Unintended Side Effects: Stress Tests, Small Businesses and Economic Dynamism

29 Jan 2020
Read Time 4 mins
Categories :
Bank Capital
, Small Business Lending

Introduction

U.S. small businesses are an engine of growth that support one of the world’s most innovative and dynamic economies.  New empirical research from the Bank for International Settlements provides compelling evidence that bank stress tests in the U.S. have reduced small business lending and stunted entrepreneurship, innovation, and productivity.  In this post, we review this research and discuss how data-based analysis should inform regulators’ views on the costs and benefits of stress tests.

In Unintended Side Effects: Stress Tests, Entrepreneurship, and Innovation Sebastian Doerr examines the impact of stress tests in the U.S. on loans to businesses with a value of less than $1 million that are secured by home equity.  Business loans that are secured by home equity are more likely to go to smaller, newer firms that don’t have a long track record to demonstrate their creditworthiness or other collateral, such as an existing factory or inventory, to secure a loan.  Also, these loans tend to be relatively small.  According to the data used in the study, as of 2018, the median loan in this category was $155,000, far less than $1 million and squarely in line with what most people would typically regard as a “small business loan.”  Accordingly, these home equity-backed small business loans benefit early stage companies that are trying to establish themselves and bring new ideas, products and services to market.

Stress Tests and Small Business Lending

The paper provides concrete evidence that banks subject to the stress tests significantly reduce their willingness to extend small business loans.  Specifically, Doerr finds that “stress tested banks reduce small business lending by 26.7% more than non-stress tested banks.”  The paper suggests that stress tested banks are less willing to make small business loans secured by home equity because the stress tests assume a very sharp decline in house prices that would significantly reduce the value of the loan’s collateral and increase loan losses in the stress test. These findings are also consistent with other research documenting that the stress tests have led to a decline in small business lending among stress tested banks.

Reduced Small Business Lending and Economic Activity

A second important contribution of this paper is that it links reduced small business lending to reduced economic activity.  Table 1 shows the impact of moving from a U.S. county with a relatively low proportion of stress tested banks to a county with a high proportion of stress tested banks on employment, wages and patents.  The interpretation of the research paper is that this reflects the economic impact of reduced small business lending in counties with a higher (versus lower) proportion of stress tested banks.  The research paper makes use of a number of control variables to help ensure that the results in Table 1 reflect the impact of reduced small business lending rather than other confounding factors such as changing trends over time.

Table 1: Impact of Stress Tests on Economic Activity

Activity IndicatorChange in Activity Indicator Between Countries with High and Low Proportion of Stress Tested Banks (%)
Employment-1.8
Wages-2.6
Patents-16.2

Source: Unintended Side Effects: Stress Tests, Entrepreneurship, and Innovation

The first row of Table 1 shows that moving from a county with a low proportion of stress tested banks to a county with a high proportion results in a reduction in employment of 1.8%.  In addition, the paper finds even larger declines in employment at younger firms that are more likely to use small business loans.  The second row of Table 1 shows that moving from a county with a low proportion of stress tested banks to one with a high proportion results in a reduction in wages of 2.6%.  Again, the effects are larger when focusing on the wages of younger firms that are more likely to use small business loans.  Finally, the third row of Table 1 shows that moving from a county with a low proportion of stress tested banks to one with a high proportion results in a reduction in patents of 16.2%.  And once again, the paper shows that these effects are concentrated in younger firms that are more likely to be impacted by reduced small business lending.

These results are important for two reasons.  First, they demonstrate in concrete terms the real economic effects of reduced small business lending.  Reduced lending reduces employment, wages, and new idea creation via fewer patents.  To the extent that economic dynamism is fueled by innovation, the patent results are perhaps the most troubling of all.  Second, these results provide an answer to a common supposition that any reduced lending from stress tested banks will simply be made up by increased lending at other banks.  According to this theory, reduced lending by stress tested banks has no aggregate economic costs as other banks “pick up the slack.”  These results suggest that this offset does not occur because on net, economic activity declines in those counties with a large presence of stress tested banks.                 

Conclusion

Recent research shows that banks subject to the stress tests significantly reduce their provision of home equity-backed small business loans.  These are loans that typically benefit younger firms that are so often the source of the new ideas and products that make the U.S. economy so vibrant.  The research also shows a link between reduced small business lending and several measures of economic activity.  U.S. counties with a high proportion of stress tested banks exhibit lower employment, lower wages, and fewer new patents than counties with a lower proportion of stress tested banks.  These results are important because they suggest that reduced small business lending by stress tested banks has aggregate economic costs and is not offset by increased lending by other banks.  As the author states: “This paper provides new evidence that post-crisis financial regulation has stymied entrepreneurship.  I show that stress tested banks strongly cut small business lending secured by collateral and that the contraction in loan supply has had real effects.”  As regulators evaluate the costs and benefits of the stress tests, they should evaluate its effects on small business formation and aggregate economic activity.

Forum Updates