Don’t Like Gerrymandering? Be Wary of the Alternatives.

Nonpartisan boards are never devoid of politics, and the Supreme Court has said that the federal judiciary is not the solution.

By
81

Scan the headlines of any major media outlet and you’re bound to see a fulsome condemnation of the way state legislatures pick their district boundaries. It’s an interesting phenomenon given the relative monotony of the process over the years. Every decade, state legislatures use the results of the census to redraw congressional and state legislative districts. Sometimes this is done with a divided government, sometimes it is done with one party in charge. In every case, it is done with political considerations in mind.

This had been the case even before adoption of the U.S. Constitution in 1787. Partisan apportionment was known to the Founding Fathers—Patrick Henry actually attempted, unsuccessfully, to gerrymander James Madison out of the first Congress. 

In 1811, Massachusetts Gov. Elbridge Gerry—a signer of the Declaration of Independence—approved a bill adjusting the state’s legislative districts to enhance the power of the Democrats and weaken the influence of the Federalist Party. And thus was born “gerrymandering.”

And so it went for the next two centuries. But Democrats have been making the case that it’s a national evil that must be extinguished since 2010, when Republicans swept state and national elections in the aftermath of the Obamacare debacle.

Keep reading with a free account
Create a free Dispatch account to keep reading Get Started ALREADY HAVE AN ACCOUNT? SIGN IN
81
By
Comments (81)
Join The Dispatch to participate in the comments.
 
  1. Avatar photo
    Ben Connelly

    Very interesting article. I remain concerned with gerrymandering, but perhaps less so than I was. Nonpartisan committees aren’t.

    Collapse
  2. Joe Socher

    What is a “good” map?
    Is it one that has geographically compact districts? What is the neutral principle that says nice shapes are more democratic? What if compact shapes divides natural communities or combines many conflicting interest groups? Or if it dilutes votes of one minority bloc in favor of another?

    If the goal is to reflect actual communities of interest it’s likely that districts would not be compact but would reflect demographics which can be really weird shapes and ethnic pockets etc.

    If you have cohesive districts representing real communities then you are also likely to have strong incumbency which is supposed to be bad.

    Do you want more competitive districts? Then you will likely be chopping up natural communities and diluting ethnic voting blocs etc

    Do you want majority-minority districts? Then you end up siloing the minorities into a few districts.

    These are incommensurable and incompatible goals that can’t really be adjudicated on neutral principles.

    Collapse
    1. Avatar photo
      MikeyDubs

      Which is why any reform effort would be useless unless the hierarchy of these principles was codified. Yes, we have to choose between these principles. Congress (or the voters in a state amending its constitution) have to say which goals come first, second, or third. This is what Florida did in 2010 and (knock on wood) it's worked so far.

      Collapse
  3. Avatar photo
    Dr. Livesey

    Here, lemme solve this gerrymandering thing in one comment (you're welcome):

    Mandate that individual voting precincts in each municipality have the shortest possible boundaries, in total, and that each have as equal a number of people in it as possible. A computer should be able to do this. If the computer comes up with different models with the same total length of boundaries, flip a coin. If there are many models, keep flipping coins. There will probably be isolated households on the other side of a river or something creating some hardship in driving a long distance to get to the polling place. Too bad. I'm here to solve the gerrymandering problem, not the getting-to-the-polling-station problem.

    Take those voting district boundaries and create districts with equal numbers of voters (or as equal as possible) with the shortest possible total length of all district boundaries. Now, I don't know if this could happen, but let's say this results in one district at the end that's wildly distended and all or nearly all the others are compact. Would that be so awful? Cry me a river. I'm here to solve the gerrymandering problem. Deal with your single distended district, citizens. Quit whining and get on with it, it's better than gerrymandering, and that's what I'm here to solve for you all. It's one district. Show a little gratitude, dammit. Take one for the team.

    Do we need districts that unify a municipality so that one state senator or representative has all or most of that political unit? No, we don't. Do we need certain groups to have their own majority district — some for Blacks, or Hispanics, or Hungarian Americans, or whatever? No, we don't. Do we need anything but compact districts, and every other consideration be damned. No, we certainly do not. We let computer models come up with district lines, maybe come up with two or three or four candidates and then let the state legislators vote on only those options. Someone will still accuse the mapmakers of unfairness and sue, and .... so what. The Supreme Court doesn't want to get involved in politics? Fine. They shouldn't have a problem getting involved in math questions. They can appoint experts in geography or math or whatever, because that's what this process would be reduced to. Civilization decides a lot of other important things that are vastly more complex, and then those decisions become the subject of lawsuits and court decisions. This ain't no bigger.

    The advantage (besides having seats safe from lawmaking creeps who cater only to some ideological dark corner of the country): We'd have one less thing to worry about. This won't bring Nirvana, but it would improve the Republic, and cheating pols would have to look elsewhere to grift the public.

    Collapse
    1. Joe Socher

      I still don’t get why compact is a goal? Aesthetics?

      Collapse
      1. Avatar photo
        Dr. Livesey

        Noncompact = gerrymandering. You can't have a gerrymander with relatively short borders. Lengthening borders is how gerrymandering grabs constituencies that the schemer knows will be more supportive (or less supportive) of a particular political view. If you always have the shortest length of borders, how are you going to gerrymander?

        Collapse
  4. Avatar photo
    Mark Stobbe

    I'm from Canada. We have "independent" election boundary commissions in provinces and for the federal government. Political parties still influence boundaries. As noted above, the sausage making is just a little more hidden.

    Collapse
  5. Avatar photo
    Bill Dyer (aka Beldar)

    Winston Churchill, in 1947:

    ===begin quote===
    Many forms of Government have been tried, and will be tried in this world of sin and woe. No one pretends that democracy is perfect or all-wise. Indeed it has been said that democracy is the worst form of Government except for all those other forms that have been tried from time to time.
    ===end quote===
    link: https://winstonchurchill.org/resources/quotes/the-worst-form-of-government/

    A derivative of this principle is that letting state legislatures draw their own districts, and their state's federal districts for U.S. House representation, is the worst way to district and redistrict — except for all those other forms that have been tried from time to time.

    In general, at least in states with several congressional districts, it's easier to flip the state's legislature and governor than it is to flip its U.S. House delegation, just because the state legislative districts are smaller in population, and grass-roots changes are reflected more promptly and precisely therein.

    In my lifetime, I've seen Texas flip from being a one-party state in which Democrats in the Texas Legislature protected their incumbent partisans in the U.S. House via gerrymandering, and did so with such effectiveness that Texas continued to send a majority-Democrat delegation to the U.S. House well into the 21st Century, even though the state's voters had already so overwhelmingly switched allegiance that no Democrat has won a statewide election since 1996. The gerrymander designed by former U.S. Rep. Martin Frost of Fort Worth to protect himself and his fellow House Dems had to be undone first — and by far the biggest impediment to that was the unending barrage of almost entirely bogus civil rights cases brought by Dems in federal courts. But the necessary precondition to changing the state's U.S. House delegation to more appropriately represent the state's aggregate voting population was the capture by the GOP of both chambers of the Texas Legislature and the Governor's Mansion — which was very much a grass-roots project.

    And that's where you WANT the action to be. That's why the Framers, in their wisdom, gave this power to the state legislatures. Sometimes you have to be satisfied with the best out of a bad set of alternatives, and this is one.

    Collapse
    1. GregS

      Indeed.
      Countries have tried forms of government other than democracy and we've seen that it turns out badly every time.
      Do you have examples of democracies that have tried methods of creating parliamentary or congressional district boundaries other than our partisan system? And have those methods turned out worse than what we've got?

      Collapse
    2. Avatar photo
      MikeyDubs

      This is an ignorant comment. In other democratic countries that draw district lines, the drawing is entrusted people appointed from either one level up in the government, or one level below. The idea that a state legislature gets to draw the lines for the state legislature is obviously an invitation for shenanigans.

      Collapse
  6. M M (Lee)

    Why don't we just assign citizens to congressional districts at random, within each state? Voila, problem solved. I'm for maximizing competition within each district and through randomization it seems like that goal would be achieved. Although politicians will never vote for this, could we not achieve this through some kind of direct vote by the citizens?

    Collapse
    1. Avatar photo
      Peter S.

      If a state has 55% Democrats and 45% Republicans, then what you suggest would result in an election of 100% Democrats, each of whom wins with 55% of their "district". Is that what you want?

      Collapse
      1. M M (Lee)

        The idea is that the current extremism of both parties is born out of the imbalanced, non-competitive districts that breed extremist politicians who have no incentive to adjust their behavior.

        If their districts go from, for example, 65% Republican to 54% Republican, that gives an opening for moderate Democrat candidates to win by appealing to say 5% of GOP voters who are uncomfortable with GOP extremism and would consider voting for a moderate Democrat. This risk of competition would create the incentive for extremist politicians to moderate their behaviors on both sides of the aisle.

        As a result, due to fewer extremist politicians in either party, the party brands themselves would change for the better.

        Collapse
      2. M M (Lee)

        In such a system, the definition and very identity of the parties would evolve dramatically. In the scenario you have posed, the Democrats would adjust their policies to persuade at least 50% of the voters. With the vast majority of districts now competitive, the parties would be incentivized to market themselves to voters in the middle rather than on the extremes. Republicans would reach out to groups other than white evangelicals, and Democrats would reach out to other groups besides their base. In the process, the party brands would change, which would also bring greater competition to senate races where gerrymandering is not an issue.

        Collapse
  7. Avatar photo
    Jeff Walden

    The article only very briefly gestures at the problem of "independent" redistricting boards and commissions ending up partisan. For a real-world demonstration of how this actually works out, look no further than California in 2011. Pro Publica did a long piece on it, that is extremely worth reading:

    https://www.propublica.org/article/how-democrats-fooled-californias-redistricting-commission

    *Even assuming everyone on the board was acting in good faith*, the board got snookered by partisans who came to them claiming non-existent constituencies their map-drawing should preserve.

    I don't for a second think that any system that ultimately is constructed from human judgment will effectively end the problem of biased maps and lines. The only solution that seems like it *would* be a solution, is codifying in law a rigorous system that *purely mechanically* draws lines taking into account only specifically enumerated factors not susceptible to manipulation. For example, city lines or something like that. (I think it highly improbable that enough cities could coordinate to redraw their own boundaries to enable effective gerrymandering of district lines. Although these days, maybe that's less and less far-fetched, given how much we all hate each other...)

    Collapse
  8. Phil P

    Good article, and right on cue, the usual outrage from the usual suspects. I may be in the minority, with the exception of the districts like Dan Crenshaw's, I am not that bothered by gerrymandering. The left is more outraged simply because the GOP currently controls more governerships, more state legislatures, more trifectas.

    As someone else commented, Michigan just enacted a new "nonpartison" board that is already causing problems. There are likely no real solutions to the gerrymandering issue, although maybe there are ways it can be mitigated a little bit.

    Collapse
  9. Avatar photo
    Peter S.

    You can't remove the incentives for gerrymandering ... or can you?

    My proposal: individual candidates don't win by plurality, they win by being in a party that earns a certain representation according to how the party did in the whole state. That is, you still elect individual candidates, but only as many as your party "deserves".

    To be specific: Say there's 100 districts, and the race comes out 63% R, 35% D, 2% Other.

    1. Toss out the 2%, as it's below a "take this party seriously" threshold. So effectively you've got about 64% R, 36% D.
    2. Since there are 100 seats, the 64 R's who got the highest percentages win, as do the 36 D's in the rest of the districts. If the map is completely ungerrymandered, then basically everybody wins with a plurality. BUT — if the map is highly gerrymandered by the D's, there may be several R's who win without a plurality and D's that lose with a plurality. And vice versa.

    That's it. Suddenly it doesn't matter how intricately you gerrymander, there's no partisan advantage in it. Now, there would still be gerrymandering on other axes, notably to favor incumbents. But PARTISAN gerrymandering would go away.

    Collapse
    1. Avatar photo
      MikeyDubs

      Your system could work. It does make sense, and it does make gerrymandering much less relevant. But it also means, in a given district, candidate A could win the seat, even though candidate B got more votes in the district. You're letting an outcome in a district change based on statewide votes. That is not going to seem fair to most people. Nor will it seem fair to any the people running for office. You may see them as interchangeable members of a party, but they are all real people with real careers who already risk losing their seat all the time due to factors beyond their control. Why would they add another factor they can't control to the list? Plus you would be increasing the role of parties, and party discipline. Mavericks would be guaranteed to get kicked out of their party, have to run as independents, where they would always lose whenever the vote balancing you describe takes place.

      Collapse
      1. Avatar photo
        Peter S.

        I agree that the optics of winning with fewer votes than your opponent would be awkward to explain, but hey, we keep having to justify the Electoral College to millions of skeptics every 4 years already. Anyway, the loser-with-the-plurality has no one to blame but their own party who (presumably) engineered the gerrymander that way. (If the gerrymander was from the other party, then the shoe's on the other foot and you could win with less than a plurality.)

        Anyway, each candidate is an individual, and they're still elected on the merits, as adjudicated by voters. It's just that instead of simply getting more votes than your opponent, you have to get more vote share than X percent of your fellow party members. Your incentive is still to get lots of votes. It's just a bit less obvious exactly how many you're gonna need. (But this can still be modeled / predicted, of course.)

        I understand the unease about elevating parties over individuals. Seems a bit parliamentarian, for sure. But my theory is based on the idea that gerrymandering is already partisan, i.e., the primary consideration is to engineer the most seats for a given party. That being the case, the remedy has to speak to parties as well. No getting around it, really.

        (And as I noted, this does not address gerrymandering from other motivations, most notably to protect favored incumbents by making their seats safer. I don't know of an obvious way to prevent that one, aside from those "neutral" commissions.)

        As for the fact that no independent could win — the simplified system I laid out above does not address that, but it's a fairly easy tweak. Between rule 1 and rule 2 above, have a rule 1.5: after excluding the "too small to count" parties, anyone who won a plurality from outside the parties that still "count", automatically wins their district. Rule 2 then only applies to the remaining votes and the remaining districts.

        Collapse
        1. Avatar photo
          MikeyDubs

          You would actually see the parties put up well funded candidates in every district when every vote counts towards the statewide total. Many of these candidates would know that they have no chance of winning a seat, yet they would campaign just as aggressively as the ones who can win. Individual ambition would have to be subsumed to the parties on multiple levels. While I agree that the disproportionate results we see from gerrymandering are a problem, to me the larger problem is how the current electorate and voting rules disincentivize consensus and cooperation. Karl Rove showed that a party does not need to move to the center in a general election, it can double down on energizing fringe voters instead. Since 2000, only Biden has broken that pattern and there's no reason to believe he's anything but an outlier. Your idea would strengthen this incentive structure for parties, such the only possible outcomes are gridlock or whiplash, both worse than we see today.

          Collapse
          1. Avatar photo
            Peter S.

            I dunno, if I live in an 80-20 district (whether gerrymandered, or merely self-sorted), it'd be nice to know my vote still counts for something, and that both parties still care about getting my vote. Whether I lean toward the 80% party or the 20% party.

            You might think that the party that expects to get 20% of a district will run an extreme candidate in order to drive base turnout, but they could just as well run a moderate candidate in hopes of picking off a few moderate votes. The fact that they have an incentive to care about the district at all is, to my view, a good thing.

            Parties already have a choice of either trying to appeal broadly, or double down on their base to drive turnout. I don't think that changes much if they have to compete in the whole state rather than concentrate on specific winnable districts.

            Now, purely from a party insider / donor perspective, probably they would prefer to not have to compete for the goodwill of an entire state, but just concentrate on their own safe races. But why should I care about that? I don't think it's in the voter's interest.

            Collapse
  10. Avatar photo
    Nick H

    One interesting note about the requirement for there to be majority-minority districts is that it often helps Republicans and hurts Democrats. There are two different strategies that can be used to create maps favorable to one party over the other: pack and crack. Creating a majority-minority district in a case where 80 to 90% of the minority voters will be supporting one party is exactly the same as the strategy of packing. That can put the GOP into a situation where they'll be sued for creating a partisan gerrymander if they do create the majority-minority district and sued for violating the law if they don't.

    Collapse
Load More