Princeton Election Consortium

A first draft of electoral history. Since 2004

House Control Meta-Margin: D+1 Also Check out the new Beta!

Presidential Debate Question: Gerrymandering

October 1st, 2016, 8:14pm by Sam Wang


Partisan gerrymandering can be done by either party. Currently the net effect heavily favors Republicans. As I wrote in the Stanford Law Review and in the New York Times in 2013 and 2015, gerrymandering has captured a net 10-15 House seats for Republicans since 2012, and dozens more seats in state legislatures. As a result, gerrymandering makes it challenging for Democrats to capture the House this year. I have proposed statistical standards that, if adopted by the Supreme Court, could remedy most cases of partisan gerrymandering in a way that is as consistent with existing judicial precedents as I can make it.

The Commission on Presidential Debates is asking people to vote on questions to ask Clinton and Trump. One concerns gerrymandering. Go vote! (Think of it as practice for voting on November 8th.)

Tags: 2016 Election · President

36 Comments so far ↓

  • Unclearnie

    Another problem with the House is the fact that other than adding a seat for both Hawaii and Alaska, it hasn’t grown since 1913. Since the lowest population state, Wyoming, must have a whole Congressman (no partial people allowed) the remaining states have to divide up the rest. This means that large population states like California, New York and Texas have congressional districts that have many more people in each of them than Wyoming has in the entire state (appx 550,000 last census). For example, if California had congressional districts of 550,000 each, they would have something like 68 instead of 53 congress persons. Definitely, not equal representation. I suggest enlarging the House of Representatives so that every congressional district has essentially the same population. Maybe there were valid reasons back in 1913 to limit the number of representatives (no place for them to sit?) but with today’s technology, you don’t have to be in the room to “attend” each session. Enlarge the House: equal representation for all!

    Note, this is not a Constitutional limit. It is arbitrary, set by congress, changeable by congress. Originally, it was thought each member of the House would represent appx. 30K voters.

    • Harold Bridges

      The Congress was structured as it was, with equal weighting in the Senate and the House stuffed with Southerners because of the three-fifths compromise, in order to establish unequal representation that favored the South. The Electoral College was designed to transfer the South’s overrepresentation in the Congress to enable them to control the presidency as well. So, it’s not through some oversight that anomalies of unequal representation arise. That was the goal.

      All of which means we are still dealing with the effects of America’s original sin.

  • Partha Neogy

    Is this solely due to gerrymandering; or is it partly due to the inefficient concentration of Democratic voters? Or is the latter cause also a result of gerrymandering?

    • Sam Wang

      Read my paper please. Gerrymandering effects far larger than clustering in individual states. Nationwide, clustering effects do add up to a total effect that is of similar size.

  • WildIrish

    In NC, it’s definitely due to gerrymandering. The Rs re-drew the districts as soon as they won enough seats and could twist enough arms, and now they have a veto-proof majority in the NC House. There is no way this state’s population willingly elected that many Rs, when the popular vote was so evenly split. We managed to elect Obama the first time around, but by the 2012 election, the Rs had put so many “protections” in place to prevent voter “fraud” that many Ds didn’t get to vote. The disenfranchisement efforts continue to this day, but we’ve won a couple of court battles recently.

    Democratic voters are actually *divided up* with cleverly drawn districts so that there are less than 50% in most districts, and there is no way to get a majority. We desperately need a non-partisan redistricting effort, so that the proportion of NC House members elected from each party more closely resembles the proportions found in the state’s voting population.

  • ru

    I can’t help but feel that multi-member districts would render the whole thing moot. (Along with its other inherent advantages)

    • Sam Wang

      This idea gets brought up a lot, but is an impossible goal. It is disallowed by legislation and judicial precedent – see this history. At one point, multimember districts were used as a way to evade the enfranchisement of black voters. I believe that led to passage of a law in 1967 that mandated single-member districts. It is hard to imagine how that law would get modified any time soon.

      There are lots of attractive-sounding solutions. The challenge is coming up with one that fits with a rather large body of law and Supreme Court rulings!

  • whirlaway

    Democrats are outwardly opposed to gerrymandering but in reality, they are fine with it as it gives them a ready-made excuse to do nothing or make deals that helps their corporate masters and then blame the Republicans for it!

    I saw an interview with a Democratic House member in a gerrymandered Florida district and she sounded very happy that she now has a seat that is guaranteed to go ‘D’. And to hell with everything else.

    • Sam Wang

      Not quite right…but this raises an important point that is not specific to either party. When a partisan gerrymander benefits one party overall, a few members of the opposing party benefit. Whirlaway is making the same conceptual error that Justice Kennedy made in LULAC v. Perry.

      It is important to keep partisan and individual benefit separate – and to have a standard that does not rely on individual lawmakers to make sure it sticks.

  • Phoenix Woman

    You’ve got a lot of readers! The gerrymandering question has over eleven thousand votes. No other potential question has even a thousand votes.

    Consider this an unscientific gauge of the size and activity of your readership.

    • Lil Sister

      I don’t think you’re seeing all the potential questions. The gerrymandering question is pretty high on the list but not at the top. Get your friends and relatives to vote too.

  • calvinhobbesliker

    Why is the PA Senate margin Toomey +1 when the last 3 polls are McGinty +5, +3, and -1? All other polls ended more than a week ago.

  • we_are_toast

    The problem is, how do you think both candidates will respond? Kind of like asking if they favor kicking puppies. Trump will blame gerrymandering on Miss Universe, and Clinton will denounce it and pivot to Citizens United.

    But this issue certainly underscores how CRITICAL the next Supreme Court nominee will be.

  • shma

    Sam, do you have the snapshot probabilities of the individual states somewhere on the site? (the ones used to build the daily histogram, I mean).

  • Marty Schiffenbauer

    Republican gerrymandering was enabled by their control of state governments. And the major factor allowing Republicans to dominate state governments, even in states traditionally voting for the Democratic presidential candidate, is the exceptionally low turnout of traditional Democratic Party voters in non-presidential elections (e.g., 2010, 2014).
    There is one strategy the Democratic Party can adopt to increase turnout of their voters in state elections. A number of states, including battleground states, allow citizens to use the initiative petition process to amend state constitutions. In such states, Democrats need to use that initiative process to switch off-year state office elections, both gubernatorial and legislative, to coincide with presidential elections.

    • Sandlot

      The argument about Democrats not turning out during mid-term elections is not accurate, rather it is the party that holds the presidency that does not show up. If you look at the last 27 mid-term elections (14 defended by D, 13 by R), there have only been 3 instances where that party did not lose seats in the House. The median loss is 28 seats. The Senate is slightly more durable. However, there were still only 5 instances of gains, 2 of no change, and 20 of losses. The median loss was 5 seats.

    • Matt McIrvin

      Hmm… maybe the modern pattern appears as it does mostly because one of the two Bush-era midterms (2002, conducted in the shadow of 9/11) was such an anomaly? So the other one, 2006, which was pretty much a landslide for the Democrats, then looks like the exception rather than an application of the rule.

  • Kevin

    Gerrymandering takes place at the state level; as far as I can tell there is literally nothing the president can do to impact it one way or another. It’s a good question, but not for this forum. What am I missing?

    • Sam Wang

      Good point, though I had in mind the fact that there’s a Supreme Court appointment pending. The topic is likely to come up in the next 1-2 years. Or is that too inside-baseball-y?

    • Pechmerle

      I went over and voted fot this Q. But I do tend to think it will sink like a rock if asked. Only a tiny fraction of U.S. public has any awareness of connection between S. Court decisions. & legislative gerrymandering. (To paraphase Stalin, how many votes do readers of the Stanford L. Review have?) The public does better when the issue is presented on its own via initiative\referendum. But % of eligible voters who vote on these measures is well below votes for president.

    • Richard

      The current high level of gerrymandered districts has happened because the Republican party has been remarkably successful relative to the Democratic party at the state level. The Republicans control about 30 states sufficiently to gerrymander, and the Democrats only about a dozen. That includes California, but the advantage is still solidly with the Republicans. Outside of the Northeast and the West Coast, Democrats are scarcely ever elected except in urban (not suburban) districts at the state level.

    • WildIrish

      @Richard
      “The current high level of gerrymandered districts has happened because the Republican party has been remarkably successful relative to the Democratic party at the state level.”

      I would say exactly the opposite: The Republican party has been remarkably successful relative to the Democratic party at the state level precisely because they have zealously gerrymandered districts at every opportunity.

      I saw it happen in NC with my own eyes: There was a Democratic majority at the state level, with a Democratic governor, and they did NOT try to gerrymander the districts. As soon as Republicans gained enough seats in 2010, they immediately redrew the map to give themselves a super-majority, and even sneered at the Democrats for being too stupid to do it when they had the chance.

      It defies logic that a state which votes 51% Democrat would have so many elected Republicans. However, they are now in “safe” seats due to gerrymandering and are unlikely to be voted out any time soon.

      And the moderate Republican mayor of Charlotte became a hard-right governor as soon as he got elected, and we’ve all seen how well that’s worked out for NC.

    • Ben Alpers

      It’s true that even truly national electoral reform wouldn’t much involve the executive branch, as it would likely have to come via Constitutional amendment. That being said, imo the current inside-baseball-y quality of all electoral reform issues could probably only change by having a President declare them a national priority.

    • Jacob

      I think it can be useful to get the candidates on the record regardless of their own ability to mandate a change (or benefit from it). They will lead their party if elected.

    • scdennis

      Richard: California used to be gerrymandered to protect incumbents. That was changed by referendum in 2008. California electoral boundaries are now established by a citizens’ commission.

      In an early result, the Democrats picked up four Congressional seats in 2012. Previously, only one congressional seat had changed hands in the previous ten years. Also, the Senate race this year is between two female Democrats as a result of the new jungle primary system.

  • kenn_sun

    I see the gerrymander is a very respectable #8 on the list. I think this is well worth asking if only to increase awareness.

  • Terry

    Is there a mistake on the Senate for PA (Toomey up)?

  • Rick Howard

    Finally, the M/M is up to Clinton +2.8. Fly, M/M, fly!

  • Matt

    Hi Dr. Wang, there’s something wrong with your 270towin.com map. It’s not showing up on the sidebar and if I click the link, it redirects to the main site.

  • Michael

    David Axelrod, Obama’s main guy, citing the extreme polarization that y0u often cite, said that in ’08 and ’12 the public polling swung more than their private polling. He also said the private polling is better resourced and has a lot more data to base their models on. So, is it a reliable clue to watch where the candidates go, and where they spend money, as a reflection of their private polls?

    • 538 Refugee

      I’m not sure if private polling is showing better in Ohio for Clinton or not but it seems they are going to start doing some higher profile visits after a long absence. The advertisements have always been here though. I’d say they could be hoping to make Trump play defense but the fund raising gap he has supposedly been closing isn’t showing up as ads here. I heard one commentator ticking off a list of basic things that Trump doesn’t seem to comprehend. I’ve wondered if he always thought he could funnel unspent funds to his charity. We know where that would end up. Maybe blow any leftover paying for events at his properties on election night?

    • Sam Wang

      Certainly in the Clinton campaign and in Senate/House downticket organizations, I expect candidate movements and ad buys to reflect private polling. The Trump campaign does not appear to be all that data-driven.

  • Mark F.

    I’m pretty sure that the smaller population states in both the North and South wanted the Senate to prevent the large population states from steamrolling things through Congress. The 3/5 compromise actually gave the South fewer House seats than they would have had if the slaves were counted as a full person.

    In any case, there is zero chance of abolishing the Senate, but it is possible to have less gerrymandered Congressional districts.

Leave a Comment