Edition: Dec 2013 - Feb 2014


Credibility is key

The shine of star state witness Peter Ghavalas has been tested under cross-examination in the criminal trial of Simon Nash. What transpired is not only of historical value. For instance, who’ll ultimately benefit from the mega-millions of rand spent ostensibly in the interests of fund members and pensioners?

Imagine that you receive a call from an investment banker. “I have an idea that can make a lot of money for you and your company,” he says. “Come in and see me.”

Sounds like an offer too good to refuse. Off you go to the bank, ushered through the protocols and enter the professional office of the great man. His personal assistant brings in tea and sits you down for a power-point presentation. The banker shows you, step by step, how he can realise value in the surplus of your company’s defined-benefit pension fund.

There’s a veritable fortune to swish hither and thither. Nobody loses. Everybody gains.

The banker is proud. “It’s my intellectual property, worked out all by myself,” he explains. “What’s more, it’s perfectly legal. It’s been approved by our in-house team and by experts in pension-fund procedures.”

When you scratch your head, he suggests that you seek your own advice. Then he rattles off the names of outside actuaries, advisors, attorneys and counsel – all prominent and reputable – who’d provided favourable opinions.

By this time you’re salivating. “Where do I sign?” you can’t ask quickly enough.

And so this scene from the mid-1990s, in the Nedbank offices where banker Peter Ghavalas worked his magic on nine pension funds, shifts to the mid-2000s when employer-executives who remained in the funds and respective administrators of these funds variously paid monies back to the funds.

By this stage, all the “Ghavalas funds” were under curatorship. Some erstwhile executives of principal employers paid multiple millions of rand, in plea bargains where they admitted participation in the illegal “Ghavalas scheme”, while former administrators Alexander Forbes and Sanlam paid hundreds of millions in settlements without admissions of guilt (see particularly TT Sept-Nov 2011).

Now the scene moves arduously to the present. In a room that’s seen better days at the Johannesburg commercial crimes court, the prosecution of Simon Nash is well into its third year (TT June-Aug). Nash had elected to defend the charges against him in respect of the Sable Industries pension fund.

It’s the first, and so far only, criminal trial on the “Ghavalas scheme”. Nash is charged with such heinous offences as fraud and money laundering, related to the “stripping” of pension-fund surpluses, which potentially carry long prison sentences.

Standing in the witness box is Ghavalas himself. He’d returned from Australia to be a state witness in terms of his plea-bargain agreement that granted him a conditional indemnity from prosecution.

There appear to be two main issues. One is what made the scheme illegal, at a time when there was no certainty as to whom a fund’s surplus belonged; it was only made certain by surplus-apportionment legislation in 2001. The other is the guilt of Nash through having participated in the scheme and benefited from it.

It’s common cause that “sales and subscription agreements” between Ghavalas and the funds weren’t made to the Financial Services Board. Was such disclosure material in persuading the FSB to grant approval, under s14 of the Pension Funds Act, for the transfer of employee members into annuities while a handful of principal-employer executives remained in the fund to benefit from surplus transaction? Yes, said the FSB. No, said Ghavalas in his pre-plea bargain affidavits.

Whose responsibility was it to decide that disclosure of these agreements was material, not merely to be considered what Ghavalas described as an “accelerated contribution holiday” for the principal employer to a fund in surplus? Who was required to make the disclosures? Did non-disclosure constitute a fraudulent misrepresentation to the FSB?

Much might turn on how the court finds the credibility of Ghavalas as a witness.

Through his evidence-in-chief, he was cool and factual in describing how the scheme worked (see box). The sales agreement, he explained, “was essentially the agreement whereby the selling company sold the shares in a dormant subsidiary to the Lifecare group (totally different in ownership and management from the JSE-listed Lifecare of today). This agreement contained “the normal commercial transaction clauses”.

Ghavalas . . . abrupt turnaround

The subscription agreement, for its part, provided for commissions by way of a dividend mainly to his private family company. Dividends would come from the selling company, he said, in a “methodology to put the actuarial surplus in a stated-benefits pension fund into the hands of the principal employer of that stated-benefits fund”.

Members of the pension fund would be moved to different funds, he elaborated, leaving two or three active members in the fund. The principal employer (dormant subsidiary) of the fund would then be sold to the pension fund of Lifecare (a holding and operating company in which Nedbank had invested).

Prosecutor: Why was it necessary in your methodology to have two or three members remaining in the pension fund that was targeted?
To give substance, or disguise if you like, the fact that we were merging simply for accessing the surplus to the pension fund.
Prosecutor: What happens once the FSB issues a s14 certificate?
Ghavalas: What should have happened was all the assets of the target pension fund should have merged with the assets of the Lifecare pension fund. What in fact happened was that the surplus in the target pension fund was paid across to the Lifecare pension fund (which then) paid for the shares in the principal employer.
Prosecutor: Including the assets that would be providing for any liabilities that the pension fund still had?
Ghavalas: That is correct. Out of those assets the pension fund’s administrator would buy annuities for the pensioners and the three or four (remaining) active members would be transferred back to their original pension fund.
Prosecutor: What should have happened with the actuarial surplus?
Ghavalas: It should have been merged with assets of the Lifecare fund.

The court then focused on Ghavalas’ presentations to clients. The power-point slides showed that the scheme had to comply with relevant legislation, including fund rules, and had to be approved by the FSB.

The court: Is there anything (in your presentation) that should have been part of the fund which is not there?
Ghavalas: Yes, the three or four active members.
The court: All that reengineering would take place in order that you can create a fund that would have only pensioners and the surplus?
Ghavalas: Correct.

Beginning his cross-examination, Willem de Bruyn SC asked Ghavalas whether he’d confirm an earlier denial under oath that he’d committed any wrongdoing. “No,” Ghavalas replied. At the “particular time it was difficult” for him to admit wrongdoing, but he now admitted it.

“The wrongdoing was that I knew that the applications to the FSB did not disclose the full facts,” he said. “I did know that, in applying for the s14 (certificates), the professional parties would be misleading the FSB. I did know that the entire transaction was a smokescreen to put...the bulk of the actuarial surplus into the hands of the controlling company of the principal employer.”

De Bruyn: Where do you get the information that the FSB was misled?
Ghavalas: I read this in the various inspection reports.
De Bruyn: So if the inspection reports are wrong, your evidence here is wrong?
Ghavalas: Correct.
De Bruyn: Then you are squarely and solely relying upon the inspection reports of Cor Potgieter? (These reports had previously been dismissed by the court for their hearsay and opinion.)
Ghavalas: Correct.
De Bruyn: At the time, when the applications were made, you did not know the content of what Alexander Forbes had compiled in the application?
Ghavalas:  Correct.
De Bruyn: You believed that the transactions were in all respects lawful, and lawfully executed? Ghavalas: Correct.
De Bruyn: You had no inkling or indication that Alexander Forbes would do anything wrong?
Ghavalas: Correct.
De Bruyn: That is the basis on which you sold the transaction to, amongst others, Mr Nash?
Ghavalas: Correct.
De Bruyn: You made Mr Nash believe that it would be a valid and lawful transaction?
Ghavalas: I believed it myself.

As the three-day cross-examination concluded, De Bruyn put it to Ghavalas that he was a liar: “Last Thursday you told us ‘I made Mr Nash believe that it would be a valid and lawful transaction’. That is exactly the opposite of what you have now just told us. Which one is true? One of them must be untrue and false?”

Ghavalas: The one I have just told you is true.

The trial will proceed sometime next year. Before it does, the High Court is to hear an application by Nash on whether the FSB and curator Tony Mostert had access to privileged information following the removal of computers from his office and a breach of client confidentiality by attorney June Marks. Should the court rule in favour of Nash, it will presumably be argued that his criminal trial is unfair and will have to be abandoned.

It’s come a long way to go nowhere, not only for Nash personally but also for the whole controversy of allegedly illegal stripping of pension-fund surpluses. Many millions of rand have been spent on legal costs, out-of-court settlements, plea-bargain payments, curatorship fees and trial days.

By the end of all this, it’s hardly conceivable that there’ll be no finding on the legalities. Or that, almost 20 years after the so-called strips, costs will eclipse the payouts claimed for fund beneficiaries still capable of being traced.


This four-page circular in 1999, after the "Ghavalas scheme" transactions had been implemented, set out the FSB's position.

The Appeal Board's decision, it pointed out, "follows many years of debate and uncertainty, both internationally and locally, about... whether or not the repatriation of surplus funds in a pension fund should be permitted to a participating employer upon the winding up of the fund. In most countries

where the problem has been addressed, it has only been possible to resolve the matter through legislation". Prior to 1994, it noted, the Registrar had consistently refused to register any rule amendment that provided for the payment of any part of the surplus assets to anyone but fund members and beneficiaries.

This was based on the view that, under the Pension Funds Act, the money contributed by the employer was owned by the fund. Nothing in the Act enabled an employer to acquire an interest in or right to assets on liquidation or proposed cancellation of registration of a pensionfund, notwithstanding that the assets were in excess of the liabilities.

The circular then discusses various important court decisions, in the Lintas and Tek matters, which indicated how difficult it would be resolve surplus disputes on a case-by-case basis:

"The Registrar subscribes to the view adopted by the Supreme Court of Appeal that the matter would best be resolved by legislation." Until there was legislation, eventually promulgated in 2001, the Registrar would

consider a rule amendment providing for repatriation.

  • There were strict conditions. Amongst them: ~ There should be full disclosure to the membersas part of the offer. /n particular, the offermust specify the expected portion of surplus tobe provided to each stakeholder group, including the employer;
  • ~ At least 75% of members who contribute, oron whose behalf contributions are made by theemployer, and 75% of pensioners must acceptthe offer.

No matter that the 2001 legislation was made retrospective. Even the Registrar's pre-1994 approach offers a context for the cleverness of the "Ghavalas scheme", operating through the then Lifecare group, to be considered.



Company X buys Company Z. Both companies are principal employers of their respective defined-benefit pension funds.

If the fund of Z is in surplus, it has a positive value to Z. This value is factored into the purchase consideration paid by X for Z (in the same way that a deficit would be factored). The value to X, having bought Z, would be in subsequently merging the two funds so that X then enjoys a “contribution holiday” i.e. as the principal employer, not to make further contributions to the merged fund until the surplus is used up. 

Once used up, contributions from X would need to recommence in order that the merged fund (where X is the principal employer) is able to honour its defined-benefit promise to members. A legal “contribution holiday” is distinct from illegal dealing in pension-fund surpluses.

The distinction comes to the fore in the Simon Nash trial.

Midmacor Industries, where Nash was managing director and a minority shareholder, had a wholly-owned a subsidiary called ProBase. This dormant subsidiary was the principal employer of the Sable defined-benefit pension fund. ProBase was sold by Midmacor to Lifecare (not to be confused with today’s Lifecare company).

The sale enabled Lifecare to merge the Sable fund with its own and thus for Lifecare to enjoy a contribution holiday from use of the surplus in the Sable fund; in other words, for Lifecare to benefit from not having to make cash contributions to the fund until the surplus was consumed.

So far, so good. But the waters became murky when Lifecare took money from the merged fund to pay for its purchase of ProBase. This had not been discussed between Midmacor and Lifecare and did not form part of the agreement by which Midmacor had sold ProBase to Lifecare.

In his evidence, Peter Ghavalas said as much. He testified that neither Nash nor the Midmacor financial director could have known about Lifecare’s intended use of the merged fund’s money because it had not been disclosed to them. Neither had it been disclosed to the Financial Services Board.

It was an internal arrangement, he said, between Lifecare and its pension fund. Ghavalas was a director of the former and a trustee of the latter. He saw this arrangement as effecting an “accelerated contribution holiday” in that Lifecare, by not paying for ProBase from its own resources, instead took an advance on the longer-term cash savings it would have made through consuming the fund’s surplus.

These transactions took place around 1994, prior to the 2001 surplus-apportionment legislation. At the time Ghavalas, a senior executive in Nedbank’s Nedfin division which he represented on a number of clients’ boards, advised Nash and the Midmacor financial director to seek independent advice from the Sable fund’s administrator.

He happened to be Aubrey Wynne-Jones. Ghavalas revealed in evidence that, unbeknown to Nash until many years later, he’d paid Wynne-Jones a R720 000 commission for having introduced him to Midmacor.

Back then, in the mid-1990s, members of the overfunded Sable comprised four Midmacor head-office employees (including Nash) and about 150 pensioners. Once the pensioners were transferred into the Lifecare fund, with their concurrence, Lifecare protected their pension promises by procuring annuities from Old Mutual and Sanlam.

As it turned out, these annuities not only protected them but also provided them with a 20% excess over their defined-benefit promise. Accordingly, the pensioners suffered no prejudice and actually gained from the transaction.

What then of the four Midmacor head-office employees? They did not participate in the Sable surplus, as alleged, but in discounted proceeds from the sale of ProBase to Lifecare. These entire proceeds were used by two Midmacor subsidiaries for the purchase of equipment to help them compete against Chinese imports. That, at any rate, is how the Nash defence presents it.

  • Note: The detail here focuses on Sable, but broadly it illustrates how the ‘Ghavalas scheme’ operated amongst all nine pension funds in various companies that were party to the scheme.