More than the current Pension Law (Pensioen- en Spaarfondsenwet), the new Pension Law will provide guidelines for the funding and balance sheet management of Dutch pension schemes. These guidelines are meanwhile known as the ‘Financieel Toetsingskader’ or FTK,which in the English translation it is equivalent to ‘Financial Assessment Framework’.
In the past year, the industry has seen white papers, consultation documents, supplementary notes, all pertaining to the FTK and all published by different organisations. The key players are of course the Department of Social and Labour Affairs (SZW), the regulator (the Dutch Central Bank or DNB)1, and a government advisory body on labour affairs called STAR.
The current state of affairs is that the DNB has published a bulky consultation document in October, which constitutes the heart of the FTK. Interested parties can submit reactions to this paper, but the general feeling is that the FTK has almost reached its final form, and that there is not much opportunity for amendments.
One of the most important issues for pension funds, the determination of necessary contributions, is however not covered in the October consultation document. This is because at that time the STAR still had to comment on a supplementary note from SZW on appropriate contribution levels. In November the DNB published temporary guidelines for funds that already want to implement the FTK ahead of schedule, and this document does specify guidelines for the determination of appropriate contribution levels. Finally the DNB has published a consultation document on the methodology it proposes to use to determine the term structure of interest rates to be used in calculating the present value of a fund’s pension liabilities.
This is really the most important change that the FTK will bring. Instead of discounting the liabilities at one fixed discount rate of 4%, each future cash flow will now have to be discounted at the relevant market interest rate. This has a number of effects. First of all, the present value of the liabilities will depend on the actual term structure of interest rates. If interest rates move up (down) the present value of liabilities will decrease (increase).
Under the old guidelines, the present value of the liabilities does not depend on market interest rates, because there funds can use a fixed discount rate. The second effect is that term structure movements will have opposite effects on the funding ratio of pension funds under the old and new guidelines. Increasing interest rates would lower the funding ratio in the old situation and would increase the funding ratio under the FTK. This happens because the interest rate sensitivity, the duration, of liabilities is in general much higher than the duration of the assets under the FTK.
Only nominal unconditional liabilities will have to be accounted for under the FTK. All conditional elements like conditional indexation, will not have to be included in determining the financial solidity of a pension fund under the FTK, as long as the conditionality is explicitly stated in the scheme’s rules and is communicated to the members and retirees.
The heart of the FTK is the solvency test which determines whether the pension scheme is able to withstand predefined (combinations) of financial shocks. The reserves that pension funds should build need to provide a cushion such that the probability that the fund will reach an under-funded status in one year’s time will be less than 2.5%. Or, stated differently, the probability that a pension fund has the capacity to cover all of its unconditional liabilities in one year’s time needs to be greater than 97.5%.
The regulator provides a standard risk model that should be used to determine the required buffer levels given the asset allocation and the population of participants. Pension funds can use their ‘internal’ risk model alongside the standard model, if this internal model is approved by DNB. If the available buffers or reserves are less than the required buffers, then a pension fund has to submit a recovery plan with the DNB, in which it has to specify how and when the required buffers will be built.
Pension funds are allowed to spend up to 15 years implementing the recovery plan and to reach the necessary buffer level. The DNB has calculated that for an average pension fund the necessary funding ratio will lie in the vicinity of 130.
Next to the solvency test, pension funds need to conduct continuity analyses on a regular basis. The first of these continuity analyses will have to start no later than January 1 2008. The continuity analysis should cover a period of at least 15 years and contain a prognosis of the fund’s evolution based on economic scenarios and assumptions.
Sensitivity analyses need to establish a fund’s ability to cope with stress situations and determine the effectiveness of the fund’s policy instruments. Since there is a still plenty of discussion and confusion possible regarding the guidelines for establishing appropriate contribution levels, we expect further clarification from the regulator.
Another question that has been the subject of much discussion and speculation is whether pension funds will change their asset allocation as a result of the FTK. The common opinion is that pension funds will hedge their duration mismatch and invest (more) in long maturity bonds and will engage in interest rate swaps. The fair valuation of liabilities makes these liabilities sensitive to interest rate movements, and this interest rate sensitivity or duration is generally different from the duration of the pension fund’s assets.
This duration mismatch is one of the financial risks a pension fund has to establish a buffer for. It is however not a risk that entails a positive risk premium, so the expectation is that pension fund will hedge this risk away, and reallocate the freed-up risk budget towards assets that do contain a positive risk premium.
In our view the FTK creates a tension between pension funds’ long-term goals, for example, to earn long-term risk premiums and the need to comply with the more short term oriented risk control guidelines. This tension will be very prominent when short-term risk control measures prohibit the capture of the necessary risk premiums. If pension funds, in their desire to hedge the duration mismatch away, completely match their assets to their liabilities then the investment return will by definition be the risk free return. The current level of interest rates however is too low for pension funs to be able to honour even very modest indexation ambitions.
We believe that long-term targets for pension funds should not change because of FTK. Defined benefit plans still have the provision of (inflation protected) pensions for their participants at a reasonable cost as the most important long-term goal. In the realisation of this goal, a long-term perspective and a strategic asset allocation are indispensable.
Asset managers and investment banks have noticed that the new risk based regulation provides new opportunities for them to sell risk protection products, which they were until now selling to insurers and banks, to a new clientele, that is, pension funds. Interest rate risks, stock market risks can be hedged away, even with the possibility to keep part of the potential upside.
Pension funds can diversify their financial risks, through asset diversification, and they can hedge or insure their risks, for which they will pay an insurance premium or face possible opportunity costs. There is a plethora of products available to achieve the exact level of protection needed, but pension funds will also need to keep some investment risks to safeguard their long-term goals.
The risk model proposed by the DNB provides a quantitative framework in which pension funds can assess the perceived risks of asset categories. If one combines this with the expected risk premiums to be collected from these assets, one can determine how to allocate the available risk budget, that is, the available buffer, to the various risky assets.
We see this as one of the main advantages of the FTK. The financial risks faced by pension funds become very transparent, and the price of these risks, that is the necessary buffer is also given, so that pension funds have a new instrument available to decide on how to allocate their risk budget. As a consequence the initial funding ratio and the strength of the sponsor are going to be the main determinants of the level of risk that a pension fund can accept. For with a high funding ratio and high buffers, pension funds can use higher risk budgets, and with a strong sponsor this budget can be replenished when necessary. Obviously with a low funding ratio and a weak or unwilling sponsor the risk budget is lower.
This could however lead to a possible conflict, because one can also defend the claim that the weakest funds should take the most risk, to at least have a decent probability of recovery. In our view the conflict between short term and long term can only be resolved by adopting an integrated approach that proceeds along the following six steps:
q Formulate the long-term goals of the pension fund and important stakeholders, for example, providing inflation protected pensions at reasonable costs.
q Analyse whether these long-term goals can be achieved with the current investment, contribution and indexation policies. If yes go to step 3, if not then first reconcile the policies with the goals, and take into account a maximum recovery period of 15 years, then move to step 3.
q Model the use and effects of risk control measures in an ALM study if the short-term underfunding probability (funding ratio less than 100) is larger than can be tolerated.
q Analyse which risk control measures:
l Reduce the probability of underfunding as much as possible and as soon as possible.
l Create the least damage towards achieving the long-term goals.
l Can be implemented easily and cost effectively.
q Select and implement one or more risk control measures and keep monitoring the underfunding probability.
q Remove the risk controls when these are no longer necessary, that is, the one-year probability of underfunding is less than 2.5%.
By integrating short-term risk controls with longer-term goals pension funds can comply with FTK and still maintain a long-term perspective.
Fred Nieuwland is practice leader, asset consulting services and Jan Luuk Roelfsema is senior consultant with Towers Perrin in Amsterdam
1 The Pensioen-en Verzekeringkamer (PVK), the regulator of the Dutch pension industry, has merged with the DNB. The new regulatory entity operates under the DNB name.

Topics