Identification of capitalists and workers

Is there any variable or combination of variables which, taken together, indicate if a person has an ownership interest in their business or place of employment? Is it possible to identify any such individuals, even some small percentage? If the latter, i.e. only some persons with ownership interest are identifiable, are there also some identifiable individuals who definitely do not have any ownership interest in their place of work? For example, if a person reports only wage income and no form of capital income, do you think we can validly infer that none of their wage income derives indirectly from business ownership?

This forum thread discusses this issue - the short answer is that you can get a rough idea of who are business owners using CLASSWKR, but it’s not 100% clear. Besides the discussion in that thread, you may also want to look for positive INCBUS in the ASEC samples.

Thanks Matthew! That is really very helpful.
There are several problems with taking this data at face value. Income from a business is typically some mix of labor and capital income, but the legal classification of such income, on which I assume the reported characterization is based, is largely unrelated to its economic character. Since 1983 or so the share of total business income that is distributed through pass-through entities (partnerships, s corporations, proprietorships) has gone from a few percent to about 60 percent. Since some of these forms apply payroll taxes income characterized as wages, there is an incentive to recharachterize that income as profits.
It seems to me that the right answer is to estimate the true labor income of these individuals via a Mincer equation, with some upward adjustment for managerial or entrepreneurial skill (capitalized and added to human capital generally), and then estimate profits as the residual. This would cause a very marked additional decrease in the reported labor share in recent years.

See, e.g., this NBER report:

1 Like