In a reversal of earlier trends, death rates among white non-Hispanic Americans in midlife increased sharply between 1999 and 2013, according to a new study by economists Anne Case and Angus Deaton, winner last month of the Nobel Prize for economics. The increased deaths were concentrated among those with the least education and resulted largely from drug and alcohol “poisonings,” suicide, and chronic liver diseases and cirrhosis. This midlife mortality reversal had no parallel in any other industrialized society or in other demographic groups in the United States.
Case and Deaton’s analysis, published today in the Proceedings of the National Academy of Sciences, also shows increased rates of illness, chronic pain, and disability among middle-aged whites. The findings have important implications for American politics and public policy, particularly for debates about economic inequality, public health, drug policy, disability insurance, and retirement income. The data also suggest why much of American politics may be taking on an increasingly harsh and desperate quality.
The recent divergence in death rates between the United States and other rich countries is striking. Between 1979 and 1999, Case and Deaton show, mortality for white Americans ages 45 to 54 had declined at nearly 2 percent per year. That was about the same as the average rate of decline in mortality for all people the same age in such countries as France, Germany, the United Kingdom, and Sweden. (See figure below.) After 1999, the 2 percent annual decline continued in other industrialized countries and for Hispanics in the United States, but the death rate for middle-aged white non-Hispanic Americans turned around and began rising half a percent a year.
The White Midlife Mortality Reversal: All-cause mortality, ages 45–54 for U.S. white non-Hispanics (USW), US Hispanics (USH), and six comparison countries: France (FRA), Germany (GER), the United Kingdom (UK), Canada (CAN), Australia (AUS), and Sweden (SWE).
The consequences of this divergence have been staggering. If the white midlife mortality “had continued to decline at its previous (1979‒1998) rate,” Case and Deaton estimate, “half a million deaths would have been avoided in the period 1999‒2013, comparable to lives lost in the U.S. AIDS epidemic through mid-2015.”
Case and Deaton’s data indicate that the white midlife mortality reversal was due almost entirely to increased deaths among those with a high school degree or less. Mortality rates in that group rose by 134 per 100,000 between 1999 and 2013, while there was little change among those with some college, and death rates fell by 57 per 100,000 for those with a college degree or more.
Death rates from suicide and poisonings such as drug overdoses increased among middle-aged whites at all socioeconomic levels (as measured by education). But the increases were largest among those with the least education and more than sufficient in that group to wipe out progress in reducing other causes of death. Deaths from diabetes rose slightly but did not account for a significant part of the white midlife mortality reversal.
None of this is to suggest that whites are generally in worse health than blacks. Among blacks, midlife mortality has been higher than among whites. But over the period 1999-2013, according to Case and Deaton, midlife mortality declined by more than 200 per 100,000 for blacks while it was rising for whites. As a result, the ratio of black to white mortality rates dropped from 2.09 in 1999 to 1.40 in 2013. Contrary to what many Americans may still believe, drug overdoses are no longer concentrated among minorities; in fact, among the 45-54 age group, drug-related deaths are now higher among whites.
Case and Deaton also cite evidence of declining mental and physical health among whites ages 45 to 54 between 1999 and 2013. According to national surveys, mental illness rose. There were significant increases as well in the percentages reporting poor health, chronic pain, and difficulties with such activities as walking a quarter mile, climbing ten steps, and socializing with others. The percentage reporting themselves unable to work doubled.
These findings suggest that something has gone deeply wrong for middle-aged white Americans—changes that have hit the less educated especially hard in the 21st century. Previous studies have pointed to rising economic pressures on low- and middle-income workers that have been especially acute for those with only a high school education. On the right, in a 2012 book, Coming Apart: The State of White America, 1960-2010, Charles Murray argued that a decline in moral virtue since the 1960s has led to the deterioration of life among low-income whites. For example, according to Murray, low-income whites work fewer hours than they used to and are more likely to be out of the labor force because they have abandoned the work ethic and prefer, in Murray’s words, “goofing off.” Although educated Americans in the upper-middle class have been working more and doing well, Murray contends that “they have abdicated their responsibility to set and promulgate standards.” For Murray, the ultimate cause of these developments is the adoption of a European-style welfare state, which he maintains undermines the moral virtues at the core of what he calls the “American project.”
To conservatives, the white midlife mortality reversal in the United States may initially seem to confirm Murray’s argument about moral decay caused by the welfare state. But that interpretation runs into an obvious objection: Similar trends are not evident in the European countries that have even more generous systems of social protection than the United States does.
Although Case and Deaton are cautious about interpreting the data, they single out two possible causes of the mortality reversal. The first relates specifically to the timing of increased drug-related deaths: the introduction and ready availability of opioid prescription painkillers (such as Oxycontin) beginning in the late 1990s, followed by a shift to heroin, both directly linked to rising death rates among whites over the 1999-2013 period. But it is not clear, Case and Deaton point out, whether rising drug use is a response to an “epidemic of pain,” or whether the introduction and distribution of new prescription painkillers played an independent, causal role. One way or the other, however, Case and Deaton’s study puts in bold relief the sheer magnitude of the consequences of today’s drug plague.
A second potential cause highlighted by Case and Deaton (and possibly related to the first) is stress from economic change resulting from slower economic growth and rising inequality. “Many of the baby-boom generation,” they note, “are the first to find, in midlife, that they will not be better off than were their parents. Growth in real median earnings has been slow for this group, especially those with only a high school education.” But they also observe that some other rich countries have seen “even slower growth in median earnings than the United States, yet none have had the same mortality experience.”
Here is where the stronger systems of social protection in other countries may play a role in both reducing inequality and cushioning people from the adverse social psychological consequences of wage stagnation. One key difference potentially affecting people in midlife, as Case and Deaton point out, is that the other rich countries have maintained defined-benefit pensions, while employers in the United States have shifted increasingly to defined-contribution pensions (such as 401(k) plans) that do not provide the same degree of security. As a result, many Americans with only a high-school education not only lack the skills in midlife to find good jobs or even to stay employed but also face the likelihood of destitution in old age.
These trends put new light on current debates about disability insurance and retirement policy. Contrary to those like Murray who attribute the growth in Social Security Disability Insurance to a decline in the work ethic, Case and Deaton’s data suggest that the increased number of beneficiaries reflects a real deterioration of health in middle age. Raising the Social Security retirement age may seem to be no problem for the educated and affluent who are in good health and do little physical labor, but delaying retirement poses a much bigger problem for workers who are experiencing increased burdens of pain and disability in midlife.
The declining health of middle-aged white Americans may also shed light on the intensity of the political reaction taking place on the right today. The role of suicide, drugs, and alcohol in the white midlife mortality reversal is a signal of heightened desperation among a population in measurable decline. We are not talking merely about “status anxiety” due to rising immigrant populations and changing racial and gender relations. Nor are we talking only about stagnation in wages as if the problem were merely one of take-home pay. The phenomenon Case and Deaton have identified suggests a dire collapse of hope, and that same collapse may be propelling support for more radical political change. Much of that support is now going to Republican candidates, notably Donald Trump. Whether Democrats can compete effectively for that support on the basis of substantive economic and social policies will crucially affect the country’s political future.