The Evolving Rationality of Rational Expectations: An Assessment of Thomas Sargent's Achievements (review)

The Evolving Rationality of Rational Expectations: An Assessment of Thomas Sargent's Achievements. By Esther-Mirjam Sent. Cambridge and New York: Cambridge University Press, 1998. x; 242 pp.

Esther-Mirjam Sent's book, in which she provides an account of the development of rational expectations theory from the perspective of Thomas Sargent's contributions, is a splendid contribution to the history of economic thought quite unlike other works in the field.

What Sent has done is to take seriously a perspective on science studies developed most thoroughly by Andrew Pickering in his book The Mangle of Practice (1995). Pickering sees science as essentially in time--the book's subtitle was Time, Agency, and Science. Pickering's perspective contrasts with that of many historians of science who focus on reconstructions of scientific activity as a sequence of rational moves by human agents in response to the resistances of nature, but it is also in contrast with many sociologists of science who reconstruct scientific activity as a sequence of moves made by agents in response to various interests outside nature. Pickering attempts to reconstruct the temporal process of science in which agents, equipment, and ideas, nature, experiments, interests, colleagues, laboratories, accounts, theories, and all manner of stuff are mangled to together in complex and unpredictable ways. One tells the story of a bit of science, then, from many different perspectives. It is not a simple story of representation nor is it a simple story of intervention but rather a complex dance of representation, intervention, reconfiguration, reconceptualization, and rerepresentation and reintervention.

Sent's Thomas Sargent is no simple, or realistic, figure. Sargent, who is real, in Sent's account does and is done by the scientific work he constructs and is himself constructed by that work as a scientist who does the work. She locates ten interdependent stories, and Sargent plays a role and can be thought of or reconstructed as contributing to work in each of these narrative lines: the stories involve the expiration of the Phillips curve, policy irrelevance, using available techniques, restoring symmetry (between agent and economist), optimizing over information, endogenizing expectations, making public predictions, countering bounded rationality, and incorporating vector autoregressions.

In Chapter 1, Sent presents all those various stories, which were involved in what we know now to call the rational expectations revolution in macroeconomics. Those ten stories are each involved, in some measure or another, in past reconstructions of the revolution. What Sent is doing though is suggesting that all these overlapping narratives must be invoked in various measures to tell a convincing story, all are involved in some mix-and-match way. But which story is the right one? Sent answers all, or none, or many, for it is the nature of the story that the real history is not a given but is constructed out of all the elements described.

Chapter 2 sees Sent providing a background to what is to follow, a discussion abut the probabilistic revolution, and how economists learned to endogenize randomness in economic theory itself, much as physics, in the quantum era, learned to explain probabilistically. Then Sent takes up, in the third chapter, the issues that had arisen with prediction and places Sargent at the intersection of many stories about how economics theorized prediction in the relevant periods.

Chapter 4 contains the material on the symmetry problem, or why the decision-making economic agent would be modeled as systematically having less information than the economist modeling the agent's decision making. Sent nicely locates this issue in a long line of writings in science studies concerning the deprivileging of the observer-scientist's account vis-à-vis the observed. This plays out in anthropology, and it plays out in two levels here as Sent problematizes her own accounts of Sargent's work by becoming a fallible actor in the narrative in which Sent questions Sargent about his...