Week 33 Day 1: Sunk Costs: Money Already Spent Cannot Be Unspent

You bought a stock for $10,000 and it dropped to $5,000. You do not want to sell because you would 'lose $5,000.' But you have already lost $5,000 -- selling just acknowledges the reality. The real question: if you had $5,000 in cash right now, would you buy this stock? If not, sell it and put the $5,000 somewhere better.

The sunk cost fallacy in investing: You hold a losing mutual fund that charges 1% per year because 'you already paid the front-end load.' But the load is gone regardless. The 1% annual fee continues to drain returns. Switch to a 0.03% index fund. You hold individual stocks that have declined because 'it will come back.' Sometimes it does. Sometimes it does not. Enron, Lehman Brothers, and thousands of other stocks went to zero. The company does not know or care about your purchase price. You refuse to sell your house at a loss even though the carrying costs (mortgage, taxes, maintenance) exceed the rental income. The down payment is a sunk cost. Evaluate the house based on current numbers, not what you paid. The fix: the 'clean slate' test. Imagine your portfolio was liquidated to cash overnight. Given your current cash balance, would you reconstruct the exact same portfolio? Every position you would NOT repurchase should be sold. The portfolio you would build from scratch is the portfolio you should own today.

The sunk cost fallacy was formally identified by Thaler (1980) and is one of the most pervasive violations of rational decision theory. Under normative economic theory, only marginal costs and benefits (forward-looking) are relevant to current decisions. Past costs are 'sunk' and irrelevant. However, Arkes and Blumer (1985) demonstrated that the sunk cost effect is robust across contexts: people continue attending a bad movie they paid for, continue eating a bad meal they ordered, and continue holding a bad investment they purchased. In investing, the sunk cost fallacy is closely related to the disposition effect (Shefrin and Statman, 1985; Odean, 1998): investors hold losers too long (hoping to break even, treating the purchase price as a reference point) and sell winners too early (locking in gains relative to the purchase price). Odean (1998) found that individual investors are 1.5x more likely to sell a winning stock than a losing stock -- a behavior that is both tax-inefficient (you should sell losers for the tax benefit) and return-reducing (winners tend to continue outperforming). The neurological basis: fMRI studies (Knutson and Bossaerts, 2007) show that realizing a loss activates the same neural circuits as physical pain (anterior insula, dopaminergic pathways), making loss realization aversive. For index fund investors, the sunk cost fallacy is less relevant (you rarely need to sell a total market fund), but it applies when evaluating whether to switch from a high-cost fund to a low-cost fund, whether to sell a losing position for tax-loss harvesting, or whether to abandon a failing strategy.