Telecommunications Policy

This paper examines differences in fixed location broadband adoption rates among households of various demographic and socio-economic characteristics and in different geographic locations utilizing the Federal Communications Commission's census tract level adoption data, demographic data from American Community Survey and the census block level broadband availability data from National Telecommunications and Information Administration. Ordered probit models are estimated and used to conduct simulations in order to analyze the determinants of the broadband adoption rate.

High speed broadband creates potential productivity gains and has a positive impact on economic growth. Achieving Europe's broadband access objectives will require large scale investment in next generation broadband networks, and it is imperative that an appropriate investment climate is created to encourage fibre network rollout. This study considers whether and how competition in the DSL market affects the incentives of operators to invest in the deployment of high-end fibre optic networks.

Recent studies have confirmed that broadband adoption (as opposed to simply having access to broadband infrastructure) is positively linked with economic growth. In light of this, federal policy efforts have switched from focusing mainly on the provision of infrastructure to more explicit adoption-oriented efforts. One of those efforts was the Federal Communications Commission's Low-income Broadband Lifeline Pilot Projects, which ran from 2013 to 2014. The program worked with 14 private telecommunications firms to subsidize household broadband adoption for low-income households by providing discounted monthly and equipment costs. Low-income households are an important component of the broadband adoption puzzle: between 2003 and 2013, the adoption gap between low-income and high-income households actually increased by 5% points. This paper focuses on two specific FCC Broadband Lifeline Pilot projects that allowed consumers to make choices among different options, such as data allowance, speed, and wireless vs. wired connections. Conditional logit models are used to develop estimates of consumer's willingness-to-pay for specific broadband attributes. The results indicate that low-income consumers have a preference for smartphone connections (versus aircards) – and that this effect is even more pronounced for those households earning less than $20,000; that low-income consumers have a preference for wired connections (versus wireless); and that there is evidence that low-income consumers are willing to pay for an extra GB of data each month – but not for the speed of their connection.

Spectrum fragmentation, a concept that describes the degree of split for a given portion of spectrum, is a newly emerging issue which can affect productivity and technical efficiency in spectrum use as the telecommunication service paradigm evolves from voice oriented to data oriented services.

This paper proposes ten indices, compares them and calculates their values for Korea and Japan. The values of all indices range between zero and one, so one can use them for international comparison of spectrum fragmentation. Out of ten indices, three are Theil indices, comparing the discrepancies between spectrum and subscriber distributions. In addition, this paper discusses policy alternatives to lessen the spectrum fragmentation problem.

There has been an extensive debate about the role of broadband access regulation on market outcomes. This paper estimates the impact that the different modes of competition have had on broadband take-up to date, using a data set for 27 European Union countries. We find that unbundled local loop (ULL), which is one of the main types of access-based competition in Europe, has had a positive impact on broadband take-up. However, the impact of ULL becomes smaller as its share increases. That is, ULL entry is less effective in areas where ULL take up is already high. Further, there is evidence of a crowding out effect between ULL and inter-platform competition. This means that ULL is less effective in enhancing broadband penetration in the areas where alternative networks already have a significant share of broadband lines.

[Commentary] The first spectrum auctions generally assigned the chosen number of licences of predetermined size to the highest bidders, but auctions now allow a greater choice of outcomes, with bidders with existing spectrum portfolios competing, with others, for multiple lots, often in different bands. Modern auctions also contain design features expressly directed at efficiency and equity objectives. In relation to efficiency, spectrum caps or set asides for new entrants can be incorporated to combat the exercise of market power downstream. In relation to equity objectives, licence conditions may specify obligations to provide coverage in non-commercial areas, or promises of the attainment of social objectives can be given a weighting, with revenue, in the determination of winning bids. The paper provides and overview of the wide use of such tools and the consequences for the operation of the auction process. Some lessons are drawn for future spectrum auctions.

[Martin Cave is associated with the Imperial Business School, London, UK
Rob Nicholls is associated with the UNSW Business School, Sydney, Australia]