› Kuwait 9.0› Barbados 12.0› Malaysia 22.4› Belgium 14.1› Mexico 16.8› Bermuda 9.9› Netherlands 9.8› Brazil 27.8› New Zealand 23.8› Chile 18.6› Nicaragua 12.3› Colombia 18.5› Norway 11.6› Costa Rica 10.3› P.R. of China 17.7› Czech Republic 19.0› Pakistan 26.7› Denmark 14.4› Panama 19.4› Dominican Rep. 14.5› Peru 16.1› Ecuador 10.3› Philippines 23.9› Egypt 17.2› Poland 24.6› El Salvador 11.8› Portugal 23.9› France 17.5› Russia 35.4› Germany 19.8› Singapore 7.6› Greece 23.0› South Africa 16.9› Guatemala 10.3› Spain 18.1› Guyana 12.0› Sweden 10.0› Haiti 30.4› Switzerland 20.6› Honduras 16.6› Taiwan 13.9› Hong Kong 7.0› Thailand 17.1› Hungary 14.4› Trinidad 14.6› India 31.2› Turkey 17.9› Indonesia 35.5› U.A.E. 7.7› Ireland 18.0› United Kingdom 13.0› Israel 8.5› Uruguay 22.3› Italy 18.2› Venezuela 23.8› Jamaica 8.7› Vietnam 24.7› Japan 19.7› Several findings emerge from the study:  TCPs are substantially below the benchmark ranges adopted by the Commission in 1992.  TCPS are generally significantly lower than U.S. carriers' current settlement rates, on average 60 percent lower.  TCPs are much closer to costs than current settlement rates.  There is substantial variation in the TCPs, ranging from 7› per minute to 43› per minute.  Current settlement rates for some countries are lower than their TCPs.  There are wide variations in tariff rates among countries.  International transmission facility tariffed components range from 2.4› per minute to 25.5› per minute.  Local distribution tariffed components range from zero to 25.2› per minute.  The range of current settlement rates, 9›-$1.30 per minute, is much wider than the TCP range, 7›-43›. II. Structural Framework The ITU-T recommendation that calls for cost-oriented, nondiscriminatory accounting rates describes the guidelines for cost elements used by telephone administrations to terminate international calls. The ITU-T identifies three network elements that are used to provide international telephone service: (1) international transmission facilities, (2) international switching facilities, and (3) national extension. International transmission facilities consist of international terrestrial transmission or submarine cables, international satellite transmission, or a combination of these facilities. The facilities that compose this network element include the links between the earth stations or cable landing stations and the international switching facilities. International switching facilities consist of international switching centers, including their associated transmission and signalling equipment. Finally, the national extension element includes that part of the national exchanges, national transmission facilities, and the local loop (if specified in the agreement) that is used to terminate international telephone service. Each network element has related costs according to the recommendation. The costs are divided into two broad categories: direct costs and indirect or common costs. Direct costs include: (1) investment costs; (2) operation and maintenance costs; (3) rental and lease costs of telecommunications facilities including direct transit leasing costs, where applicable; (4) switched transit costs, where applicable; (5) cost of access to national or local networks, if applicable; and (6) directly attributable research and development costs. Indirect or common costs include: (1) general administration costs; (2) management systems; (3) other research and development; and (4) appropriate taxes. The detailed information identified by the ITU-T that would be needed to estimate cost- based accounting rates is not publicly available and, therefore, cannot be used to establish such rates. Foreign telephone administrations' tariff rates, however, are available for most countries. Tariff rates that correspond to the network elements of the structural framework adopted by the ITU-T in Recommendation D.140 can, therefore, be used to calculate benchmarks that more closely approximate costs than current benchmarks or current settlement rates. The model used in this study to estimate the TCPs is based on the ITU-T's structural framework for cost-oriented accounting rates. The model produces TCPs that exceed foreign termination costs for several reasons. Tariff rates include costs associated with providing retail communications service to consumers which would not be included in cost-based settlement rates. There is, for example, an allowance for uncollectible billings in tariff rates that is not included in termination costs for international, wholesale service. Similarly, general overhead expenses associated with retail service are part of the cost of providing retail service, but not international termination services. Moreover, the retail tariff rates are set in most other countries by monopoly suppliers and, as a result, generally reflect monopoly rents. In short, the model produces TCPs that are well above costs. III. Data Collection The first step in the study was to collect information on foreign carriers' tariffed rates. The original study sample included seventy-one countries. Sixty of the countries were selected for the study because they represent the largest volumes of international service with the United States. The remaining eleven countries were chosen to ensure proper representation of all geographic regions. These countries accounted for 95 percent of the U.S. international telephone traffic in 1994. A questionnaire was sent to these seventy one countries seeking tariff information for international dedicated (private line) services from each country to the United States and for local and long distance service within the countries. The information used in the study came from published tariffs or directly from the telephone administrations. The data collection period was the fourth quarter of 1995 through June 1996. During this time, the original information was supplemented with revisions and corrections. Responses from sixty-five countries contained information that could be used in the study. Thus, six countries were excluded from the study due to data deficiencies. International telephone service between the United States and most countries included in the sample is transmitted over international telephone networks equipped with high speed digital circuits, either T1 (1.544 Mbps) or E1 (2.048 Mbps) circuits. Most telephone administrations offer international dedicated service to their customers using the same type of high capacity circuit. Therefore, for the international facilities component, the questionnaire requested detailed information for all available rates for international dedicated services between the sample countries and the United States, such as the tariff rates for service offerings of all pertinent bandwidths, the transport media used to provide the service offerings, the availability of multi- year pricing plans, and any volume discount options offered to users. There are exceptions. The telephone administrations in Guyana and Haiti, for example, do not publish tariffs for international dedicated circuits, but they offer dedicated service to the United States under negotiated rates. For these two administrations, the highest tariff rate for dedicated service to the United States using a T1 or E1 circuit from another country in the same region is used as a proxy for the tariff rate for service from Guyana and Haiti. For Guyana, the price of an E1 circuit available for international service in Brazil is used. For Haiti, the price for a T1 circuit in Barbados is used. For other telephone administrations that do not offer international dedicated service with either T1 or E1 circuits, the published tariff rate for the highest bandwidth circuit is used. Tariffs for international dedicated service used in the study are listed in Appendix C. For the national extension component, the questionnaire solicited detailed information on the existing prices charged to the telephone administrations' customers in each country for domestic direct dialed telephone service. For each country in the sample, information on the tariff rates for all the available rate periods and rate bands was collected, primarily from public telephone directories. This information includes the hours of the day and days of the week when the rates are in effect, and the distances for the rate bands. Information was also collected for volume pricing plans available to customers for domestic telephone service in the sample countries. In addition, information on each country's network configuration for international telephone traffic, domestic numbering plan, and other related information was collected. Tariffs for domestic telephone service are characterized by substantial structural variations. The number of rate periods, for example, varies from one to five, and the number of mileage rate bands ranges from one to fourteen. Several telephone administrations have tariffs with different rate periods for local service and long distance service. Three of the studied countries (Barbados, Hong Kong, and Kuwait) do not charge their customers for domestic calls. Some administrations offer volume discounts, e.g., Germany, Japan and Norway, which are used in the study. A detailed summary of the tariff rates for domestic public switched telephone service is provided in Appendix D. Data published by the ITU-T provide the source for the international switching facilities rate component figures used in the study. The data, which are compiled from information used by the TEUREM member countries to establish accounting rate shares for service among them, have rates for the international exchange component that vary inversely with the level of digitization. A high relative level of digitization is associated with a lower absolute accounting rate share for the international exchange component. This relationship between digitization levels and accounting rate shares suggests that costs decline as the level of digitization rises. IV. Estimation Procedures The information collected from the responses to the questionnaire, along with ITU-T data, is used to estimate a price per minute to terminate switched message telephone service from the United States. As noted above, the composite of the prices for each network element is referred to as the tariffed component price (TCP). Again, the three network elements are: (1) international transmission facilities; (2) international switching facilities; and (3) costs associated with the local distribution (or "national extension") of calls within the country. The methods used to compute the price for each element are discussed below. A. International Transmission Facility Tariffed Component Prices Many telephone administrations offer international private line service to their customers using high capacity circuits. Typically, these are 1.5 or 2.0 Mbps circuits. These circuits are functionally equivalent to the dedicated circuits used by telephone administrations to provide IMTS. For international telephony, telephone administrations use high capacity circuits (e.g., 1.5 or 2.0 Mbps facilities) to interconnect with U.S. facilities-based carriers. The rates charged by telephone administrations for dedicated private line service, therefore, are used to estimate their international transmission components. If an administration offers private line service to its customers using high-speed digital facilities, the rate charged for the service is used to estimate the international transmission facility component. If an administration offers private line service using slower speed facilities, then the tariff rate that applies to such services is used to compute the estimate. Some administrations offer multi-year service options at reduced rates to their customers. Others offer reductions from their tariff rates to customers with large billings. The rates used in the study reflect these options in those situations where they are available. The tariff rates, including the service bandwidth and time period, are shown in Appendix C. Estimating the TCP for international transmission facilities on a basis that is comparable to accounting rates requires converting the monthly private line rates to a charge per minute. The first step in this process is to calculate the number of voice grade circuits that are derived from a private line half-channel. A 2.048 Mbps half-channel is comprised of thirty 64 Kbps circuits. Each 64 Kbps circuit can be multiplexed to produce voice grade circuits capable of completing switched international calls. Digitization capabilities enable carriers to derive a range of voice grade circuits from a 64 Kbps half-channel. Typically, U.S. facilities-based carriers derive about four voice grade circuits from a 64 Kbps half-channel for IMTS although substantially more circuits are possible. Because the general practice among U.S. carriers is to derive four voice grade circuits from a 64 Kbps half-channel, a multiplication factor of 4:1 is used in the study. Thus, 120 equivalent voice grade circuits can be derived from a 2.048 Mbps half-channel. The second step is to estimate the rate per minute for the voice grade circuits using monthly minutes transmitted over a circuit. Monthly minutes transmitted over international circuits vary from country to country, from carrier to carrier, and from month to month. Recent operating experience of U.S. facilities-based carriers suggests that about 8,000 minutes of voice traffic per circuit per month represents a reliable and reasonable usage level for the countries included in the study. This figure represents a usage level of less than twenty percent and, therefore, may be a bit conservative. It suggests that significantly higher levels of usage can be transmitted over international circuits. Nevertheless, a usage level of 8,000 minutes per month is used for all countries in the study. Two examples are presented to demonstrate how this information is used to estimate the TCP for the international transmission facility component. Example 1. France. International private line service offered by France Telecom provides an example involving an E1 circuit (2.048 Mbps). France Telecom's monthly tariff rate for an E1 circuit with a lease period of five years for service to the United States is 167,300 French Francs (FF). France Telecom offers a 15 percent discount to customers with a monthly billing of more than 300,000 FF. Because U.S. facilities-based carriers generate monthly bills that exceed 300,000 FF, they are entitled to the discount. Allowing for the 15 percent discount in France Telecom's tariff rate reduces the monthly charge to 142,205 FF per circuit, or approximately $27,574 at the current exchange rate between the franc and the U.S. dollar. With thirty 64 Kbps circuits to an E1 half-channel, a multiplication factor of 4:1, and a usage level of 8,000 minutes per circuit per month, 960,000 minutes are transmitted over an E1 half-channel in an average month. Thus, France Telecom's monthly tariff rate, after the discount, is equal to an average charge of 2.9› per minute for the international transmission facility component for service from the United States to France. Example 2. Uruguay. Uruguay provides an example of the procedure used in the study to estimate the international transmission facility component for flat rate private line service transmitted over a circuit with lower bandwidth than an E1 half-channel (i.e., there is no reduction from the tariff rate for a multi-year lease and no discount for large volume customers). Service is offered with a 128 Kbps half-channel. ANTEL's tariff rate for this service, at the current exchange rate of $1=New Pesos 7.53, is $8,131 per month. There are no adjustments in this rate. With two 64 Kbps circuits to a half-channel, a multiplication factor of 4:1, and a usage level of 8,000 minutes per circuit per month, 64,000 minutes are transmitted over a half-channel in an average month. Thus, ANTEL's monthly tariff rate is equivalent to an average charge of 12.7› per minute for the international transmission facility component for service between the United States and Uruguay. Table 2 summarizes the TCP for international transmission facilities, which are also included in Appendix B. Table 2. International Transmission Tariffed Component Prices (per minute) Country TCP Country TCP Argentina 6.7› Jordan 15.9› Australia 4.8› Kenya 25.5› Austria 8.1› Korea 5.1› Bahamas 5.2› Kuwait 7.1› Barbados 8.6› Malaysia 6.6› Belgium 3.0› Mexico 0.9› Bermuda 4.5› Netherlands 2.6› Brazil 6.6› New Zealand 5.7› Chile 2.9› Nicaragua 3.8› Colombia 5.1› Norway 3.2› Costa Rica 3.3› P.R. of China 8.7› Czech Republic 8.1› Pakistan 14.7› Denmark 5.9› Panama 4.7› Dominican Rep. 3.6› Peru 5.8› Ecuador 2.9› Philippines 6.5› Egypt 10.4› Poland 4.7› El Salvador 5.9› Portugal 4.6› France 2.9› Russia 5.4› Germany 4.3› Singapore 5.0› Greece 5.2› South Africa 5.2› Guatemala 3.1› Spain 4.8› Guyana 6.6› Sweden 3.6› Haiti 8.6› Switzerland 4.4› Honduras 3.1› Taiwan 5.7› Hong Kong 5.1› Thailand 4.0› Hungary 6.1› Trinidad 3.6› India 8.1› Turkey 5.4› Indonesia 6.8› U.A.E. 3.3› Ireland 2.7› United Kingdom 2.4› Israel 4.2› Uruguay 12.7› Italy 4.8› Venezuela 3.7› Jamaica 2.9› Vietnam 9.3› Japan 6.5› B. International Switching Facility Tariffed Component Prices Most telephone administrations included in this study publish their tariff rates for international private line service and domestic local and toll service but there is little information in the public domain concerning the international switching facility component. Carriers in Sweden and the United Kingdom have termination tariffs which could serve as a reference point for international switching costs incurred by a correspondent but these arrangements may not be representative of other countries in the study. In other cases, developing a reasonable estimate is a complex procedure because a correspondent's switch is often used for domestic service, both local and long distance calls, and for international service, both originating and terminating calls. Thus, even if relevant information is available, potentially complex cost allocation and relative usage problems would need to be addressed in order to develop separate estimates for the international switching facility component. Fortunately, the ITU-T has published information used by TEUREM member countries for telephone settlements among them. These countries base settlements on accounting rate shares for each of the three network elements in this study: international transmission, international exchange, and national extension. The accounting rate shares for each network element, which are denominated in SDR, vary with the proportion of plant capacity composed of digital equipment relative to its total plant capacity. The accounting rate share declines as the digitization capability rises to reflect the greater efficiency of digital equipment. The digitization categories are: (1) 0-30%, (2) 31-60%, and (3) 61-100%. To determine settlements for the international exchange component, TEUREM countries use an accounting rate share of 0.0324 SDR (about 4.8›) for the first category, 0.0228 SDR (about 3.4›) for the second category, and 0.129 SDR (about 1.9›) for the third category. The accounting rate share figures are calculated from data filed by the member countries. The TEUREM results are based on analyses of operating results conducted by a range of member telephone administrations that provide service in economically developed and developing countries. Some TEUREM members have relatively low levels of digitization and others have higher levels. The results presented in D.300, however, do not list the countries that fall into each digitization category. This prevents identifying, for example, any correlation that may exist between the degree of digitization and the level of economic development. We believe, however, that telephone administrations providing service in developing countries are generally more likely to have communications networks that are less technologically advanced and, therefore, have lower levels of digital equipment than those in developed countries. For purposes of our study, therefore, TEUREM's highest accounting rate share figure for the international exchange component, 0.0324 SDR, is used to estimate the TCPs for the least developed countries in the study. TCPs for the most developed countries use the lowest figure, 0.0129 SDR, and TCPs for other countries in the study use the middle figure, 0.0228 SDR. Classification of the countries in this study according to their level of economic development is based on a World Bank scheme, which is also used by the ITU. This classification scheme has four categories:  low income, GNP per capita of $726 or less,  lower middle income, GNP per capita between $726 and $2,895,  upper middle income, GNP per capita between $2,896 and 8,955; and  high income, GNP per capita greater than $8,955. Table 3 lists the countries in the study by their level of economic development. Table 3. Economic Development Classification Low Lower Middle Upper Middle High Egypt Colombia Argentina Australia Guyana Costa Rica Barbados Austria Haiti Dominican Rep. Brazil Bahamas Honduras Ecuador Chile Belgium India El Salvador Czech Republic Bermuda Kenya Guatemala Greece Denmark Nicaragua Indonesia Hungary France P.R. China Jamaica Korea, Rep. Germany Pakistan Jordan Malaysia Hong Kong Vietnam Panama Mexico Ireland Peru South Africa Israel Philippines Trinidad & Tobago Italy Poland Uruguay Japan Russian Fed. Kuwait Thailand Netherlands Turkey New Zealand Venezuela Norway Portugal Singapore Spain Sweden Switzerland Taiwan U.A.E. U.K. Table 4 presents the international switched facilities tariffed price components, which are also included in Appendix B. Table 4. International Switched Facilities Tariffed Component Prices (per minute) Country TCP Country TCP Argentina 3.4› Jordan 4.8› Australia 1.9› Kenya 4.8› Austria 1.9› Korea 3.4› Bahamas 1.9› Kuwait 1.9› Barbados 3.4› Malaysia 3.4› Belgium 1.9› Mexico 3.4› Bermuda 1.9› Netherlands 1.9› Brazil 3.4› New Zealand 1.9› Chile 3.4› Nicaragua 4.8› Colombia 4.8› Norway 1.9› Costa Rica 4.8› P.R. of China 4.8› Czech Republic 3.4› Pakistan 4.8› Denmark 1.9› Panama 4.8› Dominican Rep. 4.8› Peru 4.8› Ecuador 4.8› Philippines 4.8› Egypt 4.8› Poland 4.8› El Salvador 4.8› Portugal 1.9› France 1.9› Russian Fed. 4.8› Germany 1.9› Singapore 1.9› Greece 3.4› South Africa 3.4› Guatemala 4.8› Spain 1.9› Guyana 4.8› Sweden 1.9› Haiti 4.8› Switzerland 1.9› Honduras 4.8› Taiwan 1.9› Hong Kong 1.9› Thailand 4.8› Hungary 3.4› Trinidad 3.4› India 4.8› Turkey 4.8› Indonesia 4.8› U.A.E. 1.9› Ireland 1.9› United Kingdom 1.9› Israel 1.9› Uruguay 3.4› Italy 1.9› Venezuela 4.8› Jamaica 4.8› Vietnam 4.8› Japan 1.9› C. National Extension Tariffed Component Prices The final element is the TCP for the national extension component. The estimation procedure for this component requires several steps. Tariffs and rate structures for local and toll long distance service vary widely among the countries included in the study. As discussed above, the tariff rates vary by the time of day and day of the week, call destination, call distance, and local versus toll calls. Some telephone administrations do not charge their customers for local or domestic calls. For those telephone administrations that charge for this service, the distribution of international calls from the United States has an important impact on the national extension TCP. The first step is to determine the distribution of international calls from the United States within each country included in the study. The distribution categories vary with the tariff rate classifications for local and toll service within each country. The categories could include time of day, day of the week, mileage, etc., depending upon the telephone administration's tariff schedule. Thus, international calls from the United States are distributed among service classifications, time periods, and the destination of the calls. The distribution of minutes for each country was determined from information collected on customers' calls during a three month period that began on January 6, 1996. Call distribution information was collected for the same one-hour period of the week for the sampling period. In many countries, most calls from the United States terminate in major metropolitan areas. In many cases, more than 70 percent of the calls fall into this category. Tariff rates for service in metropolitan areas are generally significantly lower than those for calls to remote areas. In Argentina, for example, calls within Buenos Aires have a rate during the normal period of 2.1› per minute as compared to $1.418 per minute for a call transmitted beyond 600 kilometers. The second step in developing an estimate for the national extension TCP is to determine the distance from the foreign international exchange switch through which the calls pass en route to their final destination. This determination is necessary because domestic rates in many countries included in the sample vary with distance. The third step of the estimation process is to select the appropriate tariff rate to use for the minutes in each distribution category. The lowest rate offered by a telephone administration for each of its different tariff elements is used in the study. In those cases where a telephone administration offers a discount available for large volumes of domestic service, the discount is used to calculate the estimates. In Japan, for example, NTT offers volume discounts ranging from 5 to 25 percent, which are reflected in the estimates of the national extension component for Japan. A complete list of the national tariffs used in the study is presented in Appendix D. The final step is to develop a country's national extension TCP. These figures are derived from the information about the distribution of international calls from the United States among tariff rate categories, the destination points within a country, and the distance from a country's international gateway switch or switches. The result is a figure that is weighted by each country's distribution of minutes from the United States among service and tariff rate categories, the particular domestic telephone service tariff schedule, and the distribution of calls throughout the country. Estimates for the national extension TCPs are summarized in Table 5 and included in Appendix B. Example 1. Argentina. Argentina presents an uncomplicated example of the process used to estimate a national extension component. Call distribution information for service from the United States to this country shows that 80 percent of U.S. service terminates in the Buenos Aires area. The local domestic charge for minutes in this area is 2.1› per minute. There are ten mileage rate bands in Argentina but only a small amount of international traffic from the United States terminates in each rate band. In order to simplify the estimation process without a significant loss in accuracy, the minutes terminating outside the area of Buenos Aires are combined into two categories and the highest rate for each category is used. Thus, all minutes that terminated in mileage rate bands 2 through 6 are combined and the tariff rates for rate band 6 (as shown in Appendix D) are used for these minutes. Similarly, all minutes terminating in mileage rate bands 7 through 10 are combined and the tariff rates for rate band 10 (as shown in Appendix D) are used in the calculation. The result is an estimated national extension TCP of 22› per minute for Argentina. Example 2. India. India has a complicated tariff rate schedule for service within the country and international service from the United States is more widely distributed throughout the country than is the case with Argentina. These two features of service with India add to the difficulty encountered in estimating the national extension TCP. In addition, there are four international gateway switches that serve the entire country. This last factor means that, in order to estimate India's national extension TCP, it is necessary to locate each city calling code in relation to the nearest gateway switch. The seven mileage rate bands for domestic service in India are plotted around each international gateway switch and the appropriate city calling code is assigned to the proper rate band based on the distance from the nearest gateway switch. The percentage of traffic in each rate band is determined by combining the appropriate city code and international gateway switch. International traffic from the United States is grouped by the seven mileage rate bands with time-of-day weighted prices. The results range from 2› per minute to 78.9› per minute. Finally, the weighted rates for each mileage rate band are weighted by the percentage of U.S. traffic terminating in the rate band. The result is an estimated national extension TCP for India of 18.3›. Table 5. National Extension Tariffed Component Prices (per minute) Country TCP Country TCP Argentina 22.0› Jordan 2.3› Australia 12.0› Kenya 12.3› Austria 21.4› Korea 4.3› Bahamas 12.8› Kuwait Zero Barbados Zero Malaysia 12.4› Belgium 9.2› Mexico 12.5› Bermuda 3.5› Netherlands 5.3› Brazil 17.8› New Zealand 16.2› Chile 12.3› Nicaragua 3.7› Colombia 8.6› Norway 6.5› Costa Rica 2.2› P.R. of China 4.2› Czech Republic 7.5› Pakistan 7.2› Denmark 6.6› Panama 9.9› Dominican Rep. 6.1› Peru 5.5› Ecuador 2.6› Philippines 12.6› Egypt 2.0› Poland 15.1› El Salvador 1.1› Portugal 17.4› France 12.7› Russia 25.2› Germany 13.6› Singapore $0.7› Greece 14.4› South Africa 8.3› Guatemala 2.4› Spain 11.4› Guyana 0.6› Sweden 4.5› Haiti 17.0› Switzerland 14.3› Honduras 8.7› Taiwan 6.3› Hong Kong Zero Thailand 8.3› Hungary 4.9› Trinidad 7.6› India 18.3› Turkey 7.7› Indonesia 23.9› U.A.E. 2.5› Ireland 13.4› United Kingdom 8.7› Israel 2.4› Uruguay 6.2› Italy 11.5› Venezuela 15.3› Jamaica 1.0› Vietnam 10.6› Japan 11.3› V. Summary and Conclusions The TCPs are substantially below the Commission's benchmark ranges adopted in 1992, and also are generally significantly lower than the settlement rates between U.S. carriers and foreign telephone administrations. The settlement rates in effect with some administrations, however, are below the TCPs for those administrations, which suggests that the TCPs exceed current costs of handling international calls from the United States. We recognize that using the TCPs as the basis for establishing benchmark settlement rates would not result in rates that are cost-based. Nonetheless, we believe that the TCPs can be used to calculate benchmarks in a manner that treats foreign carriers fairly and places some limit on the subsidies in current settlement rates. APPENDICES TO REPORT ARE EXCEL FILES CONTAINED IN ZIP FILE APPENDIX A - br484a.xls APPENDIX B - br484b.xls APPENDIX C - br484c.xls APPENDIX D - br484d.xls