Following another round of Federal Reserve stress testing in which banks withstood a financial doomsday scenario
At the heart of the discourse is a fundamental disagreement about the ultimate goal of minimum capital requirements. On the one hand, requiring banks to hold capital ensures their solvency and keeps credit available through the business cycle. On the other hand, more capital retention means less money can be lent out, thus limiting the benefits that such lending can provide to the economy.
“That's the question: What's the right level of safety versus support for the economy?” said Randal Quarles, former Fed vice chair for supervision.
This is an age-old conundrum, but one that was largely hypothetical in the early days of the 2008 financial crisis. When the economy — and specifically the banking system — was reeling from the collapse of the mortgage market, the utility of higher capital requirements was apparent, and the Dodd-Frank Act made enhanced capital requirements less of an option for regulators or banks. But several factors have made this debate more relevant today than at any time since that crisis.
One of those factors is the anticipated confirmation of Michael Barr as the Biden administration’s pick for vice chair for supervision at the Fed — the central bank’s point person on precisely these kinds of supervisory questions. Barr is expected to be confirmed by the Senate this summer, and he will assume the role with the final implementation steps of the Basel III global standards looming next year and in the context of a slowing U.S. economy and a bare-knuckle fight against inflation. While it might seem like this would be a good time for banks to hold more capital, Quarles said instead this should be a time when banks should be freer to lend.
“The folks who prioritize safety say the current environment is particularly the time when we need to worry about capital and make sure that banks are resilient,” Quarles said. “I think that is getting the balance wrong, and I think that raising capital levels now would be a contributing factor as we move towards a recession.”
Despite a more rigorous hypothetical stress scenario than last year, each of the 33 banks examined retained far more than their minimum capital requirements in this year’s stress-test results.
With a new regulatory regime coming to the Fed and heightened concern that the stress-capital framework could soon be tested in a real-life scenario — one that is not met with the same fiscal and monetary interventions of the COVID-19 era — both sides are making their cases.
Industry advocates argue the stress-test results released last week show banks are sufficiently prepared for a crisis and should not be required to retain any additional capital. In total, the 33 banks tested maintained a Tier 1 capital ratio of 9.7% — more than double the 4.5% minimum — despite losing more than $600 billion in the intentionally severe scenario.
Yet, the aggregate capital ratio for the banks fell slightly more this year than under 2021’s less harsh scenario, so some institutions will see their stress-capital buffers increased. Bank of America will likely see its buffer increase by 1 percentage point, while Citigroup and JPMorgan Chase are not far behind at nine-tenths of a point each.
Francisco Covas, executive vice president and head of research at the Bank Policy Institute, said these increases are not necessary to protect financial stability and could instead put it in greater jeopardy. By requiring banks to retain more capital now, as economic activity slows down, the Fed risks keeping capital parked on banks’ balance sheets when it should be encouraging them to put it to work, he said.
Covas added that increasing capital requirements makes funding costs for banks more expensive. That added cost gets passed on to borrowers, he said, causing more loan activity to shift away from banks and toward unregulated lenders.
“Further increases in capital requirements incentivize even more the shifting of lending out of the banking sector to nonbanks because bank capital is more costly than deposits, so loan rates must increase and demand for bank loans falls,” Covas said. “Banks will continue to adjust their balance sheets in response to increases in capital requirements, but the level of risk in the banking system is already very low."
Sean Campbell, chief economist and head of research for the Financial Services Forum, said capital requirements have not become a binding constraint on banks, but he said the Fed has generally structured them in such a way that ratios increase, at least to some extent, every year.
“In doing that, one can envision creating a set of stress tests that are so severe and, in some senses, implausible, that the requirements become so onerous that banks become less able to lend and other sectors of the economy — other nonbank lenders, whether they be finance companies or insurance companies or hedge funds that don't face such stringent capital requirements — end up being more significant organs of credit for the economy,” Campbell said. “And I think that that is a tension that is real and significant and deserves greater attention from the public."
Meanwhile, consumer advocacy groups and some Democrats in Congress look at this year’s results more skeptically. They say the fact that even the most severely impacted banks — Huntington Bancshares and Citizens Financial Group, which saw their ratios fall to 6.8% and 6.9%, respectively — did not come close to the minimum ratio is evidence that the test was not strenuous enough. In their view, banks should be retaining significantly more capital.
In a statement issued after this year’s stress-test results were released, Sen. Sherrod Brown, D-Ohio, chair of the Senate Banking Committee, said it was “past time” for tougher stress tests and higher capital requirements.
“Wall Street bank CEOs have raised the alarm bells that an economic hurricane is coming, but the reality is that the largest banks aren’t doing what they need to do to protect the economy from the next crisis. Instead of building up capital to withstand losses or investing in the real economy and workers, they’re planning to spend $80 billion in stock buybacks and dividends,” Brown said.
Phillip Basil, director of banking policy at the consumer advocacy group Better Markets, decried the “plain vanilla” scenarios in this year’s test. While the scenario was, in fact, more severe than 2021, it consisted of variables that banks have dealt with in the past, such as plummeting home prices, stock market sell-offs and commercial real estate value declines.
“Despite this being a more, at least on paper, stressful scenario, it's a very bread-and-butter scenario that the banks are well positioned to handle when it comes to the stress test,” he said.
By focusing on replicating past shocks, Basil said, the Fed’s stress tests are not looking out for current weaknesses or new issues that could be on the horizon. Because of this, he said, capital requirements are inherently lacking.
“The test results feed into the capital requirements, they are the basis of capital requirements, so if the test isn't stressful enough, then necessarily capital requirements are too low,” he said.
Basil blames efforts undertaken by Quarles for weakening the stress test. He said a shift toward a uniform set of criteria and transparency around testing models have made it easier for banks to “game” the test.
Quarles said his transparency measures made it easier for banks to know what was expected of them and to construct their loan portfolios accordingly.
“The notion that they're gaming the system because of more transparency around the models, because they will keep capital holds that our model suggests are safe — that’s not manipulation, that’s how government works,” Quarles said. “That's not gaming the system. That's complying with our rules.”
Yet, Quarles said, the criticism about the scenarios being “plain vanilla” is valid. He said he pushed for the inclusion of a more varied set of testing criteria, one that would occasionally mix in extreme and out-of-the-ordinary shocks. Doing so, he noted, would require test results to be averaged over a multiyear period when setting capital requirements to avoid unduly punishing banks for failing to prepare for specific, highly unlikely occurrences.
The 2018 test, Quarles said, included a scenario in which the yield curve moved in an unsuspected way during a period of distress. He said the inclusion yielded some interesting and useful data, but also proved to be especially difficult for some banks to cope with. Had that outcome been included in the stress-capital calculations, it would have changed the capital requirements for those banks dramatically, which would not be prudent, Quarles said.
“I’m in favor of the notion that we should run some less vanilla tests, but you can't really do that until you've solved this issue about the potentially large volatility of required capital,” he said.
Whether it be the addition of new testing criteria or a ratcheting up of the severity of current scenarios, Covas said an attempt by the Fed to design a stress test with the goal of pushing more banks to their minimum capital ratios would not be fruitful.
"To say the stress test is not stressful enough and to make it more stressful effectively makes banks hold a higher stress-capital buffer, so the capital levels go up. Then it's even more difficult to have banks reach a 4.5% minimum capital ratio" in a stress scenario, he said. “There is no convergence in this argument because capital requirements would never be high enough. The fact that banks do not end up even close to 4.5% capital ratio under an extreme scenario demonstrates that U.S. banks are already subject to very high capital requirements.”
Campbell said while the desires of an ever-stricter capital requirement regime are understandable, the end result of such a pursuit is an industry that is too burdened to function.
"It's like trying to design a safety system for a car so that in the event of some really bad and horrific crash, everybody survived,” he said. “Imagine you do that by encasing the car with three-foot-thick concrete blocks. That will work, but that will make it impossible for anybody to ever get in the car and drive to the grocery store."
