Do call termination rate interventions affect developing countries (with smaller fixed line networks) differently? : testing for the "waterbed effect" for non-linear tariffs in South Africa

The Independent Communications Authority of South Africa (ICASA) has over the last six years, in line with other regulators in many jurisdictions, reduced call termination rates (the price one operator charges another to complete a call between two networks) by more than 80%. A key question is what impact this has had on retail prices. Theoretical work suggests that reducing call termination rates could result in a ‘waterbed effect', where retail prices for mobile services increase after a call termination rate reduction, since profits from fixed to mobile calls declines and this reduces the profitability of attracting customers, and operators accordingly "soften" retail price competition. On the other hand, operators might choose a high mobile to mobile termination rate in order to exclude rivals. The extent of this "waterbed effect" therefore depends on the extent to which fixed to mobile calls matter and the extent of market symmetry. South Africa has a relatively small, and declining fixed line network and significant asymmetries between mobile operators (there are two entrants and two incumbents). Measuring retail prices over time in South Africa presents us with an opportunity to test whether, in circumstances where fixed to mobile calls don't matter as much, operators are asymmetric, call termination rates do indeed facilitate entry and cause a reduction in retail prices. A key problem with retail mobile prices in South Africa is that each of the mobile operators offer a wide variety of packages. In order to overcome this challenge, a hedonic regression approach will be used to develop an index of quality adjusted prices, and estimate the impact of call termination rate reductions on quality adjusted prices. The preliminary results of this analysis show that "waterbed effects" do indeed exist in respect of postpaid prices in South Africa, and this result is robust to using quality adjusted prices or mean prices. This means that operators would choose low call termination rates in order to soften competition, absent an exclusionary strategy. The choice of high call termination rates by incumbent operators prior to Cell C's entry in 2001, and pressure on resistance to regulatory steps to reduce termination rates when Telkom Mobile entered in 2010, are therefore consistent with an exclusionary strategy on the part of MTN and Vodacom.