2011 Nobel Prize to Thomas Sargent and Christopher Sims

According to the Nobel website: "The Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel 2011 was awarded jointly to Thomas J. Sargent and Christopher A. Sims `for their empirical research on cause and effect in the macroeconomy.'" But what does that actually mean?

The website of the Nobel organization always offers useful background information about the laureates, including a "Scientific Background" paper about the winners. This year's background paper about Thomas Sargent and Christopher Sims is going to be hard sledding for those uninitiated into academic macroeconomics--by which I mean it has a bunch of equations. But the opening  pages offer an accessible overview of why they are eminently deserving of the prize. Here are some excerpts, mixed with some of my own explanations:


How was macroeconomic analysis done before the work of Sargent, Sims, and others? 
Here's my own description: If one looks back at how macroeconomics was typically done in the 1960s and into the early 1970s, the common macroeconomic models were big sets of equations--that is, they added up relationships between elements like consumption, investment, saving, imports, exports and total economic output, along with equations for how interest rates and exchange rates affected each other and these categories. A big category like "consumption" would be broken down in to durable goods and nondurable goods, and in turn these categories would be broken down still further. The resulting models would have hundreds of equations all interrelated with each other, and adding up to a picture of the macroeconomy as a whole. But as the Nobel background paper reports: "This estimated system was then used to interpret macroeconomic time series, to forecast the economy, and to conduct policy experiments. Such large models were seemingly successful in accounting for historical data. However, during the 1970s most western countries experienced high rates of inflation combined with slow output growth and high unemployment. In this era of stagflation, instabilities appeared in the large models, which were increasingly called into question."


The key role of expectations in this analysis
Many of the public policy discussions in the stagflation of the 1970s focused on expectations. What if workers were expecting higher wages? What if firms could promise higher wages because they expected prices to rise? Were the expectations causing inflation and recession, or were inflation and recession causing the expectations, or were there feedback loops in all of these and other economic factors? The macroeconomics of that time had no clear-cut tools for dealing with these issues.


The background paper puts it this way: "In any empirical economic analysis based on observational data, it is difficult to disentangle cause and effect. This becomes especially cumbersome in macroeconomic policy analysis due to an important stumbling block: the key role of expectations. Economic decision-makers form expectations about policy, thereby linking economic activity to future policy. Was an observed change in policy an independent event? Were the subsequent changes in economic activity a causal reaction to this policy change? Or did causality run in the opposite direction, such that expectations of changes in economic activity triggered the observed change in policy? Alternative interpretations of the interplay between expectations and economic activity might lead to very different policy conclusions. The methods developed by Sargent and Sims tackle these difficulties in different, and complementary, ways."

Sargent and structural econometrics
Instead of trying to build a macroeconomic model on a pile of statistics, and how those statistics added up and interrelated, the approach of Sargent (and others) was to build a macroeconomic model starting from the idea that economic actors like households and firms were doing their best to pursue their own interests. This approach has sometimes been called "rational expectations," but that term is probably misleading. The "rationality" here doesn't mean that economic actors have all available information, can calculate everything perfectly, and always make correct decisions. It only implies that they won't make the same mistake over and over again. In Sargent's hands, at least, this approach explicitly leaves open the question of just how people form expectations and learn.

Here's the background paper: "Sargent began his research around this time [the early 1970s], during the period when an alternative theoretical macroeconomic framework was proposed. It emphasized rational expectations, the notion that economic decisionmakers like households and firms do not make systematic mistakes in forecasting. This framework turned out to be essential in interpreting the inflation-unemployment experiences of the 1970s and 1980s. It also formed a core of newly emerging macroeconomic theories. Sargent played a pivotal role in these developments. He explored the
implications of rational expectations in empirical studies, by showing how rational expectations could be implemented in empirical analyses of macroeconomic events--so that researchers could specify and test theories using formal statistical methods--and by deriving implications for policymaking. ...
In fact, the defining characteristic of Sargent's overall approach is not an insistence on rational expectations, but rather the essential idea that expectations are formed actively, under either full or bounded rationality. In this context, active means that expectations react to current events and incorporate an understanding of how these events affect the economy. This implies that any systematic change in policymaking will influence expectations, a crucial insight for policy analysis."

I would add that instead of a model of the macroeconomy with potentially hundreds of variables, Sargent and others worked with models that on the surface appeared much simpler: for example, one example in the "background" paper is a model of the macroeconomy that has only three variables: inflation, output, and a nominal interest rate. But the inferences about cause-and-effect in these models are defensible and logical.

Sims and vector autoregressions
Sims pointed out that the earlier generation of macroeconomic models were built on a series of assumptions about how certain economic factors or policies "caused" other policies. But in an model of expectations, these statements about "cause" needed to be demonstrated, not assumed. Thus, instead of having a model in which some factors caused other factors, Sims proposed that macroeconomic analysis should begin with a model in which is was possible for every factor to "cause" a change in every other factor, and in addition for past values of every factor over the last few years to "cause" a change in every factor. This approach is called a "vector autoregression," but I often preferred to think of it as starting from a position of honest ignorance.

You then plug in all your data--say, quarterly data over a period of years--and see what patterns emerge. As you might imagine, it immediately looks clear that certain factors are not affecting others. Sims proposed a process for figuring out when certain factors aren't connected. As you begin to rule out what is NOT connected, what is left behind is a model of connections that actually exist. It's sort of like the way that a sculptor starts with a block of stone, and by gradually removing pieces, ends up with an image.

The background paper puts it this way: "Sims launched what was perhaps the most forceful critique
of the predominant macroeconometric paradigm of the early 1970s by focusing on identification, a central element in making causal inferences from observed data. Sims argued that existing methods relied on "incredible" identification assumptions, whereby interpretations of "what causes what"
in macroeconomic time series were almost necessarily flawed. Misestimated models could not serve as useful tools for monetary policy analysis and, often, not even for forecasting. As an alternative, Sims proposed that the empirical study of macroeconomic variables could be built around a statistical tool, the vector autoregression (VAR). Technically, a VAR is a straightforward N-equation, N-variable (typically linear) system that describes how each variable in a set of macroeconomic variables depends on its own past values, the past values of the remaining N - 1 variables, and on some exogenous "shocks." Sims's insight was that properly structured and interpreted VARs might overcome many identification problems and thus were of great potential value not only for
forecasting, but also for interpreting macroeconomic time series and conducting monetary policy experiments."

Other thoughts and resources
Sargent and Sims were colleagues at the University of Minnesota for about 15 years. The Federal Reserve Bank of Minneapolis puts out a readable publication called "The Region," which often does in-depth interviews with prominent economists about their work. An interview with Sargent from the September 2010 issue is available here; an interview with Sims from the June 2007 issue is available here.

Also, Sims published an article in the Spring 2010 issue of my own Journal of Economic Perspectives called "But Economics Is Not an Experimental Science," on issues of how to draw defensible cause-and-effect inferences from naturally-occurring data. Like all article in my journal, it is freely available courtesy of the American Economic Association.


While no one quite knows what the Nobel committee is thinking when they choose laureates, it seems clear that one standard is whether the ideas have been important enough to launch a sustained research literature. The ideas of Sargent and Sims from back in the 1970s and early 1980s certainly meet this test. Both these authors, and hundreds of others, have built on these ideas for decades.