Cabinet endorsement of the new FTK proposals means Dutch pension funds must soon choose between a nominal or a real pension framework. But some are holding out for a hybrid option, writes Nina Roehrbein
Dutch pensions are to undergo significant change in the coming years after proposals for the new financial assessment framework, the FTK2, were finally published in July.
Defined benefit (DB) pension funds will have to choose between a nominal framework, with high levels of guarantee, and a real contract, which offers annual price indexation. While the latter is a longer-term framework with a specific ambition to compensate for inflation, its entitlements are conditional and linked to the results of the respective pension fund.
A 101% solvency ratio is required in the real contract. Pension funds will be able to handle financial market crashes and changes in longevity with the help of an adjustment mechanism for financial shocks (AFS). In case of underfunding, those with real contracts can spread their recovery period over as much as 10 years.
Nominal contracts have three years to recover. For the nominal contract, additional buffers are required to ensure that a certainty level of 97.5% can be maintained.
“For an average pension fund the buffer is currently 21.6% and that figure will increase to 26.6%, which will most likely result in higher premiums in the nominal contract,” says Corine van Egmond, senior ALM consultant at Aon Hewitt.
“With the buffer level set higher by the new framework than is probably achievable over that time period, they will in all likelihood have to adjust their asset allocation to lower their buffer requirements, especially those with low coverage ratios,” says Philip Jan Looijen, managing director for implemented clients solutions at ING Investment Management.
“Pension funds with nominal contracts are stuck in a solvency trap, which is not optimal in the long run.”
Peter Borgdorff, executive director at the Dutch pension fund for the healthcare and welfare sector Pensioenfonds Zorg en Welzijn (PFZW) notes: “We have to report to the regulator, the DNB, every year, declare our ambition and promise to the participants and how we combine this with our asset allocation. If pension funds fail to meet the promise in the long term, the regulator will tell them if they need to change their policy and promise and ensure a balance between the promise and the risk profile of the pension fund.”
Unlike pension funds that choose nominal contracts – which in the spirit of the new contracts are de facto required to de-risk and lock in nominal benefits – those with a real contract are expected to invest in index-linked bonds and other inflation-linked assets, such as real estate and infrastructure.
“But we do not have Dutch inflation-linked bonds, which is why it is difficult for pension funds to exactly match this inflation,” says Looijen. “However, if they do not have sufficient returns to match inflation, they can still pay out some inflation because the deficit can be spread over as much as 10 years. If they produce a higher return they can compensate the lower inflation of previous years. This opens the way for other assets, such as euro inflation-linked bonds, real estate, infrastructure and dividend paying equities.”
Van Egmond adds: “Real estate may be an investment idea because rental income could hedge part of inflation risk but there will still be a mismatch with regard to the inflation risk and then it remains to invest, for instance, in equity to receive an excess return.”
But Tom Steenkamp, executive vice-president and co-head of investment solutions and research at Robeco, adds: “Of course, it makes sense to have index-linked bonds instead of nominal bonds in the real contract. But beyond 20 years, pension funds have to take into account a forward rate of 4.2% and in the valuation of the real liabilities there are adjustments that will replace the long-term inflation expectations by a fixed 2% inflation target. In combination with another adjustment – a surcharge for the higher risk profile of the liabilities, standing at a maximum 1.5% – the problem is the effect of these adjustments on the valuation of the liabilities.”
Whenever a real scheme goes into underfunding, rights cuts – cuts to indexation or benefits – will have to be applied. As a result, benefit cuts are expected to occur more frequently in the real framework but to be smaller in size.
“In the current framework, we are trying to prevent cuts as much as possible,” says Jelle Beenen, business leader for Benelux at Mercer Investments. “In the new framework, cuts are almost hard-coded into the regulatory framework.”
Benefit cuts have already affected several pension funds. In April 2013, for example, around 70 pension funds were reported to have cut their pension benefits by 0.5-7%, affecting more than one million pensioners and more than two million active members. Another round of benefit cuts is likely in April 2014.
“At the moment, the majority of pension funds are in despair because they do not want to have a nominal contract and they are not able to go to the real contract,” says Gerard Riemen, director at the Federation of Dutch Pension Funds. “Pension funds simply cannot pay the price of a real pension scheme based on price indexation unless they take a lot of risk. But pension funds do not want to be forced to take more risk. Most want to keep the level of risk they currently have but with that they are not able to pay the prices for the real contract and have to de-risk for the nominal contract. And social partners do not want to completely give up their potential to have some indexation in their pension scheme, which is essentially the case with the nominal contract. This is why we need to have a third way.”
Consultation on the new proposals will continue until mid-September. The final version is expected to be unveiled at the end of the year, with the new framework coming into force on 1 January 2015, a year after the retirement age for second-pillar pensions switches to 67.
The third way
A third alternative could emerge as a by-product of the consultation, a hybrid between the nominal and the real contract, according to some market participants.
Unhappy with the obligation to choose between the two contracts, the Federation of Dutch Pension Funds is one of the many respondents to the proposals.
“The ideal solution for us would be the social partners determining the level of contributions,” says Riemen. “The pension funds would then consider their risk appetite and asset allocation to determine their ambition. There has to be a balance between those three important factors.”
Another issue of the new FTK is communication. “Pension funds have to think through whether they want to change their pension agreement towards a real contract and how they are going to communicate that to members because it is difficult to understand how, for example, excess returns are translated into additional pension fund rights or lower returns are translated into lower pension fund rights,” says Looijen.
“While the current pension system is already difficult to explain to members, shocks and spread periods will be even trickier to explain,” says van Egmond. “If different spread periods are used for longevity and economic shocks, the real contract could also be a challenge for the administration.”
Some complexity could also arise from having accrued rights in an old contract and new rights in a new contract. “This will lead to more complex administration and higher execution costs,” says van Egmond. “Pension funds have the option to choose between a closed and open AFS, and a closed AFS presents another challenge for the administration. Furthermore, there is no risk sharing over generations in a closed AFS.”
According to Beenen, the biggest problem with the new set-up is that it might reduce investment discipline because in the end everything will be solved by the AFS.
“The AFS may encourage riskier investment behaviour, which in turn can lead to higher financial shocks,” he explains. “The cuts are not necessarily smaller because you do not solve a shortage by regulation and if there is a big loss, there will be a big correction. But the transparency of the mechanism is a definite plus. The only other downside is that of multiple, consequent shocks.”
Beenen believes the main impact of the new framework will be how the discount curves will be calculated. “For the real discount curve, there is hardly any inflation sensitivity. They have taken a long-term inflation level and central economic bureau forecasts for the first two years. But from that point onwards, it is extrapolated to the long-term level of 2%, meaning that any change to inflation in 10 years’ time will not have any impact on the discount curve and therefore not on the solvency ratio of the fund, meaning there is no reason to invest in inflation swaps,” Beenen says.
“In certain scenarios, an inflation hedge can even be counter-productive. To some extent, if you protect yourself for 10 years inflation it might lower your funding ratio or introduce more risk in the official way of calculation. So the only thing that makes sense is more assets that react to short-term inflation shocks, such as commodities. But of course from a pure market point of view it is still prudent to include some inflation sensitivity in your assets in the real contract. The difficulty is that the framework is not a reflection of reality anymore.”
Frank Driessen, chief commercial officer at Aon Hewitt, says: “Indexation cannot be hedged very well at the moment, as products of this kind are already scarce and set to become scarcer, which may lead to a market disruption. There is also the possibility of a mismatch in ambition and the difference between price inflation and wage inflation, if the latter is used.”
Tim Burggraaf, principal at Mercer in the Netherlands, says: “Some people believe that if this legislation comes through, that the total of contributions in the Netherlands will go up because of it. But it was the extra contributions required in 2009 and 2010 that caused employers and unions to push for a new framework. If that is the case, it is a bomb under this specific piece of legislation and some additional measures will need to be in place.”
Jacqueline Lommen, director of European Pensions at Robeco, adds: “With the new framework, the financial risks for pensions are shifting away from the employer and the pension institution towards the individual member. This trend stems from the international financial reporting standards (IFRS) and solvency legislation. In the Netherlands, the transfer of the financial risk is happening in two ways – through the move to the new nominal or real contract and through a move away from DB to smart DC.”
There is some appetite for the real contract, especially among industry pension funds. But many smaller corporate pension funds are expected to find it too complex to move to a real scheme and all that comes with it, including governance and risk management.
“We expect that some pension funds will take this opportunity to switch to a much simpler DC-based pension scheme, which especially for corporate pension funds works better in an adverse environment,” says Looijen.
Steenkamp believes that with the exception of pensions funds, for whose participants second-pillar pensions are relatively small compared to their first- pillar benefits, in principle the only contract that should be provided for members is the real contract. “But by choosing the real contract the solvency ratio will initially go down because the liabilities are higher,” he says. “This could be a problem because then pension funds may immediately have to face benefit cuts.”
Driessen expects the pension funds opting for the real contract to be schemes where the participants can tolerate risk, in other words, pension funds representing highly educated participants with a higher salary.
“Nominal contracts will suit participants who need their pension to cover their cost of living and they cannot bear the risk of ending up with a little pension,” he says. “However, the majority of the participants of PFZW, most of which are not high-end earners, also voted for the real contract.”
PFZW decided some time ago to move to the real contract. It has also undertaken a pilot programme with the ministry of social affairs.
“The pilot programme strengthened our belief to move to the real contract, although of course everything will become more definite only when the legislation is in its final stage,” says Borgdorff. “But at present, we are still not entirely sure what the new FTK framework means for our pension fund. We have our last board meeting at the end of August, when we will conclude on the consultation documents. But our ambition to offer indexation remains.”
Borgdorff says the pilot programme found no legal objections to a move to the real contract, in other words, the existing pension rights can be transferred to the new contract.
“But first the requirements for changing to the real contract have to be met,” he says. “For example, we had to ask our pension fund members how they felt about the risks of the real contract and we have to calculate the intergenerational consequences.”
The pilot should result in a set of guidelines, containing the steps pension funds have to take to be ready and prepared for transferring existing rights into the new contract.
PFZW currently has a rather active strategic asset management strategy, with 35% in fixed income and the rest in equity and alternatives.
“It works as our performance figures – an average annual return of 8.2% since inception in the early 1970s – prove,” says Borgdorff. “By choosing the real contract we are able to continue with our asset management style as we did before. A move to a nominal contract would mean having to change our asset allocation to a more defensive portfolio, a lower return and not being able to earn the money needed for indexation.”
Ultimate forward rate
As part of the new framework, the ultimate forward rate (UFR) was introduced in autumn 2012 to discount pension liabilities.
For the average DB pension plan, the UFR increases funding status and decreases regulatory market risk because the long end of the curve is tied to the UFR.
A UFR committee is currently looking at the 4.2% level.
“There is also an equity risk premium of 3% set by the committee, which is included in the discount rate,” says Looijen. “All this will open the door for lobby groups to modify their discount rate in whatever direction they want to have it. In other words, there is little market discipline in that real contract.”
But the level of the UFR is probably the least important issue, according to Beenen, who believes it is more important how pension funds calculate liabilities and whether interest rate sensitivities are correct.
“In the consultation document, the UFR plus the three-month averaging mechanism is used to calculate liabilities,” he says. “This will be changed into a 12-month averaging, not of the liabilities but of the solvency ratio.”
“The UFR was not the Dutch government’s idea,” says Borgdorff. “It was the European Commission’s plan for European insurers. It does not affect the FTK itself. At the moment, the UFR is calculated based on 4.2%. What this will be in the new contract is not clear yet.”
The FTK proposals have for now also taken any attention away from other developments in the pension market. Premium pension institutions (PPIs) have been able to provide DC pensions, including on a cross-border basis, since 2011. However, since the public consultation in spring, talk surrounding their DB counterparts – the APIs – has gone quiet.
“We see the API as an opportunity to bring a commercial DB product to the market,” says Looijen. “The API could be a solution for smaller or closed corporate pension funds, which struggle to move the scheme to an industry-wide pension fund because of high costs, while low interest rates make it too expensive to do a buyout within an insurance company. An API could be an interesting solution to that.”