The 2007-2009 Global Financial Crisis (GFC) represents a massive risk management failure on the part of both private-sector financial firms and their regulators. The financial system itself became over-leveraged due to the growth of products like credit default swaps in the early 2000s, and lending practices became lax due to the growth of securitization in areas such as subprime mortgage lending. Many of these changes occurred, or were accelerated, by gaps in regulation. In the aftermath of the GFC, regulators have attempted to fill these gaps with new programs designed to foster resiliency, and bank stress testing is among the most important of these regulatory innovations. Although a burgeoning literature studies the efficacy of stress testing, little is known about how stress testing has changed the way banks approach risk management. Moreover, the stress testing regime has been criticized for its high degree of complexity and its reliance on detailed data and models.
Our research analyzes banks’ investment in human capital aimed at managing risk following the GFC and the advent of bank stress testing. We first document a rising demand for risk management expertise over the subsequent decade, and we show that this demand is greater among banks that suffered heavier losses during the GFC. Next, we examine how stress testing – the main innovation in bank capital requirements coming out of the Crisis – has shaped banks’ demand for risk managers. We find that banks seek to hire highly skilled stress test labor in anticipation of a test and following poor performance on a test. We then show that this higher demand translates to lower systematic risk and lower profitability. Overall, our results suggest that stress testing has catalyzed advancement in risk management through the acquisition of highly skilled talent.
Our data come from comprehensive information on online advertisements by firms looking to hire at all levels across the organization. These data capture labor demands, as they reflect the kinds of skills banks are seeking to employ. As such, they help us understand what banks are trying to achieve in their human resources policies. With these data, we document that total demand for risk management jobs climbed nearly six-fold from 12,000 job posts by banks in 2010 to over 70,000 posts by 2019, before falling during the coronavirus pandemic (Figure 1). The data clearly show that risk management has been a persistent and increasingly important factor at banks in the years following the GFC. Beyond banks, the Federal Reserve system has also increased its demand for these kinds of skills, peaking between the years 2015 – 2017 (Figure 2).
To understand these patterns, our research considers two effects of the GFC on banks’ labor demand. First, we look for, and find, evidence that banks learned about their risk exposure from the trial by fire during the Crisis and responded by demanding more risk management skills. Second, we shift our focus to stress testing, given its importance in the post-crisis regulatory regime. Stress tested banks dominate the demand for risk management jobs, accounting for over 80% of these job posts in any given year (Figure 1). Their demand for risk managers is higher not only because they are bigger, but also as a fraction of their hiring. Risk management represented about 4% of total job posts across all banks in 2010, but this share tripled to over 12% of job posts at stress-tested banks by the end of the decade, outpacing the increase to 8% at other banks.
We show that stress testing spurred bank investment in highly skilled workers, and that this led to lower risk. Specifically, banks that expect to be stress tested in the next year increase their demand for stress-test specific labor, relative to banks not tested. This effect is greatest at banks who perform worse in or fail the stress tests. Following higher demand for stress-test specific labor, banks exhibit lower systematic risk and lower profitability. This is strongest when banks demand jobs requiring higher education or advanced quantitative skills. Overall, stress testing has helped advance internal risk management practices at banks. By extension, the general tightening of regulation itself helps explain the marked upward trend in banks’ demand for risk-management skills.
In addition to strengthening internal risk management by talent acquisition, banks also invest heavily in hiring external consultants. Globally, banks’ expenditure on consultants has increased from $16 billion in 2007 to $29 billion in 2015, much of that for stress tests. Since individual banks’ expenditures on consulting services is not easily observable, the labor demand response that we estimate is likely a lower bound on the aggregate investments that banks make in their risk management practices. That is, our results likely understate the overall resources that banks dedicate towards risk management and its subsequent effect on risk outcomes (Figure 3).
Ours is the first study to analyze bank hiring patterns in the wake of the GFC, which allows us to provide a fuller understanding of how banks reacted to the Crisis. Some existing studies have emphasized the risk-management failures of both banks and their regulators. But the only research on the role of bankers themselves that we know of focuses on bank senior managers or board members and structures. While management and governance are clearly important – top executives and board members set the priorities and incentives within an organization – these few individuals cannot singlehandedly govern bank risk without the support of low- and mid-level employees, which is the focus of our work.
Our results indicate that banks have become more highly skilled and quantitatively sophisticated as a consequence of the trauma of the GFC, and that much of these changes were spurred by regulatory pressure from stress tests. These skilled professionals help banks reduce systematic risk at the cost of reduced profitability.
Recent changes in the stress testing regime, however, have reduced the severity of these tests, as well as the level of public disclosure of the results. In particular, starting in 2019 the Federal Reserve removed the qualitative assessment from stress testing, which had formerly been a core component of the regime. Prior to 2019 the results from these qualitative assessments – which depend on supervisory oversight of internal risk management policies, practices, and governance – had been publicly disclosed (and are the source of our key measure), along with the quantitative results. Our analysis suggests that this component of the stress testing regime had the greatest impact on skill acquisition. These changes, however, indicate that the tests may have lower, or perhaps no, impact of risk management hiring in the future.
Figure 1. Aggregate Trend in Risk Management Jobs
Figure 2. Stress Test & Other Risk Management Job Posting at Banks
Figure 3. Stress Test Job Postings at Consulting and other Firms
Figure 3: Stress Testing Job Postings at Federal Reserve Banks
Thomas Schneider is an Assistant Professor of Finance at Old Dominion University
Philip E. Strahan is the John L. Collins, S.J., Chair in Finance at Boston College and a Faculty Research Fellow at NBER
Jun Yang is an Assistant Professor of Finance at the University of Notre Dame.
This post was adapted from their paper, “Bank Stress Testing, Human Capital Investment and Risk Management,” available on SSRN.