The timing of the surveys means the March employment report picked up the beginning of the layoff tsunami…

COVID-19 didn't spare the labor market in March. According to the Bureau of Labor Statistics (BLS) employment report, 701,000 jobs were lost in March (from the payroll survey), ending the nine and a half year stretch of consecutive job gains. We expected a more modest 75,000 jobs lost, under the assumption that the timing of business closures is a factor, with most of the closures happening after the BLS survey week that includes the 12th of the month. This only shows how many businesses were shutting their doors well before the many state-mandated closures that followed.

The unemployment rate, which is derived from the household survey, jumped to 4.4% from 3.5%, with both a 1.6 million drop in labor force participation and an even larger 2.98 million drop in household employment--no doubt a devastating combination. The underemployment rate (U6) increased by 1.7 percentage points to 8.7% in March--the largest one-month increase on record since this measure began in 1994. The BLS reports that the "bulk of the increase in unemployment occurred among people on temporary layoff, which increased 1 million in March to 1.8 million."

…And now over 20 million jobs lost by May appears well within reach.

This doesn't change the grim reality that April job losses will be even worse. A record number of 6,648,000 people filed initial unemployment claims in the last week of March. This is twice last week's record number and almost 10 times the previous record of 695,000 from October 1982. We had expected well over 8 million jobs would be cut from the payrolls in April. But unfortunately, that looks too optimistic. Given that 10 million workers have already applied for unemployment benefits, with one more week to go before the BLS conducts its survey, we now expect almost 13 million jobs will be lost in April. 

We had earlier expected a cumulative 16.5 million jobs lost by May when all is said and done with social distancing, but job losses are coming at a faster pace and sooner than we initially thought. Unfortunately, they could be much greater, and, while once unthinkable, we now expect over 20 million jobs lost by May.

While hard to comprehend, this potential is becoming very real. Of the total 152.5 million jobs in February before COVID-19 hit, approximately 109 million were in private service-providing industries, with 17 million in leisure and hospitality while another 16 million were in retail sectors. Granted, not all in the same industry have the same risk of layoffs (for example, a people-facing floor sales person versus a computer-facing website designer of a retailer), but a hypothetical 50% laid off from each of these sectors by themselves mean 16.5 million jobs lost.

Once we include jobs beyond those at risk from social distancing, it's only matter of time before more dominos will fall. Recent St. Louis Fed research estimated that 46% of total employed (in 2018 figures) fall in occupations at "high risk" of layoffs. Assuming between 25%-50% of these people are at risk of losing their jobs, it not hard to see between 17 million and 35 million jobs lost hard and fast. A sobering prospect.

Barely any sector was spared the job declines, with service-sector jobs suffering the most, unsurprisingly, losing 659,000 jobs in March, though the goods sector still lost 54,000 jobs for the month. In contrast, government jobs increased by 12,000, thanks to a 18,000 jump in federal census jobs for the month.

Chart 1

As we expected, paychecks were spared the carnage, with average hourly earnings climbing by 0.4% over February with a 3.1% gain over last March, a bit stronger than our 3.0% forecast. The drop in the workweek, to 34.2 hours from 34.4 hours, wasn't as extreme as earlier thought. Both wage growth and hours worked are affected by compositional bias, however. For example, as more low-wage workers get laid off relative to high-wage workers, this can give a false sense of a higher-wage trend. The highest risk of unemployment falls in the lower-paid occupations, which also means the economic burden of COVID-19 will most immediately fall on the most financially vulnerable folks.

Relief To Those Out Of Work

And, while the March jobs report showed the start of a labor-market collapse, the government's larger $2 trillion COVID-19-aid package known as the Coronavirus Aid, Relief, and Economic Security (CARES) Act (passed just last week) has (i) unemployment insurance provisions (UI) that now include an additional flat $600 per week payment (above the state benefit) to each recipient for up to four months, and (ii) extended UI benefits to self-employed workers, independent contractors, and those with limited work history. This means that those with weekly wages less than $600 are going to get more than 100% of their weekly wages in the form of unemployment checks if laid off. (Workers in the leisure and hospitality industry had average weekly wages of $435 in January and February.) Besides, direct transfers of $1,200 per person is also on the way (see table 1 for more on the CARES Act).

From the employer side of the equation, small- and midsize businesses (who employ about half of the workforce) also have direct assistance to meet payrolls and incentive to keep payroll intact in the form of loan forgiveness proportional to retained employees. A paradox we observe within this mix is that for workers who make weekly wages below the unemployment insurance benefit, it may be better to be laid off than to remain on the payroll.

Quick implementation and government willingness to scale up (dependent on duration of the current distress) will ensure households remain liquid and solvent through this severe (hopefully temporary) crunch so that the economy can recover strongly. Fiscal support can only buy us time and act as a floor preventing an economic collapse rather than providing a pure boost to activity. It can only do so much when the true underlying problem is in the hands of medical scientists.

Table 1

Timely Economic Tracker Data Reveal The Beginning Of A Sudden-Stop Downturn

• Initial claims: A record number of 6,648,000 people filed initial unemployment claims in the last week of March. This is twice last week's record number (3.3 million) and almost 10 times the previous record of 695,000 in October 1982.
• Light vehicle sales: In March, sales were 11.4 million SAAR--the weakest pace since April 2010 and down 32% from 16.7 million pace in February. Year over year, sales were down 34%.
• Open Table network restaurant reservation: This has been 100% below the comparable day a year ago beginning March 21.
• STR hotel data: U.S. hotel occupancy was down 67.5% year over year during the last week of March.
• Box Office Mojo weekly gross movie receipts: Last week, receipts were down more or less 100% from the comparable week in 2019. Month-to-date, box-office receipts were down 69% from the same period last year.
• Consumer confidence: The Conference Board Consumer Confidence Index declined 12.6 points in March to 120--the largest monthly decline since February of 2011 (when there was political imbroglio surrounding the debt ceiling). Most of the decline was driven by expectations (versus current).
• Baker-Hughes rig counts: The Baker Hughes reported that the number of oil and gas rigs in the U.S. fell by 64 this week, falling to 664 and 361 fewer than comparable period last year. It is the largest single-week decline since early 2015, and now the number of active rigs are the fewest since January 2017.

Chart 2

Chart 3

Chart 4

Chart 5

Table 2

Table 3