The dissertation is comprised of three papers. In the first two Chapters, I analyze the importance of competition, preference heterogeneity, and socio-economic/country-specific factors to explain the differences in penetration rates of mobile phone services across EU Member States. Chapter 1 presents a model of demand and supply for mobile phone services in which products are perceived as homogenous but consumers are heterogeneous with respect to their valuation of the services. Once a service is purchased, consumers (temporarily) leave the market. The parameters which govern the distribution of preferences are allowed to vary by country and will be estimated as part of the demand specification. The model matches the data well and is able to replicate the observed u-shape in the coefficient of variation in penetration rates over the sample period. Using the demand parameters, consumer acquisition costs are backed out and counterfactual experiments performed. I find that preference heterogeneity and differences in the cost of consumer acquisition explain most of the variation in penetration rates across countries. Competition and other control variables, such as the price of fixed-line calls, play only a minor role. In Chapter 2 I relax the assumption that firms are perceived as homogenous and model them as differentiated products. I incorporate endogenous population weights in a standard random coefficients logit model to capture changes in the demographic composition of potential buyers over time due to the (temporary) market exit of adopters. Compared to the results of Chapter 1, I find a larger role of competition and a smaller impact of the (net) cost of consumer acquisition in explaining differences in mobile phone services diffusion. In the third Chapter, I analyze the effect of a product introduction on labor supply. I demonstrate that it is possible to overcome many of the limitations associated with the lack of individual level purchase data by focusing on teenage labor supply and the introduction of video game consoles. I find that 16- to 17-year old male teenagers significantly increase their hours of work in the months prior to video console introductions beyond the usual male-female difference.

The dissertation is comprised of three papers. In the first two Chapters, I analyze the importance of competition, preference heterogeneity, and socio-economic/country-specific factors to explain the differences in penetration rates of mobile phone services across EU Member States. Chapter 1 presents a model of demand and supply for mobile phone services in which products are perceived as homogenous but consumers are heterogeneous with respect to their valuation of the services. Once a service is purchased, consumers (temporarily) leave the market. The parameters which govern the distribution of preferences are allowed to vary by country and will be estimated as part of the demand specification. The model matches the data well and is able to replicate the observed u-shape in the coefficient of variation in penetration rates over the sample period. Using the demand parameters, consumer acquisition costs are backed out and counterfactual experiments performed. I find that preference heterogeneity and differences in the cost of consumer acquisition explain most of the variation in penetration rates across countries. Competition and other control variables, such as the price of fixed-line calls, play only a minor role. In Chapter 2 I relax the assumption that firms are perceived as homogenous and model them as differentiated products. I incorporate endogenous population weights in a standard random coefficients logit model to capture changes in the demographic composition of potential buyers over time due to the (temporary) market exit of adopters. Compared to the results of Chapter 1, I find a larger role of competition and a smaller impact of the (net) cost of consumer acquisition in explaining differences in mobile phone services diffusion. In the third Chapter, I analyze the effect of a product introduction on labor supply. I demonstrate that it is possible to overcome many of the limitations associated with the lack of individual level purchase data by focusing on teenage labor supply and the introduction of video game consoles. I find that 16- to 17-year old male teenagers significantly increase their hours of work in the months prior to video console introductions beyond the usual male-female difference.