Forecasting the COVID recession

Forecasting the economy is a near impossible task in this recession because the data have no historical precedent. We thought the last recession from 2007 to 2009 was bad; but this one makes the last one look trivial. For example, look at the change in consumer spending in April compared with the change in the past two recessions.


Because of the unprecedented nature of this recession, forecasting is a major challenge. Like many forecasters, my forecasts were far off for the unemployment rate in May. I expected it to be around 22 percent, but it came in just over 13 percent. I based my forecasts on the April number of nearly 15 percent, combined with a huge increase in continuing claims for unemployment insurance. But if you look at the chart comparing the continuing claims to the number of unemployed people, you can see that the numbers are so far out of our historical experience, that our econometric models are not very useful.


Compared with other forecasters, such as those in the Survey of Professional Forecasters, I am a bit more pessimistic in the short run but more optimistic in the long run, as my initial forecasts of real GDP growth and the unemployment rate suggested.



These forecasts were made before the May unemployment rate was released in early June. I have subsequently revised down my unemployment rate forecasts but only because I think the BLS has a lot of errors in how people are classified and that they will continue to do so. I think the true unemployment rate is substantially higher than the BLS suggests.

Of course, all economic forecasts depend on the outcomes of the coronavirus, and no one knows how that will turn out. Thus, the error bands around all of these forecasts are much larger than ever before. In fact, my forecast for the personal saving rate in May is 30% plus or minus 20%! That’s because the personal saving rate is measured very badly, on average, and the measurement errors for most variables increase in recessions. I suspect that the data we are seeing now are going to be revised substantially, with revisions that are larger in magnitude than we have ever seen before. So, I would advise being skeptical about all macroeconomic data that you see.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s