Foundations of Economic Analysis

• 1. Introduction

  ›

  Foundations of Economic Analysis by Paul A. Samuelson opens with a motto borrowed from J. Willard Gibbs: "Mathematics is a language." That line appeared on the front page of a book published in 1947 by Harvard University Press. It announced a shift in how economics would be practiced. The book had begun six years earlier as a doctoral dissertation at Harvard, submitted to the David A. Wells Prize Committee under the subtitle "The Observational Significance of Economic Theory." Samuelson's purpose was not to collect mathematical techniques but to expose the skeleton beneath them. He believed that theories of consumer behavior, market equilibrium, business cycles, and welfare economics were all expressions of the same underlying logic. Two principles ran beneath everything: that economic agents maximize, and that economic systems settle into stable equilibria. The questions the book planted were stark. Can economics produce theorems that could actually be tested against real-world data? And what is the minimum apparatus needed to generate them? The book's 353-page body would push toward an answer, running from a sales-tax increase on a single firm to the turbulence of business cycles and entire economies.

  34

• 2. One Theorem, Many Fields

  ›

  Samuelson admitted he had been prodigal in proving essentially the same theorems over and over. Moving from consumer behavior to the economics of a firm, he kept arriving at the same formal structure. The same pattern held when he turned to international trade, business cycles, and income analysis. He found the principle behind this in E. H. Moore, an American mathematician. Moore had argued more than thirty years earlier that analogies between theories imply a more general theory beneath them. Samuelson set out to trace that general theory through economics.

  Part I of the book pursues this through maximization. For an individual household or a firm, meaningful theorems about behavior reduce almost entirely to equilibrium conditions at a maximum. A consumer maximizes utility; a firm maximizes profit. The calculus behind these conditions operates at a high level of abstraction. It yields testable predictions across a wide range of problems, from production economics to index numbers and consumer surplus.

  Part II confronts a harder case. When individual units aggregate into a market or a whole economy, the symmetry conditions required for direct maximization vanish. A market cannot be said to maximize anything in the same direct sense. Samuelson's move was to replace maximization with stable equilibrium as the source of operationally meaningful theorems for systems. Two mathematical appendices totalling 83 pages gather the tools behind both parts. The first covers maximization conditions and quadratic forms; the second covers difference equations for what Samuelson called the dynamic economist. The formalization of comparative statics, the method for calculating how a parameter change shifts an equilibrium, is among the most concrete achievements of this framework.

  34

• 3. Comparative Statics

  ›

  Timothy Kehoe, writing in 1987, placed the formalization of comparative statics at the center of what Foundations achieved. The method asks a targeted question. When one parameter of an economic system changes, say a tax rate rises, how does the equilibrium shift in response? Samuelson's treatment made the method precise and general in a way no earlier economic writing had. The most celebrated result to emerge from this work is the correspondence principle.

  The correspondence principle states that the stability of an equilibrium implies testable directional predictions about how that equilibrium moves when parameters change. Stability and comparative statics are not independent; they are two faces of the same structure. Samuelson states the connection plainly: stability imposes directional restrictions on the movement of a system. Those restrictions are what allow comparative statics to yield real predictions.

  Samuelson also carried the Le Chatelier principle from physics into economics and gave it a new formulation. In a maximum condition of equilibrium, adding auxiliary constraints that leave the initial equilibrium unchanged reduces the response to any parameter change. A practical implication: factor-demand and commodity-supply elasticities are lower in the short run than in the long run. Fixed costs constrain the short-run response in a way the long run is free to work around. Tools from mathematics, including Lagrangian multipliers, were given an operational economic interpretation throughout this framework. Samuelson introduced what followed as an attempt at a brief but fairly complete survey of the whole field of welfare economics.

  134

• 4. The Social Welfare Function

  ›

  Chapter VIII of Foundations runs to 51 pages. At its core is what became known as the Bergson-Samuelson social welfare function. The function can represent any logically possible ethical belief system. The only requirement is that the system be capable of ordering feasible social configurations as better than, worse than, or indifferent to one another. No particular ethical view is assumed; the apparatus works for any.

  Samuelson noted that theorems derived in welfare economics have a different character from elsewhere in the book. They are deductive implications of assumptions that are not themselves refutable by empirical data. This means they fall outside the strict definition of meaningful theorems he applied elsewhere. He was explicit about the limit, even as he demonstrated the function's power within it.

  Pareto optimality receives its most definitive treatment in this chapter. This is the condition under which no reallocation can make one person better off without making another worse off. Samuelson described this as illuminating the germ of truth in Adam Smith's doctrine of the invisible hand. Kotaro Suzumura, writing in 1987, affirmed the Bergson-Samuelson social welfare function as logically impeccable. Kenneth Arrow's 1983 assessment of Foundations placed it in a category of one.

  35

• 5. Redirecting a Discipline

  ›

  Kenneth Arrow wrote in 1983 that Foundations was the only example he knew of a doctoral dissertation that is a treatise. Equally, he said, it was the only treatise whose originality in every part entitled it to be accepted as a thesis. The two descriptions were the same claim from different directions. Richard N. Cooper, writing in 1997, said the book had drastically redirected the advanced study of economics. The direction was toward greater and more productive use of mathematics.

  The Nobel Prize committee cited Samuelson's development of static and dynamic economic theory. It also noted his contribution to raising the level of analysis in economic science. Roger Backhouse, writing in 2015, placed Foundations at the origin of a methodological shift. As economics became progressively more mathematical, graduate students were increasingly expected to construct formal models of maximizing consumers and firms. Foundations came to be widely seen as the canonical exposition of those methods.

  Backhouse noted that Samuelson credited Joseph Schumpeter, Wassily Leontief, Gottfried Haberler, and Alvin Hansen with teaching him modern economic analysis. The main influence on Foundations itself, however, Samuelson identified as E. B. Wilson, a mathematician and physicist. When Samuelson expanded the book in 1983, the enlarged edition included a new 12-page introduction. It also added a 145-page appendix covering developments in analytical economics since 1947. Samuelson closed the original edition with an aspiration for comparative dynamics. He wanted it to aid problems as modest as the behavior of a single small commodity. Eventually, he wrote, it might even reach the majestic problems of economic development.

  245