The Great Reversal

by Thomas Philippon

  The second pitfall is that the US lacks a framework for data protection and data ownership. US banks want to keep control of their clients’ information to stave off competition. The same debate occurred in Europe, and, as you might expect, European banks lobbied hard against the idea of sharing their data. Unlike in the US, however, where legislators caved in immediately, the EU pressed ahead. Their fortitude was a consequence of the General Data Protection Regulation (GDPR), which essentially states that people own their data. According to the GDPR, since you own the information in your bank account, you should be able to decide who has access to it and who does not. This stands in sharp contrast to what happened in the US. In effect, US banks’ customers do not own their data, and US legislators have let it happen.

  I grew up in a country, France, where large banks traditionally had more influence over policy than in America. I did not imagine that one day this fact would be turned on its head. There is a growing gap between open banking in Europe and in the US, and it is not happening by chance. It is consistent with the theory of European regulation that we developed in Chapter 8. Europeans are probably less naturally inclined toward free markets than Americans are, but when they agree to regulate a market at the EU level, they opt for tough, independent regulators. I have shown why this is the equilibrium of the political game. As it turns out, following the crisis of 2010–2012, Europeans decided to move banking supervision and regulation at the EU level. Since then, banking regulators have become tougher and more independent, and banks’ lobbyists have lost some of their influence. This was clear during the GDPR debate and during the revision of the Payment System Directive (PSD2). We will return to the GDPR and the issue of privacy more broadly when we discuss Facebook and Google in Chapters 13 and 14.

  The third pitfall is that asset management can become too concentrated. Index funds and exchange-traded funds (ETFs) are great inventions. They are cheap, they are simple, and they are certainly better for 95 percent of investors than actively managed funds. They have also expanded rapidly. The share of listed equity value owned by institutional investors has increased since 2000—primarily driven by the growth of quasi-indexer institutions. These trends are encouraging, but they imply significant concentration in the management of equity portfolios. One issue is that large money managers seem to have a marked preference for share buybacks over capital expenditures. In Gutiérrez and Philippon (2017), we show that buybacks have increased faster for firms with high quasi-indexer ownership. Large investors are looking for firms protected by barriers to entry because they understand the value of market power. Investors such as Warren Buffett have been amazingly successful in part by doing just that. This is, of course, a perfectly legitimate investment strategy. However, directly or indirectly, the growth of large money managers might reinforce the trends toward high markups and low investment in the US economy.

  To conclude, there are new things in finance, and there are valuable things in finance. All too often, however, what is valuable is not new, and what is new is not valuable. There are some reasons to think that this might be changing, in large part thanks to fintech firms. But fintech innovations will not automatically enhance stability or democratize access to financial services. If we want to reap the benefits from better technology in finance, we need financial regulators who can stand up to the lobbies.

  CHAPTER 12

  American Health Care

  A Self-Made Disaster

  People in middle age now have a fair chance of getting to Longevity Escape Velocity.

  AUBREY DE GREY

  Those in midlife now are likely to do worse in old age than the current elderly.

  ANNE CASE AND ANGUS DEATON

  NOTHING BETTER HIGHLIGHTS the clash between economists and techno-optimists than the topic of health. It is a sad irony that precisely at the time when scientists are beginning to argue that it is technologically plausible to prolong human life indefinitely, the US is experiencing the first peacetime decline of life expectancy of any democratic nation since the Industrial Revolution.

  Techno-optimists view the world through the lens of what is technologically feasible. But sadly, we’ve long known that what is technologically feasible can be rendered practically impossible by bad public policy. Nobel Prize–winning economist Amartya Sen demonstrated nearly forty years ago that cases of mass starvation are, above all, political issues: “Starvation is the characteristic of some people not having enough food to eat. It is not the characteristic of there being not enough food to eat. While the latter can be a cause of the former, it is but one of many possible causes. Whether and how starvation relates to food supply is a matter for factual investigation” (Sen, 1982). The problem, Sen argues, is rarely that there isn’t enough food, but that those in power lack the will or desire to make it available to the hungry. In other words, starvation is usually a manmade policy disaster.

  The same seems to be true with health care in the US. The gap between what we could do and what we actually do is much greater than the rate of technological change over at least several decades. The US has the best hospitals and the best technologies, and yet it has mediocre health outcomes. An inefficient, oligopolistic, and sometimes corrupt health-care system is not the only reason, but it is a major contributor.

  The US is the largest economy in the world and one of the richest on a per-capita basis. Yet, compared to other rich countries, a higher percentage of its population lives in poverty, infant mortality is higher, and life expectancy is shorter.a

  Figure 12.1 shows the evolution of life expectancy at birth in the US, France, the UK, and Costa Rica. I chose Costa Rica to broaden the comparison and to emphasize the difference between income and health. In 2000, the life expectancy of French newborns was 79.2 years. In the US it was two and a half years lower, at 76.7. By 2016, the gap had increased to 4.2 years, as life expectancy reached 82.8 years in France versus only 78.6 in the US. The US has lower life expectancy than other rich countries, certainly, but US life expectancy is lower than less wealthy countries, too: a Costa Rican newborn’s life expectancy in 2016 was 1.2 years longer than that of a child born in the US, and the gap has been increasing in recent years.

  Figure 12.2 shows the evolution of infant mortality. It has been decreasing around the world. In the US infant mortality stopped decreasing after 2010 and settled at a higher level than in other rich countries, such as the UK and France. Data from 2014–2016 indicate 5.9 deaths per 1,000 live births in the US versus 3.7 in France and 3.9 in the UK. Costa Rica’s rate was significantly higher, around 8.

  FIGURE 12.1  Life expectancy. Data source: OECD

  FIGURE 12.2  Infant mortality rates. Deaths per 1,000 live births. Data source: OECD

  Health-Care Costs in the United States

  In addition to dispiriting outcomes, health-care costs are much higher in the US than in similar countries. The average cost of employer health coverage is close to $20,000 for a family plan in 2018. You need to be a bit careful with this number, because the US is a rich country and, as we have discussed in Chapter 7, the Balassa-Samuelson theory tells us to expect that nontraded goods and services will be systemically more expensive in rich countries. Health care, then, should also be more expensive in rich countries.

  FIGURE 12.3  Health-care cost versus GDP per capita in select countries. US = United States; CH = Switzerland; NO = Norway; IE = Ireland; LU = Luxembourg. Data source: Kaiser Family Foundation analysis of OECD data

  Figure 12.3 shows the Balassa-Samuelson effect for health care. You can see that per-capita health-care costs increase systematically with income per capita. However, Figure 12.3 also shows that US health-care costs are completely off the chart (or off the regression line, to be precise). Health-care costs per capita are much higher in the US than in Norway or Switzerland, both of which have similar levels of GDP per capita. (GDP per capita in Luxembourg and Ireland is biased by the activities of large multinationals.)

  Figure 12.4 shows the shar
es of GDP spent on health care for the US and for the average of comparable OECD countries. Two facts stand out. First, health-care costs are rising everywhere. Second, the increase is much larger in the US. The US has always spent more than other rich countries on health care, but the gap has increased dramatically since the 1980s.

  FIGURE 12.4  Health-care spending, share of GDP. US versus OECD, averages. Data source: Kaiser Family Foundation analysis of OECD data

  In 2018, the US spent about 18 percent of GDP, or $3.3 trillion, on health care. Where do these costs come from? Hospital care is the largest single component of health-care spending in the US. It accounts for more than $1 trillion per year. The second largest category is physician and clinical services, many of which are now provided by hospital systems as well. The third largest category is prescription drugs, at about $330 billion.

  The allocation of public versus private spending on health care is also informative. Public spending in the US is almost exactly the same as in other countries. Private spending, on the other hand, is three times higher than the OECD average. The US has several health systems within the public and the private spheres. The public sector consists of Medicare, Medicaid, the Indian Health Service, and the Veterans Administration, all of which are separate systems. Moreover, the different states are also quite different when it comes to how they organize their health systems. Similarly, the private health-care system is not just one system but many subsystems. This complexity certainly explains part of the excess costs.

  The US might soon spend 20 percent of its GDP on health care, almost double what other countries are spending. How do we understand this fact?

  Price versus Quantity

  The reason Americans spend so much on health care is because prices are higher, not because they consume measurably more care. The Health Care Cost Institute finds that the growth in insurer claims from employer-sponsored coverage from 2012 to 2016 “was almost entirely due to price increases” for emergency-room visits, surgical hospital admissions, and administered drugs. The real quantity of health care, on the other hand, “remained unchanged or declined.”

  “The marketplace is just not working,” Gerard Anderson, a health-care economist at Johns Hopkins University, told the Wall Street Journal in 2018. Insurers that must negotiate reimbursement with health-care providers for plans offered by employers pay on average 50 percent more than Medicare, and those rising costs are “the main culprit for why the U.S. spends so much on health care” (Mathews, 2018).

  Researchers have also compared the costs of specific health-care items across countries to find out why the US is so expensive.b Neither the number of doctor visits nor the length or frequency of hospital stays can account for the extra cost. These are similar to the average in other rich countries. Once again: it’s the prices that explain the difference. Annual costs for drugs were $1,443 per person in the US, compared with an average of $749 per person in Europe.

  Another significant driver of high prices in the US is the large layer of “administrative” costs, which include costs related to planning, regulating, and managing health systems and services. These administrative costs appear to be high in the US, at around 8 percent of total health-care spending. That’s more than double the average of 3 percent in other countries. Other studies consider broader definitions of administrative cost—taking into account indirect costs such as time spent on administrative tasks—and put the figure as high as 25 percent of total costs (Tseng et al., 2018).

  I put “administrative” in quotation marks because these are really rents extracted by dozens of layers of health intermediaries and providers, from insurance companies to hospitals. We know from much research in economics that lax competition leads to high “administrative” costs.

  Labor costs also explain part of the difference. The salaries of physicians and nurses are higher in the US than in other countries. For example, generalist physicians earn around $220,000 in the US, compared to $120,000 on average in other rich countries. That is a lot more than the average difference in GDP per capita and is not accounted for by the Balassa-Samuelson effect. A caveat, however, is that other countries have nearly free medical education. In the US, medical students graduate with over $200,000 in educational debt.

  US Health-Care Productivity Is Low

  One way to summarize these results, then, is that the US does not seem to be very efficient at keeping its citizens in good health: it spends more than any nation, and its citizens live shorter and less healthy lives on average.

  Of course, the health-care system is not the only culprit here. Other factors—such as genetics, behavior, social circumstances, environmental and physical influences—also contribute to life expectancy. Most studies find that medical care accounts for less than 20 percent of the observed variations in morbidity and mortality.c Tobacco, poor diet, and lack of physical activity have a large impact. In pathbreaking research Anne Case and Angus Deaton (2017) document increases in mortality and morbidity among white non-Hispanic Americans in midlife since the turn of the century. Among white non-Hispanic people without a college degree (high school or less), mortality is rising in all age groups. Mortality rates among blacks, by contrast, fell across all age groups. They find striking increases in suicides, overdoses, and alcohol-related liver diseases among those with a high-school degree or less. “Mortality declines from the two biggest killers in middle age—cancer and heart disease—were off-set by marked increases in drug overdoses, suicides, and alcohol-related liver mortality in this period.”

  TABLE 12.1

  Top-Scoring Countries for Health-Care Access and Quality

  HAQ index

  Countries

  97

  Iceland, Norway

  96

  Netherlands, Luxembourg, Australia, Finland, Switzerland

  95

  Sweden, Italy, Andorra, Ireland

  94

  Japan, Austria, Canada

  93

  Belgium

  92

  New Zealand, Denmark, Germany, Spain, France

  91

  Slovenia, Singapore

  90

  UK, Greece, South Korea, Cyprus, Malta

  89

  Czech Republic, US

  A better metric to compare health-care systems around the world would then be to use mortality amenable to health care. A recent large-scale study uses data on diseases, injuries, and risk factors to build the Healthcare Access and Quality Index (HAQ) for 195 countries (GBD 2016 Healthcare Access and Quality Collaborators, 2018). They track thirty-two diseases and injuries that are not supposed to kill you if you have access to effective care and see how many people actually survive. If everyone survives these theoretically preventable deaths, that’s a perfect score. Table 12.1 shows the first nine scores.

  The top scores are found in European countries, plus Canada, Australia, and New Zealand. The US ranking does not reflect its wealth or level of spending on health care. Two other features distinguish the US in this study. It has the lowest absolute improvement in the index between 2000 and 2016 among rich and middle-income countries, and there is relatively high inequality in HAQ indexes within its borders. Mississippi has the lowest score (81.5), while a subset of northeastern states, Minnesota, and Washington state have European-style scores.

  This study is not perfect. It still suffers from attribution problems. Any outcome measure has its own difficulties, but if we look at a broad range of studies, it is clear that the US does quite well on acute care outcomes and really badly on population health outcomes.

  It is fair to say, then, that the US health-care industry suffers from low productivity. If we dig deeper, however, we also find that the US health-care system offers prime examples of several economic illnesses, such as oligopolistic industries, conflict of interest, regulatory capture, and political capture. Unfortunately, these lead to high costs and poor results.

  Concentration

  Hospitals have increased their market
power by mergers. There have been almost seventy mergers per year since 2010. Today, close to 80 percent of Americans living in metro areas are in highly concentrated hospital areas. Recently, for example, two large hospitals based in Texas, Baylor Scott & White Health in Dallas and Memorial Hermann Health System in Houston, have announced their plan to merge. This would create a massive sixty-eight-hospital system, among the largest in the US. Although these hospitals are nominally classified as nonprofit organizations, their combined revenue is more than $14 billion.

  Managers always justify mergers by claiming they will increase efficiency, lower costs, and improve care. They carefully avoid the issue of market power. And if history is any guide, the efficiency gains are unlikely to happen, but price increases are very likely.

  Amazingly enough, the hospital sector seems to view the airline industry as a role model. “Scale done right is vital,” said John Starcher, Jr., chief executive officer of Bon Secours Mercy Health, in an interview with the Wall Street Journal. “It’s no secret to anybody … how far behind the consolidation curve the hospital sector is compared with others, such as airlines.”

  Yes, you have read it correctly. Airlines as the example that hospitals plan to follow? Yikes!

  In reality, an important reason that hospitals merge is the opportunity to acquire more bargaining power for negotiations with health insurers. Insurance company executives (who, ironically, also enjoy virtual monopolies in many US states) are quick to offer an opinion when hospitals try to merge. When there is a dominant provider in town, all insurers have to include them in their plans. When a hospital system is the only provider of in-patient services in a region, an insurer serving people in that area has no option but to include them. As a monopoly, the hospital is practically free to charge whatever it wants.