Your cover story ‘Wake Up! Pay Up!’ suggested possible ways to address the material problem of unclaimed/unpaid benefits that the pensions industry currently faces (TT April-June). The editorial presented arguments for:
- A “central fund” managed by the state, where this fund could hold a reserve for the unpaid benefits less than the full face value of the benefits;
- The release of excess assets to perform “good works”.
The approach of releasing excess assets has already been attempted by some individual funds. It led to an ongoing dispute between these funds, the Registrar and the Minister of Finance regarding the legality, rationality and validity of Regulation 35(4).
This regulation effectively freezes unclaimed benefits due to former members who were included in surplus-distribution schemes, but who cannot be traced. There are essentially two different categories of unclaimed/unpaid former members.
In the first category, there are those who have withdrawn from funds in the ordinary course and who have not been paid their benefits. This includes the dependents of deceased members who have been awarded benefits, in terms of s37C of the Pension Funds Act, but who cannot be traced.
These members have not been paid the benefits due to them in terms of fund rules. The fund has a legal obligation to them. In accordance with PF circular 126, the Registrar compelled funds to amend their rules so as to preclude them from permitting the forfeiture of these members’ benefits.
In the second category, there are former members who’d been included in surplus-apportionment schemes because the funds had sufficient data to calculate the top-ups that would be paid to them. These former members included those who’d left the fund since 1980.
So it’s not surprising that the fund had little or no information about the whereabouts of members who’d left the fund years ago, and in many cases the fund did not even have ID numbers. Neither could the fund be assured that these former members were still alive at the time of the apportionment exercises.
This created a substantial increase in the number and value of unpaid benefits to members who’d already left their funds, prior to the relevant legislation, in terms of their rules. Without debating the “fairness” of the legislation which required surplus-apportionment schemes, it’s simply pointed out that there was a resulting increase in unpaid benefits. Obviously, it’s more difficult that benefits actually be paid to the second category than the first.
In respect of these former members who could not be traced, s15B provided that the amounts due to them could be placed in a contingency account “to satisfy (their) claims”. A contingency account is an account which shall be credited or debited with such amounts as determined by the fund’s board, on the advice of the valuator, to provide for a specific category of contingency.
The valuator, as an expert, is tasked with what value to place on this contingent liability. “Contingency” means something about which you are not sure of the final outcome; in a particular case, actually finding former members and paying them.
Based on this interpretation of the law it would have been possible for the valuator, in the years subsequent to the apportionment scheme, to reduce the contingent liability of the fund in respect of those former members. It would have been based on the likelihood of them having died before the surplus apportionment date, or the likelihood of them being traced, or of them ever coming forward to make a claim.
In the years post-apportionment (and provided that the fund had made every reasonable effort to trace former members), the prospect of those former members ever being found or making a claim diminishes substantially. For many funds, more than 10 years have passed since the approval of their apportionment schemes and they are nowhere near a full payout.
A reduction in the amount of the contingency reserve does not mean that the obligation to a former member has been extinguished or reduced. It simply means that the amount provided to meet the contingent liability to all former members has been reduced. If any former members come forward, they must still be paid.
Given that funds should be fully funded (the actuarial value of the assets should meet the actuarial value of the liabilities), it does mean that the assets that may formerly have been held to balance that liability may now exceed the liability. The amounts then released will be regarded as future surplus.
If regarded as future surplus, it could be distributed in terms of s15C to members and former members of that fund. If distributed in this manner, it will be released back into the economy in a substantially more beneficial manner than currently prevails.
Should the valuator undervalue the amount held in the contingency reserve, and more former members come forward or are traced than are anticipated, then it may mean that the fund is underfunded. This, of course, depends on numerous other factors such as asset performance compared to that assumed by the valuator, mortality assumed by the valuator and the like. In this event the fund will have to submit a scheme in terms of s18 to restore the fund to a financially sound condition.
These are normal processes in the valuation of funds’ liabilities. The valuator must make assumptions about the likelihood of future events. A particular example is that of suspended pensioners.
Where the suspension happened some time ago, valuators will normally hold a low or zero reserve against the possibility of the pensioner being reinstated. If the pensioner is reinstated, the fund must still pay the benefits even if no reserve has been held. This process is no different to what the funds in dispute with the regulator are trying to do in relation to the liability for unpaid benefits.
This is what ordinarily would have happened to the amounts allocated to former members. Former members would not be prejudiced by the valuator reducing the contingency reserve for these benefits. The fund is following a normal process. The economy would benefit by excess funds, being released as future surplus, being put to use.
This natural order of things was fundamentally upset by the Minister promulgating Reg 35(4). It provided that “monies may not be released from such contingency reserve accounts except as a result of payment to such former members or as a result of crediting the Guardians Fund or some other fund established by law to include such amounts”.
This meant that the valuator could never decrease the value of the amount allocated to the contingency reserve fund as he/she would have done in the ordinary course. Given the requirement of funds to be fully funded, this meant that assets to back that liability could never be released as future surplus.
Currently then, the full value of assets is held in funds to meet the full liability to untraced former members who’re unlikely ever to be traced or paid. That the assets are included in surplus-apportionment schemes is not useful to funds or the economy.
Thus some funds have challenged the legality, rationality and validity of Reg 35(4). Two of the cases have failed. Judges found that they did not have to consider the merits of the applications, due to non-compliance with time periods for objection, since the date Reg 35(4) was imposed. However, in both cases the judges went on to find that the regulation was in any event not invalid or illegal.
These judgments are not binding. In one other matter the court did rule on the merits. It found that Reg 35(4) was neither invalid nor illegal. Two of the matters are currently on appeal to the Supreme Court of Appeal.
The issue of providing for uncertain contingencies is more complicated than a straight accounting process, and not easily understood by lawyers. There is often a misunderstanding that the contingency reserve represents assets set aside by the fund. It actually represents an estimate of a liability, where an eventual payout is uncertain.
This is compounded by the language used in Reg 35(4) as monies are never, in fact, allocated to a contingency account. Contingency reserves have nothing to do with assets. The maintenance of a full liability to former members does not necessarily mean the fund will have the assets to meet the claim.
The actual experience of a fund due to a number of other reasons may differ from the estimate of the actuary. This does not mean that the obligation is extinguished or reduced.
Ultimately, a legislative and not a court-induced solution would be preferable. It is ironic that a “central fund” — which would be able to hold a reserve less than the full face value of unpaid benefits — is proposed while the regulator is preventing individual funds from doing just that.
There may well be a case for a central fund for the unpaid benefits of small funds. But why should larger funds not be able to make their own decisions? And pay out any resulting surplus funds in terms of s15C to their own members and former members?
The forced transfer of these unpaid benefit funds to a state-sponsored benefit fund, when particular funds are quite capable themselves of dealing with them, would represent a form of expropriation. It is far better that Reg 35(4) is scrapped, assets artificially held to meet a liability that is unlikely to ever materialise are released, and funds themselves put the assets to use as future surplus.
• Graham Damant is a partner at law firm Bowmans and Mike Walker, a fellow of the Actuarial Society of SA, has 30 years’ experience as an approved valuator to retirement funds.