Can “Some College” Help Reduce Future Earnings Inequality?

Abstract

This article addresses the policy debate over “college for all” versus “college for some” in the United States and analyzes the relationship between “some college” (as a formal education attainment category) and earnings. Our evidence confirms—using data from the American Community Survey (ACS), the Panel Study on Income Dynamics (PSID), and the Survey on Income and Program Participation (SIPP)—that more (postsecondary) education, on average, is associated with higher median earnings. However, there is emerging evidence that a proportion of workers who have attained lower levels of education (i.e., “some college”) earn more than those who have attained higher levels of education (bachelor's degree).

We focus particular attention on the subset of Americans who fall into the U.S. Census official category entitled “some college.” This is a heterogeneous group who have alternate educational credentials but who have not acquired a formal associate or bachelor's degree. Instead of an unequivocal focus on “college for all” or even “community college for all,” we argue that educators and policymakers should consider “some college” as a viable pathway to future labor market success. In sum, we conclude that some types of “some college” could lead to a reduction in earnings inequality.

Notes

Some of these excerpts are from Gitterman and Coclanis (2012).

Although it is more difficult to quantify, there are many nonpecuniary gains to education; experience and skills acquired in college reverberate throughout one's life and are observed in more than just earnings (Acemoglu and Angrist, 2001; Buckles, Hagemann, Malamud, Morrill, & Wozniak, 2013, Hout, 2012; Lochner, 2004; Moretti, 2004; Oreopoulos & Salvanes, 2011).

Presence of a sheepskin effect does not negate the labor market value some college credit may have. Carnevale, Rose, & Cheah (2011) suggest that although some occupations may have narrowly defined tasks such that some college education provides no additional value, other occupations may require greater personal initiative, allowing employees with some college to be more productive and earn more (p. 17). Marcotte et al. (2005) note that the relative importance of completion is unclear, but there are substantial returns to some community college, even when a degree is not attained (pp. 170–171).

Jepsen et al. (2014) show that associate degrees and diplomas have quarterly returns of approximately $1,500 for men and $2,000 for women, with a smaller, still positive return for certificates: $300 per quarter for men and women based on data from the Kentucky Community and Technical College System (KCTCS) (pp. 35–37). Lang and Weinstein (2013) note that returns for certificates may vary across majors and between for-profit and not-for-profit institutions (pp. 236–238).

For example, see U.S. Census Bureau (n.d.).

For the distribution of “some college” by state using 2008–2012 data from the American Community Survey, see Appendix 1.

Figure 2 Educational attainment by proportion of the population, 2012. Source. U.S. ACS (2012).

See Appendix 2 for a description of data preparation.

Ruggles et al. (2014). For more on the ACS, see http://www.census.gov/acs/www/guidance_for_data_users/subjects/

PSID (2014). For more on the PSID, see http://psidonline.isr.umich.edu/Studies.aspx

U.S. Department of Commerce (2014). For more on the SIPP, see http://www.nber.org/data/survey-of-income-and-program-participation-sipp-data.html and http://www.census.gov/sipp/

Note that the results in this article are descriptive in nature and as mentioned in the literature review are subject to selection bias; for instance, we show that those earning certain degrees/certificates earn at least as much as bachelor's degree holders, but cannot make the statement that these degrees will causally increase the recipient's earnings because this is not tested.

Figure 3 Median income by educational attainment. Note. The outcome is all earned income (wage and salary income plus farm and self-employment income). The bar graphs depict the median level of income for each educational attainment category. The sample includes 25- to 64-year-olds, working 52 weeks (or 35+ hours per week in the SIPP), usually working at least one hour per week, and with positive earned income. Source: U.S. ACS (2012), PSID (2011), SIPP 2008 Wave 13 (2012).

See Appendix 3 for historical box plots from 1970 to 2010.

For further information, see Appendix 4.

Figure 4 Box plots of earned income by educational attainment. Note. The outcome is all earned income (wage and salary income plus farm and self-employment income). The box plots depict the 25th to 75th percentile, with the solid line within the box representing the median. The sample includes 25- to 64-year-olds, working 52 weeks (or 35+ hours per week in the SIPP), usually working at least one hour per week, and with positive earned income. Source. U.S. ACS (2012), PSID (2011), SIPP 2008 Wave 13 (2012).

Figure 5 Proportion earning more than BA median. Note. The outcome is all earned income (wage and salary income plus farm and self-employment income). The bar graphs depict the proportion of individuals with each educational attainment earning at least as much as the median level of income for a bachelor's degree (calculated separately for each data set). The sample includes 25- to 64-year-olds, working 52 weeks (or 35+ hours per week in the SIPP), usually working at least one hour per week, and with positive earned income. Source. U.S. ACS (2012), PSID (2011), SIPP 2008 Wave 13 (2012).