Economic analysis of selling and renting software in electronic commerce

This research explores the economic rationale behind the increasing use of software rentals in the software industry in North America. The first part uses a two period game theoretic model to analyze a monopoly\u27s pricing strategy. The monopoly produces two versions of the software, one in each period. Version 2 exhibits delayed network externalities. In period 1, the software is rented and sold where as in period 2 it is only available for sale. We evaluate the impact of software rentals on consumers and overall demand. We find that prices fall in the first period and fan in the second period with increase in the intensity of the network effect. For high levels of network intensity, profits, consumer surplus and social welfare increase upon introduction of software rentals. The second part of the thesis utilizes experimental economics to explore the differences between selling and renting a durable product. Theoretical predictions about the pricing strategy of a monopoly firm selling a durable product span a range with nearly perfect competition at one end and nearly perfect price discrimination at the other. Our results lie in between the predictions of the Coase conjecture and the “Pacman theorem” of Bagnoli et al. In our posted offer markets, we find that a monopoly earns more revenue by renting than by selling a durable product. Renting eliminates the Coasian commitment problem and transforms the complicated multi-period price discrimination problem into a sequence of simpler, identical single-period decision problems. In our data when a seller is required to set both sale and rental prices, revenue is lower than under either renting or selling alone, further corroborating the view that the more complex the decision problem, the more difficulty the monopolist has in extracting rents