Reconsidering the regulatory use of stress testing

The use of annual stress tests to set minimum capital requirements for large banks was one of the most important innovations in prudential regulation following the global financial crisis of 2007-2009. The use of stress tests for this purpose provided greater sensitivity to risk at the level of an individual company and a better assessment of the banking system as a whole than pre-crisis methods of setting fixed minimum capital requirements or the use of internal bank models. However, several years later, the robust stress testing system that justified its use in setting capital requirements has been weakened and is now a more routinized and predictable process. With the Federal Reserve’s current stress tests ending in June, Harvard law professor Dan Tarullo, who oversaw oversight and regulation as Federal Reserve governor from 2009 to 2017, is seeking to spark a debate about whether stress tests should remain a core business. large banks minimum capital requirements in the US

Advantages of the current approach include: (1) stress tests are generally more risk sensitive than retrospective, standardized risk weights; (2) unlike capital requirements derived solely from each bank’s balance sheet, stress testing provides a more complete picture of a bank’s resilience by forecasting the impact of an adverse scenario on the entire industry; (3) stress tests can reveal the degree of correlation between asset classes to a greater extent than traditional approaches to setting minimum capital; (4) the use of stress tests as a basis for setting minimum capital fits well with other reforms introduced after the global financial crisis; (5) Stress tests provide information to regulators, banks and markets. However, Tarullo points to several flaws in the way the Fed used stress tests, including the way it constructed adverse scenarios, the failure to include sales and financing stress in the stress test models, and the reliance on the book value of bank assets as adjusted under generally accepted accounting principles. In short, the use of stress tests to establish minimum capital requires greater adaptability and greater supervisory freedom. However, bank lobbying and institutional factors within the Fed significantly limited the dynamics of stress tests.

One attractive option for changing the status quo would be for the Fed to conduct stress tests in a much more dynamic manner than it currently does. The supervisory model would be regularly updated to reflect modeling progress and regularly adjusted to take into account changes in the composition of banks’ balance sheets and system-wide holdings of major asset classes. There would be many different scenarios each year, some of which would extrapolate contemporary risks (e.g. a large interest rate increase for the 2022 stress test) and others which would reflect larger and unpredictable shocks (e.g. a pandemic that shut down large parts of the economy) . Tarullo concludes that such an approach is unlikely to be adopted and, if it were, it would be difficult to sustain.

A more realistic option would be to separate the annual stress test required under the Dodd-Frank Act from the setting of regular, ongoing capital requirements for large banks. For larger banks, the fixed capital requirements set out in Fed regulations would be increased, which would then be their binding constraint. The results of stress tests could be used both to introduce changes to these regulations and to possibly issue directives requiring certain banks to increase capital ratios due to significant risks specific to them. The advantages of this approach include: (1) establishing higher capital requirements at a given point in time would protect against the decline in resilience resulting from loss of momentum in stress tests; (2) it should encourage greater dynamics in stress tests as it would free them from institutional and lobbying pressures aimed at diluting them; (3) it would remove many institutional barriers to conducting an ad hoc, individualized stress test in the early stages of a significant shock to the economy or financial system (e.g., a pandemic); (4) the information generated would be useful to regulators in setting minimum capital levels – both generally applicable standards and bank-specific requirements; (5) decoupling would significantly reduce the uncertainty in capital planning that banks and investors face when the annual stress test changes capital requirements from year to year. Disadvantages include: (1) capital requirements will likely lose some risk sensitivity; (2) if the Fed relied too heavily on the capital directive process, the same inertial tendencies that affected the stress tests could reemerge; (3) separating stress tests from regular capital requirements would likely reduce the effort currently expended by the Fed and banks to conduct them; (4) there could be confusion about the meaning of stress test results.

Tarullo concludes that separating the stress test from ongoing, regular capital requirements would be an admission that the Fed cannot maintain an approach to capital regulation that is forward-looking, dynamic, and macroprudential. He says that if he had confidence that even a modestly more dynamic version of the status quo would be implemented and maintained, he would lean toward it. However, based on stress testing practice over the last five years, it has concluded – with considerable disappointment – ​​that decoupling is currently the best option.

Read the full article here.