A Harvard Thesis
Though he had become an assistant professor at MIT by the time the thesis was submitted, it was, as Samuelson himself pointed out, essentially a Harvard thesis. This is trivially true, in that a work such as this could not have been changed fundamentally in the last few weeks before its submis- sion.33 However, there is a much deeper sense in which it was a Harvard thesis.
As Samuelson repeatedly stated, his mentor was Wilson, who guided him continually. The earliest parts of the thesis, on consumer theory, involved tackling problems that Wilson had himself discussed. It was Wilson whose inspiration lay behind Samuelson's focus on analogies with physical science— notably thermodynamics and the Le Chatelier Principle—but also biology, where, though he might cite Lotka, Wilson's interest in these topics was never far away.p. He cited Moore in this chapter (Samuelson 1940a, p. 233) and had cited Schultz, his former teacher, earlier in the thesis.
q. Most economics continued to be based on concepts of stationary equilibrium. It was not until the 1970s and 1980s, for example, that concepts of dynamic, stochastic equilibrium, such as Samuelson discussed in this chapter, became central to macroeconomic modeling.
However, Wilson was not the only influence for the thesis. In Schumpeter, Leontief, and Haberler, Harvard had a core of experts in the type of theory Samuelson was systematizing. Schumpeter might lack Leontief’s mathematical skills, let alone those of Wilson, but he was an enthusiast for mathematical economics, and though his own work took him in a more historical direction, Samuelson’s Foundations of Analytical Economics could be seen as providing the mathematical extension of Walrasian theory, so admired by Schumpeter, that he could not himself achieve. Though the thesis kept largely clear of business cycle theory, aside from some almost incidental references when Samuelson was discussing dynamics, his work with Hansen also found echoes in the later parts of the thesis.
Discussions with his contemporaries are no less significant because they are harder to pinpoint.The thesis was also a Harvard thesis in that it was a product of Harvard’s “interstitial academy”—its profusion of academic spaces that arose outside established disciplines, the most important of which was the Society of Fellows, with which Bridgman, author of the term “operationalism,” was connected. The interdisciplinary space of the Society of Fellows is reflected in the authorities Samuelson cited. Lawrence Henderson was not cited in the thesis, but as Samuelson later noted when revising the thesis for publication, his discussions of equilibrium echoed ideas about which Henderson would regularly have been talking during the three years Samuelson was a junior fellow. This is quite apart from Henderson’s interest in Pareto, whose theory of economic equilibrium Samuelson had developed. In the published version of the thesis, Henderson’s The Order of Nature was one of the first works cited.34 Samuelson’s friend, the chemist E. Bright Wilson (another junior fellow, not to be confused with Samuelson’s mentor) would have discussed the physical analogies with him (the Le Chatelier Principle, to which Samuelson attached great importance, was derived in the context of chemical reactions).35 Perhaps most important, the most heavily cited authority in the final chapter of the thesis, which offers the most general, and arguably most fundamental theory, is George Birkhoff, a senior fellow. Even if Samuelson did not attend his lectures on differential equations (though it seems very likely that he did), they were in close touch.
Of course, not all the ideas upon which Samuelson drew came from Harvard. Samuelson might be critical of Frisch in the context of consumer theory, yet when it came to discussions of equilibrium and dynamics, Frisch was the economist with whom he engaged more closely than anyone else, adopting some of his concepts. Samuelson had developed his ideas on consumer theory against the background of a literature in which the articles by Hicks and Allen were central.
His articles show that Samuelson’s ideas were already well formed before Hicks's Value and Capital was published in 1939, but he engaged with the book, particularly in discussing dynamics, where he treated Hicks in the same way as he treated other authorities: as an economist whose work needed placing on a more rigorous foundation.Foundations of Analytical Economics was written in a few months, in a hurry, drawing on the journal articles he had written while a graduate student and as a junior fellow. It was, however, far more than a compilation of those articles. It proposed a new way of doing economics, to which certain types of mathematical analysis were central. When published as a book seven years later, it came to define the way much economic theory was done." The derivation of comparative statics results, often using the mathematical methods he had pioneered, became standard practice. The same was true of his chapters on dynamics, though here the take-up of some of the methods he discussed (notably dynamic, stochastic equilibrium) was slower. Economists would turn to his book not just for its economic analysis but also as a primer on the mathematical methods they used. However, though he continued to work on this material, revising the manuscript for publication and writing new articles on dynamic theory, he was shortly to leave Harvard. Partly because of that move and partly because the United States became involved in the war then engulfing Europe, his career was to change dramatically. The specialist in mathematical economics was to become a leading figure in the development of Keynesian ideas, and he was to become the economist from whom an entire generation of students was introduced to economics. The next stage of his biography is to see how that change came about. It is probably little exaggeration to say that everything followed, in one way or another, from his move from Harvard to MIT.
r. The transition from thesis to book is described in chapter 22 this volume.