Independent Commissions Gerrymander Just as Much as State Legislators

Independent redistricting commissions may not be as politically-neutral as theorized, says new study



The reelection rate in the U.S. House of Representatives was 87 percent in 1964. In 2016, it was 97 percent. What happened?

A December 2016 New York Times editorial offered the conventional wisdom that the problem largely the result of "partisan gerrymandering—the drawing of federal or state legislative districts to benefit Republicans or Democrats." The editorial board declared gerrymandering "among the most corrosive practices in modern American democracy. It lets incumbents keep themselves and their party in power even without majority support, it deprives voters of representatives who reflect their wishes and it contributes to the hyperpartisan gridlock in the nation's politics."

The solution? "A permanent fix for partisan gerrymandering would be to take redistricting entirely out of the hands of self-interested lawmakers and give it to independent commissions," the Times' editors argued. Now Rep. Don Beyer (D-Va.) is introducing a bill that would, among other things, set up such independent commissions to draw congressional districts.

On the face of it, taking the process out of the hands of self-interested politicians might sound like it would reduce incumbent-protection redistricting. But a new working paper by researchers at UCLA and Yale suggests otherwise.

To measure the relative competitiveness of various House redistricting processes in 2010, the researchers first simulated a set of counterfactual district maps based solely on equal population and geographic contiguity. They then aggregated what the margin of victory for Republican and Democratic candidates would have been in the simulated districts, comparing them to the actual electoral results in 2016.

Next they collected a set of 1,473 proposed district maps from 13 states made publically available by state legislatures or redistricting commissions. The researchers compared the margins of victories in those proposed districts with both simulated and actual results.

The median state redistricting plans were less competitive than 99 percent of the simulated alternatives. In fact, 43 percent of the states produced final maps that were less competitive than every single simulated map.

Next, the researchers compared the enacted redistricting maps to the publically proposed maps. Not surprisingly, state legislatures' maps turned out to be safer for incumbents than 77 percent of the simulated alternatives. But the maps produced by independent commissions were only marginally less safe: 75 percent of the simulations were more competitive than most of plans adopted by the commissions.

When they compared the final redistricting maps devised the state legislators with the publically proposed alternative maps, the researchers found that the legislators' efforts were less competitive than 71 percent of the proposed alternatives. But the maps created by nonpartisan redistricting commissions were worse; 76 percent were less competitive than the proposed alternatives.

"In sum, independent commissions do not draw House maps that encourage greater electoral competition any more than partisan legislature do," the researchers conclude. "Overall, our results suggest caution in overhauling state redistricting institutions to increase electoral competition: independent commissions may not be as politically-neutral as theorized."

H.L. Mencken once quipped, "There is always a well-known solution to every human problem—neat, plausible, and wrong." Independent redistricting commissions appear to be just such a solution.

NEXT: Phoenix Taxpayers Lose $200 Million on Sale of Largest Hotel in Arizona

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

    1. “Nevermore” or “After 2016, I’d expect everybody’s mail to come up several yards short of the mailbox.”

  1. My solution has several parts:

    1. Elect the top three vote winners in each district.

    2. Each representative proxies however many votes they got, instead of just one voter per rep.

    3. Border parcels can shift to a neighboring district, as long as (a) the new district had fewer total votes, (b) they don’t split a district or force other parcels to be isolated.

  2. Can’t they create algorithms to do this?

    1. They already do that

      They feed in voter registration data, gender data, race data, past election results data

      1. What I’m asking, is if it possible to have districts created by random algorithmic simulations based on that data, with everything publicly-available so it would be known if the algorithm was manipulated, cheated, not actually used, etc.?

        1. Of course it is.
          You limit the data fed into the system to population only.
          Align district boundaries to established political boundaries when the political unit within political boundary exceeds the maximum population of a district.
          Limit the ratio of the height and width of the district.

          The real question is why would political animals ever want to do redistricting in a way that did not benefit the party in the majority and/or certain incumbents of the minority party while redistricting?

          1. The real question is why would political animals ever want to do redistricting in a way that did not benefit the party in the majority and/or certain incumbents of the minority party while redistricting?

            Conversely, any time you draw a line you’re going to be segregating people and lumping a minority in with a/the majority at the same time.

          2. The first problem is that it’s currently illegal to consider only population. It’s _required_ to consider the color of the population, and since blacks nearly always vote Democrat, that in itself mandates manipulation of some district boundaries on political lines.

        2. Most “gerrymandering” is due to the Dems pushing for minority-majority districts. We have two in Ohio where any black Dem will win 75-25 or 80-20. They also win Toledo and Cleveland- And the Repubs win all the rest at 53-47 or 55-45…

          The Repubs just said, “Please don’t throw me in that briar patch, Br’er Fox!”

  3. Let me guess these “independent commissions” are staffed by election experts. And who is more expert in elections then politicians, former politicians, bureaucrats who are appointed by politicians and campaign consultants

  4. Term limits seems an elegant solution to the problem.

  5. My solution has one easy part:

    Don’t let a computer anywhere near redistricting efforts.

    And one assumption:

    Stop pretending that you can create “competitive” districts. People have a stubborn tendency to associate themselves with like-minded neighbors.

  6. I once read the anti-Federalist papers (which aren’t so much a unified collection as they are a serious of writings countering the Federalist papers published around the same time with pseudonyms such as ‘Federalist Farmer’ and ‘Brutus’). There were a series of papers written by the anti-Federalists where they addressed this very issue. They countered that districts should be designed organically by local communities.

    Obviously, there is a set population for congressional districts, but it seems to make more sense to allow local towns or parts of cities to negotiate among themselves to design their districts. People already self-segregate when they choose where they live, so it seems like it would be easier to achieve consensus about districts within local governments, rather than state governments.

    Maybe, this might be one of those times where Reason selectively and opportunistically uses the argument of ‘federalism’

    1. There are two reasons for requiring regular censuses (censii?). One is for approximately equal sized House districts; the other is for redistribution.

      You can solve the equal sized districts very easily by having each rep proxy the votes in the last election. Yeah, it’s more work, more prone to miscalculation in the pre-computer days, but so you let the reps themselves do it, since they have a strong interest in correct tallies.

      I would also elect the top 3 vote winners from each district, simply to get some diversity and break the two party system, but that’s not necessary. WIth election vote proxying, you lose the excuse of having to equalize districts. It makes it easier to follow natural boundaries, like counties and cities.

      1. Not bad but have you ever looked at the boundary for the city of LA? It’s pre-gerrymandered.…..undary.pdf

      2. The Latin plural, since you ask, is c?ns?s, with a long vowel in the ending; it’s one of the small group of u-stem nouns like manus (hand), sp?ritus (breath), gen? (knee, plural genua).

        Even if it were a member of the larger o-stem group (in which earlier ?os became ?us), the plural would not be ?ii. Come on, is radii the plural of radus?

  7. The models need to redistrict to be biased towards libertarians. We will control the system, and make all of those bow before us, at their own discretion and choice.

  8. The paper doesn’t explore the reason this might be so, but I’d surmise that it might have something to do with the fact that most “independent commissions” are set up to give one party or the other an advantage or total control, depending on which one controls the legislature that creates them.

    1. Ric Romero school of obviousness. That’s as close as Tony the Tiger can get to self-awareness.

      1. Hey I didn’t write a paper about it nor did I write an article about that paper with the usual upshot “Don’t mess with the status quo–things are going great for Republicans/Big Oil!”

        1. Oh my God, are you dumb. The myth that Republicans have been winning because of gerrymandering will ensure that Democrats remain a minority party. How do you explain Republicans holding the Senate? Are state boundaries gerrymandered?

          1. Sure, in a way. California gets the same number of senators as Montana, after all. Congress seems to be based on the very, very unlikely premise that the more rural a person is, the better he is at making decisions for everyone else.

            1. How about we consider the corollary: By why what reasoning would it be justifiable to let populous states like CA overpower less populous states like Montana? People keep forgetting that it is not the citizens that own the central government, it is the states. The citizens own the states.

              We all agree with the premise of one citizen, one vote. Well, in the Senate, it’s one state, two votes. Also remember that the constitution originally stipulated that each state legislature would choose its senators. I think it was a big mistake to change that. The extent that Washington has direct access to every citizen blows holes through the very idea of a federal government.

              1. The reason that was changed to direct voting was because of the corruption that was so rampant in the prior system.

                I get small-r republican arguments for systems, but I think there has to be a damn good argument for why we’re letting one person’s vote count 60 times more than another’s. The fact is the senate doesn’t make policy for states, it makes it for people. The reasons states are given representation in Congress is precisely so that people in less-populous states get more representation than they really deserve. It was basically all compromise for the sake of slave-owners, which, among principles behind setting up a system of government, is not the first I would call “lofty.”

                That is, equal democracy is ideal (since everyone in a jurisdiction is affected by policy), and any deviation needs to be justified. I personally believe that even the House is egregiously anti-democratic, let alone the senate. And truly fuck the electoral college.

                1. The reason that was changed to direct voting was because of the corruption that was so rampant in the prior system.

                  Well, at least I didn’t have to read much of your comment before I ran up against the obvious bullshit.

                  1. Well, at least I didn’t have to read much of your comment before I ran up against the obvious bullshit.

                    He is right that there was sometimes corruption involved. There was also a problem of state legislatures often could not agree on who to send to fill vacancies.

                    Of course those were the public reasons.

                    The chief private reason is that the progressives in particular wanted direct election is that the Senators tended to be obstacles to the progressive agenda.

                2. equal democracy is ideal

                  Oh, I see the problem. You’ve not heard the one about it being two wolves and a lamb voting on what’s for lunch, right?

  9. The reelection rate in the U.S. House of Representatives was 87 percent in 1964. In 2016, it was 97 percent. What happened?

    People are dumb, and vote for the person who’s name they vaguely recognize as being in their tribe?

    1. Gerrymandering was over a century old in 1964. They need to look elsewhere for the reason incumbency has become an almost insurmountable advantage.

      One thing that HAS changed is campaign finance laws. These didn’t exist in 1964, IIRC, and have been repeatedly added and tightened since, while the incumbent advantage grew. It should be no surprise that such laws act to protect the jobs of those who wrote them.

  10. A lot of the gerrymandering has been to create majority-minority districts. As the overwhelming majority of blacks also vote for Democrats, that serves to concentrate Democrat voters. Having majority-minority districts and a more even distribution of voters are mutually contradictory goals.

    1. I didn’t scroll down enough before I replied… +1

  11. Here’s one I dreamed up after (more than) a beer or two:

    Each Election Day one year before the congressional and presidential elections (odd numbered years, in other words), a simple two-part referendum –

    Have the elected officials in the executive branch of government earned the right to run for another term?
    Have the elected officials in the legislative branch of government earned the right to run for another term?

    If the answer is “yes”, incumbents get to proceed with re-election plans.
    If the answer is “no”, they’re instant lame ducks for the next year.

    The prospect of imminent unemployment (and no chance to benefit from the revolving door between big business and big government) should counterbalance the current incentive for parties to put up with the hacks, poo-flingers and crooks they tolerate because it gets them to that magic 50%+1 of the vote.

    We have to stop treating presidents, vice presidents and members of Congress as quasi-royal magical beings with unique strengths that we puny mortals couldn’t possibly understand, let alone equal or surpass. We all have thousands of people living in our congressional districts who could be every bit as useful as the offices’ current inhabitants, if not more so.

    My indisputably brilliant (after a beer or two) idea includes both the stick of term limits and the carrot of no term limits, drives up turnout for off-year local elections, and renders the whole idea of gerrymandering pointless.

  12. A major problem with gerrymandering today is the Voting Rights Act of 1965. If a state is required to create “minority influence districts” that are likely to go Democrat, then the surrounding districts are likely to be both whiter and more Republican. You will also get irregular boundaries as lines are drawn to include/exclude areas in an effort to ensure these districts will meet the some judges nebulous Goldilocks test for being “just right” rather than “too minority” (packing) or “not minority enough” (cracking) — or even declared to have put the wrong minority areas in the same district. Regularly shaped districts that respect physical and political boundaries might undercut the maximization of minority representation — but would at the same time create legislative bodies that are less polarized.

  13. Since it is federal level elections and the resultant federal laws apply equally across all districts, what’s the point of a district? Save pork spending (I answered my own question) what benefit is there to my district having a better/worse official than the district down the street? The laws passed will apply the same. As such, why not make all House seats a state-wide election. If you have 5 House seats for your state population then the top 5 vote getters are now your reps. Senate is state wide, too.

    Why should me and like-minded people who live on the other side of the state not be allowed to have our rep in the House simply because we don’t live next to each other? How does physical geography play any role what-so-ever in the concept of justice and representative democracy. My living where I do versus across the street does not change my philosophical ideas of justice and governance. Why must my tribe, In an enlightened world, Be dictated by my GPS local rather than my desires and ideas?

    Better yet… maybe instead of voting for a goon to tell me what to do, my local neighbors can get bent and leave me alone.

  14. Just draw them according to existing political boundaries. Problem solved.

Please to post comments

Comments are closed.