When Banks Fail, Why Do We Keep Bailing Out Uninsured Depositors?

Michael Ohlrogge has a theory about the Federal Deposit Insurance Corporation, the agency that makes sure ordinary depositors don’t lose their money when their bank goes bust.

Ohlrogge, an associate professor at New York University Law School, argues that when banks fail, the F.D.I.C. is not resolving them in the manner that is least costly to its Deposit Insurance Fund.

If he’s right, then the F.D.I.C. is going against the explicit instructions of Congress, so this is kind of a big deal. My impression is that F.D.I.C. staffers believe they are complying with Congress’s instructions, so this is a case of two sides looking at the same facts and drawing very different conclusions.

The issue Ohlrogge raises is how the F.D.I.C. handles uninsured deposits. These are deposits that exceed the F.D.I.C. insurance limit of $250,000 per depositor, per bank, per account ownership category. It stands to reason that the cheapest way to resolve a bank failure in many cases — maybe most cases — would be to tell those uninsured depositors that their money is gone: “Sorry, see ya, wouldn’t wanna be ya.”

But in the vast majority of bank failures, the F.D.I.C. approves a resolution in which the uninsured depositors don’t lose a penny. They are treated exactly as well as insured depositors. This is typically done by finding another bank that is willing to buy the entire failed bank, which entails purchasing all of its assets (such as loans) and assuming all of its liabilities, including both insured and uninsured deposits.

The Deposit Insurance Fund usually has to pay out when it arranges for a bank to be taken over because the bidder demands to be compensated for taking responsibility for an institution whose liabilities exceed its assets.

There are times when whole-bank takeovers are the cheapest solutions for the Deposit Insurance Fund — say, if the failed bank has so much name recognition, customer loyalty and so on that the acquiring bank will pay the F.D.I.C. a lot to take it whole out of the agency’s receivership.

But it’s hard to imagine that this would be the case more than 90 percent of the time.

Now look at these side-by-side charts. The first thing you notice is that they seem almost identical, which you wouldn’t expect if the F.D.I.C. is doing its job right.

The chart on the left shows the percentage of bank failures in which uninsured depositors suffered no losses. The chart on the right shows the average cost to the Deposit Insurance Fund of resolutions of failed banks, stated as a percentage of the bank’s assets. The bars on the two charts go up and down in lock step, which is circumstantial evidence that when uninsured depositors are made whole (left chart), resolutions will be more costly (right chart).

Notice that uninsured depositors did take more of a hit in the period from 1992 to 2007. That’s because in 1991, Congress passed the F.D.I.C. Improvement Act, which imposed the “least cost” standard on the agency. In following Congress’s instructions, the F.D.I.C. cut back on deals that rescued uninsured depositors.

But then came the financial crisis. The F.D.I.C. started selecting bids that protected uninsured depositors even more than it had done before the 1991 act, and it has continued to do so years after the crisis ended.

The F.D.I.C. can get permission to ignore the least-cost rule for any particular bank if there’s a “systemic risk” that imposing losses on its uninsured depositors will cause a panic that makes more banks fail and ends up costing the Deposit Insurance Fund more money in the long run. In coordination with other agencies, Treasury Secretary Janet Yellen invoked the systemic risk exception for the F.D.I.C.’s resolutions last year of Silicon Valley Bank and Signature Bank, both of which were quite large.

But the government did not even invoke the systemic-risk exception on Friday when it approved a takeover on Friday of Republic First Bank of Philadelphia that protected all uninsured deposits.

The Deposit Insurance Fund is financed by assessments on banks, so when it loses money, banks have to pay more into it, and they pass along their higher costs to their various stakeholders: depositors, shareholders, borrowers. Another bad result is that zombie banks stay in operation longer than they should because uninsured depositors happily supply them with funds, knowing the F.D.I.C. has their back.

Ohlrogge speculates that the F.D.I.C. is experiencing “mission creep,” taking on a responsibility for uninsured depositors that it was never assigned. He has been elaborating and pressure-testing his theories for several years in scholarly presentations, including a working paper in November, “Why Have Uninsured Depositors Become De Facto Insured?” In that paper he estimated that the F.D.I.C.’s practices have added at least $45 billion to the cost of bank resolutions over the past 15 years.

In an interview, he told me he can’t prove beyond the shadow of a doubt that the F.D.I.C. is breaking the least-cost rule, and one reason is that the agency doesn’t divulge how it evaluates different options it has. The F.D.I.C. reveals its evaluation standards to the Government Accountability Office but not to bidders, for fear that they could use that information to game the system.

At times the F.D.I.C. itself has made statements that raise questions about its practices. In a footnote of a 2001 document posted on its website that’s intended to provide guidance to other nations’ regulators, it says that making uninsured depositors whole can be least costly “in rare cases.” (Which means: not more than 90 percent of the time.) Similarly, a 2015 journal article by F.D.I.C. staff members, referring to full bank liquidations in the period before the financial crisis, said “a least-cost resolution almost always includes imposing losses on uninsured depositors.” Another F.D.I.C. document says that whole-bank takeovers in which the agency promised to share losses with buyers became the dominant option in the crisis year of 2009, “and often the only one offered to potential acquirers.” It doesn’t say what happened after 2009.

I tried to get F.D.I.C. officials to respond on the record to Ohlrogge’s accusations but they declined. So I have to go by what the agency has stated in the past about its resolution practices. The agency says that when a bank fails, the agency solicits outside evaluations of the market value of the bank’s assets to gain a sense of how much money could be raised in a full liquidation. Using that as a base line, it solicits bids, some of which involve all the deposits and others just the insured ones, and determines how attractive the bids are. It says it adheres strictly to the least-cost standard. And it says its practices are vetted regularly by others, including the Government Accountability Office. Ohlrogge says audits by the G.A.O. and the F.D.I.C.’s own inspector general hardly ever assess compliance with the least-cost test.

Giving the F.D.I.C. the benefit of the doubt, I suppose it’s possible that its statements about its practices from before and during the financial crisis, which Ohlrogge cites, don’t apply to the present. It’s possible that acquirers’ preferences have changed such that they really do want whole banks, not just assets, and will happily cover uninsured depositors to get that. Here’s an example from last year in which every bidder sought to buy the whole bank. (That might have happened last week with the Philadelphia bank.)

On the other hand, Ohlrogge says it could be that bidders don’t bother with partial bids because those usually get rejected. And the charts I showed above don’t lie. If acquiring banks really did pay a premium to acquire whole banks, along with their uninsured depositors, you wouldn’t expect the cost of resolving banks to be so much higher now than when acquirers were less likely to make the uninsured depositors whole.

I asked Ohlrogge why he has put so much effort into this one issue. He said he’s worried about banks that take big gambles using uninsured deposits. “I am really concerned about financial stability,” he said. “I lived through the 2008 financial crisis. I saw the damage it did to people and communities.” He added: “Is this one issue going to completely solve all our financial stability issues? Certainly not. But it’s a small, manageable chunk.”

Elsewhere: One Reason Spending Has Stayed Strong

Americans are saving a smaller share of their disposable income than usual — 3.2 percent in March, compared with an average of 5.7 percent from January 2000 through this February. That has freed up more money for consumption, which in turn has kept the economy aloft. (Aside: The saving rate spiked during the pandemic because people were getting stimulus payments and couldn’t get out to spend the money.)

Gains in the stock market since the pandemic have made people feel as if the stock market is doing their saving for them, lessening their perceived need to save out of their disposable (i.e., after-tax) income. That’s fine as long as it lasts, but it suggests that the economy is vulnerable to a downturn on Wall Street. On Tuesday, the Conference Board announced that its Consumer Confidence Index dipped in April. One possible reason: Stocks retreated during the month.

Quote of the Day

“Women won some of their most important workplace rights in the 1960s because of a set of fortuitous events. They continued to win in the early 1970s because of a movement that gave them influence. They won yet more because groups that were supportive of their cause — college graduates, single women, Black women — expanded relative to others. They won when they had the political clout to get men, especially those in Congress and the White House, to see that women’s rights were as valid as civil rights. Yet, women’s rights had setbacks when, in light of many gains, women abandoned the movement. Women’s rights has had a truly ‘strange career.’”

— Claudia Goldin, the winner of the 2023 Nobel in economics, “Why Women Won” working paper (2023)