The Economic Value of a Law Degree (part 1 of about 5)

You may also like...

6 Responses

  1. Derek Tokaz says:

    “We also cannot determine the earnings premium associated with attending a specific law school.”

    Is this not a serious limitation of the study that deserves more attention? Prospective students are not faced with the general prospects of an at large JD — they are concerned with the prospects at a specific school and with a specific cost.

    I fear that some will try a bit of logical sleight of hand by saying “A JD is a good financial decision; this is a JD; therefor this is a good financial decision.”

  2. Ben Barros says:

    Re: Derek’s question, I agree that per school information would be useful, but it would be very complicated to do. You’d not only need to track earnings over time by law school, but the baseline earnings for the candidate pool for each school. This is one of the many things Elie missed in his rant – sure, a Yale J.D. is probably more valuable over time than a Western Michigan J.D., but the baseline earnings potential of a person who gets into Yale is going to be higher than the earnings potential of a person who gets into Western Michigan.

  3. There is likely a reasonable working correlation between lifetime earnings and school rank (i.e., a smaller percentage of JD grads will earn BigLaw salaries at lower ranked schools). So the study does in that sense track this because it looks at income at different percentiles. And it still demonstrates that for a significant majority of JDs law school is a positive NPV proposition.

  4. Lawrence Cunningham says:


    Valuable addition to the discussion! Keep up the good work.

    And welcome to Co-Op; hope you enjoy your visit. Looking forward to further posts.


  5. Did we really need a study to show that law school was a good investment for people who graduated prior to 2008? Was that ever up for debate? Isn’t the debate actually about the relative value of a law degree specifically for people who graduated, oh, 2006/7 and after, and specifically for people considering law school now? If so (and yes, that’s so), shouldn’t it be important for the authors of this paper to insert something like “PAST PERFORMANCE DOES NOT GUARANTEE FUTURE PERFORMANCE” in their analysis? I’d think so, because these statistics (even if they were applicable in a post-2008 legal industry, which they aren’t) have no predictive value at all. Also, the caveat about “unobservable” characteristics is an exception that swallows the whale. (I mixed that metaphor myself.) It is absolutely impossible to control for variables between populations with similar undergrad courses of study but different post-bacc paths.

  6. Tyler,

    If you review the paper and power points I think you will agree that the authors take your issues into account to the extent the data is available.

    The substantial positive earnings premium they find is over the lifetime of a JD holder. Further it is cyclical. The cycle went down – at a rate less steep than in the wake of the dot com crash – and has now turned back up, as it has over the past two decades.

    That upturn in the cycle is not likely to be materially impacted by lower earnings of recent graduates even if those are lower than has historically been the case. The weight of their lower earnings in the first few years after graduation is not enough to turn that positive up turn negative, although I would defer to the authors to confirm this conclusion.

    The other issue you raise is the question of the predictive power of past experience. It is true for example in the financial markets that hedge funds have blown themselves up hoping for a reversion to the mean when in fact something new is happening. However, again deferring to the authors to weigh in, my understanding of their analysis of this issue is that the legal market has shown remarkable stability and has features that are very resistant to change when it comes to human capital.

    I study the economic institutions of financial markets, particularly market microstructure, where assumptions are being made about the impact of CEO statements, for example, on stock prices at the millisecond level. There, indeed, there is real danger in making the kinds of assumptions about the past as prologue. But I think the claim is that the legal market is far stickier.