"It is safe to assume that all life assurance investment products are expensive in relation to unit trust investments."

A single premium life insurance policy can be cheaper that a similar unit trust investment. The unit trust industry is almost exclusively a single premium investment industry. Less than 3% of the inflow into unit trusts is in the form of recurring contributions.

The only fair comparison of cost between a life policy and a unit trust is therefore on single premiums, where policies can be cheaper than unit trusts.

It is true that a recurring premium RA policy with up-front commission will be more expensive than a recurring contribution to a unit trust, where commission is paid on an "as and when" basis. The truth of the matter is that virtually nobody uses unit trust as a vehicle for recurring contributions.

You can also get RA policies with recurring commission and the cost of these will compare well with unit trust costs.

Rusconi’s paper also ignored the effect of switching fees. As a person moves closer to retirement, his asset allocation should typically change to reduce exposure to the equity markets. Retirement annuities generally allow for switching between funds at zero cost, and also no bid/offer spread. With unit trusts, switches between different unit trusts would normally attract significant fees - up to 5% - and this has been ignored in the paper. Unit trusts have not been designed to facilitate long term, recurring premium wealth accumulation.

Sales people prefer to promote life assurance products because they receive commission upfront, and not on an "as-and-when" basis as they do with unit trusts.

The truth of the matter is that the financial advisor needs to be paid for the financial advice that they provide to their clients.

While some financial advisers may be able to adapt to as-and-when commission, the majority of financial advisers will find it extremely difficult to survive on this basis, particularly those who have recently joined the industry with no a client base.

In terms of the FAIS Act financial advisers must conducted a full financial needs analysis as part of the advice process. 3% as-and-when commission payable on a R200 RA contribution (as referred to in the article would amount to R6 per month. Even on a R500 contribution the commission would be R15 per month. No financial adviser will be able to spend the time required to do proper financial planning for this level of remuneration.

The good news for policyholders is that commission is fully disclosed and is also negotiable. It can therefore be negotiated to be a fair reward for the professional advice provided.

"In some cases, high cost are reducing the retirement benefits from RA’s by almost 45%."

The measure of cost comparison used in Rusconi’s paper on which this exaggerated stamens is based, is the so-called "charge ratio", which could also be called "reduction in maturity values".

The problem with this measure is that it is very dependent on the term of the investment. The longer the term, the larger the charge ratio will be, even if the regular annual charge is very low and remains unchanged.

To say that an investment has a charge ratio of 45% is a meaningless statement without considering the investment term. And to deduce that a charge ratio of 45% implies high costs in absolute terms is even more meaningless.

For instance, all people will agree that an annual fund charge of 1% per annum is a low charge. (The UK’s attempt to limit the charges on their Personal Pensions to 1% by law was a dismal failure, since no company could afford to do business at such a low cost).

However, a fund charge of only 1% will lead to the following charge ratios, depending on the term (on a single premium investment).

This table shows that, given a long enough term, even a very low 1% annual fund charge will lead to a very large charge ratio. That is the nature of compound interest.

Personal Finance only quoted charge ratios for extreme cases of 40 year terms, which gave a very distorted picture of the truth. In fact, 65% of RA’s are sold at ages above age 40, with terms less than 20 years. Less than 5% of RA’s are sold below age 25, and even then the terms are normally less than 40 years.

"Rusconi’s research compares cost of RA funds bought by individuals and RA unit trusts products for individuals."

Rusconi used the cost of his own direct investment into a rather obscure Old Mutual Unit Trust RA product. He did not use a financial adviser to provide him with advice on asset allocation, alternative investment funds available, the contribution required to maintain his living standard after retirement, etc. However, most South Africans require financial advice, and financial advisers need to be remunerated for their service. The product that Rusconi used did not provide for any inherent distribution cost. It is therefore not surprising that it was cheaper than a normal RA policy that makes provision for the cost of the advice and distribution .

This is almost like comparing the cost of milk bought at a supermarket with the cost of milk bought at the dairy farm where you must bring your own container and fetch it yourself. Of the R10 you pay for 2 litres of milk at a supermarket, about R4 reaches the dairy farmer. The rest of the cost is for packaging, transport, distribution, etc. It is, however, an essential service to the public. If buying directly at the farm was the only way that milk was distributed, very few people would get milk.

To present a product with virtually no provision for distribution cost as a solution to the problem is not meaningful. The sales of the product used by Rusconi are really insignificant – there are less than 400 members making regular contributions to this fund.

Old Mutual seems to be the only company providing unit trust investment directly as an RA product. The common way to hold unit trusts in an RA wrapper is via a LISP (Linked Investment Service Provider), which normally has much higher charges to those used by Rusconi in his research.

The milk analogy can also be used to illustrate the meaninglessness of the "charge ratio" when not used in context. If you can save R6 per day on 2 litres milk by buying the milk directly at the dairy farm, then over a 40-year period you can save R87 000. Imagine the sensational newspaper headline : "Supermarket rip-off : milk costs R87 000 more!"

"A valid concern is that the main cause of these high charges is a lack of transparency and a lack of competition between companies."

This allegation has no substance. There is certainly no "lack of transparency" in the insurance industry. All companies are compliant with the Policyholder Protection Rules which require full disclosure of all charges. Furthermore, in terms of the current Benefit Illustration Agreement (BIA), it is best practice to reduce the effect of all the charges into a single figure called the "Reduction in Yield (RIY); and most companies provide this on their quotes.

In terms of the new Code on Policy Quotations (that will replace the BIA next year), it will be compulsory to illustrate the Reduction in Yield very prominently on each quotation. This is a much better and more actuarially sound measure for cost comparisons than the "charge ratio" used by Rusconi which is extremely dependent on investment term.

There is also no lack of competition in the industry. There is in fact substantial competition between companies on cost and many intermediaries routinely perform comparisons of the cost of various products, on a reduced yield basis.

"Life assurers express cost as a mismatch of fixed-rand cost, as percentage of assets, as a percentage of premiums, on an annual basis and up-front."

The reason for these various charges is to reduce cross-subsidies between policies of different premium sizes and length of term. With the exception of fixed policy fees, unit trusts have a very similar charging structure. As with policies, they typically have a buy/sell spread (difference between buying and selling price of units) as an up-front charge and a fund charge on an annual basis and a trail fee to pay commission to intermediaries.

The LISP and unit trust industry have no agreement like the LOA Code on Policy Quotations requiring cost disclosure in any specific format. With the focus on Reduction in Yield and the changes that have been agreed to the Code, the life industry is really in the forefront on the issue of cost disclosure.

The article quotes the cost structure of an RA with a starting premium of R200 per month.

The RA policies provided by insurers are the only products freely available at such low premiums. The minimum recurring contributions accepted by unit trusts is typically R500 or R1 000 per month (and the minimum for LISP is typically R1 000 per month or higher.)

The life assurance RA is therefore able to meet the retirement funding needs of people whose tax deductibility of contributions is low, or people for whom affordability is a problem. The same cannot be said of the other alternatives. It is not very meaningful to present a product that is not available in the middle market as the solution to the retirement funding problem.

The articles create the very dangerous impression that unit trusts are a better way of saving for retirement than RA policies. The fact that on most unit trusts there is no tax deductibility of contributions (as with an RA) is not pointed out clearly. The impact of this could be far greater than any cost difference highlighted in the article. Rob Rusconi used a very specific "RA unit trust" available only from Old Mutual, but the articles may create the impression that this applies to unit trusts in general.

Should individuals switch to normal unit trusts after reading the sensational article, the effect could be disastrous. Normal unit trusts are not protected from creditors in the case of insolvency, and RA’s are predominantly used by self-employed individuals, who run a much larger risk of becoming insolvent.