Price Dispersion in Online and Offline Markets

Consumer electronics such as Liquid Crystal Televisions (LCD-TVs) can be purchased at conventional stores or through internet online stores since the late 90's. On the internet, consumers can simultaneously compare prices at online price-comparing sites (e.g., kakaku.com in Japan). Thus, the prices are expected to be lower and less dispersed on the internet. According to a quantitative analysis, the former result is supported but the latter is not. Empirical studies on the existence and persistence of price dispersion among online stores have been done since the late 90's (see Pan et al. 2004). An early study compared online and offline prices (Bailey, 1998) and found that price dispersion among online stores was at least as great as among traditional conventional stores. Baye et al. (2004) studied the online monthly prices of popular consumer electronics products listed at shopper.com and concluded price dispersion is persistent across products and time. Moreover, we found that the online minimum prices of several types of LCD-TVs are more likely to be listed by online shops in Tokyo than in Osaka at kakaku.com. This implies that the online prices are lower in a larger city. Since the paper of Stigler (1961), there are numerous theoretical studies regarding price dispersion. Stigler (1961) investigated the consumer search behavior under given price dispersion. Salop and Stiglitz (1977) consider the equilibrium price dispersion with high and low price pure strategies for uninformed and informed consumers. Varian (1980) introduced the mixed strategy to explain the persistency of price dispersion. These papers mainly studied offline retail markets and supposed there was incomplete information on prices. On the internet, to the contrary, there is no cost to search for the lowest prices. Thus, Smith and Brynjolfsson (2001) claimed that differences in service quality among online stores is the main source of price dispersion. However, Pan et al. (2002) concluded that such differences are not the main source of price dispersion. This paper considers price dispersion in online and offline markets from a geographical aspect. We investigated the linear online rural market with two big cities lacated on the edges. The access costs, which include access fees, computer skills, and set-up costs to use the Internet infrastructure such as a home computer, credit card, were assumed to be heterogeneous among consumers. We showed that the price dispersion among online stores is larger than among the offline stores, and that the larger the population in the city, the lower the equilibrium prices. The key source of price dispersion is transportation costs, internet access fees, and differences in the population.JEL Classification: D21, D43