Paper # Author Title  
We develop a model of firm size in which firms are unable to access as many consumers as they want. Newly arrived consumers match randomly with firms. Subsequently, consumers must pay search costs to be able to switch firms. This cost promotes an inertial tendency for consumers to remain with the ̄rm they currently buy from. Consequently, established firms enjoy a proprietary relationship with respect to their old customers while new entrants only have access to first time consumers, who are as yet unattached to any firm. As time goes on, firms acquire successive generations of new consumers, and their stock of loyal customers grows gradually. Thus, more senior firms command higher market shares. We construct an industry model based on this hypothesis and show that the aggregate implications of the model are consistent with empirical facts about industry dynamics (e.g., Dunne, Roberts and Samuelson, (1988, 1989) or Davis and Haltiwanger (1992)): Larger and older firms are less likely to exit than younger and smaller firms. In our model this results from the fact that the option value of remaining operative in the aftermath of a high cost-shock is greater for an older firm than for a younger firm because the value of a cost turnaround is greater for the former (which has already accumulated a large customer-base) than for the latter (which has yet to do so). This enables older (and larger) firms to survive adverse cost-shocks which force the exit of younger (and smaller) firms. For similar reasons, R&D expenditures are larger for larger firms, as per the empirical findings surveyed by Cohen and Lewis (1989). Download Paper
This paper examines the effect of competition on firms’ efforts to experiment and learn about market demand. Consumers are assumed to know prices only at sellers they have actually visited, but must bear search costs to find the prices of other sellers. Under these conditions we show that firms’ incentives to experiment are diluted by comparison with the monopoly case and that this effect is stronger the smaller the search cost. The learning environment we portray gives rise to several time paths which have been empirically documented, including penetration pricing, cream-skimming and cyclical pricing. Download Paper
We develop a model in which the value of a firm’s reputation for quality increases gradually over time. In our model, a firm’s ability to deliver high quality at any given period depends on how much it invests in quality. This investment is the firm’s private information. Also, a firm’s current quality is unobservable. Thus the only observable is a firm’s past performance - the realized quality of the products it delivered. We assume that information about a firm’s past performance diffuses only gradually in the market. Thus, the longer a firm has been delivering high-quality products, the larger the number of potential customers, which are aware of it. We show that in equilibrium, the firm’s investment in quality increases over time, as its reputation - the number of consumers who are aware of its history - increases. This is because the greater its reputation, the more it has to lose from tarnishing it by under-investing and, conversely, the more it has to gain from maintaining it. This is recognized by rational consumers. Therefore, older - and hence larger firms - command higher prices as quality premia. This in turn feeds back into firms’ investment incentives: the fact that they are able to command higher prices motivates older and larger firms to invest and larger a firm is, the more valuable an asset its reputation is. Download Paper
A striking characteristic of high-tech products is the rapid decrease of their quality-adjusted prices. Empirical studies show that the rate of decrease of QAPs is typically not constant over time; QAPs decrease rapidly at early stages of the product and then the rate of decrease tapers off. Studies also suggest that the QAP is positively correlated with the rate of product introductions: The faster new products are introduced, the faster is the rate of decrease in their QAPs. This paper presents a dynamic model of product innovations consistent with these empirical regularities. Download Paper
We compare the rates of product innovation under rental versus sales when the product is durable and the market structure is one of duopoly. Our main conclusion is that sales induce a slower and more efficient rate of product introductions than rentals. The basic reason for this is that under rentals sellers are able to extract a higher surplus from buyers, and this higher surplus is dissipated away through excessive rate of product innovation. Since the exact opposite is true under monopoly market structure this highlights the role of market structure in determining the rate of product innovations when the product is durable. Download Paper