Presentation Speech by Professor Per Krusell, Member of the Royal Swedish Academy of Sciences, Chairman of the Economics Sciences Prize Committee, 10 December 2011
Your Majesties, Your Royal Highnesses, Laureates, Ladies and Gentlemen,
Our national economies are continuously exposed to changes in the world around us. Fluctuations in the economic activity of other countries influence foreign demand for our products; variations in the prices of raw materials influence our production costs; technical innovations change the mix of goods and services demanded. Evaluating alternative responses to these sorts of inevitable economic disturbances is a central task for policymakers. How much, for example, would a particular increase in interest rates or decrease in income taxes affect the nation’s inflation and employment? And how long before those effects would be realised?
Macroeconomic questions like these are hard to answer, because they are fundamentally social science questions. Unlike natural scientists, economists cannot look for answers by conducting controlled experiments, at least not on whole economies. We must work instead with the data provided by history. In addition, economies do not behave like mechanical, natural science systems. Nature obeys its laws independently of what governments decide, but economies do not, because economies consist of thinking people. Thinking people are also influenced by expectations about the future. For this reason, economies are also influenced by expectations of future policy decisions not yet made. This makes the co-variations observed in historical data hard to interpret. Does a particular historical change in an interest rate explain the subsequent change in the inflation rate? Or is the causality the reverse: did the expected path of inflation generate the interest rate change? In general, how can we distinguish between cause and effect in a macroeconomy?
Thomas Sargent and Christopher Sims have provided powerful tools that can be used with the data to answer that question. They have developed methods that let us precisely distinguish the effect of economic policy changes from the effect of other changes in the economic environment. Their work has illuminated the effects of monetary and fiscal policy not just for researchers but also for central bankers and finance ministers everywhere.
Christopher Sims has focused on the effects of unexpected changes in economic policy. Using a particular form of time series analysis, so-called vector autoregression, he has shown how we can understand the dynamic response of the economy to an impulse like a temporary policy change. Such impulse-response analysis allows us to track how, for example, an unexpected lowering of the interest rate, engineered by a central bank, can affect inflation and employment over time. Based on this contribution, we now know that a lower interest rate implies an immediate, gradual increase in production and employment, with a maximal effect after about two years. But it implies no immediate effect on inflation at all. In fact, inflation only starts to slowly increase roughly a year and a half after the interest rate drop, after which it recedes. What we have learned about this process is critical for central bankers considering alternative changes in interest rates.
Thomas Sargent’s research has allowed us to understand the effects of more permanent changes in economic policy, such as a switch to new government budget rules. This kind of analysis requires a different approach because the historical data offer few examples of systematic policy changes. Indeed, some policy changes being considered may never have been used before. Sargent’s approach to such analysis relies on certain aspects of the behaviour in the economy being rather constant – at least, not changing when policy changes. Examples of such constant behaviours are how people trade off leisure and work and how firms trade off different kinds of production inputs. Sargent showed how statistical methods could be used to measure these “universal economic constants” in the historical data and how a mathematical model of the economy could be built around them. With such a model, economists and policymakers can now conduct artificial laboratory experiments and investigate the effects of systematic macroeconomic policy changes.
Dear Professors Sargent and Sims:
You have successfully confronted a fundamental challenge facing empirical macroeconomic research: to disentangle cause and effect in historical data. The methods you have developed are now central tools for economic researchers trying to understand how our economies work and how they react to temporary and permanent changes in the economic environment. Because your methods allow us to use historical data to identify the causal effects of changes in economic policy, they have also become indispensable tools for policymakers worldwide. Modern empirical macroeconomic analysis rests on your shoulders.
It is an honour and a privilege to convey to you, on behalf of the Royal Swedish Academy of Sciences, our warmest congratulations. I now ask you to receive your Prize from his Majesty the King.