Chapter 39 Exercises: Ethics of Prediction Markets
These exercises focus on ethical reasoning, case analysis, and the application of ethical frameworks to prediction market scenarios. Many exercises have no single "correct" answer — the goal is to develop your ability to reason carefully about ethical trade-offs.
Section A: Moral Hazard Analysis (Exercises 1-5)
Exercise 1: Moral Hazard Severity Assessment
For each of the following proposed prediction markets, assess the moral hazard risk on a scale of 1 (negligible) to 10 (extreme). Justify your rating by identifying: - Who could cause the event - What the cost of causing the event would be - What the maximum financial incentive from the market would be
Proposed Markets: a) "Will the global average temperature increase by more than 1.5C above pre-industrial levels by 2030?" b) "Will the CEO of TechCorp resign before January 2027?" c) "Will there be a data breach at MegaBank affecting more than 1 million customers in the next 12 months?" d) "Will the price of wheat exceed $400/bushel by December?" e) "Will Professor Smith's research paper be retracted?"
Exercise 2: Position Limit Calculation
A prediction market on whether a small-town mayor will be re-elected pays $1.00 per YES share if the mayor wins. The current price is $0.60. The estimated cost of a campaign sabotage operation (negative advertising, disinformation) that could swing the election is $50,000. The probability of successfully changing the outcome through sabotage is estimated at 30%.
a) What is the expected profit per share from buying NO shares if the sabotage succeeds? b) How many NO shares would a trader need to hold for the expected profit from sabotage to exceed the cost? c) What position limit (in shares) would ensure the financial incentive from sabotage is always negative? d) Discuss whether position limits alone are sufficient to address the moral hazard in this case.
Exercise 3: Conditional Market Design
You are designing a prediction market to aggregate information about cybersecurity threats. The naive approach is a market on "Will Company X be hacked in the next 6 months?" Explain why this creates a moral hazard, and design three alternative conditional markets that preserve information value while reducing the incentive to cause a hack.
Exercise 4: The Whistleblower Dilemma
A pharmaceutical company employee discovers that a drug about to be approved has serious undisclosed side effects. There is a prediction market on whether the drug will be approved. The employee considers: a) Trading on the prediction market (buying NO shares) before disclosing the information to regulators b) Disclosing to regulators first, then trading c) Only trading, never disclosing to regulators d) Only disclosing, never trading
Analyze each option from utilitarian, deontological, and virtue ethics perspectives. Which option do you recommend, and why?
Exercise 5: Assassination Market Countermeasures
Propose and analyze five distinct countermeasures against the assassination market problem. For each countermeasure: a) Describe how it works technically b) Assess its effectiveness at reducing the incentive c) Identify any costs or trade-offs (reduced information value, privacy concerns, etc.) d) Evaluate its feasibility in a decentralized blockchain-based market
Section B: Manipulation and Integrity (Exercises 6-10)
Exercise 6: Manipulation Detection Analysis
You observe the following trading data for a binary prediction market over 10 consecutive hours:
| Hour | Volume | Price Start | Price End | Number of Unique Traders |
|---|---|---|---|---|
| 1 | 150 | 0.50 | 0.51 | 45 |
| 2 | 180 | 0.51 | 0.52 | 50 |
| 3 | 140 | 0.52 | 0.51 | 42 |
| 4 | 2,500 | 0.51 | 0.72 | 12 |
| 5 | 3,200 | 0.72 | 0.78 | 8 |
| 6 | 400 | 0.78 | 0.65 | 55 |
| 7 | 350 | 0.65 | 0.58 | 60 |
| 8 | 200 | 0.58 | 0.55 | 48 |
| 9 | 180 | 0.55 | 0.54 | 44 |
| 10 | 160 | 0.54 | 0.53 | 46 |
a) Identify the likely manipulation period and explain your reasoning. b) Calculate the volume z-score for hours 4 and 5 using hours 1-3 as the baseline. c) What type of manipulation does this pattern most likely represent? d) What would you expect the "true" price to be after the manipulation effect dissipates? e) As a platform operator, what action would you take?
Exercise 7: Self-Fulfilling Prophecy Analysis
A prediction market on "Will Bank XYZ fail within 6 months?" currently shows a price of $0.15 (15% probability). A major financial news outlet publishes an article about the prediction market price. Over the next 48 hours, the price rises to $0.45 as depositors begin withdrawing funds.
a) Is this manipulation, information aggregation, or a self-fulfilling prophecy? Can you distinguish between these? b) Draw a causal diagram showing the feedback loop between market price, media coverage, depositor behavior, and bank health. c) Should prediction markets on bank failure be prohibited? Argue both sides. d) Design a market mechanism that provides information about bank health without creating the self-fulfilling prophecy risk.
Exercise 8: Wash Trading Detection
You are given trading data where two accounts (Account A and Account B) have made the following trades in a market over one week:
- A buys 100 shares from B at $0.55
- B buys 80 shares from A at $0.56
- A buys 120 shares from B at $0.54
- B buys 110 shares from A at $0.55
- A buys 90 shares from B at $0.56
a) Calculate the reciprocity ratio between A and B. b) Calculate the net transfer of shares and dollars between A and B. c) What is the total volume generated by these trades? d) Is this pattern consistent with wash trading? What additional information would you want? e) Write pseudocode for a function that would flag this pattern automatically.
Exercise 9: Price Manipulation Cost
A prediction market uses an automated market maker with the Logarithmic Market Scoring Rule (LMSR) with liquidity parameter $b = 500$. The current price of the YES outcome is $0.40.
a) Calculate the cost of moving the price from $0.40 to $0.70. b) If the true probability is $0.40, what is the expected loss for the manipulator when the market corrects? c) Under what circumstances might a manipulator accept this loss? d) How does increasing the liquidity parameter $b$ affect the cost of manipulation? What is the trade-off?
Exercise 10: Ethical Manipulation Scenarios
For each scenario, determine whether the behavior constitutes manipulation and whether it is ethically problematic. Justify your reasoning:
a) A political campaign buys $500,000 worth of YES shares on their candidate winning, then issues a press release about the prediction market showing their candidate ahead. b) A trader uses a sophisticated statistical model to identify mispriced markets and trades aggressively on their model's signals, moving prices significantly. c) A group of friends coordinates to all buy YES shares at the same time, hoping to trigger algorithmic traders to follow. d) A platform operator adjusts the liquidity parameter of a market to make it harder to manipulate, after observing suspicious trading patterns. e) A trader places large limit orders far from the current price, intending to cancel them if the price moves toward them.
Section C: Equity and Access (Exercises 11-15)
Exercise 11: Wealth-Weighted Beliefs
Consider a prediction market with 5 participants:
| Participant | Wealth ($) | Belief (P(event)) | Max Willing Stake |
|---|---|---|---|
| A | 1,000,000 | 0.30 | 100,000 |
| B | 500,000 | 0.35 | 50,000 |
| C | 50,000 | 0.70 | 10,000 |
| D | 20,000 | 0.80 | 5,000 |
| E | 10,000 | 0.90 | 3,000 |
a) Calculate the wealth-weighted average belief. b) Calculate the equal-weighted average belief. c) If the true probability is 0.75, which average is more accurate? What does this tell us about the relationship between wealth and accuracy? d) Design a mechanism that would give equal weight to each participant's belief while still maintaining incentive compatibility. e) Discuss the trade-offs between accuracy and equity in this scenario.
Exercise 12: Access Audit
You are conducting an access audit for a prediction market platform. The platform has 100,000 registered users. Analyze the following demographic data and identify equity concerns:
| Region | Population (millions) | Users | % of Platform |
|---|---|---|---|
| North America | 380 | 62,000 | 62% |
| Europe | 450 | 28,000 | 28% |
| East Asia | 1,700 | 6,000 | 6% |
| South Asia | 2,000 | 2,000 | 2% |
| Africa | 1,400 | 1,000 | 1% |
| Latin America | 660 | 800 | 0.8% |
| Other | 410 | 200 | 0.2% |
a) Calculate the representation ratio (% of platform / % of world population) for each region. b) Which regions are most severely underrepresented? c) For a prediction market on "Will there be a drought in East Africa?", how might this underrepresentation affect the accuracy of the market? d) Propose three concrete measures to improve geographic representation. e) Discuss whether geographic representation even matters if wealthy, well-connected traders in North America have access to global information.
Exercise 13: Quadratic Participation
Design a quadratic pricing mechanism for a prediction market. Under this mechanism, the cost of purchasing $n$ shares is $n^2 \cdot c$ where $c$ is a base cost parameter.
a) If $c = 0.01$, what is the cost of purchasing 10 shares? 50 shares? 100 shares? b) Compare the ratio of cost-per-share at 100 shares vs. 10 shares under linear and quadratic pricing. c) How does quadratic pricing affect the equilibrium market price compared to linear pricing? d) What are the drawbacks of quadratic pricing for market efficiency? e) Could quadratic pricing be gamed by splitting positions across multiple accounts?
Exercise 14: Play-Money vs. Real-Money Ethics
Compare play-money and real-money prediction markets across the following ethical dimensions. For each, explain which type is ethically preferable and why:
a) Moral hazard risk b) Gambling harm potential c) Equity of access d) Forecast accuracy e) Manipulation resistance f) Information aggregation from insiders g) Overall social value
Exercise 15: Subsidy Design
You have a $1 million annual budget to subsidize prediction market participation for underrepresented groups. Design a subsidy program including:
a) Eligibility criteria (who qualifies?) b) Distribution mechanism (how are funds allocated?) c) Anti-abuse measures (how do you prevent gaming the subsidy?) d) Measurement (how do you assess whether the subsidy improved equity and accuracy?) e) Ethical constraints (what should subsidized participants NOT be allowed to do with the funds?)
Section D: Ethical Frameworks (Exercises 16-20)
Exercise 16: Framework Application Matrix
Apply all four ethical frameworks (utilitarian, deontological, virtue ethics, contractarian) to the following scenario:
A prediction market platform is considering launching a market on "What will be the final death toll of the ongoing earthquake disaster in Country Z?" The earthquake occurred 48 hours ago and rescue operations are underway. The platform argues this market would aggregate information useful for resource allocation decisions.
Create a 4x3 matrix analyzing this scenario: - Rows: Four ethical frameworks - Columns: Arguments for, Arguments against, Verdict (permissible/impermissible/conditional)
Exercise 17: Rawlsian Analysis of Futarchy
Robin Hanson's futarchy proposal suggests that democratic societies should "vote on values but bet on beliefs" — using prediction markets to choose policies predicted to maximize a chosen welfare metric.
a) Describe the futarchy proposal in detail, including how it would work mechanistically. b) Conduct a Rawlsian analysis: behind the veil of ignorance, would rational agents choose futarchy over conventional democracy? c) Identify the worst-off group under futarchy and assess whether they are better or worse off than the worst-off group under conventional democracy. d) Propose modifications to futarchy that would make it more acceptable from a Rawlsian perspective.
Exercise 18: Virtue Ethics Self-Assessment
Using the virtue ethics framework from Section 39.9.3, conduct a self-assessment (or hypothetical self-assessment) of your trading character:
a) Rate yourself on each of the following virtues (1-10) and explain your rating: - Intellectual humility - Epistemic rigor - Courage (willingness to take contrarian positions) - Temperance (moderation in trading) - Justice (fairness to other market participants) - Empathy (awareness of human impact of events traded on) b) For your lowest-rated virtue, describe a concrete plan for improvement. c) Describe a hypothetical situation where pursuing profit would require sacrificing one of these virtues, and explain how you would resolve the conflict.
Exercise 19: Ethical Dilemma Resolution
Resolve the following ethical dilemmas using structured reasoning. For each, identify the ethical principles in conflict, analyze the dilemma from at least two ethical frameworks, and reach a defensible conclusion:
a) You discover that a prediction market on corporate earnings is being systematically used by company insiders. The market's forecasts are exceptionally accurate as a result. Should the platform shut down the market, restrict insider trading, or do nothing?
b) A prediction market on election outcomes in an authoritarian country is providing valuable information to citizens, but the government is using trading records to identify political dissidents. Should the platform continue operating?
c) A researcher discovers that their prediction market experiment caused significant gambling harm to several participants, but also produced important scientific findings about information aggregation. Should they publish the results?
d) A trader realizes that their large position in a catastrophe bond market creates a financial incentive for them to lobby against disaster preparedness spending. The lobbying would be legal. Should they refrain?
Exercise 20: Cross-Cultural Ethics
Prediction markets operate globally, but ethical norms vary across cultures. Analyze how the following cultural differences might affect the ethical analysis of prediction markets:
a) Individualistic vs. collectivist societies: How does this affect the moral hazard analysis? b) Societies with strong gambling prohibitions (e.g., Islamic finance): How should prediction market platforms respect these norms? c) Societies with different privacy expectations: How does this affect the privacy-transparency trade-off? d) Societies with different attitudes toward death: How does this affect the acceptability of death/health markets? e) Propose a set of "universal" ethical principles for prediction markets that could be acceptable across cultures.
Section E: Applied Ethics (Exercises 21-25)
Exercise 21: Platform Ethics Audit
Design a comprehensive ethics audit checklist for a prediction market platform. Your checklist should cover:
a) Market selection and approval processes b) User protection and responsible trading c) Manipulation prevention and detection d) Privacy and data protection e) Equity and access f) Transparency and disclosure
Include at least 5 specific, measurable criteria for each category.
Exercise 22: Ethical Impact Assessment
You are a consultant hired to assess the ethical impact of deploying prediction markets in a developing country's public health system. The markets would predict disease outbreaks, vaccine efficacy, and hospital capacity needs.
a) Identify all stakeholders who would be affected b) For each stakeholder, list potential benefits and harms c) Assess the overall ethical acceptability using a systematic framework d) Propose specific safeguards to maximize benefits and minimize harms e) Identify conditions under which you would recommend against deployment
Exercise 23: Ethical Code of Conduct
Draft a code of conduct for prediction market traders (500-800 words). Your code should address:
a) Obligations to other market participants b) Obligations to the subjects of prediction markets c) Obligations to the broader public d) Self-care and responsible trading e) Reporting of unethical behavior
Exercise 24: Regulatory Ethics
A government regulator is considering three regulatory approaches for prediction markets:
Approach A: Ban all prediction markets except those explicitly approved by the regulator. Approach B: Allow all prediction markets except those explicitly prohibited by the regulator. Approach C: Allow all prediction markets with no restrictions.
a) Analyze each approach from utilitarian, deontological, and contractarian perspectives. b) Identify the types of errors (false positives: banning beneficial markets; false negatives: allowing harmful markets) associated with each approach. c) Which approach do you recommend, and under what conditions? d) Propose a specific list of market categories that should be banned (Approach A) or allowed (Approach B) under your recommended approach.
Exercise 25: Comprehensive Case Analysis
A major technology company wants to launch an internal prediction market for employees to bet on: - Product launch dates - Revenue targets - Competitor actions - Layoff decisions - Executive promotions and departures
Conduct a comprehensive ethical analysis:
a) For each market category, assess moral hazard risk, insider trading concerns, privacy implications, and potential benefits. b) Which markets would you recommend approving, modifying, or rejecting? c) What platform features would you require for the approved markets? d) How would you handle the tension between information value and employee morale (e.g., a market predicting layoffs)? e) Draft a one-page policy memo to the company's CEO with your recommendations.