By “commanding heights,” Lenin meant the critical sectors that dominated economic activity — primarily electricity generation, heavy manufacturing, mining, and transportation. Because these industries were the foremost drivers of employment, production, and consumption in Russia — and because they were the essential growth sectors in any economy of that era that sought to be called “modern” — government control of these particular sectors meant government dominance over the economic life of the nation. A communist government could afford to permit relatively free markets in less significant sectors, Lenin thought, because as long as it controlled those industries that formed the heart of the economy, it effectively controlled the whole.
Throughout much of the 20th century, communist and socialist parties around the world continued to see government dominance of these industries as a key goal. The commanding heights of the economy became crucial battlegrounds in the struggle between advocates of central planning and defenders of market economics.
In America today, few people champion government control of the industries Lenin saw as the commanding heights. On the contrary, these sectors have been largely deregulated, and market forces have, for the most part, been permitted to govern their development for decades. Defenders of the market might therefore imagine that they have won, and that the struggles that remain are peripheral debates.
But such a declaration of victory would be dangerously premature. Over the past few decades, our economy has undergone some fundamental changes — with the result that the fight for control over the commanding heights of American economic life is still very much with us. And it is a fight that, at least for now, the free-market camp appears to be losing.
The commanding heights of our economy today are not heavy manufacturing, energy, and transportation. They are, rather, education and health care. These are our foremost growth sectors — the ones most central to employment and consumption; the ones that, increasingly, drive our economy. And it is in precisely these two sectors that the case for extensive government intervention and planning, if not outright control, is dominant — and becoming ever more so.
If there is to be any hope of reversing this trend, champions of market economics must come to see these two sectors as the front lines in the battle for capitalism. At stake is not only an ideological or theoretical point, but also American prosperity. The historical record makes this clear: In the nations where it was practiced, government control of the old commanding heights of the economy made those industries less efficient and less innovative — bringing overall economic performance down with them.
Today, America risks following the same course. Looking to the coming decades, it will simply not be possible to maintain a genuine free market — or a thriving, innovative, growing economy — if our education and health sectors are controlled by the government. Champions of the market thus have their work cut out for them. First and foremost, however, they must come to understand the central place that education and health care occupy in America’s economic life.
AN ECONOMY TRANSFORMED
To discern where the heart of an economy lies, one must identify the sectors in which employment and consumption are focused, and in which growth is swiftest. In the case of our own economy, the data over recent decades clearly show the decline of the old commanding heights — manufacturing and heavy industry — and their replacement by “softer” sectors, especially health care, education, and government work.
Economists Michael Spence and Sandile Hlatshwayo recently devised a way of breaking the American economy into industries that produce tradable goods and services and industries that produce non-tradable ones. They calculated that, from 1990 to 2008, employment in the tradable sector edged up from 33.7 million to 34.3 million. Meanwhile, in the non-tradable sector — which covers most service-based businesses — employment rose from 88.3 million to 114.9 million. Thus the non-tradable sector accounted for nearly all of the job growth during this period.
We are accustomed to thinking that our country is in the midst of a long transition from an industrial economy to a “service” economy, and these figures would seem to confirm that perception. But the service economy is not what we often think it is. The images that most readily come to mind when we think of these sectors might involve retail sales, information-technology consulting, and financial services. But Spence and Hlatshwayo’s work shows that, within the non-tradable sector, health care was easily the growth leader — increasing from 10 million to 16.3 million jobs, and accounting for almost 25% of total job growth in the past two decades. Government was second, growing from 18.4 million to 22.5 million jobs, and accounting for about 15% of total job growth. Of this expansion in government employment, nearly 70% was attributable to jobs in education. Today, the drivers of the American labor market are therefore clearly health, education, and government work; these sectors form the backbone of our post-industrial economy.
Over the past decade, this trend has only accelerated. Economist Michael Mandel has shown that, between February 2001 and February 2011, employment in the U.S. economy in health care, education, and government increased by 16%. This was not simply a function of a growing population and economy: During the same period, employment outside of those sectors decreased by 8%. Wage gains were similarly tilted toward health and education. “What we see,” Mandel explains, “is that health and education (public and private) accounted for an amazing 75% of real wage and salary gains.”
A similar picture emerges when one looks at changes in American spending and consumption patterns. Our examination of the most recent data from the Bureau of Economic Analysis concludes that spending on health care and education (public and private combined) accounted for 21% of gross domestic product in 2000 and roughly 26% in 2010. Education and health-care spending thus accounted for an astonishing 37% of the overall growth of the economy over the past ten years.
The recession of the past few years presents an especially clear picture of consumption patterns and priorities — for it is during such downturns that people must prioritize their spending, thus offering economists a sense of the relative degree to which Americans value certain goods and services over others. In the period between January 2008 and January 2009, for instance, Americans significantly reduced spending related to cars, as well their spending on clothing and food. At the same time, they increased spending on education, recreation, and, most of all, health care.
Several related factors have combined to produce this trend. When economists compare different sectors — particularly in terms of employment — they consider the relationship between demand for the goods those sectors provide and the productivity of those sectors in meeting that demand. If demand for a sector’s product grows more rapidly than that sector’s productivity can increase in order to keep up, employers in that sector will need to hire more workers to bridge the gap. Similarly, when demand grows less rapidly than productivity, that sector will experience relative shrinkage.
Changes in both demand and productivity across the economy have brought about the growing dominance of the health and education sectors. Relative changes in demand have chiefly been a function of rising incomes over many decades — a shift that has put many basic necessities (such as food, clothing, and shelter) more easily within the reach of more Americans. At the same time, this shift has provided Americans with more disposable income, which many people have chosen to spend on health care and education. This does not mean that people are spending less money in nominal terms than they once did on food, clothing, or manufactured items. What it means is that people spend a smaller proportion of their incomes on these commodities, since Americans now have significantly more money to spend than they used to.
This change is easiest to see when examining spending patterns over the very long term. Economic historian Robert Fogel compared how potential income was divided among broad categories of goods and services in 1875 with its distribution among the same categories 120 years later. He found the following:
The period Fogel examined is, obviously, an extremely long stretch over which to trace economic trends — spanning from roughly the beginning of the industrial era in America to nearly its end. But what Fogel confirmed in his research was that, across this wide window of time, the proportion of people’s incomes spent on bare necessities diminished as their incomes rose. Meanwhile, more of Americans’ wealth came to be devoted to health, education, and leisure.
It should be noted that Fogel’s findings are expressed in terms of potential income — in other words, the income that someone could earn from working full time as an adult, including income that is implicitly taken as leisure. Thus, when Americans devote greater portions of time to leisure (as they now do, thanks to developments like a shorter work week and longer lives in retirement), this is expressed as a proportion of income — which tends to inflate the numbers for leisure in Fogel’s calculations. Putting leisure aside, however, his consideration of how real income has been used by Americans over the past 12 decades provides more evidence in support of a crucial point: As our society has grown more wealthy, there has been a significant increase in the proportion of income that people choose to devote to education and health care; a significantly smaller share, meanwhile, has come to be devoted to basic necessities and durable goods.
Economists describe such trends in terms of “income elasticity of demand.” This phrase refers to how the demand for a particular good responds to changes in the incomes of the people demanding that good. Some goods, like public transportation, show a negative income elasticity of demand — meaning that people consume less of them as they get wealthier. Others, like bread and butter, show an elasticity near zero — meaning people buy them at roughly constant levels regardless of whether they are rich or poor.
Fogel has shown that education and health care have particularly high income elasticities of demand. By his calculations, the long-term income elasticity of demand for each is roughly 1.6 — meaning that, if a person’s potential income goes up by 1%, his spending on those services will rise by 1.6%. As people get wealthier, therefore, the relative portion of their incomes going to health and education will increase.
To illustrate the point, suppose, for example, that in 2030 potential income has increased over its 1995 level by 100%. If in 1995 we had $100 in potential income and spent $9 on health care (as Fogel’s figures show), then in 2030 we will have $200 in potential income and spend $23 on health care. In other words, while potential income will have doubled, spending on health care will have increased by a factor of more than 2.5. Generally speaking, this is the trend that has emerged in education and health-care spending in recent decades — and all signs suggest that it will continue.
But the growing centrality of education and health care is not only a function of public preferences and demand. Another important factor, especially to the two sectors’ growth within the labor market, is the fact that it is more difficult to squeeze labor costs out of those industries than it is in, say, manufacturing or agriculture. After all, most factory work does not require deep knowledge or complex judgment. As a result, engineers are constantly developing machines that can substitute for humans in manufacturing. Furthermore, as countries like China and India become more integrated into the global economy, an ever-larger pool of low-skill labor becomes available. The need for manufacturing labor in the United States is therefore reduced; the relative cost of manufacturing output is thus held down.
Compared to manufacturing, the delivery of services in education and health care today is relatively labor intensive. Teachers and doctors require much more training than do manufacturing workers. Everyday work in education and health care generally involves more judgment and complex decision-making than are required on a production line. These higher-level tasks are not as easily handed over to machines or outsourced to low-skilled workers abroad.
Education and health care are also more resistant to the productivity increases that have dramatically altered the manufacturing sector. Factory automation, for instance, can swiftly raise the number of widgets produced per worker; office automation has vastly streamlined supply-chain management, inventory control, and accounting. But increasing the number of operations per surgeon, or the number of essays graded per teacher, is much more difficult. Hence, productivity growth in health care and education lags behind that in other industries.
As a rule, this means that health care and education tend to be less efficient. As increased productivity has led to wage growth in other, more efficient industries, the inefficient sectors must maintain competitive wages. But without the commensurate productivity gains, they experience cost growth, an effect named “Baumol’s Cost Disease” (after the economist William Baumol, who identified it in the 1960s).
Baumol’s famous illustration of this phenomenon compared classical musicians with auto workers. It takes just as many musicians to play one of Mozart’s symphonies today as it did a half-century ago, but it takes far fewer auto workers to produce a car now than it did then. As a result, manufacturing has become much more efficient — employing fewer people, but paying each of them somewhat more. Orchestras can’t employ fewer people, but they do have to pay each of their employees more than they used to — if only to keep up with the rest of the economy, lest their musicians run off to become auto workers.
The result is that, over time, costs in less efficient industries — like the fine arts, but also health care and education — will increase in relation to costs in more efficient industries. And these increasing costs, as well as rising demand for the services these sectors offer, have combined to place both education and health care at the commanding heights of today’s economy.
PUBLIC SECTORS
If it were true only that health care and education are increasingly important sectors of our economy, there would be little cause for concern. Indeed, societies ought to desire economies that are strong and flexible enough to hum along as new technologies and other developments cause industries within them to rise and fall. The problem, rather, is that both health care and education are increasingly government-dominated industries. And this domination produces two ill effects that exacerbate the changes these sectors are already undergoing: Government’s influence artificially increases the demand for health care and education (by significantly subsidizing both), and it makes both sectors even less efficient than they would be otherwise (by heavily regulating them and shielding them from market forces).
In most industrialized countries, more than 80% of health-care spending is now paid for by third parties, primarily government, leaving about 20% to be paid directly by consumers. In the United States, however, only about 10% of health-care spending is paid for by households out of pocket. About 50% is directly paid for by government, mainly through Medicare and Medicaid. And about 35% comes from private health insurance, which is heavily subsidized by the government through the income-tax exemption for employer-provided coverage. The result is that patients rarely need to factor in cost when making choices about medical treatment, since someone else is footing almost all of the bill.
Removing cost as a consideration certainly increases the demand for medical services, although it is difficult to calculate precisely how much. In 1971, RAND began a 15-year study to examine the effects of free medical care on both health-care consumption and participants’ actual health quality. To date, the experiment is the only major controlled study of health-care spending behavior in America. And its findings clearly demonstrated that households tend to reduce consumption of medical services when they shoulder a greater share of the cost. Families whose coverage included cost-sharing components (like co-insurance) used between 20% and 30% less medical care (depending on the extent of their co-insurance). Moreover, as the RAND study put it, “In general, the reduction in services induced by cost sharing had no adverse effect on participants’ health.”
As for education in the United States, government directly provides most schooling from kindergarten through 12th grade through the public-school system that now educates about 90% of American children. Through various tax benefits, it also subsidizes some educational expenses for parents who choose to have their children educated outside of that system. In addition, there are large public universities where many students pay less in tuition than the cost of their education, pushing enormous expenses onto the taxpayer. Finally, the federal government subsidizes college education to a massive degree even in private colleges and universities, with generous student-loan guarantees and other financial-support programs.
Government also has enormous influence over the supply side of health care and education. For instance, in primary education — again, because most children attend public schools — states and localities employ the great majority of educators. Government schools face little market pressure to improve productivity, and it is difficult to change outmoded practices and fire incompetent teachers.
Another way government shapes work-force supply (and, for that matter, demand) is through its support for credential requirements. The salary scales for government employment often automatically reward people for obtaining additional educational credentials; given the massive number of jobs in the government sector, as well as the high demand for them, this helps to stimulate the demand for post-graduate education. Requirements for credentials also restrict the supply of teachers in public schools: Even though there is reason to doubt the practical classroom value of teacher education provided at the college level, a degree in education is required for employment by most public-school systems.
The entry of new providers into the health-care sector is similarly regulated, primarily through government-mandated licensing requirements. Not surprisingly, these requirements are often manipulated by incumbent practitioners in order to restrict supply. They can make it impossible for health-care providers to improve efficiency by substituting on-the-job training for formal educational credentials. Such restrictions might make sense in the case of many physicians and nurses, but they extend to every variety of service provider in health care in ways that often undermine cost effectiveness and efficiency. For example, several years ago, the state of Maryland increased the education requirements for licensed physical therapists, insisting that anyone entering the profession hold a doctorate. This sort of requirement benefits degree-granting institutions and works to the advantage of the incumbent physical therapists, but comes at the expense of consumers.
We are left to wonder how much of the salary gap between workers with college degrees and those without is artificial. We cannot tell how salaries would be determined if government did not set its own pay scales based on educational attainment (thereby setting a standard that private employers must compete with); we cannot calculate precisely where pay levels would be if government did not enforce licensing restrictions in health care, education, and other industries. Still, it is not difficult to imagine that, if a dynamic and free market were allowed to flourish in education and health services, there might be more apprenticeships and fewer degree factories; one can easily see how salaries might be determined more on the basis of performance than of educational attainment. But because there is no true free market in these industries, the usual methods we have for evaluating such patterns simply do not suffice.
The inability to apply the usual standards and measures of our market economy to health care and education points to a broader problem with government’s domination of our new commanding heights. The unique characteristics of these sectors — and especially the fact that they are subject to so much government influence and control — make it very difficult to assess them using the terms in which we usually describe our economy. It is therefore nearly impossible to compare them to other sectors, to distinguish success from failure, and to make informed consumer choices.
In health care, education, and government work, concepts like economic value, efficiency, productivity, and consumer preferences are obscured. And as these sectors continue to grow more central to our economy in the years ahead, our broader economy will therefore become more difficult to analyze and understand in traditional market terms.
THE NON-MARKET ECONOMY
To be sure, part of this shift is inherent to the nature of the “softer” sectors that now make up the commanding heights of American economic life. The “products” of education and health care are less tangible than those of heavy industry or agriculture, making them more difficult to measure and quantify. We can quantify the inputs — the workers and equipment used in medical procedures or in teaching — but their effects on outcomes are notoriously difficult to judge.
Nobel Prize-winning economist Kenneth Arrow is usually credited with first noting the information asymmetries that characterize health care — for instance, the basic fact that it is often difficult for a consumer to know whether the treatment provided by his physician is correct. Doctors, after all, have expertise that patients lack. Similarly, in education, consumers typically have to trust that the educators involved in selecting a curriculum, and the teachers who are delivering it, know what they are doing. But it is not simply that consumers may lack knowledge about the value of educational or medical services: The providers themselves suffer from biases and large gaps in their knowledge. Thus, economic decisions and the allocation of both private and public resources in these sectors are often poorly informed.
Many careful studies show little or no value added from increased health-care expenditures or additional resources devoted to education. It is true, of course, that people in the United States, and indeed all over the world, are healthier today than they were one or two generations ago; the average lifespan appears to be increasing at a rate of about three months per year, or about 2.5 years per decade in this country. And it’s not just longevity: Overall health quality throughout one’s life has improved as well. Robert Fogel has shown that the quality of life of people in their sixties is much improved in recent decades. The average number of chronic illnesses for those over age 65 has fallen dramatically. And these improvements in longevity and general health have had an enormous impact on well-being. Economists Kevin Murphy and Robert Topel estimated the economic value of the measurable improvements in health in the United States from 1970 to 2000 at about half of the value of the increase in GDP over that same time period.
Nonetheless, although there is no doubt that health has improved considerably over time, it is not clear how much of this gain can be attributed to health care. Indeed, much of the improvement comes from better nutrition, less dangerous work environments, better sanitation, and other public-health measures, as well as more informed and health-conscious citizens. The marginal impact on health outcomes of increased spending on medical procedures appears to be low. Numerous studies, including the RAND health-care experiment, find no significant effect on outcomes for more intensive medical treatment in similar groups of patients. The Dartmouth Atlas project has documented large differences in health-care spending through Medicare across different regions of the country, with no apparent effect on outcomes.
A similar pattern prevails in education. Looking at population averages, the correlation between schooling and earnings is unmistakable. Regions and countries with higher average levels of schooling have higher per capita incomes. Work by Claudia Goldin and Lawrence Katz has recently shown that, within the United States, the gap in average earnings between workers with college education and those without is large and apparently growing.
But this correlation does not demonstrate causation, and the true impact of spending on outcomes in education is as elusive as it is in health care. In 2006, the United States spent close to 40% more per student on K‑12 education than the average of the 34 developed nations of the Organization for Economic Cooperation and Development. Yet according to student-performance rankings in science, reading, and math, the U.S. ranks far below the OECD average. And in higher education, Stacy Dale and Alan Krueger have found that, controlling for factors that could be observed prior to college attendance, the actual choice of college (i.e., the choice between a very expensive and a far less expensive school) makes little or no difference in subsequent earnings.
This is true even of education programs specifically meant to compensate for the many obstacles — poor home or community environment, low income, neglect, a lack of early-childhood schooling, and so forth — that are often blamed for gaps in later income attainment. Summarizing a large body of research, James Heckman wrote in 2005,
