Among the more liberating insights available to the investor is the recognition that the market has no knowledge of, and no interest in, what any individual paid for any particular security. This sounds obvious stated plainly. Its implications for investment behaviour are profound and frequently ignored. The price at which one bought a stock is a fact about one's own history; it is not a fact about the stock. It does not influence the company's earnings power, its competitive position, its management quality, or the macroeconomic environment in which it operates. It has no more relevance to the stock's future price trajectory than the price at which a previous owner bought it—which is to say, none at all.
Yet this irrelevant number shapes the investment decisions of the vast majority of individual investors with a force that is almost gravitational. Positions are held too long because selling below the purchase price feels like admitting defeat, despite the fact that the market does not record or acknowledge the defeat in question. Positions are sold too early because reaching or exceeding the purchase price feels like mission accomplished, despite the fact that the mission—achieving the best possible long-run return—has no necessary relationship to the purchase price. Portfolio concentration is maintained in deteriorating positions because the psychological cost of crystallising a loss is higher than the financial cost of continuing to hold, despite the market's complete indifference to the distinction.
The sunk cost fallacy—the tendency to give undue weight to costs that have already been incurred and cannot be recovered—is the formal name for the cognitive error at work here. In most domains, this fallacy is taught early: the money already spent on a failed project should not influence whether to continue spending on it; the time already invested in a relationship should not be the primary reason to continue it; the food already ordered should not be finished simply because it was paid for. These applications are widely understood. The application to investing—the stock already bought at a certain price should not be held simply because the current price represents a loss—is understood in principle and systematically violated in practice.
The violation persists because investment losses have a psychological quality that other sunk costs do not. A loss in an investment portfolio is not merely a past expenditure; it is a judgment on one's competence and foresight. Selling at a loss makes the judgment official. It creates a permanent record in the transaction history. It requires the investor to acknowledge, explicitly, that the decision to buy at the original price was wrong—or that the subsequent decision to hold was wrong, or that the circumstances changed in ways that should have been anticipated but were not. Each of these acknowledgments is uncomfortable in a way that simply declining to sell is not.
The path through this discomfort is the reframing of each portfolio decision as a fresh allocation question. The investor who asks, of each current holding, "if I had this amount of cash today and no existing position, would I choose to allocate it here?"—and answers that question honestly, without reference to what was paid or what is currently gained or lost—is making the only kind of decision that the market actually rewards: a forward-looking assessment of where capital is most efficiently deployed. The purchase price is irrelevant to this question. It always was.