Do more unequal places tend to vote for Democrats?

Jim Manzi says yes, and he has some data. He says that in 46 out of 48 states, there’s a positive correlation between a county’s neighborhood-level inequality and its vote for Kerry.

P.S. Also see interesting thoughts in the comments section below.

P.P.S. This paper by Mark Frank also seems relevant to the discussion. Frank writes:

For many states, the share of income held by the top decile experienced a prolonged period of stability after World War II, followed by a substantial increase in inequality during the 1980s and 1990s. This paper also presents an examination of the long-run relationship between income inequality and economic growth. Our findings indicate that the long-run relationship between inequality and growth is positive in nature and driven principally by the concentration of income in the upper end of the income distribution.

P.P.P.S. See also the graphs here (from chapter 5 of the Red State, Blue State book).

4 thoughts on “Do more unequal places tend to vote for Democrats?

  1. Thanks for keeping us posted on this fascinating dialogue.

    Far and away the most interesting part of Manzi's post is the portion you don't mention – his equally robust finding that "found no significant relationship between change in inequality and change in vote in any sample." I found that stunning.

    So, I have to presume, did David Frum. In one memorable passage of his Times piece, he observed, "As Fairfax has evolved toward greater inequality, it has steadily shifted into the Democratic column." But if that's the case, Manzi's data provides no support for the notion of a causal link.

    The underlying problem here, I suspect, is Frum's initial formulation. By writing of a growing partisan divide between "equal" and "unequal" locales, Frum identifies the GOP with "egalitarian" principles, and by implication, suggests that the Democratic Party owes its recent ascendancy to growing inequality. It's a nifty rhetorical sleight-of-hand. It's also a red herring. Frum fails to establish any causal link, much less a causal mechanism.

    And there's a reason for that. He spends quite a few column-inches in the Times detailing all of the stresses attendant to living in the modern economy. But he conspicuously fails to link those stresses to the physical proximity of the rich and poor. Let me put that in the form of a question: If I'm the father of a middle-class family of four, struggling to pay medical bills, paying more in payroll than income taxes, and watching my inflation-adjusted income decline, why would it matter where I live? Shouldn't the stresses be the same, whether or not the folks who live nearby are getting richer? Shouldn't I be most alienated of all if I live in an area – say, Detroit – in which everyone's income is plummeting and we're all suffering together? It's a question with which Frum never bothers to grapple.

    But there's still something interesting here, even if it's not what Frum suggests. Manzi confirms the overall correlation. Gelman points out, in posts prior to this one, that this overall national correlation may well be masking some striking regional differences. So what are we to make of this?

    I'd suggest that Frum himself provides an answer. In perhaps the most richly evocative passage in his article, he writes:

    Whether in Virginia, Missouri or Illinois, there are no more egalitarian and no more Republican places in the United States than these exurbs. The rich shun them, and the poor can find no easy foothold, but the middle-income, middle-educated, white married parents who form the backbone of the G.O.P. are drawn to them as if to a refuge.

    …as if to a refuge. In other words, it has nothing to do with equality – what we're really talking about here is homogeneity. The flight from an increasingly diverse world into enclaves of carefully constructed and painstakingly policed uniformity. On a national level, these neighborhoods trend Republican. That's a regional effect showing up in national tables. The kind of development Frum describes has been most prevalent across the American sunbelt, which has experienced a tremendous sorting-out along the lines of race and income. On the coasts, however, as Gelman takes pains to note, we tend to see the opposite effect. Some of the most uniformly wealthy areas in the country are staunchly liberal. It's just that there aren't nearly as many of them, not nearly enough to compensate for the vast tracts of sunbelt subdivisions springing up across the south. The coasts and the midwest were built out long before the sunbelt, and they tend to be more classically urban. Their suburbs are often streetcar suburbs, which mix apartments and single family homes.

    So much for the correlation, which is almost certainly an artifact of preexisting cultural preferences expressed in the choice of residence, and not causally linked at all. But what of the question Frum purports to pose – the origins of the struggles of the Republican Party? Sadly, for Frum, there's simply no evidence that this trend is linked to spatial geography, much less driven by it. His piece would have been much clearer had he stuck to a simple analysis of income – reductive as it is to offer mono-causal explanations, at least shifts in income levels correlate reasonably well with changes in political preferences, and allow for a plausible causal link. But that would have left Frum to write that under Republican administrations the very rich of gained income geometrically faster than the rest of the population, and that it is this divide which accounts for much of the present mass disillusionment with his party and its policies. But that's perilously close to saying that it's the GOP which is the party of inequality, whereas citing a bunch of confusing statistics allows him to make the opposite claim. And really, that's ever so much more comforting for conservatives.

  2. Thanks for pointing to the post.


    Thanks for your very thoughtful comments.

    I tried to emphasize that none of these findings demonstrate causality one way or the other.

    I think that the lack of a time-series effect is not dispostive of lack of causation. At a minimum, it's very unclear to me over what timescale such an hypothesized effect would operate. It's certainly very plausible that it would operate over a decade or more, in which case one would not see it here. Further, it's quite plausible that there are many interacting effects that could easily mask the causal link.

    Jim Manzi

  3. One of Manzi's commentator's get at this, but I wonder if this isn't picking up that more inequality at the neighborhood level is correlated with more poor people, whom we know tend to vote Democratic.

  4. One reason I enjoyed "Red State, Blue State" so much is that it was entirely free of the type of comments made above by "Cynic". Note, this anonymous commenter is claiming not that Frum has misinterpreted the data, or even that he is mistaken, but rather that Frum is lying and knows it — throwing up a bunch of confusing statistics just to cloud the issue.

    This type of rhetoric, which is heard just as often coming from Republicans as from Democrats, is what has poisoned the well of discourse in this country.

Comments are closed.