Can Economic Forecasting Predict The Future?
STEVE INSKEEP, Host:
The consensus among leading economists, for what it's worth, is 2.7 percent growth this year - not so great, not horrible. That's the forecast. NPR's Planet Money team has been studying just how the business of economic forecasting works and how much it can actually tell you about the future. Adam Davidson and Alex Blumberg report.
ADAM DAVIDSON: The closest thing to a modern day financial oracle is probably a computer program that lives in St. Louis, Missouri. That's where the offices of Macroeconomic Advisors are located, one of the leading economic forecasters in America. Joel Prakken, the man who created the computer program, showed us on a computer screen how it works.
JOEL PRAKKEN: Okay. So you should see on the screen various components of consumer spending on services. Are you seeing that?
DAVIDSON: The basic idea here is pretty simple. Take economic data from the past and use it to predict what will happen in the future. That's where Prakken comes in. For the past 28 years, he's been figuring out the exact relationships between different kinds of economic activity.
ALEX BLUMBERG: If average salaries go up by, say, four percent, how many more cars will people buy with their new, bigger salaries? What will all that spending do to the inflation rate, and what will the inflation rate do to home sales? And how will home sales, in turn, affect salaries? It's all connected, and Joel has written down exactly how it's all connected in a series of equations.
PRAKKEN: If we scroll down a little bit further to look at some of the equations of ours...
BLUMBERG: Oh, boy.
PRAKKEN: Yeah, okay. So here is an equation - yeah...
BLUMBERG: Oh, my God. Whoa, it gets worse every day.
PRAKKEN: It does get worse.
BLUMBERG: Now, I just want to point out, also, so this one page here, there are - what? I'm looking at the 400...
PRAKKEN: You're on page 49 of 447.
DAVIDSON: Four-hundred-and-forty-seven pages of really dense math equations. That's how you forecast the economy. Well, that's how you try to forecast the economy. Prakken says there are lots of difficulties, and getting the math right is only one of them. Consider what goes into his computer model: the data.
BLUMBERG: Most of Prakken's data comes from the government, the BEA, the Bureau of Economic Analysis. He's only now getting data for third quarter 2009. That means he's getting now data for what happened three months ago. And even now, the BEA still calls it an estimate.
PRAKKEN: What's called the final estimate for third quarter GDP.
BLUMBERG: Wait. They have to estimate what happened three months ago?
BLUMBERG: They can't even get a firm number of what happened in the past?
PRAKKEN: Well, they make three - well, they make - they make many estimates of what happened in the previous quarter.
DAVIDSON: So many estimates. Let's take the quarter that just ended: fourth quarter 2009. We'll get the first estimate of what happened in a few weeks. The second estimate will come a month later, and then a third estimate a month after that. Then in July, they'll do another estimate, and a year after that another estimate and the year after that, another estimate.
BLUMBERG: I just want to point out, we're calling you to ask what 2010 is going to be like, and what I'm hearing is if we're lucky, by summertime you can tell us what 2007 was like.
PRAKKEN: Well, if we're lucky, by July of this year, we'll have our first true annual estimate of what 2009 was like. We'll have our second true annual estimate of what 2008 was like, and we'll have our third true annual estimate of what 2007 was like.
BLUMBERG: And you're not even telling us, when we actually know?
PRAKKEN: You know, you never know for sure I think is the answer, here.
BLUMBERG: They just give up guessing.
DAVIDSON: So the data's an estimate. The model is an approximation. The end result: economic forecasters, even the best, have a mixed record forecasting the economy.
BLUMBERG: On average, the leading forecasters, like Prakken's Macroeconomic Advisors, have around a one percentage point margin of error, which might sound pretty good. But let's say you forecast a two percent growth rate for the year. That means you're saying the actual rate will be somewhere between one percent and three percent.
DAVIDSON: Now those are two totally different economies. One percent, that doesn't even keep up with population growth. That's a bad year. Unemployment will go up. It'll feel bad in the U.S. Three percent is the opposite, a pretty good year.
BLUMBERG: So the forecasters, at best, can say the economy will be somewhere between bad and good. And at worst, well, they missed something huge. Economic forecasters failed to predict the 1982 recession or the 1984 recovery or the late 1990 stock bubble or the 2001 recession.
DAVIDSON: Just last year, in January 2008, we now know we were already in the worst recession in decades, but economic forecasters were predicting a good, healthy economy for the rest of the year. They missed the present and were disastrous at predicting the future.
SIMON JOHNSON: You know, economic forecasting is inherently an impossible task.
BLUMBERG: Economist Simon Johnson is with MIT and the Peterson Institute for International Economics. He used to run the economic forecasts at the International Monetary Funds.
JOHNSON: Between now and the end of next year, the world will be hit by three major shocks, probably.
DAVIDSON: You mean that's just on average over the last century or whatever, there's...
JOHNSON: Yes, exactly. Stuff happens. Okay? You know, the thing you're studying is subject to so much chaos in a mathematical sense, so much randomness, and so making any forecast is going - inherently got this problem.
BLUMBERG: Economic forecasting is big business. Any large company you can think of spends millions a year trying to get a handle on what's coming. So why are all these serious business people spending so much money on forecasts that they surely know are not all that reliable? Again, Simon Johnson.
JOHNSON: I would call it a necessary evil. There's so much imprecision. There's so much, you know, lagging in terms of our updating, that in some sense, we'd be better off without forecasts. We'd better off, you know, making up our minds afresh every day. But the problem is the businesses, the institutions, all involve thinking about the future and planning for the future. And you can't do that without taking a view of the future.
DAVIDSON: Someone at Ford Motor Company has to decide: How much steel do we buy this year? Pension funds have to figure out: Where will we invest our money?
BLUMBERG: Schools have to decide how much tuition to charge. And governments have to out how much they'll spend on road repairs.
DAVIDSON: In everyone of these decisions, someone has to take a view about what is going to happen, even if they know that view is probably not quite right
BLUMBERG: I'm Alex Blumberg.
DAVIDSON: And I'm Adam Davidson, NPR News.
NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.