In the second in an occasional series looking at life through the eyes of a fictional CEO (see the previous one on acquisition accounting), our baffled leader tries to get his head around pension accounting with the help of his Crack Accounting Team (CAT).
CEO: OK, now I am in charge of a huge European industrial company, maybe it’s time for me to understand all these balance- sheet numbers.
CAT: We’re happy to help.
CEO: Let’s start with pensions. What is this ‘defined benefit obligation’ [DBO]? And why is it exactly €28,671m?
CAT: The DBO is the present value of all the pensions we expect to pay out until the last pensioner, or their dependant, dies.
CEO: That sounds like a really complicated calculation. How does it work?
CAT: Well, first you have to decide how long your employees are going to live. Luckily, actuaries publish tables of projected life expectancy.
There aren’t many suitable bonds with multi-decade maturities
CEO: And how accurate are these?
CAT: Historically, very inaccurate indeed. The actuarial profession was blindsided by the steady increase in life expectancy, as all their data is backward-looking.
CEO: So liabilities were understated for years?
CAT: Yes, although the rate of increase has slowed, and Covid didn’t help either. And then we have to estimate the salary of the employee when they retire, which could be decades away.
CEO: This is getting flakier.
CAT: Then we model marriage rates, divorce rates and whether the pensioner’s spouse [a male, in this example] will outlive him.
CEO: That sounds bizarre. Why?
CAT: Because the widow will draw a reduced pension after the husband dies – unless she dies first.
Inflation-proofed? That sounds scary
CEO: So if an employee marries a much younger woman, our effective liability suddenly increases?
CAT: Yes, unless she divorces him later. Plus we have to model inflation rates, as UK pensions are inflation-proofed until the pension starts being paid.
CEO: Inflation-proofed? That sounds scary.
CAT: It’s not as bad as it seems. The proofing is capped at 2.5% these days.
CEO: This is starting to remind me of the acquisition intangibles we talked about last time.
CAT: When we have finished all these very robust calculations, we turn it into a present value. But you have to use the right discount rate, or the whole thing is a nonsense.
CEO: And how do you select the discount rate?
CAT: We have to use a AA-rated corporate bond of similar duration.
CEO: What does that mean in English?
CAT: Duration is the average remaining life of our pensioners, weighted by the size of their pension, which is usually many decades. And we use a sample of AA-rated bonds with similar maturities.
You make a whole series of assumptions, then use a questionable discount rate to get a present value
CEO: Well, that sounds quite solid.
CAT: Er, no, not really, as there aren’t many suitable bonds with multi-decade maturities. And bonds don’t have listed prices, as there is no central exchange, unlike shares.
CEO: So you just make the discount rate up?
CAT: No, we use specialist advisers but with significant manual intervention.
CEO: My sense of déjà vu is increasing. You make a whole series of long-term assumptions, model them out for decades and then use a questionable discount rate to get a present value that goes into our accounts?
CAT: You’ve nailed it.
CEO: But at least the assets are easy to value? Right?
CAT: Er, no, not any more. We are increasingly investing in alternative assets like private equity, where the valuations are not market-based. For those, we use sophisticated valuation models with…
CEO: …significant manual intervention.
CAT: I think you’re being unfair; everyone does it like this.
CEO: One last question. Why does this end up as a number to five significant figures?
Silence.
