Exhaustive Study Notes on Asymmetric Information and Market Failure

Conceptual Foundations of Asymmetric Information

• Asymmetric information refers to a situation where information about a good or service is unevenly distributed between the buyer and the seller. One party possesses superior knowledge compared to the other.
• Sellers of wine typically know the quality of their product better than buyers. This disparity can lead to market failure because buyers, unable to accurately assess quality, face the risk of "buying a cat in a sack." This risk reduces their willingness to pay, potentially causing the market for high-quality products to disappear.
• Under the assumption of perfect information, all consumers and producers are fully informed about the good they wish to trade. However, in reality, market participants often have limited and unevenly distributed information.
• Common examples of asymmetric information:
  • Used Cars: Sellers have intimate knowledge of defects, correct mileage, and past repairs. Buyers gain only limited or distorted information through brief inspections or test drives.
  • Hospitalization Insurance: The policyholder (buyer) knows their own health condition better than the insurer.
• Asymmetric information is categorized based on when it occurs relative to a contract:
  • Pre-contractual (Ex-ante): Asymmetric information exists before a contract is concluded (e.g., used car quality).
  • Post-contractual (Ex-post): Asymmetric information exists after a contract is concluded (e.g., insurance behavior).

Imperfect Information versus Asymmetric Information

• Imperfect information occurs when something is not known with certainty. This includes incomplete, uncertain, or incorrect knowledge.
• Imperfect information can relate to the future (e.g., next summer's weather or China's economic growth) or the present (e.g., hidden defects in a house or the exact origin of a work of art from the past).
• Imperfect information may affect all parties in a market. In a used car sale, both the buyer and the seller have imperfect information because even the seller may not perfectly reflect the effective quality of the vehicle.
• Differences between the two concepts:
  • Asymmetric information always involves at least two market players where one is better informed than the other.
  • Asymmetric information implies that at least one party must have imperfect information.
  • If all parties have the same imperfect information, there is imperfect information but no asymmetric information.
• Asymmetric information creates risk, which refers to uncertain events whose probabilities can be objectively estimated. Examples include:
  • A $10\%$ chance a job applicant is not a hard worker despite their claims.
  • A $30\%$ chance that antiques at a flea market are fake.
  • A $0.05\%$ chance of car theft in Belgium.

Ex-ante Asymmetric Information and Adverse Selection

• Adverse selection occurs when the presence of low-quality products and asymmetric information drives high-quality products out of the market.
• Products are categorized as:
  • Lemons: Inferior or defective purchases.
  • Peaches: High-quality purchases without defects.
• This concept was described by George Akerlof in 1970 using the second-hand car market, for which he received the Nobel Prize in Economics in 2001 alongside Michael Spence and Joseph Stiglitz.

The Economics of used Car Markets: Situation 1

• Consider a market with 1,000 sellers of lemons and 1,000 sellers of peaches. In a state of perfect information, reservation prices are as follows:
  • Buyer Reservation Price for Lemons: $€3\,000$
  • Buyer Reservation Price for Peaches: $€6\,000$
  • Seller Reservation Price for Lemons: $€2\,000$
  • Seller Reservation Price for Peaches: $€4\,000$
• In this scenario, $1\,000$ lemons trade at $€3\,000$ and $1\,000$ peaches trade at $€6\,000$, achieving an efficient market equilibrium.
• If asymmetric information is introduced, buyers cannot distinguish quality and estimate a $50\%$ probability for each type. The willingness to pay becomes the expected value ($EV$):
  • $EV = 0.5 \times €3\,000 + 0.5 \times €6\,000 = €4\,000$
• At a price of $€4\,500$, both lemons and peaches are traded ($1\,000$ each), and the market remains efficient in terms of total surplus, though the distribution of surplus changes:
  • Sellers of lemons and buyers of peaches are better off.
  • Sellers of peaches and buyers of lemons are worse off (buyers of lemons realize a negative surplus).

The Economics of used Car Markets: Situation 2 (Market Failure)

• Suppose sellers of peaches require at least $€5\,000$.
• Buyers, still estimating a $50\%$ probability, are only willing to pay $€4\,500$. Because the price is lower than the peach seller's reservation price ($€5\,000$), no peaches are traded.
• Over time, buyers realize only lemons are available, shifting the probability of getting a lemon to $100\%$. The expected value drops to $€3\,000$.
• Asymmetric information causes the market for good second-hand cars to disappear, leading to an efficiency loss despite a potential win-win (buyers value peaches at $€6\,000$; sellers value them at $€5\,000$).

Adverse Selection in Insurance

• In hospitalization insurance, the insured has better information than the insurer regarding health, age, and lifestyle (e.g., smoking, obesity, risky sports).
• If insurers calculate premiums based on average risk, low-risk individuals find the premium too expensive and exit the insurance pool.
• The insurer is left with only high-risk individuals, forcing premiums to rise and causing a portion of the potential surplus to disappear.

Ex-post Asymmetric Information and Moral Hazard

• Moral hazard occurs after a contract is concluded. It involves one party improving their situation through unobservable decisions (hidden actions) or manipulated information (hidden information).
• It is often described as inappropriate or immoral behavior where an individual exploits an information advantage through opportunistic behavior.
• Examples of Moral Hazard:
  • Insurance: Arson of insured buildings, cutting off an insured finger, or neglecting roof maintenance hoping for storm damage reimbursement.
  • Insurance Fraud: Secretly selling a smartphone and then declaring it stolen.
  • Corporate Management: Managers investing in ineffective but familiar projects rather than those with the highest value for shareholders.
  • Workplace: Employees spending time on social media or Netflix during working hours.

The Principal-Agent Problem

• The principal-agent problem arises when the interests of employees or managers (agents) are not aligned with those of shareholders (principals).
• The principal cannot directly observe the actions and efforts of the agents, allowing agents to pursue their own objectives at the expense of the owner's profits.

Societal Welfare and Market Failure in Insurance

• Moral hazard affects societal welfare because the individual does not bear the full cost of their actions, leading to "too much" consumption (e.g., overconsumption of health treatments or increased bicycle thefts).
• This behavior increases provider costs and insurance premiums. In extreme cases, insurance may not be offered at all (e.g., coverage for breaking spectacle lenses or all-risk insurance for short trips).

Case Study: Bicycle Theft Insurance

• Assuming no moral hazard: $1\,000$ policies are traded at a price of $€5$. This price equals the marginal cost ($MC$), which is the probability of theft ($0.01$) multiplied by the bicycle value ($€500$):
  • $\text{Expected Payout} = p \times V + (1 - p) \times 0 = pV = 0.01 \times €500 = €5$
• With moral hazard: Policyholders are less cautious, increasing the theft probability to $0.02$. The marginal cost rises to $€10$. Fewer policies ($450$) are traded at the higher price, creating an efficiency loss (represented by a triangle on a supply-demand graph).

Managing Adverse Selection

Screening

• Screening involves acquiring additional information to resolve or bypass asymmetric information.
• It often involves an expert inspection. In used cars, if an expert charges $€250$ and provides accurate quality assessment, symmetric information is restored.
• Reservation prices adjust to the expert fee (e.g., a peach valued at $€6\,000$ becomes $€5\,750$ for the buyer). Peaches can then be traded at $€5\,750$, restoring efficiency.
• Caveats for screening:
  1. The cost of the expert must not be too high; if it exceeds the surplus potential ($€1\,000$ in the car example), it won't be used.
  2. Experts can fail; if misjudgment is high, the willingness to pay remains too low to trade.
• Examples: ISO quality certifications, mandatory car inspections for sales in Belgium.

Selection (Self-Selection)

• The less-informed party offers choices that cause the better-informed party to reveal information through their behavior.
• Solomon Judgment: King Solomon proposed cutting a child in two; the real mother revealed herself by offering to give the child up to save its life, while the false mother agreed to the division.
• Airlines: Charging different prices for business travelers and tourists by adding restrictions (e.g., booking far in advance or staying a Saturday night) that only tourists will accept.
• Insurance: Requiring medical exams; high-risk individuals often self-select away by refusing the examination.

Signalling

• Signalling involves the better-informed party taking initiative to credibly convince the less-informed party of quality.
• To be credible, the signal's cost must be high enough that those with poor-quality products cannot afford it.
• Examples:
  • Art: Bernaerts auction house offers a $3$-week return policy if an object is proven to be a forgery.
  • Warranties: Guarantees on appliances signal reliability, as sellers of poor products would face prohibitive costs from high claim rates.
  • Education: Diplomas and extracurriculars (e.g., student leadership, internships, studying abroad) signal soft skills and work ethic to employers.

Result-Based Payments

• Payments are made only when a specific event occurs.
• Examples:
  • Orchards: A seller offers to take a $10\%$ share of the annual yield for $15$ years instead of a lump sum to signal the orchard's productivity.
  • Football Transfers: Resale clauses specify a percentage of future transfer fees, signalling conviction in a player's potential.
• The less-informed party can also use result-based offers to force the seller to choose, which exposes the seller's true expectations of quality.

Managing Moral Hazard

Monitoring and Verification

• Supervision/Monitoring: Checking if actions were deliberate or if an employee made sufficient effort.
• Verification/Audit: Checking the accuracy of provided information (e.g., verifying the size of an insurance claim).
• Comparisons: Measuring an employee's productivity against the average productivity of all employees.

Incentive Remuneration Systems

• Using variable pay to align agent interests with the principal.
• Scenario: A worker chooses light effort ($€0$ cost) or heavy effort ($€10\,000$ cost). Revenues depend on effort and luck ($50\%$ probability).
  • Revenues from Light Effort: $€10\,000$ (Bad luck) or $€20\,000$ (Good luck).
  • Revenues from Heavy Effort: $€20\,000$ (Bad luck) or $€40\,000$ (Good luck).
• Fixed Pay ($w$): The worker always chooses light effort because their welfare maximized at $w - 0 = w$ rather than $w - 10\,000$. Expected profit for the employer: $0.5 \times (10\,000 - w) + 0.5 \times (20\,000 - w) = 15\,000 - w$.
• Variable Pay: The worker keeps all revenue above $€18\,000$.
  • Light effort welfare: $0.5 \times 0 + 0.5 \times 2\,000 = €1\,000$.
  • Heavy effort welfare: $0.5 \times (2\,000 - 10\,000) + 0.5 \times (22\,000 - 10\,000) = €2\,000$.
  • The rational employee chooses heavy effort, and the employer's expected profit rises to $€18\,000$.

Risk Sharing in Insurance

• Bonus-malus system: Premiums for car insurance fluctuate based on the number and size of claims.
• Co-payment: The insured pays a portion of the cost (e.g., healthcare visits or civil liability claims). This limits overconsumption but must balance moral hazard containment with the need for sufficient insurance coverage.

Organized Insurance Fraud

• Organized car policy fraud costs Belgian insurers between $€120$ million and $€240$ million annually.
• Roughly $3\%$ to $6\%$ of car insurance premiums are used to finance fraud; on an average $€400$ premium, $€12$ to $€24$ is paid to fraudsters.
• Association Assuralia and anti-fraud group Alfa Belgium developed a claims database containing accident data and vehicle identification numbers to detect fraud.
• Industry estimates suggest $5\%$ to $10\%$ of all insurance returns are fraudulent.