Economic theory has a clean prediction: people decide based on future consequences and ignore the past. If you have already spent $100 on a concert ticket and the concert is tonight, but you are exhausted and would rather rest, the $100 is a sunk cost. It is gone, irretrievable, irrelevant. The only decision that matters is: Would I rather spend tonight at the concert or at home? Attend if concert > rest, skip if rest > concert. The $100 should have zero influence on this forward-looking decision.1
But humans do not decide this way. You bought the ticket, it cost $100, and that $100 haunts the decision. The thought "I already spent $100, I cannot waste it" carries disproportionate weight. You drag yourself to the concert despite exhaustion, partly because the pain of "wasting" the $100 feels sharper than the pleasure of rest would feel. An Econ would never attend. A Human does, driven by the sunk cost.1
This gap between the theory (ignore sunk costs) and behavior (cannot ignore sunk costs) is the problem that Supposedly Irrelevant Factors (SIFs) identify. SIFs are factors that economic theory says should be irrelevant but empirically are essential to human decision-making.1
The name is precise: these factors are "supposedly" irrelevant (the theory says they should not matter) but empirically manifest as deeply relevant (people cannot ignore them). Once you start looking for SIFs, they are everywhere. They are not bugs in human decision-making; they are the core of how humans actually decide. Ignoring SIFs is ignoring how humans work.1
Sunk costs are the paradigm SIF. Once money, time, or effort has been spent and cannot be recovered, economic theory says it should not influence future decisions. But humans cannot ignore sunk costs. They feel obligated to "get their money's worth," to justify the expenditure by using the good or service purchased.1
A person buys a membership to a gym for $1,000 for the year, pays upfront. In week three, they realize they hate the gym, do not enjoy exercising there, would be happier with any alternative use of time. An Econ says: skip the gym. Future utility is what matters, and alternative uses of time produce higher utility. But the Human's mind is still attached to the sunk cost: "I paid $1,000, I must go to justify the expenditure." The person attends the gym miserable, every visit a miniature loss, because the sunk cost drives the decision. The $1,000 is irrelevant to whether this gym is a good use of time going forward, but it is supremely relevant to whether the Human will use the gym.1
This extends beyond small purchases. A person invests $50,000 in a business that is clearly failing. Liquidating would save the remaining $20,000 from further loss. But the founder cannot liquidate because the $50,000 already spent feels wasted if they quit. They continue pouring money and time into the failing business, creating a trajectory where sunk costs justify further losses. Sunk costs are deeply binding on human decisions, even though they are theoretically irrelevant.1
When you buy a coffee, economic theory cares about one thing: the consumption utility (how good the coffee is, how much pleasure it produces). The utility from consuming the coffee — its objective quality — is the acquisition utility.1
But humans also care about the deal. When you buy coffee at a café for $5, if you feel the price is fair, you experience positive transaction utility (the deal felt good). If you feel the price is gouging, you experience negative transaction utility (you got ripped off, the deal felt bad). The identical coffee — with identical acquisition utility — produces very different total utility depending on whether the transaction felt fair.1
In one famous experiment, researchers offered to sell mugs at different prices: $5 to one group, $10 to another. The mug's quality was identical. The $5 group bought more mugs and reported satisfaction (fair price, positive transaction utility). The $10 group bought fewer and reported dissatisfaction (too expensive, negative transaction utility). The price itself is a SIF — it should be irrelevant to whether the mug is desirable, but it is highly relevant to whether people feel they got a good deal. Transaction utility is the pleasure or pain of the deal itself, independent of the product's objective quality.1
This explains why people feel ripped off when they learn they paid $50 for something a friend bought for $35 (negative transaction utility from overpaying), even though the product's objective value has not changed. The discovery of an alternate price became a reference point, and their transaction utility shifted retroactively. The product is identical; the satisfaction is lower because the transaction frame shifted.1
Econs should value owning a mug at its fair market price — no more, no less. The mug is worth $X whether owned or not; ownership should not change its valuation. But ownership changes everything. As documented in the endowment effect, people demand significantly more to sell something they own than they would have paid to buy it originally. Ownership itself becomes a SIF — theoretically irrelevant (the mug's utility to you did not change when you acquired it), empirically decisive (you now value it 50-100% higher).1
The current state — the status quo — should be neutral. You should evaluate available options independently: Option A has utility level 8, Option B has utility level 7. You choose A. But the status quo biases this evaluation. If you currently have Option B, you require Option A to be meaningfully better (utility 10, not just 8) to justify switching. Humans are biased toward maintaining the current state, even when alternatives are objectively superior. The status quo is a SIF — your current state should be irrelevant to evaluation of alternatives, but it is dispositive.1
This is why people keep poor investments, stay in unfulfilling jobs, remain in unsatisfying relationships longer than rational: the status quo bias makes staying require no justification, while leaving requires overcoming the loss frame of departure. The status quo is a SIF so powerful it locks people into suboptimal equilibria.1
A coffee at the coffee shop normally costs $5. You visit on a day when the café raises the price to $7 because it is busy and demand is high (pricing optimization — higher demand should support higher price). Economic theory says the higher price is irrelevant if it is what the market will bear. But humans see this as unfair gouging. The $5 price has become a reference point, a perceived "fair" price, and the $7 is a violation of fairness — a SIF. The café's demand, its costs, its market position — all irrelevant to the fairness calculation. The reference price matters more than economic fundamentals.1
Calling these factors "supposedly irrelevant" can make them sound like mistakes — departures from rationality. But they are not. SIFs often reflect sophisticated adaptations to real constraints.1
Sunk costs may seem irrational, but they reflect a principle: if you have committed resources and the commitment is uncertain to pay off, continuing to invest signals that you believe in the outcome enough to keep paying. Sunk costs create a commitment signal. Ignoring sunk costs might mean constantly starting new projects and abandoning them; honoring sunk costs means staying committed to decisions long enough for them to work out. There is adaptive value in not abandoning every project the moment doubt emerges.1
Transaction utility reflects the fact that fairness itself is a real constraint on behavior. A supplier who feels their transaction is unfair (they are being pressured to accept low prices) will reduce quality, service, and reliability. A customer who feels ripped off will reduce loyalty and recommend competitors. Fairness is not decorative; it is economic. Transaction utility is the mechanism that enforces fairness and reciprocity.1
Endowment effect reflects the fact that ownership creates genuine relationship and familiarity value that bare economic theory does not capture. A mug you own has been in your hand, holds temperature predictably, has developed micro-patterns from use. This relational value is real. Endowment effect may overshoot this value, but the underlying fact (ownership creates relational value) is economically legitimate.1
A store that suddenly increases prices when it is the only option in an emergency (the only gas station after an accident, the only restaurant when traveling) faces not just demand elasticity but fairness-based boycotts and reputational damage that economic pricing optimization does not predict. The "supposedly irrelevant" factor (that people feel gouged when prices spike in emergencies) is empirically more consequential than the economic factor (demand sensitivity). Firms that ignore fairness SIFs in pricing often face consumer backlash that costs more than the price increase earned.1
A firm laying off workers should, in economic theory, do so based on efficiency: cut the least productive workers. But the way layoffs are done is a SIF that determines long-term morale and performance. A firm that lays off workers in a way that feels unfair (no warning, no severance, visible executives kept) suffers higher exit of good workers (those with outside options leave, reducing quality), lower morale (survivors are less committed), and reputational damage. A firm that lays off in a way that feels fair (advance notice, severance, executives take cuts too) retains better workers and maintains culture. The fairness SIF is economically material.1
Customers should switch suppliers based on current price and quality. But customers' willingness to switch depends on whether they feel the current supplier's price increases are fair or exploitative — a SIF. A supplier that gradually increases prices while maintaining relationship and service feels less unfair than a supplier that jacks prices suddenly. Customers tolerate price increases from suppliers they trust more than from suppliers they feel are extracting value unfairly. The fairness SIF explains why customers stay with suppliers even when competitors are cheaper: the transaction (staying) feels fair, the switch (leaving a trusted supplier) feels like a betrayal or an admission the supplier was exploiting them.1
Psychology: Fairness Norms in Pricing — SIFs are often fairness-based: sunk costs trigger fairness obligations (you should honor commitments), transaction utility reflects fairness in exchange, endowment effect respects fairness around ownership. Fairness is not a separate domain; it is the psychological law that makes SIFs binding. People internalize fairness norms, and SIFs are the behavioral expression of those norms.
History: Machiavellian Realpolitik — Machiavelli's insights about power and loyalty operate through SIFs. A ruler cannot rule by economic logic alone (threatening execution if you disobey). Rulers must consider fairness SIFs (people tolerate bad outcomes if they feel they are fair; they rebel against good outcomes if they feel unfair). A price increase feels fair if it is explained (costs rose) and universal (everyone pays). A price increase feels like exploitation if it is sudden and targeted. Machiavelli understood that the "supposedly irrelevant" factor — perceived fairness, honor, gratitude — is empirically dispositive to loyalty and rebellion.
Cross-Domain: Bias as Adaptive Heuristic — SIFs appear to be biases (irrelevant factors that should not influence decisions) but are actually adaptive heuristics. Honoring sunk costs prevents constant project abandonment. Respecting fairness norms sustains reciprocal relationships. Endowment effects stabilize society by making people defend their possessions rather than constantly trading. SIFs are not bugs; they are features of a system optimized for reciprocity, commitment, and fairness rather than momentary utility maximization.
The Sharpest Implication: If economic theory is built on the assumption that SIFs should be irrelevant but humans cannot ignore them, then economic theory is built on a false foundation. Every policy recommendation based on economic logic that ignores SIFs will fail or produce unintended consequences. A government that raises taxes without explaining how they are fair (who benefits, how the funds are used, why they are necessary) triggers fairness SIFs and faces backlash even if the tax is economically sound. A business that cuts wages to match market rates without considering fairness SIFs (explaining the necessity, offering alternatives) triggers employee defection even if the wages are competitive. The implication is that understanding human behavior requires understanding SIFs, not dismissing them as departures from rational theory. SIFs are the actual substance of how humans decide; economic theory is the elegant fiction.
Generative Questions: