Key takeaways

• We found systemically important banks have strong positive equity and negative quality exposure, which may indicate vulnerability to market downturns.

• Compared to systemically important banks, we found that U.S. regionals are smaller and more value-oriented with their domestic focus being a large driver of relative risk.
   
• If banks continue to make headlines, we believe factor analysis may provide an important historical context of their risk exposure.

In March 2023, Credit Suisse, a struggling and systemically important (SI) bank, was set to be acquired by long-time competitor, UBS. Additionally, one major U.S. regional bank voluntarily liquidated and two failed. 

In our recent factor performance report, we discussed how risk factors were affected by this bank failure news, including the aftermath when markets rallied. However, there exists a longer-term relationship between banks and risk factors that may be equally educational. 

Using factor analysis in Two Sigma’s Venn, we can understand the historical risk drivers of both SI¹ and U.S. regional banks² beyond what we see in current headlines. This can help provide even greater context around the relationships we might expect banks to be a part of during market shocks, and what is unexpected.

Factor analysis of systemically important banks
Using Two Sigma’s Factor Lens, we are able to look at the history of SI banks on Venn to understand which market risk factors they have been exposed to and how those factors have driven their risk and return. More specifically, SI banks are proxied by the KBW Nasdaq Global Bank Index, which is designed to track the performance of leading banks that have been classified as “Globally Systemically Important” by the Financial Stability Board (FSB) and Basel Committee on Banking Supervision. Using Venn’s unique factor selection process, we found that eight factors were the most relevant (bolded in exhibit 1) for SI banks.

Exhibit 1: Factor Exposure of Systemically Important Banks

Source: Bloomberg, Venn by Two Sigma. Period from 7/14/2015–4/10/2023. Chart shows Venn analysis being conducted on KBW Nasdaq Global Bank Index.

• Positive exposure to the equity factor: For our analysis, we are using equity indexes of banks, so unsurprisingly there is a large exposure to our equity factor. More specifically, the beta to our equity factor was well over 1.00, indicating that bank stocks are even more exposed to macroeconomic growth and profitability than the average stock. 
   
• Positive exposure to the value factor: Banks tend to have favourable value characteristics such as high dividend yields and low prices to book. For example, the Russell 1000 Value index currently has a 5.1% allocation to banks while the Russell 1000 Growth has 0%.³
   
• Positive/negative exposure to foreign currency/local equity factors: Importantly, the very nature of being a SI bank means they will have a global presence. As a result, it is unsurprising to see meaningful exposure to our foreign currency factor. For local equity, negative exposure suggests that SI banks tend to benefit from periods where international stocks outperform the U.S. This highlights their international bias when considered as a group.
   
• Negative exposure to the small-cap factor: It’s an intuitive conclusion that SI banks tend to be larger market-cap companies. After all, a colloquial interpretation of being “systemically important” is “too big to fail.”
   
• Negative exposure to low-risk and quality factors: Factor analysis suggests that SI banks are not exactly safe havens in the global universe. This means that these large financial institutions have tended to look and feel more like high-risk or low-quality stocks.
   
• Negative exposure to the interest rates factor: SI banks have tended to move opposite to a long bond exposure. Put another way, typically when rates rise you would expect SI banks to benefit from higher rates. However, it is important to note that sensitivity to rates comes with many considerations, especially with rapidly rising rates or inverted yield curves (see Silvergate and Silicon Valley Bank). With that being said, in more status-quo market environments, higher interest rates may mean more income collected on loans and earned on investment securities.⁴ 

From a contribution to risk perspective (Exhibit 2), it is again unsurprising to see that a basket of bank stocks has had more than 60% of its risk explained by our Equity Factor. However, it is notable that this 100% long equity basket did have meaningful drivers of risk from other factors. For example, Value drove almost 14% of risk historically. Foreign Currency and Local Equity together drove more than 5%, and Low Risk and Quality more than 7%.

Exhibit 2: Factor Contributions to Risk of Systemically Important Banks

Source: Bloomberg, Venn by Two Sigma. Period from 7/14/2015–4/10/2023. Chart shows Venn analysis being conducted on KBW Nasdaq Global Bank Index.

When we look at rolling three-year factor exposures, we noticed quality exposure has been deteriorating and becoming even more negative. Over the most recent three-year period, SI banks are the lowest quality they have been going back to available data since 2015.

Exhibit 3: Negative Quality Exposure of Systemically Important Banks

Source: Bloomberg, Venn by Two Sigma. Period from 7/14/2015–4/10/2023. Chart shows Venn analysis being conducted on KBW Nasdaq Global Bank Index.

This lower quality exposure has been a significant drag on return historically. In Exhibit 4, we show that over the most recent three-year period, negative quality exposure has contributed an annualized -5.32% in performance. 

Exhibit 4: Quality Contribution to Return of Systemically Important Banks

Source: Bloomberg, Venn by Two Sigma. Period from 7/14/2015–4/10/2023. Chart shows Venn analysis being conducted on KBW Nasdaq Global Bank Index.

High equity and negative quality exposure may translate into vulnerability
More generally speaking, the quality factor has tended to benefit during severe market turmoil when investors may prefer profitable companies with strong balance sheets (put another way, a typical “flight to quality” environment).

There are several data points supporting this. For example, while our quality factor is long/short and constructed to be beta-neutral, it has exhibited a -0.18 correlation to our equity factor going back to 1995. Additionally, among notable historical periods where our equity factor has been down more than -10%: The Global Financial Crisis, Covid Crash, Oil Price Melt-Up, European Sovereign Debt Crisis and the Oil Price Shock of 2015, our quality factor has managed positive returns, sometimes in the double digits.

Exhibit 5: Equity and Quality Performance During Notable Historical Periods

Source: Venn by Two Sigma. 

As we observed earlier, SI banks have exhibited a high exposure to our equity factor in addition to negative quality. This combination may suggest that typical SI bank positioning is vulnerable to market downturns. In exhibit 6, we compare the historical drawdowns of SI banks (orange) relative to global stocks (green) and find that this has indeed been the case. 

Exhibit 6: Historical Drawdowns of Systemically Important Banks Relative to Global Stocks

Source: Bloomberg, Venn by Two Sigma. Period from 7/14/2015–4/10/2023. Chart shows Venn analysis being conducted on KBW Nasdaq Global Bank Index as well as the MSCI ACWI Index representing global stocks.

Shifting gears to the difference between U.S. regional and systemically important bank
Regional banks typically are smaller in AUM than SI banks and service clients in their region, limiting their global connectedness. In some instances, such as stress testing requirements, regional banks operate with less stringent regulations as well.⁵ Thus far in 2023, outright bank failures have been concentrated in U.S. regional banks. 

Using Venn, we can conduct the same analysis we showed for SI banks, except now doing it for U.S. regionals relative to SI banks as our benchmark. This helps us to define the differences between being a U.S. regional and SI bank through systematic risks. 

Exhibit 7: Active Factor Exposure of U.S. Regional vs Systemically Important Banks

Source: Bloomberg, Venn by Two Sigma. Period from 7/14/2015–4/10/2023. Chart shows Venn analysis being conducted on KBW Nasdaq Regional Bank Index relative to the KBW Nasdaq Global Bank Index.

Positive/negative exposure to local equity/foreign currency factors: While it may be intuitive that U.S. regional banks are long the U.S. dollar and exhibit a U.S. home bias relative to SI banks, it’s an important consideration nonetheless. For example, all else equal, as the U.S. dollar rises you would expect U.S. regional banks to outperform SI banks. In a similar light, all else equal, if U.S. stocks outperform international stocks, you would again expect U.S. regional banks to outperform.

Positive exposure to both value and small caps: Relative to SI banks, U.S. regional banks have tended to exhibit a value and small cap tilt. It is worth noting that quality is more lightly shaded, signifying in this case that it contributed less than 1% to the total active risk between the two. With that being said, from an exposure perspective it was statistically significant and suggests that U.S. regional banks have exhibited higher quality than SI banks. 

Similar to earlier contributions to risk analysis, we can analyse how each of these exposures contributed to the active risk of U.S. regionals vs SI banks. As shown in exhibit 8, local equity and foreign currency together contributed over 23% to active risk, while value and small cap together contributed to over 16%. This suggests that while U.S. regionals may be smaller and more value-oriented than SI banks, the fact that they are domestically rather than globally focused has been a larger driver of relative risk.

Exhibit 8: Active Factor Contributions to Risk of U.S. Regional vs Systemically Important Banks

Source: Bloomberg, Venn by Two Sigma. Period from 7/14/2015–4/10/2023. Chart shows Venn analysis being conducted on KBW Nasdaq Regional Bank Index relative to the KBW Nasdaq Global Bank Index. Roughly 57% of the active risk was defined as residual, indicating that the majority of active risk could not be defined by Venn’s 18 systematic risk factors. 

Using factor analysis to better understand the risk of banks
If banks continue to make headlines, we believe factor analysis may provide an important historical context of their risk exposure. 

Of particular note, using Venn, we found that SI banks have high equity and negative quality exposure, which may indicate vulnerability to market downturns. In the case of what makes U.S. regional banks different, domestic orientation may be obvious, but we find it could be a primary consideration nonetheless. This notion can come into play if bank failures continue to be U.S- focused and our local equity factor suffers heavy losses. 

References

¹ References to systemically important banks refer to the KBW Nasdaq Global Bank Index which allocates to banks classified as “Globally Systemically Important” by the Financial Stability Board (FSB) and Basel Committee on Banking Supervision. This index is equally weighted.

² References to U.S. regional banks refer to the KBW Nasdaq Regional Banking Index. This index is modified market-cap weighted.

³ Source: Bloomberg as of 4/7/2023. Measured using the iShares Russell 1000 Value and iShares Russell 1000 Growth ETFs as proxies. Measured by the diversified bank category as part of the GICS sub-industry classification.

⁴ https://www.stlouisfed.org/on-the-economy/2022/nov/rising-interest-rates-bring-opportunities-risks-banks

⁵ https://www.nytimes.com/article/svb-silicon-valley-bank-explainer.html

Read the original article here.