Fees are the only part of an investment return you can predict

There is a settled way of talking about investment costs. It is worth asking how much of it survives contact with the detail.

The arithmetic is brutal and certain

A one per cent annual charge does not cost one per cent of your return; it costs one per cent of your entire balance, every year, compounding. Over a multi-decade horizon that removes a substantial fraction of the final value, and the exact fraction depends on time rather than on markets. Because it is deducted regardless of performance, it is the only element of the equation with a guaranteed direction.

This is why cost is the first thing experienced investors look at and often the last thing marketing mentions.

Count every layer

The headline ongoing charge figure of a fund is rarely the whole cost. Platform or custody fees, transaction costs inside the fund, bid-offer spreads and foreign exchange charges all sit on top.

For most households, adding them gives a total cost of ownership, which can be several times the number in the advertisement. Comparing funds on the headline figure alone routinely produces the wrong ranking.

Cost predicts relative performance better than past returns do

Repeated analyses across decades have found that low-cost funds outperform high-cost funds in the same category more often than not. Past performance, by contrast, has consistently weak predictive power, which is why the warning is a legal requirement. This is not an argument that active management never works; it is an observation that the fee is a certain headwind and the skill is not certain.

Published tables also flatter the survivors, because funds that did badly are commonly closed or merged into others, so the ten-year list you are shown is not the list an investor was actually choosing from ten years ago.

Where paying more can be defensible

Genuinely specialist strategies, illiquid asset classes and some tax structures are not available cheaply and may justify their cost. Advice is a service with its own value, and paying for it is separate from paying for fund management.

The question worth asking is what specifically the extra cost buys and whether it is a thing you need. Performance fees repay close reading, since whether the charge applies to any gain or only to a gain above a benchmark, and whether past shortfalls must be recovered before it restarts, moves the real cost far more than the headline percentage does.

Reducing costs is the easiest change available

Moving to a lower-cost equivalent fund or platform is administrative rather than strategic, and takes an afternoon. Check for exit charges and, if the holdings are in a taxable account, for any tax consequence of selling before switching. It is one of the few decisions in investing where the outcome is knowable in advance.

Ask whether the receiving provider can move the holdings across as they are rather than selling and rebuying, because that avoids both a disposal and a spell out of the market, and where it cannot, the length of that gap is a real risk to weigh against the saving.

Assume any product feature can be withdrawn at renewal.

Percentage fees and flat fees cross over

A platform charging a percentage of assets costs more as the balance grows, while a flat annual fee becomes a smaller proportion of it. There is a balance at which the two swap places, and it is worth calculating rather than assuming, because the cheaper structure for a small pot is frequently the expensive one for a large pot.

Some platforms cap the percentage above a threshold, which moves the crossover point and is rarely prominent in comparison tables. Charges are deducted from the fund or sold out of the account rather than invoiced, so nothing ever arrives to tell you the arrangement stopped being the cheapest one.

The takeaway

You cannot control returns. You can control costs, and they compound the same way.

The decision is rarely about picking the best option — it is about avoiding the expensive one.