The Evolving Role of G-SIBs: Is Big Beautiful Again?

For many years, in the aftermath of the global financial crisis (GFC), a stigma was attached to extremely large, global banks. They were perceived by many, including regulators, as institutions that should be feared, closely watched and, when needed, penalized – particularly if they fell under the global systemically important banks (G-SIBs) designation.

But times have changed, and G-SIBs are no longer seen as the big, bad wolves of the financial services industry. Indeed, a recent 10-year assessment of G-SIBs by the Basel Committee on Banking Supervision (BCBS) showed that G-SIBs have declined a bit in systemic importance and demonstrated a narrowing of the gap between large and ultra-large global banks. However, that does not necessarily mean that big is once again beautiful.

Marco Folpmers

The BCBS continues to endorse prudential plans to increase capital requirements for the largest banks in the world. Moreover, in addition to needing more capital, G-SIBs also face higher supervisory expectations for risk management (covering risk management functions, risk data aggregation, risk governance and internal controls) and stiffer requirements for group-wide resolution planning.

Perhaps not surprisingly, some large, global banks are now suspected of engaging in behaviors intended to lower their G-SIB score and keep their capital add-ons in check. This, of course, is problematic, because it could potentially lead to more systemic risk.

We’ll take a closer look at the tough G-SIB requirements, as well as the changing behavior of G-SIBs and the BCBS’ response, in a minute. But we first need to understand how G-SIBs are defined and how we’ve reached this stage.

G-SIBs: Definition and Evolution

Every year, large, global banks are evaluated for G-SIB designation based on their scores five different types of indicators: (1) cross-jurisdictional activity; (2) size; (3) interconnectedness; (4) the lack of readily available substitutes; and (5) complexity. Simply put, if a bank’s total score is 130 or more, across the five indicators, it is designated as a G-SIB.

The BCBS has monitored this dashboard of indicators, which demonstrates the size and complexity of these ultra-large banks, since 2014. G-SIBs fall into five buckets, with a total score of 130 to 229 (bucket 1) at the low end and 530-629 (bucket 5) at the high end. The higher the score, the greater the potential impact the delinquency or failure of a bank could have on the global financial system.

The G-SIB measurement system itself and the necessity of closer monitoring of systemic banks originated after the GFC of 2008, when the failure of a number of large banks sent shockwaves into the financial system and the real economy.

While the BCBS created the definition of G-SIBs and oversees the measurement system, the Financial Stability Board (FSB) is responsible for the publication of the annual list of G-SIBs. (As of November 2023, there were 29 G-SIBs.)

Over the past decade, ultra-big banks have become even larger, according to the BCBS’ 10-year assessment of G-SIBs. G-SIB scores have, however, declined slightly since 2021, because of a decrease in interconnectedness and complexity. What’s more, interestingly, non-G-SIBs have become more complex and interconnected, so the distinction between G-SIBs and non-G-SIBs has become more blurred.

Are the decreasing scores of G-SIBs over the past three years the result of natural evolution or is there a more nefarious reason?

G-SIB Behavior: Gaming the System?

G-SIBs across all five buckets now face high capital surcharges – aka “add-ons.” The largest banks (bucket 5) are hit with the biggest increase – a 3.5% capital requirements add-on. In short, this means that each G-SIB in bucket 5 needs to hold extra capital that is equivalent to 3.5% of its risk-weighted assets.

Figure 1: G-SIB Buckets, with Banks and Required Capital Add-ons

Source: FSB. The bars represent banks in different buckets; the diamonds represent capital add-ons.

As depicted in Figure 1, there is a big gap between number of bucket-1 G-SIBs and the larger banks in buckets 4 and 5. It is also clear, moreover, that the distinction between G-SIBs and the largest non-G-SIBs is becoming blurred, and that G-SIBs therefore now carry less weight. But this is not the whole story.

In March, around the same time the BCBS released its 10-year assessment of G-SIBs, it also published the findings of a study about the behavioral aspects of G-SIBs. This study indicates that the measurement and monitoring system of G-SIBs is at risk, because some banks are engaging in suspicious “window-dressing” behaviors that could reduce their G-SIB indicator values.

Specifically, the BCBS report quite convincingly shows that G-SIBs tend to decrease their complexity around the end of each calendar year, when their indicator scores are calculated. They do this by reducing their exposures to OTC derivatives (and sometimes, also, to repos) just before the G-SIB measurement date. In the days immediately after, they seem to be rebuilding their positions, and, consequently, their OTC exposure curves show remarkable “V-shapes” around the year-end.

Ironically, the steep G-SIB capital add-ons depicted in Figure 1 may actually motivate large banks to investigate how they can end up in a lower bucket, or even avoid G-SIB classification altogether, rather than protect the banking industry.

To eliminate or at least reduce the window-dressing behavior, the BCBS has proposed changes that would require large, global banks participating in G-SIB assessments to “report and disclose most G-SIB indicators based on an average of values over the reporting year, rather than year-end values.” Comments on this proposal, which is still in the evaluation stage, were due on June 7.

Parting Thoughts

G-SIBs are no longer an easily identifiable group of banks: the profiles of G-SIBs and non-G-SIBs have converged somewhat over past decade. Complicating matters further, broad risk trends, such as pandemics and geopolitical risk, have impacted the broader banking landscape broadly, as opposed to only G-SIBs.

What’s more, some large-bank concentration is now desirable, because that allows for the development of building up more sensitive risk management systems and for more effective supervision, per €1 billion of exposure.

Times have certainly changed since the BCBS successfully (and rightfully) implemented G-SIB framework in response to the GFC. We’re not yet back at “big is beautiful,” but the recognition that systemic risk factors can affect the entire banking sector as a whole has at least decreased the focus on the size of banks. Indeed, the desire of regulators to reduce the size of the largest institutions is less prominent today.

Nevertheless, given that the systemic risk is still inherent in these ultra-large banks, a proper measurement and monitoring system is key. In the long run, tactical (or “window dressing”) behavior to avoid the G-SIB designation will not help these banks and could have a significant negative impact on the entire banking industry.

Dr. Marco Folpmers (FRM) is a partner for Financial Risk Management at Deloitte the Netherlands.