Over the past decade or so the banking sector has gone through a number of transformative changes that have had an important impact on the level and types of risk in the banking sector. At the same time, measuring risk is always challenging because risk is inherently intangible and abstract. Fortunately, many economists are devoted to systematically measuring these risks with empirical data and two new pieces of research from the Federal Reserve Bank of New York (FRBNY) shed important light on how risk in the banking sector has evolved over the past decade. In this post, we provide an overview of this research and discuss how the findings relate to regulatory policy. On balance, the findings point to lower risk in the banking sector.
Three Out of Four Risk Measures Have Declined
In Have the Risk Profiles of Large U.S. Bank Holding Companies Changed?, FRBNY economists examine the trends in four separate risks over the past decade. They consider 1) idiosyncratic risk, or risk that is unique to a particular bank and not shared by other banks, 2) systematic risk, or risk that is shared across the entire banking sector due to common economic and financial developments, 3) systemic risk, or risk that is shared across the banking sector due to severe economic dislocations, and 4) liquidity risk, or the risk that a bank loses access to stable funding to finance the loans and other forms of credit it provides to the economy. Each of these risks is measured using a variety of proxies that are imperfect, but informative about each type of risk over the 2000-2018 period.
Figure 1: U.S. BHCs’ Risk Measures Have Decreased in Recent Years
Source: Federal Reserve Bank of New York, Liberty Street Economics: Have the Risk Profiles of Large U.S. Bank Holding Companies Changed?
The key findings of the research are depicted in Figure 1, which presents the four risk measures separately for banks (bank holding companies) with total assets between $25 billion and $250 billion and those with total assets greater than $250 billion. Overall, the researchers conclude that “the largest BHCs (bank holding companies) experienced notable declines in idiosyncratic, systematic, and systemic risks after the global financial crisis. However, BHCs appear to be more sensitive to liquidity risks triggered by general stress in funding markets.” The decline in idiosyncratic, systematic and systemic risk is easy to identify from Figure 1. The increase in liquidity risk is more subtle, but the authors do document some increase in their liquidity risk measure for both larger and smaller banks. Taken as a whole, the results indicate a clear reduction in several risk measures and some increase in one measure of liquidity risk.
Liquidity Risk in Focus
The research cited above uses a market-based measure of liquidity risk that correlates movements in bank funding costs with bank stock prices. Such market-based measures are not uncommon in economics, but it is important to carefully consider what they do and do not tell us. Specifically, market- based measures place a lot of credence in the pronouncements of financial markets in assessing risk. It is often useful to benchmark market-based risk measures against other measures that look at more fundamental and readily observable risk characteristics. Thankfully, in the case of liquidity risk, there are a number of other liquidity risk measures to be considered.
In Banking System Vulnerability: Annual Update, FRBNY economists (different from the economists referenced above) consider a number of risk measures across the entire banking sector including a “liquidity stress ratio” that measures any mismatch between the liquidity of the assets and liabilities on bank balance sheets. A higher value of the ratio means a greater liquidity mismatch and greater liquidity risk. Unlike the liquidity risk measure considered by the other authors, the liquidity stress ratio is based on an actual measurement of liquidity risk fundamentals and is not based on the market’s assessment of bank liquidity risk.
Figure 2: Trends in Liquidity Risk
Source: Federal Reserve Bank of New York, Liberty Street Economics: Banking System Vulnerability: Annual Update
As it relates to liquidity risk, the authors find, as shown in Figure 2, that liquidity risks have declined considerably over the past decade. The authors describe the state of liquidity risk in the banking system as follows, “the largest U.S. banks are in a much stronger liquidity position, on aggregate, than at any point before or during the 2008 crisis.” In addition, the authors also consider the stability of bank funding costs and find that “while unstable funding increased in absolute terms, it was outpaced by a concurrent increase in stable funding, leading to a decline in the unstable funding share.” Accordingly, the banking sector is less reliant on unstable funding sources now than it has been over the past decade.
These findings are important for two reasons. First, they provide an important check on market-based measures. Second, the research findings recognize the significant influence that liquidity regulation has had on liquidity risk and the liquidity position of large banks. In addition to the liquidity coverage ratio, large banks are now subject to a number of regulations, including resolution-related liquidity requirements and long-term debt requirements that reduce liquidity risk and help ensure that large banks are funded by a significant amount of long-term and stable liabilities. In particular, as shown in Figure 3, the proportion of Forum member total liabilities accounted for by stable deposits has risen substantially from 37 percent to 58 percent of all liabilities over the past decade. This evidence strongly suggests that large bank funding and liquidity risk has in fact declined rather than risen.
Figure 3: Financial Services Forum Members’ Deposits as a Proportion of Total Liabilities
Source: Federal Reserve Y-9C and staff calculations
Finally, it is worth noting that all of the other risk measures considered in the FRBNY’s annual banking system vulnerability assessment point to lower and not higher risk, which is broadly consistent with the previous findings that three of four risk measures have declined over the past decade.
So, where does all of this leave us? One recent research report by FRBNY economists shows that three of four risk measures have declined significantly over the past decade while one market-based measure of liquidity risk has risen somewhat. A second research report from the FRBNY that measures liquidity risk using a fundamental balance sheet measure of liquidity risk shows a significant decline in liquidity risk. Both of these recent research reports provide valuable insight into the level of risk in the banking system and overall point to lower risk in the banking sector.