Fixing What’s Broken: The GSIB Surcharge – Near- and Long-Term Problems

22 Feb 2021
Read Time 6 mins


Over the course of the pandemic, large U.S. banks have been actively supporting the U.S. economy by extending credit quickly into the economy and intermediating financial markets. At the same time, these financial institutions, known as U.S. Global Systemically Important Banks (GSIBs), have seen a tremendous increase in a key capital metric, the GSIB score. The increase is due in large part to a surge in deposits by consumers and businesses and could result in an unnecessary and significant capital increase that would impair the banks’ abilities to support the economy. The recent increase in GSIB scores is a clear sign that there are both near-term and long-term problems with the GSIB score methodology that should be addressed to ensure that capital requirements reflect economic reality, do not disincentive economic growth, and do not distort the international competitiveness of large U.S. banks. In this post, we describe the troubling trend in GSIB scores and discuss why the Federal Reserve should revisit the GSIB score methodology in the near-term. 

GSIB Scores and Surcharges

The GSIB surcharge is an additional capital buffer that only applies to GSIBs in the U.S.– the eight members of the Financial Services Forum. The rationale for the GSIB surcharge is that large, internationally active banks should hold higher levels of capital to account for the systemic risk they pose. The GSIB surcharge depends on an underlying regulatory metric – the GSIB score – that is a complicated function of an array of balance sheet variables that relate to the following broad categories: size, complexity, interconnectedness, substitutability, and short-term wholesale funding. The GSIB score is then translated into a GSIB surcharge. A higher GSIB score results in a higher GSIB surcharge. In general, anything that increases a GSIB’s balance sheet mechanically increases its GSIB score.    

The costs of inappropriately high capital requirements are real. A standard, mainstream finding in economics is that as capital requirements rise, a bank’s ability to lend and support the economy is impaired. Moreover, one additional factor in the U.S. is that U.S. regulators set the GSIB surcharge using a methodology that is more punitive than the one applied in other countries. As a result, U.S. GSIBs are disadvantaged when competing with their international peers. 

The Near-Term Problem

As we discussed in a post this summer, GSIB scores have increased tremendously during the pandemic. And the trend has only continued as GSIBs have continued to grow their balance sheets to support the U.S. economy. As of the third quarter of 2020 (the latest period for which GSIB score data are available), the aggregate GSIB score of all Forum members has increased by 229 points, the largest three-quarter increase in GSIB scores on record. Without any adjustment to the GSIB score methodology, this increase will result in a substantial increase in required capital for a number of Forum members.

Given this large increase in GSIB scores, it is worth asking – Is this rise a result of an increase in systemic risk? The answer is an unequivocal, No. Rather, GSIB balance sheets have grown in direct response to increased demand for deposits and, importantly, as a result of Federal Reserve policy that has injected a slew of cash into the economy, increasing its balance sheet from roughly $4 trillion to $7 trillion over the course of 2020. Specifically, because banks are the recipients of the cash injected into the economy when the Federal Reserve increases its balance sheet, large bank balance sheets have no choice but to grow as well.

So, what have Forum members done with all the cash that has been injected by the Federal Reserve? Have they taken actions that would arguably increase their systemic risk? Not at all.  Rather, as shown in Figure 1 below, in 2020 Forum members have responded by increasing their cash levels by roughly $860 billion (a 74% increase in 2020) while also purchasing roughly $350 billion in U.S. Treasury securities (a 42% increase in 2020). Both of these actions clearly reduce, rather than increase, risk.  As a result, an increase in GSIB scores and surcharges runs counter to the actual changes in the systemic risk of Forum members.

Source: FR Y-9C

This perverse result is a consequence of the fact that regulators never designed the GSIB score with economic conditions like the pandemic in mind. Accordingly, the GSIB score methodology should not be blindly applied without regard to the unusual and unanticipated circumstances created by the pandemic. 

Some argue that this increase in GSIB scores is not concerning because the resulting increase in the GSIB surcharge operates with a one-year lag. This is a red herring. A key tenet of economics is that people change their behavior immediately in response to changing information and conditions. Accordingly, once GSIB scores rise, bank analysts and investors will demand higher rates of return on money invested with GSIBs immediately. The lag is simply not economically relevant.

The Long-Term Problem

The problems that we are observing with the GSIB surcharge methodology today are also a broader sign of deeper problems with the GSIB surcharge that have been well known for years. In particular, the GSIB surcharge has not kept pace with overall developments in the economy and bank regulation since it was finalized in 2015. More specifically, as we have noted in various posts as well as formal comment letters, there are now a large number of effective polices that have reduced systemic risk in the banking system that are not recognized by the GSIB surcharge methodology. These developments include:

  • the introduction of liquidity requirements – notably the Net Stable Funding Ratio (NSFR) that was finalized in late 2020 – and is designed to reduce the systemic risk of U.S. GSIBs
  • the introduction of total loss absorbing capacity, or “TLAC” requirements, which are designed to reduce the systemic risk of U.S. GSIBs
  • the introduction of resolution planning requirements, which improve large bank resolvability and likewise reduce the systemic risks of U.S. GSIBs

Additionally, over the years a number of technical issues with the GSIB score, ranging from its adjustment for underlying economic growth, to its treatment of derivatives, to its measurement of short-term wholesale funding levels, have been identified. Taken together, the broad structural factors outlined above and various important technical considerations suggest that the Federal Reserve should revisit the GSIB surcharge in the near-term. Finally, the idea of revisiting the GSIB surcharge is not controversial. In 2015, when the Federal Reserve finalized the GSIB rule, it indicated that the central bank would consider revising the GSIB surcharge over time as changes in the economy warranted re-calibration. As the final rule states:

The Board acknowledges that over time, a bank holding company’s method 2 score may be affected by economic growth that does not represent an increase in systemic risk. To ensure changes in economic growth do not unduly affect firms’ systemic risk scores, the Board will periodically review the coefficients and make adjustments as appropriate.

The time to revise the GSIB methodology has clearly come.   

Getting the construction and calibration of the GSIB surcharge right is crucial to ensuring that large U.S. banks can continue to support the economy and effectively compete on the international stage. The specific size of any adjustment to the GSIB surcharge requires careful consideration, though the evidence suggests that the required adjustment is large. According to a previous post on this issue, recognition of changes to the regulatory environment alone would suggest nearly a 50 percent reduction in the size of the GSIB surcharge.  


Forum members have been working steadfastly to support the economy during this pandemic. As the Federal Reserve has expanded its balance sheet to support the economy, the GSIB scores of Forum members have been mechanically increased in a manner that does not reflect systemic risk. This result is a consequence of a rigid regulation in an economic environment that was never considered by regulators when designing the GSIB surcharge. Moreover, the concerning development we are seeing with GSIB scores today is also a sign of more fundamental problems with the GSIB surcharge methodology that have been known for years. The Federal Reserve should take this opportunity to reconsider the construction and calibration of the GSIB surcharge in the near-term to ensure that large bank regulation is appropriate and that large banks can continue to productively support our economy and compete globally.           

Forum Updates