The Sunk Cost Trap in Investing
The sunk cost fallacy is the tendency to continue an endeavor because of previously invested resources (time, money, or effort) rather than based on future expected returns. In investing, it manifests as the refusal to sell a losing position because doing so would mean "admitting I was wrong" or "taking a loss." The money already lost is gone. It cannot be recovered by holding the position. But the psychological pain of crystallizing the loss, making it official and permanent, keeps investors trapped in positions they would never buy today.
This is not a minor behavioral quirk. It is one of the most expensive cognitive errors in investing. A study by Odean and Barber found that individual investors hold losing positions a median of 124 days before selling, compared to just 104 days for winning positions. The losing positions, on average, continued to underperform after the point where investors should have sold. The investors were not exercising patience. They were trapped by sunk costs.
The Psychology of Sunk Costs
The sunk cost fallacy operates through two reinforcing psychological mechanisms.
Loss aversion. Kahneman and Tversky's prospect theory demonstrates that losses feel approximately twice as painful as equivalent gains feel pleasant. Selling a stock at a 30% loss does not just reduce the portfolio by 30% of the position size. It triggers a psychological pain response that the investor desperately wants to avoid. Holding the position keeps the loss "unrealized" and maintains the fiction that the money is not really gone.
The distinction between realized and unrealized losses is an accounting convention, not an economic reality. A stock purchased at $100 that now trades at $70 has lost $30 per share regardless of whether the investor sells. The $30 is gone. The only question is whether the remaining $70 is better deployed in this stock or in something else. But the emotional experience of clicking "sell" and seeing the loss appear on the tax statement is viscerally different from watching the stock sit at a lower price in the portfolio.
Commitment and consistency bias. Robert Cialdini's research on influence shows that people have a powerful need to appear consistent with their past decisions. An investor who bought a stock after extensive research has publicly or privately committed to the position. Selling at a loss contradicts that commitment. The investor may add to the position ("averaging down") or rationalize holding it ("it will come back") to maintain consistency with the original decision.
This bias is amplified when the investment was discussed with others. An investor who recommended a stock to friends, posted about it on social media, or discussed it at a dinner party faces social pressure to hold, because selling at a loss would mean admitting the public recommendation was wrong.
How Sunk Costs Destroy Portfolios
The portfolio damage from sunk cost behavior extends beyond the immediate loss on the held position.
Opportunity cost. Capital trapped in a losing position is capital that cannot be deployed elsewhere. An investor holding $10,000 in a stock that has declined 40% and shows no signs of recovery is foregoing the returns that $10,000 could generate in a better investment. If the alternative investment returns 10% per year, the opportunity cost is $1,000 per year, compounding. Over five years of holding a dead position, the opportunity cost exceeds $6,000.
Emotional contamination. A large, visible loss in the portfolio affects decision-making on other positions. The investor may become more risk-averse elsewhere in the portfolio (selling winners too early to "protect" profits) or more risk-seeking (doubling down on the loser or making speculative bets to "make back" the loss). Both responses harm the overall portfolio.
Portfolio drift. A losing position that was once 5% of the portfolio but has declined to 2% still occupies a slot in the portfolio. Selling it and redeploying the proceeds into higher-conviction positions improves portfolio quality, but sunk cost thinking prevents the action.
The "Would I Buy This Today?" Test
The most effective tool for overcoming sunk cost bias is a simple question: "If I did not own this stock and had its current value in cash, would I buy it today at its current price?"
If the answer is no, the stock should be sold. The purchase price is irrelevant to this question. The only relevant information is the stock's current price, the business's current fundamentals, and the expected future return.
This test is harder than it sounds because the investor's perception of the stock is colored by ownership. A stock purchased at $100 that now trades at $60 feels like it should be worth at least $100 because "it was worth $100 when I bought it." But the market's assessment of the business has changed, and the original purchase price is sunk history, not a floor on future value.
To apply the test rigorously, the investor should imagine they are a new analyst reviewing the position for the first time. The analyst would look at the current balance sheet, current earnings, current competitive position, and current valuation. The analyst would have no knowledge of the purchase price, no emotional attachment, and no commitment to the original thesis. If the analyst's conclusion is "I would not buy this," the correct action is to sell.
Averaging Down: The Sunk Cost Trap's Expensive Cousin
Averaging down, buying more of a declining stock to reduce the average cost per share, is sometimes a legitimate strategy and sometimes a sunk cost trap wearing a sophisticated disguise. The distinction is whether the additional purchase is motivated by the stock's current merits or by the desire to "fix" the original losing purchase.
Legitimate averaging down: the original thesis remains intact, the stock has declined due to temporary factors (market-wide selloff, short-term earnings miss), and the lower price represents a better entry point for the same investment case. An investor who bought Johnson & Johnson at $170 and buys more at $145 during a market correction, based on the same analysis of the company's competitive position and earnings power, is making a rational allocation decision.
Sunk cost averaging down: the thesis has deteriorated, the stock has declined due to fundamental factors (competitive loss, margin erosion, management failure), and the additional purchase is motivated by the desire to lower the average cost and reduce the paper loss. This behavior compounds the original error by adding more capital to a weakening investment.
The difference is entirely in the motivation, and the investor must be brutally honest about which one is driving the decision. A useful test: if the portfolio did not already hold this stock and the investor were evaluating it as a new purchase at its current price, would the analysis support a buy? If not, buying more simply because the investor already owns it at a higher price is sunk cost behavior.
Case Studies in Sunk Cost Destruction
General Electric. Investors who bought GE stock in 2000 at $60, watched it decline to $40 by 2005, and held on hoping to "get back to even," saw the stock eventually fall to $6 in 2018. The sunk cost of $60 per share kept investors anchored to a recovery that never came, while the company's fundamental problems, including over-leverage, poor acquisitions, and accounting complexity, worsened year after year. Selling at $40, which would have felt like a painful 33% loss, would have avoided the subsequent 85% decline. The endowment effect compounded this error for many holders.
Blackberry (Research In Motion). Shareholders who bought at $140 in 2008 and held through the iPhone's competitive destruction of Blackberry's market position saw the stock decline to below $10 by 2013. At every price point along the decline, the sunk cost of the original purchase made selling feel like "giving up." Meanwhile, the thesis (Blackberry's dominance of the enterprise mobile market) was visibly deteriorating quarter by quarter.
Chinese ADRs (2021-2022). Investors in Alibaba, Didi, and other Chinese technology companies saw massive declines following regulatory crackdowns. Many held through 70% to 80% losses, anchored to pre-crackdown prices. The regulatory environment had fundamentally changed, making the original investment thesis invalid, but the sunk cost of the original purchase prevented timely exits.
Breaking Free
Overcoming sunk cost bias requires deliberate, systematic practices rather than mere awareness.
Conduct a "portfolio review as if new." Quarterly, evaluate each position as if seeing it for the first time. Write a one-sentence thesis for each holding. If the thesis is "I am waiting to get back to my purchase price," the position should be sold immediately. That is not an investment thesis. It is a psychological trap.
Track opportunity cost explicitly. For each losing position held longer than six months, calculate what the proceeds would have earned in the best available alternative. Seeing that the $8,000 trapped in a declining stock could have earned $1,200 in an index fund over the past year makes the cost of inaction tangible.
Set predefined exit criteria. Before buying any stock, define the conditions under which the position will be sold at a loss. "If revenue declines for two consecutive quarters" or "if the competitive moat is breached by a credible competitor" provides a rule-based trigger that bypasses the emotional resistance to selling at a loss.
Reframe losses as tax assets. In taxable accounts, realized capital losses offset capital gains and can reduce taxable income by up to $3,000 per year. A $5,000 loss that saves $1,000 in taxes is not purely negative. It is a partial recovery. This reframe does not make the loss pleasant, but it does reduce the psychological barrier to selling by associating the action with a concrete benefit.
The money is gone. The only question that matters is what to do with what remains. The investor who can separate the emotional experience of a loss from the analytical question of optimal capital deployment will consistently outperform the investor who lets sunk costs dictate portfolio decisions. The position does not know what the investor paid for it. The market does not care. And the investor should not either.
Put these principles into practice. Track fundamentals, build portfolios, and analyze stocks with AI-powered insights.
Start Free on GridOasis →