Thursday, June 4, 2020

Economic Alphabet Soup: V, U, Z, W, L or Swoosh?

Will the economic recovery be shaped like a V, U, W, L, or Z? Cases can be made for all of these possibilities. There are other possible shapes to the recovery such as a square root sign, and even a “swoosh,” like Nike’s logo. Schematic diagrams of these alternatives can be seen in a May 11 WSJ article titled “Why the Economic Recovery Will Be More of a ‘Swoosh’ Than V-Shaped.”

In the past, economic recoveries from most recessions tended to be V-shaped. The experience of the Great Depression suggests that recoveries after such a severe downturn should be shaped more like an L or W. The recovery following the Great Recession of 2008 was widely perceived to be U-shaped.

The article cited above observed: “Until recently, many policy makers and corporate executives were hoping for a V-shaped economic recovery from the coronavirus pandemic: a short, sharp collapse followed by a bounce back to pre-virus levels of activity. Now, however, they expect a ‘swoosh’ recovery. Named after the Nike logo, it predicts a large drop followed by a painfully slow recovery, with many Western economies, including the U.S. and Europe, not back to 2019 levels of output until late next year—or beyond.”

Consider the following:

(1) Real GDP cycles. The y/y growth rate in real GDP has been mostly V-shaped during recessions and recoveries since 1948 (Fig. 1). There were two nearly back-to-back recessions during the early 1980s with up and down legs that resembled a W. The recovery in the early 2000s appears like a U. While economic growth was subpar and U-shaped during the expansion following the Great Recession, the initial recovery was V-shaped.

Following the June 1 release of the latest purchasing managers survey and construction spending reports, the Atlanta Fed’s GDPNow tracking model estimated that real GDP fell 52.8% (q/q, saar) during Q2, a bit worse than the -51.2% of the May 29 estimate. This increases the likelihood of a V-shaped recovery during Q3 and Q4, with my estimated gains of 20% and 5% (Fig. 2). Beyond that, I agree that it could be a swoosh with low single-digit growth rates. I don’t expect that real GDP will recover back to its Q4-2019 record high until late 2022.

(2) Recoveries in coincident indicators. I track the monthly Index of Coincident Economic Indicators (CEI) as a useful proxy for the quarterly real GDP series (Fig. 3). This index includes four coincident economic indicators: employees on nonagricultural payrolls, real personal income less transfer payments, industrial production, and real manufacturing & trade sales.

The CEI shows that the average time that it took for the economy to recover to its previous peak during the past six economic cycles was 33 months, ranging between 19 months (in the early 1970s) and 68 months (following the Great Recession).

I think it could take 32 months to get back to the February peak in this series, i.e., by October 2022. So the initial V-shaped rebound could eventually turn out to be a swoosh. This outlook allows for the possibility of a second wave of COVID-19 infections, though not as bad as the first wave and without another round of lockdowns.

(3) The first recession in services. Increasing the likelihood of a swoosh rather than a sustainable V-shaped recovery is the fact that the current recession is the first one that has been experienced by—and indeed led by—the services sector of our economy. In the past, recessions were led by downturns in manufacturing and construction. Most services industries were either relatively unaffected or actually continued to grow during previous recessions, while goods production declined, as can be seen by comparing goods versus services in real GDP (Fig. 4).

A glance at historical charts of industrial production and housing starts shows that both typically have V-shaped recoveries (Fig. 5 and Fig. 6). They are likely to do so again this time. The same is not likely to happen for retailers, restaurants, airlines, hotels, casinos, entertainment, and recreation. The article cited above noted:

“Among the reasons for the darker outlook is that lockdowns are being eased more slowly than originally expected in some countries. Even when they do lift, some large-scale activities—such as concerts and professional sports—won’t be possible again for months. Retailers and restaurants that have reopened are allowing in fewer customers at a time due to social distancing. And consumers worried about infection risks may take a long time to return to their old habits.”

(4) Feedback loops and aftershocks. The slow pace of recovery in service-producing industries could, in turn, weigh on the recovery in goods-producing industries. If, in fact, working from home (or from smaller suburban offices) catches on after the Great Virus Crisis (GVC), there is likely to be less business travel, which will depress airlines, hotels, restaurants, and convention centers. Commercial construction of offices, hotels, and retail stores is likely to be hurt by social-distancing aftershocks from the GVC, especially if the virus remains active because an effective vaccine isn’t discovered. Demand for new commercial jets is also likely to remain depressed since the airlines industry is unlikely to be back to business as usual for at least two to three years, if not longer.

(5) Surviving the car crash. One important industry that is very likely to experience a V-shaped recovery during the second half of this year is auto manufacturing, though that also is likely to turn into a swoosh during 2021 and 2022. The Fed’s monthly data on US motor vehicle assemblies plunged from 11.1 million units (saar) during February to 0.1 million units during April (Fig. 7). That’s extraordinary: The auto industry was essentially shut down during April, along with most other businesses.

Production is likely to bounce back smartly in coming months. Sales of domestic-make autos also fell but were well above zero at 6.7 million units during April, though that was down from 13.2 million units during February. That’s actually impressive given that many auto dealerships were affected by the lockdowns. Both sales and output should recover in coming months as the lockdown restrictions are lifted, with the latter outpacing the former.

By the way, contributing to the sharp increase in April’s personal saving rate was the drop in personal consumption expenditures on new motor vehicles and parts. It fell from $814 billion (saar) during February to $528 billion during April (Fig. 8). Since the full value of new auto sales is included in current consumption, that $286 billion drop boosted personal saving by the same amount.

No comments: