The College Payoff Illusion

Sure, college pays off well in the 
job market—for those who really belong there

Edwin S. Rubenstein

[back to table of contents]

Every spring, graduating college seniors all around the country parade up to the podium to accept their degrees and then stride off into the job market with high hopes. For them and their tuition-strapped parents, the conventional wisdom reigns: a college education is the only way to get ahead in the high-tech economy. Unfortunately, what everyone "knows" is not necessarily true.

It is certainly true that the marketplace is screaming for college graduates with intellectual skills beyond those of well-paid plumbers and auto mechanics. Graduating seniors with mathematics or computer skills now command the kind of five-figure signing bonuses and six-figure starting salaries once reserved for law-school graduates. The Information Technology Association recently estimated that 346,000 high-tech jobs were unfilled because employers could not find the right persons to fill them.

What is definitely not true is the obverse: that kids without college degrees are doomed to a life of near-poverty. Anyone wishing a quick reality check on this proposition should turn to the November 1997 Monthly Labor Review, published by the U.S. Department of Labor. The featured article was a twenty-six-page analysis of job prospects in five-hundred different occupations from 1996 to 2006. The analysis predicted a net gain of 18.6 million jobs during the decade.

So what is the hottest occupation in terms of new positions? The surprising answer is cashiers (see accompanying table). The authors estimate that the economy—globalized or not—will need 530,000 more cashiers by 2006. Although the checkout counter is increasingly computerized, the work has become easier, not harder, as technology has eliminated the need to read price tags and compute change. Working a checkout line is easier than ever, hardly suitable work for a college graduate.

Occupations With the Largest Job Growth, 1996-2006
Employment (thousands of jobs)


1996 2006 Increase

Education/Training Required

Cashier 3,146 3,677 530 On-the-job training
Systems analyst 506 1,025 520 Bachelor's degree
General manager 3,210 3,677 467 Work experience/college degree
Registered nurse 1,971 2,382 411 Associates degree
Retail salesperson 4,072 4,481 408 On-the-job training
Truck driver 2,719 3,123 404 On-the-job training
Home health aide 495 873 378 On-the-job training
Teacher aide 981 1,352 370 On-the-job training
Nursing aide 1,312 1,645 333 On-the-job training
Receptionist 1,074 1,392 318 On-the-job training

Source: U.S. Department of Labor

The number of technical positions will obviously grow more quickly on a percentage basis, but the Labor Department projections show that just 4 percent of the 2006 workforce will be "Technicians and related support occupations," up from 3 percent in 1996. Most jobs will be in the low-tech service sector, and you don’t need more than a high school degree in twenty of the thirty occupations expected to deliver the most job growth. Joining cashiers in the top ten are retail sales clerks, truck drivers, home health aides, teachers’ and nurses’ aides, and receptionists/information clerks.

The Return To Higher Education

How do we measure the economic return to higher education? Typically it is calculated as the difference between average wages of college graduates and those who have not graduated from college. In 1997, for example, college graduates earned an average of $40,508 versus just $23,970 for non-college graduates. (The figures, from the federal Bureau of Labor Statistics, pertain to full-time workers at least twenty-five years old.) Based on these income levels, the economic return to a college education is approximately 69 percent, the difference between the two income levels. But this simple calculation ignores the fact that college graduates tend to come from higher socioeconomic levels, are more highly motivated, and probably have higher IQs than nongraduates. Although these factors influence incomes, they are not results of college attendance. The result is an overstatement of the returns to higher education.

More sophisticated analyses adjust for these extraneous influences. Economists Orley Ashenfelter and Alan Krueger, for instance, estimate that each year of post-high school education results in a wage premium of between 15 and 16 percent. Their study is particularly relevant because they examined the earnings differences for identical twins with different education levels, allowing them to control for genetic and socioeconomic factors. Other research puts the wage premium for college graduates at nearly 50 per cent. For workers with one to five years of experience, earnings differentials of over 70 percent are reported.

Unfortunately, you can’t spend a college wage premium. Income levels for the average college graduate have stagnated. After adjusting for inflation, the average income of college graduates holding full-time jobs rose by only 4.4 per cent between 1979 and 1997, or at a minuscule annual rate of 0.2 percent (see fig. 1). At the same time, workers with only high-school degrees saw their real income plummet by 15 percent. Bottom line: the much-ballyhooed college wage "premium" is due primarily to the fall in inflation-adjusted salaries of workers who haven’t been to college.

In fact, if you don’t go on to graduate school or are not among the top graduates at one of the nation’s elite colleges, chances are your sky-high tuition is buying you no economic advantage whatsoever. In recent decades the flood of graduates has been so great than an increasing proportion have found themselves, within a few years, working as sales clerks, cab drivers, and in other jobs that do not require a college degree. In 1995, approximately 40 percent of people with some college education—and 10 percent of those with a college degree—worked at jobs requiring only high-school skills. That’s up from 30 percent and 6 percent, respectively, in 1971 (see fig. 2).

A powerful case can be made that the U.S. has too many college graduates, not too few. In an age of abundant student loans and, more recently, a slowing in the rate of tuition increases, fully 65 percent of recent high school graduates are in college. Too many colleges chase too many marginal students. "If you have a high school diploma—and can walk and talk—you can graduate from college," says the Dean of the Harvard University School of Education, Jerome Murphy. "There are a lot of empty seats."

The U.S. surpasses all other advanced nations in the proportion of workers with university degrees, according to the Organization for Economic Cooperation and Development (see fig. 3). Apparently our students need college just to reach achievement levels other nations expect of high schoolers. Evidence? American twelfth graders finished nineteenth in math and sixteenth in science among the twenty-one countries participating in the Third International Math and Science Survey in 1996.

There two more depressing facts to consider. First, the glut of students means that approximately 30 percent of all incoming freshmen are placed in at least one remedial course. (At state universities, remediation rates are higher—36 percent. 48 percent, and 39 percent reported in New York, Kentucky, and Georgia, respectively, according to the Thomas B. Fordham Foundation. In effect, these students have flocked to college to acquire skills they should have learned in high school.

Second, and perhaps more troubling, is the corrosive impact of unqualified students on college standards. Faculty members are under pressure to lower course demands "to maintain [their] department’s enrollment and claim on university resources," writes economist Robert Costrell of the University of Massachusetts. Even elite schools are not immune to this malady. A 1996 study by the National Association of Scholars documents a long slide in the curricula at the fifty most selective colleges in the nation. There has been an extraordinary increase in the number of college courses per institution, from 304 in 1914 to 1,418 in 1993. That sounds good—more choice. But simultaneously, the proportion of courses without prerequisites—that is, those not sufficiently rigorous to demand advanced knowledge—has also leapt, from 8 percent in 1914 to 41 percent in 1993. Remedial English may be the fastest growing course offering in U.S. colleges (see fig. 4). And remember, these are the top schools in the country.

The Paradox

Nonetheless, as noted earlier, wages of college-educated workers have been increasing relative to those of their high-school-only counterparts (see fig. 1). That’s a paradox. On the one hand, it looks as if the supply of college-educated workers is outstripping the demand for them. On the other hand, rising wages suggest that there are not enough college-educated workers to go around: employers are scrambling to hire them, bidding up their wages relative to those of workers without degrees.

Another Monthly Labor Review article resolves the apparent contradiction. Swarthmore College’s Frederic Pryor and Haverford College’s David Schaffer find that there is indeed a shortage of college-educated workers—qualified ones, that is. Rather than use the number of years of education as a proxy for skills, Pryor and Schaffer use "functional literacy" as measured by the National Adult Literacy Survey administered to a sample of the U.S. adult population in 1992.

The test distinguishes three types of functional literacy:

Prose literacy refers to "the knowledge and skills needed to understand and use information from texts including editorials, news stories, poems, and fiction." One question, for example, required respondents to summarize the main argument of an op-ed article.

Document literacy comprises "the knowledge and skills required to locate and use information contained in materials such as job applications, payroll forms, transportation schedules, maps, tables, and graphs." Survey takers were asked, for example, to complete an employment application or interpret a line graph.

Quantitative literacy refers to "the knowledge and skills required to apply arithmetic operations, either alone or sequentially, to numbers embedded in printed materials, such as to balance a checkbook, complete an order form, or calculate the amount of interest from a loan advertisement." One such question asked the respondent to add up the cost of a meal, calculate the change, and ascertain the amount of a 10 percent tip.

Although the skills are different, individuals who score well or poorly in one area tend to perform similarly well or poorly in the others. The point was not to test knowledge but analytic ability; individuals who cannot analyze written material tend to be deficient in map reading and arithmetic also—they are low in functional literacy.

Pryor and Schaffer compared the survey results with education levels, occupations, and wages. What did they find? College graduates who do not have the functional literacy traditionally associated with college degrees are the ones taking jobs that had previously gone to employees with high-school diplomas only. The wages of these folks, after adjusting for inflation, have remained roughly unchanged over the years (about $15 per hour in 1994.) The big winners are college graduates in jobs—such as management analysis and financial administration—requiring the functional literacy levels a college degree traditionally represented. These workers received an average of more than $25 per hour in 1994, compared with less than $21 (in 1994 dollars) in 1970.

Implication? Functionally literate college grads are in short supply. But the conventional wisdom, reinforced by Washington policy, is that every kid can handle college work—and should be subsidized to that end. In the last State of the Union address, President Clinton pledged financial aid for every needy family desirous of putting a child through college.

That may sound nice, but the costs of recruiting and providing financial aid for a diverse student body are added to tuition. Approximately 20 percent of the average student’s tuition payment now goes for financial aid for other students, double the amount of two decades ago. The more tuition goes up, the more aid students need, and the less additional net revenue the hikes bring in. Who gets hurt? Middle-class families, many of whom have been saving for college for years and will be paying off college loans for years to come. Although data are not available, we would be not be surprised to learn that children of middle-class parents denied student loans score higher in functional literacy than children who receive generous loan amounts.

Also, as high as tuition is, it is dwarfed by the indirect or "opportunity" cost of attending college. The four years spent in pursuit of a B.A. are years the student could have spent in on-the-job training, earning money while learning skills, rather than spending money to acquire a degree. Mediocre students would receive a far greater long-run payoff from working rather than going to school during those years.

Taxpayers, parents, and college students are paying for millions of kids to sit through years of unproductive boredom that do little to improve their economic prospects. Rather than paying for college educations for cashiers, home health aides, and receptionists, we would do well to consider ending the subsidies that encourage this irrational behavior.

Edwin S. Rubenstein is Director of Research at the Hudson Institute and a columnist for Forbes. He is the author of The Right Data (1994) and From the Empire State to the Vampire State: New York in a Downward Transition (1996, with Herbert London).

[back to table of contents]

Copyright 1999 by Hudson Institute.
To leave a comment about this page, click here.