Beware of geeks bearing formulas Banks, Risk Based Pricing Models & what next

Beware of geeks bearing formulas, the legendary investor, Warren Buffet, loves to say as he cautions against the allure of complexity and smarts in fashioning one’s investment play.

For students of economics like myself, Buffet challenges us to “strip away the guff and mathematical sophistry and figure out, as simply as one possibly can, what exactly works.”

In November 2019, after three years of misfiring and wreaking havoc in private sector lending, Kenya’s National Assembly repealed caps on lending rates, a move that sent banks dashing to the market regulator for Risk Based Pricing models like Romeo in the rush for Juliet’s balcony in William Shakespeare’s 1597 tragedy - Romeo & Juliet.

With rate caps out of the way and the Central Bank approving Risk Pricing Models from each of the 39 or so players, banks argued that the liquidity squeeze that had characterised the rate caps regime would unclench and they would be well positioned to gorge on higher risk, therefore infusing momentum into the private sector.

We could all finally hold hands and sing Kumbaya, or so we thought, but they say life always has a fat tail, which is to mean unexpected events that markedly deviate from the norm, and it is only when that fat tail crystallises that the cracks in the genius of geeks bearing formulas begin to manifest.

In February 2024, after about two years of yields in the domestic market ratcheting upward in an unrelenting fashion, the market hit what appeared to be an inflection point and yields began tumbling like bowling pins.

This came on the back of Kenya’s successful refinancing of the US$2.0 billion bullet maturity of its 2024 Eurobond note, the largest single maturity in the frontier market that year and one that had hogged the limelight amongst market watchers over amplified fears that Kenya was going to the proverbial hell in a handbasket with default in sight.

It also happened that this development coincided with the cooling off of inflation in the global environment and major central banks began unwinding from tight monetary policy, cuing frontier market central banks like Kenya’s to follow in tow and cut rates.

A chart plotting the Central Bank Benchmark Rate, the 91-Day T-Bill Rate & Commercial Banks’ Lending Rate between Jan 2023 and Dec 2024

It is this twin occurrence that began to chip away at the genius of geeks bearing formulas since commercial bank lending rates remained stubbornly sticky even as yields fell off the cliff to the basement and the Central Bank unwound its benchmark fairly fast.

There appeared to be a troublesome decoupling between the conduct of the Central Bank and Commercial Banks’ lending rates. That which would ordinarily have been expected to move in lockstep was clearly out of sync and what was worse, it left borrowers gasping on the chokehold of high lending rates.

In their defence, banks argued that the high yield environment had left them raking in bucketloads of expensive deposits and they risked stepping on the live wire that is mismatched pricing between their liabilities (deposits) and the assets they funded (loans) should they cut lending rates rapidly to mimic the Central Bank and tumbling yields.

In short, banks were saying that the Risk Based Pricing Models they tabled and the market regulator approved were overweight on the cost of funds and if they went cutting rates as fast as the Central Bank did, they’d be left holding the baby.

But the market regulator would fire back arguing that this was red herring by the lenders and in the Central Bank’s post-Monetary Policy Committee briefing in February 2025, the Governor went hammer and tongs reading the Riot Act at his licensees.

Central Bank of Kenya Governor, Dr. Kamau Thugge, during the February 2025 post-Monetary Policy Committee briefing with journalists

My view is that there is a strong basis for the argument tabled by the Governor of the Central Bank.

Four points to make in this regard:

1. The argument levelled by banks around expensive deposits and the overall impact on their cost of funds needs to be interrogated with appreciation of the fact that there is a split between retail and wholesale deposits with the former being largely short-term and therefore re-pricing for adjustments in yields shouldn’t be an extreme sport. The latest available data from the Central Bank would suggest to me that, by and large, there’s a skew towards retail deposits in the Kenyan market
2. The Cost of Funds component in Risk Based Pricing Models is anchored not on isolated deposits held by the bank by the weighted average of the consolidated pot of deposits. If such is the case and Kenya’s deposit landscape is skewed towards retail deposits, there are questions to be asked about the lethargy in aligning lending rates with the declining yields in the market
3. Perhaps the way to allay concerns emerging from No.1 & No.2 above, banks should provide visibility on the proportion of their deposits that is held under CASA (Current Accounts & Savings Accounts)
4. The fact that loan re-pricing going upwards (i.e, when yields are rising and the Central Bank is tightening) far outpaces loan re-pricing downwards (i.e, when yields are falling and the Central Bank is cutting rates suggests that there is, to a large extent, a margins protection exercise here by lenders

A snapshot of the commercial banks’ deposits landscape in Kenya as at close of December 2023

So what next for bank lending & Risk Based Pricing?

Through their umbrella body, banks have tabled a proposal to overhaul the existing Risk Based Pricing regime and adopt an alternative which is expected to yield better transmission of signals from the Central Bank.

Five important things to flag about the ask from banks:

1. They want a unified Base Rate applicable across the sector, a departure from the current regime where every bank applies its own base rate
2. They want this unified Base Rate to be anchored on the two-months Interbank Rate average
3. They want disclosure of the two-months Interbank Rate average would be part of the bi-monthly Monetary Policy Committee release
4. They want a six months long pilot period for monitoring & evaluation of the proposed approach
5. They want all new variable rate & local currency loans with a tenure in excess of 12.0 months to be priced off the proposed Risk Based model

Musings on proposals by bankers: Reflections from Jimi Hendrix

Transitioning to a unified industry rate anchored on the Interbank Rate is a welcome move.

First, it creates a better environment for clarity and comparability of pricing across players. The current regime where every bank has a Base Rate unique to itself calls to mind the great 1971 song Room full of mirrors by the legendary guitarist, song writer and singer Jimi Hendrix.

The opening line of that song is one I reflect on quite often.

I used to live in a room full of mirrors, all I could see was me. Then I take my spirit and smash my mirrors and now the whole world is here for me to see. I said the whole world is here for me to see & now I am searching for my love to be

With a common Base Rate, Kenyans, in the word of Jimi Hendrix, are now out of the room full of mirrors where all they could see was their respective banks. They now have the whole world before them to see and should be better placed to compare lending rates across banks.

To be clear, even currently there has been some effort to empower borrowers to compare pricing across banks through the regular publication of lending rates done by the Central Bank.

However, the key difference here is that in an environment where banks apply Base Rates unique to themselves, announcements around the margin by which cuts have been made lack context and information value since there is little, if any, public knowledge about the high from which they are cutting.

The most expensive bank could be slashing its Base Rate by the steepest margin yet still remain the most expensive making any celebration of the slash in its Base Rate a victory lap taken too early.

Second, anchoring the unified Base Rate on the two months Interbank Rate average is another welcome move for three important reasons:

1. This is not the first time Kenya is making an attempt at a Unified Banking Base Rate, between July 2014 and September 2016 there was the abortive attempt at what was dubbed the Kenya Banks Reference Rate. The Achille’s Heel in the Kenya Banks Reference Rate of July 2024 to September 2016 was the fact that it was anchored on the 91-Day T-Bill rate and in effect was susceptible to the distortions brought about by the government’s fiscal imprudence
2. Unlike the 91-Day T-Bill rate, the Interbank Rate (the rate at which banks borrow from one another on short-term basis) is a signal of where liquidity is playing at in the market. The market could be dry in which case the Interbank Rate will be high or it could be awash with liquidity in which case the Interbank Rate will be low. Whichever the case, the Central Bank through its operations could inject or mop up liquidity from the market with a view to aligning the Interbank Rate with its desired policy stance
3. A lot has been done to improve the functionality of the Interbank Market the culmination of which was the August 2023 introduction of a corridor within which it operates with a view to anchor interest rates along the Central Bank’s benchmark rate. Think of this as having a road on which vehicles keep meandering on and off track after-which you decide to put up guard rails and reflective road studs to guide the course of drivers. That’s the thinking behind the corridor

Central Bank of Kenya, Dr. Kamau Thugge, on Sep 18th, 2023 speaking about the anchoring of the Interbank Rate

I wish to end as I began - beware of geeks bearing formulas.

In opting for the use of the Interbank Rate as the anchor for the unified Banking Sector Base Rate, bankers have benchmarked against comparables in the advanced markets including the Secured Overnight Financing Rate (SOFR) and the Euro Interbank Offered Rate (EURIBOR).

Such Base Rates have not been without flaws, including the infamous 2012 London Interbank Offer Rate (LIBOR) scandal where a number of global financial institutions colluded to manipulate the rate to their favour. How do we safeguard against such?

Also, with banks proposing the loading of the cost of operations onto the Interbank Rate as part of the risk premium, will each bank be loading its cost or will the model provide that it is the cost of the best in class/most efficient entity that is loaded on?

In the event that it’s the former (i.e, each bank loading its own cost), how do we safeguard against the perverse outcome of banks pricing in their inefficiency onto borrowers?

As Mark Twain said, what gets us into trouble is not what we don’t know. It’s what we know for sure that just ain’t so. So let’s stay alert!