From an interesting paper by Ross Levine and Yona Rubinstein:
We find that the incorporated self-employed earn much more per hour and work many more hours than salaried and unincorporated workers. After conditioning on standard Mincerian characteristics, the incorporated self-employed have median residual hourly earnings that are 25% greater and median residual hours worked that are 15% greater than their salaried counterparts. We also find that the median unincorporated individual earns less per hour than his salaried counterpart and much less than a comparable incorporated worker. This helps explain earlier findings concerning the negative pecuniary returns to self-employment: the incorporated earn more than salaried workers, the unincorporated earn less, and there are more unincorporated than incorporated individuals.
The higher earnings of the incorporated self-employed partially reflect returns to individual traits and partially the returns to activities associated with incorporation. Individuals that at some point in their lives incorporate tend to earn about 33% more on average (and 20% more at the median) as salaried workers than comparable salaried workers that never incorporate: some people have traits associated with both higher earnings, regardless of employment type, and a greater tendency to incorporate. Nevertheless, even when controlling for individual effects, the average individual enjoys a 14% boost in residual hourly earnings when switching from salaried to incorporated self-employment (while the median person receives a 4% increase).
Furthermore, the distribution of the residual hourly earnings of the self-employed, especially the incorporated, has much fatter tails than that of salaried workers. For example, people that are successful when they are incorporated (90th-percentile of the residual hourly earnings distribution of the incorporated) tend to enjoy 70 percent more earnings than their earnings as successful salaried workers (90th-percentile of the residual hourly earnings distribution of the salaried). Entrepreneurship offers the possibility of comparably enormous positive returns.
In particular, we find that (1) incorporated individuals are more educated and more likely to come from high-earning, two parent families than salaried workers, and (2) even as teenagers, people that incorporate later in life tend to score higher on learning aptitude tests, exhibit greater self-esteem, and engage in more aggressive, illicit, and risky activities than those that do not. Along most of these dimensions, the unincorporated are on the other end of the spectrum, with values lower than salaried workers. This helps account for the puzzling observation that self-employed and salaried workers have similar traits: aggregating the incorporated and unincorporated masks crucial differences about the traits of people that sort into each sub-category of self-employment.
We also discover that several cognitive and noncognitive traits are more important for shaping the pecuniary returns to incorporated self-employment than they are for influencing the success of salaried and unincorporated individuals. Learning aptitude, self-esteem, and aggressive, risk-taking traits—which are all measured when individuals are teenagers—are especially, positively associated with being a highly successful incorporated business owner later in life. While a diverse body of research argues that traits related to self-esteem induce individuals to try self-employment (e.g., Zukerman 1994; Nicolaou, et al. 2008), our work suggests that such traits yield large pecuniary returns. These findings are consistent with research documenting nontrivial returns to noncognitive traits (Bowles et al. 2001; Heckman and Rubinstein, 2001; Heckman et al. 2006; Heckman, 2000). […]
Note, we do not evaluate the causal impact of incorporation on earnings. Rather, we assess the pecuniary returns from self-sorting into incorporation, the cognitive and noncognitive traits underlying this sorting, and how these traits differentially shape the returns to various employment activities. The nature of entrepreneurs and the returns to entrepreneurship are inextricably connected.