As each new year commences, economists throughout the land unveil their forecasts of what�s in store for the economy. These prognosticators often make such precise forecasts and communicate their forecasts with such confidence that they all but create the impression that they know what the future holds.
It�s not entirely a con job. But it�s close to it.
The reason for economic forecasting is that the future is uncertain, and the purpose of forecasting is to give decision makers in business and government a better idea of what the future might hold. But the reason for forecasting is also the greatest pitfall of forecasting. No one knows what the future holds. And no one can calculate what the future holds. If the future could be calculated, it wouldn�t be uncertain.
Economists make their forecasts from three building blocks, as it were: an assessment of the current state of the economy, assumptions about the future state of exogenous factors in the economy - factors such as Federal Reserve policy, federal tax and spending policy, and so on - and economic theory. To make a forecast, economists basically ask: given the current state of the economy and our assumptions about exogenous factors, what does economic theory say will happen in the economy? Unfortunately, precise quantitative forecasts require stronger building blocks. But none exist.
Assessing the current state of the economy is no easy task. The data that come available to economists are always dated, usually by at least a month, often by a quarter of a year or more. As such, they inform about the past state of the economy, not the current state.
The current state of the economy is almost always a blur. Our economy at present is a perfect example. Though economists are sure the economy is slowing, nobody knows by how much. Some economists believe the economy is growing by slightly less than its typical 3 percent rate; others believe the economy is already in recession. But nobody really knows what the state of the economy is today and won�t know for a couple months, by which time the present economy will be a thing of the past.
The inaccuracy of economic data is an even larger problem. Economic data are often reported as if they were accurate and precise. But the data are rarely accurate and precise. They are often fraught with measurement error, and many are tabulated on the basis of some rather dubious assumptions. The estimates of gross domestic product - the very statistics that are used to calculate the rate of economic growth - are so problematic that, according to one famous economist, using them to analyze the business cycle is �nothing short of grotesque.�
Making assumptions about exogenous factors in the economy sounds innocuous enough, but those assumptions are themselves predictions. And how are economists to predict factors such as the type of monetary policy the Fed will pursue in any given year? Fed officials themselves don�t know what sort of monetary policy they will pursue three months from now, never mind for the rest of the year.
And with a new president and a new Congress, how reliable can predictions about federal tax and spending policies be?
Then, we have economic theory. Economics is the study of human action, which means economic theory is inherently qualitative. The use of reams of data and fancy mathematical techniques does not alter that fact.
Human action is never constant, nor are the circumstances in which humans act constant. And human action is purposeful, not mechanical. But the statistical models that economists use to generate forecasts treat human action as if it were constant and mechanical. They are inherently wrong.
Heady stuff? Not really. Put it this way. The U.S. economy consists of over 280 million people, over 140 million workers, over 20 million businesses, and lots of government. Each person�s assessment of the costs and benefits of any action is entirely subjective and continually changing. Considering all that, how much faith can we put in a forecast that purports to indicate precisely how fast the economy will grow this year? Not much.
Economists can provide decision makers with valuable information. They can help decision makers interpret economic data, and they can help explain why certain economic events unfold the way the do and what events might unfold in the future. But precise forecasts are not to be believed.