UC Berkeley News
Today's news & events
News by email
For the news media
Calendar of events
Top stories
Untitled Document
Web Feature

UC Berkeley Web Feature

Berkeley experts weigh whether the Supreme Court will curtail election gerrymandering — and if it should

– Gerrymandering, the redrawing of congressional districts to ensure that elections favor a particular political or ethnic group, has been around almost as long as the Constitution itself. The term was coined when Eldridge Gerry, one of the signers of the Declaration of Independence, sought to engineer his reelection as governor of Massachusetts in 1811 by rearranging voting districts so crudely that one resembled a salamander.

Redistricting in Gerry's era, however, was more guesswork than guarantee. Using today's computer software to slice and dice voters' demographic data with razor-sharp precision, state legislators can practically dictate election outcomes for their districts.

Gerrymandering 101

Packing - Stuffing most of one party's voters into as few districts as possible
Cracking - Dispersing blocs of one party's voters into several districts so that they become a harmless minority
Kidnapping - Drawing district lines so that two incumbents from the same party must now run against each other

Both Democratic and Republican parties are well-versed in gerrymandering, but many observers believe that the practice has become even more cutthroat since the 2000 election gave Republicans a slim majority in the House. In Florida, Pennsylvania, Ohio, and Michigan, states for which Al Gore and George W. Bush each took half the vote, the Republican party has gerrymandered districts to capture 51 of 77 House seats. And in 2002, only two years after the traditional U.S. Census-driven redistricting, Texas Republicans sought to redraw the lines yet again — in some cases, in shapes more like centipedes than salamanders, causing a group of 51 Democratic congressmen to stall the plan's passing by fleeing the state.

Like the mythical lizard it is named after, said to be able to endure fire unscathed, gerrymandering has survived several halfhearted legal attempts at curtailment. Until this December, the most recent was 1986's Davis v. Bandemer, when the U.S. Supreme Court ruled that political gerrymandering that violates the equal protection clause is "justiciable," meaning subject to the court's scrutiny. Now the Supreme Court is taking another swing at gerrymandering. In Vieth v. Jubelirer (opening arguments took place December 10), some Democrats argue that, as designed by Republicans, the redistricting plan for Pennsylvania following the 2000 census would cost them an "unconstitutional" number of seats in the House: on paper, the new plan would give Republicans 12 or more of 19 seats in a state where voters are almost exactly divided between the two parties.

Politicians and legal scholars alike are watching the Pennsylvania case closely. Will the Supreme Court use Vieth v. Jubelirer to smack the hands of both parties and tell them to play more fairly? To get a sense of what's at stake politically, the NewsCenter interviewed UC Berkeley professor and redistricting expert Bruce Cain. And in a second interview, to be published on Tuesday, January 20, constitutional law scholar Jesse Choper discusses the legal context of court intervention.

 Bruce Cain
Political science professor Bruce Cain, director of UC Berkeley's Institute of Governmental Studies. (Photo by Peg Skorpinski)
Part One of Two
Bruce Cain is the Robson Professor of Political Science and director of UC Berkeley's Institute of Governmental Studies. Cain has consulted on redistricting for many legislative and citizens' groups and is the author of "The Reapportionment Puzzle" (1984), and coauthor of "The Personal Vote" (1987) and "Congressional Redistricting" (1992), now in its eighth printing.

Powell: Has gerrymandering changed in the last 10 years?

Cain: What's changed is the political context of redistricting. We're in an unusual period in which the Democratic and Republican parties are at relatively even strength both in terms of their support in the electorate and of their margins in the House. And so every single seat matters. Most seats are safe or competitive because there's not much else you can do with them. You can't make a seat in Berkeley or Oakland competitive for the Republican party; similarly, in the southern part of Orange County there's no way you can make it competitive for the Democratic party. But on the margins, there are a lot of seats that could swing one way or the other depending on how the lines are drawn. And since the parties are so closely matched, the marginal differences that come about because of redistricting are more important than they were in the '60s, say, when Congress was heavily dominated by the Democratic party.

Is this why Republicans in Texas redistricted in 2002 instead of waiting for the 2010 census?

Absolutely. Whether you're talking about Texas or Colorado or Pennsylvania, you're talking about those few states where the conditions are right for the Republicans to use redistricting as a tool to increase their party size.

And those conditions are where the state's governor and a majority of the legislature belong to the same party.

'Most voters don't know anything about redistricting and don't care. They don't see the lines.'
-Bruce Cain, UC Berkeley political science professor
Yes. You need to have control of the legislature and control of the governorship. And in Texas, for example, the congressional delegation before the redistricting was relatively close between Democrats and Republicans, yet the state is overwhelmingly a Bush state. So the potential to transform to a more Republican tilt is very high, and the opportunity is there because Republicans control the legislature and the government. In most states you have divided government. And in some states like California, even though you have one-party control, there's not much potential for the Democrats to add many more seats to their congressional delegation.

In the Pennsylvania case now before the Supreme Court, Democrats are arguing that it is "unconstitutional to give a state's million Republicans control over ten seats while leaving a million Democrats with control over five." Is it unconstitutional or merely politics as usual?

Some Democrats are arguing that. Back when Phil Burton drew the lines in California [as a state congressman in 1981], it was Republicans who were claiming unfairness and wanted the court to intervene. The shoe was on the other foot in Indiana in the 1980s or Pennsylvania in the year 2003. The aggressor claims immunity from the court and the victims want the courts to protect them.

Why do you think the court agreed to hear this case now?

There are two theories floating around. The first is that the court's seriously interested in getting involved in adjudicating these gerrymandering cases. Theory No. 2 is that the court wants to drive a stake into the heart of the 1986 Bandemer decision, which extended the prohibition on racial gerrymandering to political gerrymandering. But in the intervening years since then, there have been no rulings that have used Bandemer to deem a redistricting plan unconstitutional.

Isn't that because Bandemer's standard of determining unconstitutionality - that a political party has to be "entirely shut out of the process" - is so vague as to be useless?

It's not so much vague as it is tough. Basically the court set the threshold very high - you have to show that this party has been systematically excluded from power, has no way to address this inequity, and that the gerrymandering is part of this exclusion. That's not true in most cases, even in hard-core Democratic and Republican states. California is a hard-core Democratic state, but our new governor is a Republican. So such exclusion, which was borrowed from the racial gerrymandering cases, is a very, very high standard to apply to political gerrymandering. No plan has risen to the level of constitutional significance since the Bandemer definition. So it's quite possible that the court wants to say explicitly, "Don't give us any more of this nonsense, we set the bar very high on purpose and we're not going to be drawn into this because it's a fundamentally political fight."

You don't foresee the court ruling against political gerrymandering, then.

I was surprised about its recent campaign-finance decision, so I could be surprised again. But if I were to bet, I'd bet that the court backs away from getting involved in all this.

But should it? If gerrymandering essentially lets the legislature decide who gets elected, doesn't that undermine the legitimacy of the electoral process?

I think it does, but actually, it's not the voters who want competition, it's political scientists like me who think it's better for the system to have competition. Most voters don't know anything about redistricting and don't care. They don't see the lines. And if you go to the reapportionment hearings, most voters will say "I want to be in the district with more people like me." Democrats want to be in Democratic areas, and Republicans want to be in Republican areas, and they're not interested in competitive seats.

And overall, then, gerrymandering tends to balance out on a state-by-state basis in the House?

Not completely. In 2000 there were more states where the Republicans had control than Democratic states. But on the other hand, there were more states in which neither party had control than there were ones with single-party control. There were more states in which you had a bipartisan gerrymander than states where you had a partisan gerrymander, and that's always been the case.

What do you think about the argument that gerrymandering increases political partisanship by making it unnecessary for politicians in safe seats to "appeal to the center"?

That's where we get into the confusion that people have. There are really two types of redistricting: partisan and bipartisan. In the first type, the majority party will create slightly less safe seats in order to stretch its strength more efficiently. You don't want to win seats 70-30, you want to win them 57-43 so you can spread out your voters and get as many seats as possible. So you take a little risk and create slightly more competitive seats.

In a complete bipartisan gerrymander, everyone's going to have wildly safe seats. So in effect, what's happening in Texas and Pennsylvania, if they're really doing it right, is they're creating some seats that are a little less safe in the interests of trying to expand the number of districts they get. And that has implications, because the really, really safe seats do tend to produce more ideological candidates. There's a confusion in the public discussion, however, because really safe seats require concentrating your strength, and that's not how you take seats away from the other party.

Roughly 400 of the House's 435 seats are considered "safe." Iowa, which has a neutral third party decide its redistricting, has by far the most competitive races. Why shouldn't all states use that model?

That system does seem to work pretty well in Iowa, but it would be a lot harder in New York or California. Iowa is unusual because it's such an ethnically homogenous state. I spent the last round of redistricting advising citizen commissions in San Diego, San Francisco and Arizona that were put together by initiative and involved appointing blue-ribbon panels. And yet politics were all over the process. People were influencing these commissioners; commissioners had their own agendas. Take California, where we have not only the Democrat versus Republican issue, but urban versus rural, white vs. nonwhite, voting rights issues, all kinds of things. The politics in this state are just extremely complex, and the level of suspicion toward any commission is going to be very, very high.

Why don't we just use computers to draw the most compact and contiguous districts regardless of demographics? What's the disadvantage?

This brings up the distinction between something being procedurally fair and substantively fair. Something can be procedurally fair because it's neutral, like flipping a coin. But that doesn't mean that it's going to be substantively fair, because what if your random line-drawing process really screws one party over another?

Coming Tuesday, January 20: Jesse Choper, professor and former dean at UC Berkeley's Boalt Hall School of Law, discusses the legal context of the gerrymandering case
Here's a good example: in the early 1990s, [then-California Governor] Pete Wilson hired a geographer from out of state to draw district lines randomly. Completely fortuitously, the district lines that he drew collapsed the seats of every female member of the legislature and put them in with somebody else. That's the kind of thing that can happen with a randomizing plan.

In the short term, yes. But isn't the idea that a neutrally driven process over time will end up being more fair?

It's easy for you as a writer to be that philosophical, but that's not how the politician or the politician's wife or the people who elected him are going to feel. As an academic, I can find abstractions to be appealing, but the reality is, when you do these things you can't think about the abstractions. It's not just the politicians who feel that way, it's also the voters.

More information