Ultimate Forward Rate: Ultimate consideration

Formerly just an esoteric formula buried in Solvency II legislation, the DNB’s recent reduction in the ultimate forward rate will affect the viability of Dutch defined benefit pension schemes, writes Leen Preesman

In a controversial decision criticised by the pensions sector but applauded by financial and risk experts, the Dutch pensions supervisor De Nederlandsche Bank (DNB) in July recalculated the ultimate forward rate (UFR) down  from 4.2% to 3.3%. This reduction, part of the discount rate for liabilities within the FTK framework, is adding pressure to the viability of the traditional defined benefit (DB) schemes that have dominated Dutch pensions.

As pension fund coverage ratios drop, contributions are likely to be raised to achieve sufficient pensions accrual. The alternative would be to reduce the annual accrual rate, but tax efficient accrual has already been squeezed from 2.25% to 1.875% recently.

The principle of the UFR was introduced in September 2012, as an additional stabilising mechanism within the discount rate framework. It was meant to be a transitional measure towards a new financial assessment framework (nFTK) and a rate of 4.2% was to be factored in gradually for durations between 20 and 60 years. 

The UFR not only stabilised funds’ coverage ratio, but – to the sector’s delight – also raised funding. Schemes with an older member composition saw their coverage ratio rise by up to 3 percentage points, whereas young schemes’ funding rose by around five. However, given the drop in interest rates, the UFR could not prevent dozens of pension funds from having to cut rights.

At a glance

• The ultimate forward rate derives from Solvency II. 
• It was introduced within the FTK framework in 2012 as a discount rate for long-dated pension liabilities where there is no effective market rate or the market rate is highly illiquid.
• The UFR was originally 4.2%.
• This has been reduced to 3.3%, using a 10-year average of the 20-year forward rate instead of inflation expectations and short-term rates.

From its inception, the UFR was criticised by Theo Kocken, professor of risk management at Amsterdam’s Free University. At the time, he argued that the accounting rule was destined “to camouflage risks”. In the opinion of Kocken, also CEO at Cardano Risk Management, the level of 4.2% was too subjective. He suggested that it was dictated by market players.

Accounting trick
Jan Willem van Stuijvenberg, financial services consultant for Mercer, has described the UFR as an “accounting trick”, and said that it would cause the redistribution of pensions assets at the expense of younger generations. He also warned that the UFR would be an incentive to only hedge interest risk for a duration of up to 20 years. Lex Hoogduin, professor of monetary economics and financial institutions at the University of Amsterdam, said a level of 3.25%-3.3% was “more realistic”.

In addition, the Dutch Actuarial Society said it had identified more disadvantages than benefits from the UFR as constituted. It warned against the risk of creating artificially high funding levels, with potentially undesired effects on different generations. 

 Current and revised UFR versus 30-year zero-coupon interest rate

Nevertheless, a parliamentary majority rejected a more conservative UFR. And pension funds that wanted to refrain from applying the UFR, were corrected by the regulator, which instructed them to toe the line. The pension fund of the asset manager Robeco, for example, had initially indicated that it would ignore the UFR, “as it distorted reality”.

In 2013, a year after its introduction, the DNB attempted to allay fears that the new discount mechanism would affect pension funds’ interest rate hedging policies. A survey undertaken by the supervisor made clear that interest rate hedge ratios had hardly changed since the introduction of the UFR.

In the autumn of 2013, a six-strong committee predominantly of academics and researchers – tasked by state secretary Jetta Klijnsma with assessing the UFR’s sustainability – came up with a new concept: the running average of the 20-year forward rates during the previous 10 years. The UFR would have been 3.9% in July 2013 under this method.

Random change
However, experts from the pensions sector – including Bernard Walschots, CIO of the Rabobank Pensioenfonds – dismissed the committee’s recommendations. They argued that the proposed new approach would add “unnecessary complexity”. In their opinion, this “random change, would not contribute to greater funding stability, as pension funds’ assets were still discounted against market rates”. Instead, they advocated a consistent valuation of both assets and liabilities, “in order to achieve the best pension results”.

At the request of the Dutch Senate, the DNB postponed the introduction of a new UFR, which was initially scheduled to be included in the nFTK, and take effect on 1 January 2015. While debating the nFTK last December, the Senate had indicated that it would prefer to wait for EIOPA’s proposals for a UFR for insurers within Solvency II, as it preferred to harmonise the discount rate for both insurers and pension funds. 

EIOPA, however, concluded in March that it would maintain the UFR at 4.2%. Earlier, the Dutch Cabinet had made clear that it did not want Dutch pension funds to be subjected to the Solvency II rules for insurers, including the UFR.

When the DNB finally announced the UFR amendment with immediate effect this July, the new discount mechanism consisted of the forward rate converging from the last liquid point (LLP) of a 20-year duration to 3.3%. It explained that it had not factored in the last liquid forward rate, as suggested by the UFR committee, “for reasons of simplicity”.

Increased liabilities
A lower curve means increased liabilities. “In fact, the gift that pension funds received in 2012 has now been taken back,” concludes Marco Folpmers, professor of financial risk management at Tilburg University. However, he also calls on the DNB to clarify how it achieved a rate of 3.3%, “as the current 20-year forward rate is 1.8%”. This uncertainty, he argues, increases doubts for investment and hedging, and as such, it doesn’t reduce volatility.

Folpmers’ view is echoed by Lars van den Berg, consultant at Ernst & Young. He stresses that funding ratios have been kept artificially high since the introduction of the UFR, as the discount mechanism did not take investments into account. 

“Where there is no margin for a contribution rise and the unions refuse to accept reduced pensions accrual, a goodbye to DB plans will come a lot closer” Jan Tamerus

Frank Eldersson, director of pensions at DNB, defends the new rate, claiming it is “more balanced and fairer to all generations”, and that it accounts for “as much market information as possible”. He says the new UFR stands to reduce coverage ratios by 6 percentage points at pension funds with mainly younger participants. Average funds would lose 3 percentage points, while the funding of schemes with older participants would drop by 2 percentage points.

The impact of the UFR reduction was clear immediately. The €356bn civil service scheme ABP lost 1.9 percentage points of its funding discounted against market rates. The €166bn healthcare pension fund PFZW’s coverage was negatively affected by 3 percentage points, while BpfBOUW, the €48bn scheme for the building sector saw its funding slashed by 3.1 percentage points. As a result, their coverage stood at 102%, 100% and 111.4% respectively by July 2015.

Eric Uijen

Slower recovery 
The €43bn metal scheme PME said that the new UFR knocked 2.4 percentage points off its coverage ratio, lowering it to 100%, while adding a full year to its 10-year recovery plan. “The measure will further delay indexation, and increase the possibility of further rights cuts,” comments Eric Uijen, PME’s director. He claims that the new rate will make accrual more expensive.

The Dutch Pensions Federation has made clear that it is disappointed. Gerard Riemen, its director, argues that defined benefit arrangements will “no longer be sustainable” over time if interest rates remain at the current low level. He predicts that the new UFR will cut funding by 15% on average – from 7% for younger schemes to 23% for older ones – by 2020.

The Federation also contends that the new rate will increase pension funds’ susceptibility to rate changes, and risk destabilising the nFTK. In its opinion, the regulator’s decision to opt for a more market-based rate, “just when the European Central Bank’s quantitative easing programme is strongly affecting interest levels”, made little sense.

The Federation adds that it fears annual pensions accrual will be squeezed where pension funds have capped the premium level. Riemean says he would have preferred the DNB to have stuck with the old UFR for longer, until an entirely new pensions contract is ready. A new contract is expected in the wake of the nation-wide debate about a sustainable and collective pensions system.

Premiums to rise
The regulator acknowledges that the new rate could affect contributions, and predicts an increase in cost-covering premiums of 4-5%, if calculated using current interest rates. It expects the impact on contributions, using expected returns, to be half as large.

Jan Tamerus, master actuary at PGGM, the asset manager for PFZW, suggests that heated debates between employers and workers about the pensions arrangements could erupt later this year. “Where there is no margin for a contribution rise and the unions refuse to accept reduced pensions accrual, a goodbye to DB plans will come a lot closer,” he comments.

However, ABP has made clear that the new UFR would not mean a push towards defined contribution arrangements. “Pension assets accrued in a DC plan must also be converted into lifelong annuities. And with the current low interest rates, this remains expensive,” it says.

The UFR explained

Before 2012 pension funds had to discount their liabilities against a three-month average of market rates. However, using a pure market rate caused too much funding volatility. This led to the introduction of the UFR.

As originally constituted, the UFR is based on two components – the long-term inflation expectation plus the long-term expected short-term interest rate. These were estimated at 2% and 2.2% respectively, which makes 4.2% when the two are added together. As such, this resembled the fixed 4% rate for discounting liabilities that pension funds used until the introduction of the initial FTK in 2007.

However, the inconvenient fact is that 20 and 30-year swap rates have for some time remained far below 4.2%, which calls into question the extent to which 4.2% reflects a true long-term interest rate equilibrium. The DNB has sought to bring the UFR closer in line with market realities by calculating it on the basis of the 10-year average of the 20-year forward rate, which is currently 3.3%.

The fact that interest rates were much higher a decade ago means that the UFR will continue to fall in coming years as lower rates are factored in to the rolling 10-year average. For instance, the 20-year forward rate is currently about 1.8%.

Have your say

You must sign in to make a comment