Behavioral Economics: Reasoning About Systematic Deviations¶
Standard economic tools assume rational agents maximizing well-defined objectives with consistent preferences. Real people don't work this way. They're loss averse, present-biased, anchored by irrelevant numbers, susceptible to framing, and prone to treating sunk costs as relevant. This isn't a minor footnote - it changes everything.
Behavioral economics started as a catalog of deviations from rationality. But the real insight is that the deviations are systematic. People don't fail randomly - they fail in predictable, exploitable, and correctable ways. Loss aversion isn't noise; it's a consistent pattern with measurable magnitude (losses hurt about 2x as much as equivalent gains please).
The extraction principle: these aren't just biases to avoid but tools to apply. Loss aversion isn't just something to watch out for in yourself - it's something to design for in systems you build. Status quo bias isn't just a mistake - it's a lever for influencing behavior. The tools work descriptively (predicting what people will do) and prescriptively (designing better choices).
These tools are humbling. They reveal how far human behavior strays from textbook rationality. But they're also empowering. Once you understand the systematic errors, you can correct for them in yourself and design around them in systems.
Reference-Dependent Preferences¶
- Loss Aversion and Reference Dependence
- Framing Effects
- Mental Accounting
- Anchoring
- Status Quo Bias and Default Effects
Time and Self-Control¶
Beliefs and Reality¶
Quick Reference¶
Decision Type -> Tool¶
| You're asking... | Start with... |
|---|---|
| Why won't they accept this fair deal? | Loss Aversion |
| Why does presentation change choices? | Framing Effects |
| Why won't they reallocate resources? | Mental Accounting |
| Why is this number influencing them? | Anchoring |
| Why won't they change from the current option? | Status Quo Bias |
| Why can't they stick to their plan? | Time Discounting |
| Why do they keep investing in failure? | Sunk Cost Fallacy |
| What do they really want? | Revealed Preference |
| Why did the prediction cause itself? | Reflexivity |
| Why did this become dominant? | Network Effects |
| Why does this bad pattern persist? | Evolution and Selection |
Reading Path¶
Foundations (start here):
- Kahneman, "Thinking, Fast and Slow"
- Thaler and Sunstein, "Nudge"
Going deeper:
- Ariely, "Predictably Irrational"
- Thaler, "Misbehaving"
Advanced:
- Kahneman and Tversky, "Choices, Values, and Frames" (collected papers)
- Ainslie, "Picoeconomics"
Usage Notes¶
These tools describe systematic patterns in human behavior. They're probabilistic, not deterministic.
Domain of applicability: Strong where people face unfamiliar decisions with complex trade-offs. Weaker where people have extensive experience and feedback - experts in their domain show fewer biases for decisions within their expertise.
Limitations: Not everyone shows every bias. Biases can be reduced with training, experience, and well-designed choice architecture. Calling something a "bias" doesn't mean the behavior is always wrong - sometimes biases are adaptive.
Composition: Loss aversion and framing combine - framing a choice as gain vs loss invokes loss aversion. Anchoring and adjustment connects to mental accounting - anchors stick partly because they're assigned to mental accounts. Status quo bias and sunk cost interact - the status quo is partly favored because changing feels like "wasting" prior investment.
Integration: Behavioral tools complement core economics (they explain where rational-agent models fail) and information economics (how do people actually process information?). Use them together: what would a rational agent do (core), what information problems exist (information), and how will actual people deviate from this (behavioral)?