Cash does two jobs no other holding can do: it will be there in full on a known date, and it lets you act without having to sell something first. It pays for both by losing purchasing power, reliably.
8 min read · Level 1
The conventional framing puts cash at one end of a risk spectrum and equities at the other, with everything else arranged in between, and I think that picture is responsible for more bad cash decisions than any specific piece of advice. A spectrum implies the instruments are doing the same job at different intensities, so that moving along it is a dial you turn according to nerve. They are not doing the same job. Cash answers a question about when the money is needed; equities answer a question about what it is for. Somebody who holds a lot of cash because they are cautious and somebody who holds a lot because they are buying a house in a year have arrived at the same position for reasons with nothing in common, and only the second one can say when it will end. That is the test I would apply to any cash balance: can you name the job and the date. If you can, the negative real return is a price you are knowingly paying for something specific, and it is frequently a bargain. If you cannot, you are holding the one asset in this curriculum that is guaranteed to lose purchasing power over a long horizon, for a reason you have not stated, and the safety you feel is the feeling of not being shown the cost. Nothing about this makes cash a poor instrument. It makes it a precise one, and precise instruments punish vague purposes.
Every other asset class in this curriculum is held because it might grow. Cash is not, and pretending otherwise is what makes it get used wrongly. It does two jobs, and neither of them is growth.
It is certain on a date. Money required at a known time, in a known amount, has a requirement that only this instrument satisfies. Take an obligation of 24,000 due in eighteen months.
Held in cash at 2.00% 24,000 x 1.02^1.5 = 24,723.59, and never less than 24,000 Held in equities a 30% fall leaves 16,800 - a shortfall of 7,200
The equity column is not a prediction. It is a statement about what is available, and the obligation does not move when the market does. For money with a date and a required amount, a range of outcomes is the wrong shape regardless of where the middle of the range sits.
It lets you act without selling something first. A holder with no cash who wants or needs money must sell a holding, at whatever price exists that day, chosen by circumstances rather than judgement. That is the mechanism illiquidity-and-forced-selling describes, and cash is what removes it. The value is not that cash earns something when other things fall - it does not - but that having it means nothing has to be liquidated on somebody else's timing.
And here is the price, stated plainly, because this class is the one where the cost is easiest to hide.
Over twenty years with interest at 1.50% and inflation at 2.50%:
Cumulative real change (1.015 / 1.025)^20 - 1 = -17.8054% 10,000 becomes 10,000 x 1.015^20 = 13,468.55 nominally In starting money 13,468.55 / 1.025^20 = 8,219.46
The balance grew by 3,468.55 and what it buys fell by 1,780.54. Over thirty years at 2.50% inflation, one unit of money buys 0.4767 of what it did - a loss of 52.33% of its purchasing power before any interest is counted.
That is not an argument against cash. It is the price of the two jobs above, and both jobs are worth paying for when the money actually needs them.
Because the failure in this class is almost never about the account or the rate. It is about a mismatch between the job the money has and the instrument holding it, and the mismatch runs in both directions.
Money that needs to grow, held in cash. The commonest version follows a frightening period in markets: a holder moves a long-horizon balance into cash to be careful, and leaves it there. They have exchanged an uncertain outcome for the near-certain one computed above, and because the certain loss never appears as an event, nothing ever prompts a reconsideration. Twenty years at -17.81% cumulative in real terms is a cost that was chosen, and it was chosen in order to avoid a range of outcomes whose worst case was frightening and whose middle was not.
Money that needs to be certain, held in something that grows. The less common and more acute version: a deposit for a purchase, a tax payment due, a known obligation with a date, invested because cash felt wasteful. This one does not fail slowly. It fails on the date, and on that date the shortfall is not recoverable by waiting, because the obligation has a deadline and the holding does not care.
Both mistakes come from asking the same wrong question - which instrument is better - when the right question is what this particular money has to do.
There is a third role that deserves care, because it is frequently overstated. Cash is sometimes described as something that lets you buy when prices fall. That is real as far as it goes: a holder with cash can act while a holder without it must first sell. But it should not be confused with a claim that holding cash produces a gain when markets fall, because it does not - the cash earns its rate either way, and the real return computed above continues throughout. What cash buys is the absence of compulsion, not a payoff.
And the honest limit on that is worth naming: money held indefinitely in anticipation of an opportunity is money paying the full price computed above for an optionality with no date on it. Cash held for a dated obligation has a term. Cash held against forced selling has a purpose. Cash held because the time never feels right has neither, and it is the same position wearing a better story.
Educational, not personalised financial advice. Nothing in the Investor Academy is a recommendation to buy, sell or hold anything — what is right for you depends on your own circumstances and the rules where you live.
Four pots of money belonging to one person, and what each one requires.
Pot Date Requirement Fits School fees due in 9 months known exact amount, certain cash A reserve against loss of income unknown available immediately cash A house deposit, 4 years out approximate mostly certain mostly cash Long-term investing, 20+ years none growth not cash
The first two are not investments and should not be judged as ones. Their return requirement is that the money is there. Comparing them against an equity return is comparing two instruments on a dimension only one of them was chosen for.
The third is the interesting case, because it sits between. Four years is long enough that a negative real return compounds noticeably and short enough that a large fall could not be waited out. There is no clean answer, and the useful move is to decide how much of the amount must be certain rather than to place the whole sum in one instrument.
The fourth is where the cost computed in this lesson becomes decisive. Twenty years at 1.50% against 2.50% inflation is -17.81% cumulative in real terms. That is the figure to hold beside any decision to leave long-horizon money in cash, and it is the figure nobody sees, because the balance only rises.
What this framing settles is a question people ask constantly and cannot answer: how much cash should I hold. The question has no general answer because it is the wrong question. Cash is not an allocation percentage chosen for balance; it is the sum of the specific jobs that require certainty and availability. Add them up and that is the number. It will be large for somebody with a purchase coming and a variable income, and small for somebody with neither, and the two figures being different is not a disagreement about markets.
Which also means the number changes as those jobs change - when the purchase completes, when income stabilises, when a deadline passes - and it is worth revisiting then rather than on a schedule.
Holding long-horizon money in cash for safety. It exchanges an uncertain outcome for a near-certain real loss - 17.81% over twenty years on the figures above - and the certain one never announces itself, so nothing prompts a review.
Investing money that has a date and a required amount. A 30% fall against a 24,000 obligation leaves 16,800, and the obligation does not move. This failure arrives all at once, on a known day, and cannot be waited out.
Expecting cash to gain when other holdings fall. It earns its rate either way and the real subtraction continues throughout. What it provides is the absence of compulsion to sell, which is valuable and is not a payoff.
Holding cash indefinitely for an opportunity with no date. That is the full price of this instrument paid for an option nobody has committed to exercising, and it is the position most likely to be defended with a story rather than a requirement.
Treating cash as an allocation percentage. It is the sum of the jobs that need certainty and availability, which is a number you can actually derive - and two people holding very different amounts may both be right, because they have different obligations rather than different views.
Setting the figure once and leaving it. The jobs change: a purchase completes, an income steadies, a deadline passes. The right amount moves with them rather than with a calendar.
Every specific requirement for certainty or immediate availability, with the amount and the date where there is one. The total is your cash figure, derived rather than chosen. It is the only method here that produces a number you could defend to somebody who disagreed with you about everything else.
Known date, approximate date, or none. A pot with no date is either long-horizon money that should not be in cash, or an availability reserve that should be - and writing which one it is forces the distinction that the whole lesson turns on.
Twenty years at 1.50% against 2.50% inflation is -17.81% cumulatively; 10,000 becomes 13,468.55 nominally and 8,219.46 in starting money. Putting that figure next to the balance is what makes the certain cost as visible as the uncertain one it was chosen to avoid.
A purchase completing, an income stabilising or destabilising, an obligation being paid. Those are the events that change the correct amount, and they are the moments to recompute it. A quarterly review of a number derived from unchanged requirements produces nothing.
When deciding how much cash to hold, when a known obligation has a date and an amount, when income is variable enough that selling at a bad moment is a live possibility, and before moving a long-horizon balance into cash after a difficult period in markets.
Do not use cash for money whose job is growth over a long horizon, and do not hold it indefinitely against an opportunity with no date attached - both of those pay this instrument's full price for something it was not chosen to provide.
After a severe market decline, some holders move long-horizon money into cash with the intention of returning once conditions settle. The difficulty is that no signal ever arrives to mark the moment: markets recover gradually and unevenly, each partial recovery makes re-entry feel late, and the balance in the meantime produces no event that would prompt reconsideration. The position is comfortable in a way the original holding was not, which removes the last source of pressure to revisit it. What began as a decision about one bad period becomes the permanent shape of a portfolio, and its cost is the compounding real subtraction that no statement displays.
If money is moved to cash temporarily, write down in advance what would bring it back - a date, a condition, a schedule. Without one, temporary is decided by the absence of a prompt.
Create a free account to read the rest and apply it to your own investing.