Congress often loads up major legislation with studies of all kinds, but these generally accomplish little. The Dodd-Frank Act contained dozens of studies, but one that will prove to be an exception to the rule, and very valuable, is the Federal Deposit Insurance Corp.'s study of brokered deposits, which is to be completed by the end of July. It is valuable because the current regulatory structure for brokered deposits is badly out of date. Its definitions capture deposits that should not be covered, while excluding those that should be covered.
The FDIC is to be commended on the thoughtful approach it is taking to this study. On March 18, Chairman Sheila Bair chaired a roundtable discussion attended by the five members of the FDIC board. Bankers should take this study seriously as well. The results will greatly impact liquidity, funding costs, competition, and the number and cost of bank failures. More broadly, many banks simply will not be able to fund the loan growth needed to expand our economy from local deposits alone.
First, we should stop referring to "brokered" deposits. In fact, brokered deposits are not the real issue. For the roundtable, FDIC staff framed part of the discussion as "stable" versus "volatile" deposits. I would relabel the issue as volatile deposits.
There is no doubt that such deposits were a problem in the 1980s. With the demise of most interest rate controls, depository institutions were able to attract money through brokers by offering high rates. In too many cases, institutions used high-priced, volatile deposits to grow too quickly and to fund risky loans or investments. When some failed, their franchise value was undermined because these volatile deposits ran off.
In response, the FDIC tried first to address the problem through regulation, but this approach was thrown out by the courts, leaving it with no choice but to seek legislation. The legislative solution, adopted in 1989 and amended in 1991, was unfortunately misdirected in two ways.
First, the law made capital levels the trigger points for increased regulatory sanctions, while no sanctions were applicable to well-capitalized institutions. So no matter how great the use of volatile deposits, regulators had their hands tied until an institution was already headed down a slippery slope. Furthermore, these capital triggers have often resulted in counterproductive restrictions on deposit-gathering by stressed institutions, leading to unnecessary liquidity problems for these banks.
Second, Congress mistakenly focused on the means — deposit brokers — by which deposits were gathered, rather than the characteristics of the deposits themselves. The result was inevitable: With changes in technology, competition and business models, stable deposits that were gathered through a method that met the out-of-date definition of brokered were restricted, while volatile deposits gathered otherwise were not covered. An example of the changed landscape: One of the participants cited the ability to Google "high-rate deposits" and find numerous ways to deposit money at high rates. Some of these deposits likely would be considered volatile, but they may not be brokered.
There is growing consensus that the focus should be moved from a rigid definition of the means of gathering deposits to the characteristics of the deposits themselves. The FDIC members noted that the current paradigm may be obsolete, and the roundtable discussion provided an excellent start to identifying characteristics of problematic deposits.
The following characteristics are interconnected, and all of them should be considered: volatility (the stickiness of the deposits in the institution); the customer relationship (does the depositor have a relationship with the institution that will make deposits likely to remain there?); cost (is the interest rate offered well above market?); duration (a five-year certificate of deposit is stable, particularly if there are withdrawal penalties or prohibitions); the correlation with bank problems and failures; and the impact on the ability to sell, and the price obtained for troubled and failed banks.
No one characteristic, though, is determinative. For example, both a high-cost deposit with a strong relationship or a low-cost deposit with a long duration but little relationship could be very stable in this context.
From the 1980s through the recent crisis, a good deal of information has been developed on these characteristics for various types of deposits. The FDIC and the private sector now have the ability to model these traits for both current deposits and new variations that will be developed.
The approach in existing law is misdirected, too rigid and should be replaced. The best method is to provide bank regulators flexibility to use regulatory tools in a timely fashion based on the careful consideration of specified characteristics of deposits.
