Two colleagues put €200 a month into the same boring stock fund for thirty years. Same fund, same dates, same amounts. The only difference is the envelope: one holds the fund inside a tax-sheltered retirement account, the other holds it in a plain brokerage account.
At the end, the first colleague keeps everything. The second writes a cheque to the tax office for tens of thousands of euros, out of money that was otherwise identical. They made the same investment. One of them just made it inside the right envelope.
This article is about that envelope: what it is, why it is one of the few genuinely free advantages in personal finance, and then the harder question underneath it, the one the industry would prefer you never ask. Not "should I use a retirement account", but "who should decide what goes inside mine".
The envelope is worth real money
Start with the arithmetic, because the size of the effect is the whole argument. Assume €200 a month for thirty years, growing at 7% a year, roughly the long-run average of a broad stock index; an illustration, not a promise. Assume the unsheltered account pays a flat 26% tax on its gains at the end, in the neighbourhood of what several EU countries charge on investment income. The sheltered account pays nothing.
Both colleagues contribute €72,000 of their own money over the thirty years.
| After | Sheltered account | Taxed account (26% on gains) | Cost of the envelope |
|---|---|---|---|
| 10 years | €34,617 | €31,857 | €2,760 |
| 20 years | €104,185 | €89,577 | €14,608 |
| 30 years | €243,994 | €199,276 | €44,718 |
Almost €45,000, for filling in a different form at the start. And this is the gentle version of the comparison. If instead the tax arrives every year, the way dividend taxes do, it is worse, because each year's tax is money that can no longer compound. Under the same assumptions, if just two points of that 7% arrived as dividends taxed annually at 26%, the drag alone would turn the €243,994 into €220,378 before any final tax on the remaining gains. Tax drag compounds, exactly like fees do, exactly like returns do. The wrapper removes it, and lets compounding run unhindered.
For most people who have access to one of these accounts, using it is as close to a no-brainer as personal finance gets. What follows is about everything else, but do not let the arguments ahead blur this first point.
A wrapper, not a product
Here is the thing almost nobody explains plainly: a retirement account is not an investment. It does not grow, earn, or lose anything by itself. It is a legal wrapper around ordinary investments, the same funds and shares you could buy anywhere, that changes how they are taxed. Think of it as a diplomatic passport for your money: the money itself is unchanged, but the usual rules stop applying to it.
That distinction sounds like pedantry. It is actually the key to this entire article, because once you see the account as a wrapper, two questions that the industry likes to blur together come apart cleanly:
- Should you use the wrapper? (Almost always yes, per the table above.)
- Who decides what goes inside it? (Genuinely open, and worth a fight.)
Banks and fund companies benefit enormously when you treat these as one question, because then "opening a retirement account" silently means "handing us the contents". Keep them separate. The tax break comes from the state and attaches to the wrapper. The bank's fees attach to what they put inside it.
What the wrapper is called where you live
The wrapper exists in most rich countries, under different names and rules. The details below are approximate and current as of early 2026; caps and rules change, so check the current figures for your country before acting.
| Country | Wrapper | What it does, roughly |
|---|---|---|
| UK | ISA | Gains and dividends inside are tax-free forever; an annual contribution cap in the tens of thousands of pounds; money accessible any time |
| UK | SIPP / workplace pension | Contributions get tax relief up front; money locked until near retirement age |
| US | 401(k) | Through an employer, contributions taken pre-tax; often comes with an employer match |
| US | IRA / Roth IRA | Opened individually; the traditional version defers tax, the Roth version taxes now and never again |
| France | PEA | European shares held long enough escape most income tax on gains; an annual-contribution logic with a lifetime-style cap in the low six figures of euros |
One line in that table deserves its own paragraph. Where an employer match exists, as it commonly does with a 401(k) and with many workplace pensions, your employer adds money on top of what you contribute, often euro for euro up to a limit. That is not a return, a projection, or a maybe. It is doubled money, handed over in the payslip. Under the same assumptions as before, €100 a month of your own grows to about €121,997 in thirty years; matched to €200 a month, about €243,994. Nowhere else in finance does anyone double your stake on day one just for showing up. If a match is offered to you and you are not taking it, that is the first thing to fix, before anything else in this series.
The real question: who runs the inside
Now the part the brochures skip. Once the wrapper is open, someone has to decide what it holds. There are two broad answers.
The default answer is that an institution decides. You are placed into their pension fund or managed portfolio, a professional allocates it, and a management fee comes out every year whether the year was good or bad. Most people take this path without knowing they chose anything.
The other answer is that you decide. Most wrappers, not all, let you hold what you pick: index funds, individual shares, bonds. Same wrapper, same tax break, but the contents are yours. The industry calls this "self-directed" with a faint tone of concern, the way a parent says "so you're going alone".
The account is the wrapper; the argument is only ever about the contents. Whoever wins that argument, the tax break is yours either way. Never let a quarrel about the inside cost you the envelope.
Notice what the DIY question is not. It is not "should I skip the retirement account and use a normal brokerage instead". That choice hands €44,718 back to the tax office for the privilege of feeling independent. The question is only ever what goes inside the same account.
The case against handing over the inside
Why would anyone take the wheel themselves? Not for fun. For three reasons, each of them documented rather than dark-hearted.
First, fees compound against you exactly the way returns compound for you. We walked the arithmetic in The myth of the financial expert; here is the shape of it again with this article's numbers. A management fee of 2% a year turns an assumed 7% into 5%. On €200 a month for thirty years, that is €243,994 shrinking to €166,452. The fee takes €77,542, which is roughly a third of your final pot, and closer to half of your actual investment gains, for a service that decades of SPIVA scorecards show mostly fails to keep up with a plain index over long periods. A percentage point sounds like a tip. Over thirty years it is a wing of your house.
Second, the incentives are not aligned, and this is structural rather than a matter of bad people. A salesperson paid by commission on the products they place is paid to place products, not to maximise your outcome. Funds are routinely recommended by the institutions that earn fees from them. Regulators in the UK and EU have forced fee disclosure and banned some commission structures precisely because the conflicts were real enough to legislate against. When the advice is free, you are not the customer.
Third, when the stakes were highest, the industry's conduct is a matter of public record. In the years around 2008, major institutions packaged and sold mortgage investments that failed catastrophically, ratings agencies had stamped many of them with their highest grades, and in some documented cases sellers were simultaneously betting against products they were recommending. The settlements that followed, with US and European authorities, ran into the tens of billions of dollars. None of that is a conspiracy theory; it is court filings and regulator press releases. It does not mean your local advisor is a villain. It means "they are professionals, surely they have my interests at heart" is not a load-bearing assumption, and history says so in writing.
Running the inside yourself removes the fee drag and the conflicted middleman in one move. It does not require genius. As we argued in Index funds and the spectrum of investing, the humble version of DIY is choosing one broad, cheap index fund inside your wrapper and adding to it monthly. That is self-directed investing. It takes an hour a year. The ambitious version, picking businesses yourself, demands far more, which is what Stock picking: the honest version was about, and it only makes sense if you are willing to actually read the businesses you own. That reading work is exactly what tenbagger, the workspace this blog is part of, is built for. But the hour-a-year version alone already fixes the two biggest leaks.
Do not let anger cost you the wrapper
Here is where this article parts ways with the angrier corners of the internet, and it matters enough to be its own section.
Some people read the previous section, conclude the whole system is rigged, and refuse to touch anything with "pension" or "retirement" in the name. They keep their money in a current account, or in a plain taxed brokerage, as a kind of protest. The protest is understandable. It is also expensive, and the bill is in the first table of this article: €44,718 of avoidable tax on our modest example, handed to no bank at all, simply forfeited.
Be precise about what the evidence indicts. The fee-charging, conflicted, occasionally scandalous part of the industry is the management of the contents. The wrapper is not that. The wrapper is a tax rule, and it works identically whether the inside is run by a marble-lobbied institution or by you with one index fund and a calendar reminder. Rejecting the tax break to spite the banks is like refusing your salary because you dislike your bank's adverts: the target of your anger does not receive the damage. You do.
Anger is not an investment strategy. The sharper response to everything in the previous section is not abstinence, it is control: take the wrapper, take the match if one is offered, and be deliberate about the contents.
When paying for advice does make sense
Honesty cuts both ways, so here is the other side. There are situations where paying a human professional is money well spent: an inheritance, tax that spans two countries, a business you own, equity compensation, or simply knowing yourself well enough to admit you will never do this and an unmanaged account would sit in cash for a decade, which is its own quiet disaster.
The test is not whether advice is worth paying for. Sometimes it clearly is. The test is who pays the advisor and what they are paid to sell. A fee-only or flat-fee advisor charges you directly, by the hour or by the engagement, and sells nothing else; their incentive is to be worth rehiring. A commission-based advisor is paid by product providers for placement, and no amount of personal warmth changes what the pay cheque rewards. Ask any advisor two questions before anything else: "who pays you?" and "what do you earn if I buy this?" A good one answers in one sentence and does not flinch. Evasion is an answer too.
Do this now
This week, spend thirty minutes finding out two things about your own situation. One: what tax wrapper exists in your country, what it costs to open, and whether your employer adds anything when you contribute. Two: if you already have a retirement product, find the annual management fee, as a percentage, in writing. Not the brochure adjective, the number. If you cannot find it within thirty minutes, that difficulty is itself information. You do not have to act on any of it yet. You just have to stop not knowing.
Next in the series: Diversification vs. concentration, where the safest idea in investing meets the people who built fortunes ignoring it, and both sides get their honest hearing.