The One-Dollar Company
Filing Fees and What Business Registrations Measure
Abstract
Counts of new business registrations are used to measure entrepreneurship, and governments change the fees that registrants pay. We study three changes in a formation fee of about $50 and follow companies formed around each into records that operating businesses leave. When Colorado twice charged $1 instead of $50, registrations rose by 17%; when the United Kingdom raised its fee from £12 to £50, they fell by 17%. In pre-specified tests, registrations that responded to the fee had such records less often than others. Their relative rate is 0.07 for Paycheck Protection Program loans (95% confidence interval −0.57 to 0.60; second reduction, 0.21), −0.02 for any of three lending records (−0.65 to 0.41; second reduction, 0.20), and 0.00 for non-dormant U.K. accounts (confidence set −2.5 to 0.42). Alternative comparison periods give pooled estimates from −0.41 to 0.81 for the loans and at most 0.34 for the other two outcomes. These rates describe records. They describe businesses only if businesses among such registrations are recorded as often as others, and they do not measure a registration’s value to its filer. Changes in registrations after fee changes should not be read one-for-one as changes in the number of operating businesses.
Keywords: entry costs; business registration; measurement of entrepreneurship; filing fees; selection
This draft, its analyses, and its code were produced by Claude, an AI model developed by Anthropic, working as a research agent at Ethan Mollick’s request. No person has verified the results, rerun the code, or audited the links between records. The PDF contains the main text (31 pages), references, and an online appendix.