Part B: Why US Health Care Costs So Much — and Delivers So Little
An Introduction to Bioethics.
In Part A we built four lenses for asking what a just society owes people: maximize good (utilitarian), protect the worst-off (Rawls), respect free choices (libertarian), secure what people need to function (capabilities). We argued, with Daniels, about whether health care is a right at all.
That was the theory. Part B is the machine those values have to deal with — the real US health care system. Before we can argue about how to fix it, we have to see clearly how it actually works, what it costs, and what it delivers.
Before the numbers appear, write down your best guess.
The United States spends far more on health care than any nation on Earth — and its people are not healthier for it.
| United States | OECD average | |
|---|---|---|
| Spending per person (2024) | ~$14,900 | ~$6,000–7,000 |
| Share of the whole economy | 17.6% of GDP | ~9–11% |
| Life expectancy (2024) | 79.0 years | ~81.5 years |
| Maternal deaths / 100,000 (2023) | ~18.6 | ~10 |
We spend roughly double — and live about two and a half years less (Centers for Medicare and Medicaid Services 2024; Organisation for Economic Co-operation and Development 2025).
One number moved the right way. US life expectancy rose to 79.0 years in 2024 — an all-time high, up from 78.4 in 2023 — and COVID-19 dropped out of the ten leading causes of death for the first time since 2019 (National Center for Health Statistics 2026). The single largest contributor was the steep fall in drug overdose deaths (see Chapter 5A). The gap with peer nations narrowed, but did not close: the US still spends about twice as much to end up two-plus years behind.
This gap is not a small inefficiency — it is the central puzzle of American health policy, and where Part A stops being abstract: a country that treats care as mostly a market good has built a system with extraordinary capabilities and extraordinary gaps. Everything in this lecture answers one question: why?
By the end of Part B, you will be able to:
Before we look at the US, we need a map. Every health system on earth is built from the answers to just two questions:
The journalist T. R. Reid sorted the world’s systems into four models by those two answers (Reid 2009). Almost every country picks one. Keep that word — one — in mind.
| Model | Who delivers | Who pays | Where you’ll find it |
|---|---|---|---|
| Beveridge | government owns it | government (taxes) | UK (NHS), Spain, Cuba |
| National Health Insurance | private providers | one government insurer | Canada, Taiwan |
| Bismarck | private providers | regulated nonprofit funds (payroll) | Germany, France, Japan |
| Out-of-pocket | whoever you can find | you, if you can pay | most low-income countries |
The deepest split is who pays, running from “the public, for everyone” (Beveridge, NHI) to “private but tightly regulated” (Bismarck) to “you’re on your own” (out-of-pocket). These are idealizations — real countries blend at the edges — but each builds its system around one model and makes it coherent. Hold that thought: the United States did something no one else did.
Beveridge (UK’s NHS, since 1948)
The government owns the hospitals and employs the doctors, and funds it all through taxes. Care is free at the point of use. This is what “socialized medicine” actually means — the state runs the whole thing.
National Health Insurance (Canada, Taiwan)
Doctors and hospitals stay private, but a single government insurer pays them, funded by taxes. One payer, private delivery. No marketing, no denials for profit, strong bargaining power on prices.
Bismarck (Germany since 1883, France, Japan)
Care and insurance are private — but insurers are nonprofit “sickness funds,” everyone must be covered, and prices are tightly regulated. Switzerland’s mandate, and the US ACA, are cousins of this model.
Out-of-pocket
No system at all: you get the care you can pay for, and go without the rest. This is the reality for most people in poor countries — and, as we’ll see, for the uninsured in the US.
The US has no single health system. It has a patchwork — and the easiest way to see it is to follow one person through a lifetime.
Meet Maria.
One person. Six different programs — and, as we’re about to see, three different countries’ models of health care, all without leaving the US.
US health coverage in 2023 (Keisler-Starkey et al. 2024):
| Source of coverage | Share of population |
|---|---|
| Employer-sponsored | ~54% |
| Medicare (65+ and disabled) | ~19% |
| Medicaid (low-income) | ~19% |
| Direct purchase / ACA exchanges | ~10% |
| VA / military | a few % |
| Uninsured | ~8% (≈26 million) |
Shares add to more than 100% because many people hold more than one type in a year. About 92.5% of Americans had some coverage — a near-record high, and still leaving 26 million with none.
That high-water mark has just receded. The enhanced ACA premium tax credits — the pandemic-era subsidies that drove coverage to its record level — expired on December 31, 2025. Congress did not renew them.
The effects landed immediately:
So the “near record high” above describes a peak the country has already come down from. Treat the coverage numbers in this chapter as the 2024–25 high-water mark, and expect the 2026 figures to be meaningfully worse.
Write or discuss: Part A argued about whether health care is a right or a market good. Notice that nothing about American law or medicine changed here — a subsidy lapsed, and millions lost coverage. What does it say about a system that the practical answer to “do you have a right to care?” turns on whether a tax credit is renewed?
This is the heart of “fragmentation.” Other countries pick one model. The US runs all four simultaneously — and which one you live under depends on your age, job, income, and luck (Reid 2009).
| If you are… | Your coverage | …is really this model | …like which country |
|---|---|---|---|
| A veteran (VA) | government hospitals | Beveridge | the UK |
| 65 or older | Medicare | National Health Insurance | Canada |
| Working with benefits | employer insurance | Bismarck | Germany |
| Uninsured | pay or go without | Out-of-pocket | the developing world |
A single country containing Britain, Canada, Germany, and the developing world — sorted by who you happen to be. And running four systems at once isn’t only untidy — it is expensive: every provider bills all of them at once (the paperwork tax we’ll see shortly), people fall between models into the gaps at the seams where the 26 million uninsured live, and no single payer dominates enough to negotiate prices the way Canada or the UK can.
About half of Americans get coverage through a job — the single largest piece of the system, and the one most people never chose.
Medicare — federal, for people 65+ and the long-term disabled.
Nearly universal for the old — the closest thing the US has to a right to care.
Medicaid — joint federal–state, for low-income people.
One corner of the US system looks like the rest of the wealthy world.
For people without a job-based plan, Medicare, or Medicaid, the ACA marketplaces (since 2014) sell individual plans, with income-based subsidies.
America’s job-based system was not designed. It was an accident of World War II.
That 1954 tax break is still here — and it is enormous.
Two big expansions (1965, 2010), two big failures (1948, 1994), and one unfunded promise — EMTALA, which since 1986 has required hospitals to screen and stabilize anyone in an emergency, regardless of ability to pay (United States Congress 1986). Eighty years of fights, and the core never moved: coverage is still built around the employer, and the US still has no universal guarantee (United States Congress 2010). Every reform added a patch rather than replacing the quilt — leaving the fragmented system we just mapped, and the costs we turn to now.
The intuitive guess is that Americans must overuse care — too many tests, too many visits. The evidence says the opposite.
If prices are the engine, what feeds them?
Some of the money never buys care at all — it buys billing.
Administrative complexity. With hundreds of insurers, each with its own rules, codes, and denials, the US spends enormously just managing the money.
Part of this is the price of running four models at once: every provider must bill Medicare, Medicaid, dozens of private insurers, and self-pay patients, each with its own rules. Canada has one payer to bill. We have thousands.
Underneath the prices is an incentive problem in how we pay.
Fee-for-service
Providers are paid per thing done — per test, scan, procedure. The built-in incentive is to do more, whether or not it helps. Volume is rewarded; restraint is not.
Value-based care
The alternative pays for keeping people healthy — a fixed amount per patient, or bonuses for good outcomes. It’s spreading slowly, but fee-for-service still dominates US medicine.
Note
“Let the market set prices” is a libertarian instinct from Part A — free transactions, minimal interference. It does drive innovation and rewards investment.
But health care breaks the assumptions a market needs: patients can’t shop while having a heart attack, can’t see prices, and can’t walk away. The utilitarian asks whether all this spending actually maximizes health (it doesn’t), and Daniels asks whether prices this high are compatible with fair access (they aren’t). Same facts, three verdicts.
You have a job and a health insurance plan. Your appendix bursts; you get emergency surgery and go home in two days, cured. Weeks later, a bill arrives: after everything your insurance paid, you owe $15,000 — a deductible, “out-of-network” charges from a surgeon you never chose, and coinsurance.
This is ordinary, not rare.
Discuss: A law was needed to stop hospitals from billing insured patients for doctors those patients never chose. What does needing that law tell you about the system?
Note
A distinction the coverage numbers hide:
Tens of millions are underinsured. “92.5% covered” can hide people who skip prescriptions, avoid the ER, or go bankrupt anyway. Health economists call the bankrupting effect financial toxicity — when the treatment for the bill becomes its own illness.
Part of the outcomes gap is straightforward: people who can’t afford care get less of it, later.
Here honesty matters: a large share of the US disadvantage is not something hospitals can fix.
The finding that surprised everyone.
A landmark 2013 National Academies report, U.S. Health in International Perspective: Shorter Lives, Poorer Health, found the US disadvantage runs across the whole life course — and is not explained only by the poor or by minorities (National Research Council and Institute of Medicine 2013).
Even white, college-educated, insured, high-income Americans were found to be in worse health than similar people in peer nations. The disadvantage is national, not just a problem of the margins.
The national average also hides sharp gaps inside the US.
These are the population-level injustices Part A’s theories were built to name. They are not random — they track access, history, and bias.
Genuine strengths
Where it fails
The strain shows up most sharply in the people who deliver the care — especially nurses.
Note
A shortage of nurses is not just an HR headache — it is a justice and safety issue.
The system’s failures land on patients through the workforce. Who staffs the bedside is a distributive-justice question too.
The US drifted into its patchwork by accident. It doesn’t have to stay there — and one country proves it.
In the early 1990s Taiwan had no universal system. Rather than guess, its government studied health systems around the world, then in 1995 deliberately built a National Health Insurance model — single public insurer, private delivery. It reached near-universal coverage in a few years, at a fraction of US cost.
The lesson
A country can look at the menu and choose. The US question is not “is reform possible?” — Taiwan, Canada, Germany, and Britain all answer yes. It is which model, and which value that choice serves.
Each reform path advances a different Part A value — and sacrifices another.
| Reform | Model | Serves… | Strains… |
|---|---|---|---|
| Medicare for All | NHI | equal access (Rawls, Daniels) | liberty; existing plans |
| Public option | NHI + Bismarck | a middle path | satisfies neither side fully |
| Build out the ACA | Bismarck | coverage and choice | complexity; cost control |
| Market reforms | toward out-of-pocket | liberty (libertarian) | the worst-off |
There is no value-free choice. To pick a model is to pick which of Part A’s principles you put first.
“Medicare for All” would cover every American through a single public insurer, ending most private plans and the link between jobs and coverage.
We started with a paradox: the US spends about double any peer nation, near $14,570 per person, and gets shorter lives in return. We saw why — a patchwork that runs all four models of health care at once (Beveridge for veterans, NHI for the old, Bismarck for workers, out-of-pocket for the uninsured), built by a wartime accident, with prices no one can see and an outcomes gap driven as much by guns, cars, food, and poverty as by the clinic. We named what the US does best (the world’s finest specialty and cancer care) and where it fails (access, cost, and a nursing workforce in crisis).
None of this settles the argument — it equips it. Other countries chose one model on purpose; the US can too. Which one depends on which of Part A’s values we put first: the most total health, the protection of the worst-off, the freedom to choose, or the capabilities every person needs. The model is yours to argue for.
Recall
Explain “it’s the prices, not the utilization,” and give two specific reasons US prices run higher than peer countries’.
Apply
A friend says, “Americans are unhealthy because they overuse the ER and eat badly — fixing insurance won’t help.” Using this lecture, where is your friend partly right, and where wrong?
Debate
Should the US adopt a single-payer system, a public option, or market reforms? Defend your choice using a specific theory of justice from Part A.
Work through the paired case study on Singapore’s health care system — a country that covers everyone and lives longest while spending a quarter of what the US spends. It asks what the US can, and can’t, learn from a “fifth model,” and why Singapore’s success is harder to copy than it looks.
Bioethics OER · CC BY-NC 4.0 · Brendan Shea