Higher prices are increasingly recognized as a significant cause of the outlier status of the United States in health care expenditures. At the same time, various explanations are often invoked to justify higher prices as rational or even defensible. We evaluate—and mostly counter—potential explanations of why health care prices are higher in the United States: upper-tail income inequality explains higher physician incomes; physicians need to recoup higher training costs; American patients are perceived to have different preferences, while providers face higher medical malpractice and administrative costs; health care purchasing occurs in a fragmented marketplace; and rent seeking rewards providers with favorable prices at the expense of consumers. Of these explanations, rent seeking is compelling partly because it is more consistent than other explanations in explaining higher prices across all sectors of the health care system. We also discuss why administrative costs are gaining recognition as an important factor; however, the understanding of their contribution and the knowledge of solutions is evolving, rather than fully developed. Policy solutions to address rent seeking are challenging, because they threaten provider income. Most solutions, such as price transparency, are often touted as a magic bullet, but these are likely to be effective only in combination with other solutions.
The United States spends more than any other country on health care—this is almost universally acknowledged. Until relatively recently, however, many assumed that the higher spending was due to greater utilization—that Americans simply visit doctors more often, undergo more tests, and have more procedures. Research over the last two decades has put that misconception to rest. When it comes to America's outlier status in health care spending, utilization is not the culprit: numbers of US doctor visits are comparable to those of other countries, and hospital stays are neither more frequent nor longer. The root of the American health care problem does not lie in the quantity of health care consumed, but rather in its price. The biggest problem with the US fee-for-service system, says Joseph White (2011: 776), is “the specific fees per service.” And as one team of researchers argued, “It's the prices, stupid” (Anderson et al. 2003).
This article reviews the evidence on US health care costs and prices. We first review the evidence showing that US health care spending is much higher than spending in other wealthy countries and the key role of high prices in increasing US expenditures. We then analyze seven explanations for the higher US health care prices, ranging from the expense of US medical schools to the burden of administrative costs.
Several pieces of the puzzle are beyond the scope of this article. First, we only briefly address policies that may lower prices. In the first instance, understanding prices is critical before policy support could be mobilized to change US health care, both because causal factors need to be understood in depth, and because the political support for new policies might require a greater level of consensus over the reasons. The important question of whether policy changes might both lower US health care prices and be politically viable is a vexing one that this article can only partly address.
Second, we ignore the role of the uninsured in increasing prices. If some people are uninsured, the thinking goes, prices need to be higher elsewhere, because US health care providers bear higher costs than do health care providers in other countries. However, the argument that providers shift the costs of the uninsured is not borne out by the evidence. Estimates of the uncompensated care generated by the uninsured (Garthwaite et al. 2018) are dwarfed by the gap in spending between the United States and its counterparts. The uninsured are surely a problem in the United States, but the costs they impose on providers are not large enough to explain away high US health care prices.
Third, we focus on the price of health care across countries but ignore the price of health insurance. This is a deliberate oversight, in that the price of health insurance is difficult to compare across countries. In many places, the price of health insurance is bundled into the population's tax burden. Prices of health insurance are likely obscured by cross-subsidization and government budgeting conventions. As a result, the literature has, by and large, focused on the price of care. We follow that norm.
Finally, we do not examine the role of the social determinants of health. There is no question that, to take one metric, the United States has much higher infant mortality than other wealthy countries (Chen et al. 2016). That dismal comparison applies to many other metrics of health and to factors that determine health (see Muennig, Reynolds, Jiao, and Pabayo, this issue). Yet, it is unclear how social determinants of health would influence price, even if they might lead to greater utilization of services.
The Basic Facts
To begin with, US health care spending far exceeds that of any other wealthy country. In 2016, 17.2 percent of America's gross domestic product (GDP) went to health care, versus 12.4 percent in Switzerland, and 10.6 percent in Canada (OECD 2018). Generally speaking, health care spending is equal to the price of health care times the quantity of health care consumed. And the outlier status of overall US health care spending is very much not driven by the quantity of health care provided. Utilization of nearly all types of health care is lower or similar to that of other countries, and in some cases, it is lower than the Organisation for Economic Co-operation and Development (OECD) mean. For instance, in 2010 the United States had 114 hospital admissions per 1,000 residents while the OECD mean was 161. In fact, the US rate of hospital stays is lower than that of the United Kingdom: 136 hospital admissions per 1,000 people. The mean number of doctor visits per person in the United States was 4.00 in 2011 compared with 7.18 in OECD countries (OECD 2018). Although there are some exceptions, such as utilization of MRI (magnetic resonance imaging), the statement that the gap in spending is driven by prices, not quantities, is defensible.
To begin with, we should consider the example of the prices of a hospital stay. Prices from international private health plan data show higher prices in the United States. The average price for hospital care in the United States is $5,220 per day versus only $765 in Australia. Notably, the range in the United States was substantial, from $1,494 at the 25th percentile to $17,358 at the 95th percentile (International Federation of Health Plans 2015).
These prices are consistent even when comparing relatively routine surgical procedures. For example, the price of a hospital stay and physician fees for an appendectomy in the United States is $15,930, with the next highest price (United Kingdom) almost half as much, at $8,009 (International Federation of Health Plans 2015). Americans pay higher prices not just for hospitalizations but also for physician services, and those higher prices then lead to higher physician earnings. Peterson and Burton (2007) estimate that hip replacements command sums twice as high in the United States as in Canada (about $13,000 in the United States, $6,700 in Canada). A comparison of physician fees for hip replacements in six nations found that Medicare pays far more for the procedure than do the public plans in comparison countries (Laugesen and Glied 2011). The second-highest payer, Australia, paid $1,000; US Medicare paid $1,600. US spending on this procedure is much higher than abroad not because Americans consume higher hip replacements per capita—that is not the case—but solely due to the price of those hip replacements.
A similar pattern applies to pharmaceuticals—Americans pay more for branded drugs than residents of any other country. (Interestingly, generic drugs are an exception: they are cheaper in the United States.) At the same time, there is one important difference with regard to pharmaceuticals: while it is very clear that prices are higher, utilization is also higher. Adults in the United States take more prescription drugs per capita than adults in comparator countries (Squires and Anderson 2015). To take one, albeit extreme, example, a drug used to treat chronic myeloid leukemia, nilotinib, was estimated to cost $115,000 per year in 2012 in the United States compared to $48,000 in Canada (Experts in Chronic Myeloid Leukemia 2013).
Considerations in Cross-Country Comparisons
The task of comparing prices across countries is complicated by a lack of accurate pricing data, geographic variations within countries, and cost-of-living variations across countries. In the United States, physician service fees in the private sector have, until recently, been inaccessible to researchers. Several organizations such as the Health Care Costs Institute are only now making data on prices publicly available. Sometimes prices must be inferred from complex contracts and payment arrangements. Different payment arrangements used in the United States, such as capitation and pharmaceutical rebate programs, make it difficult to calculate prices. (One source of private sector price data for countries other than the United States is the International Federation of Health Care Plans.)
Another challenge when making cross-national comparisons involves the difference between public and private payers. Journalists will sometimes compare the price of a service abroad, in the public sector, to the price charged by private American providers. One is better off comparing public to public and private to private. This is particularly true due to the murkiness of public-sector prices in tax-based systems. In the British National Health Service (NHS), for instance, prices are public. However, as a fee-for-service system, the marginal cost of providing a service is difficult to estimate, given that sometimes the fixed costs of care are funded through other budgets or global budgets. The NHS may pay a lower price than Medicare or private US insurers, but other NHS budget allocations may compensate for those lower prices.
Researchers must also grapple with differing product mix or differences in the site of care, mode of care, or services that are bundled. There may be unobserved differences in intensity of care. For example, perhaps the United States has a more intensive style of treatment, which would explain higher prices. Some services are performed inpatient in some countries and outpatient in others. To address the intensity issue, researchers try to compare services that are relatively similar across countries, such as hip replacements and office visits (Laugesen and Glied 2011). This may be even more of a concern when comparing prices for services in which pharmaceuticals are key inputs. For example, if US patients are more likely to get new, branded drugs while hospitalized, then the price of a hospitalization will likely be higher in the United States. The difference, however, would derive from the vintage and presumed quality of the products in question (Danzon and Furukawa 2008), not characteristics of the market.
Lastly, accurate price comparisons depend on adjustments of prices for purchasing power parity (PPP), which convert local prices into an index based on the cost of a basket of goods. On the one hand, this adjustment is well understood and standardized, making it a relatively minor concern. On the other hand, adjusting for PPP can introduce errors, because the quality of that basket may vary across countries, prices may vary within each country, and it is not clear a priori what basket of goods to use.
All in all, none of these challenges is insurmountable. Instead of making research on prices impossible, they force researchers to be careful. One cannot simply compare the raw price of a medical procedure in one country to the price in another country. Instead, one has to adjust the prices by PPP, take care that the services are indeed comparable, look closely to ensure that the relevant provider contracts do not complicate the comparison, and so on. However, at the end of the day, researchers who do their due diligence nearly always find that prices are much higher in the United States.
Causes of Higher US Health Care Prices
There are many potential explanations for higher health care prices in the United States. We review—and counter—some commonly offered explanations here. Many of these explanations are easy to dismiss, but we conclude that several—those related to administrative costs and rent seeking, in particular—are central to any explanation for high United States health care prices.
Explanation 1: Upper-Tail Income Inequality
Laugesen and Glied (2011) suggested one explanation might be that higher American health care prices simply reflect that inequality is much higher in the United States than in other wealthy countries. In particular, the top few percentiles of the American wage distribution entail wages much higher than in most other countries with less income inequality. Most physicians—not to mention hospital and insurance executives—would earn very high salaries in other industries, salaries made higher by the growth in income inequality. Therefore, one can recruit those talented workers into health care only if they are paid wages that approximate salaries that could be received in other industries. In other words, if physicians mostly come from the ranks of a country's most talented, a health care system hoping to draw such talent would have to offer salaries in the highest percentiles of the income distribution, and in the United States, those highest percentiles involve ever-rising salaries.
On its face, such a hypothesis has its merits. Twenty-seven percent of physicians fall into the top 1 percent of income earners in the United States (Dewan and Gebeloff 2012). Students completing premed undergraduate degrees are among the most talented college students: the mean grade point average for individuals beginning medical school is 3.77 (AAMC 2016b). It follows that such professionals need to be paid what their neighbors in finance and corporate America are paid.
Yet, this explanation takes us only so far. The income-inequality hypothesis may imply that American physicians and health care executives must earn higher salaries, but at least for physicians, it does not explain why inequality is higher among American physicians. In the United States, specialists earn considerably more than family practitioners, psychiatrists, and pediatricians, but the same is less true in other countries. Primary care physicians have incomes more similar to primary care physicians in other countries, and there are larger income differences between specialists and primary care physicians in the United States compared to other countries. Specialists earn more relative to primary care physicians in all countries, and US specialists do not train for longer than specialists in other countries (Laugesen and Glied 2011). If physician compensation reflected the competition for talent and inequality, we would also expect to have seen the rise in income inequality track physician incomes. In fact, physician incomes were higher even in periods when America's income distribution was more even.
When looking beyond physician services, however, the upper-tail income inequality argument seems even less relevant. Upper-tail income inequality for physicians and health care executives fails to explain the higher cost of every health care good and service in the United States, even for less labor-intensive services or services that do not require highly trained personnel. For instance, room and board in hospitals is more expensive in the United States, and it is hard to argue that that is driven by income inequality. In addition, if ballooning executive compensation is driving higher prices, this fails to explain consistently higher prices for decades, even before income inequality became much larger. America's higher prices cannot be explained by income inequality.
Explanation 2: The Cost of Medical Education
Many practicing physicians, especially young physicians, when confronted with the high price of medical care will simply point to their extensive student loans. Is it possible that prices are higher in the United States simply because it costs more money to become a physician? After all, American medical students do have to pay much more in tuition than medical students in other countries. In most other countries, the public sector absorbs a larger share of the cost of medical education (Laugesen and Glied 2011). By contrast, the United States does not subsidize medical education to the degree that other nations do. (For that matter, the United States also does not subsidize other professional programs, such as law school, to the degree that such programs are subsidized elsewhere.) In addition, in many other countries, students can begin medical training very soon after high school. In the United States, physicians may take on student loans not just for medical school but also for their undergraduate educations.
This potential explanation, of course, touches on the higher US prices for physician services only, leaving the higher US prices for all other services unexplained. Medical devices, branded drugs, and hospital room and board are unrelated to medical students' loans, and yet are all priced at higher rates in the United States.
More important, a brief look at the numbers suggests that it is unlikely that student loans can explain very much of the large differences across countries. In 2009, medical students graduated with a mean debt of $156,456. That loan, paid down over thirty years, with a 6 percent interest rate, would require annual payments of $11,256 a year. The difference, however, between American physicians' salaries and their counterparts in other countries is much more than $11,000 a year, especially for specialists (Laugesen and Glied 2011). This is also evident based on the 2016 average debt of $190,000 (AAMC 2016a). Over ten years, this would cost $25,308 per year to pay back, based on a 6 percent interest rate ($2,109 monthly). As Laugesen and Glied (2011) showed, the additional loan cost paid by US physicians does not explain the size of the difference between physician incomes in the United States and abroad.
In addition, if medical school debt leads to substantially higher prices, we might expect prices to reflect changes in debt levels. This also is not supported by a steady increase in the percentage of people graduating from medical school who have no debt: 26.79 percent of medical graduates fall in this category, and 15 percent graduate with less than $100,000 in debt (Grischkan et al. 2017). Nor is the distribution of debt by specialty reflected in the variance in specialty incomes.
Explanation 3: Perceived Higher Quality and Greater Choice
Another important explanation may be cultural. Americans may simply hold widely shared understandings about the meaning of good medical care and the desirable features of the health care system. Furthermore, those beliefs interact with policies and create their own feedback effects.
Historically, messages around freedom of choice—a message advanced by health care interest groups such as physicians—have resonated with the US public, and this has had an impact on the framing of health care. Some of the public discourse on the creation and repeal of the Affordable Care Act has illustrated that some members of the public view the act as a government takeover of the health care system. A similar debate was seen during the Clinton reforms.
Higher prices might reflect American willingness to pay for high-quality health care. They may point to medical tourism to the United States by the wealthy as evidence of the superior quality of care available here. But there are at least four problems with this argument. First, while many academic medical centers likely offer the newest treatments and superior quality, people travel abroad for medical care to many countries, not just the United States (Laugesen and Vargas Bustamante 2010). In fact, the true state of cross-border health care might be reversed, when one considers that many people, especially uninsured individuals, cross the border into Mexico from the United States to obtain lower-cost pharmaceuticals and other services (Laugesen and Vargas Bustamante 2010). Second, if higher prices reflected higher quality of care, we would see a stronger relationship between price and quality within the United States. Yet there is no systematic relationship between the prices hospitals charge and quality (Batty and Ippolito 2017). Third, a majority of Americans do not perceive they will actually benefit from our so-called superior system: 64 percent say they are worried they will not be able to get high-quality care when they need it (Stremikis, Schoen, and Fryer 2011). Even if Americans want high-quality care, and may even be willing to pay more for it, the American public does not currently expect they will get quality care. Finally, related to this, the argument about the United States attracting patients looking for high-quality care is flawed because it assumes the quality of care received by wealthy medical tourists is equivalent to the mean level of care received by most US patients.
Another line of argument proposes that prices reflect the convenience and choice Americans demand, or at least are perceived to demand, which could theoretically drive up prices if there are more choices and fewer restrictions on access. US patients can typically see specialists and schedule elective procedures at their convenience, and there is a perception that Americans would not tolerate waiting lists or restricted choice. In Canada, for example, waiting times for surgery are around ten weeks after the surgical consult (Barua and Esmail 2013).
This argument is problematic because many OECD countries have public reporting of waiting times by hospital (Rechel et al. 2016), and the public in all countries demand timeliness of care. In addition, in terms of health outcomes, the availability and accessibility of primary care without a wait are likely to be more important, and on this outcome the United States does poorly compared to other countries. Wait times for elective surgery in the United States are short, but there are other countries with shorter wait times. Likewise, OECD data show that the proportion of US adults who have to wait longer than four weeks for a specialist appointment is 25 percent (the highest is 74 percent in Portugal). A lower percentage of adults in some other high-cost systems such as Switzerland have to wait more than four weeks, and only 20 percent of Israelis have to wait longer than four weeks.
If American patients believe that higher prices were associated with more attentive providers and an expansive range of treatment choices, they are mistaken. Survey data from 2013 show the United States to be unremarkable, as in not substantially worse or better than other high-income countries. Eighty percent of patients in the United States felt that their doctor had spent enough time with them—which was similar to other countries (OECD 2018). When asked if they had “been involved in decisions about care or treatment by their regular doctor,” US patients were close to other countries. In the United Kingdom, 83.9 percent of respondents said they felt involved in their care.
To argue that cultural factors have made the United States a health care outlier is to propose a very particular story for how culture can shape an industry. As described above, the United States is not unusual in terms of health care utilization, so culture would have to somehow distort US prices without distorting US quantities. Moreover, somehow America's unique culture would then raise prices for nearly all health care goods. This possibility seems unlikely—why would cultural forces distort one industry in such a particular way?
Explanation 4: Malpractice
Malpractice settlements, defensive medicine, and malpractice insurance costs are often said to be responsible for the higher fees charged by physicians. For many physicians, the malpractice system looms large—malpractice insurance is expensive, and the threat of a lawsuit is, for many, daunting.
But the credible research that exists on malpractice suggests that it is a much smaller part of the US health care system than one might initially think. One study suggests that the malpractice system accounts for only 2.4 percent of total health care expenditure (Mello et al. 2010). The Congressional Budget Office (2009) reviewed the evidence and concluded that tort reform in the United States would lower total health care spending by 0.5 percent. Other countries, of course, have their own malpractice systems, so these estimates leave little possibility for malpractice to account for very much of the difference in prices across countries.
Explanation 5: Administrative Costs May Contribute to Higher Prices
Traditionally, health policy analysts have downplayed the role of administrative costs in US health care expenditure, partly because they are traditionally already bundled into the fees paid to hospitals and other providers.
At the same time, even on the payer side, the amount is nontrivial, at 8.3 percent of total health expenditure in 2015 (OECD 2017), more than what is spent on public health or even health research. While estimates vary, they may in fact be significantly higher. The earliest study to highlight the issue of administrative costs (Woolhandler, Campbell, and Himmelstein 2003) concluded that such costs account for around 30 percent of US health care costs, based on extrapolating the hospital-specific estimates to the entire US health care system. A study in California, based on revenue rather than health care expenditure, examined administrative expenses and found that administrative costs were 10 percent of insurers' revenue, 21 percent of hospital revenues, and 27 percent of physician office expenses (Kahn et al. 2005). Another analysis estimated that 39 percent of the difference in health care expenditure between the United States and Canada could be attributed to administrative costs, although notably, higher US provider incomes or fees contributed 31 percent of the difference in expenditures (Pozen and Cutler 2010: 124).
As Kahn et al.'s (2005) study suggests, the impact on physicians appears to be greatest. US physicians spend forty-three minutes a day completing paperwork (Casalino et al. 2009), and physician-payer interaction in Canada costs practices just one-quarter of the amount spent in the United States (Morra et al. 2011). Estimates of those costs range from $85,276 (Sakowski et al. 2009) to $68,274 (Casalino et al. 2009) per full-time physician in 2006.
The emerging research in this area suggests that administrative costs may be more important than researchers believed previously, especially once one includes not just the cost of administering insurance but also billing and reimbursement costs. One can imagine two physicians, one in the United States and one abroad. The American physician would have to handle reimbursement from multiple payers with different rules and reimbursement rates. Typically, reimbursement is less complex in other countries, so how much of the American physician's higher prices can be accounted for by the higher administrative burden?
A natural question, then, is whether health care prices are higher in the United States solely because of administrative costs. Crude, back-of-the-envelope calculations suggest that extreme possibility is not the case. Administrative costs amount to, at the absolute most, 30 percent of US health care costs. That upper-bound estimate comes from extrapolating the hospital-specific estimates of Woolhandler, Campbell, and Himmelstein (2003) to the entire US health care system. If we were to take all US health care administration costs to zero, without somehow affecting GDP, then US spending on health care would shrink from 17 percent of GDP to 12 percent. Since all countries expend some resources on administration, it is unlikely that we could reduce administrative costs to zero. But at 12 percent of GDP, the United States would still be spending more of its income on health care than any other wealthy country—never mind that administrative costs are surely not that much higher in the United States for pharmaceuticals, or that the growing reliance on computers in health care is surely shrinking administrative burdens.
In all likelihood, administrative costs do play some role in America's outlier status on prices but cannot explain it entirely. Cutler and Ly (2011) argue that the issue is so important that there is then a role for the government in promoting electronic standards and implementing policies that would lower the burden of paperwork in the US health care system. But before we can determine policy solutions, we need a better understanding of what and how we are counting. For example, much of the data derive from self-reported estimates of time spent. People are unreliable when it comes to reporting the use of their time unless they track it from diaries. Therefore, developing objective measures of administrative costs is essential. There needs to be more standardized reporting of data—even within the United States, because current measures use different denominators, and there is confusion on some basic measurement issues (Sullivan 2013). We also need to know more about how the United States compares to other countries. For example, Himmelstein et al. (2014) compared hospital administrative costs as a proportion of expenditure on hospitals and estimated these at 25 percent in the United States, 19.8 percent in the Netherlands, and 15.5 percent even within England's single-payer system. Discussion on this issue is ongoing, and more work—both academic research and policy crafting—is needed.
Explanation 6: Market Fragmentation
Health care in most countries involves a single insurer, solely public health care providers, or both. By contrast, the United States has hundreds of insurers and thousands of providers. The lack of dominant purchasers, in particular, makes the US health care system unique. Vladeck and Rice (2009) argue that this lack of strong, coordinated purchasers in the United States explains many of the country's health care problems.
In most markets, monopolies and monopsonies are undesirable. Some, however, make an exception for health care, arguing that more powerful health care purchasers can extract lower prices from providers and thus slow spending. The theory is that, when a local hospital industry faces many, small insurers, prices tend to rise as each insurer enjoys lower bargaining power. It is not hard to imagine in such a context that a single insurer could dramatically lower prices. After all, hospital mergers typically lead to higher prices (Gaynor and Town 2012), so a shift in market power in the opposite direction ought to lower prices.
Similarly, one explanation for the high price of pharmaceuticals in the United States is the many pharmaceutical purchasers in the United States. Most countries have a single formulary, and the pharmaceutical industry must bargain with that single entity. In the United States, by contrast, pharmaceutical companies face a much more favorable position, with many formularies operated by many payers.
Still, there are reasons to be skeptical that a single payer or that, more generally, consolidation in the insurance industry would unambiguously lower prices. Past research suggests that insurance mergers can raise the premiums that consumers face (Dafny, Duggan, and Ramanarayanan 2012). This market may be so complex that it cannot be understood simply by comparing insurer and hospital bargaining power. Patients do not like to change networks every year, so insurers cannot credibly threaten to drop all hospitals from their networks. If nothing else, prominent hospitals often are very difficult to cut from a network. And, of course, hospitals may merge in response to increased market power among insurers.
In the end, the industrial organization of health care is more complex than that of other industries. It is important to note that the US health care system is more fragmented than other countries' systems, and it is possible that this fragmentation leads to higher prices. But such an observation does not offer immediate policy implications, nor does it suggest, more generally, a way forward. The problem is that we do not fully understand the impact of fragmentation on the market. However, it does seem that this is one major structural difference between the United States and other countries. In terms of the potential for greater institutional governance of the health care system, there is a conundrum, because reducing fragmentation requires greater coordination, while greater coordination might increase prices. The issue of how health care markets should be structured and how we should balance the market power of different players is of critical importance.
Explanation 7: Rent Seeking
As Robert Evans (2002: vi) once put it, when it comes to health care, “cost control is income control,” or “every dollar of health care spending is someone's health care income” (Reinhardt 2012: 41)—meaning policy makers may view the high price of health care in the United States as a problem, but to health care providers it is the basis for their income. Thus, some would argue, the high price of health care in the United States is not driven by culture or even by market forces but, rather, by effective lobbying on the part of health care providers.
Traditionally, higher incomes that reflect returns to political lobbying are considered an example of what economists would call rent-seeking behavior, in which firms induce the government to protect their profits. Bivens and Mishel's (2013: 62) analysis of US executive and financial compensation through stock options makes the case for a broader definition of lobbying: rent means only that the income received was in “excess of what was needed to induce the person to supply labor and capital to these respective markets.” It is difficult to quantify how much rent seeking explains higher prices in the United States—as Bivens and Mishel (2013: 63) point out, observing rents is difficult. But, as many have pointed out, the United States depends to a larger extent on private actors that operate in a “submerged state” (Mettler 2011) compared to other countries. The maintenance of the health care status quo has depended on a robust lobby that uses campaign contributions, media advertising, and direct lobbying of legislators and executive branch staff to keep prices higher. The industry continuously and relentlessly protects its ability to maintain high prices for particular services through rule making in the executive branch and blocking or promoting legislative actions favorable to its interests. In addition, pressure comes from industries outside of health care that influence prices. For example, if physicians shoulder high malpractice premiums, this may be at least partly because trial lawyers are a potent lobby. Health care prices today thus reflect a long history of lobbying. Aggressive efforts on the part of the “House of Medicine” to skew reimbursement rates ever higher go as far back as the mid-1950s, but even efforts such as the Flexner Report constitute an income-related and profession-sponsored effort to increase incomes (Laugesen 2016).
The US Constitution enshrines, in a way perhaps uniquely compared to other countries, the private right to petition the government. Though some countries do incorporate interest groups within a corporatist decision-making structure, there is more of a sense of there being a limited pie.
The most compelling evidence on rent seeking and prices is in the area of physician services, where there is more information about policy capture. For decades, physician interest groups undertook anticompetitive actions to protect their turf—or the prices they charged—such as preventing advanced-practice nurses from competing with them. To this day, physicians influence the rates paid by Medicare through regulatory capture under the mantle of advice to the Centers for Medicare and Medicaid Services (Laugesen 2016).
In the hospital sector, there is less evidence of rent seeking in relation to prices, but at least beginning with the 1946 Hospital Survey and Construction Act (known as the Hill-Burton Act), which subsidized hospital construction, the hospital sector has benefited from government subsidies. According to Hackey (1999), the American Hospital Association published a white paper asserting the need for a new industry standard whereby US hospitals would be paid based on cost-based reimbursement that reflected the entire cost of treatment, and this policy approach was embraced. Unlike the medical profession, however, hospitals' nonprofit status made some regulation more acceptable. For example, there were many efforts to subject hospital prices to scrutiny, through either voluntary or compulsory all-payer rate setting (Murray and Berenson 2015).
Finally, the pharmaceutical industry has also long defended its revenue, pushing pharmaceutical prices ever higher. The industry has always advocated the prevention of reimportation from other countries. Just as important, Medicare has never been allowed to leverage its buying power to negotiate lower drug prices, and the industry has aggressively pushed for changes in how patents are awarded that would maintain high prices for branded drugs.
In short, the American Medical Association, the pharmaceutical industry, and the hospital industry have been remarkably effective through the years at preventing reforms that would have threatened their income. The only health care reform bills that have been at all politically viable have tended to be reforms that did not threaten incomes for these special interests. As a result, both directly and indirectly, the prices Americans pay for health care have steadily risen.
Lowering Prices: Neither an Easy Task nor a Simple Solution
What ought we do about high health care prices? In theory, the government could lower public reimbursement rates and then somehow impose price ceilings on the private sector. Economic theory suggests that, in a competitive market, such price controls would lead to shortages. The market for health care may not be competitive, but price controls could still lead to a variety of trade-offs. There is no monolithic fix that would lower health care prices without causing other problems.
That said, many fixes have been proposed. For instance, some argue that encouraging price transparency would make consumers more aware of prices, which in turn would induce consumers to comparison shop and thus would encourage providers to lower their prices. However, consumers do remarkably little comparison shopping in health care (Brot-Goldberg et al. 2017). Studies of interventions that increase price transparency have found that such interventions do lower prices but by very little (Lieber 2017).
Some advocate for all-payer rate setting in the hospital industry. However, there is little evidence to suggest that such a radical policy change would bring down prices without dramatic trade-offs. All-payer rate setting remains the law in Maryland, the only state that has retained such a system. And yet Maryland is not an outlier in its spending growth (Pauly and Town 2012). If all-payer rate setting were a panacea, one would expect the Maryland health care system to be a role model for the rest of the country. It is not.
More important, the effects of all-payer rate setting, and reductions on hospital reimbursements more generally, are difficult to predict. To our knowledge, there exists no unambiguous economic theory as to whether such a radical policy change would be welfare enhancing. Hospitals facing lower reimbursement rates might be less eager to launch high-tech services (Freedman, Lin, and Simon 2015). Lower reimbursement rates might also lead hospitals to close unprofitable but important services such as trauma centers, psychiatric emergency services, and alcohol and drug treatment centers (Horwitz 2003).
For a time, many believed that the answer to high health care prices was managed care. In decades past, managed care organizations succeeded in lowering the prices paid to providers, with no clear effect on quality (Cutler, McClellan, and Newhouse 2000; Glied 2000). Nevertheless, managed care eventually had to contend with mounting consolidation in the hospital industry (Lesser, Ginsburg, and Devers 2003), and state legislation greatly limited its reach (Pinkovskiy 2013). Current models of payment reform that aim to manage care differently, such as through accountable care organizations, have been less aggressive—and less successful.
All in all, it is this type of discussion that makes health policy enduringly frustrating. One can imagine policies that would be powerful enough to bring American health care prices in line with prices in other countries. We could induce dramatic cuts to Medicare reimbursement rates and all-payer rate setting. It is unlikely that blunt cuts will be tolerated politically; even the smallest payment reforms have tended to be diluted such that they lack the penalties and claw-backs that are necessary. The history of Medicare's sustainable growth rate demonstrates the difficulty of imposing cuts (see Laugesen 2009). Therefore, the policies that are politically viable are sufficiently toothless and watered-down that they have little chance of moving the needle.
In the past, health policy scholars and health policy makers focused on managed care, single-payer proposals, and some of the other policy proposals described above. Today, the focus has shifted to payment reform: accountable care organizations, bundled payments, Medicare Advantage, and other changes to the fundamental structure of how health care providers are paid. One could argue that a shift toward these new models is warranted given the fundamental problems of fee-for-service payments. It is often argued that such payments create an incentive to provide too much health care (McGuire 2000).
The discussion above suggests an alternative view. Fee-for-service payments are not a uniquely American phenomenon, though the particular rates paid in America dwarf those in other countries. Historically, American policy makers have found it politically difficult to lower those reimbursement rates. Thus, payment reform arises simply because the obvious first option, to lower provider reimbursement rates, is unavailable. Even strategies such as increasing the evidence base for treatments has, on occasion, been subject to political pressure by providers, such as when Congress reduced and renamed what was the Agency for Health Care Policy Research (see Gray, Gusmano, and Collins 2003). In other words, if policy makers cannot adjust reimbursement rates down, then what option do they have other than to change the nature of reimbursement entirely? Addressing the influence of health care sector providers on health prices is likely to require multiple strategies and, most important, greater public awareness of the role of price in the costliness of the US health care system. Ultimately, policy makers are prone to engage in “throwing darts” to solve cost-containment; what we need is public support for reigning in the price of medical care (Oberlander 2011).
The authors are extremely grateful to Gerald Anderson, Larry Brown, Herb Pardes, Eric Patashnik, Mark Schlesinger, Michael Sparer, Jennifer Costanza, and one anonymous referee for enlightening feedback and commentary. We also appreciate the assistance of Roshini Hegde and Sophia Leenay.