Skip to content

Don’t blame gerrymandering

Matthew Yglesias quotes Richard Cohen presenting a common misconception:

Reality is real. No amount of lofty rhetoric is going to change the way members of Congress are elected. Most of them come from exquisitely gerrymandered districts created by computers that could, if good taste allowed, part the marital bed, separating husband from wife if they were of different political parties. This system created districts that are frequently reliably liberal or conservative. The computer has deleted the middle.

I can’t disagree with Cohen’s first sentence above, but I part company with him after that.

When Gary and I looked at the data, we found that redistricting (“gerrymandering”) was not associated with a decline in competitiveness of Congress or state legislatures. Legislative elections have been gradually becoming less competitive, but they are typically more competitive after redistricting.

I’m not saying that “gerrymandering” is a good thing–I’d prefer bipartisan redistricting or some sort of impartial system–but the data do not support the idea that redistricting is some sort of incumbent protection plan or exacerbator of partisan division.

In addition, political scientists have frequently noted that Democrats and Republicans have become increasingly polarized in the Senate as well as in the House, even though Senate seats are not redistricted.

P.S. I’m not saying that gerrymandering is always benign; there are certainly some places where it has been used to make districts with unnecessarily high partisan concentrations. But, in aggregate, that’s not what has happened, at least according to our research. (Also, our above-cited research is over 15 years old, and it’s possible that things have changed since then, with the advent of computer-based redistricting. But I haven’t seen any evidence for such a claim. I’d certainly love to see someone replicate our 1991 and 1994 articles to include the data up to the present.)

P.P.S. Update with more recent research (by Abramowitz, Alexander, and Gunning) here.


  1. a foreign student says:

    I wonder whether it would be a good thing to move the present presidential electoral vote system from winners-take-all at state level to winners-take-all at congressional district level, with the 2 extra seats per state given to winners of the state. Is it fairer and make more people's vote count than the present system?

  2. Andrew Gelman says:

    No, I don't think that would be a good thing. AT and I have a paper on the topic.

  3. Peter says:


    Do you distinguish 'redistricting' from 'gerrymandering', and if so, how?

    When a state is redistricted, it may, clearly, become more gerrymandered, less gerrymandered, or about the same. In fact, in one redistricting, some districts may move in each direction in the same state.

    Measuring gerrymandering is tough – taking account of natural boundaries and so on – but one measure I've seen is the amount of text it takes to describe a district (e.g. start at XXX and XXX go one block (mile, whatever) north, turn east … etc). By this measure, I believe NY-12 is the most gerrymandered district in the USA. But gerrymandering in metro NY doesn't affect D and R at all….. all the districts except Staten Island are solidly D.

  4. sohbet says:

    with the 2 extra seats per state given to winners of the state. Is it fairer and make more people's vote count than the present system?

  5. Andrew Gelman says:


    I think of "redistricting" as the general term, and "gerrymandering" as a term used by people to describe certain kinds of redistricting that they don't like.

    See here for a discussion of one particular mathematical definition of compactness of districts.

    Sobhet: No, I think an electoral-college-by-congressional-district would be worse than the current system.

  6. Peter says:


    Interesting idea in the link, but I think you point out a fairly substantial flaw. One needn't even make up an example, one could look at Utah, where the districts are relatively compac, but are designed to divide Salt Lake City (which is Democratic) among all three CDs, giving Utah 3 Republican representatives.

    One possible remedy is some weighting of compactness with county lines. Thus Utah's diving of Salt Lake county into 3 districts would hur the current plan

  7. Ted Dunning says:

    Actually, real honest-to-god gerrymandering with the intent to steal elections often results in many districts being *more* balanced rather than less.

    The idea is to put your opponents into a few massively safe districts while your side gets a bunch of slightly favorable (to your side) districts.

    This can wreak real havoc for you if the overall tide shifts because a large number of your barely safe seats can switch all at once.

  8. Felix says:

    Maybe some of you will enjoy this:

    I certainly learned a lot about this issue as an international student.
    Not sure if the process depicted in the game is really what is going on with gerrymandering, but informative nonetheless, I would think.