Volume 1 (Fall 1979)
Lawrence H. White
Ludwig M. Lachmann
In the Spring 1979 issue of this newsletter Lawrence H. White comments on "a controversy sparked by Professor Lachmann's thought which has arisen in Austrian circles over the question of general equilibration." I have thus far refrained from commenting on an issue I could not understand. Outside the text-book industry general equilibrium has few defenders these days, and the more thoughtful among them now readily admit that it has little to do with the real world. It puzzles me that some Austrians seem ready to die in the last ditch for a cause that is not theirs and the outstanding exponents of which, from Pareto to Hahn, have never concealed their contempt for Austrians and their ideas.
Now, however, White has set down a number of points I think I can understand. In commenting on them I may hope to further the cause of clarification he has made his own, even where I disagree with him.
(1) White, quoting from my 1943 article on expectations,[1] has me note that expectations are "largely the result of the experience of economic processes." I added, however, on the next page: "This experience, before being transformed into expectations, has, so to speak, to pass through a "filter" in the human mind, and the undefinable character of this process makes the outcome of it unpredictable." (ibid, p. 67) Thirty-six years later, I adhere to this view.
(2) On the other hand, White makes me look a more extreme ex ante man than I actually am. "Lachmann has taken primarily an ex ante perspective."
I do not deny that often, where possible, ex ante views have to be compared with ex post outcomes and that agents have to judge the success of their actions in these terms, nor that economists have to take note of this fact. What I deny is that such judgments permit us to predict what agents will do next. Learning from our mistakes, as from other experience, is a problem-solving activity, and none of us can take his success in it for granted.
(3) Does White agree? In a significant passage he strikes a cautious note. "In serial expectational processes learning is possible, provided the sequence of decision, action, result, and interpretation takes place with speed sufficient to outrun significant changes in the objective circumstances. Learning can play an important role in providing accurate foresight and coordination of the decision maker's (amended) plan with the plans of other market participants." We note, first of all, that "learning is possible", not necessary, and that it depends on speed. Slow learners, we surmise, do not stand much of a chance.
The crux of the matter is that traders, by entering into those transactions from which they stand to learn most, may transform the "data" of our model, White's "objective circumstances." Are we to believe that in a world in which contracts are binding (no t?tonnement) disequilibrium transactions will not give rise to gains and losses which modify the distribution of wealth? Or that "false quantities" of goods produced in response to disequilibrium prices will not affect the freedom of subsequent action of their producers? Do we have to assume that in White's model good learners learn fast enough to prevent all malinvestment (presumably an objective circumstance), even that indulged in by the slowest learners?
Painless learning, alas, is possible only under the gentle guidance of the universal auctioneer.
(4) White, while dissociating himself from the Walras-Pareto model, wants us to accept the "affirmation that such an economy harbors a strong tendency toward an overall equilibrium." The word "tendency" denotes a constant general direction, while it is of the essence of a groping process that in successive periods,the groping takes place in different directions. In White's model there can thus be no trial and error. Nor is much comfort to be drawn from the statement that the said "tendency" is here to be interpreted "as the likelihood that the configuration of an economy (particularly its array of prices) will be near to a general equilibrium configuration." A configuration the data of which are continuously transformed as the result of disequilibrium transactions, can hardly be said to occupy any distinct "area," and thus does not permit us to describe any events as happening in the "vicinity," or at some "distance" from it.
(5) Where do we go from here? Without an auctioneer, what happens in each market as well as the movement of relative prices depends on the actual sequence of events, the temporal order of market processes. Theoretical generalizations about the outcome of such processes which fail to take account of the order in which events happen in markets are therefore unwarranted.
Fifty years ago, when the inadequacy of the Walras-Paretian general equilibrium model for our understanding of economic processes was first noticed, some Swedish economists, such as Lindahl[2] and Lundberg,[3] suggested process analysis as an alternative paradigm. Like much else that was of value in the 1930's, these ideas were swept away by the Keynesian revolution. Today we realize that these thinkers have paved the way for the notion of market process now widely accepted among Austrians.
But two points must be kept in mind. In the first place, a purely mechanical process model that has no room for choice, the subjectivism of expectations, and the interpretation of information, would be no improvement at all on general equilibrium. Secondly, different markets evolve different institutions which influence the sequence of events on them. There are even some real auction markets in the world today (wool). Hence the market process assumes different forms in different markets. We must study them with some care.
We can all agree with Lawrence White in seeing virtue "in encouraging diversity in the exploration of possible new opportunities," but we have to realize that such diversity implies the incompatibility of plans. There can be no competitive game without losers.
Ludwig M. Lachmann
I am pleased that Professor Lachmann has chosen to reply to my piece, for I believe that his reply clarifies his position. Part of the issue which he professes not to have understood was an uncertainty over whether he was prepared to affirm a belief in spontaneous order, i.e. affirm that markets do succeed in coordinating plans. It is now clear that he is not prepared to make any such general affirmation. We are told: "Without an auctioneer, what happens in each market as well as the movement of relative prices depends on the actual sequence of events the temporal order of market processes.
[T]he market process assumes different forms in different markets." One of my objectives was to show that Lachmann's writings should not, on the other hand, be construed as a blanket denial of belief in spontaneous order.
Professor Lachmann asks whether I agree with the last two sentences in his section (2). In the unqualified way in which they are put, I do not. It seems to me that it is not only possible for agents to learn, but in some cases it is also possible for the economist to predict the direction of plan amendment subsequent to feedback and interpretation. If pattern prediction of this sort were never possible, I do not see what sort of process analysis we could ever do. On the other hand, I would also answer in the negative to the three questions posed in the second paragraph of section (3).
In the first three sentences of section (4), Lachmann seems to impose on my discussion of the meaning of "a tendency toward equilibrium" a definition of the word "tendency"?namely, "a constant general direction"?that I explicitly eschew. I certainly did not want to suggest that a tendency toward an economic equilibrium is inconsistent with trial and error. I am puzzled that I could be so interpreted. I might add that I am puzzled also by the similar misinterpretation implicit in the last two sentences of Professor Lachmann's opening paragraph. The Hayekian equilibrium theory I and other Austrians would defend is quite distinct?in my mind at least?from the equilibrium theory of the Pareto-Hahn tradition.
Lawrence H. White
By specifying the contents of the "general affirmation of belief in spontaneous order" Lawrence White has helped me to understand at last what the quarrel is about and what it is that's wanted from me.
First a minor point: the difference between a tendency and trial and error. A point of semantics perhaps, rather than Austrian economics. To my mind one precludes the other. It may be possible ex post, in charting the course of a trial and error process of the past, to discover a tendency in it, but never ex ante. To say "Let us try and see, but only in a South-easterly direction" is to limit the scope of the search. See Joan Robinson in JEL, December 78, on the impossibility of finding equilibrium by trial & error (p. 1322).
Now the main point. Is the "spontaneous order" permanent? If not, how long does it last? Take the U.S. railroad system. In 1900 it certainly presented all the features of spontaneous order, coordinating the plans of millions of people, including shareholders receiving dividends from most of the railroads. But today? White says I can only master the ex ante perspective. But if we take the ex post perspective, was the U.S. railroad system a spontaneous order?
It puzzles me that White fails to see that, by pretending to see "spontaneous order" everywhere, we are playing right into the hands of our opponents who merely have to point to obvious instances of malcoordination to win debating points. Every case of malinvestment can be held against the market economy. Does it not show malcoordination? The "absence of universal futures markets" in Arrow and Hahn as an argument against the market economy makes sense only, but, alas, does makes sense, against such "universal affirmations" as I am now asked to subscribe to.
What has gone wrong? We have to distinguish between real phenomena and those ideal types we use to classify the former as "more or less - - - " Some markets coordinate plans better than others, and it is the job of us Austrian economists (and others) to study them and classify them. If we assert that "the market coordinates all plans" (always?), we substitute an affirmation of faith for what should be the outcome of serious study of the real world. In defending the market which certainly provides people with opportunities (they often miss) to orientate their plans to others, we must beware of claiming too much. Plans that often extend into the remote future, with details often as yet unspecified, cannot all be consistent with regard to every element. By the way, how can the plans of competitors be compatible?
In refusing to sign the blanket affirmation referred to I appeal to common sense. I know that among economists that is not much of an argument. It seems to me, nonetheless, that Austrian economics is more likely to prosper if common sense is not ignored.
I repeat as an empirical generalization: Learning from experience is a problem-solving activity, and none of us can take his success in it for granted.
General demand and supply equilibrium cannot serve as a "center of gravity", a source of permanent forces of constant strength as, under the impact of innovation, technical progress and simple changes of taste, relative demand and supply of various commodities are continuously changing. A planet whose composition and mass are undergoing continuous transformation could not exert a gravitational force of constant strength. If so, how can it be asserted that economic equilibrium forces, necessarily of varying strength over time, will always overwhelm and outlast all other forces?
Ludwig M. Lachmann
--------------
[1] "The Role of Expectations in Economics as a Social Science". Economica, February 1943. Reprinted in Capital, Expectations and the Market ,Process (Kansas City: Sheed An- drews & McMeel, 1977) p. 66.
[2] Erik Lindah Studies in the Theory of Money and Capital (London: Allen & Unwin, 1939) Part 1.
[3] Erik Lundberg Studies in the Theory of Eco- nomic Expansion (London: P. S. King, 1937).
