Behaviour
The price you paid should not be part of the decision
What a holding cost you is a fact about your own history, not about its future. It still dominates how most people decide when to sell.
Comparisons of anchoring to your purchase price usually pick a winner. This one picks the circumstances, which is more useful.
The difference in one place
- The market has no knowledge of what you paid and no return depends on it.
- Selling winners and holding losers degrades a portfolio without any single decision looking wrong.
- Rebalancing to written proportions removes the purchase price from the decision.
The anchor
People treat their purchase price as a reference point, sorting outcomes into gains and losses measured against that one number. The market has no knowledge of what you paid, and no future return depends on it in any way whatsoever. Despite this, a holding sitting below its purchase price is markedly harder to sell than one sitting above it.
The pattern is well documented in studies of trading records, which have generally found that investors realise gains more readily than losses. The evidence for the pattern itself is stronger than the evidence for any particular explanation of why it happens.
Waiting to get back to even
Holding a position purely until it recovers its purchase price is a decision being made by an entirely arbitrary number. The relevant question is whether you would buy this holding today at today's price with today's information. If the answer is no, the purchase price is not a reason to keep holding it, however uncomfortable that conclusion feels.
The useful part is this: this is far easier to state than to do, because selling below cost converts a private disappointment into a recorded one. The recorded loss feels like the moment the money disappeared, even though it disappeared when the price fell.
The opposite error
Selling as soon as something moves above the purchase price is exactly the same anchoring working in the other direction. It produces a portfolio in which everything that did well has been sold and the disappointing holdings all remain. Over time this systematically degrades a portfolio without any individual decision looking wrong at the moment it was made.
The pattern is easiest to spot in a transaction history, which is one of the better reasons to keep one. Neither error is about the investments themselves; both are about where the reference point has been placed.
Why averaging down feels different
Buying more of a falling holding lowers the average purchase price, which makes the anchor easier to reach again. That is a psychological benefit rather than an investment argument, and the two get confused with some regularity. The decision should rest on whether you want more of the asset, not on where the average ends up afterwards.
The useful part is this: regular contributions into a broad fund do this automatically without any judgement being exercised, which is much of their appeal.
Deliberately adding to one specific falling holding is a different activity that requires a different justification entirely.
Designing the anchor out
A written plan stating target proportions removes the purchase price from the decision more or less completely. Rebalancing to those proportions produces sales and purchases determined by weights rather than by your own transaction history.
Put simply, this is one of the underrated benefits of rebalancing, because it makes the reference point the plan rather than the past. Platforms that display gain and loss against cost prominently make this harder, and some of them allow that display to be changed. What you see every time you log in shapes what you think about, and that is worth arranging deliberately.
None of this is a substitute for talking to a clinician if something feels wrong.
The one case where cost genuinely matters
Purchase price is relevant for tax reporting in most systems, and that is a real administrative reason to keep a record of it. It may therefore legitimately affect the timing or sequencing of a sale, and the rules governing that are entirely jurisdictional. That is a matter for a qualified adviser locally rather than a general principle to be applied from an article.
Outside that, the number is a historical fact with no bearing at all on what the holding does next. Treating it as a record rather than as a target is most of the work involved in getting past it.
Side by side
| Consideration | What it means in practice |
|---|---|
| The anchor | The market has no knowledge of what you paid and no return depends on it. |
| Waiting to get back to even | Selling winners and holding losers degrades a portfolio without any single decision looking wrong. |
| The opposite error | Rebalancing to written proportions removes the purchase price from the decision. |
The takeaway
What you paid is history. What you would pay today is the only question the market is answering.
Small and repeatable beats ambitious and abandoned, almost every time.
Questions readers ask
Should I sell a holding at a loss?
The question is whether you would buy it today at today's price. What you paid is not part of that assessment, though tax reporting rules in your country may affect timing.
Is averaging down a bad idea?
It depends entirely on whether you want more of the asset. Doing it to lower an average price is a decision about a number rather than about the holding.
