No investment number is more misunderstood than the small ones. A fund charging one percent a year and a fund charging 0.1 percent look, to a beginner, nearly identical — a rounding error. Run them over forty years of contributions and the difference is staggering: the higher-fee fund can leave you with something like a quarter less money at the end, having delivered exactly the same market.
The mechanism is compounding in reverse. Fees are subtracted every single year, which means every fee also removes all the future growth that money would have produced. A one percent fee is not one percent of your wealth — it is one percent of your wealth, every year, forever.
Where fees hide
The visible fee is the fund's annual charge — the expense ratio. Around it cluster quieter ones: platform or account fees, trading costs inside the fund, entry or exit charges in some markets, advice fees layered on top, and currency conversion margins for international investing. Each is small; the stack is not.
The rule of thumb: know your all-in cost — everything you pay per year as a percentage of your money. For a simple index portfolio on a cheap platform, well under half a percent is achievable in most markets today. If your stack totals two percent, you are giving away roughly a third of a typical long-term real return before you take any risk at all.
What fees are worth paying
Cheap is not automatically right. Paying for genuine, personal financial advice at life's hinge moments — an inheritance, a business sale, retirement itself — can be worth many times its cost, especially when it prevents one large mistake. The problem is not paying; it is paying a percentage of your assets every year for advice you receive once.
The same logic applies to funds. An index fund charging almost nothing is not lesser — it is the market itself, wholesale. Decades of evidence show most expensive actively managed funds fail to beat cheap index funds after their own fees. When you do pay more, know precisely what you are buying, and check that it is not simply the brochure.
Three questions that protect you
Ask every provider: what is my total annual cost, in one number? What would this same portfolio cost at the cheapest reputable alternative? And what, specifically, am I receiving for the difference? Honest providers answer crisply. Evasive answers are themselves an answer.
Fees are the one investment variable you control completely. Returns are uncertain, markets are moody, but costs are a contract — and every fraction of a percent you keep compounds in your favour for the rest of your life. Guard it like the asset it is.
Working the arithmetic yourself
The claim that a single percentage point can consume a quarter of a portfolio sounds like rhetoric until you construct the comparison. Take two identical portfolios, both growing at the same gross rate with the same contributions, one paying a percentage point more in annual charges. Run both for a working lifetime.
The gap that emerges is not one percent of the final figure, nor is it forty times one percent. It is much larger, because the charge is taken every year from a balance that would otherwise have compounded, and the growth on the money removed is removed as well. The effect accelerates in exactly the way the growth does, only against you.
Doing this calculation once, with your own numbers, is worth more than any general statement about fees. Free compounding calculators handle it in a few minutes, and the resulting figure tends to change behaviour in a way that a percentage never does. A charge expressed as a percentage sounds negligible. The same charge expressed as the sum it will cost you over a career does not.
Percentage charges against flat charges
One structural distinction matters more than most people realise: whether a charge is a percentage of your assets or a fixed amount. The service delivered by a platform holding a portfolio is very largely the same whether the portfolio is small or large, and yet percentage charging means the cost rises without limit as the balance grows.
This produces a crossover point. Below it, percentage charging is cheaper because the percentage of a small balance is less than the flat fee. Above it, flat charging is dramatically cheaper, and the gap widens every year the portfolio grows. A great many people pass that crossover point without noticing, because nothing prompts a review and the charge is deducted automatically.
Working out where the crossover sits for your options, and setting a reminder to check when your balance approaches it, is a specific and finite piece of work. For a portfolio that will grow over decades, it is among the highest-value hours available, and the industry has no reason to prompt it.
What advice is worth and when
Not all charges are extractive, and the case for paying for genuine advice deserves fair treatment. There are situations where professional input can be worth considerably more than it costs: complex tax positions, business sale proceeds, cross-border arrangements, estate planning, and the transition from accumulating to drawing down, which is genuinely difficult.
The distinction that matters is between advice charged as a one-off fee for a defined piece of work and advice charged as an ongoing percentage of assets indefinitely. The first is a service with a price. The second means the cost rises with your balance regardless of whether the work does, and over decades it can consume an enormous amount for a relationship that may amount to an annual meeting.
Where ongoing advice is genuinely wanted, the question worth asking is what specifically will be done each year that justifies the recurring charge. A clear answer is a reasonable basis for proceeding. A vague one, or an answer that amounts to monitoring and reassurance, is expensive for what it delivers, particularly when the underlying portfolio is a handful of index funds that need almost nothing done to them.
The charges that are deliberately hard to total
A recurring frustration for anyone investigating their own costs is that no single document shows the total. Platform charges appear in one place, fund charges in another, transaction costs in a third, and any advice charge somewhere else entirely. Each disclosure is compliant and the aggregate is nowhere.
Building the total yourself is the only way to know it, and it is a one-afternoon exercise. List every account, find the platform or administration charge for each, find the ongoing charges figure for every fund held, weight them by how much is in each, and add any advice or transaction costs. The output is a single percentage describing what your money actually costs to hold.
People who do this for the first time are frequently surprised, and the surprise is almost always in the same direction. It is also actionable in a way that a vague sense of paying too much is not: with a total in hand, you can compare it against what a straightforward alternative would cost, and the difference is the annual sum at stake.
Legacy accounts are where the worst charges live
The single most expensive holdings most people own are usually in accounts they have forgotten about. A pension from an employer two jobs ago, an investment product bought a decade ago on someone's recommendation, an account opened for a specific purpose and never revisited.
These carry the charging structures of the era in which they were opened, which in many cases are several times current levels. Nobody has any incentive to draw attention to this, and the annual statement is typically designed to report performance rather than to make the cost prominent.
Locating and reviewing these is genuinely worthwhile. Consolidation is frequently but not always the answer, since some older products carry valuable guarantees or benefits that would be lost on transfer, and exit penalties exist on some contracts. The point is to find out rather than to act automatically. The distinction between an expensive old product and a valuable one is not visible from the outside and is usually discoverable by asking the provider directly.
What a reasonable total actually looks like
It is fair to ask what the target is, having established that the total is worth measuring. A portfolio of broad index funds on a competitively priced platform, self-managed, can be held for a total annual cost that is a small fraction of one percent, and this is available to ordinary retail investors in most developed markets without any special access.
Above that, each additional increment should be buying something identifiable. Active management, specialist exposure, ongoing advice, a platform offering something the cheap one does not. Any of these may be worth it; the test is whether you can name what you are getting.
The figure that should prompt investigation is a total in the region of two percent or more, which was common a generation ago and persists in legacy products and some advised arrangements. Over a long horizon a charge at that level consumes a very substantial share of the eventual outcome, and in most cases a cheaper alternative delivering substantially the same exposure exists a form away.
The one cost worth paying more for
Having argued consistently for minimising charges, there is a legitimate exception worth naming. If a slightly more expensive arrangement is one you will actually stick with, and a cheaper one is one you would abandon or mismanage, the more expensive one is better.
This applies most often to people who genuinely cannot leave a portfolio alone. Someone whose history is of selling during every decline may do better in a more expensive arrangement that removes the ability to interfere than in a cheap self-managed one they will damage. The behavioural cost of a badly-timed exit exceeds decades of fee difference.
The honest framing is that this is a real consideration and it is also the argument every expensive arrangement makes about itself. The test is your own record rather than the claim: someone who has held through a previous decline has evidence they do not need to pay for that protection. Someone who has not may be buying something worth having. As with everything on this site, this is educational rather than advice, and the right structure depends on the person operating it.
How the charge is actually taken
Part of why fees go unnoticed is the mechanism by which they are collected. Fund charges are typically deducted from the fund's assets daily, before any price is published, which means the price you see is already net of them. Nothing ever appears as a payment, and no statement shows a line item.
Platform charges are more visible but often taken from cash within the account rather than billed, which means they too are absorbed without any transaction that prompts attention. Advice charges deducted from the portfolio work the same way. The result is a set of substantial recurring costs, none of which ever requires you to actively part with money.
This is worth understanding because it explains why people who scrutinise every household bill frequently have no idea what their investments cost. The design is not deceptive in any regulatory sense — all of it is disclosed — and it removes every psychological cue that would ordinarily trigger review. Substituting a deliberate annual check for the absent cue is the only remedy.
What to do with the number once you have it
Having totalled your costs, the useful next step is to convert the percentage into a currency amount and then into an annual figure you would notice. A percentage of a six-figure balance stops sounding abstract the moment it is expressed as a sum per year, and comparing that sum to things you deliberate over is clarifying.
The second conversion is forward-looking: what the same charge costs across the remaining years to retirement, including the growth forgone. This figure is invariably far larger than the annual one and it is the number that should inform whether a change is worth the paperwork.
Armed with both, the decision becomes straightforward rather than agonising. A transfer that takes a few forms and saves a substantial recurring amount is obviously worth doing. One that saves a trivial amount and risks losing a valuable guarantee is not. The arithmetic settles it, which is the entire reason for doing the arithmetic rather than relying on a general sense that fees matter.