Prospect Theory: The S-Curve That Rewrote Economics

In the spring of 1969, a psychologist named Daniel Kahneman invited a colleague to speak to his graduate seminar at the Hebrew University of Jerusalem. The colleague was Amos Tversky, a thirty-two-year-old mathematical psychologist who had already earned a reputation as one of the sharpest minds in the department. The two men knew each other casually but had never worked together. They had almost nothing in common.

Kahneman had spent his childhood hiding from Nazis in occupied France, moving from attic to barn to borrowed apartment, losing his father to a stroke six weeks before D-Day. He grew up cautious, self-doubting, and tuned to the ways the mind deceives itself. Tversky had grown up in the new state of Israel, served as a paratrooper in the IDF, and earned a citation for bravery after sprinting from behind cover during a training exercise to tackle a soldier who had frozen after lighting a bangalore torpedo fuse. Tversky shielded the man with his own body and took shrapnel. He was confident, organized, and utterly unafraid of being wrong in public. Kahneman was the opposite on every dimension.

That afternoon, Tversky told Kahneman's students about research suggesting that ordinary people function as reasonably good "intuitive statisticians." Kahneman listened, waited for the talk to end, and told Tversky the whole thing was wrong. Not a detail. The premise. Kahneman's own research had shown him that even trained scientists made systematic errors in statistical reasoning. If researchers couldn't trust their statistical intuitions, how could anyone claim ordinary people were good at it?

Instead of defending his position, Tversky got curious. The two men went to lunch. The lunch became a conversation that lasted the afternoon. The conversation became a research program. The research program became the most influential collaboration in the history of the social sciences.

For the next decade, people passing their office at Hebrew University would hear shouting arguments and raucous laughter behind a closed door. They worked so closely that neither could later remember which ideas had belonged to whom. They flipped a coin to decide who would be listed as first author on their initial paper and alternated after that. Kahneman was the typist, because Tversky never mastered the skill. Tversky was the one who could hold the entire mathematical structure in his head. Together they produced a theory that explained why human beings are systematically irrational in the way they evaluate risk, and why that irrationality follows a precise mathematical shape.

They called it prospect theory. The paper they published in 1979, "Prospect Theory: An Analysis of Decision under Risk," appeared in Econometrica and became one of the most cited articles in the history of economics, referenced tens of thousands of times. It overturned a century of economic thinking that assumed people make rational calculations about risk. And the core idea, the one that matters most if you set prices, negotiate deals, or decide when to pivot, can be drawn on a napkin.

The S-Curve: How the Brain Actually Calculates Value

Classical economics assumes that people evaluate outcomes in absolute terms. A gain of $500 feels like a gain of $500 whether you're starting from zero or from a million. A loss of $500 feels like the opposite of a gain of $500. The math is clean, symmetric, and rational. It's also wrong.

Kahneman and Tversky proposed something different. They argued that people don't evaluate outcomes as endpoints. They evaluate them as changes from wherever they happen to be right now. Your brain doesn't compute "I will have $10,500." It computes "I will have $500 more than I have now." The starting point, what they called the reference point, isn't an objective calculation. It's whatever the brain has decided is "normal" at this moment. And everything that follows is measured as a departure from that baseline.

This is reference dependence, and it's the first principle of prospect theory. The same $50,000 salary feels like a windfall to someone making $30,000 and a pay cut to someone making $70,000. The number hasn't changed. The reference point has. Every price on your pricing page, every term in your negotiation, every metric in your board deck is being evaluated not on its absolute merits but on its distance from whatever your audience has decided is normal.

The second principle gives the theory its famous shape. When Kahneman and Tversky plotted how people actually experience gains and losses, the curve was S-shaped. On the gains side, it rises steeply at first and then flattens. The jump from $0 to $100 feels significant. The jump from $1,000 to $1,100 barely registers. This is diminishing sensitivity: each additional dollar of gain produces less psychological impact than the one before it. The same pattern holds for losses: the drop from $0 to negative $100 is agonizing, but the drop from negative $1,000 to negative $1,100 feels like noise.

But the curve isn't symmetric. The loss side drops more steeply than the gain side rises. Losing $100 produces roughly twice the psychological intensity as gaining $100. This is loss aversion, and it means the value function is kinked at the reference point. The brain weighs what you might lose about twice as heavily as what you might gain, because neural hardware built for survival treats losing your food supply as a bigger threat than finding a second one.

The third principle is what makes prospect theory dangerous for entrepreneurs. Diminishing sensitivity doesn't just flatten your experience of gains. It changes your relationship with risk. In the domain of gains, when you're ahead, you become risk-averse. You have something to protect, and the marginal value of gaining more is decreasing, so you play it safe. In the domain of losses, when you're behind, you become risk-seeking. You've already absorbed the worst of the pain, and the marginal cost of losing more is decreasing, so you gamble. The same person who would never bet $10,000 on a coin flip when they're ahead will eagerly take that bet when they're $10,000 in the hole. Not because the math changed. Because the S-curve changed which part of the function they're operating on.

The Experiment That Made Economists Uncomfortable

In 1981, Tversky and Kahneman published a demonstration of prospect theory so clean that it's still taught in every behavioral economics course forty-five years later. They gave participants a scenario: an unusual disease is expected to kill 600 people. Two programs are proposed. Participants had to choose one.

The first group received the options framed as gains. Program A: 200 people will be saved. Program B: a one-third probability that 600 will be saved, and a two-thirds probability that nobody will be saved. The expected value is identical. Two hundred lives either way. But 72 percent chose Program A, the safe option. When you're in the domain of gains, you protect what you have.

The second group received the same options framed as losses. Program C: 400 people will die. Program D: a one-third probability that nobody will die, and a two-thirds probability that 600 will die. Again, identical expected value. But now 78 percent chose Program D, the risky option. When you're in the domain of losses, you gamble.

Same disease. Same population. Same math. The only variable was whether the outcome was described as people saved or people killed. The frame shifted the reference point, the reference point determined whether participants were operating in the gain or loss domain of the S-curve, and the S-curve determined their tolerance for risk.

This wasn't a laboratory curiosity. A pricing page that frames the purchase as gaining a feature activates the risk-averse side of the curve, and the customer picks the cheaper option. The same page framed as avoiding a loss, "Stop bleeding $2,000/month in inefficiency," activates the risk-seeking side, and the customer is willing to pay more. The framing effect isn't a separate phenomenon from prospect theory. It's a downstream consequence of the S-curve.

Why Founders Make Their Worst Decisions When They're Losing

Prospect theory predicts something specific and testable about entrepreneurship: founders who perceive themselves as behind their reference point will systematically take larger risks than founders who perceive themselves as ahead. And the data confirms it.

Consider the pivot decision. A founder has spent eighteen months and $400,000 building a product that isn't getting traction. The rational move is to evaluate remaining capital against alternative opportunities. But the founder isn't standing at zero on the S-curve. They're at negative $400,000, deep in the loss domain. Diminishing sensitivity means the pain of losing the next $100,000 barely registers compared to the pain they've already absorbed. So they double down, pouring remaining runway into the same failing product, because the S-curve has made the incremental loss feel small and the potential recovery feel large.

Investors see this pattern constantly: founders who raised money on a thesis that isn't working, who know it isn't working, and who can't change direction. It looks like stubbornness. It's usually prospect theory. The founder's reference point is the fundraise, the version of the future they pitched. Every month of declining metrics puts them further into the loss domain, which makes them more risk-seeking, which makes them less likely to cut losses, which puts them further into the loss domain. The S-curve creates a feedback loop that looks from the outside like irrational commitment and from the inside like courage.

The antidote is resetting the reference point. If a founder can genuinely reset to "where am I now, with the information I have now," the same decision that felt like abandoning $400,000 reframes as allocating $200,000 of remaining capital toward its highest-value use. The math is identical. But the founder is now operating on a different part of the S-curve, risk-averse about protecting what's left rather than risk-seeking about recovering what's gone.

This is also why the first year after a successful exit is psychologically treacherous. The founder's reference point resets to a much higher baseline. A startup that generates $2 million in its first year would have been exhilarating before the exit. After a $50 million exit, it registers on the loss side of the S-curve. The founder who once slept on a couch for a chance at success is now risk-averse about ventures that would have thrilled them five years earlier. Same person. Different reference point. Different part of the curve.

Pricing, Negotiation, and the Pain of Paying

Prospect theory doesn't just explain risk decisions. It provides the mathematical scaffolding underneath every principle in pricing psychology.

Start with the pain of paying. In 1998, Drazen Prelec and George Loewenstein published a framework showing that the act of paying for something activates genuine neural pain, processed in the same brain regions that handle physical discomfort. This is loss aversion applied at the moment of transaction: handing over money is a loss, and the brain processes it at roughly double the intensity of the gain it's receiving. Every pricing strategy is, at its core, a strategy for managing this asymmetry.

Subscription pricing works because it decouples the pain of paying from the moment of consumption. When you pay Netflix $15.99 per month, the loss happens once. But you consume the service thirty days in a row. Each viewing session is a gain with no corresponding loss to offset it. The S-curve predicts exactly this: separate your losses and combine your gains. A single $192 annual payment produces less total pain than twelve monthly payments of $15.99, because the loss side of the curve has diminishing sensitivity. One large loss hurts less than twelve small ones. Annual billing at a discount isn't just a retention strategy. It's a prospect theory strategy.

The inverse principle applies to gains. A single large gain produces less total pleasure than multiple smaller gains. This is why video games drip rewards in small, frequent bursts rather than one massive payout. It's why the best onboarding sequences deliver a series of small wins rather than one comprehensive setup. Each individual gain hits the steep part of the curve. Bundle them together and you slide into the flat part where additional value barely registers.

Negotiation follows the same logic. If you're negotiating a deal and need to make three concessions, deliver them as a single package. Three separate concessions create three separate loss events, each hitting the steep part of the curve. One combined concession creates a single loss event that diminishing sensitivity makes more tolerable. Conversely, if you're delivering value, break it into pieces. "We'll include free onboarding, plus a dedicated account manager, plus quarterly business reviews" hits the steep gain curve three times. "We'll include our premium support package" hits it once.

The Prize One of Them Couldn't Accept

In 1996, Amos Tversky was diagnosed with metastatic melanoma. He was fifty-nine years old. He continued working from his home in Stanford until the last days of his life, colleagues later said, with good humor and a calmness that was characteristically his. He died on June 2, 1996.

Six years later, on October 9, 2002, the Royal Swedish Academy of Sciences announced that Daniel Kahneman had won the Nobel Memorial Prize in Economic Sciences "for having integrated insights from psychological research into economic science, especially concerning human judgment and decision-making under uncertainty." The work they cited was, overwhelmingly, the work Kahneman had done with Tversky. The Nobel Prize is not awarded posthumously.

Kahneman told the New York Times: "I feel it is a joint prize. We were twinned for more than a decade."

The story of the prize has a quality that prospect theory itself might have predicted. Kahneman, by most accounts, experienced the Nobel not as a pure gain but as a loss-inflected one. The person who should have been standing beside him was gone. Reference dependence means that the experience of winning depends entirely on what you expected the moment to look like. A Nobel Prize, minus the collaborator who made the work possible, doesn't sit cleanly on the gain side of the curve.

What Kahneman and Tversky built together in that decade behind closed doors at Hebrew University has shaped virtually every field it has touched. Prospect theory explains why people buy lottery tickets (overweighting small probabilities of large gains), why they hold losing stocks too long (risk-seeking in the loss domain) and sell winning stocks too early (risk-averse in the gain domain), why your pricing page converts better with loss framing, and why your best founder friends make their worst decisions when their company is failing.

The S-curve is operating on every decision your customers make, every decision your team makes, and every decision you make. You can't turn it off. But you can learn to see which part of the curve you're standing on.