Rational choice and its limits
In standard economic models, people are rational: they have clear preferences, full information and they maximise utility (satisfaction) by comparing marginal benefits and costs. Firms maximise profit. This model is simple and often useful.
Behavioural economics mixes economics with psychology. It tests how people actually choose, often with experiments. Its first big idea is bounded rationality (Herbert Simon): our decisions are limited by:
- limited information (we cannot know every price),
- limited time, and
- limited brain power to compare many options.
So people often satisfice: they choose the first option that is good enough instead of searching for the very best. Bounded self-control matters too: we know what is good for us but still fail to do it (for example, saving or exercise).
Heuristics and biases
A heuristic is a mental shortcut or rule of thumb. It saves effort but can cause a bias: a regular, predictable error.
- Anchoring: we lean too much on the first number we hear. A high 'original price' makes a sale price seem a bargain.
- Availability: we judge how likely something is by how easily we remember examples. After news of a plane crash, people overestimate flying risk.
- Herd behaviour (social proof): we copy what others do, e.g. a busy restaurant looks better; share-price bubbles.
- Loss aversion: losses feel about twice as strong as equal gains, so people hold losing shares too long.
- Framing: the same facts, put differently, change choices: '90% fat-free' sounds better than '10% fat'.
- Present bias: we overvalue now and undervalue the future (buying snacks instead of saving).
- Status quo / default bias: we tend to stick with whatever is already set.
Social norms, fairness and the ultimatum game
Social norms are unwritten rules about how people behave. We are influenced by what others do and what they think of us.
People also care about fairness, not just their own money. In the ultimatum game, player A gets 10 coins and offers some to B. If B accepts, the split happens. If B rejects, both get nothing. A purely self-interested B would accept even 1 coin. In real experiments in many countries, most A players offer 4–5, and offers below about 2–3 are often rejected. People pay a cost to punish unfairness.
In the dictator game, B cannot refuse, yet many A players still give something. This shows altruism and the wish to be seen as fair. Firms know this: a shop that raises umbrella prices in a storm may be seen as unfair and lose customers later.
Nudges, choice architecture and ethics
Choice architecture is the way choices are presented: order, defaults, wording. A nudge (Thaler and Sunstein) changes behaviour in a predictable way without banning any option or changing money incentives much.
- Default choice: automatic enrolment in a pension scheme (opt-out) raises saving a lot compared with opt-in.
- Restricted choice: limiting the number of options so people are not overwhelmed.
- Mandated choice: people must make an active choice, e.g. on organ donation when getting a licence.
- Placement: fruit at eye level in a school canteen.
- Social comparison: 'You used more electricity than 80% of your neighbours'.
- Framing of warnings: large pictures on tobacco packs.
Strengths and limits
Nudges are cheap and keep freedom of choice. But effects may fade, they may not work for everyone and they do not solve big problems such as poverty alone. Firms can also use 'sludge' or dark patterns to push people against their interests (hidden fees, hard-to-cancel subscriptions).
Ethical questions
Is it right for a government to steer choices? Supporters call it libertarian paternalism: guiding people toward what they would choose with full information, while letting them opt out. Critics worry about manipulation, who decides what is 'good', and transparency. A fair test: would the nudge still work if people were told about it?
Key formulas and definitions
- Rational choice: choose where marginal benefit = marginal cost
- Bounded rationality: limited information, time and computing power
- Heuristic: mental shortcut; bias: predictable error it causes
- Anchoring, availability, herd behaviour, framing, present bias, status quo bias
- Loss aversion: a loss feels about 2× as strong as an equal gain
- Ultimatum game: B rejects → both get 0
- Nudge: changes choice architecture without banning options or big money incentives
Worked examples
1. A phone is advertised: 'Was 30,000, now 22,000'. Another identical phone is simply priced 22,000. Shoppers rate the first as a better deal. Which bias is this?
Step 1: Both phones cost the same. Step 2: The only difference is the first number shown, 30,000. Step 3: That number acts as an anchor and makes 22,000 seem low. Step 4: This is anchoring.
2. In an ultimatum game with 100 coins, A offers B 10. A rational self-interested B should accept. Why might B reject, and what does this show?
Step 1: Accepting gives B 10 and A 90; rejecting gives both 0. Step 2: Money-wise, 10 > 0, so the standard model predicts acceptance. Step 3: B may feel the 90–10 split is unfair and is willing to lose 10 to punish A. Step 4: This shows people value fairness and social norms, not only their own payoff.
3. A company switches its pension plan from opt-in to opt-out. Membership rises from 45% to 88%. Explain using behavioural economics.
Step 1: Nothing about the plan's money changed. Step 2: Only the default changed. Step 3: Status quo bias and inertia mean most people keep the default. Step 4: With opt-out, the default is to save, so membership jumps. This is a nudge through choice architecture.
4. Evaluate: 'Governments should use nudges instead of taxes to reduce sugary drink consumption.'
For: nudges (smaller default cup sizes, placing water first) are cheap, keep freedom and can change habits. Against: effects may be small or fade; heavy consumers may ignore them; a sugar tax changes prices and raises revenue for health. Judgement: nudges work best alongside taxes and information, and should be transparent.
Common mistakes
- Saying behavioural economics claims people are stupid. It says people are predictably limited, and their errors follow patterns.
- Calling every policy a nudge. A ban or a large tax is not a nudge; a nudge keeps all options and does not change incentives much.
- Mixing up anchoring and framing. Anchoring is about a reference number; framing is about how the same information is worded.
- Thinking rejecting a low offer in the ultimatum game is irrational in every sense. It is not money-maximising, but it can be rational if fairness and reputation matter.