The solution, just to begin with, is to take the redistricting function away from the state legislatures, and from anybody else who is politically interested in the results. In Canada they have independent nonpartisan commissions to redraw the lines of ridings (districts). Let’s do it that way.
It’s not perfect, but no system could be. These concerns could be mitigated with periodic and automatic non-partisan reapportionment (e.g. plug in the population numbers to the algorithm and let it resort after every census).
I think its flaws are greater than something like my plan, but obviously opinions will vary.
I assume you think special interests don’t get enough attention today.
There have been sf stories where the legislature includes the representative from Apple, from GM, from IBM. That would surely happen in this system. Plus a representative from the NRA - though maybe just one would be better than the 100 or 200 we have now.
Back when Congress worked - in other words when more representatives had to worry about reelection - one job was to balance the various needs of their districts, try to make deals to bring jobs and pork to their districts, and worry about regional issues. Virtual districts assume that people mostly care about one thing. That’s mostly not true, and even when it is the things they care about change with time. This method would lead to the loudest mouths for any subject that can get a representative get even more power than today, and would totally destroy any possibility of compromise. if a subset of representatives have constituents who really care about topic X, and the rest have constituents who don’t, who is going to have more sway?
I’m for independent of algorithmic redistricting - it is closer to what the Constitution intended, is the least disruptive, matches demographic changes the best, and I think is likely to solve a lot of the problems.
In the early years most Representatives were elected at-large, meaning each Congressman represented the whole state even if there were more than one seat per state. That remained pretty common until, I think, about 1840, and persisted in some states even after that. Since 1967, federal law has banned at-large seats except in states that are entitled to just one Representative (currently Alaska, Wyoming, Vermont, Delaware, Montana, and North and South Dakota). That federal law could be changed in the future, I suppose. However, I’m not sure such a change would pass muster with the courts since a lot of voting-rights jurisprudence now assumes geographical districts.
Is there any mature, well-established democracy other than the US where the politicians draw the boundaries directly?
The bit of law in question is just this:
SEC. 2. No voting qualification or prerequisite to voting, or standard, practice, or procedure shall be imposed or applied by any State or political subdivision to deny or abridge the right of any citizen of the United States to vote on account of race or color.
There is a lot of meaning packed into that, and shouldn’t be thought of as solely referring to obvious restrictions like voting tests.
There was an amendment in 1982 that cleared up some of the ambiguity:
(b) A violation of subsection (a) is established if, based on the totality of circumstances, it is shown that the political processes leading to nomination or election in the State or political subdivision are not equally open to participation by members of a class of citizens protected by subsection (a) in that its members have less opportunity
than other members of the electorate to participate in the political process and to elect representatives of their choice. The extent to which members of a protected class have been elected to office in the State or political subdivision is one circumstance which may be considered: Provided, That nothing in this section establishes a right to have members of a protected class elected in numbers equal to their proportion in the
population.”.
This pretty obviously refers to gerrymandering-type processes. Note that intent doesn’t play into this at all–an algorithm which just happens to create unequal participation in the political process (due to some arbitrary decision) would run afoul of the law just as much as a politician intentionally drawing lines to dilute the influence of some disliked group.
The law doesn’t have any direct way of testing for unequal participation, so it would have to be decided in court by a weighing of the evidence.
The key point is you’re not assigned to a district by any outside agency; you choose your own district. So why would people stay in districts where they disagree with the dominant political views? The system would essentially end up where you’d have 435 members of the House of Representatives, each of whom has the support of every voter of their district. Because any voter who doesn’t agree with their representative would simply “move” to another district that has a representative they prefer.
I think special interests have a lot of power. But my system would at least make it more transparent because it would encourage groups to lobby to the voters rather than to the legislators. And if the NRA can find 350,000 people (and they probably could) who think the NRA’s agenda is the most important political issue, then I have no problem with that being a district.
I disagree with your argument that Congressmen used to have to balance the needs of all their constituents. The reality is that all any Congressman ever needed was to have the support of 51% of his constituents; the other 49% didn’t matter. Tens of millions of people are represented by a Congressman they didn’t vote for.
The other important issue is that getting elected is only the first step in the process; the real issue is enacting legislation. So a relatively small special interest might be able to get a representative elected to Congress. But that Congressman would still need to work with other Congressmen to get laws passed.
Because they could elect more representatives by splitting up. If the democrats have enough voters for 100% of 125 districts, then they have enough for 50% of 250 districts.
Mixed-member proportional representation (MMP) seems like a sensible compromise. Your district still gets to elect a representative with local ties, but the proportional component removes the gerrymandering effect.
If they move out of the district then the self-gerrymanderers will still be in control of 2 seats out of proportion to their numbers, it would just be 2 districts with no one else in them because they’ve moved out. The other choice would be to move more people in to take away their 60% majority but then the taker-overers are under represented because their district is 25% larger than the average.
Moving the redistricting to a nonpartisan commission won’t work, because where will you find nonpartisan people? What you’d actually get would be partisans who officially declare themselves independent.
And algorithmic methods are problematic, because for most standards of “ideal” districting, it’s really computationally difficult to find a solution you can know to be ideal.
The method I’m fond of is similar to the old trick of “Kid 1 cuts the cake, and kid 2 gets to choose which piece to take”. All members of the legislature take turns (perhaps in order of increasing seniority): On each legislator’s turn, he or she proposes a map. Then, all of the maps are scored according to some objective criterion (my preferred criterion is basically “shortest total length of borders”), and whichever map scores best is adopted.
Alternately, if you want to get rid of districts entirely, there’s Monte Carlo democracy: If your state has, say, 33 Representatives, then you choose 33 registered voters from the state entirely at random, and each of them can pick one Rep (they can pick themselves, if they’d like). This gives fair representation for all demographics.
Divide “districts” by Alphabetical order-last name.
District 1 would be AA-Ad, Dist 2 Ad–Bi, etc.
Mathematical division would be easy.
Might as well.
Okay. I’ll do it. For. all. 50. states.
The simplest answer without any dramatic changes to how the states do it now would be each State setting up independent boundary commissions, with a mandate that they use some sort of geographical algorithm and with some set rules. An example from Britain is their Boundary Commissions are required to draw districts within 5% of the UK “Electoral Quota” (the average number of expected voters in a constituency), and that it cannot be more than 13,000 sq. km. in area. There are some exceptions for unique situations (like island constituencies.) Obviously the areas and the percentage would have to be changed to reflect the size of the U.S. and the fact we have 435 to apportion among 320m people vs Britain’s 600+ among 60m people.
Once the Boundary Commissioners make recommendations, Parliament has to approve them. They can reject them, but cannot make amendments. State legislatures could have similar power, without the ability to amend we’re protected from the worst political shenanigans. To avoid procedural shenanigans I’d probably put a time limit on the legislature’s approval, after a certain amount of time has passed if the legislature consistently refuses to approve several versions of the drawn boundaries then eventually the first proposal just takes effect automatically.
Meh, it works fine in Britain or Canada. FPTP single-member districts are always going to struggle to be “ideal”, but you don’t draw the rules requiring ideal districts. You require a standardized population range per district, and put in rules that the commissioners have to use some form of geographic algorithm and adhere to some compactness guidelines.
With only bureaucratic authority to oversee GIS guys who study maps all day, these commissions aren’t going to attract people super interested in being political players. It’s going to attract lifetime GIS professionals and career geographers. With clear legislative rules in place the commissioners won’t have the power to gerrymander, and with the legislature disallowed amending the proposals they won’t be able to gerrymander either.
If certain constituent groups tend to “disproportionately group themselves” then that’s a systemic electoral disadvantage. Britain has some of that too, although theirs is exacerbated because their parliamentary approval process has no automatic kick-in. If Parliament disapproves then the old boundaries remain, even when demographics means some districts grow to have far more residents than others (right now this gives Labour a baked in advantage because population has been growing faster in Conservative districts than Labour ones, so a Labour vote is “worth more” now than a Conservative vote.) But the “geographic grouping” disadvantages you can’t really get rid of, that has to be seen as a feature of FPTP.
Maybe, but it’s really easy to find solutions that are vastly better than what we have now. Even dead-simple methods like the shortest-splitline method do a reasonable job; better solutions are also available and still only take a few CPU-hours to complete.
Are you suggesting that the candidate residency rule stays in place? (Which would make it the inverse of the classic British system, where a person from Mayfair could, say, represent a district in Newcastle-on-Tyne.) Where a flashmob could decide they are really angry at that guy from Joplin, so they invade his district from all over the country to drive him out? And how full can a district get before no one else can register in it?
Shortest-splitline is actually worse than what we have now. It tends to cut cities up into small slices and lump each slice in with a whole lot of rural area, which is exactly the sort of effect gerrymandering usually produces, and also tends to violate the requirements of the Voting Rights Act.
The VRA isn’t really set in stone, and it doesn’t require minority-majority districts, as is often erroneously stated. Minority-majority districts are more a creation of political agreements that didn’t pick up a lot of steam until the 1980s. Intent actually matters quite a bit in the VRA, and Supreme Court cases have said that some minority-majority districts themselves may be unconstitutional. The ruling in Miller v Johnson held that an “affirmative” racially gerrymandered district (meaning a district racially gerrymandered by politicians who intend it to benefit minorities aka deliberate minority-majority districts, that the Congressional Black Caucus and many other minority advocacy groups support philosophically), if its boundaries are primarily drawn to advance the interests of a minority race, and it doesn’t adhere to districting standards like geographic compactness and etc then the boundaries of the district must be subject to “strict scrutiny” and you must show a “compelling state interest” for creating the racially privileged district.
So not only does the VRA not require majority-minority districts, it actually says if you draw them primarily for that reason, against districting norms, you need to demonstrate a compelling state interest for them to be constitutional. The court has largely deferred to the legislatures in this area though, but has staked out a position on it that says you should at least be able to justify this stuff. A district drawn with its primary purpose being geographic compactness, in fact, would likely not violate the VRA.
The VRA is primarily, in regard to districts, concerned with ones that are intentionally drawn to dilute minority votes. You could actually argue in some sense majority-minority districts actually do this, but they were initially accepted as a sort of “affirmative action”, and later accepted by the Supreme Court but noted as requiring a compelling state interest to be kosher.
The VRA is also ‘simple legislation’, it’s not immutable etched in stone. If we were changing district laws it stands to reason if the VRA needed amended it would be.