ECON 111 Freakonomics Set 7 – Incentives, incentives, incentives

Incentive Function: Incentives serve as foundational mechanisms using carrots (rewards) and sticks (punishments) to steer behavior and encourage desired outcomes.
Economic / Financial Incentives: Direct monetary rewards or penalties, such as a $3\,\text{dollar}$ per pack cigarette sin tax or wage adjustments.
Social Incentives: Pressures derived from interpersonal dynamics and reputation, such as municipal bans on smoking in public dining spaces to prevent public disapproval.
Moral Incentives: Ethical appeals to do the right thing, such as publicizing that black-market cigarette trafficking finances terrorism.

Policy Misalignment: Interventions backfire when incentives fail to align individual self-interest with intended societal goals.
Mexico City Driving Restrictions (1989): A mandatory one-day-per-week rotating driving ban prompted households to purchase inexpensive, older second vehicles, ultimately increasing net emissions.
London Congestion Charge (2003): Implementing an $8\,\text{dollar}$ fee (later increased to approximately $15\,\text{dollars}$ ) reduced vehicle traffic by $20\,\%$ and boosted bus ridership by $14\,\%$ , despite generating lower-than-projected public revenue and reducing downtown retail foot traffic.
Deputizing Monitoring via Aligned Interests: Fast-food receipt guarantees offering free meals if no receipt is issued deputize customers to monitor cash transactions, eliminating internal register theft without added surveillance costs.
Shared Cost Distortions: Splitting dining bills evenly lowers individual marginal cost to $\frac{1}{6}$ of actual expenditure, incentivizing diners to order more expensive items than when paying individually.

Comparison of gasoline and diesel taxes across countries

Peer Influence and Herding: The peer network is the single strongest predictor of smoking behavior; an increase in peer smoking from $20\,\%$ to $30\,\%$ increases individual smoking probability by $25\,\%$ .
Fundamental Attribution Error: The systematic tendency to attribute actions to personal character traits while ignoring significant situational and social pressures.
Declared vs. Revealed Preferences:
- In residential energy trials directed by Robert Cialdini, subjects stated environmental protection was their primary motivation and peer behavior was least important.
- Revealed preferences demonstrated that door hangers displaying "Join your neighbors in conserving energy" achieved the only statistically significant energy reduction.
Perverse Norm Nudging: Displaying signs at Petrified Forest National Park in Arizona stating that theft removes $14\,\text{tons}$ of wood annually caused theft rates to increase by establishing theft as a prevalent social norm.

Haifa Daycare Study: Introducing a $3\,\text{dollar}$ fine for parents arriving more than $10\,\text{minutes}$ late doubled late pickups; paying the fee erased moral guilt, and late pickups persisted even after the fine was removed.
Charity Canvassing in Israel: High school canvassers raised the highest donations under pure altruistic appeals; introducing a small commission ( $1\,\%$ ) degraded performance below the no-commission baseline by crowding out intrinsic moral duty.
Small Stipends: Blood donations decrease when donors receive modest stipends rather than recognition for altruism.
The Cobra Effect: A bounty on dead cobras in colonial Delhi prompted citizens to breed cobras; when the bounty was discontinued, released cobras made the infestation worse.
Corporate Quotas: The Wells Fargo "Eight is Great" cross-selling quota prompted employees to open millions of unauthorized accounts, culminating in an initial $185\,\text{million}\,\text{dollar}$ fine from the Consumer Financial Protection Bureau.
Tax Enforcement (1987): Requiring Social Security Numbers for claimed dependents on April 15, 1987, eliminated $7\,\text{million}$ fraudulent exemptions in the United States.

Calculus of Fraud: Individuals cheat when benefits are high, risk of detection is low, and penalties are negligible.
Standardized Testing Cheating: Under No Child Left Behind (2002) sanctions, approximately $5\,\%$ of Chicago Public Schools (CPS) classrooms exhibited teacher cheating, characterized by altering $8$ to $10$ consecutive difficult questions for low-performing students.
Algorithmic Detection: Cheating in classrooms is identified by statistical anomalies, including identical strings of right answers on hard questions paired with incorrect answers on simple ones.
Sumo Bout-Rigging:
- Sumo rankings determine elite status, income, and privileges across the $66$ wrestlers in the makuuchi and juryo divisions.
- Achieving an $8^{\text{th}}$ win in a $15$ -bout tournament prevents rank demotion.
- On final tournament days, wrestlers with a $7-7$ record won $79.6\,\%$ of matches against $8-6$ opponents (predicted win rate: $48.7\,\%$ ) and $73.4\,\%$ against $9-5$ opponents (predicted: $47.2\,\%$ ).
- In subsequent rematch bouts with no promotion at stake, the win rate of the former $7-7$ wrestler fell to $40\,\%$ , indicating reciprocal collusion.

Algorithmic pattern detection of identical consecutive answers in suspicious classroom

Reputational Asymmetry: Society punishes admissions of ignorance and rewards confident predictions, encouraging people to conceal lack of knowledge.
Early Development: In testing children aged $5$ to $9$ , $76\,\%$ attempted to answer unanswerable questions rather than stating they did not know.
Forecasting Accuracy: A study by Philip Tetlock evaluating roughly $300$ experts over $20\,\text{years}$ found their forecasts performed no better than crude extrapolation algorithms; CXO Advisory Group found financial market experts achieved only a $47.5\,\%$ accuracy rate.
Overconfidence Bias: Approximately $80\,\%$ of drivers evaluate their driving skills as above the median driver.

Loss Aversion Principle: Losses inflict greater psychological dissatisfaction than equivalent gains provide pleasure relative to a baseline reference point.
Endowment Effect and Status Quo Bias: People disproportionately value resources they currently own, discouraging change; this mechanism underpins $30$ -day trial periods and vehicle test drives.
Chicago Teacher Incentive Experiment: Teachers given an upfront $8{,}000\,\text{dollar}$ bonus that was subject to clawback if students missed benchmarks achieved significantly better student math outcomes than teachers offered identical bonuses paid at year-end.
Pre-Paid Commissions: Sales representatives receiving commissions in advance outsell those paid retroactively, even when retroactively paid options offer higher nominal compensation (e.g., $10\,\text{dollars}$ advance vs. $12\,\text{dollars}$ retroactive).
Voltage Drop in Policy Scaling: The drop in intervention efficacy when moving from small trials to population scale, caused by administrative sludge, indirect outcome metrics, limited recipient attention, and population heterogeneity.