Investment platform RA fee structures: removing the fund manager from the equation

The uncomfortable truth about retirement in South Africa - Rands and Sense by 10X [video]

We sit down with 10X Investment Consultant lead Andre Tuck and discuss the retirement savings crisis in South Africa. We also delve into living annuities, retirement annuities, TFSAs and everything in between. Read more

Four fee structures into this series, and every provider so far charges for somebody's judgement. The fifth charges for none of it. R2,750 in the first year buys an account, a set of statements, and nobody's opinion at all on what to put inside it. No advisor. No fund manager. No recommended portfolio arriving attached to the quote. The self-directed model charges for the wrapper and leaves the judgement about where the money sits to the person paying.

A quick recap for anyone joining partway through. Five South African retirement annuity providers were set the same benchmark, a R500,000 lump sum measured over thirty years on identical assumptions. In the first year, the most expensive structure cost six times the cheapest. By year thirty, the difference between them came to roughly R1.8 million. This series has worked through , which charges upfront for a relationship with an advisor, , where the manager's own pay rises and falls with the fund's returns, and , two flat charges printed one above the other on the same quote.

A note on sourcing before the numbers. This provider is a share and unit trust trading platform with a retirement annuity wrapper attached, and, unlike every other provider in this series, it did not respond to written questions about its fee structure, despite two attempts. The figures below are drawn from its own published cost profile, the same approach flagged in the opening episode's methodology for the provider that did not respond.

The anatomy of the platform fee

This structure charges two fees for holding the investment and two more for what the investor does with it. 

• A retirement fund administration fee of 0.30% a year, charged on the net asset value of the account. This is what keeps the account open, administered, and compliant, independent of what sits inside it.
• An annual unit trust platform fee of 0.25% a year, charged on the value of unit trust holdings specifically. Added to the administration fee, the base cost comes to 0.55% a year. That is the cheapest headline of the five structures in this comparison.
• No fee for the decision itself. Neither charge covers fund selection, because the platform is not the one selecting. The 0.55% buys administration. It does not buy a view on whether what the investor has chosen is any good.

Two further costs sit outside that 0.55%, and neither appears in the headline rate.

• A broker commission of 0.25% per transaction, with a one cent minimum, charged when the investor buys or sells shares or ETFs directly rather than holding a unit trust. It does not apply to unit trust purchases, but it does apply the moment an investor builds a portfolio out of individual instruments, which is the platform's core business.
• The underlying fund's own total expense ratio, set by that fund's manager and built into its unit price, not disclosed or controlled by the platform. A fund with a TER of 1.5% adds 1.5% to the true all-in cost, on top of the 0.55% headline.

With no house-recommended fund to quote, the closest analogue on the platform is a balanced unit trust in the same high-equity multi-asset category the other four providers were asked about. It returned 30.25% over the twelve months to April 2026 (which is a very large number and, while in all likelihood not wrong, should prompt caution in an investor). Three- and five-year figures are not published for it (and would probably be much lower), a direct consequence of a structure that leaves fund research to the investor rather than compiling it centrally.

The cheapest structure doesn't finish cheapest

The projection below holds the 0.55% base rate for all thirty years and assumes no trading, which is the most favourable case this structure can be modelled under.

Even on these assumptions, the cheapest opening fee does not necessarily produce the cheapest outcome. Year 1 costs R2,750 here against R4,373 for the layered fee. By year thirty, the layered fee has finished R72,000 cheaper in total. The base rates never changed. What changed is what each rate covers. The layered fee keeps everything inside its 0.81%, while this structure leaves the underlying fund's TER and every transaction cost outside its 0.55%, compounding for three decades in a number nobody quoted.

The case for doing it yourself

• The cheapest entry point in the comparison At 0.55% and R2,750 in Year 1, no other structure in this series starts lower. For an investor holding a single low-cost unit trust and never trading, that base rate is also close to the true cost, because little else is stacked on top of it.
• No fee for a person the investor may not need The advice-led model in episode two charges 0.86% a year for an ongoing advisor relationship, whether the investor wants it or not. This structure charges nothing for that, because it offers nothing in its place. For an investor who already knows what they want to hold, that absence is a saving rather than a gap.
• The widest choice in the comparison Every other structure in this series narrows the investor to a recommended fund or a house range. This one does not. An investor who wants a specific index tracker, a specific sector tilt, or a combination across several unit trusts and ETFs can build that directly, which none of the other four structures allow within their standard offering.
• No cost to transfer, but only in cash The provider charges nothing to move a retirement annuity in or out, which is rarer than it should be. What it does not accept is a unit transfer. A Section 14 transfer into this platform has to arrive as cash, so the existing holdings are liquidated first. That leaves the investor out of the market for a stretch, watching from the beach while the tide does whatever it does. No tax is triggered by the transfer itself, though the old provider may charge its own exit fee.

Savvy RA investors let SARS fund their Tax-Free Savings Account (or their kid's one)

Being responsible with your retirement annuity (RA) could mean you get a maxed out tax-free savings account for free. Here's how. Read more

The case against

• Nobody is checking the investor's decisions The four structures before this one all included a professional in the price, whether or not that professional earned it. Here the role does not exist. That matters because the mistakes retail investors make are reasonably well documented, and has found that holding too few positions and trading more often than a plan requires both track with weaker returns. Whether oversight prevents either habit is not something that research tests. What it does establish is that both are common, and neither shows up as a fee or on a statement.
• Investors earn less than the funds they own Morningstar's annual study measures the difference between what a fund returns and what the average rand invested in it actually earns. Over the ten years to December 2024, the gap ran to 1.2 percentage points a year, more than twice this structure's entire headline fee. The size of that gap is disputed, and a argues the portion attributable to poor timing is far smaller. What is less disputed is where the gap concentrates. Morningstar found investors in multi-asset allocation funds, the kind the other four providers in this series recommend, lagged by 0.1 percentage points. Investors picking their own sector funds lagged by 1.5.
• A portfolio built without guidance can end up concentrated by accident The ten largest holdings in a Satrix Top 40 tracker , and Naspers alone takes close to nine percent of it. Build a local equity portfolio without thinking hard about it and you end up owning Naspers, a few banks, and some gold, whether that was the plan or not. Diversifying out of that is possible on this platform, and the tools are all there. It just has to occur to the investor to do it.
• The all-in cost is the one figure this structure cannot quote Every other provider in this series could state a total expense figure for R500,000 in a named fund. This one could not, because the total depends on a fund the investor has not chosen yet. The 0.55% headline is real, but it is a floor rather than a ceiling, and how far above it any individual investor ends up is not something the platform's own disclosure can answer in advance.

Retirement Annuity vs Tax-free Savings Account vs Bond: Where Should Your Extra Money Go?

Got extra money to invest? Compare three powerful options: Retirement Annuity, Tax-Free Savings Account (TFSA), or paying off your bond faster. Learn the pros and cons of each to make an informed choice for your financial future. Read more

The investor experience

The complaints about this provider are not about the platform. They are about getting money out of it. Getting in takes a few taps and a bank verification. Getting out is why 129 people went to HelloPeter to complain.

The numbers behind that pattern: 1.1 out of 5 on HelloPeter across 129 reviews in the twelve months to April 2026, with 93% of them one star, and a Net Promoter Score of -95. That is the lowest score of any provider in this series, below even the advice-led provider's -77.

There is a structural explanation available, and it is worth naming because it is not entirely a criticism. This is a trading platform first, built for high-frequency, low-stakes transactions, with a retirement wrapper added to it. Retirement withdrawals are the opposite of that, infrequent and consequential. Whether the review pattern reflects a support model built for the wrong use case, or something else, is not something review data alone can settle.

Who it's for

This structure rewards a specific kind of investor. Someone who already knows what they want to hold, is comfortable researching and monitoring it without help, and trades rarely enough that the 0.25% per-transaction cost never becomes material. For that investor, the 0.55% headline is close to the real cost, and the absence of an advisor fee is a genuine saving rather than a missing service.

Every departure from that costs more. Trading adds 0.25% each time. A more expensive fund adds its TER. Both sit outside the number the platform quotes, so neither arrives as a bill anyone has to look at. The structure asks the investor to be disciplined and charges nothing extra when they are not, which is a different proposition from the four before it, where the cost of the professional was the first thing on the page. Nobody sends a reminder. Nobody phones in March.

How it compares

The headline row and the thirty-year row tell different stories.

Nothing here is hidden. Every cost outside the 0.55% is published in the provider's own cost profile, in a table anyone can read. It just isn't in the headline, and the headline is what gets compared. That is how the cheapest structure in this series finishes third over thirty years, behind two structures that quote a bigger number and mean it.

The four structures before this one all charged for somebody's work. The research, the fund selection, the discipline to leave it alone for thirty years, all of that work now sits with the investor, and none of it appears on any statement. Whether that is a saving depends entirely on who is doing it.

That leaves one structure standing. The last episode turns the same questions on 10X, §§§§, and the provider publishing this series. Same benchmark, same thirty years, same scrutiny. No exemption for the home team.