2019 Series • No. 19–1
Current Policy Perspectives
Stress Testing Effects on Portfolio Similarities Among Large US Banks
In the wake of the 2008–2009 financial crisis, the United States enacted regulatory reforms designed to strengthen the country’s financial system. One new policy change, implemented in 2011, is the Dodd-Frank Act’s requirement that each of the largest banks operating in the United States—those with more than $50 billion in assets—undergo an annual stress test to examine the losses that might occur if there is a significant deterioration in economic conditions. After evaluating how an individual bank would fare under both an adverse and an extremely adverse scenario used in the stress testing exercises, those institutions that performed poorly on the tests, judged by having a high capital shortfall, then adjust their portfolios and reinforce their capital buffers. The goal of the stress tests is to make the largest US banks more resilient to shocks and default risk by proactively correcting for any weaknesses that are detected before a negative event actually occurs.
This paper examines how the behavior of major banks has changed in response to the stress testing requirements. While complementary studies have used stock market data to document an increased correlation in default probabilities and equity returns within the US banking industry since 2010, this paper uses the quarterly data on wholesale lending activities that banks report to the Federal Reserve to comply with supervisory requirements (the FR Y-14Q data collection). This loan-level data enables the authors to compare the largest banks based on several dimensions of their individual portfolios and to trace how these portfolios have changed over time.
Key Findings
- The banks that did not perform well on the stress tests tended to have dissimilar portfolios before 2011. However, after the tests showed which banks would experience high capital shortfalls under the extremely adverse scenario, these poor-performing banks altered their portfolios to more closely resemble the portfolios of the banks that did do well on the tests (judged by experiencing low capital shortfalls when under stress). As a result of these shifts, there has been an extraordinary increase in the amount of loss-absorbing capital buffers, as well as a reduction in balance sheet risk, for each of the largest banks operating in the United States.
- Since the stress tests began in 2011, the overall portfolio composition of the largest US banks has increased in similarity by more than 5 percent, on average, through 2016:Q4. Yet during this same period, those banks that are not subject to the annual stress testing have, for the most part, not rebalanced their portfolios. This suggests that the need to adhere to more stringent regulatory oversight enacted since the financial crisis has driven the portfolios adjustments made by the largest banks.
- The authors show that while the individual portfolios of the largest US banks have become more diversified, the greater convergence of the portfolios held by these banks may be inadvertently increasing the aggregate banking sector’s systemic risk factors.
Implications
These findings have implications for financial stability goals. While the current stress testing methodology has resulted in individual banks better managing their balance sheet risk and reinforcing their capital buffers, this improved microprudential (bank-level) management may have the unintended result of increasing the entire US banking sector’s systemic risk, and hence could be defeating the macroprudential (system-wide) goals of the post-crisis regulatory reforms. While the US banking industry has become more concentrated since the financial crisis, more analysis is needed to determine if the sector taking on risks that are not captured in the current stress test scenarios. However, one clear conclusion from this paper is that the scenario design used in the Dodd-Frank stress testing should be as comprehensive as possible. Individually diversified bank portfolios that seem well-insulated from shocks when economic conditions are relatively benign may give rise to a more sensitive aggregate banking system during periods of stress, particularly if the stress test scenarios do not capture all the potential systemic risk factors.
Abstract
We use an expansive regulatory loan-level dataset to analyze how the portfolios of the largest US banks have evolved since 2011. In particular, we analyze how the commercial and industrial and commercial real estate loan portfolios have changed in response to stress-testing requirements stipulated in the 2010 Dodd-Frank Act. We find that the largest US banks, which are subject to stress testing, have become more similar since the current form of the stress testing was implemented in 2011. We also find that banks with poor stress test results tend to adjust their portfolios in a way that makes them more similar to the portfolios of banks that performed well in the stress testing. In general, stress testing has resulted in more diversified bank portfolios in terms of sectoral and regional distributions. However, we also find that all the large US banks diversified in a similar way, creating a more concentrated systemic portfolio in the aggregate.