A flawed process generated by a flawed structure
Published by the R Street Institute.
Testimony to the Subcommittee on Oversight and Investigations
U.S. House Committee on Financial Services
Madam Chairman, Ranking Member Green and members of the Subcommittee, thank you for the opportunity to be here today. I am Alex Pollock, a senior fellow at the R Street Institute, and these are my personal views. I spent 35 years in banking, including 12 years as president and CEO of the Federal Home Loan Bank of Chicago, and then 11 years as a fellow of the American Enterprise Institute, before joining R Street last year. I have both experienced and studied numerous financial crises and financial cycles, including the political contributions to their creation and the political reactions afterward, and my work includes the issues of banking systems, central banking, risk and uncertainty in finance, housing finance and government-sponsored credit, and extensive study of financial history.
To begin, let me compliment the committee staff for their detailed, specific paper on the FSOC’s non-bank designation process. The paper embodies a very good analytical idea: it “compares the FSOC’s evaluation memoranda [of various companies] against one another to measure the consistency of the FSOC’s analysis.” This comparison, as documented in the paper, results in the conclusions that the treatment of different companies is not consistent, that FSOC did not follow its own formal guidance, and in summary, that the evaluations upon which companies either were or were not designated as systemically risky (as “SIFIs”) “have been characterized by multiple inconsistencies and anomalies on key issues.”
The paper says that “These examples cast doubt on the fairness of the FSOC’s designation process.” They do, but in my opinion, the more important point than fairness, is that the observations cast doubt on the objectivity of the FSOC’s work. Were these evaluations impartial analyses looking for disinterested conclusions, or were they rationalizations for conclusions already reached in political fashion?
As we all know, U.S. District Judge Rosemary Collyer, in her decision on the lawsuit MetLife brought against FSOC, found for MetLife and ruled that FSOC’s action was “arbitrary and capricious.” I want to focus on one of the reasons stressed by the judge: the assumptions FSOC made to arrive at its proposed designation.
Considering hypothetical losses resulting from MetLife, Judge Collyer’s Opinion pointedly observes that: “FSOC assumed that any such losses would affect the market in a manner that ‘would be sufficiently severe to inflict significant damage on the broader economy.’ …These kinds of assumptions pervade the analysis; every possible effect of MetLife’s imminent insolvency was summarily deemed grave enough to damage the economy.” [italics mine]
But the judge continued: “FSOC never projected what the losses would be, which financial institutions would have to actively manage their balance sheets, or how the market would destabilize as a result.” [original italics]
Further, “FSOC was content…to stop short of projecting what could actually happen if MetLife were to suffer material financial distress.” FSOC’s work appears pretty pathetic in this light, doesn’t it? FSOC “hardly adhered to any standard when it came to assessing MetLife’s threat to U.S. financial stability,” the judge found.
This sound and sensible judicial decision was appealed by the previous administration. I believe the current Treasury Department should immediately request the Department of Justice to withdraw the appeal, and that Justice should do so as soon as possible.
Recall that the point of designation of insurance companies as SIFIs is to give significant regulatory jurisdiction over them to the Federal Reserve Board, an institution with little or no experience in insurance regulation and which certainly cannot be considered expert in it. The Independent Member of FSOC Having Insurance Expertise, Roy Woodall, who indubitably is a true expert in the insurance business and its regulation, voted against the SIFI designation of MetLife. Coming again to FSOC’s assumptions, he objected: “The analysis relies on implausible, contrived scenarios” [my italics], which moreover, include “failures to appreciate fundamental aspects of insurance and annuity products.”
Mr. Woodall continued that “the central foundation for this designation” is the assumption of “a sudden and unforeseen insolvency of unprecedented scale [and] of unexplained causation.” He reasonably added, “I simply cannot agree with such a premise.” Can anybody?
Voting against the earlier designation of Prudential Financial as a SIFI, Mr. Woodall similarly pointed out that “Key aspects of [FSOC’s] analysis are not supported by the record or actual experience,” that it presumes “an unfathomable and inexplicable simultaneous insolvency and liquidation of all insurance companies” among its “misplaced assumptions.”
Ed DeMarco, a distinguished financial regulator who was at the time the acting director of the Federal Housing Finance Agency and thus the conservator of Fannie Mae and Freddie Mac, joined the dissent on Prudential and also observed the lack of evidence presented in the FSOC’s evaluation. FSOC proceeded “despite the acknowledgment that no institution has as disproportionally large exposure to Prudential”; it “does not fully take account of the stability of Prudential’s liabilities”; it assumes that “withdrawals at Prudential could lead to runs at other insurance companies without providing supporting evidence.” Once again, FSOC was operating on assumptions.
Of course, Messrs. Woodall and DeMarco were in the minority. But did the majority address their serious and substantial objections? Was there a meaningful, substantive exchange among FSOC members about the conceptual issues and the relevant evidence, as would be appropriate, before voting the proposal in? I am told that there was not.
Why not? The whole point of the existence of FSOC is supposed to be the combined substantive deliberation and development of insights by this committee of the heads of financial regulatory agencies. But it doesn’t seem to happen. So the designation process does not work well not only at the staff level, but also at the level of the FSOC as a corporate body.
I directly asked one former senior FSOC insider from the previous administration if the meetings of FSOC members had ever provided a new insight into financial issues. After thinking a moment, he gave me a candid answer: “No.”
Why is this? The Milken Institute, in a recent paper, proposed idealistically that although FSOC is currently nothing like this, “policy makers should convert the FSOC into a truly cooperative working group of regulators focused on risks.” To anyone familiar with the ways of Washington, this will seem an unlikely outcome.
The underlying problem, it seems to me, is the structure of FSOC itself. The shortcomings of the designation process reflect the underlying problems with the fundamental design. To begin with, FSOC is primarily a group of individuals each representing a regulatory agency, with turf to protect from intrusions by the others, and a regulatory record to defend from criticism, as principal bureaucratic concerns.
It is a big group, with 15 official members, but in addition, they all bring along helpers and allies. At the FSOC meeting of Dec. 18, 2014—which approved the MetLife SIFI designation—there were, according to its minutes, 46 people present. It’s pretty hard, indeed impossible, to imagine a real, open, give-and-take and “truly cooperative” discussion with 46 people.
Moreover, FSOC is chaired by the secretary of the Treasury, a necessarily very political, powerful senior government actor with major partisan and institutional interests always in play. No company can be taken up for systemic risk study by the SIFI staff without the approval of the secretary. Does this suggest a disinterested analytical process?
The Federal Reserve is a special case in the structural design of FSOC, because it stands to expand its power every time FSOC makes a SIFI designation. Does the Federal Reserve like power? Would it like to acquire a big new jurisdiction? Of course, and it is a party at interest in every SIFI discussion. I think it is not unreasonable to suggest that, given the Fed’s major conflict of interest, it should recuse itself from any SIFI votes.
With this context, it is easier to see why the FSOC’s SIFI evaluations had to rely on big assumptions and tended to make inconsistent analyses of different companies. It was because the decisions being made were inherently judgmental, with inherently subjective elements, made amid competing interests—that is to say, unavoidably political.
The shortcomings of the FSOC evaluations appear at least consistent with the theory that the evaluations were meant to rationalize decisions already made. Where might the pressure for such decisions have come from?
One publicly debated possibility is that commitments were already made in the setting of the international Financial Stability Board, in which two of the FSOC members, it is sometimes suspected, made deals with foreign central bankers and regulators about which companies were “global systemically important insurers.” There is dispute about whether the FSB discussions were really agreements, and whether they were thought to be binding. But there is no dispute that the international discussions and the naming of “Global SIIs” preceded the Prudential and MetLife designations of the FSOC. Roy Woodall reflected: “While the FSB’s action should have no influence, I have come to be concerned that the international and domestic processes may not be entirely separate.” A related question is whether the Treasury and Federal Reserve FSOC members felt personally committed by their international discussions. If they did, it seems that they should have disclosed that and recused themselves from the FSOC decisions. Did they feel committed to follow the FSB? Only they know.
During their research for the study of the FSOC designations process, the committee staff asked the FSOC’s executive director, Patrick Pinschmidt, what “significant damage on the broader economy” meant, in their assessments. Mr. Pinschmidt replied: “It’s up to each voting member of the council to decide for him or herself what constitutes a significant threshold.” That sounds like depending on subjective judgments to me.
I agree that it is a naturally good idea for financial regulatory agencies to get together and share information, ideas and experiences (to the extent that they will really share). But what is a committee of heads of regulatory agencies, who are acting as individuals and not even on behalf of the relevant boards or commissions, doing making political decisions? If Congress wants to have the Federal Reserve Board regulate big insurance companies, it can make it so in statute, using whatever subjective judgments it wants. In my view, FSOC is a distinctly inappropriate body to act as a little legislature.
The staff paper of FSOC’s evaluations of possible SIFIs, those recommended for designation and those not, details the inconsistencies in treatment. But these differences pale beside the huge discrepancy of those companies chosen for evaluation and those companies not evaluated at all, because the previous Treasury Secretary did not approve their being studied. The FSOC staff did not even analyze them, because of some higher, prior, political judgment. I think this could fairly be characterized as desperately wanting to “see no evil” when it comes to the systemic financial risk of some entities.
The most egregious cases are Fannie Mae and Freddie Mac, which are obviously systemically important and without question systemically very risky. To document that is simple, starting with their combined $5 trillion in credit risk, virtually zero capital and ubiquitous interconnectedness throughout the country and world. Two of the biggest causes of systemic risk are leveraged real estate and the moral hazard created by the government—Fannie and Freddie are both of these combined and to the max.
The Dodd-Frank Act gives a key assignment to FSOC: “To promote market discipline, by eliminating expectations on the part of shareholders, creditors, and counterparties that the Government will shield them from losses in the event of failure.” Fannie and Freddie are pure cases of the government shielding creditors and counterparties from losses, not only as a hypothetical, but as a vast fact. They operate entirely on the government’s credit. They represent the very essence of the problem that FSOC was supposedly created to address. But FSOC doesn’t even study them—instead, the staff was ordered not to study them.
That is an inconsistency raised to the nth power – in my view, a bankruptcy of FSOC’s intellectual credibility as run by the previous administration.
A recent article claims that “the next financial crisis that rocks America…will be driven by pension funds that cannot pay what they promised.” Whether or not it triggers the next crisis, there is no doubt that this is a looming huge risk.
In the very center of this risk is an insurance company absent from FSOC’s evaluation as a SIFI: the Pension Benefit Guaranty Corp. The PBGC is not only on the hook as guarantor of unpayable pensions nationwide, but is already insolvent itself with, according to its own books, a deficit net worth of $76 billion. Might PBGC represent a systemic risk? Yes. Do the creditors of the PBGC think “the Government will shield them from losses”? Yes. Does the FSOC staff evaluate the PBGC? Nope.
In sum, it appears that the flawed process of FSOC’s SIFI designations is generated by the flawed structure of FSOC itself.
In my opinion, structural reform of FSOC is needed as part of larger the Dodd-Frank reform legislation. But here are a few recommendations for improvements which could be implemented by the new administration in the short run:
FSOC should have regular meetings of principals only with substantive discussions of major issues and explorations of disagreements. No helpers, no staff.
The secretary of the Treasury should immediately instruct the FSOC staff to undertake systemic risk evaluations of Fannie Mae and Freddie Mac.
The secretary of the Treasury should immediately instruct the FSOC staff to undertake a systemic risk evaluation of the Pension Benefit Guarantee Corporation.
The Treasury Department should immediately request the Department of Justice to withdraw the government’s appeal in the MetLife v. FSOC suit and the Department of Justice should immediately do so.
FSOC staff should be encouraged to come up with new ideas on evolving risks for discussion among the FSOC principals.
Any SIFI evaluation should strictly follow the rules and guidance approved by FSOC, with analysis performed in a strictly consistent manner.
Assumptions about macro reactions and assumptions of implausible and contrived scenarios should be clearly identified as judgments and guesses.
International discussions and actual decisions of FSOC should be kept strictly separate. Any international agreements, even if informal, made by FSOC members, should be fully disclosed.
Again, my appreciation to the committee staff for their productive study of the inconsistent FSOC designation process and the very important issues it raises.
And thank you very much for the chance to share these views.