Skip to main content

Looking for Valuant? You are in the right place!

Valuant is now Abrigo, giving you a single source to Manage Risk and Drive Growth

Make yourself at home – we hope you enjoy your new web experience.

Looking for DiCOM? You are in the right place!

DiCOM Software is now part of Abrigo, giving you a single source to Manage Risk and Drive Growth. Make yourself at home – we hope you enjoy your new web experience.

Funding Long-Term Assets on the Cheap

May 19, 2015
Read Time: 0 min

Industry “best practices” call for us to evaluate the profitability of a long-term fixed rate loan using Funds Transfer Pricing (FTP) techniques. Various points along the transfer curve provide a funding cost for the loan’s cash flows and repricing points. Once funding costs have been assigned to the loan’s cash flows a weighted funding cost can be calculated. The most commonly used FTP curves are the Swap curve (big banks) and the FHLB Bullet advance curve (smaller banks and most credit unions).

After adjusting for credit risk, option risk, and servicing cost, a spread can be calculated relative to the funding cost. That spread, when divided by the capital requirement assigned to the loan results in a risk-adjusted return on capital (RAROC) or ROE. If the ROE doesn’t meet the institution’s goal, the loan would normally be rejected or modified to make the loan more profitable.

Most small to medium sized institutions don’t do interest rate swaps, and only a small percentage of their funding is in the form of FHLB Advances. Non-maturity deposits make up the majority of these institutions’ funding.

Industry “best practices” also have us evaluate the behavior of non-maturity deposits (NMDs) by performing a core deposit study. The outputs from the core study are pricing betas, decay rates, and the portion of the balances that are surge balances. Outputs from core studies are most commonly used in modeling the institution’s interest rate risk. However, core studies also provide evidence that a portion of the NMDs studied represent a nearly ideal funding source for fixed-rate long-term assets with the added advantage of carrying a much lower interest rate than comparable points along the SWAP curve and FHLB Bullet Advance curve.

Using output from a core deposit study, NMD balances could be divided into three classes:

  • High Beta NMDs – Those whose rates move by a large percentage of changes in market rates, making them a less than ideal funding source for fixed-rate lending.
  • Surge Balances – When rates rise NMD surge balances are likely to jump back into CDs or leave the institution, forcing the institution to replace departing funds at market rates. In either case, the cost of surge funding will move by a large percentage of the change in market rates.
  • Low Beta Non-Surge NMDs – These balances behave like long term (based on decay rates) relatively fixed-rate (low betas) funding. Conceptually, this is an ideal funding source for fixed-rate lending.


Of course considering use of the above information in making balance sheet funding decisions raises the following issues:

  • Was the core study performed in a competent manner?
  • Do you believe in the results of your core study?
  • Will these accounts behave in the future as the study indicated they did in the past?
  • How will my regulator react to my use of core deposits to fund fixed-rate long-term assets?


The extent of the risk in the first three of the above questions can be examined by performing sensitivity testing on key assumptions coming out of the study. But wouldn’t it be nice if there was some way to hedge the risk inherent in the core deposit study results?

A bit earlier in this article, I identified the FHLB bullet advance curve as the most commonly used FTP curve in small to medium sized institution transfer pricing. Of course, today, the FHLBs offer a dizzying array of advance products aimed at hedging the interest rate risk and option risk in virtually any kind of asset you may wish to place on your balance sheet. Their behavior through a variety of rate environments is very predictable (unlike NMDs), but they can be expensive.

Let’s say I wanted to portfolio a 15 year fully amortizing commercial real estate loan yielding 4.50%. I choose a 15 year amortizing advance to hedge the interest rate risk in the loan at 2.68%. Gross spread is 182 bp. After adjusting for credit risk and servicing cost, the net spread is 102 bp. Divide that by a 10% capital requirement and you have a pre-tax ROE of 10.2%, below your 15% ROE target.

Note in the following two graphs (assuming no prepayments), cash flows from the loan and from the advance are virtually the same. Spread is virtually constant through the lifetime of the two instruments. Because amortizations are slightly different, a small amount of core funding is being used to plug the gap.






On the other hand, funding with low beta, non-surge MMDAs with a beta of 0.3 increases the gross spread to 420 bp, net spread to 340 bp and ROE to 34%. The blended cost in the following left graph is MMDA cost of 30 bp.

Take rates up 500 bp with the MMDA beta of 0.3 (middle graph) and the gross spread drops to 270 bp, net spread to 190 bp and ROE to 19%. Do a sensitivity test on the beta, taking it from 0.3 to 0.6 and gross spread drops to 120 bp, net spread to 40 bp and ROE to 4.0%. That’s too much risk for some peoples’ blood.





But what if we blended funding, 50% low beta MMDAs and 50% 15 year AMMO advances. We start with a gross spread of 300 bp, a net spread of 220 bp, and a ROE of 22%. Take rates up 500 bp and gross spread drops to 225 bp, net spread to 145 bp and ROE to 14.5%. Sensitivity test the MMDA beta taking it from 0.3 to 0.6 and gross spread drops to 150 bp, net spread to 70 bp and ROE to 7.0%.






The following graph compares the ROEs for the three different funding scenarios. The only environment where the institution fails to meet its 15% ROE goal in the MMDA and blended funding options scenarios is with a 500 bp rate increase accompanied by an increase in MMDA betas from 0.3 to 0.6.


If the 50/50 blend results represent too much risk for you, change the blend to a higher percentage of advances. For each step up (60/40, 70/30, etc.) you will give up spread in exchange for a reduction in risk. Did the addition of the advances to the MMDA funding mix perfectly hedge beta and beta assumption risk in the MMDAs? No. But financial institutions make money by taking on and managing risk. So we should be looking at risk/return tradeoffs. The 50/50 blend in the above example both moderates risk and increases return. That’s what we are going for.

Author: Tom Farin

About Abrigo

Abrigo enables U.S. financial institutions to support their communities through technology that fights financial crime, grows loans and deposits, and optimizes risk. Abrigo's platform centralizes the institution's data, creates a digital user experience, ensures compliance, and delivers efficiency for scale and profitable growth.

Make Big Things Happen.