- As rates drop and profit compresses, we urge banks to embrace asset/liability management (ALM), funds transfer pricing (FTP) and the analytic intelligence of deposit behavior.
- Rapidly rising rates proved that banks with real-time deposit analytics were better at forecasting and managing the performance of their balance sheets and margins.
- For many institutions, stressed-liquidity analytics for deposits is more art than science — we see ways to improve no matter how mature the practices may be.
After a tumultuous rising rate cycle that saw four consecutive hikes of 75 basis points — which in prior years would have been viewed as hypothetical scenarios to be modeled by bank treasury functions — the Fed is now foreshadowing interest-rate cuts for the first time in nearly four years. In response, treasury and risk-management functions at banks are shifting to balance sheet optimization in a falling rate environment.
Curinos expects that this phase will be defined by margin pressures. On the deposit side, we expect continued portfolio remixing from lower-cost to higher-cost products. This will include a significant increase in CD originations that will drive portfolio costs upward even as acquisition rates tick down. Meanwhile, variable-rate loan repricing and lower yields on new fixed-rate loans will reduce yields. These dynamics, along with higher asset liquidity buffering in the aftermath of bank receiverships in early 2023, are sure to squeeze profits.
Another key ingredient for profitability will be the extent to which Treasury and Risk invest in asset/liability management (ALM), funds transfer pricing (FTP) and the analytic intelligence around deposit behavior applied to each. Robust ALM and FTP will be critical for measuring and managing interest-rate and liquidity risk and, by extension, balance sheet optimization.
ALM: Realistic Deposit Beta And WAL Assumptions
Rapidly rising rates earlier in this cycle proved that banks with more granular, real-time deposit analytics — beta and weighted average life (WAL) being the most critical — better anticipated and managed the performance of their balance sheets and margins.
Industry benchmarking from Curinos Treasury Analyzer shows that not only did betas increase in a non-linear fashion as rates rose but also that WALs declined materially across all segments of deposits, though by different degrees (Figure 1). This suggests deposit duration shortened significantly and, as a result, the liability sensitivity of bank balance sheets increased in 2022 and 2023. Banks with a better view of WAL shortening had the information they needed to set more realistic deposit forecasts, make smarter interest-rate hedging decisions and better share IRR profiles to internal and external constituencies as they saw fit.
Figure 1: Change In Trailing 12-Month WAL, 2023
Banks with a better view of WAL shortening were positioned to forecast deposits more realistically and hedge interest rates more effectively.
One reason we anticipate continued pressure on deposit portfolio costs even as rates decline is that sizable swaths of both retail and commercial deposit balances are priced well below prevailing acquisition rates. Another reason is the current competitive demand for deposit growth amid stagnant industry deposit growth — we expect this to keep competition for deposits high and thereby sustain high acquisition rates. Curinos data show that, as of December 2023, more than half of retail savings/term deposits and 40% of commercial non-operating balances were priced below 300 basis points (Figure 2).
Figure 2: Retail And Commercial Balance Mix By Rate Paid | Dec 2023
Deposit portfolio costs will face ongoing pressures due to the high volume of deposit balances priced well below prevailing acquisition rates.
For many banks, this dynamic conflicts with how falling interest rates are represented in their ALM systems. Betas in falling-rate scenarios are often shown as being positive, meaning that deposit costs will decrease immediately as market rates decrease. In addition, many banks assume that as rates fall, betas will be higher in magnitude than those set for a rising-rate scenario — that is, that deposit costs will decrease faster as rates fall than they increase as rates rise.
Senior management committees focused on balance sheet management, such as ALCO, should stress-test their current ALM assumptions to align Treasury’s views with pricing strategies of the business lines and their expectations for continued balance remixing as rates remain high. It would also help ensure that the right deposit beta assumptions are configured as the cycle turns. For example, from the perspective of customer management, immediate reductions to deposit pricing may be a non-starter given competitive pressures, or if such reductions are built into to the assumptions, they would likely trigger increased balance attrition.
FTP: Optimizing Contingent Liquidity Costs
2023 proved that understanding the short-term liquidity profile of deposits is paramount for effective balance sheet management and that profiles vary considerably across deposit segments. This was clearer than ever after some high-profile banks fell into receivership early in the year following a volatility spike in the commercial and wealth deposit segments due to high concentrations of uninsured deposits.
To some degree, all deposits are volatile, which presents a risk of large outflows in the short term. To protect against such events, Treasury holds cash and other high-quality liquid assets (HQLA) on the bank’s balance sheet, along with other sources of stress liquidity such as committed facility draws.
Heading into a falling-rate environment, in which rising deposit costs and declining variable-rate loan pricing are expected to intensify margin pressures, banks would do well to optimize the amount of cash and HQLA on their balance sheets because of the drag these assets can have on profitability. This requires solid analytics around the short-term liquidity value of deposits and a mechanism to signal it to the deposit-gathering businesses.
Even though the importance of stressed-liquidity analytics for deposits has been well understood for years, the intelligence many banks bring to the discipline is more art than science. Curinos has seen numerous opportunities for improvement, no matter how mature the practices may be. Best-in-class analytics can address major customer/client concentration risk. It can also apply segmentation techniques, driven by machine learning, to sort through large troves of data on customers’ product penetration, usage and demographics to reveal volatility across the differing types of customers or clients.
But a word of caution: These studies can’t be relegated to the shelves as academic theory to be dusted off only during one-time events. The information needs to be used continually to drive a more optimal balance sheet structure that can combat the margin pressures exerted by expected rate decreases. Specifically, contingent liquidity costs should be allocated through FTP, ideally in a way that connects costs with those customer/client segments whose deposit behaviors create the contingent liquidity costs in the first place.
Banks that don’t allocate such costs should strongly consider doing so, at a minimum to the non-relationship deposit segments with the highest volatility. Institutions that allocate costs as a single line item at the aggregate business-line level should evolve to more granular, segment-based cost allocation to better influence customer/client behavior. Those that deliberately withhold such cost allocation to spur growth of a strategically core customer/client type should ensure that the funding economics are transparent through “shadow P&L” reporting — this will ensure that balance sheet costs are communicated even if such costs remain in Treasury rather than being allocated to the business.
The ultimate goal of any funding evaluation should be to apply analytics to assess deposit quality. Beta, WAL and short-term contingent liquidity analytics can be used to not only better manage treasury-centric interest-rate and liquidity risk, but also to optimize funding strategy. Doing so will empower ALCO to make smarter, more strategic incremental-funding decisions to optimize performance across business lines. The rate cycle’s next chapter is fast approaching, so banks need to build this muscle quickly.
