No, GDP Isn’t Really Going to Shrink 30%
The U.S. practice of reporting quarterly GDP growth as an annualized number is about to make the Covid-19 downturn look a lot deeper than it is.
In the U.S., the headline gross domestic product number reported every quarter by the Bureau of Economic Analysis is the percentage change in GDP from the previous quarter, annualized. Other countries, and international economic-statistics compilers such as the Organization for Economic Cooperation and Development, tend to report just the quarterly percentage change and skip the annualizing.
I don’t know that one way is better than the other. The BEA’s version gives a more consistent view of the pace of GDP growth over quarters and years. The rest of the world’s approach gives a more realistic view of a quarter’s contribution to annual GDP growth.
The formula for calculating an annual rate from quarterly numbers involves dividing the current quarter’s GDP by the previous quarter’s, taking the result to the fourth power and subtracting by one. But you can approximate the result pretty well by multiplying the quarterly percentage change by four. So if the U.S. says its GDP grew 2% in the quarter and the U.K. says its grew 0.5%, they are in fact reporting equal rates of growth. People who follow these things for a living know this, of course, but I imagine many consumers of business and economics news (and surely some reporters of it) do not.
In normal times, the main effect of these different ways of calculating GDP is probably to give Americans an exaggerated sense of economic superiority. The U.S. economy has in fact grown faster over the past decade than the economies of the other affluent countries that make up the OECD, but the average quarterly difference since 2010 is just 0.05%.
These are not normal times, however, and I’m starting to suspect that the American way of calculating GDP is going to make a lot of people think the downturn caused here by Covid-19 is substantially deeper than it actually is. Forecaster after forecaster after former Fed chair has been predicting a 30% annualized decline in the U.S. GDP in the second quarter, and in headlines and sometimes article texts this is being represented as meaning that GDP will fall 30%.
What all these people are actually predicting, though, is that GDP will fall about 7.5% (30% divided by 4) in the second quarter. That would still be really awful — in the worst quarter of the quite terrible 2008-2009 recession, real GDP fell 2.2% on a non-annualized basis. But during that recession, GDP declined in four other quarters as well, for a total peak-to-trough drop of 4%. If the U.S. economy grows again in the third and fourth quarters, which it can do even if life is still far from normal just because it will be starting from such a depressed base, that’s a hopeful but not delusional target for 2020’s full-year GDP decline. Over the course of the Great Depression, by contrast, real GDP fell 26%. We’ll find out the actual second-quarter number on July 30; the first-quarter GDP report will be released April 29, but that will reflect growth before the bulk of the coronavirus-related shutdowns.
There are other economic statistics — mainly those involving jobs — that really will be of history-making awfulness in the coming weeks and months. This awfulness could be temporary, though, so let’s at least not talk ourselves into an even worse recession by believing that GDP is going to fall 30%.