Speeches & Testimony
Keynote Remarks by FDIC Chairman Jelena McWilliams on "Brokered Deposits in the Fintech Age" at the Brookings Institution, Washington, D.C.
December 11, 2019
On April 29, 1982, the superintendent of banks for the State of California closed down a bank in San Diego by the name of Pacific Coast Bank. In the four months prior to failure, Pacific Coast Bank, despite a capital position of less than $200,000, acquired more than $5 million in brokered deposits and booked more than $5 million in loans. The majority of the new loans were fraudulent and resulted in jail sentences for the individuals who controlled the bank. The bank failure ended up costing the FDIC approximately $5 million.
More famously, or infamously, and far more costly to the FDIC, was the story of Penn Square Bank, a one-office bank based out of a shopping mall in Oklahoma City that was closed a few months later. From 1977 to 1982, Penn Square Bank grew dramatically from around $30 million in assets to more than $400 million. The bank focused on oil and gas loans, many of them notoriously poorly underwritten, and many of which it sold to upstream banks. When oil prices began to fall, the bank suffered large losses.
Prior to January 1982, Penn Square maintained only a moderate level of brokered deposits – around $20 million. However, between January and July, the total brokered deposits at Penn Square increased to $282 million. Even as the condition of the bank deteriorated and local depositors pulled their money, Penn Square continued to churn out loans. On July 5, the Office of the Comptroller of the Currency (OCC) closed the bank, and losses to the Deposit Insurance Fund (DIF) ultimately amounted to $65 million.
The failure of Penn Square Bank shined a spotlight on the use of brokered deposits, prompting Congressional hearings and regulatory scrutiny. It was one of many bank regulatory issues Congress and regulators debated as hundreds of banks and thrifts failed across the country over the next few years. Many of these institutions exhibited a similar pattern: rapid growth in risky, poorly administered, low-quality assets, fueled by brokered deposits, followed by attempts to "grow out of their problems" by adding more of both. This was an era before minimum capital ratios, when banks could operate with very low – or even negative – levels of capital, but continue to stay afloat. When these so-called "Zombie Banks" failed, it was very costly to the insurance fund.
In 1989, Congress acted by, among other things, amending the Federal Deposit Insurance Act by adding Section 29, which generally prohibits the acceptance of brokered deposits by banks that fail to maintain a minimum level of capital.1 Congress addressed a specific problem present at both bank failures I just described – the use of brokered deposits by troubled institutions to fund rapid growth in low-quality assets.
The statute mandates that the FDIC use a blunt tool to address an important policy problem. The law permits the unlimited use of brokered deposit funding for banks that are well capitalized, but completely cuts off banks from all brokered deposit funding once capital falls below a certain level. Yet this is when banks are most in need of a stable source of liquidity to try to right the ship. While some 500 banks failed during the recent financial crisis, many others that experienced trouble were able to stabilize and survive at no cost the Deposit Insurance Fund.
Meanwhile, the language used in the statute has led to many questions regarding what is or is not a brokered deposit. The statute defined a "deposit broker" in a manner that some have interpreted very broadly, and others more narrowly, and provided nine exceptions from this definition, including one – the "primary purpose exception" – that some have interpreted very broadly, and others more narrowly. The FDIC historically has interpreted the term "deposit broker" broadly, with notable exceptions, and has interpreted the primary purpose exception narrowly, again with exceptions.
Since Section 29 became law in 1989, the industry has changed dramatically, and it continues to evolve today. In the 1980s, businesses communicated through phones and fax machines, and bank customers typically deposited money and accessed banking services by walking into a local bank branch. The term "brokered deposits" was commonly understood to refer to a specific market in which third-party brokers gathered and pooled funds and placed them in "brokered" CDs issued by banks.2
Since then, the emergence of the internet and other technological innovations has fundamentally changed how banks interact with their customers. While some depositors still walk into a local branch, many customers now access banking services solely through the internet or smartphones, and today customers are increasingly utilizing channels such as prepaid cards and third-party fintech apps. Bank accounts are increasingly available to consumers through partners or affiliates of banks.
A prime example is so-called "sweep accounts," which emerged around the early- to mid-2000s, in which broker dealers "sweep" cash balances that are not invested in stocks and bonds into FDIC-insured bank accounts or money market funds. Additionally, regulatory changes, such as the repeal of interstate banking restrictions, have increased competition for deposits in markets throughout the country.
As the banking landscape has changed, and the permutations of how deposits are structured and offered have expanded, the FDIC's brokered deposits regime has struggled to keep up. For years, the FDIC faced difficult questions regarding whether different types of deposit arrangements should be reported as brokered. The FDIC responded to each of these questions on a one-off basis, typically through confidential letters or public or non-public staff advisory opinions. The result was the development of a fragmented, opaque legal regime that exists outside of the FDIC's public-facing regulations, understood by only a select few.
It reminds me of how my good friend Randy Quarles described the Federal Reserve's treatment of "control" under the Bank Holding Company Act, which he said was understood only by those with "a long apprenticeship in the subtle hermeneutics of Federal Reserve lore, receiving the wisdom of their elders through oral tradition like the tribes of Orinoco." If English were my first language, maybe I would have a greater appreciation for subtle hermeneutics, but I suspect many of you will need to look up where the Orinoco is too. Nonetheless, it seems like a suitable description of the FDIC's brokered deposits regime.
In 2015 and 2016, the FDIC issued a set of FAQs3 in an effort to more clearly explain the agency's interpretation of Section 29, and how its interpretation applied to different types of deposits. While this was a commendable effort to provide more transparency, it did not address many of the underlying interpretative issues and inconsistencies that had arisen over the years, and many firms continue to operate under individual non-public advisory opinions.
In December 2018, six months after I became Chairman, the FDIC issued an advanced notice of proposed rulemaking seeking comment on how to update and modernize the brokered deposits regime. I promised to undertake a comprehensive review of changes to the industry and the FDIC's regulatory policy. After receiving more than one hundred comments and a year of closely studying changes to the industry, the FDIC Board will vote tomorrow on a proposal that establishes a new framework for regulating brokered deposits.
The new framework was designed with four specific goals in mind.
The first goal is to develop a framework that encourages innovation within the industry, and allows banks to serve customers the way customers want to be served. The approach is intended to reflect how consumers want to access banking services in 2020 and beyond, not how they did so in 1989. The proposal will clarify that various types of existing partnerships in which a consumer maintains a direct relationship with a bank generally would not result in a brokered deposit.
Since becoming Chairman, I have prioritized establishing a regulatory approach that encourages, rather than stifles, innovation. Striking the right balance in how we interpret the brokered deposits statute is key to this goal, as we seek to remove regulatory hurdles to innovative partnerships between banks and nonbanks, and avoid discouraging banks from offering products and services through online and mobile channels.
The second goal is to take a balanced approach to interpreting Section 29 that tracks the letter and spirit of the law. For example, the statute provides that an entity is not a deposit broker, and thus deposits it places are not brokered deposits, if the entity's primary purpose is not the placement of deposits at depository institutions. Under our proposal, the FDIC would analyze whether the primary purpose exception was met by looking at the business relationship between the third party and the customers for whom it is placing, or facilitating the placement of, deposits, consistent with the plain meaning of the statute.
The third goal is to minimize risk to the Deposit Insurance Fund, and to ensure we address the core problems Congress sought to address in enacting Section 29. In furtherance of this goal, under the proposal "brokered CDs" would continue to be treated as brokered deposits under the proposal. These were the specific products members of Congress explicitly referred to when debating the legislation, and they continue to comprise a large portion of the brokered deposits market today. Brokered CDs are used by some investors to place money in a number of FDIC-insured accounts, thus receiving FDIC insurance on large sums of money by spreading out deposits to stay within the $250,000 statutory limit at each institution.
Separately, the FDIC is also considering changes to its deposit insurance assessment pricing to address concentrations in funding that are correlated with higher losses to the DIF. This could include unaffiliated sweeps that qualify for the primary purpose exception under the proposal and certain listing service deposits, which, in many cases, are not considered brokered today. The changes the FDIC is considering to its brokered deposits regulation should not be interpreted to mean the FDIC is ignoring the potential risks associated with different forms of funding. For large banks, a modification to address these types of funding concentrations is one of several changes the FDIC is considering to make its pricing more risk-sensitive.
The fourth goal of the new framework is to establish an administrative process that emphasizes consistency and efficiency. First, our proposal would establish an easy-to-understand, bright-line standard for determining whether an entity satisfies the statutory definition of a "deposit broker" or not. Second, it would establish an application process for implementing the primary purpose exception.
Today, when a new product is developed and an institution wants to know whether to report it as brokered, there is no established process for making such a determination. At times, requests to the FDIC for clarity have lingered for years. Meanwhile, there are certain types of deposits that some institutions report as brokered and others do not report as brokered. We intend to establish a more rigorous process to give institutions clear and timely feedback on whether deposits should be reported as brokered or not.
Those are our four goals. Optimizing for all of these goals is challenging, and we look forward to feedback on the proposal, if approved by our Board. The new framework is still somewhat complicated, but this is the inevitable result of an ambiguous statute that was written in a very different era.
It is, of course, up to Congress to decide whether and when a holistic review of the brokered deposits statute is warranted. But if Congress were to undertake such a review, there might be better ways to accomplish the worthwhile public policy goal Congress intended to address. One option to consider is replacing Section 29 of the FDI Act altogether with a simple restriction on asset growth for banks that are in trouble. This would be a far easier regime for the FDIC to administer, would at the very least limit the size of the FDIC's potential exposure, and would more directly address the key goal of preventing troubled banks from using insured deposits to try to grow out of their problems. A simple limitation on asset growth would also be more durable and should retain its effectiveness as the industry evolves and as banks change the way they attract deposits over time.
If Congress were to choose not to replace Section 29 altogether, and instead maintain a statutory restriction on brokered deposits, Congress could consider repealing the primary purpose exception – which puts the FDIC in the challenging position of trying to decipher the purpose of every potential deposit broker in the country – and replacing it with a more flexible exception based on actual risk to the DIF. Congress could also consider allowing more flexibility regarding when and under what circumstances brokered deposits are prohibited or restricted.
If Congress chooses to tackle these issues, the FDIC stands ready to provide assistance. In the meantime, we will implement the law as it exists today. The proposal the FDIC Board will vote on tomorrow is the next step in the process.
According to the FDIC's most recent National Survey of Unbanked and Underbanked Households, 4 more than 20 million Americans live in households in which no resident has a checking or savings account at a bank. Yet approximately half of these individuals own or have regular access to a smartphone. Meanwhile, many banks located in low-income areas, including a number of minority depository institutions, depend on deposits sourced from outside their local areas for funding. The changes we are considering to the brokered deposits regulation are intended to help such individuals and such firms.
Thank you for your time today.
1 Initially, the statute prohibited an insured depository institution that did not meet the applicable minimum capital requirement from accepting brokered deposits, but the FDIC was given authority to waive this prohibition upon finding that acceptance of such deposits did not constitute an unsafe or unsound practice. (See FIRREA sec. 224) In 1991, the FDI Act was amended by Congress to establish prescribed capital levels under the prompt correction action framework, and the brokered deposits restrictions were tied to this framework. The statute also limited the FDIC's authority to waive the prohibition on accepting brokered deposits to only firms that are adequately capitalized and not to firms that are undercapitalized or significantly undercapitalized. (See FDICIA Sec. 301)
2 In its publication The History of the Eighties, the FDIC wrote "Brokered deposits are certificates of deposit issued by a financial institution and purchased by an investor through a third-party intermediary; the third party receives a fee or commission from the issuing institution." This is not a legal definition, but reflects the commonly understood meaning of the term at the time.
4 Available at https://www.fdic.gov/householdsurvey/2017/2017report.pdf.