Improving Regional Personal Consumption Expenditure Estimates Using Credit Card Transaction Data

• Activity Number: 632 - Using Big Data to Improve Official Economic Statistics
• Type: Invited
• Date/Time: Thursday, August 3, 2017 : 10:30 AM to 12:20 PM
• Sponsor: Business and Economic Statistics Section
• Author(s): Abe Dunn* and Ledia Guci and Mahsa Gholizadeh and Bryn Whitmire
• Companies: Bureau of Economic Analysis and Bureau of Economic Analysis and Bureau of Economic Analysis and Bureau of Economic Analysis
• Keywords: Regional Personal Consumption ; Personal Consumption Flow ; Credit Card Data

Abstract

In 2015 BEA released the first official set of Personal Consumption Expenditures (PCE) by state statistics. These statistics are largely based on Economic Census receipts of establishments, which reflect the location of sale rather than the residence of the consumer, thus adjustments need to made for cross-border purchases. To address this issue, we use credit card data from FirstData, the largest credit card intermediary in the country containing millions of establishments throughout the country. The data include the location of the establishment and information about the location of the consumer. Based on this information, we build estimates of consumption flow patterns across the U.S. for a wide variety of retail industries. We demonstrate that the consumption flows show predictable economic patterns across markets with consumers preferring nearby locations, larger economic markets, and popular vacation destinations. Next, we show how this consumption flow information may be used to correct for the border-crossing problem in our regional PCE state estimates and could even be applied to create a new product of PCE at the MSA level.