Federal Deposit
Insurance Corporation

From: Bert Ely [mailto:bert@ely-co.com]
Sent: Monday, March 13, 2006 4:54 PM
To: Comments
Subject: Large-Bank Deposit Insurance Determination Proposal - RIN 3064-AC98

To all concerned:

<>This comment letter on the FDIC’s large-bank deposit-insurance determination modernization proposal is offered as a response to a comment letter on this ANPR submitted on February 7, 2005 [sic] by Mr. Gary Stern, president of the Federal Reserve Bank of Minneapolis. Mr. Stern’s letter was posted as comment #3 on the FDIC website section devoted to comments on this ANPR. Mr. Stern’s comment letter made five points. Underlined below are Mr. Stern’s points. Following each of them is my comment on that point. <>

The FDIC must reform current insurance determination procedures which hinder its ability to carry out the least-cost resolution of a large bank.

<>This point of Mr. Stern’s should be rejected for two reasons. First, as is the case throughout Stern’s letter, he does not define what a large bank is. At least the FDIC took a crack at that by defining a large bank as one with over 250,000 deposit accounts and more than $2 billion in domestic deposits. Of course, like any definition of “large,” the FDIC’s definition is arbitrary. The arbitrariness of the FDIC’s definition is highlighted by the fact that the very largest bank, as of June 30, 2005, had 187 times as many deposit accounts as the smallest large bank. The FDIC’s size range for a large bank illustrates a problem that crops up in other areas of deposit insurance – where to draw the line between “large” and “not large” banks. <>

Second, Stern wants to “ensure effective use of society’s resources,” yet fails to note that the banking reforms of the early 1990s – FDICIA in 1991 and the authorization of interstate banking and branching in 1994 – have greatly strengthened America’s larger banks. He proposes that all large banks be treated as if they were likely to fail, yet to do so would waste “society’s resources” as that is not where the bank failure problem lies today. Stern’s letter takes absolutely no cognizance of the fact that the 70 bank and thrift failures since 1994 have occurred among the “not large” banks – not a single one of those failures would have qualified as a “large bank.” If the total assets of those 70 failures had been combined into one bank, that bank would be about the size of the 100th largest bank today, a further demonstration that banking failures since 1994 have not been a large-bank problem. <>

Large banks are not where the failure problem is, and for good reason: One, a large bank, even one as small as $2 billion in domestic deposits and with 250,000 deposit accounts, has much greater internal risk dispersion than a bank with $200 million of deposits and 25,000 deposit accounts, and therefore is much less likely to be brought down by risky lending or fraud than the smaller bank. Two, almost all large banks are publicly owned, with widely dispersed ownership which permits effective stockholder oversight and discipline. Stockholder discipline, reinforced by public disclosure of FDICIA’s prompt-corrective-action capital measures, has forced the sale, merger, recapitalization, and/or management turnaround of large, troubled banks. Stern’s comment letter fails to acknowledge the powerful role stock-market discipline plays today in preventing the failure of large banks, even the smallest large banks.

Further, Stern does not acknowledge that large banks don’t fail overnight or only on short notice – that occurs only in small banks, and only when a massive fraud, relative to the size of the bank, has occurred. In fact, large banks, as well as most small banks, fail after at least a year, if not several years, of internal problems that are readily evident to banking supervisors, including the FDIC. While the ANPR states that “generally, the FDIC has at least 30 days advanced warning to plan and prepare for failures,” in fact, as numerous inspector-general reports on costly bank failures have documented, the FDIC has know for many months, if not several years, that a bank was going down. That advance knowledge certainly gives the FDIC more than enough time to comfortably assemble the deposit-account data it needs to make adequate deposit-insurance payout determinations. So that it not get caught by a sudden failure, perhaps the FDIC board of directors should direct that FDIC supervisory personal begin the necessary deposit-account linking as soon as a bank or thrift is classified, under the present deposit-insurance pricing scheme, as a C-1, B-2, C-2, A-3, B-3, or C-3 institution – these are the FDIC’s six highest-premium-rate deposit-insurance categories. Interestingly, and not surprisingly, no large banks, as the ANPR defines them, fall into any of those categories – only “not large” banks.

The FDIC’s Board of Directors should focus on net benefits when evaluating the comments received on the ANPR and choosing which option to implement.

Mr. Stern’s second point is one the FDIC’s Board of Directors should focus on – the net benefits of the option it proposes to implement. The sounder, more defensible net benefit calculation is one based on explicit cost calculations: Net benefit equals savings to the FDIC (and hence to the banking industry since it, not the taxpayer, is liable, as a practical matter, for the FDIC’s deposit-insurance losses) minus the cost to the banking industry of implementing whichever option is chosen. Interestingly, neither the FDIC’s ANPR nor Stern’s letter quantifies any explicit benefit to the FDIC – reduction in deposit-insurance losses – yet it is readily acknowledge that this proposal will impose a cost on the banking industry. As Stern notes, “this benefit is difficult to quantify but the limited evidence available suggests that it is potentially large.” Not surprisingly, Stern offers absolutely no evidence about these “potentially large” benefits, but charitably notes that the cost of implementing this proposal “should certainly be kept to a minimum.” Given that this proposal offers no explicit benefit, any cost above zero represents a waste of “society’s resources.”

<>Stern tries, but fails, to sidestep the lack of an explicit economic benefit by offering an implicit argument that this modernizing process would “instill equity in the resolution process” and provide a “signaling effect.” In effect, Stern claims one of the FDIC’s proposed “modernization” options should be imposed on large banks, even though none of them have failed since 1994, because FDICIA’s prompt-corrective-action provisions actually have worked in the resolution of failed small banks. In fact, there has been equity in the resolution process – uninsured creditors in failed banks have been treated as Congress intended. Not unsurprisingly, though, only “not large” banks have been failing. Unfortunately, uninsured creditors have suffered losses where regulators failed to detect internal fraud (witness 2002’s failed Oakwood Deposit Bank in Ohio) or were slow to close an obviously insolvent bank, such as Superior Bank in Chicago, which was closed in July 2001, more than five months after I publicly warned that it was insolvent.

Stern’s “signaling effect” argument fails to acknowledge the substantial marketplace signaling which takes place today, under the present system for determining the application of deposit-insurance limits. As noted above, large publicly held banks and thrifts are subject to stock-market discipline, which is market discipline of the highest order. Although less timely than stock-market discipline, credit-market discipline, in the form of credit ratings and bond and debenture yields, add to the pressure on large banks to sell, merge, recapitalize, or hire turnaround management long before they become insolvent. The burden is on Stern to argue how one of the options proposed in the ANPR would improve the effectiveness of market discipline by an amount at least equal to the cost the selected option would imposed on all large banks.

<>The features of Option 2 are necessary but may not prove sufficient to correct weaknesses in the insurance determination process.

Option 2 presumably is the least costly of the three options proposed in the ANPR. As he did earlier in his comment letter, and in his other writings, Mr. Stern sets up a straw man – “where a bank’s failure is rapid” – on which to base his case for at least going with option 2. In fact, as banking history has shown time and time again, even during “the turbulence of the 1980s and early 1990s,” to use Stern’s phrasing, almost no bank or thrift failure was “rapid.” As is well known, I was able, using only relatively limited Call Report data, to flag hundreds of banks and thrifts in the 1980s and early 1990s that were doomed to fail years before they were finally closed. My later work as an expert witness in failed-thrift litigation enabled me to review supervisory documents in numerous failure situations – in every case, supervisors knew of the deteriorating condition of the institution long before it was closed, and certainly in sufficient time to begin doing the necessary account-linking to accurately apply the deposit insurance limit when closure finally occurred.

<>The FDIC should give serious consideration to implementing Options 1 and 3.

Since neither the FDIC nor Mr. Stern have made the case for the least odious option, number 2, no case can be made for the more expensive, and more intrusive, options 1 and 3. Further, as I point out in another comment letter, the more ambitious account-linking proposed in these options opens the door to a massive, unnecessary invasion of financial privacy and violations of civil liberties.

<>

A reformed insurance determination regime should apply to all large banks for which the current regime could prevent a least cost resolution; the same insurance determination scheme need not apply to all covered institutions.

Perhaps to avoid the ire of “not large” banks, Mr. Stern proposes limiting this “modernization” scheme to just those banks where it is not needed. To reiterate a point made above, the simplest, most cost effective, and fairest way to address the deposit-insurance determination issue is for the FDIC board of directors to do what it should have been done years ago – once a bank’s condition has deteriorated to a certain point, the bank (or thrift) would have to begin, under the supervision of the FDIC’s Resolutions and Receivership personnel, to commence account-linking activities in order to accurately apply deposit-insurance limits should the bank or thrift be closed. This crossover point could be, as proposed above, when an institution moved into one of the six most-costly deposit-insurance premium categories. Based on December 31, 2005, FDIC data, this requirement would have applied to 60 banks and thrifts (less than 1% of the industry’s institutions) holding approximately $6 billion of deposits, less than .1% of the industry’s domestic deposits. Alternatively, the account-linking procedure could be initiated at a bank as soon as it is placed on the “problem bank” list. At the end of 2005, there were 52 institutions on that list with total assets of $6.6 billion.

At such time as account-linking would be ordered for a troubled bank, the FDIC would become morally obligated to discourage new, uninsured deposits in the bank and to not permit the bank to resume accepting new uninsured deposits until such time as it restored itself to financial health, it had been acquired by a stronger institution, or it had been resolved as a failed bank. Whatever liquidity problems a bar on new uninsured deposits would create (which the FDIC could readily offset by advancing funds to the bank) pale in significance to the losses which depositors would unfairly suffer if they were permitted to place uninsured funds in a bank after the time when regulators knew the bank had a reasonable likelihood of failing in the near term.

__________

Please contact me is there is any aspect of my comments on Mr. Stern’s comment letter that you would like to discuss.

As is the case with my other comments letters on this ANPR, I am submitting this letter on behalf of my personal interest in this issue and not on behalf of any client.<>

Bert Ely

Ely & Company, Inc.
Alexandria, Virginia
703-836-4101
<>Bert@ely-co.com<>