De-risking the pension scheme
In recent years, corporate sponsors of pension schemes have found themselves in a constantly changing environment. Defined benefit pension commitments are now recognised as a major source of financial risk for most global businesses and companies realise that their credit ratings and their ability to finance themselves are likely to be affected by the growth in pension liabilities. In addition, the collapse of acquisition projects in the UK due to the intervention of pension fund trustees (encouraged by regulators) has brought the issue of pension funding very much to the fore.
No wonder that CFOs are taking a very much closer interest and treating the fund as an integral part of the sponsor’s capital structure.
Many European multinationals have for some time had centralised control of their pensions exposure, setting policy and providing coordination, support and guidance to their pensions managers worldwide. Even those that are traditionally decentralised seem to be moving rapidly in this direction.
For some, the development is driven, as one would expect, by a combination of funding deficits and changes in accounting standards. Another driver frequently mentioned is risk management; the move to more centralised control of pensions is at least in part a reflection of similar developments in other risk areas facing a global corporation and of a greater awareness of corporate governance.
Although it is not just UK and Netherlands-based companies that manage their pensions globally it is probably safe to say that those who are most aware of the risks are those with substantial pension funds in those countries and in the US. Clearly this is a reflection of the funded nature of pension schemes, the regulatory pressure on funding and the early move by accounting bodies in the US and UK to enforce the disclosure of pension assets and liabilities in the accounts. Now that the EU has adopted international accounting standards (IFRS), all EU listed companies will publish accounts in the next 12 months showing their pension commitments. `
The table shows the ratio of pensions liabilities to market capitalisation of the top 50 European companies in the Dow Jones index according to disclosures in their 2004 accounts and SEC filings. This survey by Lane Clark and Peacock1 excludes European companies who do not file financial information under UK or US accounting rules: the 2005 version should be more comprehensive.
The figure for Germany is perhaps a little surprising but includes DaimlerChrysler which has significant US exposure. The same report has calculated the average deficit of companies in each country and this shows German and Spanish companies in the lead followed by the UK, and then the Netherlands and France. The deficit is defined as the market value of externally held assets less the present value of liabilities.
Of course, the result is a reflection of the particular way that pensions are provided in the different countries and although the figure given is not necessarily the local measure, it is the approach used by IFRS. When the first IFRS figures are published next year, some large numbers are likely to appear in balance sheets and a few companies, such as Carrefour, have already warned of the impact. Companies that have not so far promoted pensions risk to the top of their management agenda, may soon do so.
For some companies, the move to centralised control of pensions is counter-cultural. One highly-diversified multinational with six major schemes in Europe and many smaller ones in Europe and elsewhere (but which together are small relative to the group’s market capitalisation) has a head office pensions policy team which reports into the directors of both human resources and finance. This team is responsible for monitoring global pension commitments, coordinating asset managers, providing advice on investment matters and ensuring that policy is implemented across the group. Although this arrangement is quite different to the decentralised approach adopted in the rest of the group, it is consistent with the very strong focus on risk management that is an essential part of the corporate culture.
Risk management is closely bound up with corporate governance and the increased focus of shareholders on off-balance sheet commitments. Acceptance that the pension fund cannot be considered as a separate entity from its sponsor, as had been the view in the past, has had a major impact on the willingness of sponsoring companies to take the long-term view implied in the provision of DB pensions. In the current environment, most CFO’s would not consider for one moment taking on a liability that is indeterminate in size, not controllable by the company and lasting up to 70 years. How many companies can be confident that they will exist in 20 year’s time, let alone 70?
Although company boards must plan for the long term, the pressure from shareholders for returns in the shorter term means that CFOs are generally working to a much shorter time horizon than pension fund trustees. Shareholders are looking at the potential returns to them from the company and giving more value to the early cashflows and relatively little value to cashflows in the distant future. This is reflected in the company’s cost of capital; the higher the perceived risk to future returns, the higher is the cost of capital. Investment decisions made by the company will be based on this figure, which is a blended debt/equity rate usually referred to as the weighted average cost of capital (WACC).
Pensioners (and regulators) look at things differently, giving relatively less weight to early cashflows and more to long term ones, reflected in the lower rate required by IFRS for the discount of pension liabilities. As we will see later, even small differences between the company’s cost of capital and the IFRS discount rate have big effects.
When the company and the pension fund were seen as separate the difference in outlook was not a problem. How can the CFO reconcile these irreconcilable points of view in one entity? He has to “de-risk” pension-related cashflows so that the lower discount rate is appropriate to the risk profile of the pension commitment. The trend towards closer matching of assets and liabilities by shifting the portfolio from equities to bonds, is a direct result. Raising the funding level is another way of reducing the risk in the fund but this just shifts the risk to the company to no net benefit.
As may be expected, companies with a strong risk management focus tend to have a low proportion of equities in their investment portfolios, as is the case with the multinational mentioned above. ICI (featured in IPE July/August 2005 issue) is another example, having been something of a pioneer in its management of pensions risk. The company is adamant that the driving force for the closure of most of its DB schemes to new members is the reduction of risk, not cost. Since 1997, it has also progressively reduced the equity component of its pension portfolios to 18% (2004 accounts). Of course, risk reduction is a major issue for a company with pension liabilities standing at £8.5bn (e12.6bn) (FRS17 figure from 2004 accounts) compared with group market capitalisation of around £3bn.
ICI is a company with a philosophy of decentralisation where the centralised management of pensions is an exception to the corporate rule. Other multinationals have a more centralised culture so a head office department overseeing separate pensions operations in the subsidiaries is not a new development. In one such case, the central department has for some years provided a coordinating framework for the many pensions departments around the world. However in recent years there has been a change in focus and resourcing, with HR personnel needing more support from and coordination with the finance department. One of the spin-offs has been efficiency savings and improved performance monitoring resulting from the central procurement of services, but the main driving force in this case has been changes in accounting standards.
In 2001, when John Coombe was CFO of GlaxoSmithKline and a member of the UK Accounting Standards Board during the period when FRS 17 was being introduced, he said “This accounting change, it is said, will encourage companies to abandon DB schemes in favour of DC schemes. I have to say that if UK management allows itself to make economic decisions on the basis of accounting practice then the cart is firmly in front of the horse and calamity lies ahead.2”
Unfortunately shareholders do make economic decisions based on accounting figures, as they have very little else to go on, and there is little doubt that FRS 17 has been a factor in the closure of DB schemes. As noted above, if shareholders perceive that their returns are more risky, the company’s cost of capital will go up. This is particularly annoying if the accounting figures do not reflect the sponsor’s view of the economic balance between its pension assets and liabilities, one of the reasons for the often bitter arguments between corporates and accounting standard setters.
Pension liabilities are hard to pin down and several assumptions have to be made in order to estimate a current value for accounting purposes. There will be considerable disagreement over these assumptions within companies and between companies, and even the new accounting standard allows a range of assumptions and of accounting treatment. For this reason, figures published in IFRS accounts will not necessarily (or even usually) be comparable.
A further problem for the CFO trying to explain the figures to investors, is that the value of assumptions, in particular the discount rate used to calculate the present value of liabilities, changes over time so the resulting accounting figures become potentially very volatile although the underlying liabilities and assets may have changed very little. Volatility means risk and high volatility makes shareholders nervous. If the pension liability is moderate relative to the size of the balance sheet then the CFO may decide that it can be lived with. If you are CFO of ICI, with a total pension scheme of 10 times the balance sheet total, you may well feel that volatility must be avoided at all costs.
In IFRS accounting the discount rate for the valuation of pension liabilities is the yield on AA corporate bonds (for a duration appropriate to the particular fund). Whatever the currency, this will be a lower figure than the figure traditionally used for funded DB schemes, which was a blended debt/equity rate, and it can move significantly from one year end to the next. A rule of thumb is that a change in the discount rate of 1% (eg from 6% to 5%) changes the present value of the liability by 8-12%. As we have seen in the UK, this can be a very large variation for companies with big pension schemes.
The significance of relatively small changes in discount rate is demonstrated in a joint paper by Cardiff Business School and SEI Investments, published in August 2005, which suggests that companies should discount their pension liabilities at their own WACC. This makes some sense as this is the rate that companies use to evaluate other investment opportunities, so why should pensions be different? The paper estimates the average WACC in the UK to be 6.5%, just less than 1% higher than the average yield on AA corporate bonds, but this differential is enough to wipe out the FTSE 100’s combined pension fund deficit. It seems that some DB schemes are being closed because of a variation in the discount rate of less than 1%.
Unfortunately WACC is not suitable as a discount rate for liabilities, the problem being that as a company’s credit rating falls, its cost of capital rises. As the discount rate (WACC) rises, the present value of liabilities falls, so we would find that the present value of the same quantum of pension liabilities would be lower for a company in difficulties than for a financially strong company.
Although the decision to use the AA corporate bond rate was a compromise that many CFO’s feel produces a rate that is too low, some argue that the appropriate discount rate should be even lower, for example the market risk-free rate. Again, this has drawbacks as the rate is hard to establish when there are no Government bonds (the usual proxy) outstanding for the relevant maturity (and, of course, reported deficits would be even higher). Others argue in favour of using market annuity rates since these have to be used in a termination situation but few DB sponsors could fund anywhere near this level without damaging the company’s future prospects. The closer that regulatory discount rates come to annuity rates, the greater the incentive for sponsors to wind up their schemes altogether, so the ratcheting up by regulators of funding levels may well be counterproductive in the long term.
From a financial perspective the main worry about pensions is that increased contributions to existing DB schemes means less to invest in the future growth of the business. As one group pensions executive put it “One of the things we and the regulators need to consider carefully is whether we are putting the interests of pensioners before jobs for our existing employees and those we could take on in the future.”
1Accounting for Pensions UK and Europe annual Survey (2005), Lane Clark and Peacock
2Pension Schemes – Controlling the Corporate Risk, published by The Association of Corporate Treasurers (2001).