Missing Institutional Heritage

Some think tanks style themselves “universities without students,” but most are what I call “mission-driven” institutions.  A mission-driven think tank produces (ideally rigorous) public policy research.  But it is also supposed to stand for something: a particular vision of a just society.  Last week’s high-speed train wreck at the Heritage Foundation should give pause to every supposedly mission-driven think tank that has a neglected or under-defined institutional mission.

My former Harvard classmate, Jason Richwine, resigned from Heritage last Friday amid public outcry over his 2009 doctoral thesis, “IQ and Immigration Policy.”  In it, he argues that Hispanic immigrants as a group have a lower average IQ than do U.S. citizens born in the U.S., and he recommends that IQ tests be administered to prospective immigrants as one criterion for admission to the country.

Some immigration reform supporters imbibed more than a little Schadenfreude as the news cycle imploded.  Not I.  I remember Jason as a generous and genial man who refused to accept compensation in return for patiently tutoring me for an analytical methods qualifying exam in 2007.  I find Jason’s research interests disturbing and his policy recommendations wrongheaded.  But he is not an unkind person, and I feel badly for him and for his family.

I have neither time nor inclination to digest a stack of research about what Richwine calls “the construct of IQ” in order to develop an informed opinion about the quality of his empirical work.  Reports are mixed.  Tufts Professor Daniel Drezner is unimpressed, while Harvard Professor Richard Zeckhauser, who served on Richwine’s dissertation committee, calls it “careful.” (Disclosure: Zeckhauser is a former professor of mine.)

But even if Richwine’s empirical claims were well supported, I would find his policy proposal—that aspiring immigrants to the U.S. should be admitted or denied based in part on their IQ scores—philosophically objectionable for reasons that could well indicate that Richwine is a bad ideological fit for Heritage.  Responding to Slate’s Dave Weigel, Zeckhauser usefully distinguishes the empirical from the normative:

“[M]y view is that none of his advisors would have accepted his thesis had he thought that his empirical work was tilted or in error. However, Richwine was too eager to extrapolate his empirical results to inferences for policy.”

Policy recommendations necessarily depend on a combination of empirical findings and ideological commitments about the nature of a just society.  From the perspective of justice, IQ importantly differs from acquired job skills.  Job skills reflect individuals’ choices to improve their life prospects.  It is certainly, sadly, true that not everyone has an opportunity to acquire such skills—a consideration in favor of welcoming at least some low-skilled immigrants.  Nonetheless, a policy of favoring prospective immigrants who have job skills is more intuitively defensible than favoring some over others based on one narrow measure of undeveloped talent.  Richwine’s proposal reasonably offends people who believe that our laws—even our immigration laws—should not deny important life opportunities to individuals based on prejudgments about their inborn capacities.

Heritage is a mission-driven think tank—it exists to promote a specific, conservative vision of a just society, and to develop policy proposals that are consistent with that vision.  Success in this mission depends critically on hiring scholars who share Heritage’s animating values.  If Heritage is institutionally committed to a conception of equal opportunity that precludes legal discrimination on the basis of inherent characteristics rather than choices and accomplishments, then Richwine is a poor ideological fit there.

But if this is the correct story, then Heritage should not have hired Richwine in the first place.  Opportunistic firing (or even “resignation acceptance”) practices don’t somehow excuse careless hiring practices.  Indeed, both are costly to wonks and to the institutions they inhabit in multiple ways.

Most obviously, a supposedly mission-driven think tank with no clear idea about who fits under its ideological tent doesn’t really have a mission.  I don’t mean to suggest that a think tank’s mission should be narrow, or that its scholars must agree on everything.  On the contrary, a think tank that demands lockstep agreement will never retain smart scholars.  But if a think tank has a handful of truly defining commitments, those commitments should be clearly articulated lest the institution become media-driven or money-driven in lieu of advancing its mission.

Moreover, mission-driven think tanks that lack clear standards of inclusion—a category that includes nearly all of them—can be fearful places to work.  One need not dabble in the dark arts to find oneself suddenly unemployed for straying too far from an ideological reservation with invisible fences. How far is too far?  Most of the time, wonks find out only when they are handed their walking papers, asked to abandon cherished projects, or pressured to recant unacceptable statements.

Fear of incompetency can promote research excellence.  Fear of being branded a heretic often precludes it.  Most policy wonks live with too little of the former kind of fear and far too much of the latter.  I believe that many mission-driven think tanks could achieve a better balance by reforming their hiring and firing practices simultaneously, bearing in mind a few general guidelines:

Clarity about shared ideological commitments – Think tank mission statements are usually lofty-sounding but vague.  If a think tank has a set of core ideological commitments about the nature of justice and the proper role of the state, these should be specifically articulated somewhere, so that both financial supporters and scholars have clear expectations when they decide whether to invest time or money in the institution.

No mandatory empirical beliefs or policy conclusions – Insofar as a think tank holds itself out to be engaged in public policy scholarship as opposed to mere issue advocacy, it cannot be committed to predetermined findings and conclusions.  These are the results of good public policy research, not the premises.  A wonk who fears learning the “wrong” thing will fail to learn anything at all about the world, and her think tank will fail to make a meaningful new contribution to the public policy debate in which she participates.  Unhealthy, research-spoiling fear squanders scholars’ talents and supporters’ money.

Greater emphasis on empirical skills – Good empirical skills are the best preservative of truth under pressure. Public policy scholars often add value by developing new public policy proposals, an essentially interdisciplinary task that requires empirical findings, ideological commitments, a savvy awareness of the location of the Overton window, and a working knowledge of any potential legal hurdles to implementation.  Policy wonks do not necessarily need to be able to produce peer-reviewed social science research, but I’m going to go out on a limb and maintain that they should be able to understand and evaluate it if they address empirical questions in the course of their work.

Term contracts for policy scholars – I understand why mission-driven think tanks don’t grant tenure.  Although adults’ ideological commitments are usually fairly stable, they can shift over time, sometimes dramatically.  But the hasty departures that accompany media firestorms ill serve both think tanks and scholars.  I wouldn’t blame Heritage at all for concluding that Richwine’s personal views are inconsistent with Heritage’s principled ideological commitments.  Indeed, they are inconsistent with mine.  But what we saw last week was a desperate damage control effort, not a sober reevaluation of Heritage’s staffing profile. Ubiquitous at-will employment arrangements in the think tank world diminish research quality and creative thinking by fueling the debilitating kind of fear.  They also encourage a long-term pattern of scandal mongering and sometimes-feigned outrage among partisans and in the press.  Replacing at-will agreements with standard two- or three-year employment contracts could paradoxically empower think tank executives to make thoughtful personnel decisions by tying their hands until the smoke clears.

Heritage was humiliated last week because its leaders failed to make personnel decisions thoughtfully in accordance with a clear institutional vision.  Let it serve as a cautionary tale for other mission-driven think tanks.  In a world of clear—not narrow!—standards, policy scholars could determine with more confidence where their own ideological commitments make them a good fit.  In a world without them, we can expect more embarrassing departures and, I fear, more subtly compromised research.


5 Responses to Missing Institutional Heritage

  1. Brandon M. May 14, 2013 at 2:17 am #

    Thank you for this thoughtful perspective. I’m currently a second-year public policy student at the Kennedy School and many of us here are very upset that IQ-based policy recommendations in Mr. Richwine’s dissertation were deemed worthy of meeting the standard to receive a PhD at HKS. Would you mind sharing your perspective on this point?

    • Marie Newhouse May 14, 2013 at 10:43 am #

      Brandon, thanks for your question. First of all, you should know that the Harvard Kennedy School does not run the public policy Ph.D. program. The Ph.D. program is actually governed by Harvard’s Graduate School of Arts and Sciences.

      Regarding the substance of your question, I think Professor Zeckhauser said it best when he pointed out that policy dissertations often move too carelessly from empirical findings to policy recommendations. As Zeckhauser pointed out, policy recommendations require philosophical commitments in addition to mere facts. Although I haven’t read Jason Richwine’s dissertation in full, it seems likely that he provides no rigorous philosophical argumentation in support of his policy proposals. If he does not do so, however, he is not unusual. Very, very few public policy dissertations include rigorous philosophical argumentation.

      This would be a great discussion topic for HKS students to bring to their required ethics classes. I know that many HKS students consider the ethics course a boring waste of time, but it is the *only* required course that exposes HKS students to the kinds of philosophical arguments necessary to make the value judgments which are part and parcel of policymaking. Professors Applbaum, Kamm, and Risse can offer some wisdom about how to separate out the empirical and normative questions related to public policy formation, and about how to approach the normative questions in a rigorous way.

  2. Anonymous May 18, 2013 at 7:12 am #

    I would just note that it seems in very poor taste to suggest that the problem with this dissertation is not empirical, purely on the basis of Professor Zeckhauser’s statements.

    What would it mean if Richwine’s dissertation was “careful” empirically? Would that mean that Richwine carefully examined the empirical data on IQ? And that his work, purely on IQ, was “careful” and, as Zeckhauser implies, that it was not “tilted or in error”?

    If this is what Zeckhauser means then, we can presume that Zeckhauser thinks that Richwine’s findings about IQ are sound: that hispanic immigrants have a lower IQ than “native” whites, that IQ is a reliable measure of general intelligence, and that the difference between the IQ of hispanics and whites is likely “genetic.”

    So, is Professor Zeckhauser (and by repetition, are you) implying that Richwine has a very well-reasoned scientific argument for why hispanics are genetically predisposed to be less intelligent than whites? Is that the kind of argument you want to repeat from Zeckhauser without having read the dissertation?

    The fact that Richwine was “a nice guy” (as you point out in your article) is hardly relevant to the fact that he’s advancing racism in the name of science. The weaknesses with Harvard’s Public Policy PhD program clearly run deep indeed if it’s strengths are about “facts” and not about philosophy, because any rigorous research into general intelligence would find Richwine’s claims all but unsupportable. In his dissertation he doesn’t even mention some of the most damning counter-arguments to his interpretation. For example, that “hispanic” is not a genetic group. How can one come to the conclusion that a group of people who aren’t defined by genetic similarity score poorly on a test because of their genes? What I’m saying is that you don’t have to be an expert in anything to see that this is flimsy, and the tiniest bit of up to date research on the purported genetic basis of these things would have revealed an inexhaustable wealth of literature that we could basically call a scientific consensus: IQ has historically been an unreliable test; one that may reference little or nothing of relevance; one that has been applied in culturally-specific ways, rending all historical statistical data on IQ tests unreliable.

    None of this is a fringe idea. This is the state of IQ: a historical embarassment. If Richwine had wanted to make the claims he did in an “empirically careful” way, he should have, at the very least made mention of these damning critiques of IQ, of the fact that the scholarly consensus among experts in cognitive science is against IQ, etc. And if, in spite of addressing all those overwhelming critiques, Richwine had then come up with some brilliant reason why his racist ideas were actually correctly, and the consensus of real scholarship was just politically correct posturing, then and only then might we even begin to think of the word “careful” when we describe his work.

    So the question is, why are you ready to stand behind that assessment on your blog without even having read it? Zeckhauser may have been your Professor, but that’s no excuse for repeating what is essentially a scientific stamp of approval on Richwine’s most ridiculous, racist claims.

    To put it another way, which of these two things is more racist:

    1. Hispanics are genetically predisposed to be less intelligent than white people, and this will likely continue to remain the case indefinitely. Translation out of scholar speech: Mexicans are just born dumb.

    2. We should block the immigration of people who have a genetic predisposition not to be very smart. Translation out of scholar speech: We shouldn’t let stupid people become Americans.

    Number 1 is Richwine’s “empirical” work. Number 2 is his policy claim. Please, please, do not parrot some feeble defense of point 1.

    • Wonks Anonymous May 22, 2013 at 10:27 am #

      I haven’t read his dissertation myself, but I have seen excerpts where Richwine noted that “Hispanic” is not a racial group. I also read excerpts in which he said it wasn’t very important for his policy conclusions the extent to which genetics plays a role.

      “the scholarly consensus among experts in cognitive science is against IQ”
      That would be a surprise to me, do you have a cite?

  3. Marie Newhouse May 19, 2013 at 6:48 pm #

    Anonymous, I didn’t express agreement with Professor Zeckhauser regarding the quality of Richwine’s empirical work. Take another look at what I wrote above:

    “I have neither time nor inclination to digest a stack of research about what Richwine calls “the construct of IQ” in order to develop an informed opinion about the quality of his empirical work. Reports are mixed. Tufts Professor Daniel Drezner is unimpressed, while Harvard Professor Richard Zeckhauser, who served on Richwine’s dissertation committee, calls it “careful.” (Disclosure: Zeckhauser is a former professor of mine.)”

    I wrote that I had no opinion about the quality of the empirical work, I observed that reports from people in a position to know about it are “mixed,” and then I illustrated the mixed nature of the reports by linking to Drezner, who says the work is unimpressive, and Zeckhauser, who says it is careful. I don’t know enough about the subject to have an opinion. Drezner may very well be right, and Zeckhauser may very well be wrong. I don’t privilege either assessment in this post.

    The thesis of my subsequent paragraph is that Richwine’s policy recommendation is a bad one for philosophical reasons, regardless of who is right about the quality of the empirical work.