When you see an engineer performing a stress analysis with a CAD program, it’s all thanks to Robert Hooke, a little-known 17th-century scientist.
He devised the eponymous Hooke’s Law, which states that materials bend in proportion to the stress, or pressure, applied. It’s an empirical law (a posh way of saying he made it up) but backed up by lots of experimental evidence. It doesn’t work in all situations, so engineers have devised supplementary models to fill the gaps.
I like empirical laws; give me something lacking theoretical foundations that actually works over an intellectually pure nonsense any day. Monetary theory, and for that matter most economic theories, could be cases in point.
All this came to mind when I was thinking recently about discount rates, how they are devised and where they are used. At least 12 IFRS Accounting Standards explicitly require the application of a discount rate, including pensions, insurance, leases, provisions and intangible assets. The choice of discount rate can have a major impact on profits, assets and liabilities.
Plethora of approaches
A non-accountant might therefore imagine that there would be coherent, consistent and robust guidance on how to choose the ‘right’ discount rate. The non-accountant would be disappointed.
I have seen companies use different discount rates without really explaining why
Probably the most prescriptive guidance is for pension liabilities, which require the use of ‘high-quality’ (typically AA-rated) corporate bonds of appropriate duration. This works quite well in most currencies, but can give strange results when the bond market is stressed or over-exuberant. Back in 2020, the yields on such bonds fell to under 2%, which caused liabilities to blow out in a way that made little economic sense.
Most of the other standards require some element of judgment, which could be interpreted by a non-accountant as ‘we made it all up’. I have seen individual companies use very different discount rates for ostensibly similar long-dated liabilities without really explaining why the different rates were chosen. The differences can be even greater for asset impairment tests, where the rate has to take account of the solidity or flakiness of the projected income streams.
Investors deserve a better and more transparent process
One of my fundamental takes on accounting is that companies should not be able to mark their own homework. Giving them considerable latitude to choose discount rates goes against this principle.
A key aim of the International Accounting Standard Board’s Conceptual Framework is to ‘contribute to transparency by enhancing the international comparability and quality of financial information’. Financial information will not be comparable if similar companies with similar long-term assets and liabilities use different discount rates.
Pick a team!
The whole concept of discount rates is based on the time value of money. Being given £100 today is inherently more appealing than being given £100 in 10 years’ time. This is intuitively obvious, but what is not obvious is how to choose the ‘right’ discount.
Should it be based on government bonds, often used as a proxy for a risk-free rate? Or high-quality corporate bonds? Or projected stock market returns? And how do we adjust for commercial risk such as debtor insolvency or disappointing returns? IFRS is largely silent on these points.
I would also argue that the whole idea of a risk-free rate is flawed. Government bonds are not risk free, as anyone who watched the Liz Truss car crash can attest. Around 80 countries defaulted in the 20th century and around 25 have defaulted so far in this one.
If we could reincarnate Hooke, I think he would laugh at all the rickety maths used to ‘calculate’ discount rates and immediately grasp that these are all just convenient fictions. Investors deserve a better and more transparent process.
