The employment gains attributable to the Paycheck Protection Program have been modest relative to the huge outlay of government funds, according to different teams of economists using varied statistical techniques on large data sets from private sector sources. For example, one study (with 10 authors) estimated that the PPP boosted employment by 2.3 million jobs at a cost of $224,000 per job.
After providing background, we review the tentative results of four recent economic studies’ estimates of the effect of the PPP on reducing job losses. We also highlight other notable findings from these studies about the PPP and the economic impact of the pandemic.
Background
The current recession’s effect on employment was far larger and more rapid than that of the Great Recession of 2007-2009. Figure 1 shows that from November 2007 through December 2009, U.S. employment declined by 8.6 million jobs; it didn’t return to 2007 levels until 2014.
In February U.S. employment declined by 3 million jobs. During its nadir in March, employment declined by an astounding 22 million jobs. U.S. employment has since increased by 13 million jobs. That still leaves 9 million fewer jobs compared with pre-pandemic levels.
Businesses started applying for PPP loans on April 3. A bit more than five months later, on August 8, the program closed to new loan applications, with a total of $525 billion in approved loans to 5.2 million borrowers. The average loan amount was $101,000. Those disbursed funds equal about 2.4 percent of GDP. The entire cost of the fiscal stimulus enacted into law in February 2009 was larger — equal to 5.3 percent of GDP — but the outflow of funds to the public was spread over several years.
Figure 2 shows the amount of job losses by major industry during the pandemic. Each sector had job losses from February through April, and each has since recovered from the lows of April. Leisure and hospitality (which includes hotels and restaurants) account for about a third of all losses throughout the pandemic. The retail and construction sectors have made a considerable comeback.
Millions of Jobs
A study by Raj Chetty and coauthors estimate that the PPP slowed the reduction in employment at small businesses by about 2 percent from pre-pandemic levels, implying a cost of $377,000 per saved job. Of course, estimates are uncertain. Using their statistical findings, they report that the cost per saved job exceeded $119,000, with a 95 percent probability. The average annual earnings for employees at PPP-eligible companies were $45,000. (Complete study citations are at the end of this article.)
João Granja and coauthors estimate that job creation attributable to the PPP is between 3.2 million and 4.8 million. That is between 4.5 percent and 6.8 percent of 70 million eligible workers. This implies a cost per saved job of between $109,000 and $164,000. Figure 3 illustrates how job losses differed for small businesses with low and high access to PPP lending.
David Autor and 11 coauthors conclude the PPP increased the level of employment at eligible businesses between 2 and 4.5 percent, implying an increase in employment in the United States by between 1.4 million and 3.2 million, with a midpoint of 2.3 million. This translates into a cost per saved job of between $380,000 and $162,000, with a midpoint of $224,000.
The authors of these studies carefully describe their assumptions as well as the potential shortcomings in their data and research design. But when they conduct numerous additional statistical tests to examine whether those potential problems could require a reassessment, they find very little to raise serious doubt about the thrust of their conclusions. Each study stands on its own as an important contribution to understanding the economic impact of the PPP. The sum is greater than the whole of its parts. Together the studies complement one another, and given how they independently arrive at similar results, little room is left to doubt that the job creation attributable to the PPP programs was small compared with the program’s massive disbursement of funds.
A study by Alexander Bartik et al. provides a wide-ranging overview of the pandemic’s effect on employment. Its findings about the employment effects of the PPP are consistent with the three studies noted above (although the statistical analysis is far less rigorous): U.S. states that received the least PPP money saw slower recoveries than states that received more funds.
I’ll Take Manhattan, You Take the Bronx
Along with the important results about the effect of the PPP on employment described above, these studies provide useful information that would be a shame to neglect here:
In the first month of the crisis, 30 to 40 percent of small businesses didn’t experience sales declines. (Granja et al.)
Older workers, African American workers, Asian workers, single workers, and women were more likely to lose their jobs in April and were less likely to start work again in May or June. (Bartik et al.)
Spending reductions were concentrated in services that require in-person physical interaction, such as hotels and restaurants. (Chetty et al.)
Online purchases increased by 37 percent from the first to the second quarter of 2020. (Chetty et al.)
Low-income people who worked in more upscale area codes (for example, Manhattan) remain less likely to be employed months after losing their jobs than those working in less affluent areas (for example, the Bronx). (Chetty et al.)
High-income households cut spending more in percentage terms (by 13 percent) as of mid-July than low-income households (by 4 percent). (Chetty et al.)
During the pandemic, unemployment surged in affluent areas that in previous recessions had relatively low employment declines. (Chetty et al.)
Traditional macroeconomic tools like stimulating aggregate demand or providing liquidity to businesses are less effective when consumer spending is constrained by health concerns. (Chetty et al.)
Data on consumer spending in the initial stages of the pandemic show that spending fell primarily because of health concerns rather than a loss of current or expected income. (Chetty et al.)
The recession’s effects are much deeper and more prolonged for lower-wage workers. (Chetty et al.)
An estimated 70 percent of eligible businesses applied for and received PPP loans. (Autor et al.)
PPP funds initially flowed to regions that were less adversely affected by the pandemic. (Granja et al.)
Many businesses used the loans to pay rent, make interest payments, and build up cash balances. (Granja et al.)
The fraction of businesses receiving PPP lending depended on their local banks’ degree of participation in the PPP, driven not by differences in demand for PPP loans across regions but by their exposure to banks that underperformed. (Granja et al.)
In the second round of PPP funding, nonlocal banks and non-banks provided loans to businesses that were unable to access PPP loans from local banks in the first round. (Granja et al.)
Economists Are Relevant
Empowered with big data from private sector sources, economists have been able to rapidly provide quality, in-depth analysis of this unique fiscal stimulus provision, which is one of the largest in U.S. history. This type of timely analysis from the economics profession hasn’t been the norm. If the goal is job creation, lawmakers should carefully consider results like those summarized here when allocating funds among different programs. Overall, research to date suggests that the most effective policies — from both a social welfare and macroeconomic stimulus perspective — are those that can be targeted to the relatively small subsect of individuals and businesses with the most need.
References
David Autor, David Cho, Leland D. Crane, Mita Goldar, Byron Lutz, Joshua Montes, William B. Peterman, David Ratner, Daniel Villar, and Ahu Yildirmaz, “An Evaluation of the Paycheck Protection Program Using Administrative Payroll Microdata,” MIT Working Paper (July 2020).
Alexander W. Bartik, Zoe B. Cullen, Edward L. Glaeser, Michael Luca, Christopher T. Stanton, and Adi Sunderam, “The Targeting and Impact of Paycheck Protection Program Loans to Small Businesses,” National Bureau of Economic Research Working Paper No. 27623 (July 2020).
Raj Chetty, John N. Friedman, Nathaniel Hendren, and Michael Stepner, “The Economic Impacts of COVID-19: Evidence From a New Public Database Built Using Private Sector Data” NBER Working Paper No. 27431 (Nov. 2020).
João Granja, Christos Makridis, Constantine Yannelis, and Eric Zwick, “Did the Paycheck Protection Program Hit the Target?” (Nov. 2020).