The Divisions That Tighten the Purse Strings

by Eduardo Porter, The New York Times, April 29, 2007

MANY Americans are skeptical about government spending on social programs, and they cite a litany of familiar reasons: big government programs aren’t effective, they are vulnerable to waste and abuse, and they run counter to the libertarian, self-reliant spirit of the nation’s founders.

But a growing body of research suggests that America’s antipathy toward big government has another, less-often-acknowledged underpinning: the nation’s racial and ethnic diversity.

Recent studies by economists and other social scientists have found that this mix tends to undermine support for government spending on “public goods” of all types, whether health care, roads or welfare programs for the disadvantaged.

Some of these studies suggest that America’s rich diversity — not only ethnic and racial but also religious and linguistic — goes a long way toward explaining why government spending on social welfare programs is much lower than it is in the more homogeneous nations of Europe. Other studies have found that within the United States, local support for various types of public spending falls as diversity rises.

“Racial divisions and ethnic divisions reduce incentives for people to be generous to others through social welfare,” said Alberto Alesina, a professor of economics at Harvard. “This is very unfortunate. But as social scientists, we can’t close our eyes to something we don’t like.”

In America, government spending on social transfers — everything from food stamps and unemployment insurance to health care and pensions — is about a third less than it is in Italy, France or Belgium, when expressed as a share of the economy, according to data from the Organization for Economic Cooperation and Development. And it is about half the level of Sweden’s. Moreover, Americans pay less in taxes than the citizens of nearly every other wealthy nation in the O.E.C.D.

In their 2004 book, “Fighting Poverty in the U.S. and Europe,” Mr. Alesina and Edward Glaeser, another Harvard economist, applied statistical regression techniques to correlate data on government spending with data on racial, ethnic, linguistic and religious diversity in Western Europe and the United States. The professors concluded that about half the gap between Europe and the United States in public spending on social programs could be explained by America’s more varied racial and ethnic mix. (They said that much of the rest resulted from stronger left-wing parties in Europe.)

As William Julius Wilson suggested in his 1996 book, “When Work Disappears: the World of the New Urban Poor,” many white Americans turned against spending on welfare during the 1970s because they thought that it mostly served blacks. “White taxpayers saw themselves as being forced, through taxes, to pay for medical and legal services that many of them could not afford for their own families,” Mr. Wilson wrote.

In the relative homogeneity of Sweden, by contrast, most taxpayers are confident that social spending programs will be directed to people much like themselves.

This doesn’t mean Americans are stingy. In fact, they contribute much more than Europeans to charity, selecting who they want to help. “It’s not that Americans are bad guys,” Mr. Glaeser said. “They just want to target it.”

But in drawing on a wide range of data like population surveys and patterns of municipal spending, researchers have found ample evidence of how ethnic and racial diversity has undermined support for spending on social welfare in the United States.

In a study in 2001, Erzo F. P. Luttmer, an associate professor at the Kennedy School of Government at Harvard, reported that the percentage of people who say they support welfare spending decreases as the share of local recipients from their own racial group falls. His report was based on data from the General Social Survey, a social-attitudes poll conducted across the United States nearly every year since 1972.

In another study, published in 1996, James Poterba, a professor of economics at the Massachusetts Institute of Technology, found that public spending on education falls as the percentage of elderly people in a given area rises. The reduction, he found, “is particularly large when the elderly residents and the school-age population are from different racial groups.”

In a 1997 study, Mr. Alesina, along with Reza Baqir, an economist at the International Monetary Fund, and William Easterly, an economics professor at New York University, looked at the relationship between social spending and ethnic diversity in 2,700 cities, counties and metropolitan areas across the United States.

They found that in more diverse cities and counties, the share of local government spending on public goods — in this case, roads, sewage treatment, trash clearance and education — was generally lower than it was in more homogeneous localities. “Our results are consistent with the idea that white majorities vote to reduce the supply of productive public goods as the share of blacks and other minorities increases,” they wrote.

Of course, there are some exceptions to the pattern. New York and California, for instance, are among the most highly taxed and most diverse states, although there is evidence that racial and ethnic tensions have whittled away at support for public spending. In California, to cite just one example, Proposition 187, barring illegal immigrants from access to public services, passed in 1994 on the support of 63 percent of whites, despite the opposition of four of every five Latinos.

New York’s history also has examples of ethnic tensions. But New York City, in a way, is a special case: so diverse that no single ethnic or racial majority controls the public purse. “There is no cleavage between an Anglo majority and some poor minority,” Mr. Glaeser said. “In New York, everybody is a minority.”

New York City’s experience, in fact, underscores that diversity does not automatically lead to hostility among ethnic groups or toward government spending as a whole. From public education to intermarriage to the many institutions in civil society promoting mutual understanding, there are countervailing forces acting to overcome ethnic, religious or linguistic cleavages.

Ethnic diversity doesn’t inevitably reduce spending on public goods. Rather, spending tends to fall when elected officials choose to run and govern on platforms that heighten racial and ethnic divisions. Over the long term, governments usually find it in their interest to bridge the centrifugal forces of diversity rather than to exploit them, if only to promote stability.

STILL, there is little reason to believe that the racial, ethnic, religious and linguistic antagonisms that have eroded support for social welfare programs in the United States are likely to abate any time soon. Indeed, the arrival of hundreds of thousands of illegal immigrants a year from Latin America seems to be sapping support for public welfare.

Last year, laws were enacted in Arizona, Colorado, Rhode Island and Hawaii barring illegal immigrants from access to some government programs. This month, the State Senate in Oklahoma overwhelmingly passed legislation that would bar illegal immigrants from receiving public benefits.

And these restrictive attitudes can easily turn against spending on domestic programs as a whole. Representative Tom Tancredo, a Colorado Republican running for president on an anti-immigration platform, says that Americans pay too much in taxes and that the Internal Revenue Service should be abolished. His candidacy may not prosper, but the issues on which he is running are likely to be around for some time.

Mr. Alesina certainly expects further conflicts. “One can expect public support for public goods to erode further,” he said. “Public spending in law and order might not go down. What would go down is spending on redistribution.”