UC Berkeley

This thesis consists of two empirical studies of the industrial organization of food retailing.

Chapter 1 investigates consumer behavior in response to changes in variety in grocery store offerings. Using a rich panel dataset of consumer purchases from one national retailer, we measure the effects of changes in product variety on sales of ice cream, where the assortment changes were gradually introduced to stores. We use a difference in differences approach to measure treatment effects against trends in stores prior to treatment. Our results show an adjustment to product variety causes consumers to alter their purchasing behavior, at least in the subsequent eight weeks of the change. When the rearrangement consisted of adding additional products, consumers purchased more on average in the immediate term, but sales increases mitigated over time, with smaller magnitude increases in variety experiencing sales reductions in later weeks. Product groups treated with reduced variety experienced decreased sales both in the immediate and longer term. We find that extending and reducing assortment have asymmetric effects, with a marginal change to increased variety boosting sales less than a marginal change in decreased variety. Larger magnitude reductions perform poorly in the short run, but actually perform better than both smaller variety reductions and small variety increases in later weeks.

In Chapter 2, I investigate the following question: do firms respond to quality disclosure by nearby competitors? This paper utilizes exogenous variation provided by Los Angeles County's introduction of restaurant hygiene grade cards to explore inter-firm responses in markets for differentiated goods. Under this program, a subset of cities adopted mandatory grade card posting, requiring restaurants to disclose hygiene. This study demonstrates that restaurants under a voluntary hygiene disclosure policy improved hygiene more when located closer to restaurants in mandatory disclosure areas. The proximity effect implies that mandatory disclosure policies influenced restaurant hygiene in nearby voluntary disclosure areas, thus providing direct evidence that quality-disclosure regulations affect non-targeted firms.