How to Save 40 percent of the Payroll Protection Program’s Costs

Thanks to the loan program part of the CARES Act, small businesses can receive up to 2.5 times their monthly payroll, regardless of how much their business is affected. The program ran out of money in 13 days. Need-based loan forgiveness would save 41 percent of available resources, with no negative impact on the program’s goals.

As part of the CARES Act—the economic policy response to the coronavirus—roughly $340 billion was reserved for the Paycheck Protection Program (PPP) loans. This loan program is designed to incentivize small businesses to keep their workers on the payroll, so that people may maintain their income and to prevent otherwise-viable businesses from going bankrupt.

The loans will be forgivable regardless of how much a business is impacted. This contrasts with similar policies in many European countries, where coronavirus-motivated payroll subsidy polices are need-based: they depend on the degree to which a business is affected (e.g. percentage of lost revenue compared to last year).

My estimation suggests that the PPP could save around 40 percent of its costs by implementing a need-based loan forgiveness standard while not affecting the program’s efficacy.

The PPP works as follows. From the 3^rd of April, any small business could apply to get a loan guaranteed by the government. Eligible small businesses are those with less than 500 employees, or that satisfy the Small Business Administration’s size standard, with the exception of the food and accommodation industry in which roughly any business is eligible.

If at least 75 percent of the loan is spent on payroll, the rest may be used for interest, rent, and utilities, and the loan will be fully forgiven. The maximum loan amount is 2.5 times the monthly payroll for the previous year (using a $100,000 compensation cap per employee), up to a maximum of $10 million, and must be used within eight weeks of the loan’s disbursement.

The loan duration is two years, with an automatic six-month period before repayments on the not-forgiven amount begin, all at an interest rate of 1 percent. If workers are fired or salaries and wages are decreased, the forgivable share of the loan is adjusted.

Since it is essentially free money, current loan-forgiveness will spur businesses to request and receive a gift of 2.5 times their monthly payroll, regardless of how much their business is affected.

For some businesses, the full 250 percent of monthly payroll loan is probably a decent estimate of lost revenue and they need this full amount forgiven—a lunchroom closing its doors for two full months while maintaining payroll, for instance.

Many other businesses probably face a partial loss, and some might not be affected at all—my local supermarket probably never had more business.

Since the PPP does not differentiate between these cases, it provides both much-needed help and generous handouts.

How much would be saved if the much-needed help is maintained, but the handouts eliminated?

Combining sector-level data on (1) allocation of PPP funds, (2) payroll costs of eligible small businesses, and (3) how much sectors might be affected, allows an estimate of the total savings under a need-based version of PPP where loans would need to be repaid in proportion to how much a business is affected.

Since loans from need-based PPP would be available in an identical manner, with only the ex-post forgiving criteria altered, this change would not create any administrative hurdles that worry many with ex-ante loan criteria.

First, allocations of the first round of PPP funds per sector have been released. Second, John Barrios (Chicago Booth), Petro Liswosky (BU Questrom), Michael Minnis (Chicago Booth), and William Minnis (EIU Business) launched a tool that estimates monthly payroll costs for eligible small businesses. This is considered a higher-end estimate, not accounting for the PPP’s $100,000 income cap, $10 million loan cap (violations of $10m cap have been reported), and the exclusion of some small business because they are affiliated to other businesses (think subsidiary).

Lastly, Jonathan Dingel (Chicago Booth) and Brent Neiman (Chicago Booth) recently estimated what share of jobs can be done from home, per industry. Their overall estimate of home-viable jobs is 37 percent.

Daily data from Homebase, a company providing time-tracking tools used by over 100,000 small businesses, showing total hours worked declining more than 60 percent, with over 40 percent of firms having shut down entirely, suggests this is the right order of magnitude.

The savings estimates are constructed by sector as the share of home-viable jobs, weighted by a sector’s share in total PPP eligible small business payrolls, or the sector’s share in PPP allocations to date. The estimation by Barrios et al. for sector PPP eligible payrolls can be found in their published tool, to which I add a small improvement.

Businesses with more than 500 employees are eligible for PPP in ~30 percent of 6-digit NAICS sectors, which their estimate does not incorporate. Using these definitions and the same payroll data, this improvement adds $5 billion to their $287 billion estimate of eligible small business monthly payrolls and slightly adjusts the sector shares.

Based on this payroll by sector, combined with Dingel and Neiman’s shares of home-viable jobs per sector, I arrive at an estimate of the savings of a PPP program that would base forgiving the loan on how much a sector is affected, to the extent that Dingel and Neiman’s numbers provide a good proxy for this.

If loans will be approved across industries in the same proportion they have been so far, need-based loan forgiving would result in cost savings of 41 percent. If PPP would be approved by sector in proportion to payroll expenses, the program could save 44 percent of its costs.

Loans that are not forgiven cost approximately nothing. Hence, these savings percentages should be applied to the total value of forgiven loans. The savings of a need-based PPP are thus estimated to be around $41-44 billion, per the $100 billion of loans that will be forgiven under current rules.

A point estimate for savings from a need-based PPP depends on the final size of the program, and the share that will be forgiven. If, for example, the PPP will run $650 billion—its original size plus the currently-rumored additional funding—with 70 percent of loans forgiven, total savings are around $190 billion. This equals two-thirds of the total estimated costs for the $1200 direct payments/stimulus check program.

Fast and wide-coverage relief is essential to maintain all our livelihoods and for the economy to rebound as fast as possible when the health situation allows. We should, however, always remain mindful of exactly who should and who will benefit from a policy, especially when its costs will have to be repaid collectively.

The PPP’s initial $340 billion ran out in only 13 days, while intended to last for 4 months. A $300 billion replenishment is currently being negotiated and will probably be announced this week.

Transforming the current PPP into a need-based PPP does not have to impact the efficacy of the program, while most likely it will result in very significant savings.

Joost Sijthoff is a research professional at the Stigler Center, at the University of Chicago Booth School of Business. Calculation available per request: joost.sijthoff@chicagobooth.edu

ProMarket is dedicated to discussing how competition tends to be subverted by special interests. The posts represent the opinions of their writers, not necessarily those of the University of Chicago, the Booth School of Business, or its faculty. For more information, please visit ProMarket Blog Policy.