Becker, for example—the most radical of the American neoliberals, if you like—says...that the object of economic analysis can be extended...and that economic laws and economic analysis can perfectly well be applied to non-rational conduct, that is to say, to conduct which does not seek at all, or, at any rate, not only to optimize the allocation of scarce resources to a determinate end. Becker says: Basically, economic analysis can perfectly well find its points of anchorage and effectiveness if an individual’s conduct answers to the single clause that the conduct in question reacts to reality in a nonrandom way. That is to say, any conduct which responds systematically to modifications in the variables of the environment, in other words, any conduct, as Becker says, which “accepts reality,” must be susceptible to economic analysis. Homo oeconomicus is someone who accepts reality. Rational conduct is any conduct which is sensitive to modifications in the variables of the environment and which responds to this in a non-random way, in a systematic way, and economics can therefore be defined as the science of the systematic nature of responses to environmental variables.
This is a colossal definition, which obviously economists are far from endorsing, but it has a certain interest. It has a practical interest, if you like, inasmuch as if you define the object of economic analysis as the set of systematic responses to the variables of the environment, then you can see the possibility of integrating within economics a set of techniques, those called behavioral techniques, which are currently in fashion in the United States. You find these methods in their purest, most rigorous, strictest or aberrant forms, as you wish, in Skinner, and precisely they do not consist in analyzing the meaning of different kinds of conduct, but simply in seeing how, through mechanisms of reinforcement, a given play of stimuli entail responses whose systematic nature can be observed and on the basis of which other variables of behavior can be introduced. In fact, all these behavioral techniques show how psychology understood in these terms can enter the definition of economics given by Becker....
...Becker’s definition, which, again, although it is not recognized by the average economist, or even by the majority of them, nonetheless, despite its isolated character, enables us to highlight a paradox, because homo oeconomicus as he appears in the eighteenth century—I will come back to this shortly—basically functions as what could be called an intangible element with regard to the exercise of power. Homo oeconomicus is someone who pursues his own interest, and whose interest is such that it converges spontaneously with the interest of others. From the point of view of a theory of government, homo oeconomicus is the person who must be let alone. With regard to homo oeconomicus, one must laisser-faire; he is the subject or object of laissez -faire. And now, in Becker’s definition which I have just given, homo oeconomicus, that is to say, the person who accepts reality or who responds systematically to modifications in the variables of the environment, appears precisely as someone manageable, someone who responds systematically to systematic modifications artificially introduced into the environment. Homo oeconomicus is someone who is eminently governable. From being the intangible partner of laissez-faire, homo oeconomicus now becomes the correlate of a governmentality which will act on the environment and systematically modify its variables.--Michel Foucault, The Birth of Biopolitics, 28 March 1979, translated by Graham Burchell, 269-271
While it is not the main topic of today's digression, it is worth noting how incredibly prescient Foucault is in grasping that what is now known as nudging is an eminently foreseeable, internal development of Chicago economics. In context that point is even clearer in light of the concluding remarks of the previous lecture (of March 21), which focuses on the possibility of "an environmental type of intervention" to produce or stabilize homo oeconomicus which in the Becker/Stigler version of Chicago just is MaxU.
Now, Foucault rightly notes that Becker actually relies on a relatively thin notion of agency which really amounts to responding 'systematically to modifications in the variables of the environment.' Incentives matter, after all. Let's call this the modern economist's conception of agency.
As an aside, as I have noted before Foucault actually understates the situation here. Becker's own position develops an idea from Alchian (and noted by Arrow) that even 'agents' who behave stochastically, will, if they respond to the environment -- and there is some selective winnowing -- approximate the behavior of MaxU, or, as Milton Friedman also emphasized, can be treated as if they do so.
Okay, with that in place, let's turn to Daniel Dennett's treatment of the Warden's Rule: "if it can happen, it will happen. (See Freedom Evolves, 2003, 160-61. The version I discuss below is reproduced from Dennett's preface to David Haig's excellent From Darwin to Derrida (forthcoming); recall this post by Haig.)
The Warden's Rule...is an improvement on “Murphy’s Law,” that whatever can go wrong will go wrong. It is an improvement because there are real circumstances where a version of it is true, not merely an amusing expression of pessimism. It is supposedly the maxim of every prison warden. Its rationale is that prisons are full of prisoners with time on their hands and a measure of patience and an inquisitive and competitive nature. They will exhaustively search every aspect of the system in which they are incarcerated looking for a way out, or a way of improving their condition, however arduous the path. How clever are the prisoners? Some are brilliant, others just dogged; if they share their discoveries it doesn’t much matter who gets credit for the innovations. Instead of testing them for intelligence, test their environments for opportunities. (In Elbow Room (1984), I described the existence of a bag of jewels in a trash can a few feet away as a bare opportunity—if only I had known about it, I could have been rich, but I had no information at all that would lead me to check that trash can for treasure.) According to the Warden’s Rule, even bare opportunities must be counted, since prisoners have nothing better to do than hunt for them. No trash cans go untested in a prison.
Bare opportunities abound in the world of evolution, and it doesn’t take intelligence to stumble across them; it just takes lots of time for variations to arise and get tried. If there is information in an evolutionary environment, a difference that could make a difference to something that could detect or just respond to that difference, it will make a difference (generally—this is a rule of thumb), a difference that in turn can be amplified by reproduction, making a bigger difference and, in a recursive process, making the detection of that difference more likely in the future, and so forth. Yes, this is just another way of explaining how natural selection works, but it works this way all the way down, so that even simple molecular structures can be usefully seen to be like those persistent prisoners, ever delving, ever seeking an advantage that they can exploit to further their own “interests”. The prisoners have a single overarching goal: Escape! These agents have a different goal: Reproduce!--Daniel Dennett.
The warden rule prompts a useful contrast with the modern economist's conception of agency. What the warden's rule suggest is that rather than treating agents as responding systematically to modifications in the variables of the environment, agents are better thought of as responding to systematic possibilities in the environment. The difference isn't primarily about the shift from incentives (a) to information (after all, the variables in the environment can be treated as information). Nor is the difference (b) that the warden rule stipulates away search and opportunity costs (something an economist would not do)--the warden rule effectively assumes that prisoners have infinite time (and patience, etc.). I don't mean to deny that (b) is important. That evolution takes place in time, often extended periods of time, is non-trivial. Rather it is, primarily about the shift from tracking features of the environment (c) to searching and exploiting possibilities in the environment. In particular, the warden rule (d) tacitly recognizes the agency of the agent in her shaping the environment in ways that cannot be effectively foreseen. I view (c)-(d) as connected.
Now, I could stop here.* As Foucault notes, but does not explore, Becker's analysis received a response by Kirzner in 1962 (the leader of the Austrian school of economics in exile). As I noted a few weeks ago, Kirzner was very interested in entrepreneurship. This matters for my present purposes because (d) just describes entrepreneurship (or attributes it to organisms). So, today, I am going to assume Kirzner's commitment to some analogue of (d). But I am interested in his relationship toward (c). To get to that we must note that Kirzner conceived of market processes as essentially about "the systematic way in which plan revisions are made as a consequence of the disappointment of earlier plans." (381; Kirzner is describing buying and selling plans.) In other context, it would be tempting to talk about updating here. But we can discern here that Kirzner is articulating features of (c).
The connection between (c) and the intentional stance is also made explicit by Kirzner: "only by assuming that buyers purposefully seek to achieve given goals can we predict that their thwarted market plans of yesterday will lead to their systematically offering more attractive choices to sellers today." (382) Of course, Kirzner may well think that the intentional stance is more than a mere stance, but as he goes to acknowledge one need not assume such purposeful rationality.+ He recognizes that market institutions as a system may induce behavior that can be treated as purposeful. It is worth asking if he was already aware of Vernon Smith's striking experimental results (about that some other time).
Now, one may think that I have not said sufficient amount to motivate that Kirzner is really articulating a version of (c). But the relationship between search (which is his main focus in context), the intentional stance, and the role of possibility in the environment, is pretty clear in his thought: "on the assumption of rationality, a market demand curve is seen as a set of hypothetical equilibrium positions." (I agree it is not identical to the Warden Rule, but we're not far removed either.)
Okay, let me zoom out again. Foucault, correctly noted that one way to understand the difference between classical liberalism (e.g., Adam Smith) and many branches of twentieth century neoliberalism, is that the former treated economic analysis in terms of exchange and the latter in terms of competition. Both of these are fundamentally about processes. Foucault treats Becker (with a nod to Robbins) as something radically new: now "the starting point and general frame of reference for economic analysis should be the way in which individuals allocate these scarce means to alternative ends." (223, 14 March 1979). And with the ends given (by experts or politicians, tradition, society, etc.) the individual can be nudged into being good at such allocations.
As Kirzner discerned, Becker's approach is not at all about process. By contrast he thinks, "it is primarily upon the systematic revisions of disappointed plans that the market process depends." (384) He thinks this feature of purposeful behavior generates stable outcome patterns associated with market processes. As is well known, Becker's approach was vastly more influential in economics and facilitated the manner in which it became the imperial discipline in other social sciences. So, Foucault was prescient to ignore Kirzner and focus on Becker.
Let me wrap up. I understand Vernon Smith's contribution to economics in part to teach us all that the micro-foundations of economics are unknown. (Let's not even begin with macro.) However, in other part, some institutions and norms generate remarkably stable outcome patterns without us really understanding why. With Dennett and David Haig re-articulating Darwinian theory, and, in particular, supplying a new set of ideas about natural teleology and social functions (recall Haig here; and me on Dennett here); and about which more soon), a Darwinian process oriented approach (not necessary Kirzner's) is ripe for a major relaunch in the human sciences, also in economics.
*A remark by Bas van der Vossen made me return to Kirzner.
+Kirzner rules out completely random behavior and purely habitual agents. He lacks the idea of an intentional stance, so at times he makes claims stronger than I attribute to him (as befits his commitment to praxeology). But in all the cases he does, all he needs is something equivalent to the intentional stance or in his terminology the attribution of "purposeful" behavior.
Comments