top of page
• William Webster

# Basis Risk Appetite

Updated: Jan 13

In the beginning

This story goes back to 2008, that’s when bank deposit rates (LIBOR) and Base rate (Bank rate) both went South but not together and it took a lot of firms by surprise.

One Building Society I worked with at the time lost out from tracker mortgages another gained from LIBOR linked bonds. These windfalls arose from a game of luck; not that we saw it like that at the time. Soon the regulator was interested, because for some, their losses threatened viability. This was basis risk - where different reference interest rates altered the net interest margin.

This risk is still very much alive, so how do we form judgement on what our risk appetite should be? I’ll try and answer this by considering three things:

1.    What basis risk looks like

2.    When does it, (basis risk), matter?

3.    How much control do we really have?

What basis risk looks like

I’ll assume there are four types of interest rate, Administered, Base, Libor (soon to be overnight index rate) and Fixed. For a building society or bank the simplest balance sheet will consist of all lending (assets) and borrowing (liabilities) to be of the administered rate type - where you determine the rate.

A £100m balance sheet of this nature could look like this:

Asset             Liability        Net

This would appear to have no basis risk but, as we will see later, it may or may not be profitable.

Now let’s add some fixed rate lending, (say £50m):

Asset            Liability        Net

Fixed                       £50m                               £50m

Total                                                                £100m

Now we have mismatches which can be shown as a ratio:

The net figure (if all mismatches treated as positive numbers) is £100m, compare this with the savings liabilities of £100m and we get 100/100 = 1.00 or 100%.

Let’s suppose we hedge the interest rate risk by taking out swaps paying fixed and receiving OIS (successor to LIBOR):

Asset            Liability         Net

Fixed                       £50m            £50m           £0m

OIS                          £50m                                £50m

Total                                                                 £100m

The mismatch is still 100/100 = 1 or 100%

What you’ve done is to convert fixed interest received to an overnight rate, in theory interest risk is mitigated.

Finally, because the Bank of England is a safe place to put money let’s take £25m from administered lending and park it in a Reserve account, the picture is now:

Asset              Liability        Net

Fixed                       £50m             £50m             £0m

OIS                         £50m                                   £50m

Bank rate                £25m                                   £25m

Total                                                                   £150m

The mismatch is now 150/100 = 1.50 or 150%

What does this indicate?

As we move away from administered rate balance sheets basis risk increases.

To the extent that the interest basis themselves move independently there will be an unknown future effect on margins and therefore profitability.

It’s something the Board will see in its colour coded information and presumably in our example it’s moved from green, to amber and then to red. Amber and red frequently being the preserve of ratios exceeding one.

When does it (basis risk) matter?

For most Boards it matters when your basis risk exceeds that of your peer group because you are an “outlier” and therefore you attract regulatory attention. This means that from a regulatory “peace of mind” perspective keeping the basis risk down can be an important point in the psychology of risk appetite.

But this raises a dilemma, if customers don’t want administered rate products you need to offer them something else and thereby basis risk will start to increase.

Restricting basis risk could put you in a less competitive position, the corollary of which, is that running basis risk is linked to the margins you earn and leads to the question, “should we limit this risk?”

Risk appetites seem to vary, nevertheless I like to link them to considerations of how much you can afford to lose. This in turn means making comparisons with things like:

a.    Your profitability (past, present and future)

b.    Your reserves or capital strength

d.    Your expertise – the depth of knowledge and skill you have in dealing with this risk

A straightforward way to look at this is to find out how much your earnings are affected should the basis move a little against you. For example, if Base rate moves up/down by 0.01% and all other rates remain the same, what’s the impact on (a), (b) and (c) above? This isn’t perfect but it starts to give you a “feel” for what’s at stake and lends itself to scenarios.

It’s also worth considering how far into the future your basis risk exists. Whilst beyond the scope of this article long duration assets or liabilities increase uncertainty and risk. The only way to reduce this is to limit the extent to which you accumulate long dated positions on the balance sheet.

How much control do we really have?

Probably not as much as you wish for! It’s what makes this business challenging in the current environment. Try this question:

“If we suffered a loss and needed to rebuild capital how could this be achieved without reducing the credit quality of our assets or by taking on additional interest rate risk?”

The answer will probably lie in your ability to alter Administered rates, ie charge more for mortgages and offer less attractive savings rates. This can be done only within the competitive conditions of the retail savings and lending market you operate in.

It means that Administered rates are rates where you have some degree of control and offers an important distinction between administered rates and “others”.

Changes to non-administered rates - fixed, LIBOR/OIS and base rates - are outside your control and they involve spreads or relationships that can move for or against you. In this respect, for most, such mismatches offer games of luck not skill.

We should therefore think of administered rates as offering some degree of protection in margin adjustment. Whilst they don’t guarantee profitability, (that’s more a function of the competitive position), it’s something you have expertise in, and this allows for a more nuanced judgement of risk and reward.

If you decide that you need to offer non-administered rate products, (as will often be the case), you will inevitably incur more basis risk. That’s when you need to be clear that risk and price are aligned and that a margin of safety is built in – is there enough margin built in to absorb the basis risk you are taking?

Using mismatches in base, Libor and fixed rates to sustain margins is not a comfortable place to be, that’s unless you are convinced you have better knowledge of capital market and central bank behaviour than market participants.

It all comes back to taking risk where you have expertise and pricing that risk into the products you offer - something the regulator will certainly want to know more about if you are an outlier.

Summary

• Basis risk goes back decades, 2008 is a good starting point;

• Basis risk is a mismatch in reference points for interest paid and received;

• All but the simplest firms will incur this risk;

• As basis risk gets larger it matters and simple measures can help you understand the potential cost;

• This needs to be linked to how much you can afford to lose;

• Administered rates offer you some degree of control and utilise expertise;

• Other interest basis are outside your control and their relationships can change unpredictably leading to windfall gains or losses;

• This leads to a game of luck;

• If you do take basis risk price in a margin of safety.

2 views

See All

bottom of page