Two key issues in the upcoming election are jobs and the economy. Given their importance, it is appropriate to review exactly where we are on these issues. Both the economy and the job situation have been radically impacted by the COVID-19 pandemic. In the first quarter of 2020, real GDP declined at an annual rate of -5 %, largely as a result of the first surge of COVID cases in March, when the pandemic began to have a real impact on economic activity. Corporate profits declined by $37.5 billion, and the first round of job losses – over 1.3 million – occurred in March. The decline in the second-quarter GDP of -31.4 % was even more dramatic and was accompanied by a drop in the PCE price index to 1.6% and by the loss of 20.8 million jobs in April. Rounding out the third quarter, only 7.5 million of those job losses were reversed in May and June, and only another 3.9 million were recovered in the third quarter, indicating that the virus continued to impact activity in the third quarter.
Providing further data to consider, Federal Reserve banks have started to produce what are call nowcasts, or running estimates of GDP and other economic statistics. One of the most well-known is produced by the Federal Reserve Bank of Atlanta, and the forecast of the annualized in GDP growth in Q3 was 35.2%, as of October 9. It is worth noting that in June the Bank predicted a increase of 51.8%, substantially larger than what occurred; so, like all forecasts, nowcasts are subject to error and sometimes huge errors. The New York Fed also now produces a nowcast of GDP growth as of October 17 and its nowcast is a growth of 13.8% on an annualized basis (less than half the increase predicted in Atlanta’s forecast) for Q3 and 3.6% for Q4, suggesting that a significant slowdown will persist for the remainder of the year. The NY Fed’s number seems more reasonable at this point and may be too high for Q4, given the recent announcement of layoffs by airlines and by Disney and other companies in the hospitality business.
Turning now to the job situation, we note that in February the unemployment rate dipped to 3.5% – a rate not seen since February 1969, when unemployment stood at 3.4%. Equally impressive, beginning in March of 2018, the number of job openings exceeded the number of unemployed and continued to do so until March this year, when we had 1.2 million more job vacancies than unemployed persons. The data suggest that we were probably at the point where the chief explanation for the difference is what economists term frictional unemployment, where, for example, an unemployed person lives in Seattle, while the best job for that person might be in Miami. There is also a “skills mismatch” explanation; for example, the job opening is for a nurse, but the unemployed person is a computer technician. By April, however, the story changed. There were 23.1 million workers suddenly unemployed. To be sure, that number has declined to 12.6 million as of the end of September. However, new weekly claims for unemployment continue to be high, reaching 898,000 for the week of October 10. The four-week moving average of 866,250 rose 8,000 from the previous week’s number.
The general picture we get from labor market data is that while the economy is improving, the rate of improvement is slowing. Chairman Powell and other federal officials have argued that more fiscal stimulus is needed and that the failure of the Congress and the Trump administration to reach an agreement is adding to the uncertainty. One argument advanced against providing additional support for the unemployed is that the $600 per week may act as a disincentive for the unemployed to seek work. But a study by the Federal Reserve Bank of Chicago (see https://www.chicagofed.org/publications/chicago-fed-letter/2020/441) found that people receiving the $600 weekly payments searched for employment more than twice as intensely as those whose benefits had expired. The study also found that those who had exhausted their benefits were more likely to take lower-paying jobs. Other work is also suggesting similar conclusions. This growing body of work suggests that the expanded unemployment insurance does not function as a disincentive. There are at least two logical arguments consistent with this behavior as well. First, recipients know that benefits will not go on forever. Second, people know they will be better off with a job that has benefits than they will be if they stay on unemployment. Will Congress respond? We don’t know, and it is becoming less and less clear that a deal at this point will benefit either party in the election. Meanwhile, the virus outbreak continues and shows no sign of abating.