Working Paper: NBER ID: w14166
Authors: Maris Goldmanis; Ali Hortacsu; Chad Syverson; Onsel Emre
Abstract: While a fast-growing body of research has looked at how the advent and diffusion of e-commerce has affected prices, much less work has investigated e-commerce's impact on the number and type of producers operating in an industry. This paper theoretically and empirically takes up the question of which businesses most benefit and most suffer as consumers switch to purchasing products online. We specify a general industry model involving consumers with differing search costs buying products from heterogeneous-type producers. We interpret e-commerce as having reduced consumers' search costs. We show how such reductions reallocate market shares from an industry's low-type producers to its high-type businesses. We test the model using U.S. data for three industries in which e-commerce has arguably decreased consumers' search costs considerably: travel agencies, bookstores, and new auto dealers. Each industry exhibits the market share shifts predicted by the model. Interestingly, while the industries experienced similar changes, the specific mechanisms through which e-commerce induced them differed. For bookstores and auto dealers, industry-wide declines in small outlets reflected market-specific impacts, evidenced by the fact that more small-store exit occurred in local markets where consumers' use of e-commerce channels grew fastest. For travel agencies, on the other hand, the shifts reflected aggregate changes driven by airlines cutting agent commissions as consumers started buying tickets online.
Keywords: E-commerce; Market Structure; Retail Industries
JEL Codes: D4; L1; L8
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| e-commerce reduces consumer search costs (L81) | market shares reallocate from low-type producers to high-type businesses (L19) |
| increases in online purchasing (L81) | declines in the number of small establishments (L25) |
| e-commerce use grew rapidly (L81) | faster exit of small stores in local markets (L81) |
| e-commerce diffusion (L81) | changes in market structure (D49) |
| decline in small outlets (L81) | aggregate changes driven by airlines cutting commissions (L93) |