How Proportional Representation Would Finally Solve Our Redistricting and Gerrymandering Problems

 

Douglas J. Amy
Department of Politics
Mount Holyoke College

 

(This piece is adapted from Chapter 2 of Real Choices / New Voices: How Proportional Representation Could Revitalize American Democracy from Columbia University Press.)

 

Most Americans believe that who wins political races is decided on election day by the voters. But in a single-member district electoral system that is frequently not true. Who wins is often determined before voters even go to the polls – sometimes many years before. The outcome is decided by those who draw the district lines. If they decide to create a district that is 70 percent Republican, there is little chance the Democratic candidate will win. And Republican candidates will usually lose if a district is drawn so that it is predominantly Democratic. Voters go to the polls confident in the illusion that they control the fate of the candidates. But in reality they are often only participating in the last act of political play whose ending has already been written.

In a single-member plurality system, the power of the vote pales in comparison to the power to draw district lines. The districting process not only can determine which candidates will win in specific districts, but also can determine which party ultimately controls our local, state, and federal legislatures. In a very real way, then, the political manipulation of district lines devalues the vote and undermines the democratic process.

Describing the redistricting process in unusually candid terms, North Carolina State Senator Mark McDaniel has said: "We are in the business of rigging elections." This process of drawing district lines to rig elections is known as "gerrymandering," and it is one of the great curses of the single-member plurality (SMP) electoral system. The United States could rid itself of this unscrupulous and undemocratic practice; but eliminating gerrymandering in all of its forms can only be accomplished by abandoning the single-member districts that make it possible and by adopting proportional representation elections.

Gerrymandering

After each census, a process of reapportionment and redistricting begins in the states. The federal courts require that every district contain approximately the same number of voters, so shifts in population require that U.S. House seats be reapportioned between the states. States that lose or gain seats must necessarily redraw their congressional districts to take this into account. Even in states that do not gain or lose Congressional seats, shifts of populations within regions of the state can also require the redrawing of districts.

In this redistricting process, gerrymandering – the manipulation of district lines for political advantage – has become the political weapon of choice. The two main purposes of gerrymandering are to protect the seats of incumbents and to allow the dominant party in a state to win more seats that it deserves. We will first consider the latter form of gerrymandering – which is usually called "partisan" gerrymandering. This process takes advantage of one of the main drawbacks in single-member plurality voting – the fact that many votes are usually wasted. Wasted votes are those that do not contribute to the election of a candidate. In partisan gerrymandering, the dominant party in a state draws the district lines in an effort to get the other major party to waste as many of its votes as possible and thus to not be able to elect its fair share of legislative seats.

The two main strategies that are used in partisan gerrymandering are called "cracking" and "packing." In cracking, district lines are redrawn to ensure that the opposing party's pockets of voting strength are divided up so that they become permanent minorities in most districts. For example, Democrats may eliminate a previous district where Republicans had 60 percent of the vote, and divide that party’s supporters into two new districts, where they make up only 30 percent of the voters in each – thus making it impossible for them to win a seat. The packing strategy is used when the opposing party's strength is too large to be completely divided. Instead, lines are redrawn to concentrate most of that party's voters in one district so that many of its votes will be wasted in an overly large majority. (Any votes in excess of 51% are actually wasted, since they are not needed to elect the candidate.) If, for example, the Democrats put most of the Republicans in a region into one district in which they make up 85 percent of the voters, they will win only one seat; whereas if those Republican voters had been spread around several districts, they might have helped their party win more seats.

These two gerrymandering techniques are illustrated in the two figures below, both of which represent the same city in which four members of the city council are to be elected. (In these figures, D = 1,000 Democrats and R = 1,000 Republicans. Each grouping of three letters makes up a neighborhood.  It takes eight neighborhoods to make up a district.) As is often the case, the central area of the city is predominantly Democratic, while the suburbs are predominantly Republican. Given that the Democratic and Republican parties have approximately even numbers of supporters (fifty-two thousand and forty-four thousand), it would seem fair that they each receive two of the four seats to be filled. However, gerrymandering can produce quite different results. In the version of the districting on the left, the Democratic party draws the district lines in a way that fragments the Republican vote. Each district includes part of the Democrat-dominated inner city, and the Republicans become minorities in all of them. As a result, the Democrats take all four seats. All forty-four thousand Republican votes are wasted – producing no representation. In the version of districting on the right, the Republicans draw the lines and create a central city district into which they pack the Democratic vote, where it wins one seat by a very lopsided margin, thus wasting much of its voting strength. This allows the Republicans to win the other three seats.

Democratic Gerrymander

Republican Gerrymander

RRR

DDD

DRD

RRD

RRR

RDD

RDR

DRR

RDR

DRD

DRD

DDD

DRD

DDD

DDD

RRR

DRD

DDD

DDD

DDD

RDR

DRD

RDR

RRR

DRR

DDD

DDD

RRR

DRR

RRR

DDD

RRR

 

RRR

DDD

DRD

RRD

RRR

RDD

RDR

DRR

RDR

DRD

DRD

DDD

DRD

DDD

DDD

RRR

DRD

DDD

DDD

DDD

RDR

DRD

RDR

RRR

DRR

DDD

DDD

RRR

DRR

RRR

DDD

RRR

               Dem.         Rep.                     NW       13,000       11,000                     NE        13,000       11,000                     SW        13,000       11,000                    SE         13,000       11,000
Totals   52,000       44,000

             4 seats       0 seats

                 Dem.          Rep.                         NE          11,000       13,000                         NW         11,000       13,000                       Central   19,000         5,000                       S             11,000       13,000
Totals     52,000       44,000

                1 seat        3 seats

 

As these examples illustrate, partisan gerrymandering can be a very powerful political tool. Which party wins the most seats may not determined by how many votes that party gets, but instead by how the district lines are drawn. A real life example of this can be found in the results of districting in New York State. As shown by presidential votes, about 60 percent of the voters in New Yorker State consistently vote Democratic. This is reflected in the State Assembly where Democrats routinely win about 64 percent of the seats. But the election results are dramatically different in the State Senate, where the Republicans consistently dominate with a large majority of about 57 percent of the seats. Same voters, but very different results. The obvious suspect is the districting process. The districts for the Senate races are different from those for the Assembly races and they are largely drawn by the Republican dominated Senate, which makes every effort to favor GOP candidates. This seems to be a clear example of where districting, not the voters, decides who wins control of legislative bodies.

The intended result of partisan gerrymandering is misrepresentation of the public. The party controlling the redistricting artificially inflates the number of seats it gets. In some cases, this distortion of representation is so gross that it allows a party that gets a minority of the votes to win a majority of the seats – a clear violation of the basic principles of democracy and majority rule. In the 2000 elections for the U. S. House of Representatives, for example, gerrymandering in Texas allowed the Democrats to win 57 percent of those seats, even though they received only 47 percent of the vote statewide.

Examples of partisan gerrymandering abound in the United States. Indeed, some states, such as California, Texas, and Indiana, have become infamous for their blatant gerrymanders. After one reapportionment, Republican legislators in Indiana did everything they could to use district lines to their advantage. Over the protests of the Democratic minority, the Republicans adopted a plan that took two districts with Democratic incumbents and packed them with additional Democratic voters so that GOP candidates in other districts would have less opposition. The plan also relocated three Democratic incumbents into the same district, and separated another Democrat from his hometown, placing him in a traditionally Republican area. In case anyone had doubts about the political intentions underlying this plan, one of its principal authors, Republican state Senator Charles Bosma, told reporters that the Democrats "are going to have to face the political reality that we are going to do everything we can to hurt them within the restraints of the court rulings."

These political shenanigans were much in evidence again in the redistricting efforts following the 2000 census. In Kentucky, Albert Robinson, the chairman of the committee charged with redrawing their Senate’s district map vowed revenge on the Democrats now that the Republicans were in control of that body. "I’ve been a victim of partisanship," he complained, pointing out that the Democrats "tried to destroy the Republican party" in previous redistricting efforts. He cheerfully admitted that the Republicans eagerness to use their advantage "goes without saying. Any party that’s in control, charity begins at home."

The Democrats are hardly any better when they are in control. In California, the Democratic controlled legislature has a history of creating district lines that meander around California, throwing Republicans incumbents together, and carving out new districts that favor Democratic candidates. In one famous instance of excessive creativity, they designed a district for a Democratic incumbent that had 385 sides. In the past, the electoral power grab has been so effective that Democrats have been able to win a majority of the U.S. House seats from that state, even though they sometimes have earned less than a majority of the votes.

Redistricting: Politics at Its Worst

The legislative battles over redistricting can be incredibly passionate. The political combat has been various described in the press as "bare-knuckle brawling," "cut-throat politics," and "political street fights." Speaking of the 2001 redistricting process in Pennsylvania, Professor Terry Madonna said: "I’ve studied reapportionment and redistricting for 30 years, and I’ve never seen an atmosphere so poisonous as it is today in Pennsylvania." This political intensity is caused by several factors. First, the very jobs of the state legislators are at stake. Redrawing boundaries can make them vulnerable to defeat and they naturally will fight desperately to protect their careers. It’s a matter of political survival. Second, redistricting has profound and wide-ranging political impacts on who has political power in the United States. These decisions affect not only who controls state governments, but the federal government as well. Officials in the U.S. House of Representatives are very aware of the implications of redistricting. After the last census, U. S. House Majority Whip Thomas DeLay gave a speech to the Texas state legislature urging them to use redistricting to dramatically increase the number of their Republican House seats. "I hope each of you will also remember that your vote will not simply count for Texas," he said. "The decisions will carry implications for our entire nation." The political effects of redistricting also extend over time. Once districts are drawn, the political landscape is set until the next census and reapportionment. As political scientist Thad Beyle, has observed, "Whoever sets up the legislative districts for the 2002 election basically sets up the elections of the whole decade."

One reason why redistricting political battles have become worse in recent years is the ease with which more groups can now join in the battles. In the past, redistricting was done by a few groups of experts and politicians working largely in secret. They worked over maps by hand and revealed their final plans only after the process was over. These few players were the only ones with the information and expertise to accomplish this difficult task. But with the advent of the personal computer, internet access to demographic data from the census, and new user-friendly redistricting software, nearly any organized group can developed their own redistricting plan. As Bruce Cain, director of the Institute of Governmental Studies as the University of California at Berkeley has observed: "There used to be two or three groups that were able to draw their own (redistricting) lines. Now there could be hundreds."

Redistricting battles have not only become more complicated, but now they usually do not end with the passage of a final plan by the state legislature. Increasingly, redistricting ends up in the courts, as various groups challenge the plans offered by the legislatures. This means that the redistricting mess can drag on for years and states can spend large sums of money in court fights. In Florida, Alabama, Louisiana, and North Carolina, litigation over the districts drawn in response to the 1990 census was still going on in 1999. In Chicago, in the 1990s, the city government was forced to pay over $8,000,000 in legal fees for litigation resulting from its redistricting process.

To sum up, redistricting is a political disaster – an enormous political train wreck that occurs every ten years. It is a process full of partisan rancor, naked power grabs, unfair representation, and millions of dollars in legal costs. It is an example of American politics at its worst – and all made possible by our single-member district electoral system.

Gerrymandering as Incumbent Protection

If there is one bright spot in this redistricting mess, it is that brazen partisan gerrymandering is sometimes difficult to accomplish. It works best in states where a single party controls both houses of the state government and the governorship. When control over state government is split between the major parties, this gives each party a veto over any redistricting scheme and makes it difficult for one party to grossly take advantage of another. Often, when one party can’t dominate the process, the end result is no large change of seats from one party to another. However, each party gets something out of the process: the protection of its incumbents. This is sometimes called a "sweetheart gerrymander" – where the two major parties make sweetheart deals intended to create safe districts for both Democratic and Republican incumbents. Safe districts ensure incumbents reelection by packing their party supporters into those areas. So in these situations, the redrawing of district lines becomes mainly an exercise in incumbent protection, with politicians of all parties scrambling to make sure that their districts are not eliminated or that any new district lines will encompass as many of their supporters as possible. Legislators from the minor party in a state often tend not to object to such sweetheart gerrymanders even if these arrangements result in some underrepresentation of their party. While these legislators do care about how many seats are won by their party, their first priority in redistricting is to preserve their own seats.

Focusing on these bi-partisan arrangements, defenders of single-member plurality elections sometimes argue that the problem of gerrymandering has been exaggerated. They maintain that it is not really a serious political problem because often it does not result in gross distortions in representation. But this sanguine view mistakenly assumes that incumbent protection gerrymanders are politically innocuous and it overlooks the serious political damage produced by this approach to redistricting. The main problem is that these gerrymanders create districts that are uncompetitive – districts in which voters are denied any meaningful choice. When legislators manipulate the district lines to create safe seats for Republican and Democratic incumbents, any real competition in these elections disappears. Commenting on the latest redistricting effort in California that protected incumbents of both parties, one Republican consultant concluded: "This new plan basically does away with the need for elections. These seats shouldn’t change (parties) over the decade." While such arrangements clearly benefit incumbents, they mean that voters are faced with uninspiring, no-contest elections. Instead of a healthy competition between two parties, we end up with a series of one-party fiefdoms in which a single party rules and voters feel helpless to change it. The New York Times once editorialized that as a result of gerrymandering, New Yorkers had "no more voting options than North Koreans have."

The tendency of legislatures to create safe districts is well known and well documented. After every election, the Center for Voting and Democracy analyzes the competitiveness of races for the U.S. House, and they inevitably find that an enormous number of the seats are no-contest elections. In 2000, for example, 78 percent of the U.S. House seats were won by landslides – defined as victory by a margin of greater than 20 percent. Only 9 percent of House races were defined as competitive – having been won by a margin of less than 10 percent. Naturally, most of the landslide wins took place in districts where one party had a lopsided advantage in voters. The study identified "two major culprits for our non-competitive elections: a winner-take-all electoral system that turns natural majorities into no-choice elections and typically partisan methods of redistricting that allow legislators to craft districts "safe" from competition – legislators literally choose their constituents before their constituents choose them."

Uncontested Contests

In safe districts, candidates of the minority party are often little more than sacrificial lambs being led to the slaughter. But it gets worse. These safe districts are often so one-sided that the disadvantaged party does not even bother to put up a candidate. There is no contest at all. The number of uncontested U.S. House seats varies considerably from state to state. Florida and Texas are two of the worst. In the 2000 U.S. House elections in Florida, 43 percent of the races were uncontested. That year in Texas, 30 percent of their House races were uncontested. The existence of some many uncontested races is a political disgrace, but it is also a perfectly understandable result in a winner-take-all system. Why should the smaller party bother to compete in districts dominated by the larger party? It would be a waste of valuable political resources. As the former head of the Republican campaign effort in the House, Rep. John Linder of Georgia, explained: "The GOP did not recruit candidates in districts where Democrats are shoo-ins for re-election." And according to his counter-part on the Democratic side, Rep. Martin Frost of Texas: "Both parties, I believe, had made a decision to concentrate on really competitive districts." So in an SMP system, it is a rational political strategy for a party to no contest seats in many districts.

On the level of state legislative contests, the problem of unopposed candidates has reached epidemic levels. In races for state legislatures in 2000, either the Democrats or the Republicans refused to nominate a candidate in 40.6 percent of the districts. In 1998 that figure as 41.1 percent. Thus in a large number of cases, the creation of safe seats for incumbents means that voters literally have no choice on election day. These are not so much races as prolonged victory laps by the preordained winners.

These safe districts created by gerrymandering not only deny voters a meaningful choice, they contribute to other political problems as well. For one thing, safe districts clearly encourage political apathy and low voter participation. Why bother to participate if the outcome has been predetermined by how the district lines have been drawn. As one San Francisco resident complained: "Vote? Why vote? I know who’s going to win, everybody knows who’s going to win. Pelosi always wins, with 80 percent of the vote. Nobody else has a chance." Those who support the party that always loses are not the only ones discouraged from voting in safe districts. Even those who support the dominant party have little reason to vote because they know that their candidate will win anyway.

Lack of Responsiveness

Another major political problem of uncompetitive districts is that they make the political system much less responsive to the public. One expert on redistricting, David Wells, explains this problem in the following way:

Gerrymandering invariably inflates the number of safe districts. Barring a successful primary challenge, the individual incumbent is virtually assured of continued reelection for as long as he or she cares to hold the seat. This has the effect of insulating the legislative body against the consequences of changing sentiments and circumstances, for gerrymandering has provided the individual legislator, the legislative leadership, and the legislature as a whole with rather strong guarantees of continued office and power. The political ... composition of the legislature has been effectively frozen for a decade, and changes are possible only within a limited, narrow range. The representation system, because it has been made less politically sensitive and therefore less responsive, has thus been rendered less able to perform its most fundamental task – the translation of public sentiment into public policy as accurately as possible.

A hypothetical example may help to illustrate this point. Assume that gerrymandering creates a safe district packed with Democratic voters and the Democratic candidate is elected with 70 percent of the vote to 30 percent for the Republican. Before the next election, however, voter sentiment turns strongly toward the Republican party – with 16 percent of the former Democrats now favoring the GOP. Despite the dramatic change in political opinion, the Democratic candidate will likely continue to represent this district – and be reelected this time by 54 percent to 46 percent. So despite a large change in political preferences by the public, no resulting change of representation takes place in this safe district. The protection of legislators from both parties in this way severely restricts the responsiveness of the political system.

Contrast this result with the effect of changing voter sentiments in the multimember districts used in proportional representation. Assuming that ten representatives are elected in a district, a swing of 16 percent from one party to another would result in change of at least one and possibly two seats in favor of the party that is growing in popularity. In this sense, PR elections are much more sensitive to changes in voter preferences, and these changes are much more likely to be reflected in the makeup of legislatures. In PR systems, the total number of a party's seats in a district cannot be made safe. Parties always can lose seats in proportion to their drop in popularity – which may not be true in single-member plurality systems.

To sum up, there are two different kinds of political evils created by gerrymandering. In its most egregious form, gerrymandering is used by parties to steal seats and political power away from one another. Thus it violates basic notions of fair representation. But it also acts as an incumbent protect process, where instead of voters choosing their politicians, politicians actually choose their voters and thus ensure their reelection. Clearly redistricting and gerrymandering are serious threats to fair and competitive elections in the United States. But how can this problem be solved? We turn to that question next.

The Difficulty of Reform within the SMP System

Given the pernicious and clearly undemocratic results of gerrymandering, its elimination is obviously high on the list of those who believe in fair elections. But reform in this area has been slow in coming, and the problem continues to plague our political system. Some reformers have tried to challenge this practice in court, but they have been consistently unsuccessful. While the Supreme Court has said that redistricting plans can be challenged in federal court, they have consistently refused to overturn them. Strangely, while the Supreme Court has been eager to label as unconstitutional the practice of "racial gerrymandering" – the manipulation of district lines to increase the representation of minorities – it has been reluctant to criticize partisan gerrymandering even in its most blatant and undemocratic forms.

Reformers have had some success in transferring the redistricting process out of the hands of state legislatures and into bi-partisan commissions. Hawaii, Idaho, New Jersey, and Washington have given the task of redistricting to commissions comprised of equal numbers of Democrats and Republicans, with an additional member approved by both parties. But this approach has proved far from ideal. Members of these commissions have often proved do be as rabid partisans as the legislators themselves. And some critics have pointed out that, unlike legislators, these political mapmakers are not directly answerable to the public if they draw questionable district lines. Tim Storey, a redistricting expert for the National Conference of State Legislatures, has observed that the redistricting commissions "concentrate a fair amount of power in small unaccountable groups, rather then in the legislatures, where members have to stand for elections." But the most serious problem with bi-partisan commissions is the possibility of sweetheart gerrymanders. With equal numbers of Democrats and Republicans, there is a good possibility of deals being struck that serve to protect the incumbents of both parties. The result would inevitably be the kinds of uncompetitive districts that were discussed earlier.

The most radical approach to the problem of redistricting has been taken by Iowa, which has empowered civil servants to draw the districts maps, allowing them to take into account population figures only. Some reformers, including Common Cause, have backed similar apolitical approaches that would require the drawing of districts using only political neutral criteria, such as avoiding splitting up counties and cities, ensuring compact districts with the shortest boundary lines possible, and minimizing population differences between districts. (Some have even suggested leaving redistricting to computers programmed to use only nonpolitical criteria.) But while this approach would eliminate the most obvious examples of packing and cracking and thus would be a step forward, a number of problems remain. First, most legislatures are unlikely to agree to give up their redistricting power entirely to such bodies. Second, and more importantly, while this approach would clearly be an improvement, it would not succeed in abolishing gerrymandering altogether. By eliminating political discretion from the redistricting process, this apolitical approach would eliminate the most severe cases of intentional gerrymandering, but it would not eliminate the problem of unintentional gerrymandering.

Unintentional gerrymandering is less well known than the deliberate variety, but it can also pose very serious political problems. Even districting that is done with apolitical criteria will likely produce many of the symptoms of gerrymandering – such as the creation of many uncompetitive districts where one party dominates. The reason for this is simple. Intentional gerrymandering is made possible by the geographic concentrations of Republicans and Democrats – concentrations that can be split and combined to provide advantages for a particular candidate or party. Apolitical districting does not change that geographical fact. Thus many districts created by neutral criteria will inevitably have concentrations of partisan voters that work in one party's favor. Take, for example, the situation illustrated earlier in figure 2.2. In an urban area reasonable and apolitical district lines could be drawn that respect the integrity of the boundaries between cities and suburbs. But in creating a central city district, Democratic voters are unintentionally packed into that district, forcing them to waste much of their voting power with excessively large majorities. The unintentional result is the creation of districts that are uncompetitive and that unfairly distort the representation of the parties – the hallmarks of gerrymandering. As such examples illustrate, just because apolitical districting ignores geographic partisan concentrations does not mean that these concentrations no longer have any political effects. The concentrations remain and create unintentional gerrymanders.

In short, even when mapmakers use apolitical criteria to guide their work, they are bound to produce political effects. This is exactly what has happened in Britain, where independent redistricting commissions have sometimes been offered as models for what we need here in the United States. A number of studies have shown that although these commissions have used neutral criteria to draw district boundary lines, they often created unintentional gerrymanders that biased the results of national elections. One of Britain's leading political geographers, R. J. Johnston, found that the "review of recent redistricting in Britain, undertaken by independent commissioners within the constraints of apparently neutral rules, cast doubt as to the likelihood that fair redistricting would be achieved in the U.S.A. or indeed anywhere that the plurality single-member constituency electoral system is used." He concluded that "the concept of non-partisan cartography is a myth."

This is not to say that apolitical districting with its unintentional gerrymanders may not be better than the flagrant intentional gerrymandering that occurs now. But it does mean that we should not delude ourselves into thinking that this approach will eliminate the problem of gerrymandering. Indeed it cannot, for it does not address the underlying cause of the problem: the single-member plurality system itself. Gerrymandering is a necessary, built-in characteristic of any plurality voting system that uses single-member districts. As long as we draw district lines where only one candidate wins and where large numbers of wasted are unavoidable, some gerrymandering is inevitable, if only unintentionally. As Robert Dixon, a longtime student of redistricting in the United States, concluded: "To be brutally frank, whether or not there is a gerrymander in design, there normally will be some gerrymandering in result as a concomitant of all district systems of legislative elections…In a functional sense it thus may be said that districting is gerrymandering."

PR: The Real Solution

The only sure way to eliminate gerrymandering – both intentional and unintentional – from American elections is to abandon single-member plurality arrangements and adopt proportional representation. Indeed, the whole purpose of PR is to minimize wasted votes and ensure that the parties are represented in proportion to the votes they receive. This eliminates the possibilities of unfair representation produced by gerrymandering. The key to eliminating partisan gerrymandering is the large multimember districts used in PR systems. As numerous studies have shown, as long as a PR system has at least five seats in every district, it is effectively immune from gerrymandering. These districts largely eliminate the wasted votes that make gerrymandering possible. In such districts, even small political minorities do not waste their votes and are able to elect their fair share of representatives. Thus, under PR arrangements, where voters live or how district lines are drawn makes no difference – fair representation will result.

Imagine, for example, that we have a region in a state that is 60 percent Republican and 40 percent Democratic and that it must be divided into two ten-member PR election districts. No matter how the district lines are drawn and no matter how party voters are distributed between the districts, each party will be able to elect its fair share of representatives. If all the Democrats are packed into one district, they will constitute 80 percent of the voters there and elect eight of the ten representatives in that district and none in the other – 40 percent of the total seats. If the Democratic voters are fragmented and make up 40 percent minorities in the each of the two districts, they will be able to elect four representatives in each – and still receive 40 percent of the total seats.

That gerrymandering and unfair representation can be truly eliminated only under PR arrangements is so indisputable that even political scientists who continue to support our single-member plurality system are forced to acknowledge it. Nelson Polsby of the University of California at Berkeley is no supporter of proportional representation. Yet he testified in a court case on gerrymandering that "proportionality cannot be guaranteed in a system of voting in which the winner of the most votes wins the election regardless of how many candidates run in each race. To ensure proportionality it is necessary to have a proportional representation system of elections.''

Proportional representation would also eliminate the uncompetitive districts created to protect incumbents. Under PR rules, it is impossible for one party to have a monopoly on the seats in a district and to prevent the others from winning any representation. Instead of a winner-take-all system we would have an all-are-winners system, where all parties have a reasonable chance to win some district seats. Also, as noted earlier, PR districts are much more responsive to changes in voter sentiment than one-party dominated districts. Even relatively small changes in partisan preferences are translated into changes in seats in the large multimember districts used by PR.

Thus the adoption of proportional representation in the United States would in one stroke cut through the Gordian knot of redistricting and gerrymandering. It would eliminate all the partisan brawling, endless and expensive legal challenges, stolen seats, and uncompetitive districts that plague our single-member plurality system. Rarely do we find such a clear and simple solution to such a difficult set of political problems, but in the case of gerrymandering, PR is exactly that. Indeed, this ability to effectively eliminate both intentional and unintentional gerrymandering has been one of PR's main attractions in countries that have adopted it. In these countries, redistricting is a trivial and boring political activity because it has no real political impact. We could certainly use some of that boredom here in the United States.