Back in March, I warned that the unemployment rate could hit 15% or more in the coming months, and in May I listed a variety of reasons why the recovery would be slow and drawn out. Now is a good time to provide a brief update. In short, I continue to believe that the recovery will be slow and drawn out, with the level of GDP not reaching its pre-crisis trend level until sometime in late 2022 or 2023.

The chart above shows weekly U.S. initial and continuing jobless claims data. When someone becomes unemployed, they must apply for jobless benefits through their state’s employment agency. Those whose applications are approved and who have begun receiving benefits and are counted in the “continuing claims” category. While well down from the initial spike, initial claims remain well in excess of 1 million people per week, and the rate of decline has slowed drastically. For context, even during the depths of the Great Recession, we never saw more than about 700,000 population-adjusted applications in a single week. More than 17 million people continue to receive jobless benefits; in the past 50 years, that number has never exceeded about 7 million people on a population-adjusted basis until now.

In my March blog post, I sketched out two possible scenarios for the unemployment rate. The chart above shows those scenarios as well as the updated actual unemployment rate. The initial jump I had expected to occur in March in fact occurred in April. Nevertheless, as of today, the quarterly average unemployment rate is close to my severe scenario. Yes, the unemployment rate for June was 11.1%, down from the 14.7% reported for April. Many states like Arizona, Texas, and Florida began reopening in the latter half of May and early June, so some people did return to work. However, in late June, after the survey for the June unemployment rate was completed, those states again shut down as the number of COVID cases began rising (more on that in a moment). Thus, the July unemployment rate is likely to be closer to the April figure than the June figure.

The chart above shows the total number of COVID cases for several states. The number of cases initially surged from late March through May in states like New York, New Jersey, and Pennsylvania. Seeing the national caseload level off, states including Arizona, Florida, and Texas began to lift restrictions and open for business. Unfortunately, those states appear to have moved too early. Since mid-June, those states have seen a rapidly increasing number of cases, with many reports of hospitals reaching capacity in parts of Texas and Florida. Restrictions have been put back in place, suggesting more layoffs and economic hardship. Until the number of cases is brought under control across the entire country, the pickups we have seen in retail sales, hiring, and other economic indicators will be fleeting.

The Atlanta Federal Reserve Bank currently projects that second-quarter GDP declined 34.7% at an annualized rate. The New York Fed’s real-time model suggests second-quarter GDP fell “only” 14.3%, although it had suggested a contraction of as much as 35.5%; the improvements are in part due to the recovery in the labor market since April as well as improving retail sales. Private forecasts are all over the map. In any event, we will get preliminary official numbers next week, and they will be utterly abysmal.

Most forecasters, myself included, expect that we will see a partial rebound in GDP in the third quarter as businesses reopen. The shutdowns and extensions announced this month imply that a full reopening of the economy likely will not occur until the fourth quarter at the earliest. I am beginning to doubt just how robust the recovery will be in the second half of this year. I can easily concoct scenarios where the number of COVID cases increases again in the fall, pushing any meaningful recovery to 2021.

I want to end on a positive note, in part to avoid being classified as just another “gloom and doom” economist. Several vaccines against COVID are currently in development, and the U.S. government has partnered with several of them to purchase enough doses to cover the entire population. Those vaccines are currently undergoing clinical trials, and some could be ready by the end of this year. Just this morning Pfizer and its German partner BioNTech announced an order from the US government for 100 million doses of their vaccine, pending the outcome of phase-3 clinical trials that are now starting. Johnson and Johnson, Moderna, and other drug companies with whom the US has signed agreements are also readying candidates for phase-3 trials. With rapid emergency-use FDA approval, we could see vaccines by the end of this year. While the trials are ongoing, the US is already procuring an adequate supply of vials, needles, and syringes so that as vaccines are approved, they can be put to use.