This regulation would apply to 37 banks as of the end of 2014, according to FDIC data. The largest number of deposit accounts at a single bank at that time was 84.5 million; the 10 banks with the largest number of accounts had a total of 319 million.
That is a lot of accounts to aggregate by depositor, all on the extremely slight chance that the FDIC will close and liquidate a large bank the next day and either pay off or transfer the insured deposits to another bank. Unsurprisingly, the FDIC has not estimated what it will cost banks to fully implement this regulation.
The FDIC first imposed this aggregation requirement in 2008. But it later concluded that it “has not been as effective as had been hoped in enhancing the capacity to make prompt deposit insurance determinations.” Now it’s attempting to make the initiative more workable.
This will be a difficult task, since the FDIC’s reasoning on this requirement bears no relationship to reality. Since the IndyMac failure in July 2008, the FDIC has rarely imposed losses on uninsured deposits, particularly at large, failed banks. Instead, the FDIC has used a “least cost resolution” test to justify protecting uninsured deposits against any loss.
Under this test, if the FDIC’s projected loss in a failed bank can be reduced by fully protecting uninsured deposits against any loss, the bank’s entire deposit franchise — including uninsured deposits — can then be sold to one or several banks.
Selling all of a failed bank’s deposits and related customer relationships usually reduces the FDIC’s loss in closing the bank, which leads to lower deposit insurance premiums for healthy banks. Fully protecting uninsured deposits against loss when a large bank is closed also helps to calm financial markets.
Moreover, declining to impose losses on uninsured deposits does not compromise the objective of getting rid of too‐big‐to‐fail banks. The failed bank still disappears, even if another bank acquires all of its deposits and customers.
Of the 30 failed banks since IndyMac where uninsured deposits did suffer a loss, only five had deposits exceeding $1 billion. The largest had $1.7 billion of deposits, including approximately $20 million in uninsured deposits, and less than 21,000 deposit accounts.
In sharp contrast, in 55 failures since IndyMac of banks with total deposits over $1 billion, uninsured deposits were fully protected against any loss. The largest failure, Colonial Bank, had $20 billion of deposits, including an estimated $4.4 billion of uninsured deposits. It had 790,000 deposit accounts, well below the two million account threshold the FDIC has proposed.
Interestingly, since IndyMac, uninsured deposits were protected against any loss in 94% of the 446 failed banks with less than $1 billion in deposits. Put another way, since 2008, seldom have uninsured deposits suffered any loss in banks of any size that have failed.
Two much larger bank resolutions that incurred no loss for the FDIC — Washington Mutual and Wachovia — also caused no loss to uninsured deposits. The larger of Washington Mutual’s two banks had 42 million accounts and an estimated $45 billion in uninsured deposits when it was acquired in 2008 by JPMorgan Chase.
Wachovia’s largest bank, meanwhile, had an estimated 29 million deposit accounts and $130 billion of uninsured deposits, including $23 billion of foreign‐office deposits, when Wells Fargo took it over in 2008. None of Wachovia’s depositors with uninsured deposits lost a dime!
Given this history, it is easy to see that the FDIC will not, and should not, try to impose insolvency losses on uninsured deposits in a large failed bank. Nor will the FDIC attempt to do so with even most not‐so‐large failed banks. Therefore the FDIC should admit as much and kill the requirement. It is a failed proposition.