Economics of Information – Comprehensive Study Notes

Introduction

• Perfectly competitive markets (Micro 1 result) ⇒ allocations are Pareto-efficient
  • Key assumption: complete information (all agents know all relevant facts at all times).
• Micro 2 relaxes this assumption → studies markets where one party holds more information than the other (asymmetric information).
• Two canonical cases of asymmetry:
  • Hidden (exogenous) characteristic → adverse selection.
  • Hidden (endogenous) action → moral hazard.
• Framework: principal–agent problem
  • Principal hires/contracts an agent.
  • If incentives diverge and information is asymmetric, inefficient outcomes emerge.

Key Concepts & Definitions

• Asymmetric Information: Information relevant for a trade/contract is not symmetrically distributed among agents.
• Principal–Agent Problem: One party (principal) delegates an action to another (agent) whose interests or information differ.
• Hidden Characteristic: Quality, risk type, or ability is known to agent but not to principal (exogenous).
• Hidden Action: Effort, care, risk-taking chosen by agent and unobservable to principal (endogenous).
• Pareto-Efficiency: Allocation where no one can be made better off without making someone else worse off.

Adverse Selection

• Occurs when individuals with undesirable (to the principal) hidden characteristics self-select into a contract.
• Classic illustration: Used-car ("lemons") market
  • Three quality types, equally frequent:
  • High: 15{,}000\,€
  • Medium: 12{,}000\,€
  • Low: 9{,}000\,€
  • Sellers know car quality exactly; buyers only know distribution.
  • With symmetric info, trade price e.g. 15{,}250\,€ for a high-quality car ⇒ Pareto improvement.
  • With asymmetric info, buyers offer at most the expected value of a randomly drawn car.
  • If buyers expect all three types equally, expected value = \frac{15{,}000+12{,}000+9{,}000}{3}=12{,}000\,€.
  • Sellers of high-quality cars refuse to sell at 12{,}000\,€ ⇒ they exit market.
  • Buyers revise beliefs: remaining cars are medium or low. New expected value =11{,}500\,€, etc.
  • Converges to outcome where only low-quality cars trade or market collapses.
• Result first formalised in Akerlof (1970), "The Market for ‘Lemons’".
• Core intuition: Price no longer signals quality once quality is hidden ⇒ high-quality supply disappears.

Other Real-World Examples

• Insurance: High-risk customers know more about own risk, buy more coverage, driving up premiums.
• Labour markets: Low-ability workers may apply more aggressively when wage contracts ignore ability.

Policy & Contractual Solutions to Adverse Selection

When characteristic verifiable ex-post (Akerlof setting)

• Independent experts / certification
  • Third-party inspections, e.g., mechanic checks, credit ratings.
  • Reduce info gap but costly; some agents still worse off vs. full info.
• Warranties / guarantees
  • Seller promises reimbursement or return if quality revealed low later ⇒ incentivises truthful revelation.
• Government minimum standards / regulation (e.g., TÜV)
  • Lowers transaction costs of private certification.

When characteristic NOT verifiable ex-post (Rothschild–Stiglitz setting)

• Example: General health risk; insurer cannot confirm true type even after contract expires.
• Competitive screening contracts
  • Menu of policies where low-risk types self-select into low-coverage/low-premium, high-risk into high-coverage/high-premium.
  • Still second-best inefficient (pooling impossible without subsidy).
• Compulsory insurance
  • Mandated universal coverage forces risk pooling.
  • Pareto-improvement possible: high-risk subsidised, low-risk guaranteed minimum insurance.

Moral Hazard

• After contract signed, agent’s action invisible/unobservable.
• Forms:
  • Ex-ante moral hazard (before damage): inadequate precaution, excessive risk-taking.
  • Ex-post moral hazard (after damage): inefficient claim behaviour, over-utilisation of services.
• Examples
  • Insured driver speeds, skips maintenance.
  • Employee shirks when supervision lax.
  • Patient overuses health care once insured.

Intuitive Mechanics

• Insurance makes marginal cost of risky behaviour to agent lower (some cost shifted to principal), shifting optimal effort level downward.
• Leads to higher probability or magnitude of losses than socially efficient.

Mitigation Instruments

Ex-ante Moral Hazard

• Deductibles / co-payments: Agent bears first d units of loss ⇒ reinstates marginal cost.
• Coverage limits / liability caps: Insurer only pays up to ceiling.
• No-claims bonus: Future premium discount conditional on reporting no claims.
  • Creates intertemporal link between behaviour and cost.
• Trade-off: Lower moral hazard but reduced risk-sharing; expected utility < first-best where effort observable.

Ex-post Moral Hazard

• Similar tools: deductibles, co-insurance on repair costs.
• Monitoring & audits: e.g., medical second opinions.

Empirical Evidence of Ex-ante Moral Hazard

• Higher insurance coverage ⇒ lower precaution (Breyer, Zweifel & Kifmann 2013).
• U.S. citizens turning 65 (Medicare eligibility) ↓ physical activity (Dave & Kaestner 2009).
• Naloxone access laws ↑ opioid-related ER visits & theft (Doleac & Mukherjee 2022).

Conclusion

• Asymmetric information on characteristics (adverse selection) or actions (moral hazard) yields inefficient market outcomes.
• Private or public mechanisms can mitigate but rarely restore full efficiency.
• Compulsory insurance may correct adverse selection but simultaneously introduces moral hazard, necessitating second-layer incentives (deductibles, monitoring).
• Overall theme: Balance between risk-sharing and incentive preservation defines optimal contract/institution design.

Recommended Reading

• Frank, Robert H. & Cartwright, Edward (2013). Microeconomics and Behavior, Ch. 6.
• Classic papers: Akerlof (1970); Rothschild & Stiglitz (1976).