Staff Report 568
What Do Survey Data Tell Us about U.S. Businesses?
Revised December 4, 2018
This paper examines the reliability of survey data on U.S. businesses. We examine publicly available and widely used surveys that ask questions about businesses and business owners and compare outcomes across surveys and with aggregated administrative data. We document large inconsistencies in business incomes, business receipts, and the number of business returns. We highlight problems due to non-representative samples and measurement errors. Non-representativeness is reflected in undersampling of businesses in categories of owners with low total incomes. Measurement errors arise because respondents do not refer to relevant documents when answering survey questions and possibly because of framing issues. Finally, we discuss issues with other statistics of interest, such as business valuations and rates of returns. We conclude that the predictions for U.S. businesses based on current survey data should be treated with caution.
Download Paper (pdf)