Nobel Winners Saved Macroeconomics After Keynes
In the midst of great macroeconomic uncertainty, the Nobel Prize in Economics has been awarded to Thomas Sargent, of New York University, and Christopher Sims, of Princeton University, for work on “empirical macroeconomics.” Sargent and Sims are both superb scholars whose work has molded macroeconomics. They helped destroy the false certainty of an older Keynesian orthodoxy, and did their best to build more robust tools that shed light on public policy over the business cycle.
Sargent and Sims are part of a small cadre of intellectual rebels who have pointed out the logical inconsistencies buried within seemingly impressive Keynesian models of macroeconomics. In the early 1960s, Keynesian insights had been used somewhat successfully to iron out the business cycle. President John F. Kennedy supported a Keynesian plan to lower taxes to stave off a recession, and the economic slump duly diminished.
But economists were seeing flaws in such strategies. Milton Friedman (the 1975 winner of what’s officially known as the Bank of Sweden Prize in Economic Sciences in Memory of Alfred Nobel) was perhaps the most visible early critic, but Edmund Phelps (the 2006 laureate) was also particularly prescient in pointing out that the Phillips curve, an empirical relationship found between unemployment and inflation, was unlikely to be stable over the long term.
Yet while Friedman and Phelps grasped the failings of Keynesianism, they didn’t provide an alternative set of tools. And who could blame them? With macroeconomics, there are barriers to insight. Micro-economists have vast numbers of essentially independent observations. But macroeconomists must observe only a limited number of generally idiosyncratic economic declines, such as the one we face today.
Sargent and Sims tried to build something new from the wreckage of Keynesianism, and are being honored for their contributions to empirical work. Sargent was an early leader of a movement often labeled as “rational expectations” macroeconomics, whose great contribution was to build macroeconomics from the ground up, starting with assumptions about individuals and firms and then examining the implications for the larger economy. That movement has become mainstream, partly due to the work of Sargent, who showed how to bring theory to data.
For instance, the prize committee cited Sargent’s 1971 “note” showing that reasonable assumptions about consumer rationality mean that typical tests of the connection between wage growth and inflationary expectations made little sense.
The committee particularly singled out Sargent’s 1973 paper, which “carried out the first successful economic estimation under rational expectations.” This paper responded to criticisms from Robert Lucas Jr., another Nobel-winning economist, about existing empirical work on the natural rate of unemployment, and it delivered tools for estimating a macroeconomic system. The paper elegantly illustrated the profound differences that result from assuming rational, as opposed to adaptive, expectations, and it found little support for a long-run Phillips curve. This work continued with Sargent’s “A Classical Macroeconometric Model for the United States” in 1976.
Although this year’s prize is focused on “empirical macroeconomics,” the committee also discussed Sargent’s major theoretical contributions to economic policy making, which is cited even more often than his empirical work. Together with the economist Neil Wallace at Pennsylvania State University, another pioneer, Sargent explored the implications of expectations for monetary policy, finding the surprising result, in a 1975 paper, that “in the rational expectations version of the model, one deterministic money supply rule is as good as any other, insofar as concerns the probability distribution of real output.”
In other words, as long as monetary supply growth is extremely predictable, the real economy will muddle along, no matter how quickly or slowly the monetary supply grows. Sargent and Wallace’s later “unpleasant monetarist arithmetic” argued that monetary policy, especially open-market operations, may not even be able to control inflation.
Sargent’s work has wonderfully balanced serious mathematical theory, strong econometrics and a clear passion for history. In 1982, he looked at the end of hyperinflation in Germany, Austria, Hungary and Poland before World War II, and demonstrated that these episodes of inflation stopped abruptly, rather than gradually. A sharp end to hyperinflation suggests that inflationary expectations aren’t controlled by some fixed formula based on past inflation.
Sargent has written an insightful essay on “Macroeconomic Features of the French Revolution.” He used his presidential address to the American Economic Association, provocatively titled “Evolution and Intelligent Design,” to explore the intellectual influences on the legendary economist David Ricardo two centuries ago.
Sargent has also been a great teacher and mentor to generations of young macroeconomists. Two of his lucid monographs, “Macroeconomic Theory” and “Dynamic Economic Theory,” have long been mainstays of macroeconomic education. They may be the most dog-eared textbooks I own. I had the fortune to be a colleague of Sargent’s at the Hoover Institution in 1994-1995, and was profoundly influenced by his gentle spirit of scholarship.
Theory and Data
He and Sims did their most important work simultaneously, and they even cooperated on one well-known paper, but Sims’s contributions are quite different in character. Sargent always begins with theory. Sims begins with data. Sims taught us how empirical work can and should inform theory.
Sims -- like Sargent, Lucas and Edward Prescott (another great theorist of the post-Keynesian world) -- saw that the Keynesian macroeconometric models were a thing of the past, but he understood the ongoing need for economic prediction. Perhaps one day, economic theory will make complete sense of the business cycle, but until that time, policy makers and ordinary investors will still want to have some idea of what lies ahead. Sims’s work addressed that need, free from the confining assumptions of Keynesianism.
His most-often cited work is aptly titled “Macroeconomics and Reality.” The paper catalogs the criticisms of macroeconomic models, and then provides an alternative approach -- Vector Autoregressions. VARs make no pretense of uncovering deep parameters of an economic model, but rather provide a statistical tool for integrating various macroeconomic series, and rely on assumptions, not about the whole economy, but about the independent nature of some shocks. If some series of shocks to the economy cause, but are not caused by, other series, then econometricians can create powerful predictive models. Sims’s approach is now accepted around the world.
The prize committee also referred to Sims’s earlier and often-cited work that focused on applying U.K. economist Clive Granger’s definition of causality to macroeconomics. Granger defined causality not with philosophy but with timing. If variable X helps predict the future of variable Y, usually controlling for the past of variable Y, then variable X can be said to “Granger cause” variable Y. Sims found that money supply appears to “Granger cause” changes in income, but there is no comparable causal link between income and money supply.
Sims’s contributions to time-series econometrics are enormous, and he has also inspired scores of students. His work is distinctly pragmatic and the prize committee did well by balancing him with Sargent.
The world faces big macroeconomic challenges, and macroeconomists lack a clear consensus of what should be done. But their debates are guaranteed to persist because the conditions are always changing, and natural experiments are scarce. Still, we know much more than we did 40 years ago about how the macroeconomy operates, and Thomas Sargent and Christopher Sims deserve plenty of credit for moving the field forward. Their prizes are extremely well-deserved.
To contact the writer of this article: Edward Glaeser at firstname.lastname@example.org
To contact the editor responsible for this article: Mary Duenwald at email@example.com