Suppose you are a high school senior with a lot of foresight and a bottom line outlook. Naturally, you would want to pick a college based on how much its degree will be worth in future earnings. Conveniently, there is a website called PayScale "a market leader in global online compensation data" that recently published a list of median salaries by college. They list both what graduates got right out of school and "mid-career median salary."
The methodology of the survey leaves a lot to be desired. Firstly, it is not a random sample, but based on the data entered by college graduates who signed up with PayScale. That happens to be rather a lot of people, but it is still hard to avoid the assumption that the data is skewed in some way.
Secondly, and maybe more importantly, holders of graduate degrees were excluded. As quoted in the New York Times’ Economix blog, Al Lee, PayScale’s director of quantitative analysis attempted to justify this.
“You’re thinking of buying a college. If that’s all you buy — an undergraduate degree –without having to spend more money and time and effort to get another degree,” Mr. Lee said, “you want to know what the return on that investment is.”
Except that a good reason to choose one school over another might be that it will help you get into a better graduate school so you could make more money later in life.
Quibbles with methodology aside, the results of the survey are approximately what you would expect. Five of the top ten by mid-career median salary are from the Ivy League sports conference. (Seven of the eight Ivies are in the top 30. Poor Columbia shows up at #52.) With few exceptions, the non-Ivies at the top are the usual suspects as well: MIT, Stanford, etc.
Okay, so obviously the prestigious schools make people richer later in life, presumably because they teach valuable things.
Not so fast. All this confirms is the fairly obvious conclusion that there is a correlation between salary later in life and attending a highly ranked college. Correlation is not causality.
Let me throw out a radical alternate hypothesis. There is no difference in the economic worth of curricula from one college to the next. High salaries in mid-career tend to be given to motivated and talented people. Those people were motivated and talented as high school seniors and the most able of them had a habit of attending the most competitive colleges. Later in life, if having a prestigious college on their resume helps them, it is only because it labels them as motivated and talented, not because potential employers think they learned anything there.
As upsetting as this view may be to some, particularly those few who take instruction at top undergraduate programs seriously, there’s not much evidence to the contrary.
And there is evidence in favor. Consider that roughly the same top schools have been the top schools for about as long as people have thought to compare them. Surely, if what the students learned was important then changes and innovation in instruction would have caused a meaningfully different rank order over the past hundred years. And how can it be that top liberal arts schools can refuse to teach such economically valuable subjects as accounting, IT, and medicine, as a matter of principle, and still stay at the top of the salary list?
So in terms of money, does it matter which college you go to? I think it does, but not by nearly as much as is widely believed and not for the reasons usually assumed.
One of the observations to be made from the median salary data is how small the difference between the top schools and not-so-top schools actually is in dollars. The mid-career salary difference between school #1 and #50 is $129K vs. $101K. That’s a real difference, but it is hardly life changing. And those are medians. A given high school senior might expect to do worse than the #1 median if he went there and better than the #50 median if he went to the less prestigious school. In fact, it is entirely possible that choosing between #1 and #50 will have no long-term economic impact at all.
I will not go so far as to say that it does not matter at all if you go to Dartmouth, which was #1, or Black Hills State University, which came in last at #597. But nobody makes that choice. Within a set of colleges that a particular high school senior is likely to consider, however, it’s not clear that future earning power should even be a factor.