Hedonic prices for fixed broadband services : estimation across OECD countries

This paper sets out a framework to estimate quality-adjusted price levels and price changes for fixed broadband services in OECD countries. We extend and adapt existing hedonic frameworks for international and interarea comparisons and consider the extended country product dummy approach. Hedonic pricing studies often are context and data dependent, and this study is no exception. We find that the multilevel structure of international broadband price datasets suggests modeling hedonic functions at the company level. This not only mitigates efficiency loss due to lack of subscriber information on individual plans but also allows for company costs and markups to influence estimates of hedonic function coefficients. Incorporating random variation in hedonic slope coefficients at the ISP level produces results that statistically dominate standard models where slope coefficients are the same across countries, and we suggest how price comparisons based on random coefficient hedonic models might be useful in telecommunications policy analysis.