Who wins when banks distribute government loans? Funding small businesses in times of crisis

Small and medium-size enterprises (SMEs) employ the majority of workers around the world. These companies are perhaps the most vulnerable and stand to be the most affected during economic contractions. To lessen the impact and protect jobs, policy makers often provide much-needed liquidity via special lending programs allocated through private banks. In developed and developing countries, many public credit programs rely on the private financial sector to reach SMEs (IMF 2020).

So, what issues arise when government credits or grants are channeled via private banks? In this blog, we summarize what we have learned from studying one of the world’s largest second-tier lending programs: the case of earmarked credit in Brazil.

The Brazilian government-sponsored credit program was designed to stimulate investment and capital expenditures in strategic sectors. A large portion of the program funds are transferred from the Brazilian National Development Bank (BNDES) to private banks that then select loan recipients. These government-sponsored loans, known as earmarked loans, complement “free market” commercial credit. Similar to the Paycheck Protection Program in the United States, interest rates on earmarked loans are regulated and set below market rates. Starting in 2008, the Brazilian government expanded the program substantially, to offset the negative effects from the global financial crisis, especially among the more credit constrained SMEs.

Using loan-level data from the Central Bank of Brazil for 2005 to 2016, we study how the allocation of government loans via private banks takes place. We also examine cross-selling strategies that banks might be using when giving an earmarked loan to a firm.

Before presenting our findings, here are some stylized facts from our sample. About 6 percent of the firms are granted earmarked credit. Somewhat surprisingly, most of these firms obtain multiple and frequent earmarked loans — with the median time between consecutive earmarked loans being nine months. Perhaps less surprisingly, 92 percent of the firms with multiple earmarked loans exclusively receive this credit from the same bank.

Which firms benefit from earmarked lending in Brazil?

• Banks tend to select larger firms (see figure 1) and especially borrowers with an existing credit relationship. The incentives for banks seem to be straightforward. Banks that service an earmarked loan bear part of the credit risk of this loan. Banks reduce these risks by selecting borrowers that are ex-ante less risky (larger customers and already established clients).

• We find that larger banks are more active in the program. In turn, firms with outstanding credit relationships with the largest banks have greater access to earmarked funds. It seems that who you bank with matters for whether you can access the stimulus or not.

Figure 1. Average share of earmarked credit of firms, by firm size

Note: The vertical line marks the date when the BNDES earmarked credit program expanded. Firms in the industry and construction sectors are considered micro if they employ fewer than 20 workers; small if they employ 20 to 99 workers; medium if they employ 100 to 499 workers; and large if they employ 500 or more workers. Firms in the trade and services sectors are considered micro if they employ fewer than 10 workers; small if they employ 10 to 49 workers; medium if they employ 50 to 99 workers; and large if they employ 100 workers or more.
 

After a bank grants an earmarked loan to a firm, what happens to new loans in the “free market” of the same firm-bank pair?

As some say, there is no free lunch… one interesting stylized fact that we see is that as soon as BNDES expanded its earmarked program, larger banks that participated in it more actively began raising the cost of their working capital loans (figure 2). How is this pattern related to the earmarked program?

Figure 2. Average interest rate spread of working capital loans, by type of bank

Note: The vertical line marks the date when the BNDES earmarked credit program expanded.

Banks appear to compensate for the limited expected revenue from earmarked loans by increasing their interest rates in other free-market loans that recipient firms take . Our findings suggest that banks are willing to disburse earmarked loans at below-market interest rates to riskier borrowers, as long as the banks can adjust the interest rates of other credit products in the competitive credit market to these same firms (suggesting a type of cross-selling strategy).

To conclude, similar government lending programs, in which resources are channeled by private banks, are likely to yield similar outcomes as those we observe in the Brazilian program. In particular, larger firms, those that bank with the largest private lenders, and those that are willing to bundle free-market and earmarked loans could have greater chances to access the program. 

The key policy question is how to target the funds better, but more research on this is needed. When economic contractions unfold fast, programs tend to be less focused on targeting and more focused on scale and immediacy. In such situations, our evidence suggesting that a large portion of government lending funds are not going to the most vulnerable firms might be the norm rather than the exception.

References

• IMF, 2020. Policy Responses to COVID-19
• Ornelas, J.R.H., Pedraza, A., Ruiz-Ortega, C. and Silva, T.C., 2019. Winners and Losers When Private Banks Distribute Government Loans: Evidence from Earmarked Credit in Brazil. Policy Research working paper; no. WPS 8952. Washington, D.C.: World Bank Group