How Have Monetary Regime Changes Affected the Popularity of IS-LM?

Finally, what emerges clearly from this study of Keynes's views on monetary policy are the close links between theory and policy. For every change in the problem at hand came new proposals growing out of previous theory or, if previous doctrine could not accommodate the new situation, shifts in his theoretical position. This is clearly illustrated (from a Keynesian point of view) by the evolution of Keynes's views under the impact of British (and world) events between 1925 and, say, 1934. This interrelationship was the hallmark of Keynes, the economist.

—Donald Moggridge and Susan Howson, "Keynes on Monetary Policy, 1910–1946" (1974)

1. Explaining Innovations in Macroeconomic Theory

There are several ways of explaining theoretical innovations in macroeconomics. One is to view the history of macroeconomics as a steady progression from ignorance to enlightenment. Each succeeding model builds on previous theoretical developments in a way that is if not predictable, then at least readily explicable. Others view the field of macroeconomics as endlessly cycling between one conceptual approach and another: classical/Keynesian, demand-side/supply-side, statist/laissez-faire. In many cases these fads are attributed to changes in the underlying macroeconomy. Thus Keynesian economics is often regarded as a "depression model."

There is undoubtedly some truth to both perspectives. If one views macroeconomics as a series of technical innovations in areas such as money demand or expectations of inflation, then the optimistic perspective is probably warranted. If, on the other hand, one views macroeconomics in terms of conceptual approaches to stubborn macro problems, then the picture is much less clear. It could be argued that the essence of contemporary "macroeconomics" is what the one hundred top macroeconomists think about issues such as what explains the persistently high unemployment rates in Europe, or why Japan seems to find it so difficult to boost aggregate demand, or what is the relative importance of real and monetary factors in recent U.S. business cycles. These issues remain highly controversial, and the debates have a way of continually reviving discussions from decades, or even centuries, ago.

Academic debate about current macro issues involves much more than just the recitation of findings from time-series regressions. References to events such as the Volcker disinflation or the Great Depression play an important role in theoretical discussions, even at the highest levels of macroeconomics. To be widely persuasive, it's not enough that an economic model is consistent with the key phenomena of interest to macroeconomists, it must also seem consistent with those phenomena. In this essay, I focus on the plausibility of the simplest, and hence most widely understood, versions of IS-LM. My thesis is that the changing popularity of IS-LM (or indeed any macroeconomic framework) has more to do with changes in the external policy environment than with the internal logic of progress within the field of macroeconomic theory.

In the next section, I argue that the IS-LM model is a technical apparatus specifically tailored to a particular conceptual approach to macroeconomics and, more precisely, monetary theory. Furthermore, the model is only likely to prove useful (i.e., persuasive) for economies with a very specific institutional setup. In particular, IS-LM is especially suited to an environment where there are doubts about the effectiveness of monetary policy, and where monetary analysis is based on an interest rate transmission mechanism. And these two factors are related to even more basic structural factors, such as the nature of the monetary regime, and the sophistication or complexity of the economic system. The analysis required to support these hypotheses requires an unconventional mixture of economic history, the history of thought, and monetary theory.

The conference title refers to the "strange persistence" of IS-LM, and if one views macroeconomics as a progression of better and better technical models, then the persistence is indeed strange. I hope to demystify the enduring popularity of the IS-LM approach to macroeconomic issues. To do this, I begin by examining some of the strengths and weaknesses of the IS-LM approach. Then I examine how structural changes...