Sunk Cost Anchoring

Anchoring on Purchase Price: The Sunk Cost Fallacy

You've been building a deck in the backyard for three months. You've spent $6,000 in materials and 60 hours of your time. Halfway through, you discover the wood is warping badly, the design you chose doesn't work with the yard's drainage, and finishing it will require another $8,000 and 80 hours to produce something you no longer actually want.

Rational analysis says: evaluate the decision from here. What it costs to complete versus abandon are the only relevant figures. The $6,000 and 60 hours already spent are gone regardless of what you do next. They cannot be recovered. They should not influence whether you finish.

But they do. Abandoning the half-built deck feels like admitting failure, wasting what you've already invested, and "giving up" on a project that carries a substantial prior commitment. Most people finish the deck.

This is the sunk cost fallacy—continuing an investment of time, money, or resources because of what has already been spent, rather than because of what the future investment will produce. Combined with the anchoring bias—the cognitive tendency to rely too heavily on an initial piece of information when making subsequent decisions—it produces a category of financial error that is both common and preventable.

WHAT MAKES COSTS "SUNK"

A sunk cost is any expenditure that has been made and cannot be recovered. The money is spent, the time is consumed, the resources are gone. By definition, a sunk cost is irrelevant to future decisions—it will not change based on any action taken going forward.

The economic principle is simple: future decisions should be based on future costs and future benefits, compared at the margin. What you've already spent is not a future cost. It's a past loss that will exist regardless of what you do next.

People don't feel this way. The sunk cost triggers a psychological obligation to justify the prior expenditure—to not "waste" it by failing to complete the thing it was invested in. But the investment has already been wasted if the future completion doesn't justify the future cost on its own merits. Completing the deck doesn't un-waste the $6,000; it merely adds $8,000 more to the total cost of an outcome you no longer want.

ANCHORING: HOW THE PURCHASE PRICE HOLDS ON

Anchoring is a separate but related bias: the tendency for the first number encountered in a decision context to disproportionately influence subsequent evaluations, even when the first number is arbitrary or irrelevant.

In investing, the purchase price functions as an anchor for how an investment's current performance is perceived. A stock purchased at $120 that is now at $85 feels like a loss. A stock purchased at $60 that is now at $85 feels like a win. The current price—$85—is identical in both cases. The investment opportunity going forward is identical. But the prior purchase price creates a reference point that shapes the emotional experience of the holding.

This anchoring on purchase price causes systematic errors:

The reluctance to realize losses (covered extensively in the loss aversion article) is partly an anchoring phenomenon: the purchase price is the anchor, and the current price's distance below that anchor feels like a loss that shouldn't be locked in.

The premature realization of gains is its mirror: when a stock rises above the purchase price anchor, the investor has "made money" and is tempted to lock in the gain—sometimes before the investment has reached its optimal exit point.

In real estate, the price paid for a home anchors what sellers consider acceptable offers. A home purchased for $380,000 that has declined in market value to $340,000 feels impossible to sell for $340,000—even though $340,000 is what the market is willing to pay. The $380,000 anchor makes any lower price feel like a loss that shouldn't be accepted. The result: homes sit on the market, accumulating carrying costs, while the seller waits for a buyer willing to ignore market conditions and meet the seller's anchored price.

Research by Genesove and Mayer (2001) in the Quarterly Journal of Economics documented this pattern directly: homeowners with negative equity (who bought at prices above current market value) set higher asking prices relative to market conditions, experienced longer time on market, and ultimately received prices closer to their anchored purchase prices—but at the cost of significant opportunity cost during extended marketing periods.

THE BUSINESS CASES THAT ILLUSTRATE THE SCALE

Sunk cost and anchoring errors are not confined to individual investors and home sellers. Some of the most documented business failures in history involve institutional-scale commitment to sunk costs:

Concorde aircraft: The British and French governments continued funding Concorde development for decades after internal analyses showed the project would never be commercially viable. The sunk costs were so large—measured in billions—that the psychological commitment to not "wasting" what had been spent overrode rational discontinuation. The term "Concorde fallacy" has entered the literature of decision science as the canonical large-scale sunk cost example.

The common thread across corporate and individual sunk cost examples: the larger the prior investment, the stronger the obligation to continue—even though the size of the prior investment is precisely irrelevant to whether continuation makes sense.

ESCALATION OF COMMITMENT

A specific pattern associated with sunk costs is escalation of commitment: the tendency to increase investment in a failing course of action precisely because prior investment has been made. The psychological mechanism is that additional investment reframes the sunk cost as potentially recoverable—it transforms a "past loss" into a "temporarily underwater position" that further investment might rescue.

In investing, escalation of commitment produces "doubling down" on losing positions: buying more of a declining stock to average down the cost basis, under the logic that the lower average price makes eventual recovery more attainable. Whether the additional investment in the declining stock is justified depends entirely on the stock's future prospects from its current price—the original cost basis is irrelevant to whether it's a good investment at the current price.

In business, escalation of commitment appears in companies that continue throwing resources at failing product lines, markets, or technologies rather than redirecting to better opportunities—because the sunk investment in the failing direction is too psychologically large to abandon.

THE CORRECTIVE: ZERO-BASED EVALUATION

The most effective corrective for sunk cost thinking is zero-based evaluation: periodically setting aside the history of an investment and evaluating it as if encountering it for the first time.

For a stock holding: "If I had $12,000 in cash and was evaluating investment opportunities, would I choose to invest it in this company at its current price, with its current competitive position and financial condition?" This question separates the investment decision from the ownership history—the question of whether the stock is a good investment from today's price is independent of what you paid for it.

For a project: "If someone proposed this project to me today—starting from where we currently are, with the current completion costs and projected outcomes—would I approve it?" The prior expenditure doesn't appear in this framing because it's genuinely irrelevant to the forward-looking evaluation.

For a real estate decision: "At the price I'm asking, would a buyer in this market, with access to all available comps, offer this price?" If the honest answer is no, the ask is anchored to the purchase price rather than to market conditions.

THE COST OF THE DECK

The half-built deck costs $6,000 and 60 hours, with another $8,000 and 80 hours to complete. Completing it produces something you no longer want. Abandoning it produces a tearing-down project that might cost $1,000 to $2,000.

Abandonment cost: $7,000 to $8,000 total (sunk $6,000 + demolition $1,000 to $2,000), no further time investment. Completion cost: $14,000 to $15,000 total (sunk $6,000 + remaining $8,000), plus 80 more hours, producing an outcome you don't want.

The forward-looking decision is unambiguous once sunk costs are excluded: abandon costs $1,000 to $2,000 more in future expenditure but saves 80 hours and produces no unwanted outcome. Complete costs $8,000 more in future expenditure plus 80 hours to produce something you don't want.

The sunk cost made abandonment feel impossible. Once extracted from the decision, abandonment is clearly correct.

Every financial decision involving prior investment deserves this extraction. The money already spent has no vote in what you do next.