Econometric analysis is a statistical method which provides an objective way of controlling for market and economic factors. Compared to benchmarking, there is less reliance on often imperfect comparisons between different countries. Instead, this approach uses the estimated relationships between observed spectrum prices and the explanatory factors (such as auction, demographic and economic indicators) to predict a values for the country under study.
The goal of the econometric analysis is to estimate the value of spectrum (dependent variable) based on the effects of various drivers of spectrum value (independent variables) using data from a large number of auctions. The potential drivers of value which are typically considered in econometric analyses of spectrum value is shown in the list below.
- Spectrum characteristics: frequency band or propagation, total bandwidth offered, time of auction, level of harmonisation
- Licence characteristics: licence duration, national or regional licences, coverage requirements
- Auction characteristics: ratio of bidders to winners, reserve prices, auction format, spectrum caps
- Economic and market characteristics: GDP per capita, population density, mobile penetration, average revenue per user, fixed line penetration, number of mobile operators in market
While econometric analysis can be a more robust statistical method than the direct and adjusted benchmarking approaches, its reliability is affected by data availability and data quality.