A Better Way to Vote

By John B. Anderson and Robert Richie
Published May 17th 2004 in Legal Times
 The Supreme Court's decision last month in Vieth v. Jubilirer to uphold Pennsylvania's congressional redistricting plan demands that we confront an uncomfortable fact: We must either change our winner-take-all electoral system or accept the degradation of democracy.

No justice disputed that Pennsylvania Republicans had pursued an aggressive partisan gerrymander nor that legislators around the nation draw partisan and pro-incumbent gerrymanders that leave most voters without the means to hold their representatives accountable. But the Supreme Court majority saw no constitutional violation, in large part due to the paradoxes and political consequences inherent in efforts to fix the way we draw districts. Someday, the April 28 decision in Vieth may be seen for what it truly is: the death knell of a winner-take-all, single-member district system that allows --indeed, inevitably results in -- llopsided contests, distorted representation, and gerrymandered outcomes.

A Quick Tutorial

Every 10 years the U.S. Census releases new population figures, and elected officials set to work carving up the political landscape into new legislative districts with equal numbers of constituents. Armed with increasingly sophisticated software and more precise demographic data, incumbents in most states quite literally choose the voters before the voters choose them. Most voters are locked into one-party districts where their only real choice is to ratify the party's nominee.

In 2001, both Republicans and Democrats elevated incumbent protection to new heights. In California, incumbent U.S. House Democrats paid $20,000 apiece to a redistricting consultant for "designer districts." Republicans accepted this cozy arrangement in exchange for their own safe seats. The result in 2002 was 50 incumbent landslides, with no challenger winning even 40 percent of the vote. Nationally, only four challengers defeated House incumbents, the fewest
in history. For the third consecutive election, fewer than one in 10 House races were won by competitive margins.

The lockdown of the U.S. House has major repercussions for representative government. Men continue to hold more than 85 percent of seats, and the number of African-Americans has decreased from its high. Leadership is essentially fixed, with partisan control of the House changing just once since 1954. Members are polarized, with very little compromise and negotiation across party lines.

Gerrymandering clearly contributes to this dismal state of democracy, but it's less of a problem than the fact of single-member districts themselves. Gerrymandering can be effective only when voters are predictable. While most voters and districts typically have leaned toward one party, both the percentage of partisan voters and the number of partisan districts have risen sharply since 1994, when the Republican takeover led voters to see the House "in play." Today, few voters consider candidates' personal characteristics over their party

The combination of predictable voters, hardening partisan divisions, high-tech tools, and the general free hand given to legislators creates a perfect storm for no-choice elections -- one that fairer redistricting can tame only at the margins.

Political Matters

Enter the Supreme Court. The Vieth case was its first political gerrymandering case since Davis v. Bandemer in 1986, when the Court dangled the potential of political gerrymandering claims, but set the bar so high that plaintiffs have only once been successful.

The Vieth appellants challenged a partisan gerrymander that led Republicans to win 12 of 19 seats in 2002, with the potential for two more in 2004, despite a statewide tilt that has produced Democratic presidential victories since 1988. This Republican success in Pennsylvania -- as well as in Ohio, Michigan, and Florida -- inspired House Majority Leader Tom Delay to re-open redistricting in Texas, where Republicans rammed through a plan that may cost Democrats seven seats.

The Supreme Court's decision to reconsider political gerrymandering raised the question of whether a majority might see a means to protect voting rights and democratic accountability. The historic problem for the courts has been how to pick and choose among standards that inevitably have partisan consequences and frequently clash with one another. More competitive districts encourage accountability, for example, but can produce a biased outcome statewide or prevent fair racial representation. More compact districts seem sensible, but can result in a biased outcome where one party's total vote is geographically concentrated. And judges face the specter of Bush v. Gore-type accusations of partisanship whenever they rule in one party's favor -- as indeed their 5-4 decision in Vieth could appear partisan.

In that light, the best approach for the Vieth Court might have been dramatic intervention, on the scale of the 1960s reapportionment cases that required districts of equal population. Surveying the endless machinations of redistricting, the Court could have concluded that fair elections demand nonpartisan voter registration, election administration, and redistricting, and therefore prohibited use of all partisan data in line-drawing. While this one-time edict would have required new redistricting almost everywhere, it would also have set an unambiguous standard that would have kept judges out of future redistricting.

Given that no justice sounded ready for such a sweeping ruling, however, it should be no surprise that Vieth resulted in five separate opinions. Constitutional scholars Daniel Lowenstein and Jonathan Steinberg pointed out in the UCLA Law Review in 1985 that "there are no coherent public interest criteria for legislative districting that are independent of substantial conceptions of the public interest, disputes about which constitute the very stuff of politicy." Rational justices presented with the same facts can come to a variety of cconclusions.

Lowenstein and Steinberg questioned the justiciability of redistricting
claims over single-member districts. They argued that any claim based on individual rights is flawed because winner-take-all elections always leave many individuals in districts where they have little ability to elect candidates of their choice: "So long as winner-take-all district elections are allowed, the claim that gerrymandering infringes individual voting rights can have no merit." But if anti-gerrymandering theories must instead rest on group rights, they noted that the normal arenas for clashes between groups are the political branches, not the courts.

Rethinking Elections

Although not ready to concede the case against anti-gerrymandering suits for fairer single-member districts, we agree with Lowenstein and Steinberg that redistricting claims are most coherent when challenging the very concept of winner-take-all elections. Certainly the failure to do so undercut arguments for judicial action in Vieth.

The appellants creatively argued, for example, that democracy requires that if a group of voters -- in this case, Democrats in Pennsylvania -- win 51 percent of the total vote statewide, that group should have a real opportunity to win a majority of seats. But the logic of single-member districts does not support claims based on statewide fairness. As Justice Antonin Scalia in his plurality opinion points out, "In a winner-take-all district system, there can be no guarantee, no matter how the district lines are drawn, that a majority of party votes statewide will produce a majority of seats for that party."

Indeed, winner-take-all elections regularly violate the principle of majority rule based on statewide vote totals and can easily do so even after nonpartisan redistricting. Iowa Democrats have not won more than 20 percent of House seats in districts drawn by the state's much-vaunted nonpartisan system for more than a decade, although they regularly win the state's presidential race and at least 40 percent of the total congressional vote.

Only non-winner-take-all, multiseat district systems are designed to fully and accurately represent the majority of voters. Full-representation systems avoid the Vieth appellants' questionable suggestion that the rights of a group of voters constituting about 50 percent of the electorate are more worthy of protection than the rights of smaller but still substantial voter groups. And full-representation voting would address concerns raised in the Vieth opinions of Justices Scalia, Anthony Kennedy, and Stephen Breyer.

Kennedy writes in his concurrence that the First Amendment might offer some promise for gerrymandering claims on behalf of groups of voters receiving "disfavored treatment by reasons of their views." His logic would be even stronger if the First Amendment protected the representational rights of a range of voter groups, not just those large enough to win single-member elections.

Scalia notes that even "if we could identify a majority party, we would find it impossible to assure that the party wins a majority of seats -- unless we radically revise the States' traditional structure for elections." Exactly.

Breyer in his dissent most directly addresses winner-take-all elections and "why the Constitution does not insist that the membership of legislatures better reflect different groups of voters." He states that the Constitution demands "a method for transforming the will of the majority into effective government." But his subsequent discussion reflects a primitive understanding of comparative electoral systems, suggesting that the only alternative to single-party-majority governments, elected by single-member districts, is coalition-ridden, multiparty governments like those of Italy and Israel. In fact, there are other viable alternatives.

We Have History

Scalia's use of the word "radical"and Breyer's specter of coalition-ridden Italy point to an underlying problem: Rather than interpreting the Constitution, the justices are acting as political scientists, and rather poor ones at that, in leaving undisturbed the status quo of single-member districts.

Far-from-radical, full-representation voting methods have a lengthy history in the United States. In fact, Justice Clarence Thomas discussed them quite cogently in Holder v. Hall (1994), noting that "from the earliest days of the Republic, multimember districts were a common feature of our political systems." Non-winner-take-all voting methods used here (in a growing number of cities) and in some other nations have led to largely two-party systems, yet still resolve nearly all political gerrymandering concerns -- and, importantly, all the conflicts the Court has faced in trying to ensure that racial minorities can elect candidates of their choice.

If non-winner-take-all systems would constitute no "radical" change, there is simply no constitutional reason to cling to single-member districts. Indeed, Illinois shows how alternatives to winner-take-all elections can enhance our political traditions rather than fundamentally alter them.

From 1870 to 1980, the Illinois lower house had three-seat constituencies elected by cumulative voting. Voters had three votes each, which they could give to one candidate or spread among a few. The majority party usually won two seats, often with two candidates reflecting different elements within the party. The third seat usually was won by another party with support from about a quarter of the voters.

After the system was replaced in 1980 (due to a citizen initiative that sharply reduced the number of representatives), the Illinois legislature became much more polarized. Today most longtime leaders in both parties support the return of multi-seat districts, as evidenced by the 2001 recommendation of a bipartisan commission led by former Republican Gov. Jim Edgar and former Democratic Rep. Abner Mikva. The commission argued that multiseat districts would lead to greater cooperation between the parties and fairer representation across the state.

The Illinois system's one downside -- the fact that parties often nominated only two candidates to avoid splitting the vote -- could be addressed by adopting the choice voting method used in Ireland. That system lets voters indicate their first, second, and third choices, so that voters whose first choice doesn't win a seat can still help elect their second or third choice. Also, to ensure greater accuracy of representation statewide, a few "add-on" seats could be awarded to underrepresented parties, as recently proposed in the United Kingdom.

Many students of American democracy and nearly every major newspaper, from The New York Times to The Wall Street Journal, warn that our democracy is in crisis because there is so little competition and accountability in congressional elections. But without challenging the dogma of winner-take-all districts, any reforms will fall short of addressing the real crisis. To confront the political realities of the 21st century and rebuild a vibrant, accountable representative democracy, we must turn to American systems of full representation.

John B. Anderson is a former member of Congress, one-time candidate for president, and currently professor of law at Nova Southeastern University. He is chairman of the Center for Voting and Democracy (www.fairvote.org). Robert Richie is the Center's executive director.