Issue: March/May 09


A sick prescription

There’s a certain way to reduce the returns on retirement funds’ assets.It’s by government seizing a chunk of them. Trustees and principal officers can’t sit on the sidelines. They should say whether they agree, or scream before it’s too late.

To hear it in the dulcet tones of President Kgalema Motlanthe, the tapping of pension funds’ assets for societal purposes sounds almost seductive. Couched in the velvet glove of the need to minimise job losses and stimulate development, an iron fist protrudes.

What he’s really saying is that introduction (or reintroduction, for those with long memories) of prescribed assets is on the cards. What he’s not saying is that prescribed assets mean the state commandeering a proportion of savings held in retirement vehicles.

As such, prescribeds are a selective tax by another name. Being selective, by affecting a particular category of individuals, they’re also discriminatory. Financing of state projects is properly the role of the fiscus, properly borne by the general body of taxpayers in accordance with tax policy, or by investors who can choose whether to subscribe for government bond offers.

By their nature, prescribed assets imply a lower level of return on savings than offered in the marketplace. Were it otherwise, there’d be no need to prescribe how a chunk of assets must be invested. This is in contradistinction to bonds, issued at commercially competitive rates to attract investment, which retirement funds are in any event obliged to hold as a percentage of their portfolios in terms of the Regulation 28 prudential requirements.

Motlanthe... flawed arguments
Motlanthe... flawed arguments

To dress up prescribeds as a form of socially responsible investment (SRI) is deceptive. It does SRI a disservice. There’s no shortage of SRI projects that offer market returns, often better. SRI is not synonymous with lower returns, and there’s plenty of evidence to prove it.

Companies adopt SRI policies because, in the first instance, it’s good for them. It’s in their own best interests to look after their stakeholders and the communities they serve. They ignore SRI at their peril. The marketplace prescribes.

On top of this, there’s self-imposed compulsion. Check, for instance, the provisions of the voluntarily accepted Financial Sector Charter on targeted and corporate social investments.

What makes government think it can do better by prescribing the investment of assets than by not prescribing them? The slow pace of service delivery is not due to lack of money in government departments. It is due to lack of capacity to spend what they already have, and to the wastefulness in their spending that so concerns Finance Minister Trevor Manuel. The question then begged is whether prescribeds have a predominantly ideological motivation.

Noticeably, in his latest Budget speech Manuel offered no hint of prescribed assets. He appeared comfortable with the public-sector borrowing requirement, at 7,5% of gdp, “to be raised from domestic institutions, investors, multilateral institutions and portfolio investors from abroad”.

If prescribeds are imminent, then are retirement funds to be treated less favourably than other domestic institutions and investors who can only be induced by competitive returns and not forced to accept uncompetitive ones? If government’s borrowing requirements will strain the debt markets, will retirement funds have to invest whatever punitive amount at whatever prejudicial rate government decrees?

He also said: “Our development expenditure over the period ahead will require both improved domestic saving and continued capital inflows.” Well, let’s see how prescribes will encourage domestic savings through retirement funds. Better still, let’s not see it.

Contrast Manuel’s omission of prescribeds with the remarks of President Motlanthe a few days earlier. Delivering his State of the Nation address, Motlanthe said:

In the interactions between The Presidency and leaders of various social partners, we agreed jointly to devise interventions that would minimise the impact of this (economic) crisis on our society. Under discussion are... utlisation of resources controlled by workers such as pension funds.

“Our development expenditure over the period ahead will require both improved domestic saving and continued capital inflows.”

He didn’t identify the “leaders of the various social partners”. Hazard a guess. What’s more, it’s labour-speak that pension funds are “controlled by workers”. None are. They’re controlled by trustees, beholden to the fund alone, even where most or all of the fund’s members are “workers”.

Shortly after, Motlanthe elaborated to a media briefing:

The task team includes labour. As you know, pension funds and provident funds are directed by a board of trustees which is composed of a 50-50 base representation between employers and organised labour. So that’s one of the areas they would have to look at, if a very important sector is likely to go under and that you know an injection could save jobs and position it to survive.

Some of his facts, respectfully, are wrong. First, funds’ boards of trustees do not necessarily comprise a 50-50 representation between employers and organised labour. At least half of the trustees would be drawn from employers; the other half, who might not be unionists at all, is elected by fund members who similarly might or might not be part of organised labour.

Second, in law, trustees are responsible only to the fund. They are not responsible, and have no accountability, to those who elected them. Were nominated trustees to take instructions from their employers, they’d be as much in breach of their fiduciary duty as elected trustees who take instructions from a union. The fiduciary duty of trustees is to exercise independent discretion in the interests of the fund and nobody else.

Third, occupational retirement funds, where membership is compulsory, usually comprise the widest variety of employees. Some might not be members of a union; others might belong to different unions, and the unions in turn might be affiliates of federations other than Cosatu. “Organised labour” has no authority to speak for members of retirement funds as though all were “workers” from whom it had a mandate.

Fourth, assets of retirement funds belong to the funds. They do not belong to the members, or to any grouping of them. For these assets to be deployed in saving jobs by preventing “very important sectors” from going under, deflects from the primary purpose of retirement funds, which is to provide for members’ retirements.

Were trustees to invest in businesses whose sustainability is threatened, in the normal course they’d be guilty of exposing fund assets to undue risk. Prescription will compel them to do precisely this.

Remember that retirement funds, although they cumulatively have huge assets, are not the exclusive domain of the wealthy. They are repositories of individuals’ savings to fulfil a specific social need, and amongst funds’ many millions of members the poorer far outnumber the richer.

Funds themselves aren’t “rich”. It seems they’re being confused with “the rich” as a target for sacrifice. There are fewer people who can afford a nick into their retirement nest-eggs – the richer, who come off a higher base – than those who can’t.

Whereas sentiment would overwhelmingly support government-led initiatives to counter the effects of an economy in crisis, introduction of prescribed asset requirements is an inequitable and unbalanced way to go about it.


A leading proponent of prescribed asset requirements (PAR) is Cosatu general secretary Zwelinzima Vavi. In a seminal paper to the union federation’s annual conference in 2005, he argued:

Cosatu proposed, as early as 1995-96, two key measures which would have had a major impact on the fiscus and unleashed massive resources to address the apartheid social deficit. These were the reintroduction of PAR, which the apartheid regime had used extensively to leverage investment in its projects; and the restructuring of the Government Employees Pension Fund (GEPF), which had been deliberately overfunded during De Klerk’s rule and artificially ballooned SA’s debt....Introduction of PAR, which would have required all retirement funds and the insurance sector to invest a certain proportion of investments in a government reconstruction bond, would have unleashed many billions annually for social investment.

  • Unlike the overwhelming majority of retirement funds, which are defined-contribution (where the funds, and ultimately their members, bear the investment risk), the GEPF is a defined-benefit fund (where a proportion of final salary is the member’s guaranteed pension). With DC funds, members are prejudiced by the lower returns that PAR implies. With the GEPF, being a DB fund, government as the employer would have to make good on a liabilities shortfall. The better the GEPF performs, the less the potential burden on government and hence on the taxpayer.
  • All retirement funds invest in government bonds as a prudential requirement. Were there a need for a special “reconstruction bond”, government can readily introduce it at a market-competitive rate. The need for it to be at an uncompetitive rate isn’t obvious.

Labour placed these (PAR) proposals in its Social Equity document in 1996, tabled them in the Jobs Summit in 1998, and attempted to engage government on numerous occasions. The end result of all this was a refusal, particularly by Treasury, to seriously engage on the merits. The bottom line, which emerged, was that there was no substantive basis for rejection of these proposals....It was simply argued that to reintroduce PAR would unsettle “the markets”, although no explanation could be forwarded as to why capital could live with them in the several decades preceding liberation.

  • “Capital” lived with them because they were imposed. It did not “live with them” happily.
Vavi... voice of the politburo
Vavi... voice of the politburo

There has been little progress in creating institutional frameworks to promote the collective economic power of labour, and harnessing it for social reconstruction. The most obvious area requiring state intervention relates to regulation of retirement funds, which collectively control close to R1 trillion in assets. Far from harnessing these assets for social investment, and introducing tighter investment requirements, the democratic state has introduce PAR, has allowed the shift from bonds to speculation on the Stock Exchange at great cost to pensioners, and has allowed offshore investment by the funds when internal social and productive investment is so desperately needed.

  • Assets of retirement funds belong to the funds, not to “labour” or anybody else;
  • Retirement funds exist to provide their members with retirement, death and disability benefits. These are surely among the highest social purposes. “Internal social and productive investment” is a function of bonds and equities. PAR presupposes that these investments be defined and directed, in defiance of marketplace norms, as the government of the day would have it;
  • Portfolio diversification – through equities and bonds, onshore and offshore – is fundamental. In fact, the Financial Services Board insists on it through Regulation 28, which sets out the maximum proportions of assets that a fund can prudentially hold in each category;
  • Far from investment in equities being “at great cost to pensioners”, it’s been the opposite. Despite the hammering of the past year, equities have easily proven the best-performing asset class over the longer term (which is what retirement funds are all about). Look at any retirement funds’ performance over the, say, past five years, or any longer period, and perhaps the PAR proponents will get a clearer perspective of why it’s been exceptional.