Winning Big: Scale and Success in Retail Entrepreneurship

Abstract

In 2014, Washington State used a lottery system to allocate licenses to firms in the newly legalized retail cannabis industry, generating random variation in how many stores entrepreneurs were able to own. We observe highly detailed data on all subsequent industry transactions, including prices, wholesale costs, markups, and product assortments. We find that entrepreneurs who are randomly allocated more store licenses ultimately earn substantially higher per store profits than do single-store firms, suggesting that the returns to scale in the mom-and-pop retail sector are quite large. Despite these firms having less local competition, this increase in profits does not come at the expense of consumers. Rather, retailers in multistore chains ultimately charge significantly lower prices and margins and offer greater product variety. This gap in prices is not initially present but grows substantially over time, as does the difference in assortment size and profits between stores in multistore chains and stores operating alone, consistent with firm learning. Using the full history of outcomes, we track the evolution of firms in this new market and show that multistore retailers use an initial advantage in offering larger assortments to position themselves as the low-price, large-assortment retail option and attract a larger but more price-sensitive set of customers. These results have implications for the study of retail concentration and mergers, countervailing buyer power, and consumer search. Our results suggest that policies to help entrepreneurs expand in retail may have large benefits to both firms and consumers.

Keywords: Economies of scale, retail pricing, entrepreneurship, consumer search, firm learning, cannabis policy

JEL Classification: L11, L22, L81