Behavioral economics challenges the core assumptions of traditional rational choice theory (RCT) by showing that real human decisions systematically deviate from the ideal of fully rational, consistent, self-interested optimization. Key points:
1) Bounded rationality vs. full optimization
- RCT assumes unlimited cognitive capacity and costless computation. Behavioral work shows people use heuristics (e.g., availability, representativeness, anchoring) that are fast but biased.
- Limited attention and rational inattention: people miss or ignore information, leading to context-dependent choices.
2) Unstable, reference-dependent preferences vs. stable, context-free preferences
- Prospect theory: choices depend on gains/losses relative to a reference point, not just final wealth. Value function is concave for gains, convex for losses, with loss aversion (losses loom larger than equal-sized gains).
- Framing effects and preference reversals show that how options are described or ordered changes choices, violating description invariance and sometimes transitivity.
- Endowment effect: ownership increases valuation, contradicting standard, reference-free utility.
3) Deviations from expected utility theory
- Allais paradox: violations of the independence axiom.
- Ellsberg paradox: ambiguity aversion conflicts with subjective expected utility.
- Probability weighting: people overweight small probabilities and underweight moderate/large ones.
4) Time inconsistency vs. exponential discounting
- RCT uses exponential discounting implying dynamically consistent plans.
- Behavioral evidence supports present-biased, quasi-hyperbolic discounting (β-δ), leading to procrastination, undersaving, and demand for commitment devices.
5) Bounded willpower and self-control problems
- People intend to act one way (diet, save) but later deviate. This contradicts the idea that a single, consistent preference ordering governs all intertemporal choices.
6) Social preferences vs. pure self-interest
- Fairness, reciprocity, altruism, and inequity aversion affect behavior (e.g., ultimatum and dictator games), challenging the assumption of purely selfish utility.
7) Noise and stochastic choice vs. deterministic optimization
- Random utility, limited discrimination, and choice overload produce variability not explained by standard maximization with stable preferences.
8) Market and asset-pricing anomalies vs. efficient markets
- Disposition effect, momentum and overreaction/underreaction, equity premium puzzle, limited arbitrage: psychological biases can persist in aggregate outcomes, challenging full-efficiency assumptions.
Modeling tools and alternatives introduced by behavioral economics
- Prospect theory and cumulative prospect theory (value function + probability weighting).
- Reference-dependent and loss-averse preferences (including in labor supply and consumption).
- Quasi-hyperbolic discounting and models of self-control/dual-self.
- Salience, attention, and rational inattention models.
- Social preference models (Fehr–Schmidt, Bolton–Ockenfels).
- Choice architecture and defaults: small frictions and design features (defaults, framing, partitioning) shape outcomes.
Policy implications
- Nudges and libertarian paternalism: redesigning choice environments (default enrollment, simplified forms, reminders, commitment devices) can improve decisions without restricting choice.
- Recognition that “revealed preference” may not equal welfare; welfare analysis may require identifying true long-run or “experienced” preferences.
Bottom line: Behavioral economics doesn’t reject optimization altogether; it relaxes strong RCT assumptions by incorporating psychological realism—bounded rationality, reference dependence, present bias, and social motives—yielding models that better predict actual behavior and inform more effective policy and market design.