Testing for Consistency in Firm Entry and Exit Between the National Establishment Time Series and Business Dynamics Statistics

ABSTRACT Proprietary data allows researchers to obtain additional, unique information that might not be easily available in public data. In the business establishment space, the National Establishment Time Series (NETS) is a commonly used longitudinal micro dataset of business establishments developed by Walls & Associates and Dun & Bradstreet. A comparable publicly available dataset is the Business Dynamics Statistics (BDS) from the Census Bureau. The BDS is a longitudinal business dynamics dataset of aggregated tabulations including job creation and destruction, establishment births and deaths, and firm startups and shutdowns. Both datasets are widely used to study firm entry, exit, growth, and other development topics in the social sciences. We explore the consistency of firm entry and exit patterns in the NETS and BDS. As shown in prior work, there are substantial inconsistencies in two time series. However, the cross‐sectional correlations between local market measures and firm entry and exit are highly consistent across the two datasets.