Human Capital, Labor Market Inclusion and Earned and Asset Incomes

Background:  After the passage of the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (PRWORA; U.S. Public Law 104-193), welfare recipients were faced with new restrictions and challenges as they attempt to become self-sufficient. This study investigated the effects of human capital and labor market inclusion —as structural and systemic labor market factors—on earned and asset income. While there are many studies that validate the relevance of human capital on earned income, there is relatively little known about its effect on asset income. The concept of earned income assumes labor market attachment to be able to bring home the pay from work. However, there is little evidence that labor market inclusion is relevant in terms of increasing asset income.

Methods:

The Core and Topical Module files of the 1996 panel of the Survey of Income and Program Participation (SIPP) data was used for the purposes of this secondary data analysis.  The 1996 SIPP data is a multi-panel, nationally representative, longitudinal survey collected by the U.S. Census Bureau consisting of 36,700 housing units and 46,562 working-age individuals between the ages of 18 and 65.  The panel used consists of 12 waves, covering the period from April 1996 to March 2000.  To mitigate selection bias, the Heckman Two-Step Estimator is used to analyze the relationship between human capital and earned income. Due to the distribution of asset income in the form of censored data, Tobit regression is used as the method of analysis when examining human capital and asset income.

Results: In the first stage of the Heckman model, Probit estimation of employment status (N=34,498) resulted in a statistically significant relationship with educational attainment, job training, and working limiting health conditions as human capital variables in addition to the number of years of work experience and transportation barrier ( = 6007.98, p<0.001).  OLS regression was conducted on earned income (N=29,979) in the second step with significant findings for all human capital and labor inclusion variables (λ= 2158.32, p<0.001), and 20% of explained variance in earned income (=.20).  Estimations yield positive significance of employment status in relation to education and job training, suggesting increased education and job training, and decreased work-limiting health conditions are linked with increased earned income.

(2) Tobit regression results of asset income (N=34,498) exhibited significant relationships with all three human capital variables ( = 5190.49, p<0.001).  In addition, jobs with health insurance was positively linked to asset income (), connecting better jobs with health insurance with increased asset income. However, part-time was not statistically significant and full-time status was negatively associated with asset income.

Conclusions and Implications: This study can inform researchers and policy makers on the relationship of human capital components to both earned income and asset income for future policy considerations.  Education, job training, and health along with labor market inclusion variables contributes to increased earnings income and asset income.  Going forward, public officials may be better served to gain a broader sense of the potential disproportionate impacts of legislative reform.