Reinstating Board Oversight over Failed-Bank Resolutions

Many are surprised when I tell them that the Board of Directors (the “Board”) did not vote on the FDIC’s sale of First Republic Bank. Instead, FDIC staff selected the winning bidder under authority delegated by the Board. In my view, that was not the right way to handle a significant bank failure, especially given the many significant policy issues raised by the sale, not to mention the $13 billion cost to the Deposit Insurance Fund.

After discussing this issue and potential solutions with the Chairman and other Board members, and as contemplated by our Bylaws, ten days ago I delivered notice asking that we consider at today’s meeting my proposal to require a majority of the Board to vote in favor of approving any future sale of a failed bank with assets of at least $50 billion. I thought this topic was appropriate for this meeting in light of the three other resolution-related matters we were to consider.

Today’s Board discussion on this topic began with staff’s alternative proposal that did provide for a Board vote to approve such failed-bank sales—but only if at least three Board members requested a vote. I then presented my proposal as a substitute to staff’s and placed some emphasis on the only practical difference between the two, namely that my proposal would require each Board member to go on the record “for” or “against” the staff-recommended sale even where there is a silent Board majority already in support. I am unable to support staff’s proposal because it does not go far enough to reinstate the Board’s proper role in overseeing resolutions of failed banks.

By law, “the management of the Corporation shall be vested in a Board of Directors.”1 While the Board delegates authority to FDIC staff to manage the day-to-day affairs of the FDIC, the Board has reserved for itself consideration of any matter that would establish or change policy, involve an issue of first impression, or attract unusual attention or publicity. Consistent with the Board’s reservation of these “major matters” for its consideration, the Board reviewed and approved the sales for the two largest FDIC failed-bank resolutions of the 2008 financial crisis: the $307 billion Washington Mutual Bank in 2008 and the $32 billion IndyMac Bank in 2009.

The Board generally has not approved smaller failed-bank sales. When a smaller bank fails, the Board typically invokes an existing framework for delegations of authority known as the Robinson Resolution, which was adopted in 1999 and provides for staff to be delegated expansive authorities to design and implement an auction process for the failed bank, select the winning bidder, and negotiate the terms of the sale.

Although the Robinson Resolution expressly provides that its scope is limited to failed banks with assets of $1 billion or less, FDIC staff nonetheless recommended that the Board invoke the Robinson Resolution to give staff these same expansive authorities to resolve Silicon Valley Bank, Signature Bank, and First Republic. For Silicon Valley Bank and Signature Bank, a minority of the Board resisted that approach and, thanks to the supermajority vote necessary to make the contemplated system-risk determinations, was able to negotiate a requirement that the Board approve the sales of those banks. That Board minority did not have the same leverage when First Republic failed, as that bank failure did not entail a systemic-risk determination; instead, First Republic was resolved, as staff recommended, under the Robinson Resolution. As a result, the Board did not vote on the sale of First Republic.

Given the size of First Republic and the significant policy considerations associated with its sale, the Board should have taken direct responsibility—consistent with its obligation to manage the affairs of the FDIC—by voting on the sale. Had it done so, the Board would have had to assess whether the FDIC had taken appropriate steps to satisfy the statutory mandate that the proposed sale be “the least costly to the Deposit Insurance Fund of all possible methods” for satisfying the FDIC’s obligations.2 To that end, each Board member should have had to develop an informed view as to whether the FDIC had taken steps to explore a reasonable range of “possible methods.” Each Board member also should have had to develop an informed view on the critical assumptions underlying the FDIC’s least-cost financial analysis that identified the winning bid.

Instead, as it happened, the FDIC designed and implemented an auction process for First Republic without a formal role for the Board. In my view, that auction process did not take sufficient steps to develop “possible methods” under which nonbanks could have acquired First Republic’s assets.3 The auction process also seems to not have considered “possible methods” under which a potential acquirer would have assumed only First Republic’s insured deposits, leaving the uninsured deposits behind to bear losses.4 Had a Board vote on the eventual staff-recommended sale been contemplated, even a minority of the Board might have been able to drive real-time enhancements in the auction process to remediate these gaps.

The FDIC also conducted the least-cost analysis of the bids without a formal role for the Board. Each bid’s estimated cost to the Deposit Insurance Fund was quite sensitive to critical assumptions as to the liquidation value of asset pools that would have been left in the receivership. We should not pretend there is only one “right” set of assumptions; had a Board vote on the eventual staff-recommended sale been contemplated, a more searching inquiry by Board members into the range of reasonable assumptions could have led to a different consensus view as to the least-cost bid, or a view that several bids were in effect equally least costly.

These are, of course, all counterfactuals. Ultimately, it is beside the point whether a Board vote would have led to a different outcome. The FDIC’s exercise of judgment and discretion in running the First Republic auction and selecting the winning bidder entailed significant policy tradeoffs and will have long-term implications. It is the job of each Board member to stake an informed view on consequential decisions of this sort, and hold themselves accountable for explaining their views to the public. We should not try to give ourselves a degree of plausible deniability by delegating the hard decisions to staff.

At today’s Board meeting, I proposed a simple solution. I asked the Board to adopt resolutions to provide that, going forward, in the event of a failure of a bank with at least $50 billion in assets, the Board must approve any transaction that would dispose of all or a substantial part of the assets, the deposits, or the insured deposits of the failed bank.

To address a potential objection, my approach would pose no risk of deadlock that does not already exist today as it would require only a simple majority vote. Even today, a simple majority could stop a proposed sale through a notational vote or special meeting. Notably, staff’s proposal also contemplates that a simple majority could stop a proposed sale.

My approach also would pose no increased risk of delay or use of a bridge bank or other interim solution before a sale is consummated. My proposal would require essentially no new work by staff. Even today, staff typically prepare a memo on the accepted bid that generally should be sufficient to brief the Board in advance of the vote. Notably, staff’s proposal also contemplates briefings for Board members on the auction process and results.

As I see it, the main practical difference between staff’s proposal and mine is that mine would require each Board member to go on the record “for” or “against” a staff-recommended sale even when an otherwise silent Board majority supports the sale. Requiring each Board member to go on the record is important for accountability and transparency. Ideally, requiring each Board member to go on the record also could lead to better outcomes as each Board member would have a strong incentive to actively monitor the auction process, develop an informed view on the staff-recommended sale, and present credible challenges to staff on key questions.

Going forward, today’s other resolution-related proposals would, if finalized, strengthen the need to enhance Board oversight over failed-bank resolutions. The proposed long-term debt requirements would increase the likelihood that a bank failure would result in no loss to the Deposit Insurance Fund. In those circumstances, the FDIC is more likely to find that multiple bids are equally least costly (i.e., there would be no loss to the Deposit Insurance Fund) or the statutory least-cost mandate simply would not apply. In such a situation, the FDIC would have wide latitude to decide when, to whom, and on what terms the assets and deposits of the failed bank are sold. In that world, it is critical the Board play an active role in balancing the many competing interests inherent in deciding the fate of the failed bank.

Although my proposal was not approved, I appreciate the Chairman’s constructive approach to finding common ground. While I do not think staff’s proposal goes far enough, I recognize that he and I agree on the importance of formalizing the Board’s role in overseeing failed-bank resolutions. I would like to thank the Chairman for his time and attention to my concerns, and his inclusion of this topic on today’s meeting agenda. I think the FDIC has been well served by this constructive and ongoing dialogue among the Board members, and I hope dialogue of this sort will continue into the future, both on this issue and the many other matters before the FDIC.

• 1

  12 U.S.C. § 1812(a).

• 2

  Id. § 1823(c)(4)(A)(ii).

• 3

  While some nonbanks were admitted to the auction, the FDIC did not extend to nonbank bidders the same generous financing and loss-share terms that it offered to bank bidders. Denying nonbanks the chance to compete on a level playing field with banks potentially reduced competitive tension in the auction, resulted in lower prices for First Republic’s assets, and increased the ultimate cost to the Deposit Insurance Fund. Perhaps even more importantly, by in effect excluding nonbank bidders, the FDIC might have lost out on a very real opportunity to pair one of the regional-bank bidders that submitted a competitive bid for First Republic’s deposit franchise with a nonbank that would have acquired the remaining assets at a price greater than the FDIC’s estimated liquidation value, which could have resulted in a less costly bid to the Deposit Insurance Fund.

• 4

  The FDIC apparently did not consider this option because no bidder submitted an “insured-deposit-only” bid. That perhaps begs the question as to whether the FDIC took sufficient steps to encourage insured-deposit-only bids.