Letter from America: Inequality in America

from the Royal Economic Society Newsletter,
April 2002

There is more income inequality in America than in any other major rich country, and it appears that Americans like it that way. Or if they don’t like inequality for itself, and many prominent American economists argue that, indeed, inequality is of no importance in itself, they much dislike the government tax and transfer programs that would be required to reduce it. The current administration in Washington appears to subscribe to these views, believing in the general benefits of tax cuts even if, in the short run—and perhaps beyond that—they primarily benefit the well-off. Yet income inequality is perhaps not the most significant of the many inequalities that characterizes American life, and it is not the only kind of inequality that shows signs of increasing in the six months since September 11th.

The consequences of income inequality have become a major focus of empirical research, not only among economists, but also among scholars in political science, sociology, and public health. One focus of interest is whether inequality is bad for your health. Across the states and cities of the United States, there is only a weak relationship between average income and mortality rates, in spite of the fact that data on individuals show that people in the top five percent of the income distribution have about 25 percent longer to live at all ages than those in the bottom five percent of the income distribution. Yet this relationship between income and mortality appears not to hold among the aggregate populations of cities or states. Instead, the strong correlation is between income inequality and mortality, rather than between mean income and mortality. The southern states of the US—Louisiana, Mississippi, Alabama, Georgia, and South Carolina—are particularly dangerous places to live. Worst of all (even though it is not really a state) is the nation’s capital, the District of Columbia, whose age-adjusted mortality rate is half again as high as that of (for example) Utah, a remarkably healthy state, populated to a large extent by tee-total, non-smoking, and caffeine-shunning Mormons. The District of Columbia has a more unequal distribution of income than any of the states, followed closely by Louisiana, Mississippi, and Texas. The southern states are also relatively poor—in the US, with weak social safety nets, poorer states are more unequal, something that is not true, for example, across the provinces of Canada, where a strong safety net generates a negative association between poverty and inequality—but it is their income inequality, not their levels of income, nor their poverty or unemployment rates, that is correlated with mortality rates.

Anyone who has been to DC, and seen its rich white northwestern segment surrounded by poor largely black neighborhoods that make up the rest of the central city, may suspect that the relationship between mortality and income has something to do with race, a suspicion that is reinforced by the peculiar status of the South. Inequality between the races has perhaps more to do with mortality than inequality in incomes. Indeed, this is an important part of the story. Blacks die younger than whites in the US, even when we control for differences in income and education. (Parenthetically, Hispanics live longer than either group, the so-called “Hispanic paradox.”) Indeed, there is a 20 year difference in life expectancy between whites in the richest counties of the US and blacks in the poorest counties. Because blacks also earn less than whites, places with a substantial black population are places with high income inequality and high mortality rates. But there is more to it than that. When we look at each race separately, both whites and blacks die younger in places where there is a larger fraction of blacks in the population. And once we allow for these effects of race, there is no longer any association between income inequality and mortality.

Apart from having replaced one kind of inequality for another, the mechanisms behind these correlations remain obscure. Perhaps, as is often claimed, the American health care system discriminates against blacks so that, for example, hospitals in predominantly black areas are of poorer quality. If so, not only do blacks die younger, but so do the whites who have a heart attack near a “black” hospital. It is also possible that racial diversity poisons the atmosphere, inhibiting the production of local public goods, or more generally, limiting the formation of social capital. Indeed, standard indexes of social capital, which incorporate measures of trust, are negatively influenced by both income inequality and racial diversity, more so by the latter. If racial discrimination does indeed lie at the root of these findings, the inequalities that it engenders hurt not only those who are discriminated against, but also the discriminators.

Another form of discrimination, by citizenship rather than race, has sharply increased in the wake of the barbarities of September 11th. Under the USA Patriot Act of October 2001, the US authorities can detain indefinitely, without trial and without acknowledgment, anyone suspected of terrorism who is not an American citizen, including the 20 million or so people who are long-term, tax-paying, legal residents of the US. To take an example close to home, economics departments in the US are now heavily populated by non-Americans. In my own department at Princeton, half of the faculty were born outside in the US, and of the younger (untenured) faculty, more than three-quarters were born outside the US. Universities, as well as other employers, must now open their personnel files–or indeed “any tangible things (including books, records, papers, documents, and other items”—on non-citizens to the FBI on request; they are not only not required to seek the target’s permission, they are prohibited from doing so. Indeed, they are legally bound “not to disclose to any other person that the FBI has sought or obtained tangible things under this section.” Such measures, as well as the suspension of habeas corpus for non-citizens, are widely supported among Americans, at least for the time being. Protests are dismissed as helping the enemy, and Attorney General John Ashcroft likes to say that terrorists deserve no protection. Such measures are certainly bad for the health of the economy, and of the academy, if not of the population as a whole.

Editor’s Note: The issues discussed in this letter, and more broadly the effects of social and economic arrangements on health, will be discussed by a group of American and British economists and epidemiologists (Angus Deaton, Timothy Besley, Sir Michael Marmot, Andrew Oswald, James Smith, Adam Wagstaff , and Richard Wilkinson) at this year’s Economics Section of the British Association’s Festival of Science, to be held at Leicester University on September 12th.

Angus Deaton's Letter from America appears every six months in the Royal Economic Society's Newsletter. For more information, visit http://www.res.org.uk/society/newsletters.asp.

Return to Angus' full list of Letters from America.