http://www.payscale.com/
I think the study's methodology is weak and does not prove causation. For the "return" part of ROI, they look at the median pay of grads for 30 years of working, adjusted to 2013 dollars. They project out pay increases by looking at grads from 1983 to 2012 (i.e. the earlier grads are later in their careers, and forecast what recent grads will earn at the same age). But of course that is fraught with problems, because universities could change a lot in quality and student composition over 30 years. And the economy changes a lot over time, so an older person may see very different wage changes from 1990-2010 vs. a recent grad from 2010-2030 (especially for degrees that saw a lot of industry growth or contraction). Also, I am not sure if they were looking at total earnings or just base pay, because we know that some jobs may have lower base pay but make up for it with better benefits, equity, or pensions. Students who went on to graduate/professional education were excluded, but that could be a problem too since some universities have a high % of students with those career aspirations (and those jobs tend to pay higher).
Also, are we measuring the ROI of the school, or the ROI of the students? University is not the only factor determining a person's future salary. Some Stanford students would still make similar money if they attended higher or lower ranked schools because of their abilities and ambition. But that is what the study should try to answer - if Joe Smith went to Cal vs. Stanford vs. Caltech and studied the same subject and pursued the same career path, how would his pay differ? This would be expensive and maybe unfeasible, but to control for student quality variation, they should have looked at each decile of US high school students (by general aggregated admission criteria), mapped out which college they attended, and then looked at their earnings over the first 5 years after graduation (30 years is overkill, entails too much uncertainty, and recent grad wages are a good predictor of late career wages anyway). But still, variation in regional pay and career choices would still confound the results.
This info would be tough to obtain as well, but possibly a cleaner way to answer the question would be to query employers. Do their compensation departments place a premium on various schools for certain job roles? Same with grad schools - do they give bonus points to students who went to undergrad at certain places (and convert that to a $ value)? Also repaying student loans has an opportunity cost, and there is value in the networking and relationships that one obtains in college. So it would be nice if they could incorporate those factors into their model as well. What do you think?
-----------
In this case, though, it's a little hard to separate out the reputation effects vs. the quality of education vs. network effects.
I.e.
do Stanford kids earn more money because we graduate with a good
education or because of the Stanford name premium or because we just
happened to make friends at a school where there is a lot of smart
people.
If it's because of the quality of education, then
you could say Stanford is worth it. If it's because of the network
effects, then maybe yes maybe no. An expensive networking program for
sure, but maybe not replicable by any other experience (and thus worth
it). If it's the reputation effect, then Stanford should be concerned,
as companies could potentially test for real skills some day,
undercutting the reputational aspect of our degree.
------------
I agree that those factors are heavily linked in such a data
set and it's pretty hard to draw any meaningful conclusions about ROI.
Another interesting question: are Stanford students successful because they went to Stanford, or is Stanford successful because of its students? I am sure it is symbiotic, but maybe that is why the top US schools have been locked in an "arms race" to recruit star profs, improve the student experience, etc. to woo the top talent. I guess they realize that a key to their success is the general quality of their "customer base" (but the spending is also partly irrational pissing contest). Though I am pretty sure that most students accepted at Stanford would become professionally successful no matter what university they attend (a testament to the merits of the admission process), even dropping out and going Zuck style. But if Stanford and say Ole Miss (sorry for offending any alum) do a "trading places" student body swap, I think that plenty of Ole Miss students would not be able to capitalize on the Stanford experience as effectively, leading to lower outcomes. And if the avg. quality of one's peers decreases, that may have negative effects on one's performance. But on the flipside, the sudden infusion of Stanford-caliber students to Ole Miss would have profound benefits for that institution on many levels.
So I guess college students should have much more bargaining power vis-a-vis university administrators. They could form a union haha and demand lower tuition or face a walk-out. But that threat is not credible, unless all of them can join startups the next year, or the Ivy League is able to take in the thousands of striking Stanford students en masse. Plus Stanford would replace them with eager "scab" students from US public schools and Asia - probably lower quality students, but "passable" to keep the place running. Therefore, it seems that the top schools are engaged in a certain degree of price fixing and cartelization (i.e. why is the tuition at Stanford pretty similar to that of Carnegie Mellon or Columbia, despite massive geographic and programmatic cost differences?). Unlike Wall St. and Si. Valley, they refuse get in a self-destructive price war, or steal away top talent from each other with cash/other incentives. Maybe that is illegal, but if absolute student body quality is the goal (as it pertains to national rankings or other metrics that administrator bonuses are based on), then it's worth it (and their endowments can justify it). But really, just imagine how that would affect recruiting if Stanford decided to slash its tuition by 1/3? 50% of undergrads are on some sort of financial aid anyway, so it wouldn't be such a shock to their finances. Well, no one ever suggested that higher education was an efficient market. :)
Another interesting question: are Stanford students successful because they went to Stanford, or is Stanford successful because of its students? I am sure it is symbiotic, but maybe that is why the top US schools have been locked in an "arms race" to recruit star profs, improve the student experience, etc. to woo the top talent. I guess they realize that a key to their success is the general quality of their "customer base" (but the spending is also partly irrational pissing contest). Though I am pretty sure that most students accepted at Stanford would become professionally successful no matter what university they attend (a testament to the merits of the admission process), even dropping out and going Zuck style. But if Stanford and say Ole Miss (sorry for offending any alum) do a "trading places" student body swap, I think that plenty of Ole Miss students would not be able to capitalize on the Stanford experience as effectively, leading to lower outcomes. And if the avg. quality of one's peers decreases, that may have negative effects on one's performance. But on the flipside, the sudden infusion of Stanford-caliber students to Ole Miss would have profound benefits for that institution on many levels.
So I guess college students should have much more bargaining power vis-a-vis university administrators. They could form a union haha and demand lower tuition or face a walk-out. But that threat is not credible, unless all of them can join startups the next year, or the Ivy League is able to take in the thousands of striking Stanford students en masse. Plus Stanford would replace them with eager "scab" students from US public schools and Asia - probably lower quality students, but "passable" to keep the place running. Therefore, it seems that the top schools are engaged in a certain degree of price fixing and cartelization (i.e. why is the tuition at Stanford pretty similar to that of Carnegie Mellon or Columbia, despite massive geographic and programmatic cost differences?). Unlike Wall St. and Si. Valley, they refuse get in a self-destructive price war, or steal away top talent from each other with cash/other incentives. Maybe that is illegal, but if absolute student body quality is the goal (as it pertains to national rankings or other metrics that administrator bonuses are based on), then it's worth it (and their endowments can justify it). But really, just imagine how that would affect recruiting if Stanford decided to slash its tuition by 1/3? 50% of undergrads are on some sort of financial aid anyway, so it wouldn't be such a shock to their finances. Well, no one ever suggested that higher education was an efficient market. :)
No comments:
Post a Comment