In the aftermath of the 2008 collapse, financial regulators began "stress tests" for banks. These are meant to see if a bank can handle another economic shock. Basically, the Federal Reserve simulates an economic panic of some kind, and then sees how the banks' business models perform. The latest round of tests concluded last week, and virtually all the big Wall Street players passed with flying colors. The Fed had some reservations about Credit Suisse, but otherwise the 18 biggest banks had enough capital on hand to weather a crisis.
"It's really a good year for the big banks," Adam Gilbert, who leads the financial services advisory arm of PricewaterhouseCoopers, told Bloomberg. "It's a vote of confidence from the Fed."
Can we all breathe a sigh of relief that America's financial system rests on a firm foundation? Unfortunately, maybe not. Over the last few years, policymakers have toned down the "stress" part of the stress tests.
Part of the problem is just the practical challenge of simulating an economic crisis. As Mark Whitehouse pointed out, it's never just the unexpected shock — the burst housing bubble or stock collapse — that's the problem, it's all the cascade effects that have to be accounted for in the model too. It's an open question whether the stress tests were ever robust enough to really test Wall Street's resilience.
But even if they were at first, they might not be now.
The Fed began the tests soon after the Great Recession, as a way to shore up confidence in the U.S. financial system and incentivize the banks to repair their balance sheets. The 2010 Dodd-Frank Act further expanded and entrenched the practice. Then a partial rollback of Dodd-Frank's regulations was passed by Congress and signed into law by President Trump in May 2018. That law raised the threshold of assets a bank had to have before it faced mandatory annual stress tests, from $50 billion to $250 billion — though banks between the $100 billion and $250 billion mark still face periodic tests.
As a result, the number of big banks tested fell from 35 in 2018 to just 18 this year.
The Fed itself also has a fair amount of leeway to design the tests the remaining 18 still face. Randal Quarles is the central bank's point person for financial supervision, appointed by Trump as a temporary placeholder in late 2017 and confirmed by the Senate for the full position in July 2018. He came into the office pushing for several ways financial regulation should ease up on the banks, and under his guidance the Fed has moved in that direction.
A particularly scathing recent piece by Bloomberg's editorial board lays out the changes: First, the Fed has begun providing the banks more information about how the stress tests are designed and the underlying models used. Quarles defended this change as a matter of "democratic accountability and the rights of all U.S. citizens — including the management and shareholders of the institutions that are subject to the stress tests — to understand the requirements to which they are subject." But as critics point out, it obviously raises the risk that the banks can simply game the tests.
Second, the Fed has largely done away with the requirement that the big banks pass the "qualitative assessment." The stress tests come in two parts: The quantitative assessment is the part that tries to mirror an economic crisis and measure how a bank's capital buffer and business model withstands the event through data analysis. The second part is the qualitative assessment, a more subjective analysis of the bank's internal practices and rules by Fed officials: regulators basically decide if they think the bank's institutional culture is handling its duties with enough care and responsibility.
You could see how this sort of subjective judgment by an outside group — with few clear metrics to hit — would annoy banks. But it was also an effective way to keep the banks on their toes. Now, the qualitative assessment will still happen, but it's no longer something the banks have to "pass" or "fail" — thus they won't be penalized if the assessment finds them lacking.
Finally, Quarles is also pushing to get rid of a rule that banks stay above a minimum leverage ratio. That's basically the amount of debt a bank owes relative to its assets, such as debt other people and institutions owe the bank. (Assets are different from the capital that shareholders invest in the bank.) Quarles' objection is that a minimum leverage ratio is too blunt and simple a rule and doesn't account for banks' different circumstances and risk profiles.
Beyond appointing Quarles, Trump's nominated three of the five officials currently sitting on the board, which votes on these sorts of policy changes. Another one of the five, Jerome Powell, was promoted by Trump to Fed chair. At the same time, the dilution of the Fed's stress tests cannot be blamed entirely on the influence of the president and his appointees. That rollback of Dodd-Frank regulations was passed with the help of Senate Democrats and similar changes to Quarles' were proposed by his Obama-appointed predecessors.
This is a familiar pattern in financial systems: The longer an economy goes without a crisis, the more complacent the banking industry and its regulators become. Broad rules that maximize the system's safety come to seem over-burdensome and inadequately nuanced. Stress tests meant to mimic the chaos of an unexpected crisis come to seem capricious, despite the fact that such tests should feel like an imposition by unaccountable forces — because that's what a financial crisis is.
Finally, hiding behind all this is the matter of profits: According to the law, passing the stress tests is a hurdle the banks must clear before they can pay out dividends to shareholders. The easier and less strenuous the tests, the bigger and more certain the payouts are.
The upshot is that it isn't at all clear what will happen to the banks when another real-life version of a stress test finally hits. It's not a comforting thought.