[J. Bradford DeLong] Morgue testing the US economy

US national income and output in the first quarter of 2020 was 1.25 percent below what it had been in the fourth quarter of 2019, but still 9.5 percent above what it would be by the second quarter of this year. Now that US national income has plunged 12 percent below what it was at the start of the year, what should we expect for the third quarter?

America could always turn out to be lucky; but betting on that would not be prudent. According to Austan Goolsbee and Chad Syverson of the University of Chicago Booth School of Business, it was voluntary self-protection, rather than legislated restrictions on activity, that drove most of the decline in consumer spending this spring. Moreover, they warn that, “If repealing lockdowns leads to a fast enough increase in COVID infections and deaths and a concomitant withdrawal of consumers from the marketplace,” doing so “might ultimately end up harming business activity.”

In fact, we have seen this summer that ending or scaling back lockdowns often does lead to a rapid increase in COVID-19 infections. A sound forecast for the US economy, then, must start by forecasting the future of the pandemic. Yet most of the charts, graphs, and tables concerning the virus and its impact are useless for forecasting purposes.

After all, confirmed COVID-19 deaths tell us only what the virus was doing a month ago, whereas anticipating the virus’s trajectory requires knowledge about what it is doing now. Similarly, confirmed cases catch only those who have been tested, and one must wait days or even weeks to receive test results. To get a grip on the situation, the number of tests per confirmed case would need to be 5-10 times higher than it is.

How would the past few months have played out if effective testing infrastructure had been in place? My guess is that the true (not just the measured) number of weekly new cases increased 2.5-fold from mid-June to mid-July. Assuming that we can accurately infer true cases in mid-June by multiplying mid-July’s weekly confirmed deaths by 150 (admittedly a big assumption), the cases per week rose from about 600,000 to about 1.5 million as the “Sunbelt second wave” hit.

Since then, the overall picture has been even less clear. As was the case in mid-July, 1 test in 12 still comes back positive, and there are 150-200 confirmed cases per million people. But we are testing no more now than we were then, and the reporting date of a typical test is now lagging substantially behind the administered data -- in some cases by as long as three weeks.

If the US is lucky, the fear caused by the Sunbelt second wave will have led to more vigilance in social distancing and mask-wearing, allowing for the true number of new cases nationwide to stabilize at around 1.5 million per week. But if the country is unlucky, that number will have continued to grow to around 2.5 million per week.

Tragically, we simply will not know where we stand until we can count the bodies later this month. Only then can we determine whether COVID-19 deaths “merely” doubled (from 4,500 to 9,000 per week) between mid-June and mid-July, or continued surging well beyond that level.

America’s widespread testing failure means that it is still flying blind. As such, nobody really knows what would happen if schools were reopened or indoor dining and choral music recitals resumed. The conservative commentator David Frum spent the spring in New York and is now in Toronto. “To walk around a city of 2.8 million with fewer than 5 cases a day feels like the onset of recovery,” he recently tweeted.

He’s right. Canada, the European Union, and Japan are highly likely to see substantial economic recoveries this fall. The same cannot be said of the US. At best, there will be a mechanical bounce-back from the second to the third quarter, mostly because unemployment-insurance claims had fallen from a peak of 24 million to 17.5 million by the start of July. But, given that claims rose by one million from July 5-11 to July 12-18, there is every reason to fear that they have continued to climb in the ensuing weeks. We will soon have numbers for the first week of. If claims have returned to the 20 million range and remain at that level heading into the fall, much of the mechanical bounce-back will have been reduced.

For the US to have any chance of a real economic recovery this year, it would need to focus on two overarching objectives. The first is to stop the spread of the virus, so that people no longer fear going out and spending. The second task is to put those who have lost their jobs in hospitality and other indoor high-human-contact occupations back to work, either providing online services or working in investment-goods sectors that do not require close contact with people.

It is in America’s power to do these things, in principle. We have the resources, and as John Maynard Keynes once put it, “Anything we can actually do, we can afford.”

But doing what needs to be done would require governance far more competent than anything we have seen from President Donald Trump and his Republican enablers. In the coming months, at least, we should continue to expect the worst.


J. Bradford DeLong
J. Bradford DeLong, a former deputy assistant US Treasury secretary, is professor of economics at the University of California at Berkeley and a research associate at the National Bureau of Economic Research. -- Ed.

(Project Syndicate)