5

Many western democracies limit the power of their politicians to control the electoral boundaries to avoid the possibility of gerrymandering. UK electoral boundaries, for example, are redrawn when required by an independent commission. Many US states, however, put the boundaries entirely under the control of elected politicians and this often results in some fairly odd boundaries.

Some regard this as self-evidently bad. For example, the following argument is made on www.fairvote.org:

Redistricting encourages manipulation of our elections by allowing incumbent politicians to help partisan allies, hurt political enemies and choose their voters before the voters choose them. The current process is used as a means to further political goals by drawing boundaries to protect incumbents and reduce competition, rather than to ensure equal voting power and fair representation.

Some states have had non-political control for some time (Iowa, for example) and others have recently enacted citizen driven initiatives to remove the power from politicians (eg California). (see this CNN report for some examples).

It seems likely that gerrymandering will strengthen incumbents. But what does the evidence say? How much does it increase electoral security for incumbents?

Christian
  • 33,271
  • 15
  • 112
  • 266
matt_black
  • 56,186
  • 16
  • 175
  • 373
  • 3
    I'm confused by what is being asked, sorry. This seems to be fishing for a positive outcome, any positive outcome, to an act which is against the *values* of many/most people. The act of putting political self-interest above the right to equal representation (a moral value, not a scientific fact) makes the act of gerrymandering ethically wrong (to many), no matter what the unintended outcomes. Asking the equivalent to "Surely we can think of some good outcomes of murder?" seems to miss the point. – Oddthinking Oct 18 '12 at 22:34
  • 1
    I was hoping to focus on the demonstrable effects of gerrymandering in order to avoid the debate on values which would be off topic. A concrete link to more secure incumbents or stronger polarisation should be statistically demonstrable (if true) and doesn't require a value judgement. – matt_black Oct 18 '12 at 22:41
  • Perhaps we can narrow this down then to one of those two, and lose the "ensuring minorities get (disproportionate) representation, and other positives" angle? You seem to have already answered the polarisation claim yourself, so should the question be "Is gerrymandering successful at securing incumbents in office?"? – Oddthinking Oct 18 '12 at 22:46
  • I think polarisation needs to stay in. I selected a quote to demonstrate that not all commentators agree with the obvious answer, but there are many alternative views and it isn't obvious which is right. If the search for possible positives is distracting I'm happy to remove it or for it to be removed. – matt_black Oct 18 '12 at 23:04
  • I created a chat session to see if we can improve the question: http://chat.stackexchange.com/rooms/6165/gerrymandering-question – Oddthinking Oct 19 '12 at 09:11
  • 1
    Far too many assumptions in this question, and far too much vagueness in the expected answer. Is gerrymandering "bad"? Not something we can answer here. – DJClayworth Oct 19 '12 at 15:51
  • @DJClayworth I'm not asking whether gerrymandering is bad: I'm asking whether its influence on the politics of the states that allow it can be quantified. Both the strengthening of incumbents and the polarisation of their views are, in principle, *measurable* so we can pose the question without value judgements. – matt_black Oct 19 '12 at 16:29
  • 2
    While it is not evidence in the Skeptics.SE sense, the very way that politicians squabble over redistricting suggests that *they* believe it matters. Two possible avenues of research would be looking at incumbent retention rates as a function of time since the last redistricting (probably a small effects, but there is potentially a lot of data) and compare the retention rates in the (say) decade before and after California's move to a "non-partisan" commission (if we have any idea what non-partisan means and how it can be assured on a on-going basis). – dmckee --- ex-moderator kitten Oct 19 '12 at 19:01
  • @dmckee I think you are heading in the right direction with the evidence you suggest. but i think non-partisan is easier to define than you think: any group where the incumbent politicians can't influence the boundaries or the composition of the committee will be a big change. We already have models of how to do this both in the US and from other countries. – matt_black Oct 20 '12 at 09:17
  • there is no such thing as a nonpartisan committee, there are just committees that the masses have less direct control over. putting redistricting in the hands of a "nonpartisan" commission is just sticking your head in the sand, political districts are far to valuable to ever not be politicized. – Ryathal Oct 22 '12 at 17:32
  • @Ryathal Many americans and most of the rest of the world disagree with you. And, if you really wanted to guarantee non-partisan redistricting, you could allow a computer to do it based on simple rules about fairness. There are several algorithms already available. – matt_black Oct 22 '12 at 17:37
  • [This video](http://www.youtube.com/watch?v=Mky11UJb9AY) shows some of the problems with gerrymandering. [This one too](http://www.youtube.com/watch?v=s7tWHJfhiyo&t=3m55s) – Sam I Am Oct 22 '12 at 21:41
  • @SamIAm A transcript of that video would make a very good answer! – matt_black Oct 22 '12 at 21:50
  • @matt_black not really, you asked for **actual evidence** whereas the video I linked to is just speculation, showing the potential problems, not hard data on where this has actually happened. – Sam I Am Oct 23 '12 at 00:26
  • @matt_black The committee will start less partisan (maybe *much* less partisan), but if getting the "right" people on it is an effective below-the-radar strategy, people *will* try to corrupt the process. I'm not as certain as Ryathal about the effect of it, but also not a sanguine as you seem to be. – dmckee --- ex-moderator kitten Oct 24 '12 at 17:03
  • @matt_black As for letting a computer do it, you'll have to build it as some kind of optimizer, and the parameters of the algorithm are subject to fiddling for partisan ends. So who gets to decide and how do you pick them? And we're back into the the same mess. – dmckee --- ex-moderator kitten Oct 24 '12 at 17:05
  • @dmckee I'm almost inclined to agree with you except for the simple facts that most countries seem to have solved the problem and even some american states have done it without the corruption you expect. And demonstrable, transparent algorithms already exist for computerised districting and could be implemented transparently so the public could detect any attempts to fiddle with the rules. – matt_black Oct 24 '12 at 22:09
  • @matt_black What do you mean by "how much"? What measure can be used to describe the magnitude gerrymandering's effect? –  Jun 21 '13 at 19:16
  • @Sancho One could, for example, measure the change over time in reelection probability as boundaries change. And, as rules differ state to state (with some legislating to have boundaries drawn by independent teams--as in most democracies--for example) there might be some natural experiments and actual stats out there. – matt_black Jun 21 '13 at 20:23

1 Answers1

8

Yes, strategic redistricting (gerrymandering) can make incumbents more secure.

I'm basing my answer on the Redistricting series by ProPublica.

District boundaries can have an effect on election outcomes:

The maps used in Texas’ next elections could impact the balance of power in Congress. They will likely determine whether the four new congressional seats awarded Texas via the census will be held by Democrats or Republicans. (Actual Winner Unclear in Supreme Court’s Ruling on Texas Redistricting)

Also:

[H]ow the district lines are drawn is likely to determine whether those additional seats will be won by Democrats or Republicans -- and how big an impact minority voters will have in deciding who the new representatives will be. (Will the Supreme Court Strike Down Part of the Voting Rights Act?)

The line drawing can be done in a way so as to intentionally benefit a particular party or individual:

As it has done before, the Republican-dominated state legislature drew maps that heavily favor Republicans. (Ibid.)

[P]oliticians have been drawing district lines for their own advantage since the days of the founding fathers, when Patrick Henry gerrymandered a Virginia district to try to keep James Madison out of Congress. (Is Partisan Gerrymandering Unconstitutional?)

The right lines can all but guarantee an incumbent a decade's worth of electoral success, or alternatively can help send others into retirement. (Redistricting, A Devil's Dictionary)

Community
  • 1