Thursday, September 15, 2011

Mind Reading and the District Plan

Over at Outside the Beltway, Doug Mataconis seems to know an awful lot about me. He knows how much I know about Pennsylvania politics (very little, it seems), and he knows how I would react to a Democratic attempt to switch to a Congressional district plan for electoral college vote apportionment in Florida or Texas (see his comments section). Unfortunately, what he doesn't know is who I am; he's taking aim at my Plum Line post yesterday, but he attributes it to Jared Bernstein, who he correctly identifies as a former Obama administration economist. Oops! Which I wouldn't bother with, except it does raise the question of what exactly he thinks he knows. Does he know quite a bit about Ja. Bernstein, and is mistakenly attributing his (alleged) partisanship and woeful lack of PA savvy to me? Is he a regular reader here, and has concluded that I'm a partisan hack with an inadequate background in the mid-Atlantic states, but then somehow managed to transpose our names and professional backgrounds? Or, perhaps...ah, I could go on, but you're getting bored already I'm sure.

OK, enough of that. What I find somewhat more puzzling, although a lot less fun, is Mataconis's endorsement of the district plan in general, but especially based on what it would have done in 2000:
Thus, that would have given Bush a total of 288 Electoral Votes to Gore’s 250. And, if you did give Florida to Gore, assuming no shift in the district allocation, the total would have been Bush 286 Gore 252. There would have been no hanging chads, no Constitutional crisis, no Bush v. Gore. Sounds like another reason we should consider adopting this nationwide (his emphasis)
Really? I happen to be okay with the status quo on the EC, and I'm willing to accept the possibility of mixed results (that is, a split between the vote winner and the EC winner), but I hardly consider it a feature. Yet Mataconis is so happy with a system that would prevent the plurality vote winner from entering the White House that he doesn't even bother to mention it as a potential problem at all? Never mind that it's not at all clear that he's correct even on his own terms; after all, a 288-250 win with Florida's two ECs contested means that if only 17 Congressional Districts are very close, we could easily have had the mess he hoped to avoid.

Regardless, it's a silly analysis. There's nothing magic about the district plan that would prevent contested results; that's just how the numbers fell in 2000. Indeed, checking my Polsby & Wildavsky, it turns out that the 1976 election would have been a dead heat had the district plan been adopted nationwide back then.

It's also worth noting that the district plan appears to contain a GOP bias. In the two elections Mataconis cites, Republicans would have done better with it, and the same is true for four of the six elections P & W list in the edition I'm looking at -- with Nixon beating Kennedy in 1960 and the 1976 tie (disclosure: I might have been the one who compiled some of those numbers. I don't really remember, but I think so). Part of why I'm untroubled by the EC is that it doesn't appear to have any clear or consistent partisan bias. At any rate, partisan bias aside, we have a system that, in the four closest recent elections, would have picked the "wrong" winner twice and dumped a third into the House of Representatives, at least after all the lawsuits in all the close districts across the nation were resolved.

I obviously am not a fan of the district system. But what I really want to know from Mataconis, who takes the position that all's fair in politics, is whether he would be okay with the "give the Republicans all of Pennsylvania's Electoral Votes" plan that Matt Yglesias points out would be perfectly Constitutional. Indeed, if every GOP-controlled state Suppose that in fact every GOP-controlled state just cancelled their 2012 presidential election and (legally) awarded their electoral votes to the GOP nominee. Would that be OK?

If not, then there clearly is some line where we don't want to just say "it's politics, live with it." And actual regular readers won't be too surprised to learn that I'd put that line pretty far to the "it's politics" side. But this one -- not the district plan per se, which I think is just a terrible idea, but selectively adopting it in a handful of states in order to create a lopsided playing field in presidential elections -- is clearly a step way too far, in my book.


  1. DB: I agree.

    I'd only add this: the problem with the real hardball politics plans (i.e. GOP state legislature moves to decide EC votes itself)isn't that they go beyond accepted but non-codified norms. The norms themselves create a self-correcting system: if the Pennsylvania GOP retracted the popular vote system for deciding EC allocation, they would themselves face a significant electoral backlash. And thus the norms both retrain prior action and self-correct for poor actions.

    The problem I see is that if the prior-restraint aspect of the norms isn't enough to dissuade the legislature, then any self-correction taken by the voters will come too late to save the action from affecting the immediate election. Which means the entire country may have to live for four years with a weakened President who many see as non-legitimate.

    On the other hand, I don't see a lot of room for an arms race to develop. As i wrote yesterday, the conditions that need to hold in order for a state legislature to even be in position to want to alter the EC allocation are quite rare. And the opposite conditions --- those that would offer the state legislature an incentive to go winner-take-all (unified government in a state likely to vote with the majority party) are much more common. So there's little equilibrium chance of ending up with a large proportion of district system states.

    I share your concern about each election featuring a handful of manipulated district system states, again for the reasons of legitimacy I cite above. But I'm still of the mind that it is unlikely any state legislature will try this: it is a dangerous move to assume the statewide vote is against your party, and then take action that places your party directly in the same electorate's crosshairs.


  2. Isn't there a 14th amendment issue in the scenario you are drawing?

  3. Charlie: what's the 14th amendment issue? I'm not following what you mean.

    JB: I like to take this point in the following direction when I teach the EC, so I'll do it here. What does the post-election lawsuit/recount world look like in a district election plan? Well, let's just compare how far off the district plan results are from the popular vote in the 8 elections in Polsby & Wildavsky (12th ed...I haven't gotten my copy of Polsby, Wildavsky, Schier, and Hopkins yet). The district plan is off from the popular results anywhere from 2.1% (in 1976) to a whopping 32% (!) in 1988, when an 8 point Bush popular vote win translated into a 40 point district vote win. Thus, it is actually reasonable for partisan lawyers to want contest the results in up to--wait for it--141 congressional districts! Under the EC, you don't possibly consider doing this, because you only look into contesting results in the closest of states that deliver a large enough number of electors to matter. You, maybe, contest results in a couple of states.

    Now, compound this with the INCREDIBLY nasty nature of election administration across districts (as elections are administered by counties in 42 states, townships/cities in 7 states, and "election districts" in AK, and only in the single-district states do you get nice congruence), and you're talking about lawsuits involving multiple counting authorities across multiple districts, and multiple cases like that....yikes. I'm not a big fan of the EC (I'm a direct vote man, myself), but leaving the presidential election up to the whims of whatever authority does redistricting (usually a legislature, so that just compounds problems) seems like a terribad idea to me.

  4. "Section 2. But when the right to vote at any election for the choice of electors for President and Vice President of the United States, Representatives in Congress, the Executive and Judicial officers of a State, or the members of the Legislature thereof, is denied to any of the male inhabitants of such State, being twenty-one years of age, and citizens of the United States, or in any way abridged, except for participation in rebellion, or other crime, the basis of representation therein shall be reduced in the proportion which the number of such male citizens shall bear to the whole number of male citizens twenty-one years of age in such State."

  5. Charlie,
    I'm not expert enough to know this, but I don't think that supercedes Article 2, section 1: "Each State shall appoint, in such Manner as the Legislature thereof may direct,"

    In essence, I think the case comes down to: has "the right to vote for Prez/VP been denied," or did such a right never exist to be denied? I think that, given that state legislatures can determine the manner of appointing electors, that no such right exists, until states choose to grant it. Once states grant you the right to vote for president at all, they cannot disenfranchise without the 14th amendment penalty. But if a state chooses to then not grant anyone such a right (except by way of voting indirectly through their state legislators), are they "denying a right" or are they making a new determination of what your rights are in that state? If nobody can do something, it would seem to not be a right.

    Wikipedia quotes Thurgood Marshall in Richardson v Ramirez as saying that "Section 2 provides a special remedy—reduced representation—to cure a particular form of electoral abuse—the disenfranchisement of Negroes." I think it's the interpretation that it provides for a partial sanction implies that the denial of this right is partial, applied only to a portion of the population. If you deny EVERYONE the right to vote for electors, I don't think you're in violation of the 14th.

    However, as I understand it, Section 2 is kind of a backwater of Supreme Court jurisprudence on the 14th Amendment, and as such, it's one where it's VERY VERY possible that my lay-reading and informed guess is bass-ackwards. It's an interesting question; I'm glad I asked for clarification (I was assuming you were referring to the equal protection clause).

  6. @matt; yes, section 2 is been long neglected. Rather parts of the second amendment, before Tribe started looking at them.

    But if you read it and Article 2 together it is prety clear:

    1. Each State shall appoint, in such Manner as the Legislature thereof may direct....

    2. "But when the right to vote at any election for the choice of electors for President and Vice President of the United States, ... is denied to any of the male inhabitants of such State, being twenty-one years of age, and citizens of the United States..."

    3. It comes down the silly example posited. If a state legislature just appointed the electors, clear 14th amendment violation. If they just fooled around the process with enough to voters, maybe. If they just said blacks couldn't vote, yep, clear violation.

    Nothing in the law is ever settled. Judges can say what they want. The language is sitting there, and if a state ever tried that stunt there are constitutional consequences.

  7. It's also worth noting that the district plan appears to contain a GOP bias.


    How can it be "biased"? It has to be illegitimately skewed to a political party in order to claim bias. This idea changes groupings and geometry a bit, across the board. Any "bias" would occur in the gerrymander, which the "biased" do every 10 years, on a bipartisan basis. This is business as usual... gerrymandering.

    You change groupings and geometry, and you potentially change the final counts. Is that a surprise to you? And we wind up with some "ties" in either case, just as we'd expect: 1960, 1976, 2000... whatever.

    And there is no "wrong" winner. There is only the winner, using the system in place. (Unless you're fixated on popular vote, AKA national card check, AKA the Palm Beach County jihad, accompanied by the long march to constitutional amendment).

    As I mentioned earlier, this is pretty much a "meh" thing, overall, and I suspect only fierce lefty partisans are getting excited over it, which they needn't, as it may potentially salvage for Obama 1/2 of PA's EV's this cycle. He's got a very thin needle eye to thread in 2012, and 10 EV's might be the difference, assuming he holds Florida and/or most of the Midwest.

    I do have a suspicion that given the hard-Left's political dominance in urban areas, coupled with the Left's self-imposed majority-minority redistricting requirements, which skew redistricting drastically, it is inevitable that the hard-Left will eventually bottle themselves up in those urban areas. District-allocated EV's might tend to exacerbate the effects of their self-bottling, in presidential elections. But, that's self-induced, even if it does occur. Get rid of majority-minority, and that would self correct. Move to the Center, and even more self correcting.

    The Left appears to be going extinct in rural districts, but that's a pure political phenomenon, as they've given up trying to attract rural voters, it appears, and would have to adopt policy accordingly, if they're ever to move back out of those urban strongholds. We could be headed for an era of R dominance, but heading it off is in the Left's hands totally, I believe.

  8. Anon,

    A system has a partisan bias if it regularly helps one party or the other. That's a definitional thing.

    As for "wrong" winner, I put it in scare quotes for a reason; I mean a candidate who wins by the rules despite a plurality of the votes for another candidate, but I don't want to write all that every time. As I've said, I'm mostly OK with that when the plurality is very small, but it's hardly a positive feature of a system.

  9. Mr. Bernstein,

    You're saying "regularly helps", but I think what you really mean is that the outcome has changed because the system changed, and the counts changed accordingly.

    It's not "partisan"... unless you view the original system as the "correct" system. Obviously, we've all swapped out electoral law again and again, so we've never quite hit upon a "correct" system, PA being the latest swapper it seems. So that standard doesn't exist, and can't be used to project a "partisan bias" onto a changed system.

    No system is going to "regularly help" any party over another. It can't. They'll adapt. As long as it's fair, discretely, to both... it's fair, and cannot help one party over another, certainly not over the long term.

    Again, the only thing jacked up about district EV's is that it involves another aspect of gerrymandering... which we already do... with gusto.

    But again, it may have a negative effect on the Left longterm, but that's entirely self induced. The Left seems to prefer caging themselves up in urban bantustans, in congressional districts and now potentially presidential districts. So be it, if that's what they want. It won't be good for our politics, if they keep on with that self-caging, however.


Note: Only a member of this blog may post a comment.

Who links to my website?