Thursday, September 23, 2010

Does Education Pay?

By: Matthew Denhart and Jonathan Robe

Tuesday the College Board released its triennial study: Education Pays. The study's opening line reads:
Students who attend institutions of higher education obtain a wide range of personal, financial, and other lifelong benefits; likewise, taxpayers and society as a whole derive a multitude of direct and indirect benefits when citizens have access to postsecondary education.
And thus with the following line it concludes:
Accordingly, uneven rates of participation in higher education across different segments of U.S. society should be a matter of urgent interest not only to the individuals directly affected, but also to public policy-makers at the federal, state, and local levels.
The authors' argument is quite clear: higher education has tremendous benefits, and therefore should be encouraged. In short, "Education Pays," and more specifically, higher education easily passes a cost-benefit analysis.

Yet, while the report provides useful data in a well displayed fashion, it lacks the clear-cut evidence needed to support its thesis that the benefits overwhelmingly outweigh the costs (see further CCAP discussion of this study on YouTube). The authors use a tremendous amount of data that show that, on average, college graduates earn more, live healthier lives, and vote at higher rates than do their counterparts possessing only a high school degree.

However, this study suffers from a confusion of correlation with causation. The authors clearly acknowledge the difference between the two and concede that the mere comparison of people with different educational backgrounds "do not reliably determine causation or measure the exact size of the effects [i.e. the effect of college on outcomes]." Nonetheless, the authors continue to suppose that such comparisons are best treated as: "broadly gauged evidence of the powerful role that higher education plays in the lives of individuals and in society."

It strikes us as odd that comparisons that are unable to reliably predict the size of the effect of higher education on these beneficial outcomes, can concurrently provide reliably "broad gauged evidence" of a "powerful role" for higher education.

Perhaps the greatest problem with our higher education system today is that the public is entirely unaware of the real value of a college degree. Comparisons such as those used in the College Board study ignore the fact that the same types of students who are disciplined enough to complete college are likely predisposed to also have traits that will lead them to earn more money, participate in elections and live healthier lives. Furthermore, as Charles Murray noted in his insightful book, Real Education, these comparisons can be highly misleading, even damaging, to those people who are poor fits for occupations requiring college degrees but who would perform excellently in other careers. Charles Miller and others have pointed other problems with the College Board's approach, as the Chronicle pointed out in its story.

Being able to conduct an honest cost-benefit analysis of higher education is essential to informing our public policy priorities. Better efforts need to be made to measure the benefits that can actually be attributed to college attendance itself. Likewise, cost estimates must go beyond the simple monetary and opportunity cost of attendance to students to include the cost of taxpayer resources that subsidize the system and the opportunity cost of those funds which are thus made unavailable for other, perhaps more productive, uses.

Neglecting to tease out the specific benefits added from college attendance and failing to factor in societal costs of subsidizing higher education, means that the College Board's study grossly overstates the ease with which higher education passes a cost-benefit analysis. Rather than encouraging private citizens and policy makers at all levels to take "urgent interest" in promoting college attendance, we prefer to encourage them to strive to better understand the true costs and benefits of higher education. Only then can an honest analysis take place.


Glen S. McGhee said...

Even to suggest improved cost analyses misses an important point: if, as studies show, on-the-job training is the norm, why not view these graphs as raw, institutionalized, discrimination? Why not view this study as evidence of a legal and institutionalized means of excluding workers from labor markets, and its success?

As the blog suggests, these results are not necessarily causal. In fact, by sampling on the dependent variable in this way, what you are really looking at here is the interplay between credential markets and SES. These effects are largely artificial, based on the exclusionary sorting / signaling effects of credential markets -- where "The employer pretends to need a degree; the employee pretends to have one."

To ignore this deeper issue amounts to begging the question because these effects are largely what degree-producing institutions set out to accomplish more than one-hundred years ago, even earlier. But we forget this, and in the present case, take another look at the results, and then remark on how successful we've been. Duh! Isn't this what we call playing with a stacked deck?

I disagree that "the report provides useful data in a well displayed fashion." Averages are used in the beginning charts which mask "fat tails" and mask what the normal distributions really look like. Even the later graphs that do show ranges are normed-out.

And worse yet, these aren't based on actual earnings, but only estimated earnings! Give me a break!

"However, this study suffers from a confusion of correlation with causation." This is true, for example, when looking at pathways to employment, where, without detailed longitudinal studies, it is impossible to determine what efficacy a certain major has for generating job offers in that field, and to what extent human capital trumps social capital (the hidden assumption of the study) -- especially in a deep recession, such as we now have.

And not a word on how these supposed beneficial effects are all caused by possessing a sheepskin.

Maybe, if I am not being overly cynical, the reason we get flawed and damaging reports like this is because they are created by those produced by the very same institutional sectors now being evaluated, and they are unable to break out of the cognitive bubble kept aloft by sector rhetoric.

But this raises moral questions, doesn't it? How can those in the educational sector produce objective studies about their own sector? They cannot. Rather, this is the view from the inside -- that's what this shows: the view from the inside. And this is the extent of its validity.

Parag said...

Parents nowadays have higher expectations when it comes to their children’s education unlike before so there is a tendency to send them to private or international schools and foreign universities. So education becomes more expensive.
Rising Cost Of Education