The nonsense of pension fund accounting

Last October, my colleague Tom Dobell wrote about the absurdity of how companies are forced to treat their long-term pension liabilities.

After another couple of reporting periods characterised by falling discount rates, I wanted to lay out in a slightly more mathematical fashion why I think Tom is right and the accountants aren’t. . . as well as highlighting the impact a seemingly small difference in the discount rate makes.

To do this I will use a fictitious company pension scheme with £2bn of liabilities, £1.5bn of assets and, therefore, a £500m deficit. It is worth understanding where these numbers come from. The asset side of the equation is easy – it is simply the market value of the scheme’s investments, be they stocks, bonds, cash, property or whatever. It is the liabilities that cause the problems. The £2bn is the ‘present value’ of all payments the scheme will have to make to members over its life.

For those unfamiliar with ‘time value of money’ calculations the concept of present value can be a little daunting. The easiest way of thinking about it is the value you would need today to make the promised payments and end up with nothing, given an assumed rate of interest. In other words, in each year you will pay some cash out, but build some back up in interest. The nub of the whole problem is the rate of interest (known as the ‘discount rate’) that you assume.

In this case I have made up a payment profile over 40 years as shown in the chart below:

Although the total value of payments over 40 years is about £3.2bn, the present value is only £2bn once you apply the discount rate of 2.5%. This is based on corporate bond returns and is typical of current assumptions.

The deficit is simply the difference between the assets and liabilities – in this case £500m. If this were a real pension scheme the company responsible for it would have to make significant additional payments in order to close up the deficit over the next few years.

The underlying (flawed) assumption in this calculation is that scheme assets will be invested in corporate bonds. If a more flexible approach is taken instead you get a very different outcome.

Let’s assume that the assets are invested 45% in equities, 25% in corporate bonds and 30% in cash with long-term returns assumed to be 9.5%*, 2.5% and 0.5% respectively. If we run this scenario out for 40 years we find that, far from running out of money, the scheme has over £750m of assets remaining after making its last payment. The other way of looking at it is that the liabilities are only worth £1.3bn when discounted at the blended asset return of just over 5%.

It is easy to dismiss all of this as arcane accounting noise that has little bearing on reality. After all, the members of these schemes are not paid any more or less as a result of differing accounting treatments so in theory the companies sponsoring them should pay out the same either way. The problem is that companies are expected to make ‘catch-up’ payments to schemes while they show a deficit. This drains capital from the business with results ranging from under-investment to company failure.

From an investor’s point of view the lesson is to think carefully about the underlying economics of pension deficits. The market is quick to treat these figures as equivalent to debt, but as we have seen above that can be very misleading.

* The MSCI world returned 9.7% p.a. between 1970 and 2016