Coordination in the History of Economics

This Thursday, Daniel Klein will be presenting his paper (co-authored with Aaron Orsborn), “Coordination in the History of Economics,” at the Workshop on Philosophy, Politics & Economics (hosted by Peter Boettke). Dan and I have discussed his ideas on coordination for a while. His paper directly indicts the way Kirzner uses coordination. I won’t be at the workshop on Thursday so I will provide some comments online. If you are interested in the discussion, open the link below.

Let me make two general points.

The distinction the paper makes between concatenate and mutual coordination makes sense from the perspective of the history of economic ideas and from an analytical perspective. The paper treats these two aspects of the issue in a somewhat interrelated fashion. They should perhaps be more clearly separated. As far as the analytical aspect is concerned, I disagree with the authors’ interpretation of the work of Kirzner and other “latter-day” Austrians.

The authors do not discuss the following point. One could argue that the notion of concatenate coordination has not disappeared from contemporary economics. Instead it has been subsumed under the notion of equilibrium (as in general equilibrium), which basically means complete concatenate coordination (i.e. no better pattern of concatenate coordination can be more pleasing (to whomever) than the equilibrium one). This would mean that in contemporary economics, equilibrium can be used to cover the two meanings of coordination (concatenate and mutual).

Here are comments on the paper.

The authors make the point that we should resist calling the spontaneous market order a system of cooperation (p. 7). This may overlook the sense in which Hayek (and others) used the term “cooperation” in this instance. It is used in a looser sense to mean that in the liberal political order individuals engage in activities of their own accord without being forced to do so by anyone. In that sense, competition is a system of cooperation. This is no surprise that other liberals such as Bastiat also used the term in the same way.

Here are some remarks on the section on the “confusion among the ‘Austrians’.”

First (and this is a minor point), it seems strange that Klein and Orsborn have elected themselves as the purveyors of truth in economics and can define what is acceptable as an idea from what is not (p. 26). Whether the idea of coordination as defined by Kirzner is acceptable to them or not is irrelevant. Who cares? It matters to know what the concept of coordination is about, not whether it is acceptable to the authors.

The authors claim that Kirzner (and Hayek to some extent) has established two new meanings of coordination (see figure 6: columns II and III). I would argue that this is not the case. While it is true that Kirzner is perhaps not sufficiently clear in his different uses of coordination, he has not introduced two new views. He mostly uses the notion of coordination as concatenate coordination and, at times, he may also use it as mutual coordination. [I am presenting my view of Kirzner’s work out of memory because I don’t have access to my library at the moment. If need be, I will provide the references.]

Kirzner’s work enriches our understanding of the concept of concatenate coordination. The crucial Kirznerian distinction that is missing in the paper is that between underlying variables and induced variables of the market. This distinction can also be found in Mises (but not in those terms). Mises refers to false prices, which means that market prices reflect only the current available knowledge of market participants (e.g. see Mises HA pp. 337-8). This knowledge is present in current market prices (which are induced variables) but there is more knowledge that is yet to be known (which is in the underlying variables of the market). False prices are not equilibrium prices because there is more knowledge to be discovered at any moment in time. From that perspective the notion of dispersed knowledge present in Hayek’s work is also present in Mises’s. Kirzner builds on both views to theorize about the entrepreneurial discovery process. The entrepreneurial process can only be understood with reference to the existence of false prices and thus with regard to the underlying/induced variable distinction. Coordination in Kirzner’s work takes place with regard to these two sets of variables.

Kirzner talks about coordination in the market from the perspective of an omniscient being who would have access to the knowledge that is yet to be discovered (this mind would know what the underlying variables are). Coordination in Kirzner’s work is generally discussed with this benchmark in mind. In the “living economy”, plans are discoordinated with regard to the knowledge contained in the underlying variables of the market, not with regard to the induced variables. Opportunity discovery is about rearranging resources taking into account the new discovered information that was contained in the underlying variables of the market. This is why profit opportunities can go unnoticed without anyone experiencing any frustration (p. 29), but prices will still be false (ships are passing by each other in the night).

Regret (p. 29) may arise when one realizes that some other course of action could have been taken at zero cost. Kirzner discusses regret and explains that the very notion of regret is, in fact, only conceivable in a world where discoveries are true discoveries (i.e. they are costless). Had someone been alert, she would have noticed the opportunity, but she wasn’t, so she can truly regret not having noticed what was staring at her in the face.

I disagree with your claim that “there is no clear reason why state regimentation of economic sectors… cannot satisfy Kirzner’s explicit definitions of coordination” (p. 30). If opportunity discovery is about bridging the gap that may exist between underlying and induced variables such that gains from trade are captured, it results that a more pleasing (to the omniscient mind) state of concatenate coordination could occur (i.e. more knowledge has been revealed in prices). Kirzner argues that this discovery process cannot be attained through any other process than the entrepreneurial discovery process. While some (all?) individuals may not experience frustration or disappointment under regimented markets, discoveries will be hampered and thus, from the perspective of the omniscient mind (who knows the underlying variables), a more pleasing state of concatenate coordination won’t be achieved (because knowledge cannot be revealed).

Kirzner makes the claim that successful entrepreneurial activity brings greater concatenate coordination. Allocation of resources by entrepreneurs is concatenate coordination at the individual level (i.e. the entrepreneur coordinates resources). When successful (i.e. when it discovers an actual gap between underlying and induced variables), it leads to greater concatenate coordination at the systemic level (better knowledge of individuals’ preferences is conveyed by prices). Klein and Orsborn claim that Kirzner does not address the issue of successful entrepreneurial activity upsetting some people’s plans (p. 30). This is not true. Kirzner directly addresses this issue in his distinction between underlying and induced variables. Successful entrepreneurial reallocation of resources may frustrate the plans of some (based on the knowledge currently conveyed by false prices), but from the perspective of the underlying variables of the market, it offers greater perspective of contentment. In other words, (false) market prices had not yet revealed a better way to use society’s resources—better in the sense that it was satisfying more urgent needs than the current way of using resources. From this perspective, every instance of successful entrepreneurship is necessarily (concatenately) coordinative (not from the perspective of the knowledge revealed in the false old prices but from the perspective of the new knowledge that has been discovered and is now reflected in the new false prices).

Last, I would like to address footnote 20 on p. 31. Kirzner separates the development of institutions from market activity for the reason that there is no entrepreneurial process that would come to systematically select and improve institutions and rules. There is no market process because there is no entrepreneurial discovery and monetary calculation associated with the rules that the market rests on. This does not mean that there is no new institution ever emerging. Indeed, people may come to discover new rules and implement them. But there is a difference between innovation in rules and the existence of a systematic entrepreneurial process of discovery. Rules and institutions are selected via other mechanisms than that of the market process. This does not mean that the market process won’t influence rule selection. It just means that the two mechanisms are different.

It is true that Kirzner’s thought is dispersed over at least four decades of writing and it makes it difficult to pin point his exact view. I believe that the notion of coordination that Kirzner uses is mainly that of concatenate coordination. The main reason why it doesn’t seem obvious to the authors is because they fail to take into account Kirzner’s view of the market process as a process of discovery of knowledge that was hitherto not contained in (false) market prices. The Misesian notion of false prices and the Hayekian notion of dispersed knowledge are key elements to understand Kirzner’s view of coordination. This view of coordination is best understood by taking into account the distinction Kirzner makes between underlying and induced variables.

I have learned a lot reading Dan's work and this paper is no exception. I have tried my best to understand Kirzner's position and to defend it when I think he is right. I do not claim to have all the answers nor that Kirzner is always right. However, I still believe Kirzner's economics is the best theory we have explaining the truly dynamic character of the market system.

Comments

[Alex Max]

Where is the best source to read up on Kirzner's thoughts on underlying & induced variables?

thanks

[Peter Boettke]

The Meaning of Market Process (1991).

[dg lesvic]

This will sure inspire the masses.

Capitalism is saved.

[Mario Rizzo]

This paper appears to be total madness. Mutual coordination begins around 1960???? I invite the interested reader to look at my detailed treatment of Hayek in "Hayek's Four Tendencies Toward Equilibrium" on my website. Click below.

[Daniel Klein]

Mario, Sounds like you are equating mutual coordination and "mutual compatibility of plans/expections" stuff.

If I'm selling pencils in a competitive market, there may be a sense in which my expections and plans are fulfilled--the Nash requirement may be satisfied. But it is not the case that the other pencil sellers are doing strategies that are best for me given what I'm doing. I would prefer that they charge higher prices, withdraw from the market, etc.

In equilibrium terms, coordination is stronger than the Nash requirement. Every coordination equilibrium is a Nash equilibrium, but the reverse is not true. The market equilibria that Hayek's generation focused on are examples of the latter.

Schelling coordination really had practically no presence in economics at all prior to The Strategy of Conflict. Now it dominates--most usage of "coordination" in mainstream literature is Schelling coordination.

[Daniel Klein]

I appreciate Fred's commentary. I'm going to hold off on responding.

The Austrian section of Klein and Orsborn is being removed, so as to simplify the paper. The Austrian section will become a separate paper. I will use Fred's comments in thinking about the development.

[Islam Hussein]

The problem of coordination in multi-agent (mechanical robotic) systems has received much interest in recent years, especially from a mathematical point of view. Methods include graph theory, networks, control theory, etc. As a new comer to the literature on the economic side, I wonder if there are economists who have strongly approached this problem from a mathematical point (I came across Payton Young's work), especially a decentralized, spontaneous order or methodological individualism point of view (yet is still mathematical is approach).