Chart of the day, College tuition edition

By Felix Salmon
September 9, 2009
John Caddell, and it shows the cost of attending Rensselaer Polytechnic Institute as a percentage of US median income. Scary stuff. But not as scary as David Leonhardt's column today, which demonstrates a nasty ghettoization effect at state colleges, many of which are turning into failure factories:

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">


This chart comes from John Caddell, and it shows the cost of attending Rensselaer Polytechnic Institute as a percentage of US median income. Scary stuff. But not as scary as David Leonhardt’s column today, which demonstrates a nasty ghettoization effect at state colleges, many of which are turning into failure factories:

Only 33 percent of the freshmen who enter the University of Massachusetts, Boston, graduate within six years. Less than 41 percent graduate from the University of Montana, and 44 percent from the University of New Mexico.

There are serious problems with incentives, here: colleges get paid according to their enrollment, not according to the number of students they graduate. And with freshmen cheaper to teach than seniors, it actually benefits a college to have more of the former than the latter.

The first order of business here is to level the playing field: at the moment, poor kids have an alarming tendency to attend colleges with low graduation rates, even when they’re more than capable of getting into colleges with higher graduation rates. They thereby essentially give their rightful place at the better schools to richer kids, who are much more likely to graduate in the first place. That’s why Joe Weisenthal is wrong here:

The authors cite students who go to Eastern Michigan University (39% graduation rate), but who could have gone to University of Michigan (88% graduation rate). But UMich is already at maximum capacity — as are other elite schools — so for one thing, an influx of new applicants, would just displace students, and we’d be back to ground zero. But beyond that, how do we know that the the UMich graduation rate would stay constant given an influx of students who used to go to Eastern Michigan? That’s a gigantic variable.

The point is that the displaced students would be richer students who would be much more likely to graduate in any case, even if they went to Eastern Michigan University. And even if the UMich graduation rate fell from 88%, it would still be much higher than 39%.

This, from Weisenthal, is also silly:

While the educators complain that high schools aren’t doing enough to prepare students for college, the goal of “improved matriculation” sounds just as silly. Just as graduating from high school doesn’t automatically make you prepared for college, graduating from college doesn’t automatically make you ready for the real world.

No, but it makes you much more employable, and it does wonders for your lifetime earnings. Lots of companies simply won’t employ a college drop-out, no matter how qualified they are, and will employ a less-able college graduate instead. It’s often the first filter applied: companies won’t even look at people without a degree.

Improving the matriculation rate therefore improves the range of choice presented to employers, and improves the overall quality of the white-collar workforce. It benefits everyone, except for maybe dubiously-competent graduates who right now don’t really need to compete, in the job market, against smarter applicants who unfortunately dropped out of college. If those smarter applicants get themselves degrees, the less-competent will find it harder to get jobs.

Update: Ryan Avent seems to think that dropouts are dropouts, wherever they attend university. I don’t think that’s true. For one thing, the presence or absence of a campus makes a big difference: students are much more likely to drop out if there isn’t one. And for another thing, there are all manner of peer effects: if most of your peers are dropping out, you’re more likely to follow suit than if nearly all of them are graduating.

Update 2: Oops, Charles Kenny just pointed out that Caddell was using the wrong numerators in his chart, he was using the top line when he should have been using the second line. So I’ve deleted the chart, as it stood it was very misleading.

Update 3: Caddell has fixed the chart, so it’s back.


We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see

With all due respect, both phenomena are rational and explainable from the perspective of less wealthy students.

First, in every way — from dating to housing to hockey tickets — UM is more expensive than EM — leave aside tuition.

Second, the attrition rate at EM-type schools is also rational given the opportunities available both to drop outs and to graduates.

But most importantly, where higher ed is concerned, Marshall McLuhan was right: the medium is the message. The medium is debt; the message is you better have a plan to find a conventional job that pays a lot.

Talk to college dropouts from EM-type schools. Very few in my experience mention anything ahead of expense — and the unlikelyhood of financing and repaying it — as a reason for leaving.

The EMs of the US provided teachers, engineers, nurses, accountants, etc. In short, our lower- and middle- middle class infrastructure.

The ever-rising college costs and the debt required to finance them is driving social change. To understand why our competitive/economic future is so bleak one need look no farther than EM and the now-gone infrastructure of low cost, public institutions that existed into the 1970s.

Posted by Peter D. Kinder | Report as abusive

sigh. this is, as usual, the government’s fault. like most things that are irrationally overpriced, the cause is subsidies, in this case, much as with housing, through artificially easy credit. the solution is to abolish federal student loans and most of the rest of the ways federal money skews the cost of education. if colleges had to face a real market, they could never charge multiples of the median income for their degrees.

Posted by Aaron Davies | Report as abusive

Felix, thanks for sharing my chart with your readers. I agree with Aaron Davies but would go farther. The government is only one factor skewing the cost of education. The colleges themselves, with grants, loan programs and tuition abatements, also contributes to a murky, uneven market.

This contributes to their automatic assumption that they can increase their charges above the inflation rate every single year.

regards, John

That chart shows the “list price” of attending RPI, the cost without merit money or need based financial aid. Such aid is substantial at schools like RPI, and even greater at richer schools like the Ivies. If your family income is under $100K, going to Harvard (and possibly RPI) is still quite affordable. Harvard, in fact, is listed as a “Best Value” in college guides for that reason. The ones who take it on the chin are the lower echelons of the upper middle class, ie., income > $100K, who get little or no aid, and in fact, subsidize the tuition of the students with whom they competed (insanely) for admission. Nice system…

Posted by maynardGkeynes | Report as abusive

For the most part I found this to be an interesting article, but I was surprised to read you say a degree does wonders for a graduate’s lifetime earnings when you wrote an article less than a week ago questioning that very idea. ( 009/09/05/the-value-of-a-college-educati on/)

why the change of heart?

Felix, you seem to be arguing in circles. It simply can’t be the case that graduation rates at Eastern Michigan are low despite the fact that the students are poor, while at the same time graduation rates at the University of Michigan are high because the students are rich, which you seem to suggest. It’s an issue of socioeconomic status, cultural standards and adequate primary education; there’s nothing magical about Ann Arbor over Ypsilanti.

Posted by Michael | Report as abusive

Felix, you have some great comments here; how about a response?

My own comment is:
>> the presence or absence of a campus makes a big difference: students are much more likely to drop out if there isn’t one.
>> if most of your peers are dropping out, you’re more likely to follow suit than if nearly all of them are graduating.

I don’t think you can support these statements, but would appreciate it if you had some evidence here.

Posted by Curmudgeon | Report as abusive

The chart is seriously misleading. It is dividing the cost in 1950 and 1980 by the current median income not the median income in 1950 and 1980.

Thanks to the readers who pointed out the error in the chart. I have fixed it in the original post (link here: 9/09/a-brief-explication-of-the-problem- of-rising-us-college-tuition/). The revised chart, to me anyway, still says that college is radically less affordable than it was in 1980 or 1950–but apparently I got a way better deal in 1980 than I had realized!

regards, John

It’s a “natural” result of growing income disparity in this country. The artificial shortage of those with 4 or 6 year degrees has caused their services to increase in cost faster than the median income. Maybe eventually we’ll have enough people with degrees, so that the wages wouldn’t be so inflated.

Posted by Markus Robinson | Report as abusive

I’m sorry, Felix, just where are you going with this? Or have you moved on? I think there is something relevant here, if you would just, well, comment.

Posted by Curmudgeon | Report as abusive

Here’s a thought: students who transfer to other schools are listed as “dropping out”. When I transferred schools in undergrad, my original school never wrote me asking why I didn’t show up that fall, didn’t ask if I transferred elsewhere, and certainly didn’t follow up several years later to find out if I graduated elsewhere. I bet that a sizable percentage of the “dropouts” are transferees. Here’s a question: how many undergrads transfer into U of M every year, and how many of those come from EMU?

Posted by Incentives | Report as abusive

Eastern’s graduation rate is due to a high percentage of non-traditional older students, a high percentage of students who transfer in to EMU and a high percentage of students who must work to support their educations. Comparing the U of M to EMU is a ridiculous comparison.

Posted by Ralph Pasola | Report as abusive

I have to agree with Michael. More selective colleges admit students who in general are more likely to be ready for the academic standards of that college. These colleges have the ability and luxury to filter out students they don’t think will make it through in 4-5 years. Less selective institutions have to deal with a much wider range of preparation. If these institutions conduct remedial classes (as many community colleges do), they know that they will have a low success rate at high cost to the institution. Less selective colleges have a more difficult educational task ahead of them vs. the more selective colleges, which can let its students run on autopilot, so to speak. Why do you think professors at research universities (like UM) have such terrible reputations as teachers?

Tuition is an important factor for a student choosing where to go, and would involve socioeconomic factors very directly. Here’s a comparison of EMU and UM tuition:

EMU, resident tuition, undergraduate: $238.50/credit hour; EMU estimates tuition and fees cost for freshmen as $4188.50.

UM, resident tuition, undergraduate: $805/first credit hour, $449/add’l credit hours; full time enrollment for lower division undergrad (12-18 credit hours) $5735 + 95 in various fees. The total increases by about $700 per term for upper level undergraduates.

Either is expensive, but EMU would be an easier choice for someone without a large pool of financial and social support.

Another factor to consider for students who must work to support their education is to what degree institutions accommodate people who work during the day and/or have to get childcare. Institutions like EMU and community colleges have much more available in those regards than more selective places like UM or the University of Iowa, where I teach. Getting an undergraduate degree at either one of these places would be extremely difficult to do for someone with a full-time job. The undergraduates I have who are returning students leave full-time work when they get to upper division courses, which are scheduled at the convenience of the college and faculty, not the student.

Posted by Rosemary | Report as abusive