PPP and the Unemployment Rate

by kinzaman

BLS released its unemployment numbers last week, showing a decline in unemployment month over month to 13.3%. The number of unemployed is 21 million. Absent from this statistic is the number of “employed” individuals who are still on payroll through PPP, but are not actually working or who would be laid off in the absence of PPP. So PPP is masking the extent to which labor is being underutilized and the extent to which demand shortages exist.

For those not familiar with the Paycheck Protection Program (PPP), this program provides loans to small businesses so that they can continue to pay workers and certain business expenses. If at least 75% of the proceeds of the loan are used for payroll, the loan will be forgiven. Otherwise, the loan must be paid back in full with 1% interest in two years.

The first round of PPP was launched April 3 at the heigh of the lockdowns. The budgeted $310 billion, intended to last for months, was doled out in less than two weeks. A second round of PPP was approved on April 27. As of May 30, A total of 4.48 million loans had been approved for a total of $510 billion. Assuming one loan per business, this means that 8.96 million businesses took out PPP loans with an average loan size of $113,800. In 2018, there were an estimated 30.2 million small businesses with 58.9 million employees, implying 1.95 employees per small business. So 8.96mil*1.95 = 17.5 mil employees being covered by PPP.

The average amount spent by employers per hour to keep an employee on payroll in December 2019 was $34.72. Average weekly hours worked by an employed individual in the same month was 34.3. So per week, the cost of keeping the average employee on payroll is $1,191.

Let’s assume that all borrowers are using 75% of the loan for payroll to meet the loan forgiveness requirements.

With an average loan size of $113,800 per business (75% of which is for payroll) payroll of $1,191 per employee-week and 1.95 employees per business, PPP should cover payroll for the average business for 37 weeks. So roughly 29 weeks until PPP starts to run out for the first wave of borrowers.

Assuming PPP loans were only taken out by businesses who actually needed them to afford to keep their employees on the payroll, there are 17.5 million “employed” who are either not working and continuing to collect a paycheck (effectively unemployed, but receiving the equivalent of unemployment benefits from their employer) or who may no longer be employed once PPP runs out if macro conditions do not improve. In the absolute worse case scenario that none of these jobs came back and the unemployed in May remained unemployed, unemployment would be (21 mil + 17.5 mil)/158 mil = 24.4%.

24.4% is certainly too high; some of these jobs will be coming back once PPP expires. But 13.3% is too low; some of those on PPP are “employed”, but not working (the BLS concedes that accounting for this number increases their May unemployment estimate by about 3%) and many of those who are working are working in jobs that continue to exist only through government intervention. I have seen many anecdotes on Reddit about employees who expect to be laid off once PPP runs out.

TLDR: I don’t think the BLS is manipulating their unemployment rate, but it greatly understates the lack of demand for labor and just how much trouble the economy is in.