Determinants of non-farm self-employment in rural Virginia

Abstract

This research uses primary cross-sectional non-farm-household data collected from Virginia’s non-metropolitan areas for the 1989 year. The primary objective of this research is to identify the factors that influence the probability of being self-employed. Using an effective sample of 851 household heads, a model consisting of two dichotomous probit equations are specified and developed to determine the statistically significant factors that influence, first, the probability of labor force participation, and then the likelihood of being self-employed. The probit equations are estimated by the maximum likelihood estimation (MLE) procedure, using LIMDEP, an econometrics program. The Statistical Analysis Systems (SAS) is used for descriptive analysis and comparisons.

The empirical results reveal that human capital characteristics influence the likelihood of labor force participation. A gender-disaggregated analysis illustrates that presence of pre-schoolers, and an employed spouse all decrease the probability of being in the labor force for women. For men, the variables have an increasing effect.

Overall, two key findings emerge from the analysis with respect to identifying the determinants of being self-employed: That human capital investments in the form of education and particularly, labor market experience, play an influential role in determining the probability of being self-employed; and that access to and availability of financial resources, for example unearned income, are important factors in determining an individual’s likelihood of being self-employed.