The Art of Marginal Cost Analysis and Assessing Your True Cost of Funds
Part Two: Defensive Funding Strategies

“If it doesn't make sense, you should find for the defense.” –Johnnie Cochran

In the Q3 2008 issue of What Counts, we detailed “Marginal Cost 101”—the art of determining the incremental cost of a deposit. We also applied this analysis to the development of offensive strategies, specifically the introduction of a new high-yield checking account. We learned how to create a break-even pricing analysis of the new product using advances as a wholesale funding benchmark. We learned that the ultimate pricing structure depends heavily on: (1) how much new money would be raised from new depositors, and (2) how much cannibalization of current depositors might occur with the new product. Respectively, we want more and less. Three months ago, we speculated that some of this new money might be derived from money market depositors who may have been thinking about the high expenses relative to yields of money market funds, not to mention the gathering storm on the credit front. At that time, who would have thought of a world in which money market funds would be accorded the same government guarantee as bank deposits?! Guarantees on money market funds are designed to be temporary; still, consumers will likely have long memories of these funds “breaking the buck.”

With unprecedented, global central-bank intervention and coordinated rate cuts, short-term interest rates appear to be getting as close to street-level as they can. In this environment, marginal-cost-driven pricing discipline is always beneficial when issuers of lower credit quality try and preserve deposits with higher-than-market rates (notwithstanding temporarily available government guarantees).

Matching the Competition vs. Lowering Deposit Rates More Aggressively
Let’s assume that quite a few depositors remain shell-shocked, and in spite of temporarily available government guarantees, there is a strong bias to deposit with strongly capitalized institutions, even if it means sacrificing some yield. With the Fed continuing to ease, you believe that rates are heading down, yet you know that some of your competitors may not be reducing their rates on a one-for-one basis with Fed cuts.

You could find yourself in the situation depicted below, offering your CDs at rates that are 25 basis points below the Seattle Bank’s advances with a comparable maturity. We’ll use the respective figures of 3.25% and 3.50%. We’ll also assume that you would like to follow the Fed on the way down on a one-for-one basis and court customers that favor your institution’s reputation, with a bias toward capital preservation rather than above-market yield. For this analysis, we’ll further assume that six-month CDs comprise $10 million of your funding base. Now, let’s attempt to get answers to the question: “Can I get away with dropping my deposit rates at a faster rate than a competitor in a declining interest rate environment?”

Table 1 indicates that your competitor would like to drop deposit rates by only 20 basis points after a hypothetical 25-basis-point Fed ease. Why would they want to do this? Perhaps they are sustaining credit issues and desire to preserve deposits, irrespective of the cost. On the other hand, you would like to drop rates by a full 25 basis points—in line with the Fed.

Table 1. Hypothetical Current and Prospective Six-Month CD Pricing Levels

Funding Source Current Prospective
Seattle Bank 6-Month Advance Rate 3.50% 3.25%
Your 6-Month CD Rate 3.25% 3.00%
Competition’s 6-Month CD Rate 3.25% 3.05%

What’s the Cost of Dropping Deposit Rates at a Faster Rate than a Competitor?
If you elect to lower deposit rates in a greater proportional amount than your competitor, how much deposit retention might you expect? If you priced below your competition, you might sustain a moderate degree of retention—let’s assume 80% and that your depositors are not capital-preservation-oriented and are not rate-sensitive or “yield hungry.” Alternatively, as we will see, an assumption of much lower retention levels would imply rate sensitivity and perhaps relative indifference to underlying credit quality (irrespective of government guarantees). In the case of a say, 50% rate of deposit retention, your total deposit balances would drop by $5,000,000. Your cost of dropping rates at a level too far from the competition would be represented by your need to replace any lost funds through alternative sources, ideally at the lowest marginal cost.

What’s the Cost of Dropping Rates in Lock-Step with Your Competitor, and Not at a Faster Rate?
Let’s give an 80% retention scenario a look. There are two costs associated with matching the competition’s response to a rate drop that is less-than-lock step with an overall market rate cut. First, you would need to pay five basis points above projected levels to preserve the remaining balance of $8.0 million ($4,000). You would also need to pay an opportunity cost of retaining the $2.0 million at 3.25% (the prevailing wholesale funding rate) ($65,000). So, your total cost of proceeding with a strategy to drop rates by five basis points, rather than by the full 25 basis points, is $69,000.

Any pricing changes you make would represent the combined incremental or marginal changes that result from either arriving or departing deposits. If you decide to match your competitor’s measured response to a rate cut, you would need to identify the increase in total deposit costs relative to the retained funds that otherwise would have been lost. Here’s the formula:

Marginal Cost of Not Dropping Rates in Lock-Step with the Market and Matching a Competitor’s Measured Response to a Rate Cut = ((Incremental Cost of Preserving Remaining Balance) + (Opportunity Cost of Retaining Balances) / (Forecast Reduction in Balances if Deposit Rates are Reduced beyond Competitor’s Levels))

In our particular example of 80% retention, the marginal cost of dropping our CD rate by only 20 basis points, instead of 25 basis points (i.e., matching the competition as rates drop, but not lowering as much as we otherwise might) would be: ($4,000) + ($65,000) / ($2,000,000) = 3.45%.

Suffice to say, a marginal cost of 3.45% is significantly higher than your incremental source of funding (e.g., the pending Seattle Bank advance rate of 3.25%). Why so high? Because 20% of your depositors would have bolted, and you unnecessarily gave up being able to fund at five basis points less from 80% of your depositors. Under this attrition assumption, there’s no need to fear the competitor’s drop of only five basis points. Why chase the competition with this loyal a deposit base? Go ahead and drop your rates by 25 basis points under this scenario.

Back to the $64,000 Question
At what retention rate does it make sense to match the competition (dropping rates by 20 basis points, rather than taking advantage of a full wholesale rate drop)? Your answer is directly dependent upon one observation: when electing to match the competition at the expense of unnecessarily paying depositors that would never have left you in the first place, the greater the proportion of non-rate-sensitive money to rate-sensitive money, the higher the marginal cost of attracting new funds.

Let’s now assume that a much higher percentage of your depositors are rate-sensitive and that by reducing your deposit rates at a faster rate than your competition, you would expect a much lower rate of retention, say 50%. In this case, you would continue to calculate the marginal cost of matching the competition and foregoing an incremental savings of an additional five basis points during a rate drop:

Incremental Cost of Preserving Remaining Balance = ($5,000,000 * .0005) = $2,500 Opportunity Cost of Retaining Balance = ($5,000,000 * .0325) = $165,000 Forecast Runoff in Balances if Full 25 bps CD Pricing Cut: (.50*$10,000,000) = $5,000,000 Marginal Cost of Dropping Rates at Competitor’s Level: ($167,500) / $5,000,000) = .0330

So, if you were faced with losing 50% of deposits as a result of failing to drop rates by only 20 basis points, instead of the 25 basis points that the wholesale markets were dictating, the marginal cost of matching your competitor’s strategy would be 3.30%. At this attrition rate, the marginal cost remains only a tad higher than the cost of incremental funding (e.g., the current six-month Seattle Bank advance rate of 3.25%). At this level, perhaps the case for failing to match your competitor and offering a CD product at a cheaper rate would be less compelling. Table 2 indicates that the excessive marginal cost of not following the competition becomes far less pronounced in the vicinity of retention levels of 50%.

Table 2. Hypothetical Sensitivity of Runoff Assumptions vs. Marginal Cost

Estimated Retention % Estimated Retention $ Marginal Cost of Matching Competitor’s Pricing Strategy Decision
80% $8,000,000 3.45% Drop 25 bps
70% $7,000,000 3.37% Drop 25 bps
60% $6,000,000 3.33% Drop 25 bps
50% $5,000,000 3.30% Indifference / Drop by only 20 bps and match competition
40% $4,000,000 3.28% Drop by only 20 bps and match competition
30% $3,000,000 3.27% Drop by only 20 bps and match competition
20% $2,000,000 3.26% Indifference / Drop by only 20 bps and match competition

The bottom line: If you believe that you have a deposit base that is relatively non-rate-sensitive, why match a competitor that refuses to match deposit rates during a downward market movement? In this example, if your deposit retention rate was 80%, at the margin, you would have saved 20 basis points by not jumping the gun and matching your competition.

Estimating Your Attrition Rate
As we discussed, in a low-rate environment, deciding whether or not to match the competition would depend on your comparison of the marginal cost of matching the competition with the cost of alternative incremental funding (i.e., the wholesale advance rate). The wild card in the process is estimating your deposit retention rate—in this case, how many depositors you would lose if you dropped the CD rate by 25 basis points, instead of your competitor’s drop of only 20 basis points? This is a complex and inexact process, but there are several variables that you can use to help you estimate the degree of rate sensitivity within your deposit base.

  • Types of deposits. CD holders are often more rate sensitive than holders of MMDAs, DDAs, and other products without defined maturity dates and that have “sticky” features, such as direct deposit or online banking.
  • Length of depositor relationship. Generally speaking, the length of a customer relationship is inversely correlated with rate sensitivity.
  • Historical frequency of roll-overs at previous or lower-than-previous rates. Past propensity to rollover at previous rates during historical rate rises indicates less rate sensitivity.

Historical roll-overs can be monitored by tracking specific deposit balances by product type, along with their corresponding rate histories during periods of varying rate stability. Local market assumptions, which can vary greatly, can be assisted with the use of external econometric models.

Remember the Name of the Game: Minimize Your Funding Costs at the Margin
Whether it boils down to effectively pricing a new promotional deposit product or effectively addressing the pricing moves of a competitor during uncertain times, the name of the game is to minimize your cost of funds at the margin. You can do this by ensuring that key decision makers within your institution understand that the acts of simply matching competitors’ strategies and using the concept of average costing lead to suboptimal profits. Finally, it pays to effectively estimate the likely behavior of your rate-sensitive and non-rate-sensitive customers.

By John P. Biestman CFA, Vice President/Director of Business Development