UCLA

This dissertation aims to advance our knowledge of long-run economic changes. It consists of three essays on strategic industry dynamics in retail services, agricultural commodities, and high-tech manufacturing, respectively. Although creative destruction is commonly understood as the replacement of old technologies by new ones, its true significance lies not in the transition of technologies per se but in either the reluctance or inability of old winners to innovate when faced with new challengers. Hence I emphasize the incumbent-entrant rivalry as the common theme across these essays.

Essay 1 assesses the impact of the entry of large supermarkets on incumbents of various sizes. Contrary to the conventional notion that big stores drive small rivals out of the market, data from Tokyo in the 1990s show that large supermarkets' entry induces the exit of existing large and medium-size competitors, but improves the survival rate of small supermarkets. These findings highlight the role of store size as an important dimension of product differentiation and caution against size-based entry regulations.

Essay 2 studies the impact of international market structure on commodity prices, using a standard oligopoly model and exploiting historical variations in the structure of the international coffee bean market. The results suggest that, of the 75% drop in the real coffee price between 1988 and 2001, the end of a cartel treaty explains 49 points and the emergence of Vietnam as a major exporter explains another 9 points. I then discuss policy implications for competition, trade, and aid.

Essay 3 investigates why incumbent firms innovate more slowly than entrants. Theories predict cannibalization between existing and new products delays incumbents' innovation, whereas preemptive motives accelerate it, and incumbents' cost (dis)advantage would further reinforce these tendencies. To empirically quantify these three forces, I develop and estimate a dynamic oligopoly model using a unique panel dataset of hard disk drive (HDD) manufacturers (1981&ndash98). The results suggest that despite strong preemptive motives and a substantial cost advantage over entrants, incumbents are reluctant to innovate because of cannibalization, which can explain at least 51% of the incumbent-entrant timing gap. I then discuss managerial and public-policy implications.