Five myths about the policy-academy gap

My FP colleagues Steve Walt and Daniel Drezner have recently weighed in on a topic near and dear to my heart: the gap between academic political science and the policy world. This is a longstanding concern that has been the subject of much talk and some action my entire professional life. Since I am also a card-carrying yakademic (a term one of my National Security Council colleagues introduced to me back when I worked in the Clinton White House — he always smiled when he said it so I am sure it was a term of endearment), I sympathize with much of what they and numerous others (see here and here) have to say.

Yet I believe that the enduring discussion would be stronger if we let go of some myths.

Myth 1: Academics bring bias-free insights into a policy debate cluttered by biased policymakers. This is a myth that tends to be peddled by folks who have had a heavy foot in the academy and have only dipped a toe or two into the policy world. In my experience, academics are just as biased as, and in some cases more biased than, policymakers. Policymakers have a partisan bias, a worldview bias, and a policy commitment bias (eg., if a policymaker advocated for a policy initially he may be predisposed to support it despite evidence that it is faltering, and vice-versa).

Academics also have a partisan bias and a worldview bias. The chief difference is that academics tend to give jargon terms to their worldviews and they tend to reinforce them with an additional layer of method bias — that is, viewing the world primarily through the narrow lens of whatever methodological tool they have mastered in the academy. Academics also have policy bias — if they supported or opposed a policy, I have found they have tended to have just the same commitment to their prior stance as any policymaker.

Academics have a further source of bias that is not as prevalent in the policymaker world: a skewed marketplace of ideas where the people they spend the most time arguing with come from a very narrow slice of the ideological spectrum. Academic debates are vigorous and vicious, but rarely do academics have to try to persuade people from outside their small epistemic community within the academy (a self-selected pool within a self-selected pool), let alone someone from all the way on the other side of the policy spectrum. Put another way, many academics spend too much time in an echo chamber and as a consequence tend to develop the kinds of prejudices that they rightly decry when they see it in others (e.g., the view that people who disagree with them must be stupid or venal).

Myth 2: Academics bring theory, method, and factual expertise to policy debates, but only factual expertise is valuable. This is a myth that tends to be peddled by folks who have had a heavy foot in the policy world and have only dipped a toe or two into the academy. Thus, a policymaker might consult a regional specialist for insight into, say, the tribal complexities of Pakistani society, but his eyes will glaze over if the academic mentions theory or method. Those other things, I have heard too many policymakers say, are just esoterica that detract from real knowledge.

Most of my academic colleagues do have a great deal of factual expertise but not, in my experience, significantly more factual expertise than the top-line policy experts that populate administrations (Republican or Democratic). In fact, most of my academic international relations/security colleagues (that is people who do not focus on a specific country or region) do not have any firmer grasp on factual knowledge than the best people I have worked with in government; indeed, for reasons I get to below, they might even have less. What they can bring to the table, however, is a rigor that comes from making theoretical claims explicit and using appropriate methods to evaluate them. The very best academics also can bring to the table a healthy respect for the limits of their methods, and thus for the limits of the claims they and others can make. The chief difference in this regard between policymakers and academics is that policymakers use theory and method implicitly whereas academics use them explicitly (or at least they should).

Myth 3: Theory and method can settle many policy debates, so political scientists should be policy judges and juries. Some people get the message on Myth #2 and replace it with Myth #3, the belief that academic findings, precisely because they derive from rigorous and explicit attention to theory and method, provide dispositive answers to policy conundrums. Thus, the political scientist tells herself, I have researched a question using the right methods and gotten my results through peer review to publication, therefore my finding (e.g., that economic sanctions rarely by themselves cause a target state to capitulate) settles once and for all a thorny policy question (e.g., whether we should sanction Iran or offer a major concession).

Academic research is a good contribution to the policy debate, but it hardly settles the matter (and, in this case, it would hardly surprise the policymaker who knows full well that Iran is unlikely to capitulate through economic sanctions alone). Policymakers have to factor in many more considerations than even very thorough academic research tends to weigh, such as: the need to cultivate political support for a policy line; the need to demonstrate that all avenues, even unpromising ones, have been explored; the need to sequence and/or link measures that an academic study might treat in isolation (or try to "hold constant" in the analysis); the need to consider the possibility that the immediate case might be the exception rather than the rule; and so on.  

Even the most ideological policymakers I know tend to have more of a pragmatic streak in them than many of the ardent academic contributors to policy debates. Academics would fare better in the arena if they circumscribed the claims they make for the policy implications of their research.

Myth 4: The Problem is with a specific method, specifically whichever method I do not use. Any academic worth his or her salt can tell you the limitations of the methods they tend not to use. And all methods do have limitations. But, in my experience, no method has cornered the market on policy relevance. I was trained primarily as a qualitative social scientist and I gravitate much more to historically oriented social science. Yet some of the most policy-relevant research I have done has been the statistical work I did with my colleagues Chris Gelpi and Jason Reifler. Today, one of the IR scholars making the most direct contributions to policy debates is the quintessentially quantitative Bruce Bueno de Mesquita, who has developed a quantitative forecasting model that is used extensively by the intelligence community. And I can think of numerous other examples of cutting edge quantitative work that has great policy relevance.

At the same time, most of the passion and concern in the academy-policy gap discussion comes from qualitatively oriented scholars, and there is no question that qualitative scholars and qualitative scholarship have an easier time conversing across the divide to policymakers. If there is a correlation between method and policy distance, it may come about because of the opportunity costs associated with some advanced quantitative methods. Students who invest a lot of time learning the math they need to use advanced quantitative methods appropriately have less time for other courses, especially courses in diplomatic and military history, the classics of political philosophy and political science, and the substance of the topic areas the academic researches. I believe this not only impoverishes them as academic political scientists, but also gives them less context and common ground on which to build a dialogue with policymakers. And in some cases, it opens them up to silly mistakes that cause them to lose the ear of a policymaker — a policymaker is not going to stop listening to an academic expert just because they didn’t use the latest trendy statistical modeling technique, but may if the academic expert reveals a dodgy understanding of, say, the basic diplomatic history of the Cold War, not to mention the background facts and context of whatever policy issue the academic is opining on.

Myth 5: The problem can be solved on one side of the partisan aisle, the Democratic side. The vast majority of the academics worried about the policy-academy gap are Democrats by virtue of the fact that the vast majority of academics are Democrats. Thus, when they seek to build bridges, they naturally reach out to their friends and the people they trust and, lo and behold, they tend to be Democrats, too. There is a danger that the bridge will tilt dramatically to the left. I have seen initiatives focused on this problem where the partisan diversity consists of the full spectrum of Democrats currently serving in government, Democrats who used to serve in government, and Democrats who hope to serve in government, plus a few people too liberal for the Democratic party. In a pinch, a quirky libertarian will be added, but only if he will denounce the imperialism of Republican policymakers more loudly than he denounces Democrats. Such a gathering could involve brilliant people who make interesting observations, but as a bridge-building exercise it seems to me doomed from the start. And yes, Republicans need to get over their knee-jerk disdain for academics; the bridge should be balanced and be a two-way street.