کاهش قیمت خدمات تلفن عمومی: یک روش توزیع درآمد

Liberalisation usually requires price rebalancing. In low-income countries, this often leads to residential rentals being priced are much higher than before, which can threaten the affordability of basic telephone service. A quantitative model is provided for assessing the likely effects of price changes on affordability, using local data. Representative data from a range of countries are provided. Using illustrative parameters, the model is applied to show the effects on telephone take-up of economic growth, different pricing strategies, and different degrees of inequality in income distribution. It is also used to estimate price elasticities of demand for lines in low-income environments.

All over the world, countries that are liberalising their telecoms industries face a dilemma. Price rebalancing tends to raise residential rentals, often very substantially; while making basic service affordable points to keeping rentals low. The notion that basic telephone service should be affordable has received widespread assent, but this has not generally been backed up by policy guidelines designed to achieve affordability.1 In a competitive market, prices will reflect underlying costs. Incumbent telephone companies therefore need to “rebalance” their tariffs to a pattern which will enable them to meet competition, and which at the same time will permit efficient competitive entry. Overall, rebalancing should be revenue neutral, so that customers as a whole do not lose out. Of course, liberalisation itself should eventually deliver major customer benefits. However, rebalancing inevitably favours some customer groups over others. In particular, it usually favours high (typically business) users, especially those who make international calls, over low (typically residential) users.2 Often, the level of residential rental that would be needed to cover the cost of line provision is several times higher than the existing rental. Existing customers may be unwilling or unable to pay for line rental at a cost-recovering level. This can have major political and commercial repercussions. The higher the level of national per capita income, the easier it is to deal with this dilemma. Most early liberalising countries have been relatively affluent OECD countries. In general, they have handled the problem by a combination of: • Restraints on the speed of rebalancing,3 and • Special tariff schemes to make service affordable to low-income users.4 But now we are seeing the spread of liberalisation to many countries with moderate or low levels of per capita income. For them the problem is much more acute. Hard cost data are scarce, but overall there is little reason to expect that per line costs will be much less than in richer countries. Typically, labour costs are lower, but capital costs may be higher, and of course line densities and economies of scale may be much lower.5

The conclusions of this paper are of two kinds. The first relate to the model which has been presented, and the second to the results of applying the model using assumed and illustrative data. First, regarding the model: the paper has provided a way of exploring the affordability implications of different pricing policies, using readily available computational tools and local data. This analytic tool should permit an objective formulation of affordability constraints and goals, which in turn will permit fuller and more balanced consideration of the costs and benefits of different approaches to liberalisation and price rebalancing. Ideally, cost–benefit analyses should give due weight to the interests of all sectors of society, including marginal telephone subscribers and non-subscribers.27 Second, regarding the results of applying the model using assumed and illustrative data: the aim has been to choose reasonably representative values for the various parameters. The data presented in Appendix B support the basic assumptions of the model and provide guidance on the key affordability parameter. It is therefore believed that in many (if not most) low- and middle-income countries, the following results of applying the model are likely to hold good: