It is the rare wonk who follows the twists and turns of financial market reform (a.k.a. “finreg”). It’s gnarly stuff, but man, is it ever important. The last two economic expansions ended because an investment or financial bubble imploded (dot.com, real estate), and seven years later, we’re still recovering from a housing bubble born of “innovative” finance, underpriced risk, and excessive borrowing.
In other words, we ignore finreg developments at our peril, which is why we must consider a recent legal decision that favored the insurer MetLife over financial regulators. The judge’s decision was misguided in a couple of key ways that pose potential threats to our economic future.
The case grows out of the Dodd-Frank law’s designation of certain financial firms as SIFIs (contrary to the way some folks on Capitol Hill pronounce it, that’s “siffy’s,” not “Sci-Fi’s”—these firms aren’t selling derivatives to space aliens, though I’m sure they’d welcome the business). A SIFI is a “systemically important financial institution,” one that’s too big to fail. Should it falter, especially amidst broader financial market instability, it would pose a cascading risk to the financial system and thus to the broader economy. Think Lehman Brothers and AIG and you’ll get the gist.
The challenge for finreg is to identify a SIFI well before its failure is wreaking havoc on global markets. Dodd-Frank sets out guidelines for regulators to do so, and once a financial firm is identified as a SIFI, it must lower its leverage ratio (debt/equity) by increasing the equity it holds against losses, like defaults on its debt holdings. Firms typically hate such rules because leverage amplifies their profitability. SIFIs must also have a workable plan (a “living will”) for their recovery when faced with threateningly large losses or even resolution in cases where they must be wound down to protect the broader system.
As you can imagine, no one wants to be a SIFI.
MetLife was so displeased to get that label that it sued the government, arguing that the regulators did not abide by their own interpretation of the SIFI-designation protocols in Dodd-Frank. The judge agreed, and last week, she released the opinion underlying her rationale.
I’m no lawyer—I don’t even play one on TV—but as I read her opinion, I was struck by the gulf she creates between what regulators need to do to protect the system and what, according to the judge, the legal system requires. Part of this gulf is her creation in ways that I suspect will not withstand the planned appeal by the Treasury Department. But part of it is more nuanced and gets to some unique aspects of contemporary financial risk.
For example, consider the concept of “tail risk,” which means a very small probability of a very bad outcome. The judge faults the regulators for not computing this probability, arguing that they failed to assess MetLife’s “vulnerability to financial distress.” This sounds reasonable, and the regulators could likely cook up such a probability—based on this decision, they may well have to do so in future SIFI designations—but it is likely to be tiny and imprecise. As Treasury Secretary Jack Lew correctly noted in response to the court, “the failure of AIG or Lehman Brothers would have been considered highly unlikely before the financial crisis.” Yet their interconnectedness was integral to the crash and ensuing deep recession.
Moreover, suppose the regulators were to quantify MetLife’s vulnerability to severe distress or failure (insolvency) and came up with the number X. What would anyone make of X? To argue that X is too small to warrant SIFI designation is to fundamentally misunderstand tail risk and the whole point of this part of finreg.
Two factors (at least) make an institution too big to fail, i.e., systemically risky. First, their interconnectedness to the rest of the system is such that their failure would have large ripple effects. Second, their risk management, leverage ratios, liquidity access, and regulatory scrutiny are insufficiently calibrated to prevent failure. In this spirit, the regulators argued correctly that their job was to assess “whether, and how, the company’s vulnerabilities, in a distress situation, could impact the broader financial system—not to assess whether distress could occur.”
This is the right standard by which to protect the rest of us, and the one I understand to be consistent with Dodd-Frank. However, the law is still young and it’s neither surprising nor devastating for a court to find some gray area. I don’t believe a faux-precise probability is warranted given the nature of tail risk. But neither do I view this part of the decision, even if it stands up on appeal, as a fundamental challenge to finreg.
On the other hand, the judge went well beyond Dodd-Frank by faulting the regulators for not conducting a cost-benefit analysis of their designation of MetLife as a SIFI. Such a requirement is simply not in the law, so, with the caveat that I know squat about case law in this space, those who do tell me that this part of the judge’s decision is highly unlikely to withstand appeal.
Practically speaking, cost-benefit analysis in this context would also be a highly unrealistic endeavor. While I can imagine coming up with a plausible estimate of the costs to MetLife of SIFI designation, e.g., the hit to profits from maintaining a larger capital buffer, the central point of finreg is that the downside of failing SIFIs is potentially recession. One could try to back out p*costs, where p is the probability of failure and costs are losses in recession, but I doubt the former of these factors (p) can be meaningfully calculated, and I’m certain that the latter—the costs of recession—cannot be.
Why not? For starters, the trillions of dollars in lost output during a recession are a function of the length and depth of the downturn, which are wholly unknowable a priori. But there are also incalculable long-term impacts of recession, as those who start careers in down economies can be permanently disadvantaged, as can be those who lose their homes or their wealth or who experience long-term unemployment.
So, when the court argues that it was “arbitrary and capricious” of the regulators to assume “the upside benefits of designation…but not the downside costs of its decision,” it is both going beyond requirements in Dodd-Frank and asking for an incalculable analysis.
What we need to know in the case of a SIFI designation is whether financial firms are large, interconnected, leveraged up, and underregulated. If so, no one’s going to shut them down, break them up, or impinge on their business models beyond the enforcement of a set of reasonable standards designed to protect the broader economy in which SIFI’s function.
For years, regulators consistently erred on the side of doing too little to the point where they snoozed at vital switches. Now that they’re finally trying to do the right thing, we really don’t need “arbitrary and capricious” judgments to undermine them.