If you don't have one, you need to be asking, "What if"

Interest Rate Risk Management is Back in US Banking

In my early career I worked as an ALM manager for a regional savings bank experiencing high growth through acquisition.?It was in the 1990’s and the ALM function was focused on ensuring that as rates rose and fell, and they did, that the bank would be positioned to continue an acceptable net interest margin throughout the cycle.?The process included a lot of modeling and often a lot of guesses as to how the balance sheet would react to changes in the rate environment. ?It was during that period in history that “refinancing” became an everyday word in American households. ?Even as bankers, largely responsible for fueling the trend, we were shocked at the level to which the process was adopted by households. ?As the bank grew the ALCO became more astute in its understanding price/volume relationships, where and how to market products at the right price, and how to make decisions that did not “bet the bank”.

If you move forward twenty years to the post global banking crisis period, the US experienced a prolonged period of near zero interest rates.?A quick look at the Fed Funds rate shows us a range of between .05% and .65% over that period from late 2009 through mid-2016. ?From this point on there was a very gradual period of rising rates to about 2.5% that took place until the pandemic reset rates to near zero levels again.?Over this 12-year period, setting rates for the retail deposit base had been a question of how many basis points from zero is best.

?A generation of low rates caused dramatic changes to the liability side of bank balance sheets, particularly in retail funding.?There is now a generation of younger risk managers that may be unfamiliar with navigating through the choppier waters of fast-moving interest rates.?At a recent conference, I spoke to a thirty-year bank risk management executive and asked her what her key post-pandemic concern was for the banking industry.?She stated that after the financial crisis, so much focus had been given to regulatory compliance processes, that much of the management process of asking, “what if” had been abandoned at many banks.?Therefore, the theme of this paper is in looking at what could happen beyond the current consensus view and reminiscing on how a world of real interest rate risk was managed when we lived with uncertainty every day. ?Hopefully, it benefits some who may not remember it.

Dramatic Shifts in Retail Funding

In the long near-zero rate period, US banks saw the death of the retail CD as a significant product.?To illustrate changes between then a now, I looked at the balance sheet of a regional bank.?The chart below shows that shift.

The death of the CD as a major funding product for banks occurred naturally in a zero-rate environment.??Banks could not pay meaningful interest rates and consumers had no desire to lock up money for little return.??The focus here should not be on the product but rather on consumer desire to get the best return as they navigate the new financial reality of inflation. Looking at customer real return and their potential behavior if rates do not fall shortly is a key, “what if” for retail banks. ?Inflation has been with us for over a year, and although now falling, looks to be with us for much of 2023 at significant levels. ?The question is one of how reactive consumers will be to better alternatives to earn a higher yield. As the chart indicates, last year consumers had a real return on savings of -6% or worse. ?Banks should really be asking themselves at what point do consumers start voting with their feet should 2023 end up in a different position than the consensus.

Riding the Consensus at Your Peril

In the past month, I’ve seen several articles by firms on the sell side.?The consensus seems to be that banks should be extending their asset durations now as the Fed is nearly done its rate rise cycle.?The current treasury yield curve certainly endorses this sentiment as it is inverted which now makes going long for that yield harder. ?While I understand this strategy, I wonder if banks are looking at the “what-ifs” about what happens if the consensus is incorrect.?Remember, recent economic consensus included statements such as, “transitory inflation”.?I also remember that in run up to the “Great Recession” of 2007-2009, the consensus among bankers was that there would be a “soft landing” which of course saw the near complete collapse of the global banking system.?So, consensus thinking is often incorrect and planning on alternative scenarios is critical.?In fact, it is really the core function of the interest rate risk manager.

The age-old bank strategy of lagging deposit rate increases as far behind market rate increases for as long and at the greatest magnitude possible is the current practice for most banks. Indeed, most of the top 20 banks are paying existing savings account customers less than .2% interest.?The questions banks should be asking of their models now include:

1.?????What if inflation proves not to be under control or as importantly what if there are new doubts that it is subsiding? ??After a long period of inflation during the late 1970’s and early 1980’s, inflation appeared to be settling at the 3-4% level by 1984. ?These were very attractive levels as compared to previous years. ?Yet fear of inflation caused the 10-year treasury to peak at 13.8% during 1984.

2.?????What are other banks doing???Currently, there are a number of high growth banks offering between 3 and 3.5% on savings accounts and over 4.6% on CDS. Clearly, these banks see the opportunity to attract new retail customers and replace wholesale funding.??More importantly, how much will rate competition erode the funding base of those not playing the game???Modeling deposit erosion will be a real guessing game since many Millennials and Gen Z’s will not remember a time when you could earn meaningful interest. Will they become what we used to refer to as “rate tarts” in the 1990’s?

3.?????What is my price/volume relationship – i.e., what price do I need to offer on a product to minimize interest expense while ensuring my franchise value (on core deposits) does not erode? In other words, how do I optimize over a range of possible scenarios? This will involve guessing and experimentation.

4.?????If the non-consensus scenarios play out – what is the optimum asset side strategy given changes in the retail funding base?

5.?????Has the Available for Sale designation been used to the best interest of the bank over the full interest rate cycle??How can it improve??This is a question a number of banks must be asking themselves now.

Changes in Retail Funding are Already Happening

The FDIC tells us that during Q2 of 2022, insured deposits decreased in US banks by more than $350 billion. ?Much of this decrease results from the consumption by consumers and businesses of the funds pumped into their accounts from the pandemic relief programs.?So, there are less funds in the system.?At the same time, through Q2 more than $500 billion exited savings accounts but more than $75 billion flowed into time deposits.??With some banks offering rates above 4% now this trend is likely to accelerate and continue through the higher interest rate environment.

Preserving and Expanding the Franchise

The value of a bank is largely embedded in its deposits. ?It is not uncommon for acquiring banks to pay 10% or more premium on the deposit balances. ?Of course, this franchise value is evaluated on the stability and rate sensitivity of the deposits acquired. ???If the consensus forecast plays out, banks will not need to struggle with re-evaluating what is core in their deposit base and what is “hot money” as we used to call it.??This is because the cycle will have been a very fast one and the impact of prolonged higher rates will not have to be dealt with in the retail funding base (i.e., it will be over before the repricing lag runs out).

?Should the consensus outlook of moderating rates and the rapid end of inflation be correct, we will end up in a soft-landing scenario. In such a scenario, banks will reverse their deposit pricing strategy, dropping rates as quickly as possible.?As the average savings rate is still about .2% in the larger banks, they will, in fact, have very little adjustment to make.?

?S&P reports that Q3 saw further deposit erosion in the top 100 US banks (by asset size) but margins widened and so the strategy has worked to date for most.?Banks extended their assets to capture higher yields and held deposit pricing at bay.?In the soft-landing scenario, deposit erosion to date will have been a minor casualty. Notably, however, during Q3 of 2022, the worst impacted of the top 100 banks lost 5.9% of their deposit base in the quarter.?Equally as notable, one bank that grew deposits by offering competitive rates, saw their NIM contract by 8pbs and their share price decline by 15%. The question for interest rate risk modelers in the bank is one of whether these were strategies implemented with the outcomes anticipated or not.

What if we haven’t peaked? ?What if the yield curve remains inverted for a protracted period of time.?In August of 1978 the curve inverted for 21 months and after just a few months inverted again for a further 11 months. What if oil returns to $150 or more per barrel and what if the rate hikes implemented to date manage to kill off growth and yet inflation stays with us??The point here is that if banks are not running such scenarios and taking a hard look at how they’ll cope with them, then they are lacking.

Conclusions and Some Tips from the 1990’s

If your organization is not asking the what if questions, modeling the deposit erosion that could occur if the rate environment becomes more challenging, and having a good hard look in each of these possible rate scenarios about the relationship between deposit rates and volume, then perhaps it’s time.?At a recent conference, I spoke with the head of ALM for a global bank.?I asked how they modeled the market value of their deposits (what was core, what time buckets do they sit in, what is the beta?) ??She replied they were modeled as very long term and very rate insensitive, as in their experience these core deposits had only ever grown.?

In the 1990’s we were disembarking from a fifteen-year roller coaster ride.?The runaway inflation and stagflation of the 1970’s into the 1980’s was becoming a thing of the past.?However, within a more moderate bandwidth, we managed the balance sheet through a lot of interest rate and consumer behavior shifts and learned some lessons that may be of use to young managers today.?

Some of these lessons learned are very basic. Since we've had no interest rate competition for a very long time, I thought they might be insightful for the new generation of modelers.

1.?????Modelers tend to look at NII over the next twelve months and use MV for longer term risk analysis.?In today’s uncertain markets, look at how key scenarios may impact your NII over two-three years.?It will provide valuable insights. It will tell you how decisions made today on both the asset and liability side impact your ability to navigate unexpected environments twelve and twenty-four months out.

2.?????Be realistic.?Ask what would motivate you and your peers as consumers. For instance, if you are paying .30% on your savings accounts, will raising that rate to .50% make any difference in retaining your existing customers??If the answer is no and it likely is, don’t do it.

3.?????Test the waters.?If you are considering offering a new product (for instance high rate CDs) and you want to know what impact it will have in eroding your low rate savings balances, test it on a limited basis by restricting it to a specific market or channel.?Be prepared to turn it off as quickly as you turned it on.

4.?????I hear endless stories of machine learning and artificial intelligence successes in banks.?Yet, I seem to be receiving the same inappropriate, il-targeted and cookie-cutter offers from banks as I did thirty years ago.?Modelers get to know your deposit customer base!?Who, from within those 100’s of thousands or millions of accounts is rate sensitive and who is not? ??Who is likely to have the funds to do more with you if have the right product and price? ?If you can operationalize this information, you can make the bank a safer and more profitable venture.

5.?????To prevent the cannibalization of your existing customer balances, consider going out of market to attract new customers.?Perhaps you are considering an expansion at some point.?Use the opportunity to attract new customers in this market.

6.?????Create new “high yield” products and switch complaining customers to these accounts as they complain.

7.?????If you are offering CDs at attractive rates and trying to attract new customers, offer them a higher rate if they move their checking ?or credit card account to you as well.

8.?????On the asset side, do not base your lending off your cost of funds.?Base it off sound financial principles.?You should not be doing loans priced below the alternative risk-free rate.

9.?????On the lending side, banks must work equally as hard to find and sell products that work for both the bank and the consumer.?For instance, banks have recently been re-invigorating sales of 10/1 ARMs that save consumers up to 1% over the 30-year fixed rate. In these times we must all think a little harder.

10.??Be very wary of purchasing any security with volatile cash flows at a premium.?Consider MBS pass throughs, CMO’s and the like that are being constructed off recent mortgages closing at 6.5% or higher.?When you model the effective yield on these using a 5-year cash flow, you have a nice return. When the prevailing mortgage rate falls back to 4% and the entire MBS is gone in a year, you’ll be carrying these at a negative yield.

11.??Make sure that you communicate.?If your assumptions are guesstimates, make sure that your ALCO knows and inputs into these estimates. Make sure that if you are experimenting with products, prices, or any other feature in the model because it is truly an unknown, that all parties understand this.?Risk management is not about having the crystal ball.?It’s about making things work in a world where there is no crystal ball!

At yet another conference I attended in 2022, one speaker opened his session with a slide that stated, “May you live in interesting times”.?In his opening remarks he stated that this phrase is believed to be a Chinese curse.?As risk managers, if we did not live in a world of volatility, changing economic climates, shifting labor markets and occasional pandemics, not to mention natural disasters, we would have nothing to model.?So, the curse is our reason for being.???

The Opinions in this Paper

This paper represents the views of the author only.

About the Author

Jim was educated in accounting and finance at Temple and Drexel Universities in Philadelphia. By the time his graduated from Temple he’d already been working in Treasury for several years at a local insurance company.?He went on to work in trading operations/accounting, and then treasury and ALM and funds management at several local banks.?The 1990’s was the era of analytical automation in bank treasury, and he worked with some of the earlier ALM models.?At a fast growing regional bank,?what-if was on the daily agenda of the c-suite so ALM modeler had a busy role!

In 1996 Jim was recruited to work for an ALM software firm in their London office.?As a consultant he got to see how processes worked at banks across the world.?Later, he would join Oracle as they expanded from database company into financial services application company.?As his career expanded, he looked after a team of sales consultants/industry experts embedded in Oracle’s largest bank clients.?Charged with helping clients shape their most strategic digital, operational improvement, financial and risk management, analytical and customer relationship management visions with Oracle, the experience was both broad and deep.?

Most recently, Jim returned to his first calling by joining Mirai Advisory.?Mirai is an ALM specialist consulting firm that worked with clients across the globe, helping to improve or close gaps in their ALM systems and processes.?In 2016, the firm decided they could do it better and built the first “from scratch” ALM tool in decades, designed to operate in the cloud, offer end-to-end capabilities and run lightning-fast simulations.?Jim is working to raise awareness of the brand as he believes it truly has the ability to make balance sheet management a more relevant, timely and profitable undertaking.

Contact Jim at:

[email protected]

+1 267 515 2079