Dear Mr. Feldman:
The American Bankers Association (ABA) appreciates the opportunity to comment on the Federal Deposit Insurance Corporation’s (FDIC) notice of proposed rulemaking (NPR or the Proposal) on interest rate restrictions on institutions that are less than well capitalized. The Proposal would amend the methodology for calculating the national rate and the national rate cap and simplify the local rate cap determination. The Proposal also seeks comment on alternative approaches to setting the national rate cap, including setting it at the higher of the current rate cap or 120 percent, or 130 percent for wholesale deposits, of U.S. Treasury securities, plus 75 basis points (the pre-2009 rate cap).
ABA applauds the FDIC’s initiative to review and modernize its brokered deposits framework, including efforts to amend the methodology behind the national rate cap. In particular, we appreciate the FDIC’s reiteration that the interest rate restrictions should not be applied to well capitalized institutions, and the modification of the Risk Manual of Examinations Policies to this effect. While the interest rate restrictions may only be officially applied to less than well capitalized institutions, the national rate is used in various internal and supervisory stress scenarios. It is important, therefore, that FDIC employment of a national rate and rate cap be transparent and financial market based, not supplanting markets.
Overall, while we are supportive of the FDIC’s reform efforts, we do not believe that the Proposal goes far enough toward correcting the problems of the current methodology. In our view, the Proposal fails to achieve an essential goal: establishment of a robust and transparent market based methodology that produces a rate that accurately reflects the cost of deposits through varied business models and economic environments.
The purpose of the national rate cap is to create a benchmark against which the FDIC is able to gauge deposit rates that “significantly exceed” those offered by other banks so as to limit the risks from rates that less than well capitalized banks may offer on deposits. Under the current rule, however, the national rate cap too often stands significantly below banks’ actual cost of deposits, which is clearly contrary to the purpose of the cap. Today’s below market rate is largely a result of the FDIC’s reliance on imperfect data and a methodology that is not responsive to and therefore not reflective of the economic environment or financial markets. While the Proposal nominally improves upon the rates offered by the current approach with respect to longer maturity deposits, it does not actually solve the methodology’s core problems. In fact, the Proposal could prove even more problematic for less than well capitalized banks and lead to a procyclical restriction of rates, potentially causing an artificial liquidity stress.
We recommend that, instead of implementing the proposed methodology, the FDIC return to one that considers the interest rate and business environments in which banks raise deposits. A market based approach would naturally produce a rate that factors in all bank business models and non-bank competition and include currently excluded products such as promotional CD products (e.g., ‘‘off-tenor’’ terms), specials offered (e.g., cash incentives), rewards checking products, and products that are available only in the online marketplace. One possibility is using the higher of (1) the pre-2009 rate cap or (2) a new market rate that incorporates the federal funds and Treasury markets data. This approach would be more transparent, precise, and dynamic than either the current or the proposed methodologies. As described more fully below, the pre-2009 approach is a blunt proxy that in practice works through most scenarios. We present a second rate, which refines the pre-2009 methodology by creating a floor and adjusting for the shape of the yield curve.
With respect to the local rate, ABA is generally supportive of the modifications and simplification of the process. As insurance against abnormal events in the Treasury markets, ABA encourages the FDIC also to establish a periodically reviewed list of allowable alternatives for use by all institutions. As the FDIC is aware, the industry is substantially less homogenous than it was in the early 1980s when the concept of a national rate cap was introduced. Significant changes in technology and enhanced competition for deposits mean that banks of all sizes source their funding through varied mechanisms and markets. Thus, ABA supports a flexible approach that is consistent with the purpose of the cap, while allowing banks and their examiners to use the benchmark most appropriate for the institution’s unique circumstances and that prevents any unintended liquidity stress an artificially low rate cap may cause.
The Proposal does not solve the flaws of the current national rate calculation
In addition to setting restrictions on brokered deposits, Section 29 of the Federal Deposit Insurance Act directs the FDIC to limit the amount of interest a less than well capitalized institution may pay on deposits. In general, a less than well capitalized bank may not offer an interest rate on brokered deposits that is higher than the “national rate” established by the FDIC plus 75 basis points. The national rate is currently calculated by taking a “simple average of rates paid by all insured depository institutions and branches for which data are available.”
Under the proposed approach, the FDIC would modify the methodology behind the national rate by finding the average rate per institution for each product. Products would include transaction accounts, savings accounts, CDs, and money market deposit accounts (MMDAs). A weighted average would then be calculated for specific products by considering a bank’s market share of total domestic deposits. The national rate cap would be set to the higher of (1) the rate offered at the 95th percentile of rates weighted by domestic deposit share or (2) the proposed national rate plus 75 basis points.
We support the intent to include a broader range of bank business models, but the Proposal does not go far enough toward solving the current approach’s material flaws, including the problem that the base dataset is not robust enough for the purpose. The data used in calculating the national rate, under both the current and proposed methodologies, are not an accurate or complete enough data set to achieve the desired result of a true and practical rate that is reflective of the competitive environment banks are operating within. The underlying data are gathered by a third party, which only collects advertised bank rates, excluding any specials, deposits for which the bank and customer have negotiated the rate, and deposit like products offered by nonbank competitors. Use of these data to calculate the national rate cap will continue to result in an inaccurate, incomplete rate that does not reflect the cost of deposits at any given time.
Moreover, because the 95th percentile of rates is still within what is the normal market, the percentile approach does not achieve a rate that “significantly exceeds” the market. In order to maintain and attract deposits, less than well capitalized banks need to offer rates aligned with or slightly above normal market rates. Should the FDIC implement a percentile approach, we recommend adding a sufficient buffer, of at least 75 basis points to the 95th percentile.
We also urge the FDIC not to use a weighted approach. The Proposal, which continues to overweight the largest institutions with a traditional branch based model, amends the methodology for calculating the national rate and the national rate cap, primarily through weighting the same limited deposit rate data by deposit share rather than by the number of branches. We remain concerned about the continued use of weighting, which whether it be by branch, market share or size, tends to misrepresent actual market rates.
The FDIC should adopt a more market based methodology
The rates that banks choose to offer on deposits are determined by reference to a variety of factors across multiple tenors and products, their competitors’ rates, and various benchmarks including Treasury rates, Federal Home Loan Bank (FHLB) advances, swap rates, and money market rates, among others. However, neither the current nor proposed methodologies employ or consider these common benchmarks or indexes to set the national rate cap. ABA suggests creating a market based national rate scheme that combines the pre-2009 approach with another method that considers and adapts to different rate environments.
The pre-2009 national rate cap, which was in use from 1992-2009, was a blunt but simple and transparent methodology that proved accurate a vast majority of the time. To accommodate various interest rate environments and financial market conditions, we urge the FDIC to adopt a second market-based rate. Institutions would pick the higher of the two rates.
Alternative market based rate
ABA presents a formula, the output of which is a curve for the national rate cap that is more refined than the pre-2009 approach, yet it is transparent and relatively simple. Because the formula includes spreads over both overnight and longer-term horizons, the curve would appropriately function as a gauge for above market deposit rates. For simplicity, we used 75 basis points. This alternative market based rate consists of two parts: a fed funds rate-based variable for establishing an upper limit for non-maturity deposits and a Treasury adjusted variable for longer maturity deposits.
Effective funds rate plus a spread. The effective fed funds rate is the interest rate depository institutions charge one another to lend reserve balances overnight on an uncollateralized basis. The effective federal funds rate is a market rate and driven by the supply and demand of money, monetary policy, and expectations of credit and liquidity in interbank lending markets. Because the rate is an overnight uncollateralized bank lending rate, it is a robust and transparent starting point for gauging the cost of non-maturity deposits. To set the rate, a spread above the fed funds rate is needed to add a sufficient upward adjustment to reflect the higher cost of deposits relative to fed funds (which also acts as a floor). To adjust for longer-term deposit products, we start with the spread above the effective federal funds rate and add an adjustment for the shape of the Treasury yield curve (described more fully below as the Treasury Adjusted Variable Spread).
Treasury adjusted variable spread. Treasury yields are reflective of Fed policy, market conditions, and future expectations. Banks must price deposits competitively with Treasuries, as they serve as the risk-free benchmark rate at different time periods. Customers use these benchmarks to evaluate competing financial instruments vis-a-vis their risk tolerances. Rather than simply adding a fixed 75 basis points (bps) to the natural rate for each tenor CD, a variable yield spread based on Treasuries would be more responsive to expectations and changes in the term premium. The formula’s Treasury adjusted variable spread takes the ratio of corresponding Treasury yields against the average of yields on 1-to-5 year Treasuries. This ratio creates a data point for that tenor that, taken together with the range of tenors, creates a curve that mirrors the Treasury yield curve.
The combination of money market rates with a fixed spread and a variable spread based on Treasury yields provide a robust base for calculating the rate cap. Key advantages of this approach are that it is responsive to different interest rate environments and that it creates a floor that prevents the rate from ever falling below the effective fed funds rate.
Appendix I provides additional information and comparisons of where this rate sits relative to other metrics, including the proposed methodology, over the past 18 years, across tenors.
The local rate cap should be a competitive rate
We generally support the proposed changes to the local rate option. In particular, we support the simplified process, the inclusion of credit unions, and the lack of weighting, all of which better reflect deposit market dynamics for institutions with smaller footprints. The Proposal would simplify the local rate cap calculation and process by allowing less than well capitalized institutions to offer up to 90 percent of the highest rate paid by both banks and credit unions on a particular deposit product in the institution’s local market area. While this is clearly an improvement, we have the same concerns with respect to the proposed percentile approach—we do not believe taking a rate that is within the bounds of normal market rates yields a rate that “significantly exceeds” the market. The result of the approach would be a rate that prevents less than well-capitalized banks from raising deposits at a competitive market rate.
To achieve a better result, we recommend that the FDIC allow an institution to use 125 percent of a competing rate as the upper limit of what less than well capitalized banks may offer on deposits. This allows less than well capitalized banks to offer competitive rates on deposits, while not going so far above normal market rates as to exacerbate safety and soundness concerns.
The FDIC should provide a list of alternative rates
ABA encourages the FDIC to expand on the proposed national and local rate approach by establishing a periodically reviewed list of allowable alternatives. A market based approach, such as the one described above, can be expected to be an accurate gauge of deposit costs the vast majority of time. For those uncommon times when a market based national rate might not the be the best gauge of a bank’s cost of deposits, we encourage the FDIC to allow all banks to opt for the most appropriate alternative rate. For example, eligible rate could include FHLB rates, 125 percent of a competitor’s rate, or a percentage of the top rates provided by a listing service. Of course, banks would need to defend their selection to their examiners, demonstrating why the national rate cap is not the best choice and providing solid evidence as to why the preferred rate is more appropriate. This added flexibility further ensures that less than well capitalized banks are not prohibited from raising deposits at a competitive rate.
The proposed treatment of non-maturity deposits for purposes of interest rate restrictions is operationally unworkable
The FDIC proposes an interpretation of when non-maturity deposits are considered “accepted” or “solicited.” Under the proposed interpretation, balances already on deposit in money market demand accounts or other savings accounts would not be subject to the deposit restrictions at the time an institution becomes less than well capitalized, but with funds deposited into the account after the institution becomes less than well capitalized the entire balance would be subjected to rate restrictions. This would require banks to maintain parallel products and systems to be able to track accounts and multiple rates in the event the bank becomes less than well capitalized. We also note that forcing a customer’s rate down, should he or she deposit an additional amount in the account hurts consumers and will likely cause liquidity stress as customers move their balances elsewhere. We urge the FDIC not to finalize this component of the Proposal as it would be operationally infeasible and therefore not a practical solution.
Instead, we recommend that once an institution falls below less than well capitalized the FDIC exempt, or grandfather, all existing deposit accounts from the rate restrictions, restricting only new deposits to new accounts opened with the bank. We also suggest allowing a 5-7 day transition to approve and market new deposit products, and process the new accounts.
Conclusion
We appreciate the FDIC’s step toward modernization of the national rate cap methodology. ABA stands ready to continue this important dialog with the FDIC for the benefit of our members’ customers and to reinforce the strength of the banking system. If you have any questions about these comments, please contact the undersigned at (202) 663-5147 or email: [email protected]
Sincerely,
Alison Touhey
Vice President, Banking Funding Policy
