The Unequal Distribution of Death
America’s Great Divergence is caused by economic
forces. But it is having profound effects outside the economic realm. The labor
market differences between the three Americas have become so large that they
are now generating a growing divide in many other aspects of our private and
public life. We’ll look here at four striking examples: health and longevity,
family stability, political participation, and charitable giving.
Life expectancy is one of the best available
measures of people’s health and overall well-being. It reflects not just
genetics but lifestyle, economic circumstances, and many other factors. Map 4
shows just how different male life expectancy is across U.S. counties. The East and West Coasts, together with parts of the
northern Plains, tend to have higher than average life expectancy, while the
South and Appalachia tend to have lower than average life expectancy. And even
within each region, there is wide variation.
These differences are not surprising, per se.
There is no country in the world with the same life expectancy in all of its
regions. But what is striking about the United States is the magnitude of the
differences. Male residents in counties with the longest life
expectancies—Fairfax, Virginia; Marin and Santa Clara (where most of Silicon
Valley is located), California; and Montgomery, Maryland—tend to live until
they are about eighty-one years old. By contrast, male residents of counties
with the lowest life expectancies tend to die at age sixty-six. In other words,
the typical man in Fairfax lives fifteen years longer
than the typical man in Baltimore, just 60 miles away. The gap for women is
equally large. This degree of geographical inequality in life expectancy is
truly staggering, and a comparison with other countries indicates that it’s
substantially larger than what we see in Canada, the United Kingdom, and Japan,
presumably because the economic divide between our communities is deeper than
that in other countries.
Incredibly, a county like Baltimore has a life
expectancy well below that of developing countries such as Paraguay and Iran.
Indeed, if the bottom 10 percent of U.S. counties comprised a stand-alone
country, that country would have a male life expectancy of 69.6 years and would
appear very low in international rankings, squeezed between Nicaragua and the
Philippines, well below China and Mexico. By contrast, if the top 10 percent of
U.S. counties were a stand-alone country, it would rank near the top of
international comparisons, just below Japan and Australia. (The United States
as a whole ranks thirty-sixth. Although Americans spend twice as much on health
care as residents of other industrialized countries, their average life
expectancy is significantly lower than that of people in many other rich
countries.)
Probably the most remarkable fact about life
expectancy trends in America is that the vast geographical differences are not
fading with time. Instead, they increase with every passing year, reflecting
and possibly exacerbating the effect of growing socioeconomic differences. When
David Breedlove, the Silicon Valley engineer mentioned in the introduction,
moved from Menlo Park to Visalia in 1969, life expectancy in the two
communities was comparable. Today life expectancy in San Mateo County, where
Menlo Park is located, is almost six years longer than in Tulare County, where
Visalia is located—a remarkable change.
The rising inequality in life expectancy among
American communities is shown in Figure 7, which plots gains in male life
expectancy since 1987 for the ten counties with the highest and lowest life
expectancies in each year. Between 1987 and 2007, the
top group gained 5.8 years while the bottom group gained merely 1.8 years. The
end result is that the gap in life expectancy for the top ten counties and the
bottom ten counties is much larger today than it has been in decades.6
The geographical sorting of individuals with
different educational and income levels is likely to exacerbate the longevity
differences resulting from these disparities. The reason is simple: poorly
educated individuals who live in a community where everyone else has low levels
of education are likely to adopt less healthy lifestyles than poorly educated
individuals in a community where there is a mix of educational and income
levels. Economists call this a social multiplier effect.
For example, the probability of a person’s
smoking or exercising is apt to depend both on her own traits and on whether
those around her smoke or exercise. The economist and former air force officer
Scott Carrell has measured the importance of the social multiplier effect using
data on physical fitness of members of the U.S. Air Force Academy. Members of the Air Force Academy are randomly assigned
to squadrons of approximately thirty individuals with whom they are required to
spend the majority of their time. The randomization of the assignment makes
Carrell’s data particularly useful, because it allows researchers to measure
the causal influence of peers separate from all other
possible confounding factors. Carrell and his coauthors have found definitive
evidence that individuals who are assigned to a squadron where others are less
fit tend to become less fit over time. The effect is very strong: poor fitness
spreads like a contagious disease, with the largest effects caused by peers who
are the least physically fit. The Yale economist Jason Fletcher has found
similar effects for smoking. Increasing the number of
smokers in a person’s social network by 10 percent increases the likelihood
that that person will smoke by approximately three percentage points. (As a
former smoker, I can attest to the fact that the urge to light up is much stronger
when I spend time in East Coast cities, where I see many people smoking outside
buildings, than in California, where I rarely see people smoking.) The
availability of nutritious food also varies greatly depending on the
socioeconomic characteristics of each community. In low-income neighborhoods,
fast-food restaurants are more prevalent and fresh food is harder to obtain
than in mixed-income communities.
The social multiplier effect matters because
it increases the difference in health between individuals with the same levels
of income and education living in communities with different average levels of
income and education. Effectively, it means that socioeconomic segregation of
the type we are now seeing has an indirect effect on people’s health and longevity
over and above the direct effect of their own education and income. This leads
to a startling conclusion: where you live has to do with how long you live.
The Moving to Opportunity program, one of the
most ambitious social experiments ever attempted in the United States, is
particularly interesting in this respect. From 1994
to 1998, the federal government gave thousands of public housing residents in
Baltimore, Chicago, Boston, New York, and Los Angeles vouchers to leave the
projects and move to private housing in the same city but in significantly
better neighborhoods. Like the Carrell study, this was a randomized experiment,
with 1,788 families randomly selected to receive the voucher and 1,898 families
randomly assigned to the control group. When researchers visited the two groups
ten years later and measured their health, their findings were astonishing.
Although before the experiment the two groups were identical, the group that
had received the vouchers and moved to better neighborhoods was in
significantly better physical shape. Its members had improved their diet and
were exercising more. They had a significantly lower incidence of obesity,
diabetes, and depression. In general, they were healthier and happier, the
effect being particularly strong for young women. There are many possible
explanations for these findings. But a plausible one is that the place where we
live and the people who surround us play an important role in shaping our
health.
Aucun commentaire