Racial Polarization and the Challenge to the Voting Rights Act

by John Sides on February 27, 2013 · 12 comments

in Campaigns and elections,Law

There can be no doubt that the covered jurisdictions differ, as a group, from the noncovered jurisdictions in their rates of  racially polarized voting. There can also be no doubt that voting the covered jurisdictions is becoming more, not less, polarized over time.

From a new paper by Nathaniel Persily, Charles Stewart, and Stephen Ansolabehere.  It is obviously relevant given today’s arguments before the Court which, at first blush, suggest that a majority is skeptical of this aspect of the VRA.  For example, John Roberts asked:

Is it the government’s submission that the citizens in the South are more racist than the citizens in the North?

Well, there is this:

General Social Survey and National Election Studies data from the 1970s to the present indicate that whites residing in the old Confederacy continue to display more racial antagonism and ideological conservatism than non-Southern whites. Racial conservatism has become linked more closely to presidential voting and party identification over time in the white South, while its impact has remained constant elsewhere. This stronger association between racial antagonism and partisanship in the South compared to other regions cannot be explained by regional differences in nonracial ideology or nonracial policy preferences, or by the effects of those variables on partisanship.


Nicholas Valentino February 27, 2013 at 2:57 pm

What is not yet clear, however, is whether section 5 actually does any good when it comes to the current disenfranchisement efforts. It is a tricky question to answer, because (as usual) we do not have a good control group.

Steve February 27, 2013 at 2:59 pm

Race is a significant predictor of voting behavior? Big surprise! The empirical relationship between race and voting behavior in 2008 and 2013 increased? Again, big surprise. Despite some reservations regarding the standard set of control variables, I suppose it is noteworthy (in all seriousness) that there is some difference in racial polarization for covered versus non-covered areas. But in the context of a discussion regarding the VRA, this finding is hardly relevant unless it is associated with the variance in voting requirements/restrictions. Side-note: if we are worried about districts where a racial minority will be unable to elect a favored representative, should we start passing similar laws for non-covered districts with a minority of white voters?

Stephen June 27, 2013 at 12:25 am

I don’t think I like the tone and implication of your “sidenote.” This isn’t affirmative action for voting so no need to play the “defend whites” card. This is about districts with a record of disenfranchising voters, generally along racial lines, in order to maintain electoral dominance and, consequently, dominance over policy. No one is trying to give *electoral minorities dominance or preference, just the ability to exercise their constitutional right to vote and have it count.

Tracy Lightcap February 27, 2013 at 6:15 pm

What surprised me is that the SG let that go essentially unanswered.

My answer: “Mr. Chief Justice, the government’s contention has nothing to do with whether citizens in the South are more racist. Our contention is that, as the Congressional record shows and as Congress has found, there have been continuing efforts to pass laws and make administrative decisions in areas covered by Section 5 that have a discriminatory effect on voting that falls on racial minorities. As long as Congress has made that determination, I fail to see why this Court needs to reach the issue in your question.”

Of course, I don’t have to be before him every other week either, but that is what the law on the matter is about. I’ll need to listen to the colloquy, but it sounds like the SG let a lot of stuff get by him today. Scalia’s remark about how these questions shouldn’t be left to Congress should have drawn a Samuel Jackson moment (“Well, allow me to retort!”) and apparently didn’t.

Anderson March 5, 2013 at 3:41 pm

“it sounds like the SG let a lot of stuff get by him today”

He does that. Not going down on anyone’s list of great SGs, I suspect.

Jerry Skurnik February 27, 2013 at 6:25 pm

Manhattan, Brooklyn & the Bronx & parts of California, New Hampshire & Michigan are also covered jurisdictions. Do Persily & the others refer to them?

Carolyn February 27, 2013 at 7:06 pm

How I wish that the authors of this blog could take over the Sunday morning talk shows, instead of the pundits we have now.

John H February 27, 2013 at 8:10 pm

Many liberals would probably prefer the court to expand preclearance coverage under the VRA to a wider array of states and counties, essentially to raise the transaction costs to manipulation everywhere (think voter ID in PA). But admitting this I think gives up the basic justification for Section 5 regarding previous history of discrimination. On this point, I think liberals and conservatives would agree, albeit for different reasons, that the current geographical configuration is odd.

The counterfactual needed anyway is to observe AL with and without preclearance. From today’s questions, I think this is the experiment the court is about to run.

Nicholas Valentino February 27, 2013 at 8:39 pm

I agree. We will have at least a good interrupted time series. I think the article is correct as far it goes: the covered jurisdictions are more vulnerable, on average, to racially motivated disenfranchisement. But the authors cannot, by their own admission, show that the actual or attempted disenfranchisement is worse in these jurisdictions than in other places. And if this distracts us from the general trend around the country to target minority voters. . .

Bruce Anderson February 28, 2013 at 7:59 am

… and yet, the very careful and perhaps conclusive study by Bullock and Gaddie (“The Triumph of Voting Rights in the South”; 2009) suggests that pre-clearance perhaps should be a thing of the past. The authors point out that the major divisions in American politics are generalized — that its not as though race has no effect, but that that effect is a national phenomenon, and not something exclusive to the South. As 96% of African Americans regularly vote for the Democratic party, it seems that those features of election law alteration that threaten minority voting are aimed at party, not race, and that “party” is, through the ages, as it were, “fair game”.

del2124 February 28, 2013 at 12:23 pm

Is that really polarization or is that just sorting? An electorate is “polarized” if voters are increasingly drawn to extremes — the right getting more conservative, the left getting more liberal, and moderates dwindling. That doesn’t seem to be happening.

An electorate is “sorted,” however, if voters are increasingly settled into ideological camps, that is, conservatives are almost all Republicans, liberals almost all Democrats.

wrrowvndgjdc March 11, 2013 at 7:33 am


Comments on this entry are closed.

Previous post:

Next post: