During the past two decades, the importance of education beyond high school in determining economic well-being has grown substantially. By the early 1990s, economists understood that, for decades, changing skill demands in the labor market have favored workers with more education. Although the large influx of college-educated workers (created by the entrance of the Baby Boom generation into the workforce in the 1970s) temporarily suppressed the wage advantage associated with postsecondary education, by the early 1980s, the large supply shift abated, and the relative wages of workers with a college education began a dramatic and persistent increase.

Though the technological and economic changes affecting the shift in labor demand are signs of a vibrant economy, they raise concerns about the economic reality faced by workers without any postsecondary education whose real wages have stagnated for the past 30 years. In response to these shifts, most policy-makers and analysts have focused on efforts to encourage high school students (and perhaps those already in the workforce) to enroll in postsecondary education. Unfortunately, policy and research on postsecondary education have tended to focus on education within four-year postsecondary institutions. Both in the arena of higher education policy and in the general perception of what postsecondary education means, two-year or sub-baccalaureate education receives little attention, and even less attention has been paid to occupationally oriented education.

At the bachelor's degree level over the last two decades, the number of college students majoring in liberal arts or academic subjects such as English, sociology or chemistry has declined both in absolute numbers and relatively while applied fields such as health, business, and education have increased. In fact, occupational fields now account for a majority of sub-baccalaureate enrollments (Bailey, Leinbach et al. forthcoming; National Center for Education Statistics 2002). What are the implications of this trend? Do occupational fields at the sub-baccalaureate level offer students good opportunities or do they lead them into educational pathways and occupations that limit their options and provide more restricted earnings potential?

Because this report is part of the National Assessment of Vocational Education (NAVE), we have a particular interest in the economic returns to occupational education. Occupational programs prepare students more or less directly for work while academic programs presumably provide a broader education. Many employers want to be able to hire workers who can start work without initial extensive training. Thus, one hypothesis might be that sub-baccalaureate students who pursue a more academic program might be more easily trained whereas those in an occupational program would need less training to be immediately productive. In evaluating occupational education, we would like to know what employers pay for workers with these different types of education. If employers value the specific preparation of an occupational program, are they, at least in the relative short run, willing to pay extra? An academic education at the sub-baccalaureate level may be useful but only as preparation for completing a bachelor's degree. A wage premium for an occupational education would suggest that it would make sense for students to pursue an occupational course of study if they were not going to go beyond an associate degree.

Data are available to address many of these questions, yet higher education researchers and labor economists who analyze the returns to various levels of education have paid little attention to a sub-baccalaureate education in general and even less to an occupational sub-baccalaureate education. Many studies of the earnings effect of education simply use an "undifferentiated years of schooling" variable or categorize any person without a bachelor's degree but with education beyond high school as having "some college." Only a handful of studies have sought to estimate the economic effect of a two-year or sub-baccalaureate education, with or without a certificate or degree, and even fewer have attempted to estimate the returns for occupational students. Moreover, those few analysts who have addressed these questions have used data from the 1970s and 1980s. The higher education system, the community college role within that system, the demographics of the student population and labor force, the nature of the labor market, and the technological characteristics of jobs have all changed in the last 30 years, casting doubt on the conclusions implied by these studies.

This report presents an analysis of the economic returns to postsecondary education. It adds to the current knowledge in this area both by focusing on the returns to occupational sub-baccalaureate education and by using a newly released dataset, the National Education Longitudinal Survey of 1988 (NELS). NELS comprises a nationally represented sample of eighth grade students in 1988. To get some understanding of trends, we also conducted a similar analysis of the High School and Beyond (HS&B) dataset that is composed of students who were high school sophomores in 1980. HS&B follows this cohort of young adults until 1992, when most were in their late 20s. Both of these datasets provide detailed information with respect to investments in postsecondary education and economic outcomes from a representative sample of more than 7,000 men and women who finished high school in the early 1980s and 1990s. Our analysis is based on data on annual earnings collected eight years (NELS) and 10 years (HS&B) after scheduled high school graduation.

We also used the Beginning Postsecondary Students Longitudinal Survey (BPS89) dataset, which tracks a cohort of first-time beginning students in the 1989-90 academic year for four years. With BPS89, short-term (one to two years after college) returns to education are estimated, but there are several caveats that will be explained in the following discussion. Unlike NELS and HS&B, a major advantage of this dataset is that education and work experiences of older students can be observed (27 percent of the sample are age 24 and older). However, to allow for at least one year of work experience after leaving college, we had to restrict the sample to those who completed or left college after six semesters, which is two semesters fewer than what is typically needed for a bachelor's degree. We recognize that the estimates from the restricted sample do not necessarily represent those of all first-time beginners in BPS89. Therefore, caution must be taken when reading the returns from this analysis.

In this report, we present findings from our analyses of economic benefits using the HS&B and NELS data and briefly discuss our analysis of immediate economic outcomes of special populations using BPS89. Our analyses have been designed to answer the following questions:

1. What are the average earnings benefits associated with sub-baccalaureate degrees?

2. How do the earnings of sub-baccalaureate students (those who complete degrees and those who do not) compare with the earnings of individuals who have no more than a high school diploma?

3. Do occupational students who earn degrees earn more than those who complete an equivalent amount of education but do not earn degrees (referred to as the sheepskin, or program, effect)?

4. Is the sheepskin effect stronger or weaker for occupational students than it is for academic students?

5. What is the effect of enrollment in a high school vocational track on the subsequent earnings of sub-baccalaureate students, and does the high school track have a different effect on the earnings of academic and occupational sub-baccalaureate students?

6. Is occupational education beneficial for subpopulation groups such as students who are older, students who belong to a racial-ethnic minority, or students who are academically or economically disadvantaged?

This report is structured as follows. First, we provide a brief review of the literature, with a specific focus on the returns to different aspects of occupational education. Second, we describe the datasets and the samples studied here, highlighting both educational experiences and economic outcomes. Next, we describe average earnings and employment differences among students in the NELS, HS&B and BPS89 samples who engaged in various forms of postsecondary study. Finally, we present results from a variety of multivariate models estimated to identify returns to enrollment and credentials from sub-baccalaureate programs and to answer the questions posed above.

To maintain terminological coherence with the other reports we have done on this subject, sub-baccalaureate refers to those students who began at a two-year college or those who had a less-than-bachelor's degree goal at their initial postsecondary institution. In all analyses, a college student can be either a baccalaureate (attending a four-year college) or a sub-baccalaureate. Sub-baccalaureate students are further divided into occupational and academic, based on their initial program of study. Students who did not declare a major at their first postsecondary institution are retained in the sample and identified as such. Throughout this report, when we refer to a high school graduate, we mean a student who graduated from high school but who has no postsecondary education. In most of our analyses, high school graduates are the reference category; therefore, the coefficients on educational variables can be interpreted as the earnings difference between whatever group the variable represents (say, those who completed an associate degree) and high school graduates with no postsecondary education.