Home > News & Events > Speeches, Testimony & Articles

Speeches, Testimony & Articles






10:00 A.M., JUNE 20, 2001

Mr. Chairman, Senator Gramm and Members of the Committee, thank you for the opportunity to testify on behalf of the Federal Deposit Insurance Corporation (FDIC) regarding the condition of the bank and thrift industries and the deposit insurance funds.

I am pleased to report that the banking and thrift industries continue to exhibit strong financial results. However, we are seeing signs of stress that indicate that this continued strong performance will be more difficult to maintain in the future. I will highlight three of these warning signs in my testimony today-subprime lending, vulnerabilities in the agricultural sector and funding and liquidity challenges.

Perhaps the most important message that I will leave with you today is that there are flaws in the current deposit insurance system and the best time for constructive debate on changes to deposit insurance is now, during a period of financial health for the banking and thrift industries, rather than in the charged atmosphere of a crisis. Today, depository institutions are strong and profitable. The deposit insurance funds also are in good financial condition and the FDIC stands fully prepared to fulfill its commitment to depositors. We should not, however, assume that these good times will last another decade. As you know, depositors in all walks of life have come to rely on FDIC insurance to guarantee that their insured deposits are absolutely safe. The financial strength of the FDIC and its ability and commitment to honor its responsibility to depositors are beyond question. I urge this Committee to take advantage of this most opportune moment and to move forward on reform to ensure that the strength and stability of our deposit insurance system remains unquestioned.


The banking sector continues to experience strong financial performance. Commercial banks recently completed their eighth consecutive year with an industry return on assets above one percent. A return on assets (ROA) of one percent or higher has traditionally been a benchmark of superior earnings performance. Prior to 1993, the commercial banking industry never had an annual ROA as high as one percent. Almost 60 percent of all insured commercial banks reported an ROA of one percent or higher last year.

Three main sources of strength drove bank earnings during this period of prosperity. First, the improvement in asset quality following the last recession has meant that expenses for credit losses have been less of a drain on banks' revenues. Second, noninterest revenues have been growing rapidly, as the industry has diversified its sources of income. Third, banks have had strong growth in assets, particularly in loans, as they have provided necessary credit to a record-breaking economic expansion.

Many indicators of trouble - unprofitable banks, "problem" banks, undercapitalized banks, bank failures -- all remain near their cyclical lows. Banks' capital has kept pace with the industry's growth. Today, more than 95 percent of all banks are in the highest regulatory capital group. The number of "problem" banks -- 78 banks, with $17 billion in assets at the end of last quarter -- is near its cyclical low point.

Our most recent earnings data, which we released earlier this month, show that net income of commercial banks set a new record in the first quarter of 2001. However, this record was made possible by nonrecurring gains on sales of securities. The industry's net operating income, which more closely reflects the strength of banks' ongoing core business, was $565 million below the level of a year earlier.

Sustaining these very high levels of profitability has become increasingly difficult for the banking industry. There is evidence that many banks have taken on more risk as they have sought to maintain profitability. At the same time, some of the most important factors that have contributed to the industry's relative prosperity are becoming less favorable.

Net interest margins-the difference between what banks earn on their loans and other investments and what they pay for deposits and other liabilities-reached a 14-year low in the first quarter. The margin decline stemmed from increased competition, which has put downward pressure on loan pricing and upward pressure on funding costs, and a relatively flat yield curve.

The volume of problem loans has been growing for almost two years, mostly in loans to commercial and industrial (C&I) borrowers at large banks. Only one-third of all banks are showing deterioration in their C&I portfolios, but together they account for more than two-thirds of all C&I loans held by commercial banks. Moreover, most of the deterioration is centered in larger banks, particularly those with large and middle market corporate loan portfolios. This deterioration is reflected in the interagency Shared National Credit review program, which has reported two straight years of significant increases-albeit from a very low base-in classified and criticized credit volumes, a 53 percent increase in 1999 and another 44 percent increase in 2000. The 2001 Shared National Credit review is currently in progress and results will be available later this year, but early indications are that this trend will continue.

Credit-card loans, which the FDIC identified as a potential concern in our 1997 testimony on industry condition, have shown an improved trend in loan losses since 1998. Up until the first quarter of this year, this improvement has paralleled an improving trend in personal bankruptcy filings through the end of 2000. However, personal bankruptcies in the first quarter of this year were 18 percent higher than a year earlier, raising the possibility of higher write-offs of credit-card loans later this year.

As the percentage of troubled loans has risen from cyclical lows, banks have had to apply an increasing share of their revenues to provisioning for loan losses. Last year, loss provisions absorbed 8.2 percent of banks' net operating revenues, the highest proportion since 1992. In the first quarter of this year, loss provisions were 36.1 percent higher than a year ago.

Concentrations of traditionally higher-risk loans as a percent of capital also have been on the rise. The forthcoming issue of the FDIC Regional Outlook, which we will release shortly, shows that the percent of insured institutions with moderately high concentrations-that is, commercial and construction loans totaling between 400 and 700 percent of capital-has increased by more than half since 1995. A greater percentage of insured institutions, 17.1 percent, has concentrations in this 400 to 700 percent range now than at any time since at least 1984. This fact is troubling as history shows that banks with concentrations such as these consistently tend to fail more often than banks with lower concentrations-as much as two to three times as often by some measures. It is important, therefore, to recognize that the higher capital levels we see are accompanied in many cases by higher portfolio risks.

The FDIC is addressing the increase in credit risk in several different ways. The FDIC employs a risk-focused examination approach that enables examiners to prioritize risk and allocate staff to those areas of the bank that represent the most risk. Enhanced examination software tools give our examiners the ability to perform more sophisticated loan reviews with special emphasis on the higher risk C&I and construction/development loans. In addition, the FDIC recently instituted a large bank supervision program that provides more on-going supervision throughout the year for many of our largest institutions. Our offsite monitoring programs provide current data on loan growth and performance trends that are closely reviewed by staff assigned to monitor each insured bank. We also monitor the industry and local real estate markets through other vehicles such as the Report on Underwriting Practices and the Survey of Real Estate Trends. We continue to work closely with other regulators to improve the information exchanges and interagency cooperation that are necessary in today's rapidly evolving banking system. An example is the recently issued additional guidance to banks on risk management practices for leveraged financing.

As we contemplate further weakening in asset quality and slowing revenue growth in the near term, we should recognize that the banking industry today is far stronger than when it entered the last economic downturn more than 10 years ago. Banks now have more opportunities for geographic diversification and new sources of income. Banks also have been able to control growth in their overhead expenses, and to steadily improve efficiency.

Many of the observations made about commercial banks apply to insured savings institutions as well. While the profitability of insured savings institutions has been somewhat lower than the profitability of commercial banks, the past few years have brought strong earnings and growth for the thrift industry as well. Reflecting their historical role as providers of financing for home ownership, more than two-thirds of all loans held by insured savings institutions are home mortgage loans. At commercial banks, home mortgages account for less than one quarter of all loans. The large share of home mortgages in their loan portfolios means that most thrifts have lower net interest margins and lower credit risk than commercial banks. However, thrifts are subject to the same competitive pressures, and exhibit many of the same trends in performance and condition that we see at commercial banks.


The two deposit insurance funds managed by the FDIC reflect the favorable condition of the bank and thrift industries. The Bank Insurance Fund (BIF) reported a balance of $31.4 billion (unaudited) as of March 31, 2001, compared to $31 billion at year-end 2000. One BIF-member institution failed in the first quarter of 2001, and there have been just 22 BIF-member failures over the preceding five years. The BIF balance has grown in each of the last five quarters, but these increases failed to keep pace with strong growth in BIF insured deposits. As a result, the BIF reserve ratio1 has drifted downward, from 1.36 percent of estimated insured deposits at the end of 1999 to 1.32 percent as of March 31, 2001. From March 2000 to March 2001, BIF insured deposits increased by $180 billion. Nearly one-third of this amount ($57 billion) can be attributed to two organizations that have been sweeping brokerage-originated cash management funds into insured-deposit accounts at BIF-member bank affiliates. The insured deposit growth at these two organizations-without additional contributions to the insurance fund-has been enough to account for a 3-basis point decline in the BIF reserve ratio.

The Savings Association Insurance Fund (SAIF) also has reported steady growth, resulting in a balance of $11 billion as of March 31. No SAIF members have failed thus far in 2001, and only three SAIF members failed in the preceding five years. Recent insured-deposit growth has been relatively strong for the SAIF, although less so than for the BIF. SAIF insured deposits grew 1.7 percent during the first quarter of 2001 and 5.8 percent during 2000, compared to average annual growth of 0.6 percent in the preceding five years. The SAIF reserve ratio stood at 1.43 percent on March 31, which was unchanged from year-end 2000 and down slightly from 1.45 percent at the end of 1999. Brokerage account sweeps added an estimated $2 billion to SAIF insured deposits, accounting for a one-half basis point decline in the SAIF reserve ratio.


A transition from a decade of rapid economic growth to the slower growth the U.S. economy is now experiencing will, to some degree, adversely affect bank earnings. The impact is likely to be greatest on institutions that have been most aggressive in their selection of risks. In this regard, as they develop risk management strategies, insured institutions need to allow for the potential for economic conditions to be less favorable than prevailed during the 1990s.

Experience suggests that a weakening economy takes some time to affect banks. I would like to devote some attention to two issues that are more immediately before us, namely those posed by subprime lenders and lenders dependent on the agricultural economy. I also will discuss an issue that is extremely important to many banks today, that of funding and liquidity.

Subprime Lending

The FDIC continues to have concerns regarding subprime consumer and mortgage lending. We are closely watching approximately 150 institutions that have subprime lending programs, i.e., programs that purposely target subprime markets, in volumes that equal or exceed 25 percent of capital.

Subprime lending can be-and indeed, has been-beneficial to borrowers with blemished or limited credit histories and is an acceptable activity for insured institutions, provided that the institution has proper safeguards in place. Without these safeguards, mistakes can be costly, as evidenced by the role subprime lending has played in recent failures. Subprime lending figured prominently in six of the 20 bank and thrift failures in the past three and a half years. Further, since most subprime lenders in the bank and thrift industry have not been tested in an economic downturn, it is realistic to expect additional problems for institutions with concentrations of subprime loans should economic conditions deteriorate further.

Several factors that are often associated with subprime lending can create problems for the lenders, their regulators, and for the FDIC as receiver for failed institutions. One factor is the nature of the assets created as a by-product of loan securitization. In a securitization, the subprime lender sells packages of loans to another party or institution, but often retains the right to receive a portion of the cash flows expected from the loans. The expected value of these cash flows is generally referred to as the retained interest, or residual.

The residual holder's right to receive cash flows is generally a deeply subordinated position relative to the rights of the other security holders (as such, they serve as a credit enhancement to the other securities). To determine the value of this residual, the lender must make a variety of assumptions about the underlying loans, including delinquency rates, charge-off rates, and discount rates. As a result, and particularly with subprime loans, the accurate valuation of the residuals can be extremely difficult, making the residuals a highly illiquid and very volatile asset. In institutions with excessive concentrations of residuals, the safety and soundness of a bank or thrift may be threatened if the valuations turn out to be overly optimistic.

The complexity of subprime loan securitizations also means that accounting deficiencies are more likely. In some of the failures involving subprime lenders that securitized loans, accounting statements were deemed inadequate or inappropriate by bank supervisors.

Finally, subprime lending programs may use third parties for loan origination, servicing or other activities. The use of third-party originators and servicers is a standard business practice that can reduce bank costs and enhance efficiency. However, poor analysis and monitoring of loans purchased from third parties have contributed to the failure or near-failure of a few institutions due to misrepresentation, and even apparent fraud, on the part of the originator.

We have intensified our supervisory attention to the roughly 150 banks and thrifts with subprime lending programs. The banking agencies released the March 1999 Interagency Guidance on Subprime Lending. In January 2001, the agencies distributed the Expanded Guidance for Subprime Lending Programs. The focus of our supervisory attention is on the need for more intensified risk management procedures and internal controls for such higher risk lending programs, as well as the need for appropriate levels of reserves and capital.

Vulnerabilities in the Agriculture Sector

Farm banks remain in a vulnerable position as their profitability is linked so strongly to the uncertain economics of farming and the continuance of government support payments. Without government payments, many farmers would have significantly more difficulty meeting loan payments.

Today, more than 1,900 banks hold more than 25 percent of their loans in farm loans. While these farm banks constitute some 23 percent of all commercial banks, these banks tend to be smaller, rural community institutions, and hold less than 2 percent of all bank assets. Farm banks are highly sensitive to local economic conditions, being less diversified in their lending and sources of income. For instance, noninterest income contributes less than 15 percent of farm banks' revenue compared to over 43 percent for other commercial banks.

The FDIC is not predicting serious near-term problems in the farm bank sector. In spite of the well-publicized stress in the agriculture sector, the performance of farm banks, on average, remains quite steady with loan quality and capital positions remaining relatively strong. Only two percent of farm banks lost money in 2000. Most farm banks are currently well capitalized and well managed and generally are in much better financial condition than they were before the 1980s farm crisis.

Over the longer term, farm banks face the difficult issue of rural depopulation. U.S. Census data indicate that the Midwest has most of the counties in the U.S. that have lost population since 1970. Farms have been consolidating for decades, resulting in larger farms and lower populated rural areas.

To date, two sources of income have helped farmers, and thereby farm banks, avert a more serious financial crisis. In aggregate, farm households have come to depend more on off-farm income, mostly wages and salaries, for their livelihood. In addition, federal assistance remains significant, providing 49 cents of every dollar farmers earned in 2000.

However, the FDIC must remain vigilant for further declines in the agricultural economy. The U.S. Department of Agriculture currently forecasts a decline in net cash farm income in 2001 to under $51 billion, down from $56.4 billion last year (assuming no supplemental assistance for the 2001 crops). Higher energy costs also play a role in the forecasted decline.

Funding and Liquidity

During this record economic expansion, loan growth in the commercial banking industry has been exceptionally strong while deposit growth has failed to keep pace. This raises questions of decreased liquidity and continued credit availability, especially at community banks.

Since 1992, loans held on bank balance sheets have increased by $1.8 trillion or at an 8.3 percent compounded growth rate. In contrast, core deposits grew by only $709 billion, which translates to a 3.6 percent compounded growth rate. As a result, the share of commercial banks' assets funded by core deposits has declined steadily from its peak level of 62 percent at year-end 1992, to 46 percent at the end of 2000. During that same period, the percent of banks' assets that consists of loans increased from 56 percent to 60 percent.

Pressures stemming from the need to fund rapid loan growth are particularly evident at community banks, which traditionally have relied almost exclusively on core deposits to fund balance sheet growth. In this environment of strong loan demand, the balance sheets of banks with less than $1 billion in assets have undergone shifts in the composition of their assets and liabilities that have increased many community banks' exposure to interest-rate risk, credit risk and liquidity risk.

Many small banks appear to be liquidating securities to fund loan growth, and increasing the proportion of higher yielding, higher-risk loans in their portfolios in order to offset the increased cost of funding. This has helped to limit the erosion in community bank profitability in recent years. But these changes have left many small banks more vulnerable to rising interest rates and a slowing economy.

The ongoing loss of liquidity in banks' balance sheets is evidenced by the industry's historically high and rising loans-to-assets ratio. Loans are less liquid, that is, they are harder to convert into cash than assets such as U.S. Treasury securities or other marketable securities. Similarly, core deposits are important because they are not as volatile as many alternative sources of funds. They do not reprice quickly when interest rates rise, and because they tend to be fully insured, they do not flow out of banks when concerns about an institution's health arise. The loss of liquidity is also shown by the declining ratio of core deposits to assets, as banks have increased the share of loans in their asset portfolios and funded a growing share of their assets with nondeposit liabilities.

Increased reliance on liabilities other than core deposits implies potentially higher and more volatile funding costs for banks. Banks' inability to fund asset growth exclusively with core deposits has led to a growing dependence on large certificates of deposit and Federal Home Loan Bank (FHLB) advances. At the end of 1992, only 4.6 percent of commercial banks had any FHLB borrowings; these advances provided only 0.2 percent of commercial banks' funding. By the first quarter of this year, 45 percent of commercial banks had FHLB advances, which supplied 2.9 percent of the industry's funding.

There is no question that FHLB advances and other nondeposit funding sources play an important role in depository institutions' liquidity and funds management strategies. New Call Report data showed that, at the end of March, 52 percent of banks' FHLB advances had maturities in excess of three years. This suggests that many banks are attempting to use these advances to hedge interest-rate exposures of their longer-term assets. However, FDIC examiners have raised supervisory concerns in certain cases when a large concentration of an institution's funding needs were being met by FHLB advances or other wholesale funds and management did not fully understand the risks associated with those funding sources. Late last year, the FDIC issued guidance to our examiners for reviewing FHLB advances. Finally, on May 11, 2001, the FDIC and the other federal bank and thrift regulatory agencies issued a joint advisory on the risks of brokered and other rate-sensitive deposits that outlined prudent risk identification and management practices for deposits.

There is some evidence that liquidity pressures are easing. The past two quarters have seen a pickup in growth in core deposits, led by increases in money market deposit accounts. These savings accounts, which offer access to funds while paying interest on balances, can represent "safe havens" for investors seeking risk-free short-term investments. Growth in banks' domestic deposits has surpassed growth in loans for two consecutive quarters. But, two quarters is not a trend, and it is much too early to determine if recent strong deposit growth is more than a temporary blip.


Last year, the FDIC initiated a comprehensive review of the deposit insurance system. Our review identified some important flaws in the system, which we described in an Options Paper issued last August. I will describe the flaws and our recommendations for fixing them. A consensus appears to be emerging in support of several of the FDIC's recommendations, but some important implementation issues remain. I urge the Committee to take up these issues with my successor as soon as practicable, to ensure that we take advantage of the opportunity to enhance the deposit insurance system in good times, when the industry is strong.

The Case for Reform

One of the key flaws in today's system is that deposit insurance premiums do not reflect the risk that individual institutions pose to the system. Although the FDIC Improvement Act (FDICIA) mandates a risk-based deposit insurance premium system, other provisions of law prohibit the FDIC from charging premiums to institutions that are both well-capitalized, as defined by regulation, and well-managed (generally those with the two best examination ratings) when a fund's reserve ratio is at or above the Designated Reserve Ratio (DRR) of 1.25 percent. As a result, over 92 percent of insured institutions are in the FDIC's best-risk category and currently pay no deposit insurance assessment. All institutions pose some risk, and there are significant and identifiable differences in risk exposure among the institutions in the best-rated premium category. Indeed, even institutions with different CAMELS ratings (CAMELS "1" or "2") pay the same amount for insurance-zero. Having institutions with different risk characteristics all paying nothing for insurance renders the risk-based premium system ineffective, reduces the incentive for banks to avoid risk and forces safer institutions to subsidize riskier institutions.

The inability to price risk appropriately has had a number of other negative effects. Since very little in premiums has been collected since 1996, the deposit insurance system is financed almost entirely by those institutions that paid premiums in the past. There are currently over 900 newly chartered institutions, with over $60 billion in insured deposits, that have never paid premiums.

In addition, deposit insurance that is underpriced creates an incentive for institutions to grow rapidly. Financial institutions outside the realm of traditional banking recently began to make greater use of FDIC insured deposits in their product mix. Large dollar volumes of investment firm brokerage accounts were swept into deposit accounts in their FDIC insured subsidiaries. To the extent that these institutions are in the best-rated premium category, they pay no insurance premiums for this rapid growth. Since they are not paying for insurance, new institutions and fast-growing institutions are benefiting at the expense of their older competitors and slower-growing competitors. Rapid deposit growth lowers a fund's reserve ratio and increases the probability that additional failures will push a fund's reserve ratio below the DRR, resulting in a rapid increase in premiums for all institutions.

The second flaw in the current deposit insurance system identified by the FDIC study is that premiums are volatile and are likely to rise substantially during an economic downturn when financial institutions can least afford to pay higher premiums. By law, when a deposit insurance fund's reserve ratio falls below the DRR, the FDIC must raise premiums by an amount sufficient to bring the reserve ratio back to the DRR within one year, or charge all institutions at least 23 basis points until the reserve ratio meets the DRR. However, during a period of heightened insurance losses, both the economy and depository institutions in general are more likely to be distressed. A 23 basis point premium at such a point in the business cycle would be a significant drain on the net income of depository institutions, thereby impeding credit availability and economic recovery.

In addition to these two key flaws in the deposit insurance system, our review addressed two other important issues. The first is the existence of two separate deposit insurance funds. As long as the FDIC maintains two funds, whose assessment rates are determined independently, the prospect of a premium differential with its attendant inefficiencies and inequities exists. Separate funds also are not as strong as a combined deposit insurance fund would be. Moreover, because each insurance fund now insures both banks and thrifts, there is little justification for maintaining separate funds.

The second issue is the erosion in the real value of deposit insurance over time. Deposit insurance coverage is an important component of the federal government's program to promote financial stability, yet there is no mechanism for regular adjustments to maintain its real value as the price level rises.

The FDIC's Recommendations

The FDIC published the following recommendations for reforming our deposit insurance system on April 5, 2001.

  • The current statutory restrictions on the FDIC's ability to charge risk-based premiums to all institutions should be eliminated; the FDIC should charge premiums on the basis of risk, independent of the level of the fund.
  • Sharp premium swings triggered by deviations from the designated reserve ratio should be eliminated. If the fund falls below a target level, premiums should increase gradually. If the fund grows above a target level, funds should be rebated gradually.
  • Rebates should be determined on the basis of past contributions to the fund, not on the current assessment base.
  • The Bank Insurance Fund and the Savings Association Insurance Fund should be merged.
  • The deposit insurance coverage level should be indexed to maintain its real value.

Collectively, these recommendations will result in a deposit insurance system that will allocate the assessment burden more smoothly over time and more fairly across institutions. They are not designed to increase the long-term assessment revenue to the FDIC.

These reforms are designed to be implemented as a package. Picking and choosing among the parts of the proposal without focusing on the interaction between the various recommendations could weaken the deposit insurance system, magnify macroeconomic instability, and distort economic incentives.

At a general level, a consensus appears to be emerging in support of several of our conceptual recommendations. There is broadening agreement that:

  • The deposit insurance system must be less procyclical. That is, premiums should not rise sharply during an economic downturn taking funds out of the banking system when they are needed most to help fuel a recovery.
  • The FDIC must be able to charge appropriately for risk, both because the current system creates perverse incentives and because riskier institutions should shoulder more of the assessment burden for deposit insurance.
  • Reform must address the issue of deposit growth, to lessen the impact of rapid growers on the rest of the industry and to bring a measure of fairness to the funding of the deposit insurance program.

Some important implementation issues remain to be resolved. These are the issues on which the FDIC will need to focus its discussions and build consensus going forward. One is how to set the target level for the fund. It is important to note, however, that a target level, be it a point or a range, should probably not be fixed permanently. It would be wise to revisit the performance of the fund and general economic conditions every few years and adjust accordingly. Another issue is how to differentiate among institutions on the basis of risk and charge premiums accordingly. A third issue is how to determine the size and allocation of rebates.

The FDIC's reform proposals were accompanied by various illustrative examples of ways of addressing these issues. These issues require policymakers to weigh and balance important policy goals. For example, in determining how to price risk across banks, actuarial judgments must be balanced against public policy goals. On an actuarial basis, banks with substantial loan concentrations pose a greater risk to the insurance fund, other things being equal. From a public policy point of view, however, it may not be desirable to over-penalize lenders in communities that happen to be dependent upon particular industries. As the examples illustrate, none of these issues are insurmountable, and working together we can implement meaningful deposit insurance reform.


I appreciate the opportunity to testify regarding the overall strength and prosperity of the banking industry. Today's strong economy and banking system also provide a window of opportunity to improve the deposit insurance system. It would be a missed opportunity to wait until the economy and the banking industry are suffering and the results of the weaknesses in the deposit insurance system have become all too evident. The FDIC's recommendations will strengthen the deposit insurance system, promote economic stability, enhance safety and soundness, and make the system more equitable.

These reforms will work best if implemented as a package. In particular, the ability to price for risk is essential to an effective deposit insurance system. Picking and choosing among the parts of the proposal could weaken the deposit insurance system, magnify macroeconomic instability, and distort economic incentives. Trying to address other issues without addressing risk pricing does not solve one of the most fundamental flaws in the current system.

I would like to thank Chairman Sarbanes, Senator Gramm, and Members of the Committee once again for the Committee's interest in this important issue and for the opportunity to present the FDIC's reform proposals. I hope that this Committee and the Congress, working with my successor, will be able to bring about these much needed reforms.

In closing, I also would like to thank my colleagues at the FDIC who produced the reform recommendations I have discussed and who work so hard at insuring a safe and sound financial system for the American people. It has been a pleasure and a privilege to work with them.

1 The reserve ratio is the fund balance divided by the dollar volume of estimated insured deposits.

Last Updated 06/20/2001communications@fdic.gov