Wednesday, July 24, 2013

The $1,000,000 Law Degree Kerfuffle: Getting in Touch With My Inner Kahneman

Posted by Jeff Lipshaw 

ValuefunUntil the phrase "$1,000,000 law degree" started filling my RSS feeds, I had paid about as much attention to the issue as I did to the Royal Baby (which, by the way, a new study will show shortly has a .00000000112% chance of being named "Jeff" but is twice as likely as that to be named "Geoff").

Something Brian Tamahana wrote in his most recent post caught my attention, however, so I went back and skimmed Mike Simkovic's paper just to confirm what I'm about to observe.  Brian asked in so many words, knowing this data, even if accurate, would you advise somebody to go to law school?

Funny, because I have two children in their mid- to late twenties, both of whom are in the midst of making a job or career change.  I occasionally joke about their going to law school.  (The older one had any desire to be a lawyer whipped out of her by way of the year she spent as the typical "just out of Ivy League school - going to go to law school next" litigation paralegal in a mega-mega NYC based firm that will remain unnamed.)  I think I said, "Only go if you go to Harvard, Stanford, or Yale."  That reflects mostly elitism and arrogance on my part, and not my inner Kahneman, which I've gotten in touch with, but unless I've missed something (let me know if I have) I don't see in the discussion that anybody has gotten in touch with theirs.

 The point is that even if we take all of Mike's data as saying what he purports it to say, it still doesn't say anything about how people look at the prospects of gain (income) and debt (loss).   I'm a skeptic about whether understanding prospect theory actually helps you make a decision (i.e., if I understand my own heuristics, will that counter the bias and cause me to calculate my expected utility - I don't think so), but I don't think you can debate this issue with at least acknowledging that people don't make decisions involving prospective risk or loss by stepping back and viewing the final outcome - the expected utility - over their entire lives.

I will let others, if they want, spend more time explaining the relevance of the above graph to the issue, but it has to do with how being risk-adverse or risk-seeking affects your decision depending on whether you are faced with high probabilities of losses relative to gains, or high probabilities of gains relative to losses.  (The exercise I do every year with my students is take a secret ballot vote on whether they would prefer $1,000 in cash or a one in ten chance of $10,000 at the end of class - risk aversion being such that invariably close to 100% chose the former even though the expected utility is exactly the same.)

I've been noodling around with my update to How Not to "Retire and Teach," and have been thinking the odd hybrid of explanation and advocacy that arises when lawyers argue either about what is or what should be.  The fancy phrase is that all knowledge beyond pure perception is theory-laden; the equivalent is "lies, damned lies, and statistics."

https://lawprofessors.typepad.com/legalwhiteboard/2013/07/the-1000000-law-degree-kerfuffle-getting-in-touch-with-my-inner-kahneman.html

| Permalink

Comments

Jeff, I did the same thought experiment and came to the conclusion -- which, if I'm wrong, I'm hoping someone will correct -- that the study provides only one secondary piece of data for the person I'd be counseling about deciding on law school.

Posted by: John Steele | Jul 25, 2013 11:14:53 AM

" (The exercise I do every year with my students is take a secret ballot vote on whether they would prefer $1,000 in cash or a one in ten chance of $10,000 at the end of class - risk aversion being such that invariably close to 100% chose the former even though the expected utility is exactly the same.)"

You're jumping from 'expected value' to 'expected utility'; this assumes a linear utility value for the money.

Posted by: Barry | Jul 27, 2013 5:33:30 AM

Also, people have a hard time with orders of magnitude and longer lengths of time. Their perception of what $1,000 will do for them is much clearer than their perception of what $10,000 would do.

In addition, this is a one-off event; expected value treats it as if it were recurring.

Posted by: Barry | Jul 27, 2013 5:36:17 AM

Post a comment