Loss aversion and the reference point
People do not evaluate wealth. They evaluate changes from a reference point, losses hurt more than equivalent gains, and below the reference point they become willing to take risks they would otherwise refuse.
Chapter 2 · Beginner
The foundation of the subject, and the one idea worth genuinely understanding, because four later chapters are consequences of it.
Changes, not levels
A standard utility function is defined on levels of wealth: how much you have. The behavioural alternative — prospect theory, from Kahneman and Tversky's 1979 paper — proposes a value function defined on gains and losses instead.
That one substitution has large consequences, because a gain or a loss only exists relative to something. The reference point is doing the work, and it is not given by the world. It is chosen, usually without being noticed.
The shape
Odean states the properties compactly: the function is "concave in the domain of gains and convex in the domain of losses" and "steeper for losses than for gains".
Unpacked, three claims:
Steeper for losses. Losing ₹10,000 registers as worse than gaining ₹10,000 registers as good. This is loss aversion proper, and it is the one most people have heard of.
Concave in gains — risk-averse when ahead. Having made money, you prefer locking in a certain gain to gambling for a larger one.
Convex in losses — risk-seeking when behind. This is the part that gets missed, and it is the dangerous one. Below your reference point, you prefer a gamble to a certain loss. The same person who would not accept a fair coin-flip for ₹50,000 will, while behind, accept far worse odds for a chance to get back to even.
Kahneman and Tversky put it in a sentence Odean quotes directly: a person who "has not made peace with his losses is likely to accept gambles that would be unacceptable to him otherwise".
Where the reference point comes from
Usually the status quo — what you had before. But the paper is explicit that it need not be: gains and losses are sometimes "coded relative to an expectation or aspiration level that differs from the status quo".
For investors the reference point is very often the purchase price, which is where the trouble starts. The purchase price is a fact about your past, not about the asset. The asset does not know it. No future cash flow depends on it. Yet it determines whether you experience the present as a gain or a loss, and therefore which half of the value function you are standing on.
Working the problem
Both investors hold ₹80 of the same stock. Same company, same prospects, same everything that can affect a future return.
The textbook answer: the purchase price is irrelevant. The only question is whether ₹80 of this stock is the best available use of ₹80. Both investors face identical decisions and should do the same thing. Economists call the purchase price a sunk cost, and sunk costs do not enter forward-looking decisions.
The behavioural prediction: the investor who paid ₹50 is up, standing in the concave, risk-averse region, and is relatively willing to sell and bank the gain. The investor who paid ₹120 is down, standing in the convex, risk-seeking region, and will hold — and would need a worse opinion of the stock's prospects to sell it than the first investor would.
Odean states the asymmetry precisely: the belief about expected return "must fall further to motivate the sale of a stock that has already declined than one that has appreciated".
The asymmetry has a cost even when the holding is fine. Suppose the stock genuinely is worth holding. The ₹120 investor is then right by accident, having reached a sound conclusion through a mechanism that would have produced the same answer regardless of the facts. A process that gives the same answer whatever the evidence is not a process.
The practical handle
You cannot stop having a reference point. You can notice which one you are using, and ask the forward-looking question deliberately: if I held cash today, would I buy this at this price? If no, the only thing keeping you in is a number from your own past.
The point
Prospect theory defines value over gains and losses relative to a reference point rather than over levels of wealth, and that function is steeper for losses, concave in gains and convex in losses. So people are cautious when ahead and risk-seeking when behind — and because the reference point is usually the purchase price, a fact about your own history decides which of those two people you are today. The discipline is to ask whether you would buy it now, which is the only question the asset itself can answer.
Check yourself
4 questions. Every answer is explained afterwards, including the ones you get right — guessing correctly is not the same as knowing. Score 70% or more and the chapter is marked done.
Question 1 of 4
0 of 4 answered. You can submit with questions unanswered — they simply score zero.
Now do it with your own numbers
Two investors each hold a stock now worth ₹80. One bought at ₹50, the other at ₹120. Neither has new information. Explain why they are likely to behave differently, and say what the textbook answer would be.
Nothing about the stock differs. The only difference is a number from the past, which standard theory says should be irrelevant.