Higher Education Bubble

02.20.13

Higher Education and Earnings: Now With Graphs

There's a big premium on college degrees--but it's biggest at the top

In my last post, I mentioned the college earnings premium.  It's real, and it's huge.  Just look at the difference that education makes to earnings potential.  The difference applies to everyone from entry level workers to the highest earners; in fact, the top earning high-school dropouts don't average much better than the bottom-earners with advanced degrees:

You can see the big discontinuity between education levels: high school dropouts, high school graduates, and those with some college tend to be clustered at the bottom, while the incomes of the educated pull away.

Here's the data organized by educational level, instead of income quartile:

Which shows the discontinuities in a different way; there's a big premium on completing high school, and another for completing college.  In fact, I had to change the scale on the graphs halfway through; otherwise, you could barely see the wage changes for high school dropous, since they tend to get stuck with permanently low earnings. 

But you can also see something interesting about income inequality: it's more dramatic among people with advanced degrees.  The high school dropouts in the 9th decile make about 3.1 times as much as those in the 1st.  The 9th high school graduates make about 3.75 times as much as the bottom. Top earners with some college make 3.9 times as much as those in the bottom.  And for those with college or an advanced degree, the ratio is about 4.4.  The higher the education level, the more dramatically the top pulls away from the bottom.  

Why does this matter?  Well, it suggests that college is to some extent a lottery; the averages are concealing a lot of variance.  A lot of the benefit of getting a college diploma is the chance to be one of the very high earners (or enter the advanced degree lottery, which is even more lucrative).  

But it's probably not true that everyone has an equal shot in the lottery.  Going to a second-tier law school is a very different proposition for a first-generation college student then it is for a guy whose father is a partner in a law firm, or a well-respected prosecutor.  If college completion is a proxy for things like cognitive skills, prior academic preparation, or social capital, then adding marginal students to the system won't convey the benefits that schooling is giving to the people who are already getting it.

It's completely true that a high school dropout would be much better off if he'd finished college.  But most people aren't choosing between dropping out of high school, or getting a PhD.  They're choosing between the next step: finishing high school, enrolling in college, finishing if they've started. Assuming that the marginal students we're adding to the educational system are in some way less likely to "win the lottery" may change the calculation of the value of the degree.  If you assume that the median person who decides to get additional education will move down a quartile in doing so (because they don't have the connections or skills of the people who've already got one), then suddenly, the payoff doesn't look so good.  A 3rd quartile candidate with "some college"earning $1085 a week becomes a second-quartile BA holder earning . . . $1066.  

Of course, that's just an illustration; people don't mechanically drop down a quartile by getting more education. But it's something we should keep in mind: as a society, are we preparing people for better jobs? Or are we just slowly moving their old jobs into the "college" basket?  

One thing I'm not arguing against is social mobility.  Rather the opposite: I'd like to see more social mobility!  I just don't like making college the primary vehicle for that movement.  

Yes, yes, college has been a great engine for mobility over the last fifty years.  But the reasoning on this is a bit circular: it's an engine for mobility because we're requiring college degrees for the better jobs, and we can require college degrees for the better jobs because  . . . more and more people are going to college.  In the process, we've shut down other, traditional sources of mobility, like on-the-job experience.

In the 1990s I worked for an IT firm that employed people from all sorts of random backgrounds--there were former telephone linemen running server design, a former building porter installing network backbones, and the lead information architect we worked with had majored in Religious Studies at a small liberal arts college. I was one of several English majors designing and administering networks.  Nor was that unusual--then.  My understanding from people in the industry is that that's all changed; now they want an industry-specific degree. And some of the older guys who got hit with the 2000 and 2008 rounds of layoffs had trouble getting hired, despite a ton of great experience, because they didn't have a stupid piece of paper.

I don't think that's an improvement.    No matter how valuable a college education is, it is not the only valuable thing.  And people shouldn't have to pay a bunch of professors $50,000 just for a chance to earn more than $50,000 a year.