Sections

A better approach to solvency

The Dutch regulator is using the wrong measure to assess pension liabilities, argue Piet Duffhues and Anton van Nunen. They offer a different approach

Markets reached new lows on an almost daily basis until spring 2009. Insiders speak of a total surrender by investors; some claim pension funds are virtually bankrupt. It is indeed remarkable that not only the general public but also funds themselves are confused about the real position of pension funds.

However, it is easy to overlook that an abnormal weakening of the funding ratio in a recession was to be expected as a consequence of the new regulation. In the Netherlands, the nFTK regulation of the DNB, the supervisor of Dutch pension funds, prescribes liabilities to be valued at market value, using a specific discount factor: the swap rate. Now is the first time that pension funds are confronted with this new regulation but it should not be unexpected. The degree of pain is accentuated, more or less by coincidence, by the fact that the present economic crisis is very deep.

In this article we try to demonstrate that Dutch pension funds are not bankrupt. That some commentators claim the opposite is due to the fact that liabilities are calculated wrongly and that the static funding ratio is mistakenly viewed as the correct indicator of long-term health.

Present value of liabilities
The pension liabilities determine the right side of the balance sheet. Calculating them on the basis of present value is in accordance with financial theory that argues that this represents the opportunity cost, which historical prices do not. When pension funds have to lend money for future liabilities to be paid, the interest rate offered should reflect the risk profile of its investments and its specific ‘redemption conditions' (pensions are not 100% secure).

This would no doubt result in an interest rate that would be higher than comparable government bond yields. A higher discount factor implies lower debt and, therefore, a higher funding ratio. Lower debt is only logical because nominal debt remains the same, irrelevant of interest rate levels.

‘Fair value' accounting is based on the actual situation in financial markets and is a well-known academic conviction. By valuing assets and liabilities at present market prices, superior information is delivered to all participants. Pension liabilities, after all, are nothing less than short assets, and as such they represent ordinary financial contracts, as do assets. Equities and bonds have this recognition as financial instruments; it is obvious that both assets and liabilities should be treated equally where valuation is at stake.

Essentially, information is about the question of how a pension contract should be valued. Daily practice shows that pension funds' financial situation, measured in terms of the funding ratio, is very volatile and for the first time that represents the implication of ‘fair value accounting' by Dutch pension funds.

Some have warned that this situation could happen, but people should get used to it as it is an unavoidable aspect of a justified change in regulation. In this light there is no reason for panic after the recession hit in 2008-09: the abnormal general weakening of funding ratios was to be expected as a consequence of regulation.

It is a double blow: the first experience with the regulation co-incides with a heavy recession. Funding ratios below 90% when 105% is required are no exceptions today. Although the market value approach is justified, that does not mean there is no problem in its implementation. There is, and that is why funding ratios have declined far too much.

‘Fair value' calculation
In our view, the guidelines for calculating liabilities' market value do not deliver a consistent fair value. There is no market for determining the actual price of pension contracts, therefore models must be used. Unprompted, in 2004, the DNB stated that pension liabilities should be valued using a credit risk free discount factor and opted for the swap rate. This approach is wrong and out of date.

As is obvious these days, pension liabilities are not without risk. Therefore, a discount factor incorporating a modest spread over variable government yield is much more appropriate. In practice, low swap rates for longer durations have caused increases in liabilities' market value and severe declines in funding ratios. Even worse, it has set off a self-reinforcing process in that pension funds buy long fixed derivatives, causing the long swap rate to decline, which leads to the same activity by more pension funds.

It should be stressed that our condemnation of the swap rate as unfit for the job is not because of the negative effects it has on funding ratios in the present situation. There are two sound and fundamental reasons.
•Theory: its risk premium is too low; and
•Practice:    the rate is too dependent on the behaviour of the pension funds themselves.

Buying frenzy has already led to a situation where the swap rate has become lower than sovereign rates. The time has come for the regulator to enter into consultations with the industry about present day views and experiences.

Swaps versus bonds
The swap rate is the fixed leg of a swap, mostly bought by pension funds that pay variable yields. The swap spread is the difference between the swap rate and the yield on comparable government paper. Normally, this spread is positive and comparable to the spread of AA corporate paper. However, recent practice has changed that and for good reasons.

The swap spread is indeed incomparable to a corporate spread as a swap carries no capital risk. It is a derivative whereby capital is only notionally rearranged. Moreover, collateral is often present when swaps are in the money. Therefore, interest rate swap risk cannot be put  in the same league as that of corporate bonds. It is much lower.

Pension funds' liabilities, especially from 2007-09, have been more comparable to AA corporate bonds. Therefore, the swap rate used in present day regulation is not fit for purpose. When pension funds experience weak investment results, pensions may be cut, implying that debts are not redeemed fully. Moreover, in these circumstances, sponsors will not provide extra contributions. Therefore, the implicit yield on pension funds' ‘quasi' bonds should be considered higher than the swap rate as the risk profile of a fund is higher than suggested by the swap rate.

Swap rates and spreads are determined in a distinct market that surely is not comparable to the AA corporate bond market. Prices reflect essential differences in risk, and supply and demand are also different. To put it another way, when the swap rate has its own determining factors it should not be used as a discount factor to simulate yields on corporate bonds.

Logically, another rate is better suited to provide that simulation: the yield index of an AA corporate bond portfolio. This yield has fared totally differently, rising strongly where swap yields have declined. February 2009 shows 130-155 bps compared to swaps in the six-to-nine year segment.

Discount rate conclusion
The differences between the swap rate and our preferred government-plus rate are so huge, both in principle and in practice, that it is no longer sound policy to use the swap rate as a discount factor for pension liabilities.

This is a design fault in today's regulation. Replacing it with an AA corporate bond yield would raise many funding ratios strongly. Of course, that is not the goal of this operation, although it is a nice side effect in the present crisis.

Disquiet over the impact of the crisis, after correction, will be reduced to manageable proportions, namely to the disquiet that accompanies the volatility of equity prices in general. Pensions can be sustained more easily and panic will subside on sound economic grounds. This will have positive effects on spending which should not be underestimated in these days of possible macroeconomic underspending.

Supervision
Our second issue is whether pension fund supervision is structured optimally as long as funding ratios are core. Today, the funding ratio is calculated on the basis of an accounting balance sheet, the outcome of which determines whether a recovery plan is necessary.

The funding ratio is defined as the ratio of assets to pension debts (normally defined as the market value of pension provisions). However, under this one-dimensional supervision approach not enough attention is paid to the long-term goals of the funds.

That is why this liquidity-inspired number should be supplemented by information on solvency. This is advocated by both finance theory and practice. Banks, other creditors and investors have reviewed corporations using figures relating to both liquidity and solvency for a long time already. Both characteristics are as important for pension funds.
Indicated (see box) are five relevant financial definitions originating from business literature.
• Fund liquidity: liquidity indicates short and medium-term payment capacity. The current ratio is well known and expresses the market value of the investment portfolio when those assets can be sold within one year, compared with one year's liabilities. This implies a strong restriction on liquidity, which is less stark in ratio [2] where medium-term assets and liabilities are also considered. Concerned citizens, fearing acute payment problems, can be satisfied with these figures (almost no Dutch pension funds have no short or medium-term problems).
• Fund solvency: solvency is all about equity versus debt. In concrete terms, it deals with the cushions used to cope with losses compared with liabilities or balance sheet total.
The regulator prescribes a certain level of the (static) funding ratio. But we would abandon this ratio as it contains a mixture of liquidity and solvency aspects. Moreover, a definition comprising these elements does not exist in the financial literature or practice. In addition, the funding ratio is too static as it does not give due consideration to a fund's return potential as a source for partly financing future pensions.

Therefore, we would replace it with another definition [4], which resembles the funding ratio to a certain extent, but which is strictly geared toward solvency. Extending [4] to a longer period [5] gives more structural information on the degree of solvency. 

The conclusion is that [5] can provide an important supplementary insight into whether a fund needs to cut pensions. The outcome of [5] being favourable with [4] showing bad marks is that short-term cuts are not necessary. These would only apply when expectations of future returns would have to be lowered structurally.

Summary and conclusions
Regulation and supervision based on funding ratios only does not give enough insight into the balance sheets of either companies or pension funds, the liquidity of a fund is not recognised at all and the solvency not sufficiently. Yet, these concepts are indispensable in financial markets. The basic question for solvency is whether the world economy will recover. Only where the answer is absolutely negative would definition [5] have no added value compared with [4].

The DNB aims to improve the stability and integrity of the Dutch financial system. Departing from this notion, regulation should be restructured by replacing the swap rate by a government-plus yield and by taking into consideration the set of five definitions in an extended supervision of pension funds.

The regulator itself should provide the industry with permitted inputs with respect to expected returns.

Piet Duffhues is emeritus professor of corporate finance at Tilburg University and Anton van Nunen is director of the Dutch consultancy Van Nunen & Partners

 

Have your say

You must sign in to make a comment

IPE QUEST

Your first step in manager selection...

IPE Quest is a manager search facility that connects institutional investors and asset managers.

  • DS-2497

    Closing date: 2019-01-09.

  • QN-2498

    Asset class: Fixed Income Investment Grade.
    Asset region: Global Developed Markets.
    Size: $50m.
    Closing date: 2019-01-07.

  • DS-2499

    Closing date: 2019-01-02.

  • DS-2500

    Closing date: 2019-01-10.

Begin Your Search Here