INTERNATIONAL TELECOMMUNICATION UNION Rome, 25 March 1996 How will the accounting rate system need to be modified in a liberalised market? Liberalisation & Privatisation of the European Telecommunications Sector Preparing for 1998 & Beyond Dr Pekka Tarjanne Secretary-General, International Telecommunication Union (ITU) An International Conference arranged by IBC UK Conferences Ltd. Ladies and Gentlemen, It is a pleasure to join you in the historic city of Rome to debate the liberalisation and privatisation of the European Telecommunications Sector. At first sight, the reform of the accounting rate system might appear to have little to do with the latest trends in the Sector that we are discussing here today. After all, the system of revenue sharing is a venerable tradition which has served the industry for many years. Indeed, one of the main reasons that 20 European countries came together in 1865 to form what eventually became the International Telecommunication Union was that they needed to develop a means of dividing the revenues from international services between origin, destination and transit countries. The methodology that they developed is still with us today, albeit in a format which has been progressively modified. Today, revenues from international telephone services form a market in excess of US$50 billion. It is the most profitable and the fastest growing segment of the telecommunications market and is the one which is generally first to be targeted by new market entrants. But some would argue that the accounting rate system itself is a barrier to entry because of the necessity for new players to negotiate bilateral correspondent agreements with 200 or more countries. Is it a relic of a bygone era or can it be adapted to a competitive marketplace? The accounting rate system was developed under a paradigm of international telephony which might be characterised as “nation shall speak unto nation”. It worked well providing certain conditions held true: • international services were jointly-provided by monopoly partners; • the price charged for a call was approximately equal in different directions (the principle of symmetry); • incoming and outgoing traffic was approximately in balance for each main bilateral relationship between countries; • collection charges, even for off-peak discounts and volume discounts, were never lower than accounting rates; • inflation rates and exchange rates were relatively constant between countries. However, in the modern era these conditions are breaking down. Consumers are now confronted with a confusing range of options. Technological change has permitted a significant reduction in the cost of constructing the infrastructure which carries international telephone traffic, particularly over submarine fibre optic cables, or via satellite. But the pace of change has been uneven. In particular, until recently the advantages of network modernisation were not being reflected in accounting rate cuts. These trends, exacerbated by exchange rate fluctuations, have generated significant differences in the level of international telephone call charges between countries. Consequently, imbalances in the traffic flow between countries began to grow. In order to interpret these changes, it is useful to consider how the paradigm of international telephony has changed. The old paradigm of jointly-provided services has been progressively eroded and some countries now view telecommunications as a traded service in which individual companies compete to provide call origination and call termination services. Furthermore, the public switched telephone network now provides just one option of many for the delivery of service. And the service being traded is no longer just plain vanilla telephony, but is increasingly also fax, data, enhanced voice, video, graphics or “point and click” commands sent to a remote World Wide Web server. Is it possible to recast the bilateral correspondent relation system in a multilateral, multimedia, multi player environment? In order to answer that question, it is useful to take a closer look at how the accounting rate system actually works. It is based on a dual price arrangement whereby, for each call, one price is charged to users by the originating Public Telecommunication Operator or PTO (this is the collection charge, usually set in local currency units) and a second price is agreed by the terminating PTO and the originating PTO (the accounting rate, usually set in international currency units such as US dollars, gold francs or SDRs). This is used to determine the fees paid by the originating PTO to the terminating PTO (the settlement rate, usually half the accounting rate). If there is an imbalance in the volume of incoming and outgoing traffic, then the originating PTO which generates more traffic pays for the difference to compensate the terminating PTO (the net settlement payment). An alternative way of looking at the relationship is to think of the price agreed by the PTOs as the wholesale price for bulk capacity, while the price charged to the user is the retail price. The scriptures teach us that it is more blessed to give than to receive, but under the accounting rate system it is actually more blessed to receive. That is because those countries that have a surplus of incoming traffic -- countries that listen more than they talk -- will receive net settlement payments from their partner countries. Those countries which are “efficient”, in generating traffic or reducing their call charges, are penalised under the accounting rate system while those countries which maintain inefficient networks, keep prices high, or deliberately block outgoing calls, are rewarded. If market distortion were the only fault with the accounting rate system, it could probably survive. After all, economists usually agree on only one thing, namely that no market is ever perfect. The difficulty is that there are a growing number of other pressures for reform. An increasing share of traffic bypasses the accounting rate system completely because it is carried by just one operator instead of two (end-to-end service), because it travels over private networks, or because it travels over the Internet. Increasingly, owners of infrastructure wish to provide service directly to end-users instead of relying on correspondent partners. Furthermore, at the local level, callback operators and resellers exploit the fact that tariffs are not cost-based by arbitraging different prices between countries. None of this represents a crisis. Indeed, anything which brings lower prices to end-users is very much to be welcomed. The international network is not collapsing, indeed it is growing faster than ever. And PTOs seem to be as profitable as ever. So what is the problem? The problem is the growing imbalance of calls on the international network., The situation is summed up in this slide which shows US outgoing and incoming international traffic between 1975 and 1994. As telecommunications has become a traded commodity, those nations that have a comparative advantage in producing it -- perhaps because they can produce it at a cheaper rate than other countries -- end up exporting more of it than they import. In the case of the United States, in 1994 some 13 billion minutes of telephony were exported while less than 6 billion minutes were imported. Several other countries, including Sweden, Australia and Japan, are thought also to have growing deficits on settlement payments. That surplus of seven billion minutes translates into a US settlement payment deficit of US$4.3 billion. This is not all bad for US carriers because the collection charges they collect at home are now much higher than before. Over the last decade they have increased their share of global traffic from 21 to 25 per cent. Indeed, it could be said that US companies are largely responsible for engineering this deficit because of their pioneering use of calling cards, country direct services and callback services. In each of these cases, the traffic looks as if it is originating in the United States whereas it is actually originating in the partner country. But for the US government, this deficit of US$4.3 billion seems bad. It exceeds, for example, the US deficit for trade in telecommunication equipment. For this reason the US government is using all available fora -- the ITU, the OECD the WTO -- to push for a reform of the system. The FCC has recently ordered all US carriers to suspend settlement payments to Argentina’s long-distance and international telephone company, Telintar, which it regards as using its market power to discriminate against individual US carriers, a process known as “whipsawing” Traffic imbalances are not the only reason for reform of the accounting rate system but they are a symptom of a much bigger malaise which is the fact that accounting rates are not cost- based. Many of the distortions we currently see in the international telephone network -- such as refile, hubbing, bypass arrangements, or call turnaround -- would be reduced if accounting rates were closer to costs. There are many reasons why the United States has become such a significant exporter of telephone calls. Some of these reasons are: • Historical — US citizens are used to unmetered local calls and therefore tend to make longer calls even when they are metered. The average US outgoing call is 20 per cent longer than the average US incoming call; • Cultural — the US is an immigrant society and its citizens call home more often than they are called. Social telephone calls last longer on average than business calls; • Economic — until recently, US outgoing calls were cheaper than those of most other countries. Ironically, as this graph shows, the United States can no longer be regarded as offering cheap international telephone calls. Since the late 1980s the official rates of international calls from the United States (measured in purchasing power parities) have risen relative to the cost of calling to the United States from other countries. It must be noted, however, that an increasing number of US subscribers, especially businesses, benefit from some type of discount on the published tariff. The average price per minute paid for US outgoing traffic is considerably below the published tariffs of US carriers. What principles should we adopt in considering the reform of the accounting rate system? The first principle must be the continued viability of the international telephone system which allows virtually any one of the 700 million or so telephone users around the world to talk to any other. As I said earlier, the accounting rate system has served the industry well and we tamper with it at our peril. If we follow the “logic of liberalisation” that has guided Sector reform in many countries and which will reshape European telecommunications in the run-up to 1998, any reform of accounting rates should attempt to bring international telecommunications into a trade liberalisation framework. At the end of next month, the talks currently taking place in Geneva in the context of the Negotiating Group on Basic Telecommunications (NGBT) are due to reach their conclusion. Any new system would need to be transparent, non- discriminatory, and cost-based to meet the World Trade Organisation’s requirements. A further necessity is to ensure that the benefits of reform are not spirited away to enrich the bank balances of the PTOs but are passed on to end-users. Any reform process must recognise and endorse the development of a competitive marketplace make it easier for new market entrants to become established. Finally, any new system must provide for an adequate transition period, in particular for those developing countries which are heavily dependent on the current accounting rate system. It is worth considering the requirements of developing countries in more detail if only because they are the main beneficiaries of the unintentional largesse of the US operators. In the group of countries in Latin America and the Caribbean shown in this chart, settlement payments constitute a significant share of total telecommunications revenue, in some cases more than 50 per cent. Furthermore, because settlement payments are paid in hard currency, they are particularly welcome in those developing countries which have a soft currency or which are plagued by inflation and exchange rate instability. The business plans for network development of many PTOs in developing countries are predicated upon settlement payments. The majority of settlement payments are recycled into orders for telecommunication equipment imports. If settlement payments are reduced, it will have a negative effect on the health of the telecommunication equipment manufacturers as well as on those finance houses which have lent money to PTOs in developing countries, on network development and on consumers. But such dependence on settlement payments is an unwise strategy. Experience shows that traffic stimulation and creating an attractive investment climate are more effective strategies for telecommunications development. By keeping charges high, developing country PTOs create incentives for callback and other forms of bypass which erode their competitive position. Furthermore, a new threat is emerging in the form of Internet telephony. The Internet famously does not employ the usage-based tariffing schemes on which the financial structures of PTOs are based, but instead employs flat-rate tariffs. Furthermore, the Internet has developed without any revenue-sharing mechanism between operators. In so far as there are payments from end-users, they are retained by service providers on a “sender keeps all” basis. Internet telephony is based on packet switched rather than circuit switched networks. It would probably cost more to trace and bill the precise route taken by each data packet across the network than it would to send the call in the first place. The current state of the art in Internet telephony is quite primitive, attractive mainly to hobbyists and enthusiasts. But one can envisage a rapid evolution over the coming months. Already callback operators are offering to terminate calls originating from computers. Soon, those callback operators and resellers will use the Internet itself as a backbone for their calls. If we lived in a rational world, few consumers would choose to have their conversations garbled by computers. But the prevailing price structures in international telephony are not rational. The ultimate commodity being sold is bandwidth. Voice traffic uses tiny amounts of bandwidth but is charged a high price. Data traffic uses huge amounts of bandwidth but is charged a low price. Consequently, “cross-over” technologies, such as voice over data networks, exploit these economically irrational tariff structures. What is the ITU doing to reform the accounting rate system? Within the ITU’s Telecommunication Standardization Sector, Study Group 3 is tasked with creating a framework within which the international carriers can establish revenue-sharing mechanisms. I should stress that ITU has no role in setting or negotiating accounting rates which are agreed bilaterally between PTOs. However, ITU members have agreed to a Recommendation (D.140) which sets a timetable of five years for bringing accounting rates closer to costs. The Recommendation also defines those costs which can legitimately be attributed to international service, so as to distinguish between costs and subsidies. The good news is that accounting rates are coming down. A survey of average accounting rates to which some 37 countries or regions responded showed an accelerating downward trend. Since 1988, the average accounting rate has fallen from 2.12 SDR to 1.73 SDR, a fall of almost 20 per cent. Among economies with competitive markets, such as those of the OECD, not only are accounting rates lower but they are also falling faster. Indeed, since 1990 accounting rates have been falling much faster than call charges, suggesting that carriers have not been passing on the benefits of technological change to users. By publishing aggregate figures such as these for accounting movements, the ITU is helping to promote transparency. Nevertheless, despite this progress, the search is on for an alternative system. Fortunately options abound. Many new networks, such as the Internet, are based on “sender keeps all”. If accounting rates continue to fall, then that may be the logical direction for the public telephone network to follow. Within the European Union, it has been proposed that accounting rates will be superseded by facilities-based interconnection agreements after 1998. Many mobile operators already use a version of this methodology for instance in Scandinavia. But maybe for future direction, perhaps we should look to one of the oldest networks: the public telegram service which uses call termination fees. Call termination fees offer a methodology which fits well with the World Trade Organisation’s trade liberalisation framework. They fulfil many of the principles defined earlier for accounting rate reform. The idea would be that each country, or operator, would define one standard charge for terminating calls, irrespective of where those calls come from. The call termination charge would be comparable to the national interconnection charge levied, for instance, on mobile operators interconnecting with the public telephone network. The system would be transparent, flexible, non-discriminatory and (hopefully) cost-based. The latter will probably depend on the degree of market competition which is allowed in each national market. Call termination fees have received the blessing of the OECD and are currently under discussion in the WTO and the ITU. They are opposed by some carriers who are unwilling, or unable, to disclose their cost structures for terminating calls. They are also opposed by those carriers who feel that they should not be required to pay more for having their calls terminated in foreign countries than they themselves charge for terminating calls. But the fact is that providing telephony service is more expensive in some countries than others. The accounting rate system, which is based on a 50/50 revenue sharing agreement, implicitly assumes that costs are equal in all countries. This is patently not the case. The possibility exists for countries to agree on a split which departs from 50/50 but in practice this is never adopted except in “sender keeps all” arrangements. A system, such as call termination fees, which does not pretend that costs are equal in all countries would be much to the advantage of developing countries. The major alternative to accounting rates and call termination fees is a system of facilities-based interconnection agreements. It is likely that such a system will replace accounting rates in Europe after 1998. Such interconnection agreements would be regarded as commercially confidential and would be based on competition for traffic. However, interconnection agreements meet few of the criteria defined earlier. In particular they would reinforce the position of dominant carriers who would be able to negotiate from a position of strength. Where national carriers create consortia amongst themselves, such as the GlobalOne alliance or Unisource, they would be able to negotiate favourable terms for their partners which are not available to competitors. Such agreements would be fully discriminatory, non transparent, and asymmetric. Developing countries can expect to be squeezed under such a regime in that they will not be negotiating on equal terms. In conclusion, I have argued in this presentation that the accounting rate system is urgently in need of reform because it does not fit with the “logic of liberalisation”. The paradigm which prevails in international traffic has shifted decisively in favour of competition between carriers rather than co-operation. If the system is not reformed, it won’t necessarily collapse, it will simply be bypassed as more and more traffic travels over networks which do not use accounting rates. Insofar as accounting rates help to support price distortion or act as a barrier to competitive market entry, then they should not have our support if we are committed to market liberalisation. While it must be recognised that settlement payments have provided a significant support for developing countries in the past, they will not be able to do so in the future, for the reasons I have given. We must find another way. In this paper I have presented three alternatives: • the progressive reduction of accounting rates to the point at which they are so low that countries are willing to move towards “sender keeps all”; • the transition towards facilities-based interconnection agreements; • the transition towards a system of call termination charges. Whichever alternative is adopted, and they are not mutually exclusive, it is unlikely to provide the same level of support for developing countries that accounting rates currently provide. Consequently, parallel measures will almost certainly be needed to attract investment, to generate traffic, to reduce waiting lists and to sustain network development. In my paper, I outlined six criteria for reforming the accounting rate system. The call termination fees approach appears to come closest to matching those criteria, but currently no option is favoured by all carriers. Over the coming months, I can assure you that the ITU will give its highest priority to accounting rate reform as we approach the World Telecommunication Standardization Conference in Geneva in October. The most likely short-term scenario is probably that the world will divide into three groups: • relations between liberalised markets, which will move towards alternative systems such as those described in this paper; • relations between liberalised and monopolistic markets that will retain accounting rates in the short-term but will come under increasing pressure for reform. Traffic passing through the accounting rate system will account for a declining share of overall traffic as operators in the more liberal market seek alternative bypass arrangements; • relations between an ever shrinking number of monopolistic markets where the accounting rate system may survive for a little longer. The ITU remains committed to finding an inclusive, multilateral solution in which all countries move forward together. Only then will the benefits of competition and technological change be extended to all the world’s inhabitants. *******