Too good?
Selling mortgages with interest rates of between 1% and 4% may seem like an opportunity too good to miss, particularly now Bank rate is 0.10%, but is this too good to be true? Let’s see.
As a lender if you want to build a mortgage pipeline it’s straight forward, all you need to do is offer lower rates than the market. Every day the market asks you the question “do you want to deal?”
Four things will influence your decision, three of which you have no control over:
1. The current level of interest rates
2. Your cost of borrowing
3. The price competitors charge
The starting point is the cost of money, the Gilt yield curve, each bank then has an additional funding spread, a further margin is then added, and this gives a loan rate. How much business you then do is constrained by what your competitors offer.
To illustrate let’s see some numbers:
Gilt yield..................0.10%
Funding spread......0.60%
Margin.....................2.50%
Loan rate.................3.20%
The maximum gross margin you can earn is 2.50%, note this is gross and all your costs and expenses need to come out before you are looking at a net earnings figure.
Keeping things simple we will assume your competitors undertake a similar exercise, their costs are identical, but they alter their loan margin to 2.30% giving them a lending rate 20 basis points lower at 3.00%. What do you do?
A fourth factor now comes into play:
4. Discretion – you can now choose to re-price at a lower rate or walk away from the market.
What’s more this is the only point in the proceedings where you have control, the choice is entirely in your hands, do you want to add to your loan book or not?
Warren Buffett talks about a similar choice in stock selection – it’s like baseball or, as I prefer, cricket. The bowler offers the batsman a choice, play or leave and by choosing which ball to strike the batsman uses his skill to accumulate a high score. Picking the wrong ball leads to a long walk.
In this way mortgage lending is surely no different. Good loans give you small incremental returns, bad ones soon wipe them out. This prompts the question:
“What is a good loan?”
A good loan doesn’t really exist in isolation it’s more whether a portfolio or tranche of loans pays both interest and principal on time, something we only know with hindsight.
The important question therefore is:
“Are we sufficiently using our discretion to build loan books where the probability of a good outcome is in our favour?”
This is all about pricing credit risk properly. Whilst you can’t determine what may happen in the future you can see what the market is telling you about its expectations for credit losses. It does this by adjusting the “credit spread” (the margin in our earlier example).
If your margin is 2.50% each £100m lent will earn £2.5m per year, if you assumed a recovery rate of nil this margin would cover credit losses of £2.5m. Similarly, a 50% recovery rate would allow £5.0m to default and so on.
There is an interplay between credit spreads, recovery rates and default rates (if you know two you can derive the third) and insight into what credit spreads tell you about how the market prices default is valuable information.
This introduces the idea that a loan doesn’t have one “fair-value” but a range where value may be fair. This isn’t an exact science, it’s an approach whereby you can look at the market and see whether you think you are being adequately compensated for the risk incurred and then add a “margin for safety”. This won’t stop defaults from happening but does use skill to tip the probability in your favour and that’s what playing games is about - knowing when to strike and when to walk away - the cornerstone of competitive edge.
I don’t believe this needs to be complex after all we are dealing with approximations but without doubt there needs to be a process that gives you some comfort that decisions are sensible.
Does Covid change things?
I believe it does. There is a trap that’s easy to fall into. As interest rates have fallen the margin you can earn may have widened. On the face of it mortgage loans are more profitable. In cricketing terms, the fast bowler has been changed for a spinner. What looks tempting to the batsman could prove his undoing.
Until you understand the new conditions (after all the cut to interest rates was brought on by extreme economic uncertainty) your margin for safety needs to increase. This is the compelling reason why:
As I write the 30-year gilt yield is 0.74%, the implication for investors is that by investing in UK government debt they will lose money in real terms. How much? If you estimate CPI as 2.1% the real loss is 1.36% each year - £100 invested today will worth £66 in real terms in 30 years time.
Whilst Gilt investors could get further short-term gains, does a loss of 34% in real terms look like a great long-term investment? (To underline this point Gilts have a P/E ratio of 135 and offer no prospect of growth).
We know Gilt rates affect wider credit markets, dramatically lower Gilt yields make other credit risks look cheap but whether their cheapness is illusory depends on believing the default and loss levels that are baked into loan pricing compensate you for taking a swing or should I say selected shot?