After the Bell: Retirement roulette - is a GEPF payment holiday a smart bet?

One suggestion to fix the government budget shortfall, made oddly by trade union federation Cosatu, is for the government to take a payment holiday from the Government Employees’ Pension Fund (GEPF). Is this a great idea, or a terrible idea?

From just a maths point of view, it’s more than a good idea; it’s perfect. The employer contribution to the GEPF was R59.4-billion in 2023/24, which is coincidentally almost exactly the amount the government needs to balance increased spending pressures.

The suggestion arises for two reasons. First, the fund is actually in a strong state financially. As at the last statutory valuation in 2021, its assets exceeded its liabilities, as estimated by Alexander Forbes, by 110.1% This is an increase to its previous valuation in 2018 of 108.3%, and even that is pretty healthy if you consider the fund’s minimum valuation is 90% of assets. 

Second, this is a defined benefit fund. In other words, the benefits of members are fixed, and not dependent on its performance. Beneficiaries of the fund would therefore not be prejudiced in any way by a payment holiday.

The level of outperformance of the fund means that its surplus in numerical terms is R187-billion, which means the government could take not a one-year, but a three-year, funding break if it wanted to. One is slightly cautious about mentioning this in public because there have been times the fund has been undervalued and, in those cases, guess who gets to make up the shortfall? You guessed it: us, the taxpayers. 

Cosatu’s logic is simply that this would be a quick and simple way to fill the budget shortfall. Cosatu parliamentary coordinator Matthew Parks told Business Day: “Right now the GEPF is funded at 110% of its capacity, which means that it can afford for everybody to retire tomorrow, not receive another contribution and still have change. So it’s overfunded and, look, you can’t do it forever, but there is space to have a one-year contribution holiday, that’s R53-billion. The political parties have not expressed themselves on the issue, other than the DA, which has said the idea is not part of its proposals.” (Not quite, but anyway).

But, of course, there are problems with the idea, too. 

The first is political. Public servants are, at the moment, furious about the GEPF’s decision to grant a meagre 2.9% increase to their pension income. The reason is both technical and rather weird, as Maya Fisher-French explained in her column on News24. Oddly enough, pension funds base their increase on the Consumer Price Index (CPI) figure in a particular month. The GEPF set its increase at CPI for November, which, because of declining inflation, was 2.9%. But if you averaged it out over the financial year, it would have been more like 5%. 

So, as it happens for this year, pensioners lost the lottery and they are furious about it. The idea that the government would increase the risk of future increases by taking a payment holiday would probably go down like a ton of bricks. 

The second problem is that although the GEPF is actuarially better than solvent, it has two measures for its actuarial health and, on the second measure, it’s a little below the required level. This is something called the balance of assets after contingency reserves measure and this has declined from 75.5% in 2018 to 74.3% as at the 2021 check. 

Contingency reserves try to take account of things like mortality improvement and pension increases for past service or future service. But this is a lesser measure because it’s really building very unlikely contingencies into the system. But still. It’s there for a reason.

The third problem is that the GEPF’s actual investment performance since the statutory valuation has been poor. Let me put it bluntly: a child with a dart taking random shots at a page of investment options could probably have done better. In the 2023/24 financial year, the GEPF managed to lose R57-billion on domestic- and foreign-listed equities in a year the JSE All Share Index was up +5.26% and the S&P 500 was up 26.6% (!).

The GEPF was saved by its foreign collective investment schemes investments and, to a lesser extent, by its domestic unlisted equities, but these proportions are both small, so ultimately it turned in a 4.9% improvement.

The problem is that this is less than the assumptions made in the valuation, which is anticipating returns from equities for the next 20 years of expected inflation plus a real return of 5.3%. The fund is just not getting that, partly because its investment agent, the Public Investment Corporation, keeps on investing in defunct, politically inspired projects and partly because the GEPF is somewhat tied to local investments, meaning it’s yoked to the local economy, which is, as we all know, performing badly. 

The Alexander Forbes report actually says: “The fund holds a lower percentage of foreign assets than might otherwise be suggested, purely in terms of the risk diversification of assets.” This tells you something.

Fourth, and this is really the biggest problem, without some expenditure moderation, this is just not sustainable. The constantly increasing funding demands, combined with lower-than-expected economic growth, is just a highway to hell. For a decade now, the ANC has plugged the budget hole with this and that, and the next thing. As I keep saying, the ANC has not yet confronted the reasons for its underperformance as a government and, until it does, we are going to keep coming back to the “plugging the hole” solution, until there is nothing left to plug it with. DM