What Changed in American Self-Employment
The US local bias of entrepreneurship is a historical pattern with two moving parts: where self-employed people live relative to their birthplaces, and what kind of businesses they are creating. This chapter traces both from 1970 through 2018. The goal is to see whether the fading hometown advantage reflects more mobile people, a less capital-intensive business mix, or both—and to keep a changing relative measure from being mistaken for the disappearance of local firms.
A trend measured against a moving benchmark
The US study combines the 1970, 1980, 1990, and 2000 Decennial Censuses with the 2006–2010 and 2014–2018 American Community Surveys. It examines US-born people aged 18 to 65 and asks whether their current residence is in their state of birth. Self-employed people are compared with wage-and-salary workers after adjusting for observable characteristics such as age, education, marital status, number of children, and region of birth; the analysis for all US-born people also controls for sex and race.
The result is not one smooth estimate of how many entrepreneurs move. It is a series of relative gaps. Among white men, self-employed people were about 8 percentage points more likely than comparable wage workers to live in their birth state in 1970. The gap fell to about 3 points in 2000 and around 2 points in 2014–2018. For all US-born people, it began at about 6 points in 1970, fell below 1 point by 2000, and approached zero in the latest period. The most recent point estimate for the broader population is slightly negative.
The subgroup contrast is a substantive result, not a footnote. The white-men series retains a positive, statistically significant gap in the latest period, while the estimate for all US-born people is near or below zero. The study establishes that the long-run change is not identical across populations. It does not, in the reported result, identify one definitive reason why the broader group converged more sharply. The cautious interpretation is heterogeneity in the location patterns of different groups, not a license to attribute the difference to a single demographic or cultural cause.
The decline also appears in both incorporated and unincorporated businesses and when the analysis is restricted to full-time status. That matters because it reduces the chance that the result is driven only by casual side work or by one legal form of business. It does not eliminate selection into self-employment, changes in the composition of occupations, or differences in the kinds of firms included over time. Those possibilities become central when we turn from the location gap to the business mix.