In 1990, Daniel Kahneman, Jack Knetsch, and Richard Thaler ran an experiment at Cornell University that has since become one of the most-cited findings in behavioral economics.
They gave half the students in a classroom a coffee mug — a plain ceramic mug with the Cornell logo, available at the university bookstore for around $6 — and gave the other half nothing. Then they ran a market. Students with mugs could sell. Students without could buy. The price was determined by demand on both sides.
The expectation, from any rational economic model, was straightforward. The mug had a market value of about $6. Roughly half the mugs should change hands at something close to that price.
What actually happened was that almost no mugs moved.
Students with mugs demanded, on average, around $7 to give them up. Students without mugs were willing to pay, on average, around $3 to acquire one. The gap was so large that the market essentially failed to clear. Owning the mug for a few minutes had roughly doubled its perceived value to the owner.
The researchers called this the Endowment Effect, and it was the empirical demonstration of one of the most consequential findings in behavioral economics: humans don't have stable preferences. We have reference-dependent preferences, and the moment something becomes ours, our reference point shifts. Loss aversion kicks in. We don't want to give it up — even if we never wanted it in the first place.
The Endowment Effect is one of the most useful behavioral principles in commerce. It is also, in its modern application, one of the most dangerous. The line between using it and abusing it is much thinner than most product teams acknowledge.
The Useful Side: Free Trials and Returns
The cleanest commercial application of the Endowment Effect is the free trial. Once you've used a SaaS product for 14 days, integrated it into your workflow, customized your settings, invited your colleagues, the product is yours. The reference point has shifted. Letting it go now feels like a loss.
Free trials work because they activate the Endowment Effect cleanly. The user genuinely gets to try the product. The conversion happens because the trial creates legitimate value the user doesn't want to surrender.
Money-back guarantees work for the same reason in physical retail. Bring it home, try it, see how it fits in your life. By the time the return window expires, the item is yours. The Endowment Effect is doing the conversion work that no salesperson could.
Costco's notoriously generous return policy is, in this framing, not a customer-service expense. It's a commitment device that lets shoppers experiment with confidence. The Endowment Effect ensures that most of what gets bought stays bought.
Patagonia's lifetime warranty does the same job. Apple's 14-day return window does the same job. Every "try before you buy" mechanism in modern commerce is, behaviorally, an Endowment Effect deployment.
This is the principle used well. The customer gets real value. The brand earns trust. The mechanism is reversible if the customer genuinely wants out.
The Dark Side: Dark Patterns
In 2010, a UX researcher named Harry Brignull coined the term dark patterns to describe a related but distinct category — interfaces designed to manipulate user behavior using the same psychological mechanisms, but against the user's interest.
Brignull's original taxonomy included things like:
- Confirm-shaming: an opt-out button labeled "No thanks, I hate saving money"
- Roach motels: easy to sign up, deliberately hard to leave (subscription cancellation that requires phone calls, account deletion buried in submenus)
- Forced continuity: free trials that auto-charge without notice
- Misdirection: visual hierarchy that obscures the user-friendly choice
- Disguised ads: native-styled ads designed to look like organic content
- Privacy zuckering: opt-in privacy decisions presented as opt-out
- Bait and switch: clicking the obvious button triggers a different outcome
What unites all of these is that they exploit the same behavioral mechanisms the ethical applications use — Endowment Effect, Loss Aversion, Reciprocity, Default Bias — but turn them against the user instead of with the user.
In May 2023 the FTC fined Amazon $25 million over the Alexa data-deletion dark pattern — the deletion option was deliberately buried multiple menus deep, and Amazon had retained children's voice data for years after deletion requests. The FTC found this was not an accident but a deliberate engineering choice to reduce deletion rates. Amazon had used the Endowment Effect against its customers: their data was yours until you fought through enough friction to get it back.
The 2023 Bringing Dark Patterns to Light report from the FTC documented systematic dark-pattern use across multiple consumer-facing industries: streaming subscriptions, mobile gaming, dating apps, e-commerce, telecommunications. The pattern was consistent — the easiest behavioral economics interventions to deploy are the ones that exploit the user. The harder ones to deploy are the ones that respect them.
The Patagonia Counter-Example
The cleanest opposite of dark-pattern design is Patagonia's "Don't Buy This Jacket" ad. Published in The New York Times on Black Friday 2011, it featured a photograph of one of Patagonia's R2 fleece jackets with the headline asking customers not to buy it because of the environmental cost of textile production.
Sales of the jacket increased.
The mechanism is interesting. Patagonia was deliberately reducing the Endowment Effect's intensity — they were telling you not to acquire the jacket, not to make it yours. By doing so, they were also signaling something else: we are willing to take an action that costs us in service of a value we hold. This is what Robert Cialdini calls a costly signal of credibility, and in the language of brand trust, it produced enormous returns.
Patagonia's market position has been built largely on the same mechanism. Their lifetime warranty isn't a marketing claim — it's a structural Endowment Effect deployment that says we trust you, and we want this to be yours forever. Worn Wear, Patagonia's repair-and-resale program, extends the same principle: their products don't need to be replaced, because the Endowment Effect is too valuable to walk away from.
This is what ethical use of behavioral economics looks like. The principle is the same. The intent is opposite.
The Diagnostic Question
If you're designing anything that uses an Endowment Effect, Reciprocity, or Loss Aversion mechanism, there's a single diagnostic question worth asking:
If the user clearly understood exactly what you were doing and why, would they consent?
A free trial where the conversion math is transparent passes this test. The user knows they'll be charged after 14 days. They know they can cancel. They're consenting to the mechanism.
A "free trial" that auto-charges without notice, buries the cancellation flow, and obscures the renewal date fails this test. The user, if shown the design philosophy explicitly, would not consent.
The same goes for almost every other behavioral economics deployment. Reciprocity used to build a real customer relationship passes. Reciprocity used to extract one transactional commitment that the customer wouldn't have made under deliberation fails.
I find this diagnostic more useful than any abstract ethical framework. It cuts through the rationalizations. If the user knew what we were doing, would they say yes? If the answer requires the user to not know, you're operating in dark-pattern territory.
What I Take From All This
The hardest thing about working with behavioral economics in product design is that the principles work whether or not you're using them ethically. The Endowment Effect doesn't care if your product genuinely earns the user's loyalty or if you've trapped them in a subscription roach motel. Both will produce reduced churn. Only one will produce trust.
The economic gap between the two strategies compounds over years. Dark-pattern brands typically have higher short-term retention metrics and substantially worse long-term LTV. The trust deficit eventually shows up in support costs, organic acquisition decline, and category-wide regulatory pressure.
The Patagonias of the world look like they're leaving money on the table in any given quarter. Over twenty-year time horizons, they're not.
If you're building anything, take the longer view. The Endowment Effect, used cleanly, is one of the most powerful trust-building mechanisms in commerce. Used as a dark pattern, it's a slow-burning fire that takes down the brand from the inside.
Same principle. Same machinery. Different intent.
The difference, in practice, is everything.