The Bullock-Gaddie studies

<x-flowed>I didn't mean to suggest that Bullock and Gaddie were playing "hide and seek" with the data. I have respect for both of them and their ultimate conclusion -- that racially disparate rates of turnout have declined markedly since 1965 -- is unassailable. For those who are interested, the 2004 census survey turnout data are available at the following link:
http://www.census.gov/population/socdemo/voting/cps2004/tab04b.xls

As these are self reported, they undoubtedly overstate turnout, as the authors recognize. So I agree that the state provided registration stats are a better source.

Nate

Rick Hasen wrote:

Keith Gaddie sends along the following response to one aspect of Nate's posting:

"Prior to 1998, the Census Bureau did not report separate figures for non-Hispanic whites. Therefore if one is to have
comparable data from 1980 - 2004, you cannot separate out the Hispanics. In AL, LA, MS and SC (along with TN), the difference in turnout in 2004 is less than 1.5 percentage points between White and White non-Hispanic.
The only states in which comparing the White non-Hispanic with black turnout leads to a different conclusion than comparing white with black turnout are GA and TX.

Nate brings up an interesting criticism, though the problem is one that is inherent to attempting to analyze the time series, because of how the census bureau at certrain points in time. And, it is not problemmatic in years where we are able to disentangle the white and non-Hispanic white participation reports. We also direct readers to our use of state-reported racial turnout data in addition to the census bureau estimates whenever available.

We'll also be working up a set of tables for Judiciary that corrects for Nate's concern, as we are not trying to play 'hide the ball' here.

Keith"

nathaniel persily wrote:
I have posted the following to the electionlawblog:

I. Highlights from Wednesday�s hearing

Wednesday�s hearing did not have many surprises. Senator Leahy started the hearing, alone, and each witness delivered their testimony. Fred Gray talked about the history of the VRA and its continuing relevance as a deterrent in Alabama. (I have to say it was pretty impressive to have Rosa Parks and Martin Luther King�s former lawyer testify at these hearings.) Drew Days talked about the constitutional issues, but I don�t think he added anything to Pam�s discussion the day before. Abigail Thernstrom was as advertised � meaning she spent most of her time criticizing �racial sorting� and �apartheid� that results from the intentional creation of majority-minority districts. (She got into an entertaining tussle with Senator Durbin during the question period.) Armand Derfner gave testimony similar to Gray�s, attesting to the continued need for section 5 in the South. I tried to focus my discussion on what the �ability to elect� standard might mean and how it would be applied. (Consider my testimony directed to the election law and redistricting nerds out there.) Warning that overconcentration (packing) of the minority community will likely prove the greater threat in the proposed 25 year span of the bill, I wanted to make sure that the legislative history was clear that this new retrogression standard prevented that as well. I also wanted to emphasize that the standard is not code for majority-minority districts and would not require that minority percentages in districts be frozen for the next 25 years. I also tried to give a list of factors that would go into this new retrogression determination. My testimony is available at the following link:
http://judiciary.senate.gov/testimony.cfm?id=1894&wit_id=5362

Two interesting developments at the hearing �

(1) John Lewis sent a letter to the hearing room, which Leahy read and inserted in the record. The letter expressed Lewis�s sadness that his testimony in GA v. Ashcroft had been taken out of context by some witnesses to suggest that he was against this bill or reauthorization more genrally.

(2) After a recess once we delivered our testimony, Senator Hatch chaired the remainder of the meeting. He focused his questions on the Thernstrom critique. He did so by referring to Stuart Taylor�s recent National Journal article that was critical of majority-minority districting and the coverage formula. His main concern, however, was the relationship between the creation of such districts and polarization in the House. He has asked us (and I suppose any of you who are interested in submitting testimony) to submit additional testimony on the issue whether section 5 is leading to greater political polarization by creating safe minority districts and more heavily conservative districts in white areas. (I didn�t want to burst his bubble and ask what might then explain rising polarization in the Senate if gerrymandering is to blame?) Incidentally, Washington Post reporter Juliet Eilperin�s new book � Fight Club Politics � blames gerrymandering for polarization in the House, but she spends only a few pages on majority minority districts.

II. Politics

I think David Epstein has the politics about right when it comes to the reauthorization debate. I am not so sure that either the civil rights groups or the Democrats� hand will be any weaker though, if this stretches beyond the midterms. If this bill becomes unraveled on the floor because of amendments etc. (an unlikely event), I could easily see the plug being pulled, the issue being used in the midterms, and then a resolution afterwards.

The most likely change I could see happening on the floor would be a shortening of the 25 year period. That would be the �compromise� offered to the few Republicans who object, perhaps in the name of trying to save it from being struck down by the courts.

III. The Trigger�s the Thing

I think everyone involved recognizes that the trigger may be the most vulnerable part of the bill to constitutional attack. I could easily see a Supreme Court decision (i.e., Justice Kennedy) saying � well, the bill otherwise may be constitutional but not with this incongruent and disproportionate coverage formula. The best justification for the current trigger, I think, would be that racially polarized voting in the covered areas is � in general � greater than in the non-covered areas. I should say that my gut tells me that must be right but I have no idea whether that is proven. Sure, that wouldn�t justify covering townships in Michigan and New Hampshire, etc., but the trigger has always been over and underinclusive. (Incidentally, the fact that �party� may explain this polarization in the South � per the Bullock & Gaddie studies � does not dispose of racial polarization as the justification for the coverage formula. Whether racially polarized voting would be sufficient to justify the trigger as a constitutional matter is a separate question, of course. More on this below.)

One thing that has become clear to me is that a potential debate over the trigger really becomes a proxy for the larger debate over what the VRA is really about � barriers to access, vote dilution, racially polarized voting, discriminatory and/or partisan enforcement of election law etc. Having that debate would certainly lead to the unraveling of this particular bill.

So why not just change the coverage formula? Leaving aside the political difficulties and the potential constitutional pitfalls of expanding coverage, it is extremely unclear what kind of trigger akin to �test plus low voter turnout� would work these days. Having a record of section 2, 5 or other violations in the last decade or two would probably cover too few jurisdictions. The data presented in Michael McDonald�s excellent article � which does the best job of trying to use historic methods for a contemporary coverage formula � suggests that 2004 election data could be used to supplement rather than substitute for the current coverage and doing so would capture more of Florida (as well as a lot of other jurisdictions, like counties in Idaho, that I found surprising). But I doubt most people think current barriers to minority participation and influence manifest themselves in lower voter turnout jurisdiction-wide and probably don�t manifest themselves in racially disparate rates of voter turnout. (Incidentally, the Bullock-Gaddie articles on the AEI website attempt to evaluate rates of turnout by race to show that African Americans often vote at higher rates than whites. There is a big mistake throughout their studies though at least with respect to their reference to census data. They accidentally lumped in Hispanics with whites instead of using non-Hispanic whites as the proper comparison group.)

Since expanding coverage nationwide is also administratively impossible (or at least really, really expensive) and would likely run into political and judicial constraints, it is no wonder that the bill sticks with the currently covered jurisdictions. At least that is the devil we know, and who knows whether the status quo is more or less likely to be declared unconstitutional than any given proposal. If I had my druthers and was forced to work within the current section 5 regime, though, I would legislate two coverage formulas � a primary one that was more expansive and a backup � akin to that used in the BCRA�s issue advocacy provisions. The primary definition of coverage would be updated for 2004 while the backup would be the one currently in the bill.

That said, of course there are problems with the section 5 architecture in general, and like everyone who has weighed in on the debate, I would throw myself behind national election reform, which is, alas, politically impossible at the present time for reasons David Epstein explains. At some point in the next 20 years though, I suspect that something along the lines of the following deal will be brokered. (This is just a prediction, not an endorsement):

1. National rules of voter registration that give a right to vote in federal elections to all citizens of voting age who are not incarcerated.

2. A national voter I.D. card (probably not with a photo or fingerprint but maybe) affirmatively provided by the federal government (or maybe just a spiffed up social security card) that will be a legal minimum for voting � i.e., no one could be refused the vote if they bring this card with them, but states could allow for other forms of ID (or just a signature) as well.

3. Repeal or modification of section 5 and/or section 2 of the VRA.

4. Perhaps a uniform ballot to be used for federal elections. (Yes � I understand the federalism questions/concerns/dangers with all of these.)

5. Some attempt to craft a body like the FEC or EAC to be more independent and effective with respect to oversight of federal elections.

Like I said, this is not even on the horizon right now (nor in some respects should it be), but after the next electoral meltdown it might be.

III. /GA// v. Ashcroft /and the /Ashcroft-/fix in the current bill

I think academics have had a healthy debate over the potential virtues or drawbacks of the holding of /Georgia v. Ashcroft/. My own view is that the biggest problems with /Georgia v. Ashcroft/, alas, are its facts and its holding. The facts are a problem because they are unique � a Democratic gerrymander (really egregious by the way) supported by African American elected officials, with control of the state senate at stake, and a denial of preclearance (or really a declaratory judgment) to districts right around the 50% mark that are decreased ever so slightly. The holding is a problem because it cannot possibly be limited to those unique facts and as far as I can tell, allows jurisdictions to choose packing or cracking as a strategy for gerrymandering but merely justify it as creating �control� or �influence� districts. By my read, the decision basically told DOJ to preclear almost any kind of redistricting plan or, alternatively, provided some justification for objecting to almost any kind of redistricting plan. I think this conclusion is bolstered by the behavior of both the political appointees and the line attorneys in the Texas gerrymandering case � both of which have a plausible reading of the decision in my view. Therefore, I think some clarification of the decision was necessary if section 5 was going to keep DOJ in the business of preclearing redistricting plans.

Now the question is whether the /Ashcroft/-fix fixes /Ashcroft /and even if so, does it do other kinds of damage. The bill under consideration would prevent �diminishing [minorities�] ability . . . to elect their preferred candidates of choice.� One plausible interpretation is that no reductions in minority percentages in current districts will be allowed for the next 25 years, or that this forces the creation or freezing in place of majority-minority districts. I reject that interpretation and do not think it is the intent of the drafters of this law. See my testimony for a fuller discussion of this. More importantly though, I think it is important that DOJ look at this bill as preventing the kind of overconcentration that will likely prove to be the greater threat to minority influence in the coming years as racial polarization declines. (See my testimony on this as well.) It would be a cruel result, indeed, if the new retrogression standard prevented the most common form of Democratic gerrymandering (cracking) but gave its blessing to the most common form of Republican gerrymandering (packing).

IV. Constitutional Questions

I recognize why there has been so much discussion of the constitutional questions regarding this bill and the importance of developing the record. But I think I part company with my fellow travelers in these debates in that I just don�t think any of this will ultimately matter to Justice Kennedy, who will be the decisive vote. I think he already knows what he is going to do.

In the end, he (and maybe even the Chief Justice) will need to decide whether they feel comfortable striking down THE VOTING RIGHTS ACT, previously the gold standard or point of comparison for the other decisions interpreting the enforcement clauses of the 14^th and 15^th amendments. If they do strike it down, it will be the most significant law stricken down by the Court at least since the New Deal if not before. (Look for a good named plaintiff that can be converted easily to a verb like � /Lochnerizing/.) I think this is probably true even if they just say they are striking down the coverage formula, the time period (25 years) or �the ability to elect� standard.

It will be the federalism revolution realized. The other cases concerning the enforcement clauses (Morrison, Garret, Kimel, Florida Prepaid, etc.), commerce clause (Morrison, Lopez), commandeering (Printz, NY v. US), or the various 11^th amendment cases are not remotely even in the same league. Of course, none of these other laws have anything approaching the federalism costs that the section 5 regime does either. In the end, I suspect Justice Kennedy will simply need to decide whether he is willing to take the heat, how important the precedent in this area is, and whether the Court wants to overturn a law that will almost be unanimously supported by Congress.

--
Rick Hasen William H. Hannon Distinguished Professor of Law
Loyola Law School 919 Albany Street Los Angeles, CA 90015-1211 (213)736-1466 - voice (213)380-3769 - fax rick.hasen@lls.edu http://www.lls.edu/academics/faculty/hasen.html http://electionlawblog.org

</x-flowed>