Macro Corner

Since the beginning of the pandemic crisis and ensuing lockdowns, we’ve been tracking the weekly jobless claims form the US Department of Labor and developed a highly predictive model (96% accuracy) which allows us to forecast where the trend is likely to go through the end of the year.

The model suggests that even with the recent plateauing of the decline in claims at around 900k per week and barring a fall resurgence to further stall recovery, we should be back at pre-peak levels of weekly claims by early November. While the unparalleled level of Americans filing for unemployment during this crisis is unnerving and tragic, ultimately it is the number who aren’t hired back as lockdowns end and reopenings (slowly) begin — i.e. the long-term unemployed — that matters most.

Economists toss around heuristics like 40–50% of jobs lost to the pandemic won’t be coming back. A May 2020 University of Chicago paper asserted the figure, ceteris paribus (i.e. no autumn resurgence stalling the recovery), is more precisely 42%; that is, only 58% of furloughed workers will get their jobs back as the economy regains strength. Applying that 58% rehire rate to our jobless claims model implies a year-end 2020 unemployment rate of 9.6%; however, the August rate published by the Labor Department on Friday was just 8.4%.

Given the predictive reliability of the jobless claims model, this disconnect between actual and forecast unemployment is likely due to the applied rehire rate of 58%. If the 8.4% rate holds steady to end of year, it implies that the rehire rate is significantly higher than the (gloomy) economists predicted — closer to 64% versus 58%. We can take some consolation from that evidence that the fledgling recovery is more robust than expected.

The downside, of course, is that even 8.4% unemployment is extraordinarily high. As a matter of fact, the US economy has only hit that level of unemployment 3 other times in the past 80 years. The last was in the housing/banking “Great Recession” ten years ago, and it took the economy 4 years to return to “full employment” (<5% unemployment rate), never mind the exceptional 3.5% unemployment rate the US hit prior to the pandemic shock. The time before that (the early 80s — another “self-imposed” recession, that time to crush 1970s stagflation) it took 13 years to get back to “normal.” We don’t even want to contemplate the 22-year timeframe in the early 70s it took to come back from that 9% unemployment rate peak.

So, while the numbers are perhaps not as grim as those in “the dismal science” may have led us to believe, and with prudence as a country with a strong economic foundation, we have the hope of clawing back from what this virus has done to us. We may just need to be a little patient and not expect a miracle overnight.