This page documents our estimated requests by small businesses for the new $284 Paycheck Protection Program. To see our analysis of the original allocation of the PPP, go our Original PPP estimates page.
Our baseline estimates suggest that approximately $185 billion of the $284 billion PPP will be requested. To see our detailed calculations, see this Google Sheet. This is a view only link, so you need to download the spreadsheet locally to edit the sheet to make your own estimates.
Our key assumptions for the baseline estimates are:
Our estimates start with the full SBA data set of PPP loans in the original PPP provided here. We further assume that no companies will apply for this round of the PPP which did not apply for the first round. One can also view this assumption as: any requests by new companies applying this round are offset by fraudulent companies that applied last time NOT applying this time.
Only small businesses with less than 300 employees are eligible, which are represented by the companies which reported "jobsreported" in the SBA data of 300 or fewer. (We also include companies with missing "jobsreported" figures.)
Only companies with revenue declines of more than 25% for any one quarter are eligible to apply for PPP. To estimate how many companies are eligible, we start with the weighted average percent changes in quarterly gross output by industry provided by the Bureau of Economic Analysis (sourced here). We then assume that small businesses performed worse than the larger businesses by an average of 6 percentage points, and therefore subtracted 6% from the average growth in the industry to derive an estimated average growth for small businesses by industry. We further assumed that growth across companies within each industry is distributed normally with a standard deviation of 25%. While a normal distribution is not a perfect representation of growth rates across firms within an industry (growth rates do not typically follow a normal distribution because they are truncated at -100% by definition), it allows us to estimate how many firms leave the sample by further assuming that all firms with growth rates estimated to be below -100% are out of business and will not apply. In sum, we:
Use mean industry-level quarterly growth rate changes from BEA.
Subtract 6% from the weighted industry growth averages to estimate small business mean growth rate changes.
Assume a normal distribution of growth rate changes with 25% standard deviation.
Assume any company with an estimated growth rate change of less than -100% is out of business and will not apply for a new round of PPP.
Each company which we estimate as being eligible (i.e., less than 300 employees and more than 25% reduction in sales) applies for the exact same PPP amount as the original PPP except:
PPP amounts are maxed out at $2 million per company.
Companies in industries 71 and 72 can apply for 3.5 months of payroll rather than 2.5 months.
Assume that companies do NOT manipulate their sales figures. Our spreadsheet does allow for scenarios which consider manipulation.
Please email any comments, questions, or suggestions to: michael.minnis [AT] chicagobooth.edu
Co-developers: John Barrios (Washington University), Michael Minnis (University of Chicago Booth School of Business), William Minnis (Eastern Illinois University), and Joost Sijthoff (University of Chicago Booth School of Business).