Quantifying the Effects of Gerrymandering

All over the country we’ve seen the effects of partisan gerrymandering by Republicans after they took control of so many state legislatures in 2010. A new study out of Duke University quantifies the effects of that gerrymandering and the results are exactly what one would expect.

In 2012, Democratic U.S. House candidates in North Carolina received 81,190 more votes that Republicans. Republicans received just under half of the votes earned by the two parties. And yet, the GOP walked away with 9 of the state’s 13 congressional districts. So, despite the fact that they earned just over 49 percent of the two-party vote, Republicans won nearly 70 percent of the state’s congressional seats.

Common sense dictates that the legislative maps that could produce such a result must be deeply flawed — and that they must be biased towards Republicans, the same party that controlled both houses of the state legislature when these maps were drawn (although the state had a Democratic governor at the time of the redistricting, the governor has no veto power over congressional maps). A new study by Duke Mathematics Professor Jonathan Mattingly and undergraduate Christy Vaughn seems to confirm this insight. Their study confirms that it is highly unlikely that a fair redistricting process would have produced a map as skewed towards one political party as North Carolina’s congressional map is.

Mattingly and Vaughn’s study redrew numerous random congressional maps, all of which complied with three rules: the districts must be “connected,” they must “come as close as possible to having [an] equal number of people,” and “they should be as compact as possible.” They then ran eight different simulations, some of which gave greater preferences to compact districts over equal population, while others placed greater emphasis on maintaining exact population. Seven of the eight simulations did not produce a single map where Democrats won less than five congressional seats, assuming that every voter who cast a vote for a Democrat or a Republican in 2012 would have cast the same vote under the simulated maps. The one simulation that did produce a handful of outlier maps where Democrats won only four seats did so “in less than 5% of the samples.”

Thus, the actual result of the 2012 elections — four Democratic congressional seats in North Carolina — did not even show up in all but one of Mattingly and Vaughn’s simulations. In the simulation where it did arise, it did so only in a few unusual cases. It is exceedingly unlikely that North Carolina’s GOP-friendly maps could have arisen organically. Rather, as Mattingly and Vaughn demonstrate, they are almost certainly the product of a legislature that carefully designed the maps to produce a desired result. The study’s authors argue that this result cries out for an independent check on redistricting — “The fact that the election outcomes are so dependent on the choice of redistrictings demonstrates the need for checks and balances to ensure that democracy is served when redistrictings are drawn and the election outcome is representative of the votes casted.”

The same thing happened here in Michigan, where Democratic candidates for the U.S. House got 515,716 votes (49.15%)

while Republicans got 1,463,854 votes (47.47%), but Republicans nonetheless control 9 of the state’s 14 seats in Congress. In the statewide election for Senator, where gerrymandering doesn’t matter because it’s only the total state vote that counts, the Democrat won easily. This is why the redrawing of districts should not be done by elected officials at all. It should only be done by a non-partisan board of demographics experts.

"I've seen many trolls trying to distract from Moore by saying what about Franken"

How to Think Critically About the ..."
"I have seen zero (serious) people who claim Franken and Moore are both equally as ..."

How to Think Critically About the ..."
"Pretty much every word of this sounds like a verbatim defense from a Roy Moore ..."

How to Think Critically About the ..."
"If that's the kind of story he has to tell his wife to get in ..."

Warning: Alex Jones is Going to ..."

Browse Our Archives

Follow Us!


What Are Your Thoughts?leave a comment
  • Trebuchet

    It should only be done by a non-partisan board of demographics experts.

    With major aid from computers, like in the studies.

  • Hercules Grytpype-Thynne

    a non-partisan board of demographics experts

    Here’s the problem: you’ll never get Republicans to go along with this, because to them the word”expert” means “liberal elitist.”

  • Pierce R. Butler

    Our esteemed host is apparently quoting from this ThinkProgress report, typos and all.

  • http://en.uncyclopedia.co/wiki/User:Modusoperandi Modusoperandi

    This is why the redrawing of districts should not be done by elected officials at all. It should only be done by a non-partisan board of demographics experts.

    How can it possibly be “non-partisan’? I mean, from the examples above it would clearly shift the advantage to the Democrat party!

    In any event, the GOP needs something to counteract all those Democrat voters. Isn’t that fair? You Liberals are for being fair, right?

  • Childermass

    Someone needs to make an open source program that takes the basic demographic and geographic data of any state and spits out fair districts without respect to protecting any party, ideology, or incumbent. And then people should demand that those be the actual districts. Given that many states both red and blue have means of getting measures on the ballot by means of petition, this could be done.

  • http://www.facebook.com/profile.php?id=1246980039 caseyboucher

    There already is an idea for mathematically drawing congressional districts. It’s called “shortest splitline”.

    Here’s what Michigan would look like:

    http://rangevoting.org/Splitline2009/mi.png

  • Crimson Clupeidae

    Oh, the software already exists. I use something very similar in what I do (finite element modeling). Just a few tweaks, like what are mentioned in the article, to make it favor equal population splits, or something similar, and it would be easy to implement on a nationwide basis.

    There are some potential drawbacks that I have seen mentioned, but none of them outweigh the current mess.

  • mildlymagnificent

    Someone needs to make an open source program that takes the basic demographic and geographic data of any state and spits out fair districts without respect to protecting any party, ideology, or incumbent.

    Even without super, fancy computer programs Australia’s Electoral Commission has been doing this since forever. The political parties can make submissions when boundaries are under review. When the results come out, they can appeal to the courts on the grounds that the result is unfair. When the appeal is decided, it’s over.

  • Gvlgeologist, FCD

    This is also why the attempts of some Repub state legislatures to award Electoral College votes on the basis of legislative districts are occurring.

    http://thinkprogress.org/election/2014/12/02/3598333/michigan-vote-rigging-hearing/

    Of course, the suggestions to do this are only in Democratic-leaning states. A state like Texas would still be winner-take-all.

  • scienceavenger

    Clearly the fair and balanced solution would be halfway between the current arrangement and one done by “nonpartisan” “experts”. Someone has to stand up to experts.