Health, Medical Care, and Insurance

What Produces Health?

There is no honest pie chart that assigns one permanent percentage of health to doctors, behavior, genes, or income. The importance of each source differs across people, places, conditions, and time. The categories also interact.

Consider heart disease. Inherited risk may matter. Smoking, diet, sleep, and exercise may matter. Air pollution, workplace conditions, stress, and neighborhood safety may matter. Public-health information and food-safety rules may matter. Medical care may prevent, detect, and treat disease. Chance still matters after all of those influences are considered.

The same checklist applies across many health problems: biology and inherited risk; behavior; the physical and social environment; income, education, and other constraints; public health; medical care; and chance. These influences overlap. Income can affect living conditions and access to care. Environment can shape behavior. Medical care may matter greatly for one condition and much less for another. The list organizes questions; it does not assign fixed shares of responsibility.

This way of thinking prevents two common errors. The first is to assume that health insurance or medical spending determines health by itself. The second is to notice that behavior matters and leap to the conclusion that health is entirely a matter of willpower.

Behavior belongs in economic analysis because people respond to incentives and constraints. A safe place to walk, the price of cigarettes, the time required to prepare food, addiction, work schedules, and information can all change choices. Explaining a behavioral channel is not the same as blaming a person.

Obesity, Risk, And Changing Medical Technology

The long-run rise in obesity in the United States illustrates why health cannot be reduced to access to doctors. Figure 19.1 reports measured obesity prevalence among adults ages 20 to 74. The estimate rose from 13.4 percent in 1960-1962 to 40.8 percent in August 2021-August 2023.²

The trend is a fact to explain, not an explanation by itself. Food, work, transportation, recreation, sleep, biology, income, stress, and medical care all changed over these decades. Obesity also is not the same as metabolic syndrome, a cluster of risk factors associated with diabetes, heart disease, stroke, and other health problems.³

The health and budget stakes are large. Chronic disease can require more physician time, medicines, and hospital capacity. It can raise private insurance premiums and public spending. CBO projects federal outlays for major health programs to grow from about 6.0 percent of GDP in 2026 to 6.7 percent in 2036.⁴ Unless taxes or borrowing rise, faster growth leaves less room for other government priorities.

Two programs account for most federal health spending. Medicare is federal health insurance mainly for people age 65 and older, along with some younger people with qualifying disabilities or conditions. Medicaid is a joint federal-state program covering eligible people with low incomes and some people with disabilities. CBO estimates that federal outlays in 2026 will be about $1.1 trillion for Medicare and $708 billion for Medicaid. The Medicaid figure counts only the federal share, not additional state spending.⁵

A straight-line forecast could therefore look grim: more chronic disease, more treatment, and ever greater pressure on families and government. But that forecast would assume that medical technology stands still. A long-run health-spending forecast is also a forecast about future medical technology, whether it says so or not.

It does not. The pace is visible even in ordinary news coverage: new compounds, new uses for older medicines, and new trial results seem to appear constantly. GLP-1-based medicines are a striking example. These drugs act on hormone signals involved in blood sugar and appetite. In major trials of selected adults without diabetes, semaglutide produced average weight loss of 14.9 percent over 68 weeks, while the two highest tested doses of tirzepatide produced average losses of 19.5 and 20.9 percent over 72 weeks.⁶ A later trial found that semaglutide also reduced major cardiovascular events in a selected high-risk group.⁷

These results do not guarantee the same benefit for every patient. Price, side effects, continued use, and long-run health effects still matter. But they demonstrate the larger point: forecasts of future illness and spending must allow for treatments that do not yet exist.

If an innovation prevents expensive disease at a reasonable cost, its value may reach far beyond the patient. Private insurers, Medicare, Medicaid, employers, families, and taxpayers may all benefit. Such innovation could ease future budget pressure and leave more resources for other purposes. But the result is not automatic. An expensive treatment with small or short-lived benefits could increase spending instead.

Why Reward Medical Innovation?

The value created by a successful medicine is not the same as the profit earned by its producer. Patients may live longer or feel better. Families, employers, insurers, and taxpayers may also gain. Economists Kevin Murphy and Robert Topel once estimated that a 1 percent reduction in cancer mortality would be worth about $500 billion to the U.S. population in their model.⁸ The estimate is not about one drug. It illustrates how large the social value of better health can be.

Developing a medicine can require years of research, failures, clinical trials, and regulatory review. Once the formula is known, manufacturing may be far cheaper than discovering and proving the treatment. A patent offers a temporary period without direct imitation so that a successful innovator may recover those earlier costs.

The trade-off is familiar from Chapter 15. Patent protection encourages research, but the resulting market power can raise prices and restrict access. It also misses some valuable ideas. Recent working-paper evidence on new uses for existing drugs suggests that research falls sharply when firms cannot protect the knowledge created by proving that a new use works.⁹

Public research funding, grants, prizes, and other rewards can fill some gaps. None is perfect. Society needs some way to pay for the search for tomorrow’s treatments, and the design of that reward affects what researchers choose to pursue.

Why Medical Care Is Different

Buying a routine physical exam is not like buying a sandwich. A patient may not know whether care is needed, which treatment is best, whether waiting is dangerous, or what the final bill will be. The provider has specialized knowledge, but the patient must trust that knowledge. Illness can make delay costly and comparison shopping difficult. Insurance separates the price of care from the price paid when care is used.

Kenneth Arrow made uncertainty the center of his classic 1963 analysis of medical care.¹⁰ Illness is uncertain. The success of treatment is uncertain. Knowledge is unevenly divided between patients and professionals. Institutions such as insurance, licensing, professional ethics, and long-term trust can be understood partly as responses to those problems.

Arrow’s point is even more useful today than a simple argument over whether health care should be called a market. Medical care already operates through a mixture of prices, professional rules, insurance contracts, employers, public programs, regulation, and organizational trust. The question is how those institutions affect choices and outcomes.

Insurance Pools Risk

Imagine 100 people. During the coming year, one person is expected to have a medical loss of $50,000, but no one knows in advance who it will be.

The expected claims for the group are $50,000. Dividing by 100 gives expected claims of $500 per person:

\[ \$50{,}000 \div 100 = \$500 \]

Why might someone willingly pay a premium above $500? The reason is risk aversion. A certain cost of somewhat more than $500 may be easier to bear than a small chance of losing $50,000. Insurance is valuable because it replaces a large uncertain loss with a smaller predictable payment.

The premium must also cover more than expected medical claims. Insurers process claims, prevent fraud, maintain reserves, hire workers, use capital, and bear the risk that actual claims exceed expectations. A private insurer also needs a return on capital. A premium above $500 is therefore not by itself proof of exploitation.

This arithmetic explains the value of insurance, but not how a successful pool is formed. People know different things about their likely risks, and coverage changes the price paid when care is used. Those two facts bring us back to Chapter 12.

Adverse Selection: Who Joins The Pool?

Suppose an insurer offers the same premium to people who privately know that their expected medical costs differ. People who expect high costs will find the policy more attractive. People who expect low costs may decide that the premium is not worth paying.

If enough lower-risk people leave, average claims among those who remain rise. The insurer raises the premium to cover those expected claims. The higher premium gives more lower-risk people a reason to leave. The cycle can continue.

This process creates pressure, not a law that every insurance market must collapse. Employer group coverage can bring people with different risks into one pool. Subsidies can keep lower-risk people enrolled. Risk adjustment can transfer funds toward plans with higher-cost members. Participation requirements or waiting periods can change the gain from waiting until sickness appears. Underwriting can price some observable risks, although it may weaken protection for the people who need insurance most and may be restricted by law.

Every response has a trade-off. A rule that stabilizes a pool may limit choice, require taxes, create administrative costs, or expose people to different risks. The economic task is to compare those costs with the selection problem the rule addresses.

Moral Hazard: Coverage Changes The Price Of Care

After someone has insurance, the price paid at the point of care may be much lower than the resource cost of the service. A visit that uses $200 of labor, equipment, and supplies might require only a $20 copayment from the patient. The other $180 does not disappear. It is paid through the insurance pool, taxes, or some other source.

When the price facing the patient falls, use tends to rise. That is the law of demand applied to medical care.

The welfare question is not whether use rises. It is whether the benefit of the added care exceeds the value of the resources used to provide it.

Suppose insurance makes a lifesaving medicine affordable. The added use may create gains far above its cost. Now suppose the lower price leads to a test that provides almost no useful information. That added use may cost more than it is worth. The graph alone cannot tell us which service is which.

Patients, Providers, And Payers

Medical decisions are rarely made by one fully informed buyer facing one visible price. The patient experiences symptoms and bears health consequences. The provider supplies advice and care. An insurer or public program may pay most of the bill. An employer may help choose the plan and contribute toward the premium.

Changing who writes the check does not make care free. The cost still appears through premiums, taxes, borrowing, or resources taken from other uses.

That separation creates information problems. The patient may not know the negotiated price, expected out-of-pocket bill, or likely quality. The insurer cannot directly observe every detail of medical need. The provider knows more about medicine than the patient but may not know the patient’s preferences, resources, or complete history.

The patient-provider relationship is therefore a principal-agent problem. Patients rely on clinicians partly because patients cannot diagnose and treat themselves. Professional ethics, reputation, licensing, second opinions, malpractice rules, and insurer review are attempts to support trust and limit mistakes or misuse. None is free or perfect.

Payment rules matter too. That does not mean doctors ignore patients or professional duties. It means that, as elsewhere in economics, rewards affect choices at the margin.¹²

Competition And Market Power In Medical Care

Not all medical services are bought under the same conditions.

A person with chest pain cannot spend three days comparing hospitals. Delay may be dangerous, and the patient may be unable to judge what treatment is needed. Emergency care is a poor setting for ordinary search and posted-price competition.

A scheduled imaging test or routine prescription is more shoppable. Patients may have time to compare locations, prices, and quality. Even then, comparison can fail if negotiated prices are hidden, insurance networks differ, or quality is hard to measure.

Local scarcity matters. A region with one hospital, one specialist group, or few insurers may have substantial market power. Consolidation may coordinate services or spread fixed costs, but it may also strengthen bargaining power and raise negotiated prices. A large building is not proof of monopoly, and a merger is not proof of efficiency. The mechanism and evidence matter.

Provider networks can give insurers bargaining leverage and direct patients toward lower-cost providers. They can also restrict choice. Licensing can protect patients from unqualified providers while limiting entry. Price transparency can help in services that are actually comparable and shoppable. But a posted price that few patients or insurers actually pay is not much of a price signal.

This is why the ordinary competitive benchmark remains useful but incomplete. It tells us to look for prices, entry barriers, substitutes, information, and incentives. It does not tell us to pretend those conditions are the same in an emergency room and a scheduled clinic.

Public Health, Externalities, And Shared Risk

Some health choices affect other people directly. Vaccination can reduce the chance of transmitting infection. Antibiotic use can contribute to resistance that makes future infections harder to treat. Disease surveillance and sanitation produce information or protection that many people can use.

These are applications of Chapters 10 and 11. Vaccination can create a positive externality. Excessive antibiotic use can create a negative externality. Disease surveillance can have public-good features because excluding people from its benefits may be difficult and one person’s use of the information need not reduce its value to others.

Private choices may therefore produce too little prevention or too much use of a shared medical resource. Education, subsidies, rules, property rights where feasible, and public provision are possible responses. As always, identifying a market failure begins the policy analysis; it does not finish it.

Every System Rations Scarce Care

Medical resources are scarce. There are only so many nurses, physicians, hospital beds, operating rooms, medicines, and tax dollars. No health-care system can provide every possible service immediately to every person at a price of zero.

Systems differ in how they decide who receives what. Some rely more on prices and private insurance. Some rely more on public insurance, budgets, waiting, or administrative coverage rules. Most use a mixture.

With little insurance, ability and willingness to pay do much of the rationing, and patients bear more financial risk. Private group insurance spreads risk but limits access through premiums, cost sharing, provider networks, and coverage rules. Public insurance spreads risk more broadly through taxes, but eligibility, covered services, payment rules, and public budgets still limit care. A tightly budgeted public system may charge patients little when care is used, but waiting, capacity, and clinical priorities become more important.

These examples show why arguments about whether care should be “rationed” miss the point. Scarcity guarantees rationing. Money prices ration. Waiting rations. Insurance networks ration. Budgets and coverage rules ration. A physician’s decision about which patient needs care first rations time.

No arrangement wins on every dimension. Broad coverage can improve financial protection while increasing taxes, spending, waiting, or administrative controls. Strong cost sharing can limit spending while exposing sick people to more risk. High provider payments can support capacity and innovation while raising premiums or taxes. Tight budgets can control spending while limiting access or weakening incentives to expand supply.

The comparison must be institutional, not imaginary. An ideal market with fully informed patients is not the alternative to an imperfect public program. An all-knowing public program is not the alternative to an imperfect market. The real choice is among arrangements run by people who face limited information, incentives, and scarcity.

Bringing The Chapter Together

Health economics is not one exception after another to ordinary economics. It is ordinary economics applied carefully to a setting with unusual uncertainty, information, and institutions.

Begin with the outcome. Health is produced by biology, behavior, environment, public health, medical care, and chance. Medical technology can change what is possible, but innovation must be financed and rewarded somehow.

Then identify the risk. Insurance pools uncertain losses and protects people from financial catastrophe. That protection is valuable, but it changes participation and the marginal price of care. Adverse selection concerns who joins before coverage. Moral hazard concerns how choices change after coverage.

Next identify the decision-maker. Patients, providers, employers, insurers, and public officials do not have the same information or face the same prices. Provider payment rules reward different margins. Emergency and scheduled care provide different opportunities for competition.

Finally, compare realistic institutions. Each solves some problems and creates others. The central question is not whether health care should involve markets or government. It already involves both. The question is which mix of rules, prices, insurance, professional trust, competition, and public action handles the trade-offs best.

Chapter Summary

1. Health is an outcome. Medical care is one input into health alongside biology, behavior, environment, public health, and chance.
2. Access, medical use, medical spending, financial protection, and health outcomes are different measures and need not move together.
3. The long-run obesity trend shows a large change in measured health risk but does not identify its causes.
4. New medicines can change the technology of health production. Their value depends on outcomes, harms, continued use, price, access, and alternatives.
5. The social value of medical innovation can exceed the profit captured by the innovator.
6. Patents can support research by creating a temporary reward, but they also create market power and do not reward every valuable project well.
7. Medical care differs from an ordinary spot market because illness, treatment success, need, price, and quality are uncertain, while medical knowledge is unevenly distributed.
8. Insurance pools uncertain losses. It makes individual financial exposure more predictable but does not eliminate the resources used for care.
9. Adverse selection occurs before or when coverage is chosen. People with higher expected costs may be more likely to join at a common premium.
10. Moral hazard occurs after protection changes incentives. A lower point-of-care price can increase medical use without implying fraud or bad character.
11. Cost sharing can reduce use and spending, but it may reduce valuable as well as low-value care.
12. Third-party payment separates the patient receiving care, the provider supplying it, and the insurer or public program paying much of the bill. Changing who writes the check does not eliminate the cost of care.
13. Fee-for-service rewards more billable services, while capitation rewards economizing, sometimes too much. No payment rule produces perfect incentives.
14. Competition works differently in emergencies and in scheduled, comparable services. Networks, licensing, consolidation, local scarcity, and whether posted prices are actually paid can affect competition.
15. Vaccination, antibiotic use, sanitation, and disease surveillance connect health care to externalities and public goods.
16. Every health-care system rations scarce resources. Systems differ in whether they rely more on prices, cost sharing, networks, waiting, budgets, or coverage rules.

Key Terms

• Health outcome: A measure of a person’s or population’s health, not a measure of medical care alone.
• Medical care: Services intended to prevent, diagnose, treat, manage, or rehabilitate illness or injury.
• Determinants of health: Biological, behavioral, environmental, social, public-health, medical, and chance influences on health.
• Medical innovation: A new treatment, device, process, or use of knowledge that changes what medical care can do.
• Medicare: Federal health insurance mainly for people age 65 and older and some younger people with qualifying disabilities or conditions.
• Medicaid: A joint federal-state health program for eligible people with low incomes and some people with disabilities.
• Metabolic syndrome: A cluster of risk factors associated with greater risk of diabetes, heart disease, stroke, and other health problems.
• GLP-1-based medicine: A treatment that acts on hormone signals involved in blood sugar and appetite.
• Social value: The total value an innovation creates for patients and others, whether or not the innovator receives it as revenue.
• Private return: The reward captured by the person or organization making an investment.
• Patent: A temporary legal right to exclude others from using an invention covered by the patent.
• Risk pool: A group whose uncertain losses are combined for insurance.
• Expected loss: The probability-weighted average loss before the outcome is known.
• Premium: The price paid for insurance coverage.
• Claim: A request for payment under an insurance contract.
• Risk aversion: A preference for avoiding large uncertain losses, even at some cost.
• Adverse selection: A participation problem caused by hidden information before or when a contract is chosen.
• Moral hazard: A change in behavior after protection changes the price or risk facing a decision-maker.
• Deductible: The amount paid by the patient before ordinary insurance payments begin.
• Copayment: A fixed patient payment for a covered service.
• Coinsurance: The share of a covered bill paid by the patient.
• Out-of-pocket maximum: A limit on covered patient payments during a period.
• Third-party payment: An arrangement in which a patient receives care, a provider supplies it, and another party pays much of the bill.
• Utilization: The use of medical services.
• Fee-for-service: Payment tied to each billable service.
• Capitation: A fixed payment per enrolled patient for a period.
• Provider network: A group of providers covered under specified terms by an insurer or program.
• Provider market power: The ability of a provider to affect prices or contract terms because patients or insurers have limited alternatives.

Figure Reading Guide

• Figure 19.1: distinguish a long-run prevalence trend from a claim about its cause.
• Figure 19.2: calculate expected claims and explain what pooling does not eliminate.
• Figure 19.3: separate the patient price from resource cost and avoid calling all additional use waste.

Review Questions

1. Why are health and medical care not the same thing?
2. Name six broad sources that can affect health.
3. Why should behavior be analyzed without treating it as a complete explanation or a statement of blame?
4. What does the U.S. obesity trend establish? What does it not establish?
5. Why is metabolic syndrome not simply another name for obesity?
6. How can chronic illness affect the medical system even when the direct health effect begins with one person?
7. How do new medicines change the technology of producing health?
8. Why can the social value of a medicine exceed the profit earned by its producer?
9. Distinguish Medicare from Medicaid.
10. State the main benefit and main cost of using patents to encourage medical research.
11. Why might research into a new use of an existing drug receive too little private funding?
12. According to Arrow’s framework, why does medical care rely heavily on trust and institutions?
13. What does insurance pooling accomplish? What does it not accomplish?
14. In the 100-person example, why might a premium above expected claims still be worthwhile?
15. Distinguish adverse selection from moral hazard by timing.
16. How can adverse selection raise the premium in a voluntary insurance pool?
17. Why does adverse selection not guarantee complete market collapse?
18. Why does the term moral hazard not imply immoral behavior?
19. What is the difference between the patient’s point-of-care price and the resource cost of care?
20. Why is additional care caused by insurance not automatically wasteful?
21. Define deductible, copayment, coinsurance, and out-of-pocket maximum.
22. What did the RAND Health Insurance Experiment show about cost sharing?
23. What is third-party payment?
24. Why is the patient-provider relationship a principal-agent problem?
25. Compare the main incentives created by fee-for-service and capitation.
26. Why does no provider-payment rule create perfect incentives?
27. Why is ordinary price competition more difficult in emergency care than in a scheduled service?
28. How can provider networks both help and harm competition?
29. Give one positive and one negative health-related externality.
30. Why does every health-care system ration care?
31. What does it mean to compare institutions rather than ideals?

Economic Reasoning Questions

1. In a pool of 200 people, two people are each expected to incur a $40,000 claim. What are total expected claims? What are expected claims per person? Why might the final premium be higher?
2. A plan’s premium is based on an average of low-risk and high-risk members. Explain what happens if many low-risk members leave. Identify the information problem and its timing.
3. A prescription costs $120 in resources. Insurance lowers the patient’s price from $120 to $10. Use marginal reasoning to explain when the added use creates gains and when it creates losses.
4. A plan raises its deductible. Trace the likely effects on financial protection, medical use, and spending. Why can health move differently from spending?
5. A clinic is paid for every visit. Identify one useful activity the rule rewards and one activity it may reward too strongly.
6. The same clinic instead receives one annual payment for every enrolled patient. Identify one useful activity this may reward and one new risk it creates.
7. A hospital posts a list price for a scheduled scan. Explain why that number may not equal the insurer’s price or the patient’s final payment.
8. Compare a person choosing an emergency department during chest pain with a person scheduling a routine scan. Which conditions for ordinary competition differ?
9. A new vaccine protects the vaccinated person and reduces transmission to others. Explain why the market quantity may be too low.
10. A government expands insurance coverage. List four outcomes that should be measured separately rather than summarized as whether the policy “worked.”
11. A patented treatment costs little to manufacture but required costly research. Explain why marginal-cost pricing and innovation incentives point in different directions.
12. A drug company cannot prevent competitors from using evidence that an old drug treats a new condition. Explain why a socially valuable study may not be financed.
13. A proposal promises unlimited medical care with no taxes, premiums, waiting, or restrictions. Use scarcity and opportunity cost to evaluate the promise.

Optional Research And Discussion Questions

1. Does paying donors reduce blood or plasma shortages? Distinguish whole blood used for transfusion from source plasma used to manufacture medicines. Compare at least one field experiment with cross-country evidence. Consider donor response, safety screening, collection capacity, regulation, and the possibility that rewards shift donations across time or place.¹³
2. Choose one scheduled medical service. Find, if available, a posted charge, a negotiated insurer price, and an estimated patient out-of-pocket price. Explain why the three numbers differ and what each does not tell you about quality.
3. Select one medical innovation. Separate its effect on health, medical use, spending, financial protection, producer profit, and social value. Do not assume those outcomes move together.
4. Compare two realistic ways to reduce low-value medical care. For each, identify the information required, the decision-maker whose incentive changes, and the risk of reducing high-value care.
5. Compare two health-care arrangements described in this chapter. Which handles catastrophic financial risk better? Which relies more on prices, waiting, networks, or budgets? What evidence would you need before recommending one change?

Source Notes