Turning Against Chicago Economics

Samuelson's later recollections of what he was taught at the University of Chicago were mixed. He praised the education he had received—both the Hutchins curriculum (though not the Great Books element) and his train­ing in economic theory, the rigor of the latter equipping him much bet­ter than his contemporaries for what they were to encounter in Harvard's graduate program.

Handbook on Money, of the Fisher-Marshall-Pigou MV = PQ paradigm can be found in my class notes and memories.²

In an unpublished note he was even stronger, adding Oskar Lange to those he had conversations with on returning to Chicago. He claimed that “With the exception of eclectic Jacob Viner, there were essentially no advances on 1911 Fisher or 1924 Keynes-Robertson.”³ They were taught that in an ideal world, in the long run, money did not matter: it was “neutral” in that changes in the money supply would have no effect beyond leading to higher prices.^a It was recognized that “under dynamic conditions,” before a new equilibrium was established, there might be effects on output, but these would be “relatively transient aberrations.” However, these qualifications to the simple quantity theory tended to be taught in courses on applied topics, with the result that there was a dis­connect between such discussions and the theory they studied.

From 9 to 9:50 A.M. we presented a simple quantity theory of neutral money. There were then barely ten minutes to clear our palates for the 10 to 10:50 discussion of how an engineered increase in M would help the economy. In mid-America in the mid-1930s, we neoclassi­cal economists tended to be mild inflationists, jackasses crying in the wilderness and resting our case essentially on sticky prices and costs, and on expectations.

Returning to the 9 o'clock hour, we thought that real outputs and inputs and price ratios depended essentially in the longest run on real factors, such as tastes, technology, and endowments. The stock of money we called M.... An increase in M—usually we called it a dou­bling on the ground that after God created unity he created the second integer—would cause a proportional increase in all prices (tea, salt, female labor, land rent, share or bond prices) and values (expenditure on tea or land, share dividends, interest income, taxes).⁴

This memory was an important part of Samuelson's story of his own life: his precocity had enabled him, as a teenage undergraduate, to understand what pre-Keynesian monetary economics was really like—unlike those who had started only a little later than himself and had to use their imagination. He

a. Of course, as Samuelson had known when he wrote the international economics essay discussed in chapter 4 this volume, the Keynes of 1923 explicitly denied that one could live in the long run. This account does an injustice to Fisher's 1911 book. argued that other economists who had tried to construct a picture of clas­sical economics were “in the position of a man who, looking for a jackass, must say to himself, ‘If I were a jackass, where would I go?' ” In contrast, he wrote:

Mine is the great advantage of having once been a jackass. From 2 January 1932 until an indeterminate date in 1937, I was a classical monetary theorist. I do not have to look for the tracks of the jackass embalmed in old journals and monographs.

In writing this he no doubt had in mind that he had arrived in Chicago before even Milton Friedman, three years older but who had arrived only as a graduate student in the autumn of 1932, and that he had stayed there longer.⁶

However, though Samuelson stuck by this picture of his Chicago teachers as being blinkered in their theorizing, he sometimes acknowledged that his teachers had gone beyond this. In the 1930s it was obvious, he claimed, that the pure classical theory, according to which money was neutral and there could be no deficiency of aggregate demand, was useless in explaining the Depression.

I knew 100 people without jobs in 1931—1934 and 100 with jobs. The groups would never voluntarily change places: the latter felt very lucky. The former, about equal in ability, felt unlucky. That's not what happens when auction markets equate supply and demand.^b

Samuelson credited Simons with having sensed the “liquidity trap”—the notion that there may be some rate of interest at which the public is willing to hold whatever money is issued, and hence a floor to the rate of interest. And he credited Viner with being eclectic and for being responsible for the

b. When Samuelson wrote these words he will have been aware that many economists with whom he disagreed on macroeconomics modeled unemployment as voluntary—as an optimal response to perceived wages and prices.

empirical research that substantiated the idea that people were willing to hoard money, making it hard for people to borrow.⁷

While there is no reason to doubt that he was taught the classical theory in which equilibrium in competitive markets results in a world in which prices are proportional to the money supply, it was an oversimplification of monetary economics as found at Chicago at this time: there were attempts to develop a theory in which expansion of the money supply would have effects on production and employment.⁸ That Samuelson was aware of this at the time is shown by his international economics paper, in which he went through theoretical arguments about why a rise in the money supply would not raise prices proportionally.

Keynes's monetary economics was a significant part of his education. The closing paragraphs of that essay, about how reducing the fraction of the money supply backed by gold might be inflationary, links to what came to be known as the “Chicago Plan” for monetary reform, signed by several Chicago economists. Their main proposal was for “100% money.”⁹ If banks were required to hold reserves to back all their loans, it would be possible to sepa­rate the functions of commercial bank lending from the creation of money. The creation of money could become the sole responsibility of government, which could use it to counter depression.

This proposal, made in a pamphlet signed by a group of Chicago econo­mists in 1933, though mostly written by Simons, was part of a much wider debate over monetary policy to stabilize the economy.¹⁰ The previous year, a similar proposal had emerged from a conference organized by international relations specialist Quincy Wright and involving economists from Columbia and Harvard, as well as Viner, held in January 1932 and published in July.¹¹ The following year, Harvard's Lauchlin Currie made a similar proposal, and Simons incorporated the idea into another, more wide-ranging pamphlet, A Positive Program for Laissez-Faire (1934). Here, monetary proposals were accompanied by an extensive program of anti-monopoly policies designed to restore competition. The view that monopoly was an important cause of the Depression was widespread, but whereas other economists considered that the growth of big business was a feature of American capitalism that policy­makers needed to take into account, Simons proposed to remove the prob­lem. His vision of liberalism required a strong government to break up large firms in order to bring about a world in which competitive private enterprise could operate effectively.

A teacher who was rarely mentioned in Samuelson's recollections of Chicago monetary theory was Paul Douglas.

Unemployment with Aaron Director, but whereas Director subsequently became more skeptical about labor unions, moving into the Knight circle, Douglas became stronger in his support for collective solutions. At the begin­ning of Samuelson’s sophomore year, he produced The Coming of a New Party (1932), a book that began by arguing that, while individualism may have made sense in the early frontier society, American industry had changed and with it American society. Industry was much more hierarchical, and owner­ship more concentrated, with the result that opportunities for social mobility were much fewer. The notion, supported by Harvard’s Thomas Nixon Carver, that workers would, through their savings, come to control industry was “almost a grotesque misunderstanding”: the top 2 percent of the population controlled 70 percent of property.¹²

The direction of industry is concentrated in the hands of a comparatively small number, while the major portion of the wealth and the surplus income is held by substantially the same group joined with a fringe of wealthy idlers. It is still not absolutely impossible for an individual “out­sider” to break into the charmed circle, but this tends to become ever more difficult because of the great head start which the sons and daugh­ters of the wealthy “insiders” increasingly possess.¹³

Thus, given that it was impossible for most people to rise to the top, the great majority of people should focus on improving their position as workmen and should combine with others to improve their collective situation. They should turn to trade unionism and political action. It was an attack on the individual­ism to which Director was turning. After outlining the needs of various groups of the population, Douglas turned to the political means by which these might be achieved.

Two years later, Douglas completed The Theory of Wages (1934), published when Samuelson was taking his intermediate economic theory course.

It is not known how much of this book's contents Samuelson would have been exposed to, though it would seem unlikely that a course taught by Douglas with the aim of exposing students to systematic economic theory would not cover the very standard theoretical ideas on which it relied, or that Douglas could have failed to mention the idea that concepts such as elas­ticities of demand and supply could be quantified. However, Samuelson did acknowledge a debt to Schultz, then working on empirical demand analysis, as the economist who introduced him to the idea of operationalism. This notion was to become central to Samuelson's PhD thesis and his Foundations of Economic Analysis (1947a), interpreted as the idea that meaningful theorems were ones that were capable of being shown to be false.^d In the preface to The Theory of Wages, Douglas also cited Bridgman in a context that is significant, given Samuelson's later commitment to the analysis of economic aggregates.⁶

Douglas was confronting the criticism that he erred in trying to explain the overall rate of wages when he should, instead, be seeking to explain wages for “an indefinite series of labor groups” and, similarly, the payments to dif­ferent types of capital and land. His first response was that there was suf­ficient substitutability, or “transferability,” between different types of labor and between different types of land (and capital was fairly homogeneous) that his approach made sense. In addition, it was legitimate to explain wages by starting with a theory of the basic rate of wages and then combining it with a theory of wage differentials that explained differences from the basic rate. He then turned to Bridgman:

It would be almost impossible to measure the incremental productivi­ties of this infinite series of sub-groups or to determine their supply curves. In view of the present inability to test the validity of this great sub-division of the factors, I can only consider this suggestion to be at present, in the words of Professor Bridgman, a non-operational con­cept. From the standpoint of scientific progress, we should primarily concern ourselves with problems which we can solve.¹⁴

c. An important analytical device in this work was what came to be known as the Cobb- Douglas production function, worked out with the assistance of Charles Cobb, a mathematician. Biddle 2012 provides a history of this function.

d. This is slightly different from Bridgman's definition. This is discussed in chapter 14 this volume.

e. As was mentioned elsewhere, Samuelson took Labor Problems with Douglas after having taken it with Director, making it very likely that he would have read this.

Not only does this remark defend aggregate analysis as being operational, it also suggests that Bridgman’s operationalism, implicitly defined in terms of testability much as Samuelson was later to do, was well known to Chicago economists at this time. No citation or explanation of the remark was thought necessary. Given Samuelson’s later attachment to the idea of being his intel­lectual grandson, it is interesting to note that Douglas went on to reinforce this point by citing a claim made by the physicist Willard Gibbs:

It will be noticed that I have treated the marginal productivity and supply curves for labor and capital in society as a whole and not for particular industries and plants. This has been done in part because as Willard Gibbs once remarked “the whole is simpler than its parts” and because it has seemed to me to be the more significant problem.¹⁵

Douglas was citing Gibbs to justify looking at the economy as a whole with­out looking at the individuals of which it is made up, an approach Samuelson was to take when he turned to the problem of the business cycle.^f

The Theory of Wages was a work for economists; its findings, such as its esti­mates of the responsiveness of labor demand to wage rates, might have policy relevance, but it did not directly address the problem of mass unemployment. However, halfway through Samuelson’s final year in Chicago, Douglas did produce such a book, Controlling Depressions (1935).¹⁶ This offered an eclectic survey of theories of the cycle. His analysis rested on a distinction between “initiating” and “cumulative” causes, the latter being most important.^g If, for some reason, there was a downturn, there could arise a cumulative break­down. When businesses cut back production, incomes of their workers and suppliers were reduced. Retail sales then fell, causing retailers to purchase less, producing an accelerated decline in production. Such a downturn could be initiated by a variety of factors. For example, structural factors such as the course of invention or population growth might cause a slowdown in the growth of consumption, which would lead (through the accelerator) to fall­ing investment and hence to depression. Though he did not use the phrase, Douglas clearly saw aggregate demand, and hence monetary and fiscal policy, as playing a role. While it might be right to balance the budget over the

f. Using the terminology that became fashionable in the 1970s, Douglas was saying that macroeconomics did not need micro foundations.

g. This distinction was not uncommon in the business cycle literature. Its most famous supporter was Ragnar Frisch (1933), who distinguished between the “impulse” and “propagation” problems: between the shocks experienced by the system and the mechanisms whereby those shocks were propagated through the system. entire business cycle, it was, Douglas claimed, perfectly proper for it to be unbalanced in depressions. It might be correct that “every past depression ha[d] sooner or later turned into the spring of revival,” but this might take a very long time.¹⁷ He criticized strongly the view that it was right to leave recovery to private enterprise.

If we have been successful in muddling through all of these past depressions, say the apostles of laissez-faire, we will have similar luck with the present. Let us simply keep quiet and events will ultimately right themselves.... There are two answers to this optimistic appeal to history. The first is that even though we have frequently managed to get out of these depressions, it frequently took a long time to do so, with much accompanying misery.. The second answer is that if

one studies these past depressions more closely, recovery in many cases may well have been more accidental than inevitable.. On several

occasions the proximate and immediate cause of the pick-up was the apparently accidental coming of an external savior.¹⁸

These external saviors could be new inventions, stimulus from the rest of the world, or war.

Picking up a view that was widely held by economists at the time, Douglas questioned the health of the capitalist system, challenging the prevalence of theorizing about equilibrium.

There are present in our present economic system latent tendencies which may, and more or less periodically do, result in a cumula­tive disequilibrium. This feature of our system has been too much ignored by the orthodox economists, who have tended to concentrate their attention upon the forces making for equilibrium in the field of prices, value, and the distribution of the national income among the factors of production. The economists have used up their vital energy in explaining how the economic system works. They have not devoted nearly as much attention to how it fails to work, or how it operates in a viciously cumulative fashion. The forces of equilibrium are real, but they are only half the story. There are also forces making for break­down. And intellectual interest in or emotional enthusiasm for the smooth-running features of competitive capitalism should not blind us to the other side of the story.¹⁹

Samuelson may not have read these particular passages, but they show that, despite his portrayal of Chicago as a place of darkness, this view was repre­sented among the teachers to whom he was close. He was later to confess that, as an undergraduate at Chicago, he had read related arguments in a book by a friend and former colleague of Douglas, John Maurice Clark; some of the ideas he had learned from the book proved very important to him, even though he was at the time very critical of them.^h Samuelson was reading very widely and was encountering and reacting negatively to work toward which he would later be very sympathetic.