Macroeconomics, it is said, is an art, not a science. That means most of it is guesswork, and whatever governments do is based as much on hope and a prayer as hard analysis of data. Which prevails, of course, is what everyone wants to know. And this is what makes this year's Nobel for economics important.

For their efforts at systematising macroeconomics and giving it the rigour of a science — well almost — Americans Thomas J. Sargent and Christopher Sims have been awarded the Economics Nobel Prize for 2011. Both received their Ph.D from Harvard University in 1968.

As so many treasuries, finance ministries and central banks may have realised, the problem is not of finding out how to steer the car; it is of knowing what to do when you hit a huge bump: do you brake, accelerate, swing hard to the right or to the left, or simply do nothing?

Dealing with bumps

“One of the main tasks of macroeconomic research is to comprehend how both shocks and systematic policy shifts affect macroeconomic variables in the short and long run,” the Royal Swedish Academy said in a statement. “Sargent's and Sims's awarded research contributions have been indispensable to this work.”

Macroeconometrics, as its composite name suggests, is the field of applying statistical techniques to the analysis of macroeconomic data. But before the numbers can be crunched, economists need a working model to test.

Simply put, Professor Sargent's contributions were to develop these frameworks, while Professor Sims aided in the development of the necessary statistical techniques. Sargent tells us where to go and Sims tells us how to get there.

The two American economists conducted separate research in the 1970s to model cause and effect in economies, including the complex interplay of tax and interest rate policies with the expectations of people and businesses.

This is of some importance every time policy is called into play to help out a flagging economy.

Rational Expectations

The point is this: government spending can be less effective when people see the limits to state finances and expect stimulus spending to run out. They behave, in that case, differently from how they would if they did not expect any limits to stimulus spending. The same is also true of interest rate changes. If anticipated, people react differently to them than if they come as a surprise.

Prior to the work of Sargent, macroeconomic models were based on the assumption that agents in the economy react only to past events, or what is called adaptive expectations. That is, they change their beliefs only after events have happened.

So, for example, if the government announces a policy that involves increasing money supply, revisions would only be made not on the announcement but only after the increase in money supply has occurred, and even then agents would react only gradually.

However, agents were not observed to make such systematic errors in their expectations. This was made even starker by the failure of policy based on adaptive expectations in the 1970s to deal with the high inflation — the high unemployment trap.

In response to this, economists developed the notion of rational expectations (RE), where agents respond not only to the past but also to their beliefs about how events will unfold in the future. Agents are, in other words, rational.

Sargent was a keen proponent of rational expectations. Applying these ideas, Sargent and his co-author Neil Wallace produced the policy ineffectiveness proposition (PIP), according to which the government could not successfully intervene in the economy if attempting to manipulate output.

They argued that agents would foresee the effects of monetary expansion, adjust their wage and price expectations, leaving the economy exactly where it was in real terms.

And that is why, they said, US policy had remained thus far ineffective. What it would take instead was a “stochastic shock” — that is, an unanticipated change in policy.

Vector autoregression

Still, having a clearer understanding of how the economy works is not of much use. You also need a sophisticated way of dealing with large samples of data, particularly when the variables involved are all mutually interdependent.

Enter, in 1980, Christopher Sims and a model called vector-autoregression (VAR). It is a ghastly thing, actually — linear model of multiple variables, in which each variable is a function of the past and present values of the other variables as well — in addition to its own past values.

In the typical macro case, it assumes that the rate of inflation today is the result of not only past inflation, but also the past interest rates, unemployment rates, and anything else one feels is relevant. And, to add to our misery, each of the other variables is influenced by all the others.

Making sense of this and, in particular, disentangling causal and correlated effects, is tricky business.

The simple VAR framework provides a systematic way to capture rich dynamics in multiple time series, and the statistical toolkit that came with VARs was easy to use and interpret.

But people will be people, and none of all this seems to be getting us anywhere in navigating the bumps.

But, then, that's the story of economics, isn't it?

(The author is a doctoral candidate in economics at the Indian Statistical Institute, New Delhi.)

Related Topics
comment COMMENT NOW