Commentary: Returning market sense to UFR
The new UFR is an uneasy compromise that requires sophisticated decision-making on the part of pension funds, argues Jens van Egmond
On 14 July, the Dutch Central Bank (DNB) announced a revision of the regulatory discount rate for pension funds, with immediate effect. The ultimate forward rate (UFR), used by pension funds to calculate the value of their liabilities in the far future, has been revised downwards from 4.2% to 3.3%, but what are the consequences for pension funds?
Economically, nothing really changes in the short run. After all, the amount of money in the pension pot does not change as a result of this calculation. However, the lower UFR could lead to a different distribution of the pension pot between stakeholders, for three reasons.
First, funding ratios have dropped by 2-6%, depending on the average age of the participants of a fund. The consequence is a lower probability of indexation, as this is conditional on funding ratios. This is disadvantageous for the elderly, causing a wave of criticism directed towards the regulator.
Second, the lower discount rate will cause pension premiums to increase, which is advantageous for the elderly (and a pain for premium payers – the employer and/or employees). Because premiums are often capped at a percentage of income, the accrual of new pension rights might be lowered. This is detrimental for the expected pension welfare of the young, but positive for the elderly.
A third, more limited effect is a lower capital requirement for most funds, depending on the structure of their interest rate hedge. This will reduce pressure to cut pension rights in extreme economic conditions, but is diminished by the direct impact of the measure on the regulatory funding ratios.
These three effects were of course anticipated by the DNB. But how did it come to its decision? The debate revolves around the level of market interest rates for maturities where such a market does not really exist. The market is usually defined by its liquidity – trading in interest rates shorter than 20 years has a much higher volume than in longer interest rates. This stimulates a debate about the interest rate used to calculate the value of a pension payment 60 years from now. In other words, the current market level of the 20-year rate is reliable and tradable, but the current market level of the 60-year rate is probably an indicator of lesser quality when it comes to reflecting economic conditions.
The DNB is trying to solve this problem by averaging the long interest rate over 10 years. This relieves pension funds from the pressure of daily market fluctuations of illiquid interest rates without ignoring the market information of traded interest rates. As such, the measure responds to the most important criticism of the old method, which ignored market information decreeing a 4.2% UFR, regardless of the worldwide economic situation. The level of 4.2% should have reflected the equilibrium interest rate in the long run.
However, it has become painfully clear that market rates can be miles away from this supposed equilibrium for many years. The DNB therefore devised this compromise between relief for pension funds from daily market fluctuations and the retention of a regulatory environment in line with economic reality.
The good news is over now, as this compromise turns out to be a highly complex solution. The pension sector has already been relieved from daily market fluctuations as funds are permitted to report their funding ratio as a 12-month average. It is unclear why a 10-year average of interest rates should be stacked on top of this rule. To illustrate the complexity of the new system, consider the following. The level of the UFR (now 3.3%) is not static, as it is an average of the past10 years. The fact that interest rates were much higher 10 years ago than they are today implies a steady decline in the UFR in the future. The magnitude will be about 0.25 percentage points per year for the next four years and the consequence will be a slow convergence of regulatory and market interest rates. We know that the step from 4.2% to 3.3% cost an average pension fund about 3% of funding ratio and from that we can derive that steps of 0.25 percentage points in the coming years will cost over 0.5 percentage points of funding ratio per year.
The result is that an average Dutch pension fund will have to generate a return of at least 0.5 percentage points above the market interest rate just to maintain its funding ratio at the same level. In this way, the gap between regulatory bookkeeping and economic reality is being erased. The unfortunate conclusion is that the regulatory framework is a complex, lagged proxy of economic reality. Managing this situation creates a need for sophisticated policymaking at pension fund boards. It is common for pension funds to mitigate interest rate risk using financial instruments, but this newly introduced risk cannot be hedged. Potential further rate declines will be dampened but any benefits from a possible rise in interest rates will obviously also be delayed as well. This will delay indexation further into the future once more, risking a new debate about the regulatory framework when rates rise.
The only risk that a pension fund board can control is the economic risk on the balance sheet. The good news is that the regulator is moving towards an economic approach with this measure. The measure itself, however, reveals the political risk that underlies financial supervision based on non-market parameters. A pension fund board is therefore advised to focus on the true economic risk on behalf of participants, and use the regulatory framework purely as a precondition. As complex as the financial world may seem, politics knows no limits. Compared with controlling political risk, being in control of the economic risk is simple.
Jens van Egmond is strategic risk adviser at Cardano