Editor’s note (December 4th): Since publication of this article the latest data release shows that jobs growth in America has slowed sharply. The economy added 245,000 jobs in November, down from a monthly average of 1.9m during the summer and autumn.
IN THE SUMMER and autumn America’s economy roared back. After peaking at nearly 15% of the labour force, unemployment fell like a stone, while in the third quarter GDP bounced from its lockdown-induced slump. The recovery of the world’s largest economy seemed oddly impervious to a second and then a third wave of coronavirus infections, even as economic activity in other parts of the world took a hit.
Yet there are growing concerns that the run of surprisingly good economic news is over, at least until a vaccine becomes widely available. In congressional testimony on December 1st Jerome Powell, the chairman of the Federal Reserve, said the recovery was slowing, while the decision on the same day by a bipartisan group of senators to release a proposal for a stimulus package reflects the same fears. The jobs report for November, which was to be released shortly after The Economist went to press, will probably be a downbeat one by recent standards—and whatever it shows, it is old news, since the surveys for the report were taken some weeks ago. More up-to-date figures show that the recovery has lost steam. That is bad news for the millions who remain out of work, as well as the rapidly growing share of Americans who are living in poverty.
Official statistics tend to be produced with long lags. So during the pandemic economists have turned to “high-frequency” data, largely produced by the private sector and generated by consumers’ and firms’ transactions, to measure the economy in real time. Wall Street banks now routinely provide clients with updates on everything from weekly electricity consumption to daily hotel bookings. The high-frequency data do not map onto the official kind perfectly. But they are useful for finding turning-points. They pinpointed the start of the downturn in March long before the official statistics could.
America is at another turning-point. STR, a data provider, finds that in the week ending November 21st hotels were running at 40% occupancy, down from 50% only weeks ago. The number of diners in restaurants has sharply declined in recent weeks, suggest data from OpenTable, a booking platform, with the fall even steeper in the states hardest hit by the virus. A recovery in air-passenger numbers appears to have ground to a halt as well.
Other real-time measures capture economic activity more broadly. The share of small firms which have temporarily closed is probably rising. Consumer spending in the week ending November 22nd was down by 5% compared with the one before, according to Cardify, a data provider. Using mobility data from Google, The Economist has constructed an economic-activity index measuring visits to workplaces, transport hubs and places of retail and recreation. After rising steadily during the autumn, the index has fallen back—though America still looks better than Europe, where the economic-activity index has crashed as governments have imposed another round of lockdowns. JPMorgan Chase, a bank, produces an estimate of monthly American GDP growth from a range of real-time data. In a report published on December 2nd, it suggests that output stopped growing in November.
Three factors explain the slowdown. To some extent it was inevitable. Loosening lockdowns had allowed millions of people to return to work and start spending again. But there was no comparable loosening of coronavirus restrictions after that. So it was never realistic for America to repeat the 7.4% quarter-on-quarter GDP growth that it saw in July to September.
Fiscal policy is the second factor. Another reason the economy bounced back so quickly in the summer was the enormous generosity of the stimulus packages agreed by Congress in the spring, worth some $3trn (or 14% of GDP). Yet Congress has so far failed to agree to another one, even though the most bullish forecasters still reckon a package worth over $500bn is required to help the economy back to some semblance of normality. A programme set up by President Donald Trump to raise unemployment-insurance (UI) payments by $300 a week, which had boosted aggregate household incomes by 1.5%, wound down in October. States and local governments, facing a severe budget crunch, cut over 1m jobs in the first six months of the pandemic, more than they lost even during the financial crisis of 2007-09.
The third and most important reason for the slowdown is the virus itself. Until recently many Americans, especially in Republican-leaning areas, seemed oddly happy to go about their business as normal. In South Dakota in September and October, for instance, visitors to sites of retail and recreation were 1.5% higher than usual for that time of year, even as coronavirus infections surged. Analysis by The Economist, drawing on Google data and work by Mark Muro and colleagues at the Brookings Institution, a think-tank, found that in the summer and autumn people in pro-Trump areas were half as likely to avoid public places as people living in areas that had voted for Joe Biden (see chart 1).
But now even people in the most pro-Republican areas appear to be getting skittish, too. In the week before Thanksgiving attendance at South Dakotan recreation-and-retail was 8% lower than normal. The continued increase in coronavirus cases may partially explain this, but a rise in death rates may be more significant. Research by Austan Goolsbee and Chad Syverson, both of the University of Chicago, finds that local deaths from coronavirus have a big impact on a local economy, perhaps because they bring home the seriousness of the situation. Deaths lag behind cases, and the share of American counties with at least one death from coronavirus in the previous week is soaring (see chart 2). Surveys suggest that a growing share of people worry about catching the virus.
The economy will rise again once a vaccine becomes available. Roughly 40% of the country should be vaccinated by March, suggests a recent paper by Goldman Sachs, a bank, putting America behind only Britain in terms of the speed of the rollout. And the vaccine-induced boost could be bigger than many expect. So far the pandemic has left surprisingly few scars on America’s economy. Business bankruptcies and the number of people in long-term unemployment remain lower than during the financial crisis of 2007-09.
Until then there will be further drags on the economy. Two further provisions related to UI, one which expanded eligibility to include the self-employed and gig workers, and one which provided extra weeks of benefits for recipients, are due to expire at the end of the year. A number of emergency lending programmes are also likely to end at that time. And the pandemic remains out of control. America, and especially its poorest folk, face a tough winter.■
Editor’s note: Some of our covid-19 coverage is free for readers of The Economist Today, our daily newsletter. For more stories and our pandemic tracker, see our hub
This article appeared in the United States section of the print edition under the headline “You must believe in spring”