Uncertainty in Microeconomic Decision-Making: Risk, Insurance, and Risk Pooling

Uncertainty in Microeconomics

Introduction to Uncertainty in Economic Behavior

Uncertainty is a fundamental aspect of many economic decisions. This topic extends the utility-maximization model to account for situations where outcomes are not known with certainty. Understanding how individuals cope with uncertainty is essential for analyzing real-world economic behavior.

• Utility-Maximization Model: The standard framework for modeling consumer choices, now adapted to include uncertain outcomes.

• Key Concepts: Insurance and risk pooling are introduced as mechanisms for managing uncertainty.

• Applications: Insurance contracts are only one example; implicit risk-sharing occurs throughout the economy.

Consumer Theory and Uncertainty

Basic Model of Consumer Choice

The consumer choice model consists of constraints, preferences, and utility maximization. Uncertainty is incorporated as a key application.

• Constraints: Budget and resource limitations.

• Preferences: How consumers rank different bundles of goods.

• Utility Maximization: Choosing the bundle that provides the highest utility.

• Applications:

  • Demand Functions

  • Income and Substitution Effects

  • Consumer Surplus

  • Revealed Preference

  • Uncertainty

Uncertain Outcomes and Risk

States of the World and Quantifiable Uncertainty

Economic decisions often involve multiple possible outcomes, each corresponding to a different "state of the world." The likelihood of each outcome can sometimes be quantified using probabilities.

• Examples of Uncertain Outcomes:

  • Outdoor party: good if sunshine, bad if rainy

  • Stock market investment: prices may go up or down

  • Buying a house: good if prices rise, bad if they fall

• Quantifiable Uncertainty (Risk): If probabilities can be assigned to outcomes, the uncertainty is called risk.

• Objective vs. Subjective Probabilities: Probabilities may be based on data (objective) or personal belief (subjective).

Decision-Making Under Uncertainty

Lotteries and State-Contingent Consumption

Choices under uncertainty can be modeled as lotteries, where each alternative yields different payoffs depending on the state of the world.

• Lottery Example: An outdoor party pays $90 with probability 2/3 and $30 with probability 1/3; an indoor party pays $70 with probability 1.

• State-Contingent Consumption Bundles: Consumption depends on the realized state.

Building Blocks of the Model

• States: Numbered 1, 2, ..., n.

• Probabilities: $p_1, p_2, ..., p_n$ with $\sum_{i=1}^n p_i = 1$.

• State-Contingent Bundles: $x_1, x_2, ..., x_n$.

• Utility Function: $u(x)$ gives utility from consuming $x$ in any state.

Binary State Example

• States: 1 (rain), 2 (sunshine)

• Probabilities: $p_1$ and $p_2$, with $p_1 + p_2 = 1$

• Bundles: $x_1$ with probability $p_1$, $x_2$ with probability $p_2$

Expected Value vs. Expected Utility

Ranking Lotteries

Individuals can rank lotteries using either expected value or expected utility.

• Expected Value (EV): $EV = p_1 x_1 + p_2 x_2$

• Expected Utility (EU): $EU = p_1 u(x_1) + p_2 u(x_2)$

• If $u(x) = x$, then $EV = EU$.

Risk Preferences

Definitions and Characterization

Individuals differ in their willingness to accept risk, which is reflected in the shape of their utility functions.

• Risk Neutral: Indifferent to fair bets; utility function is linear.

• Risk Averse: Unwilling to accept fair bets; utility function is concave.

• Risk Loving: Willing to accept fair bets; utility function is convex.

Mathematical Characterization:

• Risk neutral: $u''(x) = 0$

• Risk averse: $u''(x) < 0$

• Risk loving: $u''(x) > 0$

Certainty Equivalent and Risk Premium

The certainty equivalent is the guaranteed amount an individual considers equally desirable as a risky lottery.

• Certainty Equivalent ($x_{ce}$): $u(x_{ce}) = EU$

• Risk Premium (RP): $RP = EV - x_{ce}$

Investing Under Uncertainty

Investment Decision Model

Individuals decide whether to invest in risky assets based on expected utility.

• States: 1 (good outcome), 2 (bad outcome)

• Returns: $(1 + r_1)$ in state 1, $(1 + r_2)$ in state 2

• Investment Decision: Invest if $u(w) < p_1 u(w + r_1 z) + p_2 u(w + r_2 z)$

• Optimal Investment (divisible): $p_1 u'(w + r_1 z^*) + p_2 u'(w + r_2 z^*) = 0$

Insurance and Risk Pooling

Insurance Contracts

Insurance allows individuals to reduce risk by paying a premium in exchange for compensation in adverse states.

• States: 1 (no loss), 2 (loss occurs)

• Consumption: $w$ in state 1, $w - L$ in state 2

• Insurance Policy: Pay premium $P$ in both states, receive claim $C$ in state 2

• Consumption with Insurance:

  • State 1: $x_1 = w - P$

  • State 2: $x_2 = w - L - P + C$

Budget Line and Fair Insurance

The budget line shows the tradeoff between consumption in different states, determined by the insurance contract.

• Fair Insurance: Expected net profit for insurer is zero: $p_1 P + p_2 (P - C) = 0 \implies P = p_2 C$

• Full Insurance: Consumption is equal in both states: $x_1 = x_2$

Optimal Insurance Choice

Individuals choose the amount of insurance to maximize expected utility.

• Optimality Condition: $p_1 u'(x_1) = p_2 u'(x_2)$

• With fair insurance, full insurance is optimal for risk-averse individuals.

• With unfair insurance (premium exceeds fair value), only partial insurance is optimal.

Risk Pooling and Mutual Insurance

Reducing Risk Through Pooling

Risk pooling allows insurers to reduce overall risk by combining many independent risks. As the number of participants increases, the average outcome becomes more predictable.

• Mutual Insurance: Participants share risk, reducing individual exposure.

• Requirement: Risks must not be perfectly correlated; pooling is ineffective if all risks move together (e.g., flood insurance).

• Real-World Example: The 2008 financial crisis demonstrated the failure of risk pooling when housing markets across cities became highly correlated.

Table: Risk Pooling Example

The following table illustrates how risk pooling affects per capita payoffs as the number of participants increases.

Number of Persons	Probability of Outcome	Total Payoff	Per Capita Payoff
1	1/2	100	100
2	1/4	200	100
3	1/8	300	100
4	1/16	400	100

Additional info: As the number of participants increases, the probability that the average payoff is close to the expected value approaches one, illustrating the law of large numbers in risk pooling.