Most people who manage their own investments have put real work into it. They have read the factsheets, compared the returns, checked the fees, and deliberately spread the money around. A global equity fund. Something in Asia. A technology fund. A dividend fund. A bond fund.
That is a reasonable list. But a portfolio is not a list.
The question that decides what happens to your money is not what funds do I own. It is what am I exposed to once they are all put together. Those sound like the same question. They are not, and only the second one tells you anything about a bad year.
What the second question looks like when you answer it
Here is a sample portfolio I ran a risk analysis on. The figures are illustrative and the holdings are not anyone’s real money, but the shape of it is very common.
If global equities rallied 10%, it would gain about 7.6%, against 6.2% for its benchmark. Slightly ahead. Good.
If 2008 happened again, it would lose about 30%, against 21% for the same benchmark.
So it earns roughly 1.4 percentage points more in a good year, and loses 9 points more in a bad one.
Nobody would take that trade if it were put to them plainly. Offered a little more upside in exchange for a great deal more downside, anyone sensible declines. But it was never put to them plainly, because that comparison does not appear on a statement, a holdings list, or a year-end return. It only appears if someone goes looking for it.
Why it stays hidden
That portfolio passes most of the checks a careful person would think to run.
It is not over-concentrated in any single holding. Its overall risk level suits its stated objective. It matches its own target allocation. Anyone reviewing it sensibly would conclude it was well built, and they would not be careless for thinking so.
What it fails is a kind of concentration you cannot see from a list of names. Its equity exposure to two sectors, industrials and healthcare, is more than double the benchmark’s. Its bond exposure leans heavily into a single region.
Not because anyone decided to bet on healthcare. Because several perfectly reasonable funds, bought at different times for different reasons, each happened to hold some, and nobody ever added them up.
Diversification in form, not in substance
The instinctive test for diversification is counting. Twenty funds feels safer than three.
But funds overlap, often heavily, and the overlap is invisible from the outside. A global equity fund, a technology fund and a thematic healthcare fund can share a surprising number of the same underlying companies. You hold three things and own one idea three times.
It is not only sectors. A global fund and an Asia fund look like two different decisions, and can turn out to be sensitive to the same move in emerging market currencies. A bond fund bought for income can carry meaningful exposure to lower credit quality, or to rates, which is a different bet from the one most people think they are making when they buy bonds.
None of that is automatically wrong. A deliberate overweight is a legitimate position. The problem is the accidental one, where you are carrying a concentrated bet you never decided to make and would probably not defend if someone asked you to.
Which is why concentration has to be measured against something rather than eyeballed. Not “do I own a lot of different funds” but “how far does my exposure to any sector, country or style sit from a sensible reference point, and did I mean it to?”
What would this do if 2008 happened again on Monday?
Stress testing is not forecasting. It is not a view on whether a crash is coming, and anyone presenting it that way is overselling it.
It asks something narrower and more useful: given exactly what this portfolio holds today, what would have happened to it during a specific historical event? Lehman. The 2015 China selloff. February 2020. The 2022 invasion.
The value is entirely comparative. A number like “down 30%” means very little on its own. Down 30 when the benchmark is down 21 means something quite specific: nine of those points came from choices rather than from the market.
Most people have never seen that number for their own money. Not through negligence, but because nothing in ordinary life prompts the question. Your statement shows what you own and what it is worth today. It has no opinion about what would happen to it in a bad year, and it will never volunteer one.
Where the risk is coming from
This is the one almost nobody has met, and it is the most useful of the lot.
Total risk can be broken into sources. How much comes from simply being in the market at all. How much from country exposure, sector exposure, or style tilts, meaning characteristics like favouring growth over value, or larger companies over smaller.
In that sample portfolio the total expected variability was about 13.8% a year, against 11.1 for the benchmark. The extra did not come from market exposure, which is the part you are compensated for. It came from country and style tilts that were, as far as I could tell, unintentional.
That distinction matters more than almost anything else here.
Market risk is the price of admission. Over time you expect to be paid for taking it. Risk from accidental tilts is not reliably paid for at all. You are carrying extra volatility with no particular reason to expect extra return in exchange.
Which is the technical version of the asymmetry at the top of this article. A little more upside, a great deal more downside, and nothing compensating you for the difference.
What this is not
I want to be careful here, because this sort of analysis is easy to oversell.
These are models. They rest on assumptions about how asset classes move together, and those relationships change, sometimes at exactly the moment you would most like them to hold. A stress scenario estimates what would happen if a past event repeated against today’s holdings. It is not a prediction, and the next crisis will not be a repeat of the last one.
None of it tells you what will happen. What it does is convert questions usually answered by feel into questions that can be answered with a number, so you can decide whether you are comfortable with the answer.
It also cannot tell you what the portfolio is for. A portfolio that looks aggressive is entirely appropriate for someone with twenty-five years and a secure income, and completely wrong for someone drawing on it in four. The analysis describes the portfolio. Only you can say what it has to do.
Why I bother with any of this
Not because it is impressive, and not because a number decides anything for you.
Because the alternative is judging by feel, and feel is least reliable exactly where it matters most. It tells you a portfolio is fine because it went up last year. It tells you twenty funds is diversified. It tells you nothing about what happens if markets fall 30% in six months, which is the scenario your plan actually has to survive.
I saw this recently with a client whose old fund had slightly outperformed the portfolio we built together, for reasons that only became clear once we looked inside it. I wrote about it in He Would Have Earned More If He Had Stayed.
And the real cost is not the loss itself. It is that a portfolio structured in a way you did not understand is a portfolio you are far more likely to abandon at the bottom, which is where the permanent damage gets done.
If you are going to hold something for twenty years, it is worth knowing what you are actually holding. Not the names on the statement. The risks underneath them.