Working Paper: NBER ID: w8667
Authors: Florian Zettelmeyer; Fiona Scott Morton; Jorge Silvariso
Abstract: This paper addresses the question of how much the Internet lowers prices for new cars and why. Using a large dataset of transaction prices for new automobiles and referral data from Autobytel.com, we find that online consumers pay on average 1.2% less than do offline consumers. After controlling for selection, we find that using Autobytel.com reduces the price a consumer pays by approximately 2.2%. This suggests that consumers who use an Internet referral service are not those who would have obtained a low price even in the absence of the Internet. Instead, our finding is consistent with consumers choosing to use Autobytel.com because they know that they would do poorly in the traditional channel, perhaps because they have a high personal cost to collecting information and bargaining. This group disproportionately uses Autobytel.com because its members are the ones with the most to gain. We estimate that savings to consumers who use Autobytel.com alone are at least $240 million per year. Since there are other referral and informational sites that may also help consumers bargain more effectively with dealers, we conclude that the Internet is facilitating a large transfer of surplus to Internet consumers in the retail auto industry.
Keywords: internet pricing; auto industry; referral services
JEL Codes: L1; O3
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| internet referral services (L86) | car prices (P22) |
| using autobytel.com (L62) | prices paid by consumers (D19) |
| internet referral service (L86) | lower prices for consumers who perceive themselves as poor bargainers (D11) |
| internet referral service (L86) | division of surplus in retail auto industry (L81) |
| specific make and model quotes (L62) | prices paid by consumers (D19) |
| internet use characteristics (L96) | prices paid by consumers (D19) |