The fees debate is too important to be sensationalised, says Martin Steward.

IPE strongly supports initiatives that highlight the impact of high fees on individuals' savings assets and eventual pension income. That fees are high cannot be disputed. Here in the UK especially, individual defined contribution schemes and the underlying investment products in them saddle savers with charges well in excess of what they would be paying in collective or defined benefit schemes.

But for that very reason, it is also important that those who campaign for fairness on these issues do so responsibly. Unreliable numbers put out to generate sensational headlines are counterproductive. Not only are they likely to frighten people away from saving for retirement at all, they also enable industry lobby groups such as the Investment Management Association to issue statements dismissing their "sensationalism" and "discredited research".

This week has seen a great deal of this going on in response to the Royal Society of Arts' "Seeing Through British Pensions" report, which criticised the extent to which the UK's savers are being misled about costs.

Among the more egregious responses was one from SCM Private, a company that offers ETF-based multi-asset funds for wealthier investors for fees ranging from 0.3% to 0.75% plus VAT, and which initiated the 'True and Fair Campaign' to highlight the issue of investment fee transparency in the UK. 

The 'True and Fair' objective is admirable. Less so was the following offhand statement from SCM Private co-founder Gina Miller: "Of those fees that are disclosed, like the annual management fee, a difference of 1% or 2% doesn't seem like much. But over 25 years, a pension that charges the extra 1% AMC will cost the pensioner 28% of their original investment in these extra charges."

The charge, then, is that a pension pot would be 39% larger if the saver paid no fees than it would be if she paid a 1% fee.

The essential point is a good one: the important thing about fees is the way they compound over time. It is unfortunate, therefore, that Miller makes a basic error about compounding. The figure of 28% is what you get if you simply compound 1% growth over 25 years. But we're not talking about growth, are we? We're talking about a reduction. If you are charged 1% each year, clearly the base sum on which that charge is calculated next year is 1% lower than the base sum on which it has been charged this year (and so on for 25 years). Your pot would have been 28% bigger had you not been charged 1% (not 39% bigger), but it has "cost" you 22.2% of your "original investment" (not 28%).

Well, 500 basis points - what's the big deal, right? That's fair enough, but it's also important to model some realistic assumptions about how people save for their retirement, too, because, otherwise, you leave yourself open to criticism from the industry that your examples don't adequately reflect the costs they face to meet those savers' needs.

So, who on earth puts a single lump sum into their pension pot to save for 25 years? Most people would save a bit each year over the course of that time. Put a lump sum of £10,000 in, pay 1% per annum in fees and assume zero real return, and you lose 22.2% of your money, or £2,221.79. Put £10,000 in over the course of 25 years as 25 annual instalments of £400, pay 1% per annum in fees and assume zero real return, and you lose "only" 12% of your money, or £1,201.73.

Under this scenario, a saver paying zero fees ends up with a pension pot that is 14% bigger than the saver who has paid 1% - not 39% larger.

The Royal Society of Arts' report also puts forward a scenario comparing the pension pots of a fee-paying saver and a saver who gets the service for free. It argues that the latter gets a pension income that is 60% larger than the former's.

Specifically, it says that a 25 year old who puts £1,000 away each year, raising that amount by 3% each year to account for inflation, and who enjoys a 6% nominal return on his investments, will have £248,170 at the age of 65 and an income of £16,080 until he reaches the age of 85, if he pays no fee. The same saver paying a 1.5% fee would only get an income of £9,900 between the ages of 65 and 85, the report argues.

I scratched my head long and hard over this finding. My calculations suggested that the fee-paying saver would have a pot worth about £181,000 at age 65, and if he bought an annuity on the same terms as the RSA's non-fee paying saver, he'd get an income of about £11,600 - meaning that the non-fee paying saver's income would be 40% higher, not 60% higher.

David Pitt-Watson, co-founder of Hermes Equity Ownership Services and a Fellow of the RSA, called me to explain that this model does not assume annuitisation. Instead, the annual incomes described were what the savers could take if they assumed an ongoing 6% return on their pots (and for the fee-payer, an ongoing 1.5% fee), until they ran out of money at the end of their 85th year.

So, even under the RSA's assumptions, the size of the non-fee paying saver's pot at age 65 is indeed 40% higher, not 60% higher. The 60% higher income appears thanks to the unrealistic assumptions about what happens in retirement.

Let's leave aside the issue of annuitisation versus income drawdown and go down the RSA's route of assuming drawdown. Pitt-Watson suggests that the saver keep his pot invested in a relatively high-risk (6% return or 3% real return) fee-paying vehicle throughout retirement - or at least that he keeps it in the same vehicle as the one he used to build his pot. In absolute terms, this actually makes the fee-paying saver's situation look a bit better because, obviously, there is a bigger-than-realistic return to absorb the fee (he is still making 1.5% per annum after fee and inflation).

But relative to the non-fee paying saver, it makes his situation look worse. I would think it's more realistic to assume that the saver would put the money into a low-cost inflation-linked bond portfolio. I assume a fee of 0.25% for this linker portfolio (iShares' sterling linker ETF has a TER of 0.25%), and a conservative zero real return - which is actually not far off reality at current real yields.

Under those assumptions, the fee-paying saver gets an income of £8,394. If we also assume that the saver who paid no fees in the saving period also gets his linker portfolio with a zero fee, then we find that he gets an income of £11,973, which is 43% higher than the fee-paying saver's retirement income.

And this is without challenging the realism of the 'something-for-nothing' assumptions made for the non-fee paying saver, of course. If we assume he has to pay 0.25% for his linker portfolio in retirement and 0.3% for his investment portfolio in accumulation, he gets an income of £11,194. This is 34% higher than the saver who had to pay fees of 1.5% in accumulation and 0.25% in retirement.

Are these more realistic assumptions? I think so. And if we turn to NOW:Pensions, the division of Denmark's ATP that is making a play for the UK defined contribution market, we find something similar. In the scenario NOW:Pensions paints in its pitch material for employers, it assumes a 25-year-old saves 8% of her annual salary for 40 years with a starting salary of £26,000, with 4% annual salary increase and annual fund returns of 7%. Seems reasonable. It compares the pot she would get paying NOW:Pensions' fee of 0.3% plus a £1.50 annual administration charge with the one she would get paying a fee of 1.5%. The difference? 31%.

That is a model of how to present the fees argument: realistic assumptions, sensible numbers.

And, as I said at the top, it is important to present the argument in this way - and unnecessary to sensationalise. It is no more acceptable to have 12% lopped off your pension pot than 28%. It is no more acceptable to lag a low-fee saver by 30% than 60%.

Even if I'm selling my house, I'd struggle to get all the official fees, legal fees, estate agent fees and taxes to add up to more than 5% of the value of that house. Why on earth should I expect to pay out so much more than that just to invest my retirement money?