September 20, 2004

Redistricting

Posted by Brian

Kevin and Matt are talking redistricting, with Matt favouring proportional representation on the grounds that it would introduce intraparty competition into American politics. This is rather odd - it’s only been a few months since the Presidential primaries, which are the most vigorously and open contested intraparty political fights in the world outside of the New South Wales Labor Party. And any experience with internal Labor (or Labour) Party fights does not immediately make one think it would make the world a better place to expand that kind of fighting.

But I didn’t want to make a substantive proposal, just ask a procedural question. To the best of my knowledge there are only two classes of country where the electoral system, from drawing boundaries to determining ballot order to deciding whether there will be recounts and so on, is run by partisan appointees.

Class One: China, Cuba, etc., i.e. countries where it is known in advance how the results will turn out.
Class Two: The United States of America (with the honourable exception of Iowa).

Are there any other countries in Class Two, or is America unique in being a democracy where one of the prizes of victory is getting to be the umpire next time the game is played?

Posted on September 20, 2004 11:44 PM UTC
Comments

With a reelection rate (In the House of Representatives.) of over 98%, are you sure there are really two classes?

Posted by Brett Bellmore · September 20, 2004 11:52 PM

run by partisan appointees

Actually, I think in most places it’s voted on by state legislatures, which are not technically appointees (with federal judges occasionally doing the work, as in Texas between 2000 and 2002). The people in the House of Representatives have no direct input into the Congressional Districts, though as DeLay shows the indirect input can be fearsome. And I only just now started thinking about the districts for state legislatures and state senates—I assume those are probably drawn by the same legislatures they apply to, but I have never given a second’s thought to this fairly important process. Anyone know anything?

Also, some states don’t gerrymander because they only have one representative—though if you look at Utah’s map, you’ll see that three districts is enough for a pretty gruesome map. (Most of the population is in the little pseudopod that reaches into Salt Lake.)

Anyway, I agree with your points, just engaging in some freelance pedantry.

Posted by Matt Weiner · September 21, 2004 12:06 AM

Is it really 98%? As in less than 10 incumbents ousted per cycle? Seems awfully high.

Posted by djw · September 21, 2004 12:43 AM

In ‘98, there were only six incumbant House members defeated in their races for reelection. That’s not the total turnover, of course, because some members did retire, or die, rather than running for reelection.

Posted by Brett Bellmore · September 21, 2004 12:59 AM

As an example of perhaps a third class see http://www.elections.org.nz/esyst/repcomm.html, specifically the report from the Representation Commission. There are representitives of the Government and Opposition, and members from the administration - technically appointees but really part of the bureaucracy. I would suggest that the New Zealand model of MMP has been a good example of increasing third party representation and maintaining civil government.

Posted by Are Bee · September 21, 2004 01:13 AM

1998 represented the recent high-water mark of the incumbency advantage. In other years the percentage of incumbents reelected is between 87% and 98%. One other thing to remember: incumbency reelection rates don’t include House Members who don’t run, for whatever reason. So the net number reelected is closer to 80%. I can’t find the numbers per election at the moment, but in 1998 395 of 435 from the previous Congress were reelected. While I agree that the rates of reelection are dispiriting, they are not a new phenomenon and therefore don’t explain current political problems. Incumbency rates have been high for decades; the lowest rate of reelection is 87% (since 1962).

Posted by Sunita Parikh · September 21, 2004 01:24 AM

Actually, I think that this aspect of the American system to some degree accounts for its stability. It probably also explains the dominance of two parties. Elections in the United States can’t be won suddenly. Rather, they are the culmination of a long-process that involves winning local elections, controlling state legislatures and governorships, influencing redistricting, electing the state election officials (Kathleen Harris) and so forth. The current Republican domination is the result of a thirty-year recovery effort rising from the ashes of Nixon’s resignation. In 1974, the Republicans controlled a small fraction of the state legislatures, they had something like 12 governors (that number may be a bit low), and the Democratic majorities in Congress were so dominent that it was said to be “veto proof.” That process was not reversed with Reagan’s election in 1980 — it took until 1994 for the Republicans to win control of the House of Representatives (which they had not controlled for more than 40 years!), and that victory was to some degree the product of Republican progress in the state legislatures before the 1990 redistricting. The point is, it takes a very long time to build a true electoral majority in the United States, and those who want to change the system to speed the process of electoral change do not, it seems to me, generally acknowledge that it will make American politics less stable.

Posted by Jack · September 21, 2004 01:25 AM

1998 represented the recent high-water mark of the incumbency advantage. In other years the percentage of incumbents reelected is between 87% and 98%. One other thing to remember: incumbency reelection rates don’t include House Members who don’t run, for whatever reason. So the net number reelected is closer to 80%. I can’t find the numbers per election at the moment, but in 1998 395 of 435 from the previous Congress were reelected. While I agree that the rates of reelection are dispiriting, they are not a new phenomenon and therefore don’t explain current political problems. Incumbency rates have been high for decades; the lowest rate of reelection is 87% (since 1962).

Posted by Sunita Parikh · September 21, 2004 01:25 AM

1998 represented the recent high-water mark of the incumbency advantage. In other years the percentage of incumbents reelected is between 87% and 98%. One other thing to remember: incumbency reelection rates don’t include House Members who don’t run, for whatever reason. So the net number reelected is closer to 80%. I can’t find the numbers per election at the moment, but in 1998 395 of 435 from the previous Congress were reelected. While I agree that the rates of reelection are dispiriting, they are not a new phenomenon and therefore don’t explain current political problems. Incumbency rates have been high for decades; the lowest rate of reelection is 87% (since 1962).

Posted by Sunita Parikh · September 21, 2004 01:26 AM

Sorry about the triple post. Long time lurker, first time poster. Obviously unfamiliar with the time lag.

Posted by Sunita Parikh · September 21, 2004 01:29 AM

What Matt Weiner said, the US is a very big country, and I see no reason that State politics should not have a very large second-order influence on national politics. If you want to influence redistricting, you should take strong interest in who your State Representative and State Senator are.

Real representation should not be automatic, but a reward for political involvement and activity.
….
The 98% figure is horrible, but I am not sure how horrible. I opposed term limits. If we were to prefer our congressmen serve for say, at least six years on average, for the sake of experience, then we should look at turnover over ten or twenty year periods, not two.

Exactly how competitive would we like it to be? If the figure were a 50% turnover every election, we would have a nightmare.

Posted by bob mcmanus · September 21, 2004 01:29 AM

An interesting question is what level of turnover do we think would indicate a fair system? 98% retention rate suggests to almost everyone that there is a problem. I agree with Bob that a 40-50% retention rate would likely create chaos. What would a good target for the real world (in normal circumstances) be? 70-75% retention?

Posted by Sebastian Holsclaw · September 21, 2004 02:40 AM

There is a lot more variation in state laws governing redistricting than the discussion here presupposes. For example, Kentucky has a constitutional provision that limits the degree to which counties can be divided for purposes of representation. The 1992 state redistricting scheme was struck down by the Kentucky state supreme court, and a new scheme was created by the courts for the 94 elections.

Posted by David M · September 21, 2004 04:20 AM

Not redistricting, but Francois Mitterrand changed the system for French legislative elections on partisan grounds from 2-round ballots in 1-member constituencies to PR when the Parti Socialiste was weak on the ground that the Front Nationale would garner enough votes to fracture the right. He was right about that, but France has been living with the consequences ever since.

Posted by Chris Bertram · September 21, 2004 09:29 AM

The US state of New Jersey is another interesting example. In NJ, districts are formed by a redistricting commission formed of six appointees of each party, plus a thirteenth “tiebreaker” who is elected by a majority vote of the other twelve. I believe that in practice the thirteenth member has been an officially non-partisan academic or judge.

My understanding is that in the past, this has worked out by the Republicans and Democrats separately working out their own proposals, which are presented to the non-partisan member, who then chooses the plan he or she finds to be the most “fair”. At very least, this system will encourage the parties not to go overboard when formulating their plans, though they can probably get some considerable agreement on protecting the status quo…

Posted by Alex R · September 21, 2004 03:20 PM

Chris : it’s true that the move gave the Front National a short-term boost, by the virtue of suddenly having many (35)representatives in the French National Assembly.

But it was only between 1986 and 1988 : the law was quickly changed back to the old 2-round ballots and has stayed the same ever since. I’m not sure that Mitterrand’s original sin can really explain the continuing success of the FN.

Posted by Emmanuel · September 21, 2004 06:52 PM

Regarding the two classes of country: It is an article of perverse pride among certain conservatives and libertarians here that “the U.S. is a republic, not a democracy.” I’ve never quite understood why they think that that is a good thing, since the list of nations which are demonstrably republics but not democracies is similar to your class one, above, and congruent with the conservative/libertarian definition of evil.

Posted by HP · September 21, 2004 09:28 PM

the law was quickly changed back to the old 2-round ballots and has stayed the same ever since.

I thought that was so, but didn’t want to post that because I wasn’t sure. More grist to the mill, though, because the pattern was of Mitterrand switching to PR when it suited him and then back again when the old system became advantageous once more. As to the FN, of course there’s more than one explanation for their continued success, but the shift in voting system for tactical reasons enabled Le Pen to get beyond the “joke party” threshold.

Posted by Chris Bertram · September 21, 2004 11:36 PM

Although I’ve never heard a conservative or libertarian voice pride in the US being ‘a republic, not a democracy’ perhaps they mean a) the federal structure, which is hardly unique, though they may think it is, or b) more checks and balances on majoritarianism and more limited power to the federal gov’t?

I don’t know if the latter is actually true compared to other countries though … surely there must be other federal republics out there… Germany? Canada? How are their checks on majoritarianism?

Posted by Dubious · September 22, 2004 12:15 AM

Chris : I should have been clearer. In fact, the law was quickly changed back to the old 2-round ballots by the then right-wing majority in Parliament. Mitterand’s fault it wasn’t.

I don’t think we disagree that much on the reasons of the far-right electoral success in France. Mitterand’s ploy was important (though the FN was already on the rise, garnering 10% of votes in the 1984 European elections) but so were other, mainly economic and social, factors.

Posted by Emmanuel · September 22, 2004 09:10 PM
Followups

→ Proportional Response to Redistricting.
Excerpt: Geez, I went to all that trouble a minute ago. Pressed ‘Save’ and everything. Then I checked the news where I find this interesting snippet from Matthew Yglesias: [Washington Monthly’s Kevin Drum is] pondering redistricting reform. The best way to fix ...Read more at The 80/20 Club
→ Changing the House of Representatives.
Excerpt: The endless discussion of how to "fix" the U.S. electoral system continues at Crooked Timber. As usual, there is a call for some form of proportional representation, or at least, for more access by third parties. I've already stated my is...Read more at Left Oblique

This discussion has been closed. Thanks to everyone who contributed.