GDP Suffered Horribly Under Pandemic Lockdowns But Not as Bad as Headlines Will Suggest

NEW YORK, NY - JULY 20: The Statue of Liberty is seen over a wind blown American flag scarf on Liberty Island on July 20, 2020 in New York City. Liberty Island partially reopens months after the attraction was shut down due to the coronavirus pandemic. Access to Liberty Island …
(Photo by Jeenah Moon/Getty Images)

The lockdown of the U.S. economy and China virus pandemic this spring triggered the sharpest economic contraction since the government began keeping track of national output after World War Two.

The Atlanta Fed’s GDPNow real-time estimate shows gross domestic product declining at an annual rate of 32.1 percent. The median forecast of Wall Street economists surveyed by Econoday is for a 35 percent contraction. Some analysts think the economy may have shrank at a pace closer to 40 percent.

The biggest recorded quarterly decline in modern U.S. history was 10 percent in the first three months of 1958. The worst drop during the Great Recession was 8.4 percent.

Confusingly, however, this does not mean that the pandemic caused the U.S. economy to shrink by one-third. The pandemic has been bad for the economy but it has not been anywhere near that bad. If our economy had suffered a contraction anywhere near that size the unemployment rate would likely be closer to 30 percent instead of the 11 percent or so reported last month.

Economic output in the April-through-June period likely 7 to 11 percent lower in the second quarter than in the first. That’s still a terrible contraction but far less scary than the headline number suggests.

In the U.S., the official gross domestic product number released every quarter by the Bureau of Economic Analysis gets reported as an annualized percentage change in GDP from the previous quarter. In other words, what gets reported is how much the economy would contract if it shrank at the same rate for a full year.

A lot of our economic statistics get reported as annualized rates. Last week, for example, we reported that existing homes were sold in July at a seasonally adjusted annual rate of 4.72 million. This does not mean that Americans are on track to buy 56 million homes in 2020, which would be the equivalent of 40 percent of American homes changing hands in a single year. Instead, it means that home sales would amount to 4.72 million if they sold at the July pace for the entire year, so that just 3 percent of existing homes would change hands.

Annualizing the rate of change is useful, especially in normal times, for tracking the direction of data over time. It allows us to compare, for example, a quarter’s growth rate to annual growth rates in the past on an apples-to-apples basis. But in at times of abrupt and extreme changes, or short-lived change, it risks creating an exaggerated impression of the change.

Most of the rest of the world reports economic output differently. Instead of an annualized rate, other countries report the change in the quarter’s GDP compared with the prior quarter. That gives a pretty good snapshot of a particular quarter’s contribution to annual growth but makes it harder to compare a quarter’s output to years past. Journalists and analysts often annualize growth numbers in Europe to get a version consistent with the U.S. method.

Sometimes this leads to confusion. Americans in good times may have overestimated the strength of their economic growth compared with other counties because it got reported at a 2 percent annualized rate while, say, France got reported at 0.5 percent quarterly growth. These would have been more or less equivalent GDP growth figures. And this year, some pundits falsely claimed the U.S. was doing worse in the first quarter than South Korea, apparently not realizing the two countries report GDP differently.

Making things even more confusing, some people prefer to use the change from the year-ago quarter. This is how the Federal Reserve, for instance, looks at GDP when forecasting its growth rates.

It would probably be a good practice for journalists reporting U.S. GDP in the second quarter to report both the annualized rate and the nonannualized percent change. And then add in the change from a year ago numbers where appropriate.

One thing that is clear already is that analysts were way off the mark in their early estimates of the economic impact of the pandemic.  Wall Street forecasters were initially predicting closer to a ten to 15 percent annualized decline in the first quarter. JP Morgan predicted 14 percent annualized decline in March, which Marketwatch described as a “stunning decline.” Goldman Sachs was the most pessimistic in March, predicting a 24 percent decline, five times its earlier estimate of a five percent contraction. (Wall Street quickly became far more pessimistic. By April, J.P. Morgan said the economy would contract 40 percent.)

The same annualization method will likely produce an exaggeration the other way in the reports of third-quarter GD. Wall Street analysts are forecasting growth rates north of 20 percent—and some as high as the mid-thirties.

.

Please let us know if you're having issues with commenting.