What happens if you pick six investment trusts, promise yourself you will barely touch them, and then actually keep that promise for a decade and a half? I keep coming back to that question because most of us say we are long-term investors and then tinker the moment a holding looks ugly. Fourteen years of a simple, all-weather investment trust portfolio offers a rarer kind of evidence: not a backtest, not a slogan, but a living experiment in patience, discounts, gearing and the quiet cost of swapping one good idea for another.
What Fourteen Years Of One Trust Portfolio Really Taught Us
The idea was almost stubbornly plain. Build a small basket of listed closed-end funds that could survive different market climates, keep costs sensible, allow gearing when managers earned the right to use it, and change names as rarely as possible. That last part is the one people skip. In my experience, the hard work is not picking the first six names. The hard work is leaving them alone when the narrative around them gets noisy.
From mid-2012 through the middle of 2026, the basket delivered about 248% in share-price terms if you never rebalanced, or roughly 9.2% a year. The domestic all-share benchmark managed closer to 5.2% a year. A global, US-heavy world index sat near 9.7%. Dividends would have added a little extra, though this was never designed as a high-yield machine. Equal-weighted, the pack still yields around 2%. Satisfactory rather than spectacular, which is often what real-world compounding looks like when you refuse to chase last year’s story.
About half the switches added value. The other half did not. That split is the most useful statistic in the whole exercise.
Only six changes in fourteen years. Two of the original holdings are still there. If that sounds boring, good. Boring is frequently the point.
Why Investment Trusts Were The Vehicle In The First Place
Open-ended funds have their place. Closed-end trusts have a different toolkit. Managers can gear, hold less liquid assets, and ignore daily redemptions. Investors can buy and sell on the exchange instead of forcing the manager to raise or return cash at the worst moment. Discounts and premiums become a second price, which is both a risk and an opportunity. I have found that people either love that second price or they never quite trust it. The portfolio treated it as part of the job.
The original six names were not a random grab bag. Each was meant to cover a different job: capital preservation, high-conviction growth, quality compounding, uncorrelated macro exposure, infrastructure income, and a multi-asset family office style sleeve. That mix still reads like a grown-up asset allocation, just wrapped in listed vehicles you can hold inside a normal brokerage account.
- A defensive core that cares more about not blowing up than beating a benchmark
- A growth engine willing to look odd and concentrated
- A quality or income sleeve with a recognisable process
- Something that might zig when equities zag
- Real assets with contractual or operating cash flow
- A diversified holding company that can own public and private assets together
That architecture matters more than any single ticker. Markets change. The jobs do not change as fast.
The Original Six And The Jobs They Were Hired To Do
Personal Assets was the shock absorber. Wealth preservation, a serious gold sleeve when needed, and a long-standing habit of trying to keep the share price close to net asset value. If you wanted one holding that would not demand heroics from you during a panic, this was it. It is still in the book. That continuity is not nostalgia. It is an admission that true defensive vehicles are harder to replace than growth stories.
At the other extreme sat Scottish Mortgage. High conviction, public and private growth, and a willingness to let winners run until the position looks uncomfortable. It remains the portfolio’s standout, with a share-price gain in the region of 940% since 2012. Yes, there was a brutal drawdown of nearly 56% from the 2021 peak. Anyone who held through that stretch already knows the emotional invoice this style sends. The rewards did not arrive without that invoice.
Finsbury Growth & Income arrived as a quality compounder with little interest in hugging a domestic index. The process looked elegant in 2012 after a strong prior decade. Elegance is not a permanent edge. Later valuation stretch and a long patch of style pain would eventually force a rethink, and that rethink became one of the messier chapters in the story.
BH Macro was the uncorrelated sleeve, a listed wrapper around a global macro hedge fund that had earned its reputation in the crisis years. Access was the appeal. Cost was the problem. Fees in a 1.5% to 2% band, depending on performance, ask a lot of the underlying skill. When the skill does not clearly outrun the fee, the wrapper starts to look like a luxury good.
3i Infrastructure was meant to be the absolute-return income pick, with a preference for operating assets rather than contracts that decay on a timetable. Real assets can steady a portfolio. They can also change character when management resets return targets. That reset later became the reason the position left.
RIT Capital Partners was the old favourite: public markets, private markets, hedge funds and real assets under one roof. Useful diversification on paper. Over time the private and venture sleeve grew heavier, from about 24% to around 45%. That drift mattered because the rest of the portfolio already had growth and illiquidity elsewhere.
The First Swap: When Access Stops Being Enough
A little over a year in, the macro wrapper made way for Caledonia Investments. Hindsight is cheap, but the original choice already looked expensive relative to what it delivered. Access to a strategy normally reserved for very large cheques is not the same thing as a good expected return after fees.
Caledonia came in as a family-controlled investment company with a long horizon, a mix of funds, direct private holdings and listed equities, and a discount around 20%. That discount was part of the thesis. Family control can be a blessing and a drag. It keeps the clock long. It can also leave a stubborn overhang when a large block sits with one dynasty.
From inception of the original book through July 2026, the discarded macro trust returned about 114%. The switch into Caledonia produced about 124%. Not a life-changing gap. Still a win, and one of the cleaner wins in the whole series. Sometimes the best trade is simply leaving an expensive complexity for a cheaper, slower compounder.
Income, Value And The Temptation To Tidy The Sleeve
Late 2015 and early 2017 brought two more moves. Infrastructure left after the return objective was reset. In came Law Debenture, a curious hybrid: a UK equity portfolio plus a professional services business that does unglamorous work such as pension-scheme administration, escrow and bond documentation. That operating arm has often covered about a third of the cash dividend. Flexibility follows. Managers can own growth names that pay little today because the services income helps carry the payout.
Law Debenture was bought around a 14% discount. That gap has since all but vanished. Over the full fourteen years the sold infrastructure name returned about 118%. The replacement path delivered closer to 200%. That is the sort of result that makes people overconfident about activity. It should not. It was one good call, not a licence to keep rearranging the furniture.
The other 2017 move looked neat at the time. Quality growth looked expensive after a 98% five-year run, so the holding was sold for Temple Bar, a lower-cost, contrarian UK value trust. Value can look cheap for a reason and stay cheap for a long time. Pandemic pressure on UK value names, plus a manager departure, turned the position into a 42% loss by May 2020. That hurts. It also created the next problem: serial replacement of the same sleeve.
The Cost Of Not Leaving One Idea Alone
Temple Bar was sold for Mid Wynd International, framed as a resilient global quality compounder that might balance the more aggressive growth trust. Then management changed. Out it went in April 2025 for JPMorgan Global Growth & Income. Follow that chain from the original quality name through value, through another global compounder, into the current income-growth hybrid and you get about 65% to July. Hold the first name the whole way and you would have had about 142%, even after a rough recent stretch. Hold the value trust after the first switch and you would have had about 236%.
Read that again slowly. The sleeve that was edited the most produced the weakest path. I do not say that as a sermon against all selling. I say it because the record is sitting there in plain sight. Activity has a hurdle. If you cannot clear it more often than not, you are paying tuition.
The inability to stick with one trust for that part of the portfolio has hurt returns more than any single market crash inside the sample.
Perhaps the most interesting aspect is how reasonable each individual decision looked in isolation. Valuation stretched. Style broke. A manager left. A new process appeared. None of those reasons is silly. Together they became a leak.
The Last Big Change: Private Markets And A Value Remit
In March 2023 the multi-asset favourite made way for AVI Global. The old holding’s private equity and venture weight had climbed too high for comfort beside an already growth-heavy book. The replacement offered global value, family holding companies, discounted trusts and a meaningful Japan sleeve, with look-through United States exposure around 13% and Japan around 23%. That is a very different bet from a world index stuffed with mega-cap technology.
So far the two paths have been broadly similar. That is not a failure. A diversification swap is allowed to look dull in the first few years. The point was not to out-sprint the old name immediately. The point was to own a process that still made sense if public market valuations stayed rich and if private marks stayed harder to trust.
How The Book Looks Now And Why Tinkering Still Looks Costly
The current six still map to the original jobs, just with different faces in a few chairs.
| Role in the book | Current holding | What you are really buying |
| Defensive core | Personal Assets | Capital preservation, gold ballast, discount discipline |
| High-conviction growth | Scottish Mortgage | Public and private winners held with unusual concentration |
| Global income-growth | JPMorgan Global Growth & Income | Flexible payout near 4% of NAV, growth bias inside an income wrapper |
| UK hybrid income | Law Debenture | Equity portfolio plus fee-earning professional services |
| Family multi-asset | Caledonia | Listed, fund and direct private exposure under family control |
| Global value and activism | AVI Global | Holding companies, Japan, discounts, a non-index shape |
Personal Assets has lagged inflation over five years after the 2021-2023 price shock and the way markets digested rate hikes. Over ten years it is still ahead. Over three it has climbed back in front. Closest cousins in spirit are other wealth-preservation trusts that try to beat inflation without behaving like a disguised equity fund. Gold helped recently. Timing still matters, which is exactly why a defensive sleeve exists.
The global income-growth name has not set the night sky on fire since it arrived. Previous trading in that sleeve argues against another swap. It is the largest trust in its peer group, with a strong record and a yield near 3.9%. Management aims to pay about 4% of NAV a year from income and capital. In theory the payout can fall. In practice the growth overweight has so far protected the distribution. That flexibility is the feature. Peers that lean harder on classic yield stocks have less room to manoeuvre when the cycle turns.
Law Debenture’s ten-year record in UK equity income is in a different league from most straightforward rivals, with a gap on the order of 100% versus a close competitor that has itself recovered under new management. The professional services engine is easy to ignore until you notice how it changes the investment constraint. If you want a plain UK equity income trust instead, there is no shortage of styles: deep value, dividend aristocrat, special situations, larger-cap steadiness. Just know you are giving up that extra cash engine.
Caledonia has lagged over five years because the discount has sat in a stubborn 30% to 40% range. Splits to improve liquidity, buybacks and a higher dividend have not fully solved the overhang from a family stake near 48%. That is irritating. It is also part of the original bargain. You bought a family-controlled mix of private funds, direct companies and public equities. That mix is still one of the cleaner versions of the category. A heavier direct book still gives it a slightly different flavour from the old multi-asset favourite it replaced years earlier.
Scottish Mortgage And The Price Of Conviction
No other holding comes close to the growth trust’s contribution. The edge has been picking a handful of extraordinary businesses and living with the concentration. Plenty of vehicles try to copy the public-plus-private shape. Very few have repeated the outcome. Volatility is not a side effect here. It is the admission fee.
I have found that investors either accept that fee in advance or they become forced sellers in the middle of the lesson. There is no clever middle path. If a near-halving of the share price would make you abandon the thesis, the holding does not belong in a set-and-forget book, no matter how pretty the long-run number looks from 2012.
Why AVI Global Still Earns Its Seat
The value-and-activism trust was never meant to look like the world index. Missing part of the recent technology surge is therefore not an automatic indictment. Low look-through exposure to the United States is a feature when the rest of a portfolio already leans that way through the growth engine. Japan, family compounds and discounted closed-end holdings give the book a second engine that does not need the same narrative weather.
Simpler one-stop global trusts exist. Some will be easier to explain at a dinner table. Ease of explanation is not the same as complementarity. In a frothy tape, a process that hunts neglected holding companies can still be the adult in the room.
What The Next Fourteen Years Could Demand
Are these still the right six? Maybe. Markets do not owe continuity to any line-up. The better question is whether the jobs are still funded. Defence, asymmetric growth, flexible global income, UK hybrid cash flow, family private-and-public compounding, and non-index value. If a holding stops doing its job, replace the job-holder. If it is merely unfashionable, wait.
Defensive peers will keep circling the same problem: how to stay ahead of inflation without sneaking in equity beta you did not ask for. Global income names will keep arguing about whether a capital-and-income payout is honest flexibility or a soft promise. UK income trusts will keep splitting between pure equity processes and odd hybrids. Family vehicles will keep trading on discounts that refuse to close. Growth trusts will keep teaching the same lesson about drawdowns. Value activists will keep looking early, then late, then early again.
- Write down the job of each holding in one sentence before you buy.
- Decide in advance what would falsify that job, not what would bruise your feelings.
- Treat discounts as a second valuation layer, not a moral verdict on the manager.
- Count every switch against a stay-put alternative, including the ones that later look clever.
- Accept that a 50% hit rate on changes is not a reason to change more often.
That list is not sophisticated. It is operational. Most portfolio damage I see in ordinary accounts comes from skipping step two and improvising step four.
Discounts, Gearing And The Mechanics People Skip
Closed-end pricing can widen for reasons that have nothing to do with the portfolio. Liquidity, a large family holder, a style going cold, a sector scare. Buybacks and splits can help. They cannot repeal supply and demand. If you need the discount to close next quarter for the thesis to work, you do not have a long-term thesis. You have a trade.
Gearing cuts both ways. In a rising market it is a gift. In a falling market it is a megaphone. The original design liked the option of gearing in skilled hands. That still seems right, provided you know which trusts use it as a tool and which use it as a habit. Tools can be put down. Habits rarely are.
Costs hide in performance fees, in wide bid-offer spreads on smaller names, and in the mental energy of monitoring six processes. A set-and-forget book is not a no-think book. It is a low-trade book. Those are different standards.
Income Without Pretending This Is An Income Portfolio
A two percent average yield will not thrill anyone who needs the portfolio to pay the bills this year. That was never the brief. The income that does appear is useful because it arrives from different engines: a services business, a flexible NAV payout, ordinary equity dividends, and the occasional distribution from more mature private holdings. Diversity of income source is underrated. Yield hunting in one sector is overrated.
If you turned this into a retirement paycheck machine, you would raise the income sleeve and shrink the high-conviction growth sleeve. You would also change the drawdown profile. There is no free lunch where you keep the 940% growth path and also enjoy a fat, stable yield. People try. The market eventually collects the contradiction.
A Human Way To Read A Fourteen-Year Scorecard
It is tempting to flatten the whole story into “activity is bad.” That is too neat. One infrastructure-to-hybrid switch was excellent. The macro-to-family switch was modestly good. The private-heavy multi-asset swap looks neutral so far. The quality-to-value-to-quality-to-income chain was poor. The lesson is statistical humility, not monastic inactivity.
I’ve found that investors remember the winning switch and forget the chain of “reasonable” sells that never recovered the compounding they interrupted. Memory is biased toward drama. Compounding is biased toward time. Those two biases do not shake hands.
Is 9.2% a year good enough? Compared with cash, yes. Compared with a world index that rode a long American bull market, it is merely competitive. Compared with the emotional cost of holding a concentrated growth trust through a 56% peak-to-trough slide, it is honest. Satisfactory compounding plus survivable behaviour beats a prettier backtest you would have abandoned.
Practical Rules If You Want To Steal The Method, Not The Tickers
Do not photocopy the six names and call it a philosophy. Photocopy the constraints. Six is small enough to understand and large enough to split economic roles. Changes should be rare enough that each one can be written up like a board paper. Rebalancing can wait unless one holding becomes the whole book. Dividends can be spent or reinvested, but they should not become an excuse to own weak businesses.
A working checklist: 1 job per holding 1 falsification test per holding 1 discount range you can live with 1 maximum annual change budget 1 rule for what happens after a manager leaves
Manager departures deserve a special note. Sometimes the process is the firm. Sometimes the process is one person. If you cannot tell the difference before the announcement, you will decide in a hurry later. Hurry is how the 42% loss and the subsequent chain of replacements get born.
Family control deserves another note. It can protect a long horizon. It can also freeze a discount. You are not entitled to both the long horizon and a quickly closing gap. Pick which part of the bargain you actually want.
What This Experiment Quietly Says About Ordinary Investors
Most of us do not need more ideas. We need a smaller number of roles and a longer clock. The fourteen-year book is not magic. It is a reminder that closed-end funds can package gearing, private assets and specialist processes in a form a regular saver can actually hold. It is also a reminder that the market will keep offering reasons to break your own rules.
Will the next fourteen years rhyme? Inflation shocks, rate cycles, private market marks, technology concentration and discount politics will show up again in new clothes. The trusts may rotate. The jobs should not have to. If you take nothing else, take the hit rate on the changes. Half the time, staying put was the better trade. That is an unfashionable sentence. It may also be the most profitable one in the file.
So here is the uncomfortable close. A set-and-forget portfolio only works if you can forget on purpose. Not because you stopped caring, but because you already decided what would make you care enough to act. Fourteen years is long enough to prove that most of the noise never met that test. The next decade will produce new noise. The question is whether your process is still listening to the job, or to the week.
