Two investors put £10,000 into a Stocks & Shares ISA on the same day, aiming for the same broad global exposure. One buys units in a Vanguard FTSE All-World ETF. The other buys shares in an investment trust with a similar mandate. Twelve months later their returns can differ by several percentage points even though the underlying holdings overlap heavily — and the reason has nothing to do with stock-picking skill. It's down to structure, and most people choosing between the two have never had the difference explained properly.
Two different legal animals
An ETF is open-ended. When you buy in, the fund issues new units backed by fresh cash going into the market; when enough people sell, units are cancelled and the underlying assets are sold to fund the redemption. The share price tracks the value of the underlying assets almost exactly, because supply expands and contracts to match demand. An investment trust is a completely different legal structure — it's a company listed on the London Stock Exchange with a fixed number of shares in issue. You're not buying a slice of a pooled fund; you're buying shares in a company whose business happens to be holding other companies' shares. Because the share count is fixed, the trust's share price is set by ordinary buying and selling on the market, and that price can drift away from the actual value of the assets it holds.
Discounts and premiums — the trade-off nobody explains upfront
That drift is called a discount or premium to net asset value, and it's the single biggest structural difference an ISA investor needs to understand. Scottish Mortgage Investment Trust, run by Baillie Gifford, has traded at discounts of 15% or more during periods when sentiment towards its growth-heavy tech holdings turned sour — meaning you could buy £115 of underlying assets for £100. That sounds like a bargain until you need to sell during the same period of gloom, because the discount can widen further before it narrows, and you crystallise a loss on the wrapper even if the shares inside it recovered. ETFs don't carry this risk in any meaningful way. Authorised participants — large institutional traders — are financially incentivised to create or redeem units the moment an ETF's price strays from its underlying value, which keeps the gap to fractions of a percent almost all the time.
Gearing changes the risk profile completely
Investment trusts can borrow money to invest more than the capital shareholders put in — something called gearing, and something an ETF structurally cannot do. City of London Investment Trust, one of the oldest income-focused trusts on the market and best known for fifty-plus consecutive years of dividend growth, typically runs modest gearing of 5–10%. Amplify that further and gains get amplified on the way up, but so do losses on the way down, and in a sharp correction a heavily geared trust can fall faster than the index it's loosely tracking. This isn't a flaw to be avoided outright — used sensibly, gearing is one of the reasons well-run trusts have beaten comparable open-ended funds over long stretches. It just means two products holding near-identical shares can behave very differently in a 20% market drop, and anyone comparing past performance charts without checking gearing levels is comparing apples to a fairly leveraged orange.
Costs: the gap has narrowed, but hasn't closed
Passive global ETFs are relentlessly cheap — Vanguard's FTSE All-World ETF charges an ongoing charge of around 0.22% a year, and some rivals undercut even that. Investment trusts, being actively managed in most cases, typically charge somewhere between 0.4% and 0.9%, occasionally more for specialist or smaller trusts where fixed costs are spread across a lower asset base. On £10,000 over twenty years, the difference between a 0.22% charge and a 0.65% charge compounds into a meaningfully different end balance, all else being equal — which is exactly the point: all else is rarely equal, because you're paying the higher charge for the possibility of genuine stock-picking skill and structural features like gearing that a tracker can't offer.
Income investors: why the revenue reserve matters
If you're building an ISA for income rather than pure growth, investment trusts have a structural advantage ETFs simply cannot replicate: the revenue reserve. Trusts are legally allowed to hold back up to 15% of the income they receive in a given year and keep it in reserve, rather than paying all of it out to shareholders immediately. That reserve gets drawn down in lean years to keep the dividend growing even when the underlying portfolio's income has actually fallen — which is precisely how City of London Investment Trust has managed over fifty consecutive years of dividend increases, including through 2020, when dividend cuts across the wider market were widespread. An ETF has no equivalent mechanism. It has to pass through whatever income the underlying shares actually generate in that period, which means an income ETF's distribution can and does fall in a bad year for corporate dividends, with nothing smoothing the ride for the investor holding it.
Liquidity and the smaller trust trap
Not every investment trust trades as easily as Scottish Mortgage or City of London. Smaller, more specialist trusts — a trust focused on a single emerging market, or a niche sector like renewable energy infrastructure — can have thin daily trading volumes, which shows up as a wider bid-offer spread: the gap between what you'd pay to buy and what you'd receive to sell at the same moment. On a heavily traded large trust that spread might be a fraction of a percent. On a small specialist trust it can run to 2–3%, which is effectively a cost you pay twice, once going in and once coming out, on top of the ongoing charge. Check the average daily trading volume and the typical spread before committing a meaningful sum to any trust with a market capitalisation under roughly £200 million — it's disclosed on every major platform's fact sheet, and it's frequently the reason a seemingly attractive discount never quite gets realised when you actually try to sell.
Platform fees change the answer too
The wrapper you hold either of these inside can matter as much as the product itself. Percentage-fee platforms like Hargreaves Lansdown charge 0.45% annually on funds held in an ISA, which stings noticeably on a large ETF holding but barely registers on a small one — while flat-fee platforms like Interactive Investor, charging a fixed monthly amount regardless of portfolio size, tend to work out cheaper once an ISA passes somewhere around £30,000–£40,000. Investment trusts, being traded like ordinary shares, usually attract a dealing charge per transaction rather than the ongoing platform percentage that applies to fund holdings on some platforms — AJ Bell, for instance, charges share dealing separately from its fund custody fee. Anyone regularly topping up an ISA with small monthly amounts should check whether their platform charges a flat fee per trust purchase, because paying £9.95 in dealing charges on a £100 monthly top-up erodes the return far more than the same £9.95 would on a £1,000 lump sum.
Which one actually suits an ISA
For the core of a Stocks & Shares ISA — the bit you want to leave alone for fifteen years and not think about — a low-cost global ETF is the better choice for most people. It's simple, the tracking is tight, there's no discount risk to manage, and the charges stay low regardless of market conditions. Investment trusts earn their place as a satellite holding once the boring core is sorted: pick two or three with a strong long-term track record, a sensible gearing policy, and a discount that's wide relative to its own five-year average rather than wide in absolute terms, and you're buying genuine diversification rather than a second copy of the same index.
Don't buy a trust purely because the discount looks big on the day you're reading about it, though.
A wide discount can mean the market has spotted something you haven't — a manager change, a sector going out of favour, or a dividend that isn't as covered by earnings as the yield figure suggests. F&C Investment Trust, the oldest of the lot, dating back to 1868, has weathered manager changes and market cycles that would have sunk a lesser structure, precisely because its board has been willing to act on persistent discounts through buybacks rather than ignore them. Check whether a trust's board has a track record of managing the discount actively before assuming a cheap-looking price is a genuine opportunity rather than a warning sign the market is pricing in correctly.
What to do with your next ISA contribution
If you're adding to an ISA this month and you don't already hold a low-cost global tracker, start there — it should be the foundation, not the afterthought. Once that's in place, allocate no more than 20–30% of new money to two or three investment trusts you've actually researched, checking gearing, discount history, and ongoing charges rather than trailing twelve-month performance alone. Skip anything trading at a premium to NAV unless you have a specific reason to believe the manager's edge justifies paying more than the assets are worth — and if a platform's fact sheet doesn't state the current discount or premium clearly, that's a fact sheet worth being suspicious of.