Frictionless e-commerce implies that price dispersion for identical products sold by different e-tailers should be smaller than it is offline, but some recent empirical evidence reveals the opposite. A study by Smith et al. (2000) suggests that such a phenomenon may be due to heterogeneity among e-tailers in such factors as shopping convenience, consumer awareness, and trust. These hypotheses, however, remain untested. In this paper, we extend previous research by developing a comprehensive framework of the drivers of online price dispersion that includes market characteristics such as number of competitors, consumer involvement, and product popularity, in addition to e-tailer characteristics and product category differences. We also empirically test our propositions in a more comprehensive manner than prior research by using a range of price dispersion measures covering 6,739 price quotes for 581 products from 105 e-tailers in a variety of product categories including books, CDs, DVDs, desktop computers, laptop computers, PDAs, computer software, and consumer electronics. Specifically, we (1) identify some key dimensions of e-tailer heterogeneity using factor analysis; (2) identify clusters of e-tailers on these dimensions using cluster analysis; (3) analyze how market factors affect price dispersion using regression analyses; and (4) examine how heterogeneity among e-tailers is related to their prices using hedonic regressions by category and by cluster. Our results show that e-tailer services can be characterized by five underlying factors, namely, shopping convenience, reliability in fulfillment, product information, shipping and handling, and pricing policy. There are three clusters of e-tailers who target different consumer groups and position themselves differently along these five factors. Even after controlling for e-tailer characteristics, online price dispersion is large. Market characteristics drive a large portion of this e-tailer price dispersion. Specifically, price dispersion increases with involvement or average price level of items, albeit at a decreasing rate, and decreases with the number of competitors, but at a diminishing rate. The models explain over 92% of the variance in price dispersion. E-Tailers charge prices in line with their characteristics, but do not necessarily command higher prices for superior services. The drivers of e-tailer prices also vary significantly by the cluster to which the e-tailers belong
