On the discontents of health systems: Part 1 Why are US healthcare costs so high?

I have been thinking for some time about writing various pieces on health economics, focusing particularly on comparisons of health systems in different countries. Health economics is an increasingly important specialism in economics, as befits a sector which accounts for at least 10% of GDP in most rich countries and is likely to continue to grow rapidly. When considering health policies, many people rely heavily on parody versions of how health systems work. In addition, all of us tend to rely on individual experiences and examples as we have neither the time nor interest to consider systematic differences between different ways of managing health care.

As background, I have been interested in and taught health economics over several decades. Much of this involvement was from the perspective of public economics – i.e. how should we decide how to finance and allocate public spending on health issues as well as the choices between, for example, spending on health, education and public infrastructure. From a personal perspective, many members of my extended family are or have been medical professionals or are employed in the wider health sector. We have direct experience of health systems in the UK, Europe and the US. In these articles I want to discuss systematic differences rather than the random aspects of individual experience.

Finally, I would like to emphasize one point. I think that the extent to which medical practitioners are allowed to dominate discussions of health policy in the media and politics is quite astonishing. As an illustration, the current UK government, shortly after it took office, commissioned a review of UK health policy that was headed and substantially written by a very prominent medical academic and practitioner. In parallel, it initiated a strategic defence review that was headed by a former politician who had been Secretary-General of NATO but who had never been a military officer. Naturally, medical professionals should contribute to such reviews, but discussions of health policy in developed countries have been captured to a substantial degree by producer interests. That is rarely good when major and difficult changes are required.

Many are aware of the high costs of healthcare in the US. Still, few – even in the US – think carefully about the factors that lead to high costs. There has been a strong reaction to the claim that many people will lose access to Medicaid health coverage because of provisions in the recent Budget Bill. As usual, the logic and what may happen are quickly lost in outrage and misrepresentation. Just as Obamacare turned out to have less of an impact on access to health insurance and services than advertised in advance, so too the effects of the new provisions will be influenced by many factors that we cannot predict now.

The primary responsibility for health policies in the US lies with states. The Federal Government, though tax and other financial arrangements, can encourage states to adopt various health insurance and other arrangements, but it is the states that decide. Healthcare providers operate within regulatory and financial frameworks that are set and enforced by state authorities. Calls for some kind of universal national healthcare system – such as a single-payer version of Medicare for all or the health insurance offered to federal employees – are little more than fantasy. The Supreme Court has repeatedly thrown out Federal legislative requirements mandating state action when these are seen as trespassing on the constitutional autonomy of states in making policy decisions within their jurisdictions.

Notwithstanding the profligacy of a few states like California, most US states operate under tight fiscal constraints. Many have strict balanced budget requirements and the range of tax instruments which they can use to collect substantial revenue is limited. All states levy payroll taxes on employers to fund state unemployment insurance and a few levy payroll taxes on employers or employees to fund disability payment and training. The costs of Medicare are covered by a uniform Federal payroll tax of 2.9% nominally split between employers and employees.

To finance single-payer health insurance at state level would involve replacing existing employer-provided health insurance schemes. This would be extremely controversial both on grounds of cost and the impact on employers with a presence in multiple states. Under Obamacare - the Affordable Care Act (ACA) – employers with more than 50+ employees are effectively required to pay at least 50% of a minimum health insurance plan for their employees, but not necessarily their dependents. This leaves a huge number of people out of the health insurance safety net. It inhibits labour mobility, and it means that employees who lose their jobs may find themselves in dire financial circumstances.

The reliance on employer-provided health insurance is strongly encouraged by the provision of the federal income tax which allows the cost of such insurance, including employee contributions in most cases, to be excluded from pre-tax income. However, this is only part of the story; an equally strong incentive arises from the absurd complexity of the US tax system. The costs of health insurance purchased by the self-employed and those without access to employer-provided schemes can be deducted under many circumstances, provided – and this is crucial – the individual itemises their deductions. Since itemisation is cumbersome and is more likely to lead to tax audits, it is heavily discouraged.

Overall, current incentives are most valuable for well-paid employees in stable employment. It is hardly surprising that the healthcare that is supported by employer health insurance tends to be expensive and focused on meeting the expectations of better-off and long-term employees. Employers such as the federal, state and local governments compete by using their scale to match the insurance offered by large private companies. Similarly, private universities form consortia to take advantage of the links between training institutions and hospitals, clinics, and other medical providers. The federal government runs a network of hospitals and medical facilities for veterans through the Veterans Administration (VA).

In 2023 about 60% of the US population under the age of 65 were covered by employer-sponsored health insurance, which is required to meet minimum standards of coverage. In addition, most of those aged 65+ are eligible for one form or other of Medicare, accounting for 18% of the US population. While not everyone will be happy with the costs and coverage of their health insurance, between 60% and 70% of the US population have some reasonable level of access to insurance-based healthcare. Equally important, the population with employer-sponsored health insurance is heavily skewed to better-paid workers, who are likely to have higher consumption of healthcare services.

This drives up the costs of health insurance, so the average total premium in 2022 for employer-sponsored health insurance was about $7,500 per person covered. The median level of annual earnings was $48,000 for all workers and $60,000 for full-time workers. It is hardly surprising that most workers without employer-sponsored health insurance cannot afford to pay for equivalent insurance out of post-tax pay.

From the perspective of political economy, this is a stable arrangement. Any attempt to move toward universal coverage can be vetoed by existing insured voters if this would involve either a reduction in the services that they receive or an increase in their out-of-pocket healthcare costs. To fund universal coverage at current levels of service and costs would cost roughly $500-$600 billion per year, based on the distribution of the working age population who are uninsured some or all of the time plus those who are under-insured.

While this sum is “only” about 2% of GDP, it would require more than doubling the current payroll tax used to fund Medicare. As the debates around Obamacare demonstrated, there is no political will to incur such a large spending commitment, nor is it possible to impose the cost of extending mandatory health insurance coverage on small businesses and the self-employed.

The current system does include hidden transfers that support the provision of healthcare for uninsured population, but arguably these are perverse and certainly they are inefficient. All hospitals have an obligation to provide emergency medical care to people who are taken to or attend an Emergency Room (A&E in UK parlance) requiring assistance. The costs of providing such assistance are included in the general charges levied by hospitals. However, the unintended consequences of this policy are that (a) hospitals may try to close their Emergency Departments, and (b) uninsured and underinsured customers pay a disproportionate share of such costs because insurers negotiate fee discounts that, in effect, exclude such costs.

Most people coming from countries that provide universal healthcare coverage find the US system both inefficient and morally offensive. Many Americans would agree, but such judgements do not alter the large barriers to change, while many of the assumptions that underpin arguments against the US system are either wrong or muddled. There is a widespread belief among advocates of change that healthcare services should be provided exclusively by the public sector or that, at least, it should not be provided by for-profit entities.

What such claims mean is rarely thought through. Should all pharmaceuticals be developed and supplied by public enterprises? What about IT equipment and services? Or scanners and x-ray machines? In the UK all kinds of outsourcing and supply contracts are subject to complaints about the “privatisation” of health services.

Such arguments invariably raise large and highly contentious issues in identifying and analysing evidence. Many of the best regarded hospitals and clinics in the US are run by non-profit entities, not only universities but a whole range of foundations, trusts and other arrangements, often as an evolution of what were religious establishments. In many cities, private non-profit institutions compete with for-profit hospitals and clinics, while elsewhere they may collaborate with each other in wider networks.

It is unclear what the harm is supposed to be from relying on private clinics that provide specialist services from scanning and radiotherapy to day surgery. Indeed, is a medical partnership, whether providing specialist or general practitioner services, anything other than an organisation operating “for-profit”? The distinctions between partnerships, cooperatives, non-profit foundations, and for-profit entities with owners or shareholders are often blurred or invisible.

As medical practice and technologies have become more complex and capital-intensive, it is necessary to address the issue of how investments in buildings, laboratories, medical and IT equipment as well as training staff are to be financed. One vision – the Canadian or UK ideal – is that everything should be run as a public monopoly with capital provided or underwritten out of public funds. There is no doubt that this model can work if demands are constrained and public funds are plentiful, but such favourable conditions rarely last long.

Ultimately, the public monopoly model rests not only on controlling competing suppliers of healthcare services but, more important, on the use of the monopsony - or sole buyer – power to set wages for medical staff and payments to providers of medical services. That buying power comes with a cost: the public sector may control prices, but it cannot control quantities. Inevitably, monopoly systems are forced into capacity and manpower planning, but they are usually subject to severe constraints on capital spending as current needs have priority over future needs.

In contrast with systems dominated by public monopoly provision of healthcare services, the US system encourages competition. In most of the country, health insurers have a choice as to which healthcare providers they sign up to offer services to their customers. In practice, this means that large insurers can negotiate discounts off the charges that would be charged to customers covered by smaller insurers or without insurance. Critics of for-profit healthcare appear to believe that shareholders capture most of the benefit of such discounts. That is far from obvious, because there is competition between insurers, while non-profit insurers and healthcare providers can compete with for-profit operators.

As in any other sector, the key issue is whether either the structure of the healthcare sector or the way in which it is regulated and financed gives rise to monopoly power which enables the providers of services to earn what economists call monopoly rents. The answer in the US case is that there are large monopoly rents arising in parts of the sector. One source of such rents is well known – the patents on new drugs which allow pharmaceutical companies to earn very large, though temporary, profits on such drugs. However, that case is complicated by the way in which regulation and other parts of the system have adapted to pharmaceutical profits.

What is less well understood – and rarely discussed by the critics of for-profit healthcare – is that the primary direct beneficiaries of monopoly rents in the US are doctors and other medical staff in professions for which entry is strictly controlled. For decades, the American Medical Association (AMA), other similar organisations and their state equivalents have rigorously opposed what they consider to be “excessive” expansion of medical training. In addition, there are state requirements whose (perhaps intended) effect is to inhibit the mobility of medical and other staff between states and from outside the US. The whole structure of medical training and qualifications is designed to restrict the supply of qualified personnel. The combination of competition in demand and monopoly control over supply invariably leads to rents being earned by suppliers.

To be clear, this does not mean that individual doctors all earn a share of these rents. Systems adapt. Universities as the gatekeepers for professional qualifications capture a significant share of rents by pushing up tuition fees, extending the length of training programs, and paying trainee doctors poorly during their compulsory training periods. Lawyers and insurance companies get a share of the rents by pushing up claims for medical negligence and the costs of indemnity cover.

As an illustration, I had a graduate student whose wife was an ob-gyn specialist. The cost of her indemnity insurance was greater than her net income. The consequence was that many trainee doctors were reluctant to go into the specialism, which further reduces the supply of ob-gyn specialists and pushes up what they could charge. In turn, this fuelled suspicion that doctors were exploiting their patients, which encouraged even more negligence claims. This was a vicious circle that was almost impossible to break.

The consequences of competition between health insurers and service providers extends beyond doctors’ pay. Many hospitals and clinics choose to invest in their facilities and equipment to attract both insurers and patients. Academics in many countries know that universities often prefer not to compete on price, because they think that fee levels are treated as a signal of quality. Instead, they focus on student accommodation, support services and non-academic facilities. So, it is with many US hospitals which tend, for example, to have a higher ratio of scanners and similar equipment per patient than equivalent hospitals in other countries.

None of this can easily be changed. There are many reasons to argue for the provision of wider or universal access to healthcare in the US. However, a rapid transition would provoke strong resistance because of the potential impact on those who already enjoy what they regard as good but essential healthcare. Boosting demand for healthcare by 20% or 40% without increasing the supply of doctors as well as the capacity of hospitals and ancillary services would push up costs and certainly cause considerable political upset. Lobby groups play on such concerns but are reluctant to advocate immediate measures to increase supply, as that would threaten their market power.

There is an aphorism in macroeconomics which says that politicians know what the correct policies to adopt are (e.g. take away the cookie jar) but they don’t know how to get elected after implementing them. In dealing with healthcare, politicians are frequently urged to spend more money by people who have no responsibility for raising the money required. The same people often resist the changes in supply arrangements necessary to bring down the costs of providing healthcare services.

This is the central problem of healthcare in the US. For as long as costs are so high it is almost impossible to find ways of moving towards universal provision. Indeed, any steps in that direction are likely to exacerbate the cost crunch and provoke a hostile reaction from those who enjoy reasonable access and decent services. Even more, attempts to regulate costs or exercise monopsony buying power are fiercely resisted by those who might thereby lose.

Something as (apparently) simple as reforms to tort laws to replace lawsuits for medical negligence by compensation schemes like arrangements in New Zealand have repeatedly failed because of opposition from lawyers who might thereby lose. It is clear to almost everyone that a blame-based system of addressing the consequences of medical accidents and mistakes is empirically the least effective and most expensive approach that can be adopted, but that is what we have in most developed countries.

The answer to the high costs and incomplete coverage of healthcare in the US cannot be just to throw more money at the existing system. The policymakers who advocated the changes that ultimately became Obamacare – and the attempted Clinton reforms before that – knew that. They were trapped between a widespread resistance to any changes that might be presented as jeopardising the services that most of the population already receive, and the hostility of medical associations and other interest groups to changes that might reduce their incomes or profits. Of course, the opposition was not presented in such naked terms, but it is easy to convince oneself that wider principles are at stake when one’s income is threatened.

While interest groups such as the AMA, lawyers and pharmaceutical companies are easy targets, it is equally hard to excuse the behaviour of many academics and other vehement opponents of either for-profit or private sector involvement in the supply of healthcare services. The US public sector is not known for its efficiency, competence, or innovation. Specifically, hospitals and clinics run by the VA are neither less costly nor more efficient than comparable non-profit or for-profit institutions.

A case can be made for a public monopoly in the provision of healthcare services on grounds of equity and universal provision. Still, there is no example in the world of such a monopoly being more efficient, innovative and generally cost-effective than comparable systems that allow competition. The cost advantages of public monopoly arise from (a) squeezing the incomes paid to employees and suppliers, and (b) rationing access to services. Users may accept such trade-offs but to pretend that they do not exist in the long run is just dishonest.

In summary, the US healthcare system is caught in a high-cost trap. It works well for a substantial portion of the population who enjoy employer-sponsored health insurance and the elderly whose health costs are substantially covered by Medicare. To break out of the high-cost trap would require a large increase in the supply of healthcare services – many more doctors, support staff, hospitals, clinics, etc – which would take a decade or more to achieve. The changes required could not be implemented unless the Federal Government cooperates consistently with the state governments which have direct responsibilities for financing and providing healthcare services.

Such changes would likely be painful and strongly resisted by many interest groups. The pay-off is unlikely to be quick enough to attract the level of support necessary to overcome the obstacles. On the other hand, just injecting more money into the existing system will worsen supply constraints and increase monopoly rents.

In these circumstances, it is hardly surprising that healthcare reform is seen by many policymakers as something close to the third rail of US domestic policy. Yes, it would be nice to reduce US healthcare costs, but many who have tried to introduce major changes have been badly burned by the experience. Attempts to save money by incremental reforms that focus on efficiency and eligibility requirements are equally strongly resisted.

The sad result is a behemoth that can neither change direction nor even mend its own ills. One of the Founding Fathers of economics, Adam Smith, commented that “there is a good deal of ruin in a nation”. Only a country as rich as the US can tolerate a healthcare system that is as inequitable and inefficient as the US system. Still, we should not forget the all too large number of victims of the failure to find a route to reform the system.