The Role of Triple- and Quadruple-Play Bundles: Hedonic Price Analysis and Industry Performance in France, the United Kingdom and the United States

Abstract

Communication providers may use bundles of different services (voice, data, pay-television, mobile voice and data) to leverage market power or increase switching costs for consumers but also as an efficient way to allocate fixed costs across services, reduce the complexity of their offers or provide unified billing for all services and innovative features (e.g. home security services, online music). Communication bundles are increasingly important and raise challenges for regulators and policy makers.

Economic literature has extensively addressed bundling issues, in particular in the area of industrial organisation, price discrimination and consumer welfare analysis (Rey and Tirole, 2006; Adams and Yellen, 1976). The literature on hedonic price analysis for communication services, relatively limited but expanding, builds on previous application of hedonic prices for automobiles, personal computers or houses (OECD, 2006) and explores the relationship between prices the bundle’s quality characteristics. A recent study by the Portuguese regulator and provides some insights on how quality parameters affect triple- and quadruple-play bundle pricing (Anacom, 2013).

This paper is novel in that it tries to map firm level industry performance data (revenues, profits, etc.) and pricing behaviour by those companies for triple-play and quadruple-play bundles (fixed voice, broadband, pay-television and mobile voice and data). On one hand, it uses hedonic price analysis of triple- and quadruple-play communication bundles of the largest operators in France, United Kingdom and the United States and, on the other hand, financial indicators of these operators corresponding to the 2009-2014 period (quarterly data).

A hedonic price analysis model is specified using OLS econometric analysis of some 300 offers from 15 operators in France, the United Kingdom and the United States (including standalone, double-, triple- and quadruple-play offers), for prices in April 2014. The collected variables include service price per month, technology, download speed, contract length, data allowance, local, national, international and/or weekend calls, the number of TV channels, premium content, mobile minutes, SMS and MB. The quality of the television content has been modeled using a quality index, constructed with the number of channels and whether the TV component includes premium sports or movies content. This quality index is in turn mapped into a series of dummy variables. The nature of the bundle, e.g. 2-play, 3-play, 4-play, has also been include through the use of dummy variables.

Quarterly financial indicators from 2009 to 2014 of 15 communication/pay-television providers in France, the United Kingdom and the United States, including revenues, profits, investment, indebtedness, number of subscribers, etc., are used in an econometric model that uses bundle penetration as the dependent variable. The model, not fully specified yet, could be a simple OLS, simultaneous equations or logistic regression (logit), using bundle penetration (% of customer taking up bundles) as the dependent variable. The independent variables would be the financial indicators, market characteristics, such as the operators’ market shares, relative prices (obtained from the hedonic price analysis), competition and/or regulatory variables, or even socio-economic parameters, such as GDP per capita, which could also be included in model as control variables.

Keywords: bundles, pay-television, triple-play, quadruple-play, econometrics, hedonics