[Skip to content]

Sign up for our daily newsletter
The Actuary The magazine of the Institute & Faculty of Actuaries
.

Reconciling the warring factions

The new accounting standard covering retirement benefits, FRS17, is highlighting the long-running dispute between the financial economists and conventional actuarial wisdom. Until recently, the arguments have seemed academic, with little obvious effect on actual pensions practice in the UK. The issue became front page news in November 2001 when John Ralfe, head of corporate finance at Boots, used the arguments put forward by financial economists with great eloquence to explain and justify switching the £2.3bn Boots pensions scheme from a typical UK investment strategy of 75% equities to one composed entirely of high-quality corporate bonds (see p28).
I have approached the issues from the perspective of a practitioner, more concerned about what to advise my clients than about the niceties of the debate. The following is therefore my own personal reconciliation of the arguments. I believe both sides are ‘right’, the main areas of contradiction arising from the different perspectives of the protagonists.

Actuarial values
The actuarial orthodoxy taught to me was that defined benefit pensions liabilities should be valued on a set of long-term assumptions and the value of the assets then adjusted to an ‘actuarial value’ to be consistent with those liabilities. It was recognised that the value placed on the benefits would not be the ultimate cost of the benefits. That would depend on the actual experience.
The method’s main use was in allowing companies to budget in a smooth and sensible way from year to year. There was almost always a large discretionary element in the benefits, which meant that unfavourable experience could be absorbed without large cash calls on the company.
The method also allowed positively encouraged investment in assets such as equities, which have a volatile market value but are expected to provide higher returns than those assets with more certain cashflows, such as government gilts.
These actuarial valuations gave management and members the information they felt they needed. It also enabled a wide portion of the UK working population to participate in the equity growth of the 1980s and 1990s, the surpluses generated in the pension funds being converted into higher benefits for many members.
The method continued to be used notwithstanding the increased conversion of discretionary benefits to guarantees arising from legislative changes and benefit improvements. The result has been that this approach continues to be used today, even when the majority of benefits became guaranteed.
There has been a change to ‘market-based valuations’ in recent years, with an allowance for the equity risk premium (ERP). Although expressed in market value terms, the use of the ERP means that this method essentially retains the flexibility and subjectivity of the old actuarial values. Underlying it remains the belief that it is reasonable to assume that in the long term equities will outperform gilts and that it is reasonable for funds to be invested in equities to gain the benefit of that outperformance.

New line of thought
In 1997 Exley, Mehta, and Smith published an Institute paper ‘The financial theory of defined benefit schemes’ that challenged the actuarial orthodoxies and pointed out lots of awkward technical flaws in the ‘actuarial values’ methodology.
For me, the two key arguments they put forward were that:
– the value of defined benefit pension liabilities does not depend on how any assets are invested; and
– from a shareholder perspective, investing in equities does not add value.
They felt that the actuarial techniques in use at the time were opaque, in particular hiding the risks being run and who was bearing those risks. Also, they wanted actuaries to have the tools and the desire to answer the question ‘Does the defined benefit scheme add shareholder value?’

Advising companies and trustees
When a significant proportion of UK benefits was discretionary, then the benefit was linked to the performance of equities so there was a strong link between the value of the assets and the value of the liabilities.
Guaranteeing high benefits has broken the link between the assets and liabilities. Pension funds are no longer with-profits investment arrangements for employees, with only a low level of company guarantee. Members on 1/60ths accrual with guaranteed LPI are unlikely to get discretionary increases. The pension promises have become bond-like (unless inflation takes off again, which is possible but not something on which we should be relying).
As equities have greater risk, they should provide greater returns the famous equity risk premium. For budgeting, the company management may want to take this expected extra return into account, rather than risk putting too much money into the pension scheme. Trustees may also feel that the pension scheme is an appropriate way to expose its members to equities. Members, they can judge, would rather aim for risky high benefits than have lower benefits with certainty. Even if the equity risk does not pay off, the company may top up the benefits to the higher level in any case.
Pensions actuaries have generally been supportive of these actions by company management and trustees. The valuation techniques used have aided in this by illustrating the benefits of the ERP while being a bit hazy on the risks and who is bearing them. This is not necessarily a bad thing for the parties involved

But what about shareholders?
... except for shareholders, say the financial economists. Shareholders get a raw deal from the equities invested in the pension trust. They have to share the gains with members if the gamble is successful and pick up the full cost if the gamble fails. If shareholders have a range of investments, they would be better off investing those with more certain return in the pension scheme and investing in equities somewhere where they get the full value of any gain. (The current route is also tax-inefficient.)
The pension scheme is a set of assets and liabilities of the company. If the liabilities are all guaranteed, and so do not depend on the performance of the assets, then their value is not linked to the assets. Investing in equities does not reduce their value unless you believe the company will be able to renege on its pensions promises if equities perform badly.
More fundamentally, £100 of equities is worth the same as £100 of bonds. From a shareholder’s perspective it is not the job of a normal company to invest in a diversified portfolio of equities of other companies. The company is not adding value for the shareholder by doing so. Worse, it is risking the core business by exposing the business to the performance of other companies. For some small companies with large pension schemes, this gearing is extreme.
At a macroeconomic level, history says equities outperform bonds; logic also says they will outperform. However, can we really say that UK companies have pulled each other up by the bootstraps by investing in each other’s equity rather than in themselves or each other’s bonds? Financial economists think not and, put that way, it is hard to disagree.

Back to basics
Pensions are deferred pay. They are part of the benefits package. In particular, defined benefit schemes are designed to reward employees who stay for a long time with the employer and to reward progression within the organisation. For all the talk of labour market flexibility, defined benefits can be an important tool for companies in recruiting and retaining staff.
On a wider level, this is part of the battle between providers of capital and labour. For financial economists a gain to one party is a loss to the other. At a simple mathematical level this is true, but it is not always possible to maximise profits by driving down labour costs. For many activities, a well rewarded, well motivated, secure workforce will be more productive, and therefore be more profitable, for shareholders. Sharing the rewards of a successful business is not a zero sum game.
The role of the actuary in informing members, trustees, company management, and shareholders in the battle between capital and labour is interesting, to say the least. It might be useful for the actuarial training to make it clearer that this is what we are doing!

Getting the message
I think it is vital that actuaries take on board the ideas that financial economics provides us. Boots will not be an isolated case. Companies are going to adopt the risk-management ideas that financial economists put forward because they make sense for their businesses the financial economists are ‘right’.
Financial economists need to focus on getting across the message of what creates real value in the business. It is not the case that switching from equities to bonds is always the right answer. Nor does taking out the defined benefit provision make the company more profitable, although this seems to be the naïve message that is supporting a mad rush to defined contribution provision in the UK. For one thing, employees with the economic power to do so will demand compensation for the extra post-retirement risk they are taking on. That cost may be considerably greater cash up front.
For me, the interesting question is whether defined benefit schemes are an effective form of reward compared to the alternatives, not just looked at in isolation. Even after reflecting on all the financial economists are saying, I believe the answer is that in many circumstances they are, which is what pensions actuaries have known all along.

02_03_01.pdf