Is Political Diversity Necessary at the U?

Happy Mother's Day!

And thank you Nicholas Kristof for writing a column that has nothing to do with the campaign for President.  Kristof's column is about Liberal bias at the university.  I want to take on some of the arguments there.  But before I do let me note that Kristof's fellow columnists at the New York Times who write for the Sunday Review:  Frank Bruni, Ross Douthat, and Maureen Dowd, each wrote about Donald Trump today.  This seeming lack of diversity in topic may be justified by the current political moment.  But as a reader my view is, enough already, can't we move onto something else?

One of the disappointments that seems to have come out of this campaign is that the need to attract eyeballs is a big driver of the news and opinion that gets published in the major outlets.  Diversity of topic may indeed be necessary for the inquiring mind.  But harping on the same old thing over and over is better for attracting eyeballs, as the masses can't get enough of something with a prurient interest.

Let me turn to Kristof's piece. He notes that my discipline, economics, has both Liberal and Conservative practitioners.  So I want to start there and focus on that.  Does the economist's politics matter - for the formal research that is done, the classes that are taught, and for how the economist interacts with non-economists, within the university and in the general public.  If politics does matter, is the effect first order or second order?  (Second order effects can be ignored in a first pass at considering the question.)

If you look at sub-discipline, politics almost surely matters more in macroeconomics than in microeconomics.   When I was in grad school, there were two schools of thought in macro, Cambridge versus Chicago, or Keynesian versus Monetarist.  My macro instruction at Northwestern was largely Keynesian.  In the first quarter there my teacher was Robert Eisner and in his class we read The General Theory.  We also read several articles on the consumption function, including Modigliani and Brumberg's work on the Life Cycle Hypothesis and Milton Friedman's work on the Permanent Income Hypothesis.  While at the time of Eisner's course I may have divined the important empirical distinctions between the two approaches, at this point my memory of that is dim.  All that survives is that Modigliani was at MIT (and before that at Illinois where I believe Brumberg was his student) while Friedman, of course, was at Chicago.

Did Friedman's work get its due in Eisner's class?  If so, did his politics otherwise matter in a first order way?  Many of my classmates had a strong reaction to Eisner, but not because of his politics.  Rather it was because he gave very tough exams, something they weren't used to, and he conducted the live class session in a style akin to Kingsfield in the Paper Chase.

Indeed, for whether politics matters in teaching the question seems to be whether the instructor teaches the class in a thinking gray sort of way.  I wrote about this a while back in a post called On Social Issues Is There Ever A "Right Answer"....  The relevant passage from that post is reproduced below. 

The best articulation of the principle I’ve seen is by Steven Sample in his book The Contrarian’s Guide to Leadership. The first chapter is on Thinking Gray, which means several things all at once. First, don’t make a decision before you have to and don’t tip your hand as to how the decision will eventually come out to encourage others to provide you with evidence that you will weigh fairly. Second, actively encourage argument and debate about the decision so different points of view can be well articulated. Third, while the first two are really external behaviors this one is truly internal to yourself. It’s not that you have a quickly formed opinion that you are not sharing because of the first two reasons. It’s that you maintain neutrality on the issues until when judgment is needed. You do this so you can make the best and therefore unbiased judgment when it’s time for that. As Sample says, this is contrary to the way most of us behave because we’ve been taught to make snap judgments.

F. Scott Fitzgerald once observed something similar to thinking gray when he observed that the test of a first-rate mind is the ability to hold two opposing thoughts at the same time while still retaining the ability to function.

If the instructor does take a thinking gray approach to the subject, does the instructor's politics matter nonetheless?  If so, why?  I would guess that politics would only be a second order thing in this case. This recasting of Kristof's argument then might be - many instructor's don't take a thinking gray approach.   Let's say that is true.   The question is why.  I am thankful that most of my instructors at Northwestern were first-rate minds.  Eisner definitely was and I do think he gave Friedman's essay on the consumption function its due.  If other college instructors who don't embrace thinking gray do that because they are not first-rate minds, then we have something else to worry about beyond their politics.  I'd argue that something else is the larger concern. 

I've had rather few situations as an instructor where politics mattered in a way that was obvious to me.  The one time where it really mattered was in spring 2011 in a course on Behavioral Economics.  I wrote about that in a post called, On the role of economic rationality in teaching undergraduate economics.   I believe I took a thinking gray approach in that class, but many of my students were having none of it and it was the one time where I thought the class got away from me because of the content I had chosen to cover.   Based on that experience alone, I would argue that something else is missing in Kristof's piece.  That is the political orientation of the students.  On my campus at Illinois, Business students and Economics students tend to be Conservative.  Perhaps at NYU where Jonathan Haidt is these students may be more Liberal.  But if the students are Conservative and the instructor is Liberal is that diversity enough?

Actually, however, I would like to point out the unintended consequence of that diversity.  I gave up on the Behavioral Econ course because I didn't want a repeat experience of the class getting away from me.  Instead I've taken a safer route where the political issues largely don't manifest.  (Really they do but in a way that is more invisible to the students.  If you read my post, The Liberal View of Capitalism, you will see how the politics does show up.  I teach both approaches, but indicate to the class my prior disposition to Akerlof's Gift Exchange model, a collegiality based form of motivation on the job, as distinct from a performance based form of motivation. )

In general, an unintended consequence of the sort of diversity that Kristof argues for is to move to only safe topics where the diversity doesn't matter.  The controversial stuff gets snubbed, because there is a need to be functional and carry on as normal.  The idea that there will be ongoing and vigorous debate of the controversial issues in a way that will not harm any of the participants seems to me something of a pipe dream.  And every time there is such a debate that gets out of hand, it serves as incentive to move to safe ground the next time around.  There is thus some burden to show examples of the ongoing vigorous debate approach, before arguing that political diversity will provide a cure.

Let me take on one more point in the Kristof piece.   This is about empirical research done on political bias in academe.  He writes:

A study published in The American Journal of Political Science underscored how powerful political bias can be. In an experiment, Democrats and Republicans were asked to choose a scholarship winner from among (fictitious) finalists, with the experiment tweaked so that applicants sometimes included the president of the Democratic or Republican club, while varying the credentials and race of each. Four-fifths of Democrats and Republicans alike chose a student of their own party to win a scholarship, and discrimination against people of the other party was much greater than discrimination based on race.

I find evidence of this sort not very relevant at all on the question of whether the bias is first order or second order.  The reality is that resume information is very limited in what it says about the person.  It does matter, of course, for things like admission.  But if I know a student, either because the student took a class from me or because I am mentoring the student, will the student's politics matter to me in my evaluation of the student?  Likewise, I was a campus administrator and have had many staff members report to me.  Did my evaluation of them depend on their politics?  Or was it mainly about their intensity and competence in doing the work coupled with their good judgment?

In trying to recall incidents that illustrate, I thought of when I was directing the Center for Educational Technology and one of the staff sold Girl Scout Cookies at the office on behalf of her daughter.  I have no idea what the employee's politics were/probably still are,  but this behavior was inappropriate and bothered me, because it is moderately coercive and other employees shouldn't have to face that at the office.  When an employee makes one poor judgment, perhaps you give the person the benefit of the doubt.  After a few of them, however, you do put such a person in a box.  Ditto for students.  And I'm sure that from time to time students put their instructor in a box as well.  Maybe politics matters for this in a fundamental way in some cases.  Mostly, however, I think it is secondary.

Chicago does hire MIT or Harvard PhD's from time time and vice versa.  There is some clustering of like minded views on the politics, but some diversity as well.  When I was involved with Econ recruiting (this is now more than 20 years ago) candidates were evaluated by their research potential - regarding how much they'd produce and the likely placements of that output.  That was the sole criterion then.  The candidates, in turn, wanted to know whether they'd have somebody to talk to about their research.  I suspect it is not that much different now.  If you have thinking gray researchers they can accommodate diverse scholars as a matter of course.  If you don't, encouraging political diversity as an additional criterion for hiring looks to me like you're asking for trouble.

Kristof likes to be a Liberal contrarian from time to time.  That is welcome.  But on this one maybe he should write another column about skepticism from others in academe about promoting a political diversity agenda.