« A Reply to Gopoian and Whitehead | Main | Vets Boost Dems '06 Chances »

Gopoian and Whitehead Reply to Bartels

By David Gopoian and Ralph Whitehead, Jr.

We have checked our data and stand by the findings we reported. Our sample consists of whites who voted and have family incomes under $35,000. Whatever discrepancies appear between Bartels's descriptions and ours must reside in unknown differences between his coding of the data set and ours.

We'd also note that Bartels has leapt to a conclusion as to what we regard as the electoral-strategy implication of our attempt to flesh out the demographic characteristics of his 'white-working class.' To be fair to him, our original post failed to offer an explicit statement of the implication as we see it. By failing to do so, perhaps we were inviting him to get it wrong. So here goes:

We don't think that the Democratic Party can become a majority party unless it improves its share of the votes cast by (among others) people who fit this profile:


Ages 25 through 62

Attached to the work force, or sharing a household with someone who is attached to the work force


Nonunion, and living in a nonunion household

These voters are a significant presence in near-miss states like Nevada, Iowa, Ohio, and New Mexico, as well as in slipping-away states like Florida, Missouri, Louisiana, and West Virginia.

In order to do better with these voters, we would argue, the Democratic Prty doesn't necessarily have to give ground on social issues. But it does has to offer them a much more sweeping and aggressive economic agenda. This agenda must respond to the cumulative crises that beset those who hold a strong stake in the world of work, and aren't necessarily well-equipped to be able to protect it, let alone enhance it. As it happens, however, the content of this agenda is unlikely to appeal strongly to two groups of people who don't fit the profile but are among Bartels's white people with household incomes of less than $35K -- and, more to our point, are already drawn to the Democrats in ways that the profiled voters are not:

Those who are old enough to be eligible for Medicare and Social Security.

Young college graduates, who currently have low incomes (because they've recently entered the world of work), but have relatively good economic prospects. This is partly because they are college graduates, and are therefore on the upside of the college gap in earnings. And it is partly because their status as college graduates makes it highly likely that they are the children of college graduates, who are also on the upside of the college gap, and thus likely to have money to provide to their adult children. These young people certainly feel a stake in the world of work, but are relatively well-equipped to protect it, and even enhance it. Also, since many of them are socially liberal, the social liberalism stressed by Bartels will presumably keep them in the Democratic fold.

The voters who do fit the profile are likely to feel a need for national health insurance, unlike the people who are 62 and older and unlike those among the young college grads who currently feel bulletproof. They also have a stake in salvaging the social contract -- de facto lifetime employment, a pension, health care coverage -- or establishing a full-fledged successor to it. They have a stake in strengthening (or modernizing) the safety net: As pension coverage continues to vanish, more and more workers will have to rely on Social Security as their sole pension. (Thus, the objective economic need will be for larger Social Security checks.) Unemployment insurance was designed to tide people over during bona fide layoffs, rather than euphemized firings, and between jobs, rather than between occupations. It needs to be modernized. And the safety net will have to be extended to include long term care. (Unless the profiled voters are all able to follow the advice of a recent lead editorial in The New York Times and rely on "serious financial planning on an individual basis.") The voters who fit this profile also need to experience a sustained increase in their real earnings -- something it has been hard for them to do in recent years, and something that might well get even harder for them to do in the future.

We hate to leap beyond the evidence to impute a view to Bartels, since he has leapt beyond the evidence to impute a view to us.. But we hope he isn't implying that the Democratic Party can become the majority party merely by sticking with a low-protein economic agenda. The crucial question for Democratic electoral strategists is not: To what degree are the voting decisions of noncollege voters driven by economic concerns, as opposed to (or as well as) social concerns? Rather, it is: To what degree are Democratic issue positions clearly and strongly responsive to these economic concerns?

Also, in suggesting 'households who are in the middle of the income distribution' as a definition of 'the middle class,' he offers what seems at first glance to be an admirably simple and sensible idea. However, if we accept this definition of 'the middle class,' we are implying that a nation like Haiti OR Bangla Desh, if it has a bulge in the middle of its income distribution, has a large middle class. But this implication might be misleading. Closer to home, of course, the question is: Are the median earnings of a working-age household, some $48K, enough to enable it these days to afford a 'middle class' standard of living? Particularly if many of the people who are close to the middle of the income distribution -- such as those in the profile, and millions of otherwise similar Hispanic households and African-American households-- now have to use a growing share of this sum to plug the growing holes that have been created by the demise of the social contract, the obsolescence of the safety net, and persistently flat pay?