About Me

I am Nanxi Fu (pronounced nahn-shee fuh), a PhD candidate in Economics at the University of Toronto.

My research examines how the retail environment and competition shape consumer demand and inequality. I combine detailed household purchase data, retailer scanner data, and structural demand models to study consumer behaviour and evaluate the effects of retail and food policies.

I am on the 2026-2027 job market.

Outside of work, I am committed to bringing people from different cultural backgrounds together and building inclusive communities.

Research Interests

Industrial Organization, Urban Economics, Consumer Demand, Retail Competition, and Food Policy.

Job Market Paper

Assortment Policy: How Far Can It Move Consumption? (Draft Available Upon Request)

Abstract

This paper examines whether policies requiring retailers to expand their assortment of healthy foods can increase household consumption, motivated by the 2026 updated stocking requirements of the Supplemental Nutrition Assistance Program (SNAP). I develop a two-layer demand model that captures households' expenditure allocation across product categories and retailers. Combining household purchase data with store-level prices and assortment across nine food categories, I find that greater assortment substantially increases expenditure at individual stores but has limited effects on total category expenditure. This suggests that assortment expansion primarily reallocates purchases across retailers rather than increasing overall consumption. Counterfactual simulations of minimum fresh-produce stocking requirements show that these mandates increase purchases at affected stores but generate only modest increases in total fresh-produce consumption, including among low-income households. These findings suggest that assortment mandates alone are unlikely to substantially narrow nutritional disparities and highlight the importance of distinguishing changes in shopping patterns from changes in overall consumption when evaluating retail policies.

Work in Progress

Would Consumers Do Grocery Shopping Differently with Better Vehicle Access? (Draft Available Upon Request)

Abstract

This paper explores how zipcode-level vehicle ownership can relate to consumers' grocery shopping demands in terms of the bundle purchased, frequency of shopping and size of bundles. For households living in regions with 0.1 more likely to have at least one vehicle access, households on average shop 2.1 percent fewer trips to form an average 1000-dollar bundle. This result suggests less frequent shopping and larger bundles in terms of quantity and variety for households that are more likely to have vehicles. This result can also be related to the patterns of these households are more likely to purchase more non-perishable products. The mechanism behind these characteristics can be related to the different quality of stores in terms of the number of varieties supplied around the households. There is no evidence in the data that consumers with better vehicle access can carry more products or search less to pay higher prices for the same products. The role of vehicles in consumers' grocery shopping can have important policy implications on the food desert issues.

Teaching

I served as an instructor for ECO220: Introduction to Data Analysis at the University of Toronto in Summer 2025. I have also been serving as a teaching assistant for economics courses at various levels, including principles, intermediate-level micro, macro and metrics, and upper-year industrial organization course.