Samuelson’s Activities

Samuelson moved straight from teaching undergraduates at Harvard to teaching at MIT, where in his first year he taught mathematical statistics and, probably, courses in economic theory and mathematical economics.³⁴ Samuelson remembers being assisted in his statistics teaching by his research

table 16.1 Teaching Commitments Listed in MIT Course Catalogue (MIT 1941,

p.

assistant, Leonid Hurwicz, and their introduction of a new and controversial grading system.^e

Even more melodramatic was the new Hurwicz-Samuelson grading system for my first regular statistics course. MIT engineers have always been notorious whiners. They are grade chasers beyond Philadelphia barracks lawyers anywhere. One of us—I will point no finger— said: “Let's add a hard extra credit exam question, with the proviso that it can only raise, but not lower, your grade.” All hell broke loose when undergraduate commerce course nerds learned that their exam mark of 115 put them below the median of the class grades.

Samuelson remembered the unpopularity of the Hurwicz-Samuelson grading system as potentially jeopardizing Hurwicz's future. However, Samuelson later wrote, “Leo had little to lose. It was my tenure and future life-time career that dangled on the razor's edge.”³⁶

The following academic year, as shown in table 16.1, his teaching load remained low, possibly because he was also teaching mathematics to Navy officers, a teaching commitment not evident from the course catalog, but which he remembered doing and which is consistent with the priority attached to military training during the war. This might have been where he began to think about the mathematics of ballistics and controlling gunfire that he was to work on later in the war.

By 1942—43, any concessions he had received as a new assistant professor were over, and his load rose to four and a half courses: he took on a second

e. Hurwicz's appointment is discussed in chapter 17 this volume. course in economic theory, a course in business cycles, and he shared a course in statistics with Harold Freeman.^f The year after that, he dropped the math­ematical economics and took on public finance. Also, starting in the fall of 1942, Samuelson taught a course in international economic relations, run­ning throughout the year, at the Fletcher School of Law and Diplomacy. The first graduate school in international affairs in the United States, the Fletcher School was based at Tufts University, in Medford, which was not far from MIT, with many of its teachers coming from Harvard. The course focused on the relationship between politics and trade: “the ways in which economic life is affected by the existence of political boundaries and of the ways in which political relations between nations are affected by economic factors.”³⁷ Though the theory of international economics, covered in the first semes­ter, related closely to Samuelson’s published work, he would have had to engage in a different type of analysis in order to cover how political relations depended on economics.

A major attraction of MIT was that Paul and Marion could continue to live on Ware Street, just two blocks from Harvard Yard, and he could main­tain contacts with friends and colleagues still there. He grew increasingly close to Hansen, attending the Fiscal Policy Seminar and, from August 1941, joining him in a fortnightly commute to Washington, D.C. Though there were occasional activities at Harvard, the university had become increasingly empty. In a rare account of his life once war had broken out, he wrote to Bergson:

I hope very much that I will be seeing you pretty soon. Both Marion and I have been wondering just what you all have been doing since we saw you last fall. I suppose that Judy must be running around in

f. Note that it is possible the catalogue, for which material would have been prepared several months in advance of publication, may not reflect who actually taught the courses. However, student notes confirm that much of this is correct.

g. It is tempting to link his interest in teaching potential diplomats to his own interest, as an undergraduate, in a diplomatic career, though it is as likely that he welcomed an additional source of income.

h. See chapter 21 this volume.

the Texas sun by now. It looks as if Harvard will be a morgue by next year—as somebody in Washington put it, “Pretty soon there will only be Paul Sweezy and enemy aliens left.” The graduate enrollment is expected to shrink to nothing, and all the Ec. A staff are moving down to Washington in the hope that this will improve their draft status. At the same time they are working very hard in other directions with the result that a bumper crop of babies is already on the way.

Close friends with whom he lost contact for a very different reason were Shigeto and Masako Tsuru. In June 1942, Shigeto wrote to him, “It is with the bitterest regret that we must leave without seeing you two. By June 12 we shall be on the Atlantic headed toward Japan.”³⁹ As a result of conver­sations with Harry Dexter White, a high-ranking official in the Treasury whom he had known from his days at Lawrence College, Shigeto expected Japan to lose the war and wanted to be there so that he could take part in its postwar reconstruction.⁴⁰

The opportunity came suddenly, on June ι, when he received a telegram saying that they could be part of an exchange between America and Japan. This offer came when Shigeto was in the middle of grading term papers for Seymour Harris, in which some of the students had reminded him of his status in the United States by writing “Remember Pearl Harbor!” on the first page of their answer books. He and Masako had five days to sort out their affairs, and though they had managed to see some of their friends, spending the previous evening with Paul Sweezy and Leontief, they were able to see neither Paul nor Marion, who were away visiting Marion's family in Wisconsin, before they caught the train at midnight from Boston's South Station. The Tsurus were among 1,500 Japanese who traveled on a Swedish ship, the Gripsholm, to Lourenςo Marques in Portuguese East Africa (now Maputo in Mozambique), where they were exchanged for a group of Americans. They then sailed on a Japanese ship to Japan, arriving in August 1942.

Having too little time to sort out their possessions, Shigeto and Masako left Samuelson with several tasks, to which end Shigeto enclosed a letter, “To whom it may concern,” authorizing Paul to act on his behalf “on all matters during my absence which is to extend at least for one year from to-day.”⁴¹ The implication that the war might last at least one year suggests that the advice he received from White might have been overly optimistic. Shigeto asked Paul to retrieve the camera, which as part of the restrictions placed on Japanese nationals, he had deposited at the Cambridge police station; and he asked Paul to look after money owed him by the Harvard Coop and various fees he was owed by Harvard University.

One of the events that did take place at Harvard was a conference held on March 5, 1942, at the Littauer Center, on urbanism and the problem of towns and cities. The first session dealt with economic determinants of urban devel­opment, to which Samuelson presented a paper on “The Business Cycle and Urban Development.”⁴³ He acknowledged that he was no expert on urban problems, but he believed that his work on the business cycle enabled him to challenge some beliefs about urban problems that had arisen during the Depression years. One of these was that the long-term trend toward urban­ization had been reversed. Another was that unemployment was caused by problems that were specific to large cities: that “urbanism” was the problem.

There was, Samuelson argued, no evidence for a change in the long-term trend for people to move from the countryside to cities. There had been a movement the other way during the 1930s, but this was no evidence of a change in trend, for it was the result of people being unable to get work in the cities, moving back to the countryside to stay with families and help out on farms. Once back on the farm, they would be classified as “in work,” even though they probably contributed nothing to farm output. It was thus an illusion to think that unemployment was an urban problem, for rural unem­ployment was disguised. As soon as prosperity was restored, as would happen after the war if a full-employment policy were followed, people would return to the cities.

However, the expertise Samuelson brought to this topic was not just that of a statistician, for his conclusions also reflected his theoretical work on unemployment. The notion of disguised unemployment was theoreti­cal, discussed in his article with Nixon, but more important was the idea that Hansen and others had developed: that unemployment was the result of economy-wide factors, notably the balance between saving and investment. This determined the overall level of unemployment, and the only question was how that was distributed, whether between open and disguised unem­ployment or between rural and urban locations. He argued that unemploy­ment was not a problem pertaining to individuals.

[A] single individual, if he studies Dale Carnegie, takes a correspon­dence course, works over-time, and does the thousand and one things which from time immemorial have been thought of as leading to success, can without doubt succeed in finding employment. But all cannot do so. If everybody became a go-getter, none would be much better off than before. One man can see the parade better by standing upon a chair, but what works for one will not work for all simultane­ously, and the single individual who secures a job by self-improvement tends to do so by displacing another worker.⁴⁴

It was because unemployment was not an individual problem that work and Social Security programs had been implemented. Cities were like individuals in that advertising by one chamber of commerce might improve the situation for an individual city, but such activities would not reduce national unem­ployment. He presented clearly the view that what is true for the individual is not necessarily true for the whole, of which the individual forms a part.

Later in March, Harvard was the place where Samuelson had his only encounter with John von Neumann, then working with Oskar Morgenstern on what was to become The Theory of Games and Economic Behavior (1944).⁴⁵ In January, Haberler had written to Samuelson describing the topic on which von Neumann, who had been invited to give a lecture on some aspect of mathematical economics, had proposed to talk:

He suggested a talk on a system of equations or rather inequalities of production and distribution contained in an article he wrote a few years ago in Karl Menger's mathematical colloquium. He says that that would give him an opportunity to show that mathematical prob­lems of a much different nature than in physics are apt to be raised by economic problems.⁴⁶

Though Haberler doubted whether this was the best topic for the lecture, it is hard not to think that Samuelson would have been excited by it, for it directly challenged the approach that he, following Wilson, had adopted in his thesis and in the book he was writing, which involved making use of the common mathematical structure they believed underlay problems in physics and economics. He remembered Schumpeter being “pleasantly excited” by von Neumann's claim.⁴⁷ Samuelson recalled the incident in his Nobel lec­ture, likening the encounter to that of David and Goliath:

This sets the state for my encounter with Goliath.... [V]on Neumann gave a lecture at Harvard on his model of general equilibrium. He asserted that it involved new kinds of mathematics which had no rela­tion to the conventional mathematics of physics and maximization. I piped up from the back of the room that I thought it was not all that different from the concept we have in economics of the opportunity­cost frontier, in which for specified amounts of all inputs and all but one output society seeks the maximum of the remaining output. Von Neumann replied at that lightning speed which was characteristic of him: “Would you bet a cigar on that?” I am ashamed to report that for once little David retired from the field with his tail between his legs. And yet someday when I pass through Saint Peter's gages I do think that I have half a cigar still coming to me—only half because von Neumann also had a valid point.”⁴⁸

In a second account of the incident, Samuelson claimed he was, perhaps uncharacteristically, overawed by von Neumann's eminence.

When he claimed it [his equilibrium model] meant economics had to find a completely new mathematics, I objected, saying it sounded to me like constrained-maximization theory a la Newton and Weierstrass (and, one would add today, a la Kuhn and Tucker). Von Neumann retorted belligerently, “Will you bet a cigar on that?” I was a brash young man but not so foolhardy as to grapple with the world's greatest mathematician. Still, on leaving the seminar room, I was heard to whisper Galileo-like, “Nevertheless, the world does move: cherchez la maximization.” After these many decades I claim one cigar.⁴⁹

He summarized his position, in 1989, by saying that though von Neumann's innovations had led to the development of “indispensable modern methods” (nonlinear programming, convex set theory, game theory, and optimal-con­trol theory) the only real novelty he could see in von Neumann's work was “the philosophical complications introduced by games involving more than one person.” Beyond that, he could see no novelty in this “so-called non­physics mathematics.”