Working Paper: NBER ID: w17067
Authors: Jessie Handbury; David E. Weinstein
Abstract: This paper uses detailed barcode data on purchase transactions by households in 49 U.S. cities to overcome a large number of problems that have plagued spatial price index measurement. We identify two important sources of bias. Heterogeneity bias arises from comparing different goods in different locations, and variety bias arises from not correcting for the fact that some goods are unavailable in some locations. Eliminating heterogeneity bias causes 97 percent of the variance in the price level of food products across cities to disappear relative to a conventional index. Eliminating both biases reverses the common finding that prices tend to be higher in larger cities. Instead, we find that price level for food products falls with city size.
Keywords: spatial price index; urban economics; price measurement; barcode data
JEL Codes: L81; R12; R13
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| City Size (R12) | Price Levels (E30) |
| City Size (R12) | Product Availability (L15) |
| Product Availability (L15) | Price Levels (E30) |
| Heterogeneity Bias (C21) | Price Levels (E30) |
| City Size (R12) | Elasticity of Grocery Prices (Q11) |
| Retailer and Purchaser Heterogeneity (L81) | Price Levels (E30) |