Recently we’ve been working on expanding the capability of one of our mobile network models – the iterated cost (or bootstrap) model. In a nutshell, this model treats buyers and sellers of mobile data as rational economic agents. When the cost of supplying data falls, consumers buy more of it. Similarly, operators invest in additional network capacity if the expected revenues exceed the costs. The model finds an equilibrium where data supply and data demand are in balance. A full discussion of the model, and these findings, can be found here.
We have used this model to analyse the benefits of releasing additional spectrum for mobile (more spectrum expands network capacity, lowering costs and boosting data consumption) and to model the impact of a data ‘tax’ on traffic and spectrum value (the tax raises the price of data, suppressing data demand). We’ve also used it to explore the impact of various network parameters on spectrum value, and to generate data traffic forecasts.
However, the model can also be used to assess other policy options. In this analysis piece we use the iterated cost model to briefly explore the impact of mobile site costs, and the impact of competition, on the mobile data market.
A reduction in site costs has been identified by the GSMA as a step towards achieving European leadership in mobile. The model suggests that a policy that leads to a 2% reduction in site costs per annum will lead to 38% higher data consumption by 2030, compared to constant site costs. Spectrum value also grows significantly (in stark contrast to conventional avoided cost models, which suggest that spectrum value declines as site costs fall). The iterated cost model implies there could be substantial socioeconomic benefits from policies to reduce site costs.
The model can also be adapted to reflect competition dynamics in the mobile market, by using an economic model of oligopolistic competition. These assumptions bring in a mark-up of price over the marginal cost of data (under the ‘base case’ assumptions, firms are assumed to price at the marginal cost). The size of this mark-up depends on the number of operators: it will be higher in a two-player market than in a four-player market.
This adaptation allows us to explore the implications of a merger on data price, data traffic and spectrum value. In the model, a merger increases the mark-up of price over cost, raising prices for end-users and reducing data consumption. However, it also allows operators to consolidate spectrum holdings, increasing the theoretical capacity of the average mobile site. This lowers costs and prices, increasing demand for data. Whether a merger results in a net increase in data consumption depends on the strength of these two competing effects.
As an example, we examined two hypothetical mergers: from four to three operators, and from three to two operators. The model suggests that, with the given set of assumptions, the former would lead to an expansion of the data market and a net benefit to society. Conversely, the three-to-two merger would lead to higher data prices and reduced data consumption, and is likely to be to the detriment of society.
Of course, it should be noted that these are not general results; rather, they reflect the set of input assumptions about our hypothetical mobile market. Nevertheless, they demonstrate that the model provides a fresh way of thinking about a variety of policy issues.
 There is some debate about whether and to what extent mergers necessarily result in higher prices for consumers (for example, see Valletti et al 2015). Operators are also likely to experience some competitive pressure from Wi-Fi offloading.