Dieter Elixmann
Elastizitäten im Telekommunikationsbereich - Angewandte Nachfrageanalyse in den USA und Kanada
Nr. 125 / März 1994
(long version only in German)
Summary
Decisions of regulatory authorities as well as telecommunications carriers in the U.S. and Canada increasingly are based on empirical evidence provided by demand analysis. A focal point of these econometric estimates are elasticities, in particular price elasticities. The importance of the knowledge of structural features of demand rests on the one hand on the regulatory policy and requirements in these countries. On the other hand, an increasingly competitive environment forces carriers to focus their tariff and product policy on the requirements of the market, especially to meet the needs of customers.
The present paper presents stylized facts about the methodology and reviews the empirical evidence of research from North America centering on a broad scale of issues in telecommunications demand analysis. Among them are the analysis of driving forces of the telephone penetration of private households, the estimation of toll price elasticities, an examination of the features of carrier access demand, an assessment of the effects of regulatory and competition policy in the U.S. on prices and competition, the evaluation of customers' preferences for the price and product policy of a carrier, and the derivation of properties of international telecommunications demand.
In the past decade the empirical estimation of elasticities has undergone a change of paradigm with respect to the methodology applied. In earlier times demand analysis rests primarily on time series data. Today, cross section aspects often are taken account of by pooling time series-cross-section data (e.g. panel data). Moreover, applied demand analysis concentrates more and more on discrete-choice models leading to the set of probit or logit approaches.
The price elasticity of household telephone penetration is very small. A rule of thumb value is given by -0.05 (in Canada lower than in the U.S.). The order of magnitude of the (aggregated) toll elasticity is -0.5. The intraLATA toll elasticity in the U.S. as well as the short-haul toll elasticity in Canada (for traffic between adjacent provinces) is lower, however, the interState toll elasticity in the U.S. as well as the Canadian long-haul toll elasticity (e.g. for calls from coast to coast) is higher than this value. The size of the toll price elasticity varies directly with lenght-of-haul. Apart from distance, call elasticities are sensitive to time-of-day and they reflect also to a certain extent characteristics of specific user groups.
Empirical evidence suggests the presence of a reverse calling effect, taking account of the "I call you if you call me" principle. Canada-Canada point-to-point models reveal that around 50 % of traffic in one direction is generated by traffic from the opposite direction. However, on an international scale the reverse calling effect is much lower exhibiting an order of magnitude of around 10 %.
Discussion Paper is no longer available.