Wednesday, July 13, 2011

MBA earnings, cost, and debt

If you're reading this in an RSS reader you may have to click through to see the graphs.  Email me or leave a comment if this is an issue for you so I can debug 

Poets and Quants just posted some data on the long-term salaries and earnings of MBAs by school.  Conveniently, seven of the top eight school by 20 year earnings were the "top 5" schools I've called out.   So one might guess that there's a relationship between a school's rank and its students' earnings, which would be logical.

But I'm less concerned about rankings and more interested in return on investment.  So I threw together some quick graphs that look at earnings against cost per year of the program and average student loan debt at graduation.  Do you see any trends?









On all three of the earnings data points (Total earnings over 20 years, Median pay 20 years out and Median pay 10 years out), Harvard comes out on top.  Sometimes by a little, sometimes by a lot.  Stanford and Wharton are Harvard's closest competitors.  So congrats to all the Harvard folks out there, you're in the money.

What also struck me though is that neither cost or average debt were significantly correlated with any of the earnings metrics. Debt correlated best with median pay 20 years out with an r^2 of .05, not linearly related at all, and was worse for the rest.  Total cost fit better with an r^2 of .6 for median pay 10 years out and around .2 for the two other earnings metrics.  But basically it's hard to say you get what you pay for with an MBA.  

Instead, average ranking was a better fit with an r^2 of .6 for median pay 20 years out and .5 for pay 10 years out and earnings over 20 years.   

 
Kellogg is a big exception here, but overall rankings may actually have some predictive value.  Better than I thought certainly.  So maybe I need to revise my idea that Harvard and Stanford are the best MBA deal and just go out and say Harvard is the best deal for an MBA.    Though the difference isn't as significant as it looks in these graphs since the scales have been changed to fit the data, on a scale from zero the differences would look a lot smaller.

What do you think?  What trends do you see in this data?

If you like this please Link Back to this article...



Related Posts by Categories



7 comments:

  1. I couldn't see the graphs in Google Reader.

    ReplyDelete
  2. A sample size of 8 schools is pretty small. Do you think this correlation will still apply for lower ranked schools? What if you expanded the data to include 50 or 100 schools instead?

    ReplyDelete
  3. Danielle - Thanks for the feedback. I'm not sure what the issue is but I noticed the problem in Google Reader as well. Hopefully I can sort it out.

    Anyone have it happen outside of Google Reader?

    Anonymous - 8 is definitely a small sample size, but my guess would be that earnings would on average decrease as you went down the rankings. In general, I don't pay much attention to schools below the top 20 since the ROI gets sketchy. There's more thoughts on ROI and ranking in this Poets and Quants article:
    /2011/06/27/when-is-an-mba-worth-it%E2%80%94when-is-it-not/

    FWIW, I didn't mean this post as any sort of robust analysis, just wanted to throw the data together and see if anything stuck out. I think it might be hard to do a more robust analysis since a top school is so different from a school ranked around 100 and a lot of it is based on intangibles.

    ReplyDelete
  4. Two observations:

    1) The Harvard brand rocks. It just does.

    2) Haas is an "also ran" with Stanford. It's kind of like how Pepsi sells a ton of soda, but Coke still cleans their clock.

    Also, there's probably some bias in the degrees at each university. Wharton, for example, probably has more finance MBAs than Stanford. Stanford's MBA grads are more likely to start their own businesses, which, assuming they do well, brings up the average, but hardly the median earnings.

    Very cool post, NDM.

    ReplyDelete
  5. JT - I was thinking it'd be cool to see the 25th and 75th percentile data for each, standard deviation and mean. Then you'd really see the effect of start ups vs financiers vs non profits. I'm betting Wharton has a lower variance than Stanford or HBS for example.

    ReplyDelete
  6. What does the average mba from these schools make compared to the average school where they charge just as much to go to a lesser named MBA program? Why go to a lower ranked school full time?

    ReplyDelete
  7. LAL - Here are some reference points for median starting salary at big state schools:
    University of Virginia (Darden) - $100,000
    University of Minnesota (Carlson) - $90,000
    University of Maryland (Smith) - $$85,625
    For more data points see:
    http://www.businessweek.com/interactive_reports/bs_2010_US_FTMBA_TAB_1111.html

    People go to lower ranked schools (my guess) because they can't get in elsewhere or they are offered merit aid. The ROI is sketchier but with significant aid or an employer sponsorship it may still make financial sense. However, I would bet that there are also students at these programs who haven't seriously looked at ROI or aren't concerned with it.

    ReplyDelete