Democrats and Middle America: What’s the Real Problem?

Why won’t the Democrats learn from Kansas?

Why is it so hard for Democrats to get votes from people whose economic interests should lead them to support the party? Is Hillary Clinton right to identify the root cause as the party’s lack of credibility as a protector of national security and to seek a remedy in a hawkish approach to foreign policy? Is the problem primarily cultural, a feeling that liberal elites are hostile to the values and religious beliefs of ordinary Americans? Or is it simply that Democrats have pursued an economic agenda of free trade and deregulation—against the interests of working people—leaving the party defenseless in the face of Republican demagoguery on whatever issue comes to hand?

Debate on these questions was reinvigorated by Thomas Frank’s best-selling What’s the Matter with Kansas?, which asks why attitudes of ordinary Kansans have shifted so far to the right. With Democrats unresponsive to the economic troubles of Middle America, Frank asserts, cultural anti-elitism becomes the outlet for resentments engendered by growing hardship. This call for a renewed populism has aroused a storm of objections. Some critics doubt that working-class voting habits have changed. Others accept the reality of the shift, but take issue with Frank’s diagnosis of the cause. Still others, the most numerous, do not dispute Frank’s account of the facts, choosing instead to challenge the legitimacy of any effort to change those facts.

The challenge to Frank’s facts comes in a widely circulated paper by Princeton political scientist Richard Bartels, who cites polls showing that Democrats have lost more votes in the last fifty years in the upper third of the income distribution than in the lower third. (Bartels’s figures are for white voters outside the South.)

Bartels’s argument is based on a flawed definition of the working class. Whether one seeks the working class of Marxist theory or the ordinary people of American populism, they are to be found in the middle of the income distribution and not just at the bottom. Indeed, polling experts David Gopoian and Ralph Whitehead found that only 19 percent of the bottom third voters in Bartels’s sample were over thirty years of age and actively working. Thirty-five percent were retired, and substantial numbers were disabled or unemployed.

Gopoian and Whitehead point out that only 40 percent of whites with less than a college education voted for Kerry in 2004. This is a closer approximation to the white working-class vote than Bartels’s; whites without college degrees make up nearly half the electorate. But it still tells only part of the story.

Class voting patterns are hard to sort out because income is only one determinant of social class. Education, religion, occupation, and residence all matter. In recent decades, education and income have come to work at cross-purposes. Among voters with the same level of education, those with more income are more Republican; amon...