ICI Viewpoints

If regulators are going to impose strict rules and heavy burdens on a business, should they have to demonstrate that those rules and burdens address an actual and probable risk? For most regulations, that’s the law. But in the long-running debate over asset management and financial stability, we’ve often noted regulators’ assertions that funds or their managers should face new strictures based solely on the potential consequences of some bad event—regardless of whether that event is likely, improbable, or even impossible.

So it was fascinating to see how U.S. District Court Judge Rosemary Collyer zeroed in on the failure of the U.S. Financial Stability Oversight Council (FSOC) to conduct a realistic risk analysis in its decision to designate MetLife Inc. as a systemically important financial institution, or SIFI. In recent oral arguments on MetLife’s legal challenge to that decision, Judge Collyer pressed the FSOC’s attorney to defend the Council’s presumption that MetLife would be “at the brink of collapse” in a financial crisis. “That’s not risk analysis,” the judge said, according to press accounts. “That’s assuming the worst of the worst of the worst.”

But it’s not the first time that stability regulators, pursuing their “prudential” mission, have ignored evidence on the chances that adverse events might occur and instead assumed “the worst of the worst of the worst.”

• The Financial Stability Board (FSB), in its latest consultation on methodologies for designating funds or their managers, states that it has no interest in the probability that any of its hypothetical risks could occur.
• Economists at the Federal Reserve Bank of New York, speculating recently about risks in bond funds, assert: “We are agnostic about the likelihood of [the assumed] scenario actually materializing.” In another exercise, the Fed’s economists plugged hyperbolic assumptions—positing 50 percent redemptions from all high-yield bond funds instantaneously—into their equations in attempts to produce systemic events.

As ICI President and CEO Paul Schott Stevens remarked at the Boston University/ICI Conference on Financial Stability and Asset Management last March: “Apparently, if you can imagine that a systemic event might happen, then you must assume that it will happen—and regulate accordingly.”

But that’s not how regulation is supposed to work—in the United States or abroad. In a friend-of-the-court brief filed in the MetLife case, four legal scholars, including professors Jonathan R. Macey of Yale and Tamar Frankel of Boston University, point out that:

Every accepted form of risk regulation requires an assessment of not only the consequences of a possible contingency, but also its likelihood. There is much more to risk regulation than simply assuming that everything that can go wrong will go wrong and, simultaneously, treating that worst-case scenario as the baseline for regulation. Rather, an essential part of risk regulation is an objective assessment of which risks to regulate, based on empirical evidence and not just on the limits of the pessimist’s imagination. The mere fact that a risk is hypothetically conceivable is not enough. [Emphases in the original.]

Or, as Stevens wrote in a letter to Treasury Secretary Jacob Lew, Federal Reserve Chair Janet Yellen, and Securities and Exchange Commission Chair Mary Jo White:

U.S. regulators are required to examine relevant data and articulate a satisfactory basis for their actions, including a rational connection between the facts found and regulatory choices made. Mere conclusory statements and unsupported conjectures do not suffice, nor may agencies simply ignore contradictory evidence in the record before them. They may not impose substantial new burdens on regulated entities to guard against illusory or wholly improbable risks.

Judge Collyer appears to understand that—and her questions to the FSOC will make sure that others, including Congress, will take note as well. Financial stability regulation could be in for a bracing shot of reality.