Nordic Region: Bouncing off the floor
A temporary reprieve from the regulator to deal with ultra low rates has largely been rebuffed by Swedish pension funds, finds Nina Röhrbein
As interest rates for 10-year Swedish government bonds tumbled to a historic low of 1.125% on 5 June 2012, the country’s financial regulator took action.
The Swedish FSA, Finansinspektionen, proposed a temporary, one-year floor on the discount rate used by pension insurance companies and occupational pension funds. The companies were given the opportunity to calculate discount rates according to the current method but based on the closing rates as of 31 May 2012.
“When bond yields were quickly dropping to low levels, we became concerned that some pension funds, driven by the regulatory framework rather than the long-term outlook, might start to sell their equity portfolios and buy long-term bonds at high prices, effectively locking in very low yields for the future” says Otto Elmgart, the divisonal head of large insurance companies supervision at the Swedish regulator Finansinspektionen (FSA).
The purpose of the floor was ultimately to protect pension policyholders, through the means of relieving the pension funds’ solvency situation and reducing the risk of a spiralling reallocation that could potentially deteriorate the situation further.
“Lower discount rates increased not only the value of the pension funds’ liabilities but also the interest rate sensitivity of such liabilities,” says Leif Lindahl, head of Danske Markets Sweden. “Due to the solvency based regulation in Sweden, these companies still needed to be hedged against a further drop in rates, thus prompting them to hedge even more, which in turn drove rates even lower.”
Since the floor’s introduction, 10-year government bond yields have bounced back to well above the level, despite some slumps along the way. This is due to the fact that the Swedish discount curve constitutes the average of government bond and covered bond yields, the latter of which did not contract to the same extent. At the end of September, 10-year government bond rates stood at 1.47%.
“Therefore, nobody is currently using the floor to calculate discount rates,” says Elmgart. “Another reason why it is not being used is that the regulations state that anybody applying the floor will be bound by restrictions on value transfers and dividend payments. And life insurers and pension funds are reluctant to give the supervisor power over such decisions.”
But the floor serves as protection in the event that rates fall again.
“Rates have been ultra-low for years and just because they have recovered a couple of tenths of basis points does not mean that with new turbulence in the market they will not go down again,” says Staffan Hansén, CIO of Norwegian Storebrand Life including its Swedish subsidiary SPP Life.
“The floor has become in-the-money only as recently as October,” says Lindahl. “In other words, it only propped up pension funds’ solvency ratios in October. However, the floor is likely to have had an indirect relieving effect in that it put some upward pressure on market rates after the announcement.”
Many pension funds have not made use of the temporary floor at all, deeming the administration of it cumbersome and their solvency level sufficient. Others use it for formal reporting and accounting only, while neglecting it in their decision-making of economic matters, such as hedging considerations, because of its temporary nature, according to Lindahl.
However, when introducing the temporary floor, the FSA did not change the stress test of the Swedish traffic light system. In other words, the stress test was not capped at the discount curve.
“My impression is that the FSA did not want to create some form of moral hazard,” Hansén explains. “They still wanted life insurance companies and pension funds to be aware of the risks and the stress tests are still working in the direction that companies need to focus on the risk picture. But the traffic light system merely reflects the risk status of a company and is not as strict as the solvency ratio, which has much greater demands.”
Capital flows from Swedish pension funds into other asset classes have been balanced lately with a tendency to buy domestic mortgage bonds.
The need for regulatory short-term reallocations was reduced following the introduction of the temporary floor. “We are not fully convinced it had any noticeable effect on strategic allocations from an economic perspective, since the floor is temporary and it is uncertain what will succeed it,” says Lindahl. “Should yields on Swedish 10-year bonds continue to fall, solvency levels will continue to deteriorate, with a painful and brutal drop in June next year, when the temporary floor expires, assuming it is not extended or replaced by a more long-term solution similar to what has been done in Denmark and the Netherlands. Much will also depend on how other markets such as equities perform.”
But as the FSA pointed out when it introduced the temporary floor, it is up to the life insurance companies and pension funds to address the long-term structural problems – such as long-dated guarantees – that persistent low interest rates will cause.
“Although the temporary floor gave some relief, it is up to the life insurers and pension funds to come up with a solution” says Elmgart, “from that point of view it does not make sense to introduce a permanent measure.”
“If upon expiry of the temporary floor in June next year Sweden follows the Danish and Dutch examples and introduces a Solvency II-like discount curve, pension funds’ duration and convexity problems will be reduced artificially,” says Lindahl. “This will not necessarily sort out the solvency problem as after all it depends on market levels. What such an introduction will add though is a significant de-coupling between the real world risk and regulatory/accounting risk, as we have seen in Denmark, whose short-term effects on hedging considerations and long-term effects on policyholder protection risk being highly unfortunate.”
One of the drivers prompting pension funds and life insurance companies to address their structural problems is Solvency II.
Bond investors that are, or will become, regulated by either Basel III or Solvency II are likely to face downward pressure on yields of the bonds that are attractive from a regulatory perspective as they will all chase the same paper. Less extensively regulated bond investors can take advantage of this through selecting new and existing bonds that are unattractive and hence potentially offer relatively higher yields, believes Lindahl.
“Some companies realise that it will be more difficult to have long-term guarantees under Solvency II,” says Elmgart. “Legally, the old contract is between the insurer and the policyholders and it is difficult to change it in a short time span. However, some life insurers are already modifying the guarantee level or proposing to their clients to switch from traditional life insurance to unit-linked policies, and pension schemes move from defined benefit to defined contribution.”
Lindahl adds: “We are beginning to see a trend in the pension offering market towards products in which the policyholder and not the pension fund assumes the market risk. Unfortunately, unless policyholders actually switch existing insurances into such products, this will only affect new and future pension inflows, with the bulk of the current market risk remaining pension funds’ concerns. Asset allocation strategies themselves cannot remedy a pension fund’s damaged solvency situation in the short or mid-term. But long-term, preferably with an equally far-sighted and consistent strategic hedging strategy, circumstances will be less unfavourable.”
With Solvency II there will be a new methodology of calculating the discount curve based on an ultimate forward rate, which is more macro than market-call based. “That means the dynamics of calculating liabilities will change and I do not expect the FSA to introduce any other, longer alternatives before they see in what direction Solvency II will finally develop,” says Hansén.