The valuation of pension obligations can be an important component in determining the value of an equity investment. But should you include in your analysis the pension surplus or deficit based on the accounting liability or, as some argue, the lower actuarial ‘funding’ valuation?
It is all about the discount rate. The problem is that there are very different opinions about the appropriate rate for pension obligations and what measurement approach is most relevant for investors. We examine a view expressed by many, including BAE Systems.
There are essentially two ways to think about the discount rate for a pension liability. The rate used could either reflect the characteristics of the assets held by the pension fund that are designated to meet the liability or, alternatively, could purely reflect the characteristics of the liability itself. An actuarial funding valuation is often characterised by an asset-based discount rate. Such a valuation is most commonly used by actuaries as a basis for planning contributions to a pension fund. However, the accounting liability under IFRS and other accounting standards would generally be based on a bond yield, irrespective of the composition of assets held.
In the accounting measurement, the discount rate reflects the bond-like characteristics of a pension obligation. In effect, the company has borrowed from employees with the debt repayment being the pension payments in retirement. This means that the present obligation reported under IFRS does not take into account any expected return on assets in excess of that bond yield. However, this future expected investment gain is, in effect, included in the actuarial funding liability which, as a consequence, would be lower.
A recent question from an analyst to us about BAE Systems highlights the problem for investors – which liability should you include in a business and equity valuation?
Pension liability measurement: The case of BAE Systems
Like all companies reporting under IFRS, the BAE Systems pension liability in its financial statements complies with IAS 19. It also reports a separate actuarial funding valuation. When the management presented the most recent actuarial valuation of their defined benefit pensions in 2017 they announced a change in methodology. At the same time management indicated that, in their view, this revised funding valuation may be a more realistic measure than that presented in the financial statements. This view is not uncommon, we have often heard companies, and sometimes actuaries, complain that the accounting liability does not, in their opinion, reflect the ‘underlying economics’.
The BAE Systems 2017 annual report includes an explanation of the revised funding valuation methodology. This refers to the valuation being based on “prudent assumptions”, which might suggest to some that the valuation methodology itself is also conservative. However, in our opinion, this is not the case.
BAE Systems actuarial funding valuation
The change in funding valuation methodology involved the discount rate used. The switch is from a ‘risk-free plus’ approach used in the previous funding valuation (which is similar to that used in measuring the balance sheet amount) to a higher expected return on assets. The impact is to reduce the actuarial liability. In previous years, although the BAE Systems actuarial deficit was lower than the accounting deficit the difference was not that significant. However, this is no longer the case and it raises the question of what amount you should use in your analysis and valuations.
The date of the BAE Systems funding valuation does not coincide with the full or half year end so comparison with the accounting figures is difficult. However, the further extract below shows that the “funding deficit is approximately £3bn lower” than the accounting amount at the 31st December year end.
BAE Systems explanation of the actuarial (funding) valuation discount rate
BAE Systems does not say what funding valuation discount rate tit has used, but based on the disclosures of the gross liability, the IAS 19 discount rate and the liability duration, we estimate it to be about 1.2% higher than the accounting figure. This would primarily be due to a higher rate applicable to the equity assets because the expected return on the bond assets in the fund would not likely differ significantly from the IAS 19 rate. Considering the approximate 50% allocation of the fund to equities, this means that the assumed equity risk premium over the high quality corporate bond rate used in IAS 19 is about 2.4% which, at the time of the valuation, equates to an equity risk premium over the relevant government bond rate of about 3.2%.
The comment by the company that the assumptions used for the expected asset return are “prudent” could in one sense be true, as an equity risk premium of 3.2% may be considered by some to be on the conservative side. But the more important question is whether a funding valuation methodology is prudent in the first place. We do not believe it is and certainly do not think the resulting liability is suitable for use in equity valuations.
The most recent balance sheet pension deficit at 30th June 2018 is about £3.0bn. No separate funding valuation is given for this date but, if the difference between the funding and IAS 19 deficit is maintained, that would make the funding valuation about zero. This difference of about £3bn is significant in the context of a company with a market capitalisation (at the time of writing) of around £20bn.
Why you should not use the actuarial funding valuation
The idea behind an asset-based discount rate is that the liability is stated at an amount equivalent to the assets that are required to generate a sufficient sum to settle the obligation. Say, for example, a pension plan has a single payment in 10 years’ time of 100. If the company invests in a portfolio of securities today, which is expected to produce an annual income of, for example, 7.2%p.a., then discounting 100 at 7.2% gives that required investment; in this case exactly 50. If the actual assets held are, for example, just 40, then the deficit on an actuarial funding basis would be 10.
This all seems relatively obvious, and it does have some merit for planning the funding of a pension scheme, which is how BAE Systems correctly describes the purpose of the funding valuation above. However, an actuarial or funding valuation done in this way is, in our view, wholly unsuited to the measurement of pension liabilities for the purpose of valuing an equity investment in the sponsoring company. The problem, and we believe the reason why you should not use such actuarial values in equity analysis, is that it does not take account of the valuation consequences of investment risk.
Pension obligations are largely fixed. There is clearly some uncertainty regarding the eventual payment to retired employees due to uncertain factors such as inflation, salary increases and life expectancy. However, these factors are not linked to investment returns. Any investment risk taken in funding pensions is largely borne by the company and shareholders. It is not generally passed onto the employees. But why does this matter? It is basic financial economics … Continuing with the data above. Let’s say a company (or even an individual) borrows 68 at a rate of 4% (with interest compounded) and invests 50 out of this 68 in a portfolio of risky assets at an expected return on 7.2%. The loan of 68 will compound to an eventual cash outflow of 100 in 10 years, while the asset of 50 will be expected to grow at the higher asset return rate to also be 100, and hence can be used to repay the liability. But this seems to leave a ‘profit’, from the arrangement, of 18.
The problem is that this ‘profit’ is an illusion. While there may be such an expected gain in cash terms, it comes at a price, which is the added risk of the company itself due to the leveraged position it has put itself in. Higher risk means a higher cost of capital and lower valuation multiples, which will offset any ‘gain’ from the apparent expected cash flow advantage.
For BAE Systems the £3bn difference between their accounting and actuarial funding valuations is their equivalent of the 18 ‘profit’ in our simple example above.
There is nothing intrinsically wrong with an actuarial funding valuation based on an expected asset return discount rate when used to plan the funding of a pension scheme. However, in our view, such a valuation does not reflect the present economic position of the fund.
Pension liabilities should be measured (for the purpose of equity valuation) in a way that reflects the bond-like characteristics of that liability and not based upon the expected return of assets that may be held in order to fund the repayment. For pension liabilities where investment risk is not shared with the employees (which is most defined benefit pension schemes) this means discounting at a rate that does not include a premium for investment risk. The accounting liability measured under IFRS would get close to the appropriate liability in most cases.
Insights for investors
- Do not let management persuade you that the IAS 19 pension liability is unrealistic and does not reflect the economics of pensions or that it should be ignored.
- Use the IAS 19 pension liability in your analysis or equity valuation models and not the so-called actuarial liability, particularly where this is based on an asset return discount rate.
- Investing in higher expected return but higher risk assets, such as equities, might reduce the expected cash flow needed to fund defined benefit pensions. However, it does not reduce the overall risk-adjusted expected cost of pension promises.
- Defined benefit pensions create multiple risks for sponsors. Check the fund asset allocation; if the assets do not match the liability then the added investment risk or leverage may curtail profitable risk taking in other parts of the business.
No advertising, no spam, just new articles. Unsubscribe anytime.
Or see suggested related articles below