SAN FRANCISCO — This city is full of parks that invite exercise and bike lanes that make commuting a workout. It's home to social services that tend the poor, and taxpayers who willingly fund them. Smoking is banned at restaurants and bars — as well as in workplaces, at bus stops, throughout public housing, at charity bingo games and even inside stores that sell tobacco.
These factors may help explain why the poor live longer in the San Francisco area than they do in much of the rest of the country. According to a large study published today about how income and geography shape life expectancies, a poor person living in the San Francisco area can expect to live about three years longer than someone making the same income in Detroit. That difference is equivalent to how much national life expectancies would rise if we eliminated cancer.
"If you think about the cancer comparison, having cancer is not just about having a shorter life. It's also about having an unhealthier life, a much lower quality of life," says Stanford economist Raj Chetty, the lead author of the study, published in the Journal of the American Medical Association. Seven economists, from MIT, Harvard, the U.S. Treasury and McKinsey and Co., were co-authors.
Think about their new data as if the poor in Detroit get cancer and the poor in San Francisco don't. "Then," Chetty says, "you can see that this is a big deal."
The research, based on the tax and Social Security records of everyone in America between 1999 and 2014 with a valid Social Security number and earnings, gives the most precise look yet at a pattern that has long troubled health experts: In America, the richer you are, the longer you live. But what's especially striking is that the poor live even shorter lives in some places than others. They have longer life expectancies in affluent cities with highly educated populations, such as San Francisco, New York and Los Angeles. Among the 100 largest commuting zones ranked by the researchers, six of the top eight for low-income life expectancies are in California, a state with a strong safety net and a history of regulating where you light your cigarette or what comes from your car's tailpipe.
The poor live shorter lives in Las Vegas, Louisville and industrial Midwest towns, such as Gary, Ind. Geography also matters much more for the poor than the rich. The health behaviors of the wealthy are similar wherever they live. For the poor, their likelihood of risky behaviors such as smoking depends a great deal on geography, on whether they live in a place where smoking is common or where, as in San Francisco, cigarettes have been shunted out of view.
The findings pair with earlier research led by Chetty, using the same fine-grained tax data, that illustrates how crucial life outcomes — including economic mobility, adult earnings and now health — vary greatly with geography in America.
"It is as if the top income percentiles belong to one world of elite, wealthy U.S. adults," Nobel Prize-winning economist Angus Deaton writes in a JAMA editorial accompanying the new paper, "whereas the bottom income percentiles each belong to separate worlds of poverty, each unhappy and unhealthy in its own way."
Overall, the new study offers the most exhaustive account yet of the rich-poor gap in American life expectancy. The data reveal that life expectancies continuously rise with income in America: The modestly poor live longer than the very poor, and the super-rich live longer than the merely rich.
For men, the gap between the top and bottom 1 percent nationwide is nearly 15 years. For women, it's 10 years. And these disparities have widened since 2000. People in the top 5 percent have gained about three years of life expectancy. People at the bottom have gained almost nothing.
The economists don't have definitive explanations for the relationships they observe between income and health. It's not that more money alone buys better health. Poor health can lead to lower incomes, just as low incomes can lead to poor health. And it's hard to imagine, for instance, that the richest Americans can buy things that make them healthier that people who are just slightly less rich can't afford, says Harvard economist and co-author David Cutler.
But among the poor, the differences in what money can buy probably explain an important part of the longevity gap.
"At some point, it does become causal: You can buy better housing, you can buy better food," Cutler says. "You own a car that actually works, so you can drive to the grocery store more regularly."
But it's also true that the stress of poverty can have lifelong health consequences.
"An emerging body of science is showing that adversity itself will shorten your life span," says Nadine Burke Harris, a pediatrician in San Francisco and a founder of the Center for Youth Wellness, which tries to treat the role of toxic stress in undermining health in the city's poorest neighborhoods. Her patients, who deal with adversity from neighborhood violence to childhood abuse, come in with high rates of asthma and cardiovascular disease, or even repeated cases of uncommon autoimmune diseases.
Her health center — overwhelmingly funded by local philanthropy — is precisely the kind of resource that's harder to find outside of an affluent city like San Francisco or New York.
What's driving early death
Across the country, shorter life expectancies for the poor, measured at age 40, were most closely correlated with places that had lower exercise rates and higher rates of smoking and obesity. For both men and women, stretches of the industrial Midwest promised the lowest life expectancies. Some of the results also align with the findings of a recent Washington Post analysis of mortality rates that found deteriorating health, particularly for whites, outside big cities in rural America where unhealthy behaviors are on the rise.
Below is a map of where poor women live the longest and shortest across the United States. Poor men's life expectancies are mapped above.
Across these maps, places with high shares of college graduates, high population density, high home values and high government expenditures per capita were correlated with better life expectancies for the poor. In some ways, this finding is surprising: Health experts suspect that higher inequality may be connected to poor health. But the authors found little correlation between income inequality and life expectancy. And wealthy San Francisco, New York and Los Angeles, among the places in America with the widest income inequality, look here like the healthiest.
Burke Harris offers her own caveat to that finding: "Poor people only have better health status when they’re living in a wealthy environment if the people in that wealthy environment believe in investing in the safety net."
That is certainly true in San Francisco, a city with both private wealth and major public expenditures. To the extent that fostering health requires money — either to fund a clinic like Burke Harris's or to invest in parks and city services — San Francisco has it.
Rajiv Bhatia, the former environmental health director in San Francisco's Department of Public Health, points to other policies the city has pursued, such as universal preschool, paid sick leave and paid parental leave. These policies aren't aimed at shaping health behaviors in the same way that a smoking ban is, but they may well ease the stresses associated with poverty that affect health, too.
Policies like these, and New York's universal pre-K, are notably the result of more than the money needed to fund them. They're also the product of a particular idea about the role of government in supporting the poor and shaping behaviors linked to health.
"It's no coincidence that New York City banned trans fats," Cutler says. "New York is a very rich city. People there feel like, 'Look, socially it’s not doing anything for us, we know it’s harmful. Why not get rid of it?'"
New York Mayor Michael Bloomberg also famously tried to ban oversize sodas. He pushed for calorie labels on restaurant menus, higher taxes on cigarettes and tighter restrictions on buying them. He even issued an executive order trying to nudge city employees to take the stairs more.
Critics outside of New York called Bloomberg's public-health policies draconian, and less out-there ideas like expanding Medicaid have been met with public skepticism, too. But in the new research, New York City ranks as the place where the poor have the longest life expectancies, ahead of all the California cities. Life expectancies between the rich and poor there still vary widely. And it's still better to be rich in Detroit than poor in New York. But something about the environment there, as in San Francisco, is counteracting at least some of the massive national disparities by income.
Perhaps in highly educated cities people are savvier about health, Cutler suggests, more likely to advocate for smoking bans or to push doctors to adopt the latest medical advance. Or perhaps it's that more educated, wealthier cities also tend to be more politically progressive, believing in the power of government to make people healthier and to shape what are fundamentally personal decisions about lifestyle. New York and San Francisco also are more walkable and transit-friendly, factors that have been associated with health as well.
Burke Harris suggests there is a lot going on in San Francisco that the economists did not directly measure. "It’s the living wage," she says. "It’s the greening of the city, it’s the bike lanes, it’s the universal preschool. It’s all of these things that together add up."