It's been 11 weeks since Congress passed the CARES Act, which, among its many other provisions, established the Paycheck Protection Program (PPP). Nearly three-quarters of U.S. small businesses have so far availed themselves of this program. But it turns out that fewer small businesses got a PPP loan in the states hardest-hit by the Covid-19 pandemic than elsewhere. And whereas it seemed early on that small banks had done a better job than larger reaching the smallest businesses, the most recent public data contradict that initial finding.
The PPP consists of government-guaranteed Small Business Administration (SBA) loans, of which the SBA will forgive any part that businesses use to pay wages, rent, and utilities. The SBA reports that banks and other authorized lenders have made $512 billion worth of PPP loans so far, out of the total $659 billion Congress appropriated. The total number of loans was 4,576,388, for an average loan amount of $111,900.
Whether the PPP achieved its main objective of preventing small businesses from firing their staff and closing their doors permanently will take some months to determine. If the same proportion of PPP borrowers and non-borrowers ends up shutting down, then the PPP will have been a waste—albeit not completely, as enabling firms to keep workers on their payroll will have reduced claims for unemployment insurance, if only temporarily. But failed businesses will still dissolve, destroying the intangible capital that the PPP was meant to protect. If, on the other hand, a significantly larger proportion of borrowers than non-borrowers survives, and assuming both sets of firms are similar, the PPP will have succeeded.
Table 1. Percentage of Small Businesses in Receipt of a PPP Loan, by State (as of June 6)
Answering other questions about the program's efficacy, such as what the extent of borrower fraud was, will require data that the SBA hasn't yet published. But the agency's regular reports on the PPP already offer insights into the program's reach, its distribution across small businesses, and the types of lenders that have primarily catered to borrowers of different sizes. Although many small businesses nationwide (72.6 percent) have received PPP funding, the share varies considerably from state to state, with 57.3 percent of Delaware small businesses borrowing from the PPP, compared with 96.5 percent in Mississippi. In general, Southern and Midwestern states have had particularly high rates of PPP loan penetration, while Eastern states and those in the Pacific Northwest have had lower rates.
States with high loan penetration rates in the PPP's first round, which closed on April 16 after heavy demand, tend to also have higher overall penetration rates. This pattern raises the question of whether fewer small businesses took PPP loans in laggard states because the second round of funds came too late. While the PPP's nationwide penetration rate is consistent with an April survey that found 70 percent of small businesses anticipated applying, the pandemic first hit New England, Washington state, and California, which have lower loan penetration rates than the U.S. as a whole. Some of these states' small businesses, lacking any options for short-term funding, may have shut down for good. But it's not yet possible to tell.
The size of the average PPP loan has gradually declined, with loans under $150,000 taking up just 17 percent of the total amount in the first round, but 45 percent subsequently. As of June 12, loans under the $150,000 threshold accounted for 26.6 percent of the total amount for both rounds. Smaller loans' growing weight partly reflects the public browbeating of large firms that followed the first round, when it came to light that some firms got PPP loans, crowding out smaller applicants, when they might have raised capital elsewhere. This criticism led some of the largest recipients to return their loans. It may also have discouraged other eligible large firms from taking advantage of the program in the second round.
Responding to news reports that large banks had favored their large commercial clients over other PPP applicants, Congress, in authorizing $310 billion of additional PPP funding, set aside $60 billion for small banks, credit unions, and community lenders to allocate. Specifically, Congress earmarked $30 billion for banks with between $10 billion and $50 billion in assets, and another $30 billion for banks and other lenders with assets of less than $10 billion. These congressional set-asides were based on the conventional wisdom that small banks lend more to the smallest businesses. But the truth, as I've shown elsewhere, is that large banks' average business loans actually tend to be of smaller size than those made by small banks perhaps because large banks issue most of the business credit cards on which many of the smallest businesses rely.
Does the pattern of PPP lending also belie the conventional wisdom? An early study by economists at the New York Fed seemed to suggest that it didn't: states where small banks account for a large share of retail deposits tended to have high rates of PPP loan penetration in the program's first round. (These states also happen to be among the few that were never under statewide stay-at-home orders.) But the finding may be spurious, because small banks are more common in agricultural states with sparse populations, and other Fed research suggests that firms in rural areas are more likely to have established bank credit relationships. It follows that rural firms may have found it easier than most to get PPP loans in the early days of the program. As it happens, small-bank market shares no longer seem to predict PPP loan penetration (Figure 1).
Figure 1. PPP Loan Penetration and Small Banks' Market Share, by State (as of June 6)
Despite the congressional set-aside, the PPP market share of banks with under $50 billion in assets actually declined after the first round, from 70 percent to 50 percent of the total amount. Small banks did make somewhat smaller PPP loans, on average, than larger banks: the average loan by lenders with less than $10 billion in assets was $103,500 as of June 6, compared with $141,700 for lenders with between $10 billion and $50 billion, and $119,800 among those with more than $50 billion. But there are considerable differences among the largest banks. J.P. Morgan Chase, America's largest bank by assets, is also the top PPP lender, with an average loan of $111,041. But three other megabanks (Bank of America, Wells Fargo, and U.S. Bank) that also feature among the top 10 had much smaller average loans (Table 2). Also of note is Cross River Bank's outstanding performance, which, as the small size of its average loan hints, owes much to the $3 billion bank's strong relationships with fintechs.
Table 2. Top 15 PPP Lenders (as of June 12)
In summary, banks with under $10 billion in assets made slightly smaller loans, on average, than their counterparts with more than $50 billion. But banks in-between those thresholds, despite their inclusion in the congressional set-aside, made the largest average loans of all. And some of the nation's largest banks had much smaller average loans than small banks. The PPP experience therefore still doesn't validate the cliché that it is small banks that mostly lend to the smallest businesses. Instead, it seems that fintechs focus on the smallest loans (under $50,000), with small banks catering to somewhat larger firms (asking for a PPP loan between $50,000 and $150,000), while larger banks serve larger businesses, and megabanks serve all.
It's too early to determine whether the PPP has met its own objective of helping small businesses survive the economic ramifications of the Covid-19 pandemic. But the SBA's reports point to two developments that policymakers should keep in mind as they evaluate the program's success: small businesses in the hardest-hit states got a less-than-proportionate share of PPP loans, and small banks didn't lend more, or lend to the smallest businesses, after Congress set aside funds ostensibly for that purpose.
 Of that total, Congress appropriated $349 billion under the CARES Act and $310 billion under the Enhancement Act passed one month later.
 This post uses data from the SBA's June 6 and June 12 reports. While there were some slight changes in the numbers from one report to the other, none of the patterns identified here is substantially different.
 The New York Fed researchers used $1 billion as the asset threshold for a small bank, whereas the more common threshold (and the one used in this post) is $10 billion.
I thank Nick Anthony for his excellent research assistance.