February 2021

Sun Mon Tue Wed Thu Fri Sat
  1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28            
Blog powered by Typepad

« New Blog to Bookmark --- The Sociological Imagination | Main | Interview with Richard Ebeling »

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

And let's not forget Alchian's 1950 paper on economic evolution, nor Vernon Smith's recent work on ecological rationality. I would argue that these fit in that same tradition.

If there are no online links then the articles don't exist.

Did you mean this article ("Fritz Machlup and marginalism: a reevaluation")?:
http://www.helsinki.fi/filosofia/inem/methodus/pdf/v3n2/v3n2p86.pdf

Since the debate on other threads to which Pete is referring involves me, my take on the Lester-Machlup confrontation is on my home page:

http://www.criticalreview.com/crf/jeffreyfriedman.html

For the PDF, you'll have to scroll down to the 1996 area, or just search for "Introduction to The Rational-Choice Controversy: Economic Models of Politics Reconsidered. Yale University Press, 1996." The Lester-Machlup part is on pp. 11-13. (And for a critique of Mises in his confrontation with Weber over Gresham's Law, see end note 4!)

Jeff

I wrote in another thread that just because some fact would have, in hindsight, provided a good incentive for some action, it is irrelevant unless the fact was actually taken into consideration by those acting.

While I think this is quite true for rare or unique actions, when a problem-situation is regularly iterated, it is possible for actors to unknowlingly adjust their choices in accordance with a relevant fact.

In many primitive groups, the act of washing has evolved to take place at times and in places that account for the facts of microbiology, even though members are not taking into consideration those facts. The incentive structure they are responding to is predicated upon entirely false assumptions, but evolutionary pressures have forced them to unknowingly adjust their actions to the facts.

The point I am trying to make here is a nod toward the theory impregnated character of observations, (and the under-determination of theories by evidence). Evolution selects for beliefs that work as far as they need to, and not necessarily beliefs that are true.

Recent analysis suggests that Lionel Robbins wrote most of his "Nature and Significance" book under the influence of Wicksteed, and only later added references to Mises.

Care to comment.

Isn't is the case, however, that Hayek _isn't_ analyzing individual action in "Econ. & Knowledg", he's putting together a purely _logical_ picture of multi-person economic coordination, premised on the idea that planning and choice require some degree of marginal evaluation.

Evidence here includes Hayek use of the expression "the pure logic of choice". More confirmation comes in first couple of chapter of _The Pure Theory of Capital_, which goes over the same ground.

This a _logic_ assumptions by definition are going into an analysis. The marginal logic built into an equilibrium construction is suppose to capture something true about the ordering of both individual plans and the overall market order.

Marginalism is both a logic cranked out from assumptions going into an analysis, and it captures a systematic order built into decision making and into the global economic order -- whether individual actors are aware of it or not.

Hayek's claim is Mises claim and Wittgenstein's claim -- people's mental structures are structured like our own, and people may have no ability to articulate or map that structure in terms of logical relations or in natural language, but the fact that people share common structure of mind is what allows us to "go on together".

See Hayek's "Scientism and the Study of Society" for his statement about common structures of mind. I believe Mises' talks of this in his _Human Action_. Wittgenstein makes this point in several places in his later work.

Peter wrote:

"But it is also a position consistent with Hayek's presentation in "Economics and Knowledge", where the marginal conditions are not assumptions going into an analysis, but by products out of decision making within the market process."

Greg,

You raise very important points. But I think what Hayek was getting out in that paper is that while the pure logic of choice is a necessary component of an explanation of the market economy, it is not sufficient. Instead, we need to be explicit about the institutional environment and how alternative institutional environments impact the learning that takes place through market transactions that produce market clearing. This is, in fact, the empirical content of the analysis of the market. This is an important part of his criticism of the model of market socialism -- as those models assume what it is that must be proven.

Alchian's 1950 paper develops this argument, and Vernon Smith's Nobel address is right along these lines as well. In a very simplistic language, rationality is omnipresent, but manifestations of rationality are institutionally contingent.

I also liken Hayek's argument to the one presented in Nozick's Anarchy, State and Utopia describing an invisible hand explanation --- we must identify filter processes and equilibrium tendencies.

So I don't consider the 1937 paper to be a call for "pure logic", but instead an argument that while choice theory is an exercise in pure logic, market theory belongs in the realm of applied theory --- combination of pure theory with empirically contingent assumptions about institutional environments and their epistemic properties.

The critical issue in the historiography of the Austrian school is whether or not this position serves as a break from Mises. Despite the weight of evidence suggesting that it is, I actually argue that it is NOT.

At the end of the day, whether it is _the_ Austrian position or not is not my concern, it is whether thinking about it this way helps us better understand the market process and its systemic tendencies. I think it does, and it finds antecedents in the non-Ricardian British economists of the catallactic tradition --- economics is about exchange, and the institutions within which exchange takes place. And it is carried forward by Buchanan, Coase and Smith, and this perspective goes a long way to explaining why these thinkers differ so fundamentally from others that follow a more conventional equilibrium approach to market theory and the price system.

I must confess that I am unclear what the Lester-Machlup was about, or how recent discussions are related.

Lee,

Richard Lester did surveys asking economic actors whether they consciously follow economic optimality rules --- such as price equal to marginal cost in a competitive market, or produce that level of output which minimizes average cost, etc. Businessmen answered basically, I have no idea was marginal cost is. In some cases they might say things like we price by looking a cost and then providing a mark-up.

But what Machlup pointed out was (as the Goethe quote to start this post suggests) that economic actors due to the constraints and filters of the market will behave in ways that result in marginal cost pricing or profit-maximizing even if they don't know it. It is not that they act "as if" (standard interpretation of Machlup's position), but instead that their behavior is guided by the filter mechanisms in place (the systemic effects) even though they are unaware that they are doing it (hence again Goethe's words).

This argument was not new to Machlup (hence the reference to Wicksteed).

So go back to the statement by Matěj Šuster in the earlier post about the link between incentive environments and the *discovery* aspect of the market. Incentives and ignorance are intimately connected. Some incentive environments do not alert us to information that would be vital for us to make *correct* decisions from a birdseye view. In fact, if we go further into epistemic communities, outside of certain context it is not just that we do not collect the information, the information literally does not exist because it wasn't created due to the incentives in place.

Lee,

a useful link for you:
Philippe Mongin: The Marginalist Controversy (he seems to be a post-keynesian):
https://studies2.hec.fr/jahia/webdav/site/hec/shared/sites/mongin/acces_anonyme/page%20internet/O13.MonginMarginalistHbk97.pdf

Pete writes:

"Greg, You raise very important points. But I think what Hayek was getting out in that paper is that while the pure logic of choice is a necessary component of an explanation of the market economy, it is not sufficient."

Right, Hayek uses the pure logic to 1) help us see an the global economy; 2) he uses the pure logic as a foil (same as Marshall and Mises) to highlight _non-logical_ explanatory elements like entrepreneurial learning and patterns of rule governed behavior upon which institutions rely; 3) he uses the pure logic to give us a window into the structural logic of the operation of changing relative prices. (Interestingly, Hayek rarely talks about how the pure logic of choice helps us to seen the order in individual behavior.)

The role of Hayek's "Economics and Knowledge" essay in his economics is helpfully fleshed out in the first 2 or 3 chapters of Hayek's _The Pure Theory of Capital_, emphasizing most of the points listed above.

Pete writes:

>>Instead, we need to be explicit about the institutional environment and how alternative institutional environments impact the learning that takes place through market transactions that produce market clearing."

Right. Hayek began to do this in _The Road to Serdom_, and then went "Wittgensteinian" in his analysis of moral rules, behavior pattern, law, and institutions in the 1950s, esp. after reading the work of the neo-Wittgensteinian R. S. Peters, themes Hayek was already picking up from Gilbert Ryle in the 1940s.

Pete writes:

>>This is, in fact, the empirical content of the analysis of the market. This is an important part of his criticism of the model of market socialism -- as those models assume what it is that must be proven.>Alchian's 1950 paper develops this argument, and Vernon Smith's Nobel address is right along these lines as well.>In a very simplistic language, rationality is omnipresent, but manifestations of rationality are institutionally contingent.>I also liken Hayek's argument to the one presented in Nozick's Anarchy, State and Utopia describing an invisible hand explanation --- we must identify filter processes and equilibrium tendencies.>So I don't consider the 1937 paper to be a call for "pure logic" <<

Right. It's not. As made clear by the first 3 chapters of Hayek's _The Pure Theory of Capital_.


Pete writes:

but instead ["Econ & Know." is] an argument that while choice theory is an exercise in pure logic, market theory belongs in the realm of applied theory <<


More later.

Peter writes:

"So I don't consider the 1937 paper to be a call for "pure logic" "

Right. It's not. This is made clear also by the first 3 chapters of Hayek's _The Pure Theory of Capital_.

Pete writes:

"but instead ["Economics & Knowledge" is] an argument that while choice theory is an exercise in pure logic, market theory belongs in the realm of applied theory --- combination of pure theory with empirically contingent assumptions about institutional environments and their epistemic properties."

First, lets point out that the word "theory" can be used in all sorts of different ways, for all sorts of different purposes. Sometimes it helps to leave the word on the sidelines (one reason Hayek introduced the language of "the pure logic of choice".)

Second, you are leaving out a step. There is 1) the pure logic of choice and equilibrium theory; there is 2) identification of the empirical problem of order, and the identification of its emprical / causal explanation in entrepreneurial learning and institutions; and there is 3) the application of this empirical explanatory strategy to particular cases (i.e. "applied theory"). Most of Hayek's contribution to economic science consists of 1) and especially 2).

Pete writes:

"The critical issue in the historiography of the Austrian school is whether or not this position serves as a break from Mises. Despite the weight of evidence suggesting that it is, I actually argue that it is NOT."

It's not a "break" but it's a huge advance in terms of getting clearer about the empirical problem / causal explanation nexus, and in terms of getting a clearer idea of the place of "pure logic" in this causal science, and in terms of getting a deeper understanding of human behaviorial institutions and the "autonomy" of human learning, esp. in relation to the "facts of the social sciences".

I'd also count Hayek's work on non-permanent production goods and equilibrium theory and its relation to causal explanation as an advance -- in some ways even a "break" toward a more Mengerian / Bohm-Bawerkian position, and less of a classical / Ricardian "land, capital, and labor" non-marginalist equilibrium picture (Mises Ricardian nodding neo-classical "steady state".)

Agreed.

Pete writes:

"At the end of the day, whether it is _the_ Austrian position or not is not my concern, it is whether thinking about it this way helps us better understand the market process and its systemic tendencies. I think it does, and it finds antecedents in the non-Ricardian British economists of the catallactic tradition --- economics is about exchange, and the institutions within which exchange takes place. And it is carried forward by Buchanan, Coase and Smith, and this perspective goes a long way to explaining why these thinkers differ so fundamentally from others that follow a more conventional equilibrium approach to market theory and the price system."

Pete wrote:

"In fact, if we go further into epistemic communities, outside of certain context it is not just that we do not collect the information, the information literally does not exist because it wasn't created due to the incentives in place."

Not sure if I've understood this correctly, but I've often wondered along similar lines while thinking about fiduciary duty. To me the presence of fiduciary duty is always a red flag for skewed incentive structures. The way I've come to view it, fiduciary duty is designed to make managers act as if they have the same incentives as owners. But even if a manager wants to act like an owner, the manager will not see the world the same way an owner will. Interpreting the data like an owner or forming plans on the basis of and acting on the data like an owner won't help, as the very data will be different. For an owner, data will always be shaped much more by considerations relating to the middle- and long-term than it will for a manager.

The fact that so much is now owned by corporate persons rather than by real persons I think means that a whole bunch of rich data that would be available if owners ran companies simply doesn't exist for their corporate managerial successors.

Permit me to return to the original post. The article by Koppl and Langlois is terrific. (I hadn't read it in many years.) I wonder how it applies to behavioral economics today. It seems that today's behavioral economists are arguing in the tradition of Lester. Thanks to Matěj Šuster for posting it.

Pete and Matěj,

Thanks for the clarification. This is quite confusing for me, because my default mode of thinking is evolutionary; I tend to discount subjective intentions and beliefs, and focus upon the objective consequences of action. It might seem otherwise when eading my comments from the previous thread, but for me there was no underlying conflict.

Mario,

Slightly off-topic.

Suppose there is a group of people who are opposed to the models used by evolutionary biologists; it is unrealistic, they claim, to suppose that animals strive for fitness maximisation. When interviewed, animals give a totally different explanation of how and why they make decisions than the models assume, and animals are so bad at simple logic and math puzzles, how would they even know how to maximise their fitness?

What animals need, the behavioural evolutionary biologists say, is some government program to help them maximise their fitness, because they're just too irrational to do it otherwise.

It seems to me a matter of how you interpret economists' models.

Some people interpret the models to describe the subjective decision-making of market participants. Other people interpret the models to describe the objective consequences of market processes.

The latter group of people, of which I am a member, are less interested in how or why individuals make decisions, than they are in how alternative laws and institutions reward or punish different decisions.

What is important, economically, is not how or why some businessman decided upon a price, but that market processes select for those who get the "right" price, i.e. that which best maximises profits. They are doing it, but many do not know they are doing it.

The fact that market participants might be irrational is irrelevant, because, in principle, market participants need not even be conscious deliberating agents.

Subjective knowledge is overrated.

This is a fun thread. I think it is worth noting that Machlup and Edith Penrose were opposed to Alchian's famous 1950 article. They thought it bled human action out of the picture. Today, we think all the pieces fit seamlessly. That is my view. But in the 1950s, it could look to as broad a thinker as Machlup as if Alchian's argument needed to be answered and eliminated. I regret that I do not recall where I learned this stuff. I think it may have been in the introduction to Penrose's 1959 book, but I really do not recall. I also seem to recall that Alchian thought his article was addressing both sides of the Machlup Lester debate.

Hayek's reaction to Alchian's article was different that Machlup. Hayek thought Alchian's article was terrific -- and an important contribution to the explanatory picture.

Alchian, of course, was himself heavily influenced by Hayek's "The Use of Knowledge in Society".

See Alchian's "UCLA" interview of Hayek from 1978, linked at "Taking Hayek Seriously".

Hayek cites Alchian's 1951 essay in L, L & L. and elsewhere.

I think this is particularly interesting for Public Choice.

I think the debate between Public Choice advocates and Jeffrey Friedman relates to these problems. When politicians are supposed to be rationally maximizing in Public Choice what exactly does that mean?

The comments to this entry are closed.

Our Books