"In the late 1980s, I once enjoyed the privilege of being in the office next to yours [Joseph Stiglitz--ES] for a semester. We young economists all looked up to you in awe. One of my favorite stories from that era is a lunch with you and our former colleague, Carl Shapiro, at which the two of you started discussing whether Paul Volcker merited your vote for a tenured appointment at Princeton. At one point, you turned to me and said, "Ken, you used to work for Volcker at the Fed. Tell me, is he really smart?" I responded something to the effect of "Well, he was arguably the greatest Federal Reserve Chairman of the twentieth century" To which you replied, "But is he smart like us?" I wasn't sure how to take it, since you were looking across at Carl, not me, when you said it. [Rogoff was a child chess-prodigy.--ES] My reason for telling this story is two-fold. First, perhaps the Fund staff who you once blanket-labeled as "third rate"—and I guess you meant to include World Bank staff in this judgment also—will feel better if they know they are in the same company as the great Paul Volcker. Second, it is emblematic of the supreme self-confidence you brought with you to Washington, where you were confronted with policy problems just a little bit more difficult than anything in our mathematical models."--From an Open Letter by Kenneth Rogoff to Joseph Stiglitz. [I thank Hülya Eraslan for calling my attention to it.]
Beyond the clash of personalities (and even institutions), this decade old letter raises several issues of significance to economics and philosophy. There are several important issues lurking in Rogoff's letter none of which he develops: (a) the gap between the most sophisticated mathematical models and the complexities faced by policy-makers; (b) the informal/social metrics by which entry in the very top of the discipline is governed (we philosophers have similar conversation, alas); this ((b)) matters because (c) these departments are (i) both the guardians of professional journals [recall] as well as (ii) the credentialing place for entry into the highest policy regions Stateside (and, thus, internationally); (iii) these metrics create opportunity for all kinds of biases (as we philosophers are also starting to recognize). If (a) is not well understood, then with certain personality types, it also leads to (d) expert-0verconfidence, and (e) the use of scientific reason as a species of cultural imperialism (recall this post); philosophers are not immune to this (see below). In order to tackle these issues in a sophisticated, even philosophical fashion from within economics, one could do a lot worse than by way of engagement with the writings of the mathematical economists, Ali Khan (Hopkins). [Disclosure: I'll be speaking at a Fest in his honor in May.] This week, I'll focus on Khan's confrontation with Peter Singer's ethics. (For a good introduction to this exchange, see here.)
"a world of academia, of theory and theorems, of the arcane and ruthlessly-petty politics of publishing... [and] a world of international organisations, of practice and policy as the back-and-forth of statistical indices, of the power-play of real, big-time politics." (355)
Each of these two worlds, is "local and global in their own way." Khan recognizes that there are people that are able to live within each world at the same time (Amartya Sen is called to mind in context) and intermediaries or traders between these two worlds (and, of course, within each world). In order to avoid confusion (we philosophers really dislike multiple worlds, except when we do modal logic or philosophy of quantum mechanics), I'll call these "spheres" not "worlds." Note, however, that Khan links the academic sphere to "theory and theorems" and the policy-oriented sphere of international organizations to "statistical indices." That is, we're dealing with a conceptual, even normative distinction; Khan is not unaware that plenty of academic economists deploy statistics or help develop statistical indices. Either way, from the vantage point of the issues introduced by Rogoff-Stiglitz exchange, we are most interested in the migration of the sphere of theories and theorists into the sphere of organizations.
In fact, Khan lines up the distinction between these two kinds of spheres of activity with a distinction between "theoreticians" and "theorists," which he borrows from Oakeshott. (361; 370 n. 65) So-called "theoreticians" are to be found in policy-areas. Near the end of his paper, Khan helps us see what he has in mind with the distinction:
between the worlds of theory and practice, of the academic and the non-academic, of the unconditional pursuit of conditional learning and the inevitable need for certifying, policing and monitoring, of the theorist and the theoretician, compromising and moving on in the exercise of power, persisting and getting stuck in attempted understandings. (381)
Again, Khan is not blind to the fact that there are plenty of people with academic training to be found in the policy environment of "practice" and that some of these maintain academic appointments. (This is why I treat his distinction as a conceptual one.) In his lecture, Khan allows that the very same human being can be theorist and theoretician--so I understand the distinction as capturing a functional role (in some sense).
Notice, however, that Khan denies that the theorist is genuinely free--she is policed and monitored, while (by contrast) the theoretician exercises power. (This is not to deny that Khan recognizes other freedoms in the life of the theorist.) To what degree Khan's perspective on the theoretician fits the phenomenological and social facts of the theoretician's life is, of course, open to debate. But, again, it is worth noting that Khan also implies that theoreticians get "stuck in attempted understandings." Given that Khan also charges theoreticians (Summers [recall here] and Peter Singer in the lecture) with basic lack of courtesy in trying to understand the peoples they try make economic and institutional policy for, we should understand Khan as treating "theorist" and "theorist" as character-types with their own virtues and vices. (Khan draws on a Wittgenstein-ian ethic along the way.) Even so, while Khan implies little hope for theoretician's ability to achieve genuine understanding. Khan does not explain if this is so (I) because of the kind of human beings that are attracted to the life of the theoretician or, (II) if as constituted the selection-mechanisms of theoreticians select for the wrong personality types (recall Rogoff's open letter above), or (III) if Khan has a deeper skepticism about the functional roles of the theoretician engendering failures of understanding, or (IV) if Khan has fundamental skepticism about the very possibility of translating between different cultures, languages (spheres, etc.). Within his lecture one can find evidence for all four of these possibilities.
Now, the crucial point here is that in order to even reflect on the limitations of theorist and theoretician alike, Khan has no resources to draw from within mathematical economics or economics more generally. He has to import distinctions from the "human sciences." (In addition to drawing on Wittgenstein and Oakeshott, especially, Khan's lecture also engages with Badiou.) This is not deny that the practice of economists and economic tools can't help inform his reflection on these matters (Khan aptly notes how Peter Singer has made a business out of marketing his ethics (364-5)). But one general point to recognize is that economics is very impoverished at theorizing its own limitations--it lacks (general) equivalence(s) to Kantian critique or Gödelian impossibility. (Within philosophy one can recognize this impoverishment in the touching religiosity of our decision-theorists.) So, in an age-of-specialization, economically trained theoreticians have no systematic vocabulary to reflect on the possible limitations of their own practice. (I first explored this problem in looking at the doctrines taught to Chilean Chicago Boys.) This is one reason why I have been advocating (a) a more important role for history of economics in the training of policy economists and (b) the need to put the modeler inside the model.
Let's call this failure to think a variety of limitations about one's tools from within one's theoretical or theoretician's perspective, a "first-order-failure." This first-order failure can create disaster in the theoretician's policy interventions--interestingly enough, Khan seems to be more concerned with the lack of mutual understanding and the associated lack of respect from the theoretician toward the beneficiaries of his institutional benevolence (the treatment of Peter Singer's One World: The Ethics of Globalisation is especially illuminating on p. 374) than, say, the economic costs that may be inflicted by poor policy choices that result from such a first-order failure.
In his lecture, Khan also diagnoses a second-order failure that theoreticians are especially prone to--that is, they may not really understand the tools they import from theory. This second-order failure is especially striking when it is exemplified (by Peter Singer) in the context of rule by expert, or the "best minds," backed up by international bureaucrats (366-7). In analyzing Singer's advocacy of emissions trading, Khan calls attention to Singer's non-trivial ignorance of economics (368-9). The more important issue, of course, is to become clear on how such second-order failure is facilitated by the first-order failure. Here certain branches of philosophy and applied welfare-economists become nearly indistinguishable. Khan puts the point as follows:
The point is that a certain type of economic theory, one of its more mainstream and prominent brands, explicitly advocates the dissociation between questions of the distribution and the efficiency of resources, and focuses entirely on the latter, on situations where everyone gains, and there is no entanglement with conditions under which a policy benefits one and harms the other. This is the impulse that leads some economists to avoid the discussion of values that are not marketable, an impulse that has led to the enviable autonomy of their subject matter, prevented what can be seen in some circles as inter-disciplinary promiscuity. (369-70)
So, second-order failure is facilitated when theory has taken some crucial real-world issues off-its-analytical-table. What matters is that the policy-oriented theoretician does not call attention to and is not transparent about the controversial assumptions employed in the "jargon" (370). These remain "veiled" as long as the theoreticians agree with each other, or until a theorist with high enough institutional prestige temporarily forgoes the "unconditional pursuit of conditional learning" and enters the fray out of some higher duty or loyalty to affected communities (by which time it may be too late to prevent policy implementation). The autonomy of (monitored and policed) theorists and theoreticians has non-trivial (ahum) externalities.
Although there is more to be said about Khan's analysis of the role of philosophy in such matters, this has gone long enough for now.