Chapman Law Review
THE RED AND BLUE GOLDEN STATE: WHY CALIFORNIA’S PROPOSITION 11 WILL NOT PRODUCE MORE COMPETITIVE ELECTIONS
Copyright (c) 2011 Chapman Law Review; Anthony E. Chavez
In November 2008, Californians approved a redistricting reform measure, Proposition 11, which, despite the promises of its supporters, will have little effect on the competitiveness of the state’s elections. The initiative shifted responsibility for the redrawing of state legislative lines from the Legislature to an appointed commission. Supporters promised that by taking the process from self-interested legislators, the resulting districts would be more competitive. However, an analysis of the state’s demographics and the experience of other states suggests that a significant increase in the competitiveness of California’s legislative districts remains unlikely and may even be undesirable. Indeed, because of the give-and-take nature of the redistricting process, California would be served best by keeping the initial responsibility for redrawing legislative lines with the Legislature. A redistricting commission would be most helpful if it focused on reviewing and revising plans developed by the Legislature, rather than actually drawing the initial plan itself.
This article first reviews the legal standards applicable to redistricting in California. It then discusses the most recent redistrictings and the five failed attempts that have been made to alter the redistricting process through ballot initiatives. Section II addresses Proposition 11, the campaign, and the 2008 general election. Section III explores the benefits and detriments resulting from more competitive district elections, the much-touted benefit of Proposition 11. Section IV analyzes the decline in competitive elections in California and the non-redistricting causes of this decline. Section V looks to the experiences of other states that have employed commissions and imposed competitiveness standards on the drawing of their legislative districts. Finally, the last two sections discuss the likelihood that Proposition 11’s redistricting commission will be able to draw more competitive districts and present an alternative approach that would provide redistricting roles for both an independent commission and the Legislature.
I. Redistricting, “California” Style
Proposition 11 is the first successful salvo in the ongoing redistricting battle in California. To best appreciate the measure’s significance, a review of the state’s redistricting history is helpful. But, first, this article will look at the legal constraints impacting redistricting prior to the passage of the measure. Then, it will review the decline in competitiveness in California’s district elections which has prompted most of the concern over redistricting reform. Moreover, the article will examine the battles over the state’s past redistrictings, both in the courts and on the ballot.
A. Legal Constraints on Redistricting in California
Redistricting is the process of revising the geographic boundaries of congressional or state legislative districts to account for population shifts between decennial censuses. The U.S. Constitution requires the federal government to conduct a census every ten years for the purpose of apportioning congressional seats among the states. The states typically redistrict their own state legislatures in conjunction with this process. The California State Constitution, for instance, requires that the state redistrict its election lines once every decade in the year following the national census. Historically, the California Assembly has shouldered responsibility for redrawing state legislative and congressional district lines.
Certain federal and state standards apply to the drawing of election districts. First, the Supreme Court has determined that congressional districts must have approximately equal population. In Reynolds v. Sims, the Court held that political equality under the Constitution “can mean only one thing–one person, one vote.” This means that district populations should be as nearly equal as possible. Courts have applied this standard rigorously in congressional redistrictings. In the context of state legislative plans, the Supreme Court has allowed state legislatures greater latitude. Indeed, the Court has indicated that deviations of at least sixteen percent are acceptable.
The other federal law that controls redistricting is the Voting Rights Act of 1965 (Act), as amended and codified in Title 42 of the United States Code. Depending upon the state involved, two provisions of the Act may have implications for redistricting. Section 2 of the Act applies to all states. Where certain preconditions exist, section 2 prohibits dilution of minority voting strength. Line drawers typically use one or more of the following techniques to dilute minority voting strength through redistricting: “packing,” “cracking,” or “stacking.” “Packing” involves concentrating as many minorities as possible into as few districts as possible, thereby creating larger minority populations than necessary to elect their candidates of choice and minimizing the impact of minority votes. “Cracking” consists of splitting concentrations of a minority population and dispersing them among other districts to increase the number of districts containing white-voting majorities. Finally, “stacking” refers to combining concentrations of a minority population with larger concentrations of a white population to ensure that the districts contain white voting majorities.
Section 5 of the Voting Rights Act requires that changes involving voting (such as the passage of a redistricting plan) must be approved (“precleared”) by either the U.S. Attorney General or the U.S. District Court for the District of Columbia. Section 5 prohibits “retrogression in the position of racial minorities with respect to their effective exercise of the electoral franchise.” Section 5 extends only to nine states in their entirety, and to portions of seven others. Four counties in California fall under section 5. Section 5 applies when a covered jurisdiction adopts a redistricting plan. Because four counties in California are covered, the effects of any statewide redistricting plan (congressional or legislative) on those four counties fall within the preclearance requirement of section 5.
California law also imposes certain requirements upon the redistricting process. The California Constitution sets forth several basic constraints. It reiterates the federal requirement that the districts shall have reasonably equal population and also requires that they be contiguous. In addition, the California Constitution provides that the geographical integrity of cities and counties be respected to the extent possible, without violating other requirements.
In addition, California recently passed its own state voting rights act. While the California act is similar to the federal statute, the state law explicitly removes geographic concen-tration of the minority group as a requirement for finding a violation. Of course, some level of geographic compactness is necessary to establish an interest in redistricting. For these reasons, California’s voting rights act does not appear to create any new constraints on its redistricting process.
B. Recent Trends in California District Elections
Despite the tradition in California (and elsewhere) of legislative control of the redistricting process, the placement of this responsibility in the hands of the Legislature has long been a target of criticism. In general, critics have charged that legislators are inherently self-interested in the outcome of redistricting. Legislators have a number of incentives to exploit the process for political gain, including protection of individual incumbents, expanding partisan statewide majorities, and punishing those with differences from the line drawers, regardless of their party affiliation. Critics complain that “politicians get to choose their voters, rather than the reverse . . . .” Or, as North Carolina State Senator Mark McDaniel rather candidly admitted about the redistricting process, “We are in the business of rigging elections.”
While these arguments have tremendous emotional appeal, of greater interest is whether proof of such practices can be found in California. Critics of legislative redistricting identify several indicia of these practices. First, one trend pointed out by proponents of Proposition 11 during the 2008 campaign was the decline in changes of party control of California’s Assembly districts. Figure 1 tracks the number of seats in the Assembly that have changed parties between 1960 and 2010:
The first striking fact about this chart is the two spikes in the trendline. These spikes represent a high exchange of seats between the parties in 1974 and 1992. Both of these occurred in the first elections after the implementation of new redistricting plans. Of course, these are not the only elections held after the passage of new plans. California conducted elections pursuant to new plans in 1962, 1982, and 2002. Why were 1974 and 1992 the only years to have such extraordinarily high changes of seats between the major parties? Possibly because the authors of those plans were judges and not the Legislature.
The other trend that Figure 1 highlights is the overall decline of party turnover during the period analyzed. In the 1960s, the yearly exchange of seats averaged eight per election. By the 2000s, the average turnover was down to less than two per election. As was frequently noted during the Proposition 11 campaign, in the two elections prior to the 2008 election, no Assembly seats changed party hands.
Another possible indicator of legislative misuse of the redistricting process is the decline in competitive elections. Scholars have noted the decline of marginal districts over the past thirty years. California’s elections have followed this pattern; the number of competitive assembly seats has steadily decreased. For instance, Passantino tracked the number of assembly seats where the winner won no more than fifty-three percent of the vote. Figure 2 presents the results:
In many respects, Figure 2 presents a picture quite similar to that of Figure 1. In both Figures 1 and 2, the highest peak occurs in 1974, which is immediately subsequent to the adoption of that year’s plans. However, the second peak in Figure 1 occurs in 1992, but in Figure 2 it does not arise until 1996. In other words, in the 1990s, the highest number of districts changed party hands in the year immediately after the redistricting, while the number of competitive contests continued to increase. This discrepancy may suggest that factors other than redistricting affect competitiveness. Although Figure 2 also suggests that the number of competitive districts had been in decline since the 1992 redistricting, as with Figure 1, it illustrates a significant, though short-lived, rebound in the 2008 general election.
This decline in competitive districts during the past two decades coincided, not surprisingly, with the rise in sophisticated tools for the line drawers. Indeed, the most significant change occurred between the 1981 and 1991 redistricting cycles. For instance, the architect of California’s 1981 redistricting, Congressman Philip Burton, “used teams of individuals to analyze massive hard-copy reports of voter registration data, election results, census data, and precinct maps using simple calculators and colored markers.” One decade later, the available technology had changed dramatically. The Supreme Court described the capabilities of the software REDAPPL, which the Texas redistricters used in 1991:
REDAPPL permitted redistricters to manipulate district lines on computer maps, on which racial and other socioeconomic data were superimposed. At each change in configuration of the district lines being drafted, REDAPPL displayed updated racial composition statistics for the district as drawn. REDAPPL contained racial data at the block-by-block level, whereas other data, such as party registration and past voting statistics, were only available at the level of voter tabulation districts (which approximate election precincts). The availability and use of block-by-block racial data was unprecedented; before the 1990 census, data were not broken down beyond the census tract level.
Thus, redistricters who sought to minimize competitiveness suddenly found themselves armed with new, high-tech tools with which to accomplish this objective.
C. California’s Redistrictings and Redistricting Ballot Initiatives
In the past forty years, redistricting has been especially contentious in California. Both the process and the resulting plans have been the targets of litigation and of ballot measures. These experiences helped to shape Proposition 11 and its successful campaign.
Three of California’s past four redistrictings were the subjects of litigation. In 1971, the Democratic-controlled Legislature and Republican Governor Ronald Reagan failed to agree to a redistricting plan. Accordingly, the California Supreme Court adopted temporary redistricting plans for the 1972 election. When the Legislature did not enact redistricting plans in 1972, the California Supreme Court appointed special masters to develop the plans, which it eventually adopted.
In 1981, the Democrat-controlled Legislature passed redistricting plans that Democratic Governor Jerry Brown signed. Republicans, outraged over what they thought was blatant partisan gerrymandering, commenced two separate attacks to overturn these plans. First, they placed three referenda on the June 1982 ballot; each proposition sought to replace one of the redistricting plans (assembly, senate, and congressional). Republicans hoped that new plans could take effect immediately, but the California Supreme Court ordered the state to use the 1981 plans for the 1982 congressional and legislative elections. The three plans used for the 1982 elections each lost the referenda vote by an average margin of sixty-three percent to thirty-seven percent. Moreover, in November 1982, a Republican was elected to replace Governor Brown. Accordingly, the lame duck Democratic governor called a special session of the Legislature to commence in December 1982, one month before Governor Brown was to step down, to develop new plans. The Democrats drew plans that offered sufficient protections to Republicans and garnered the necessary two-thirds vote to receive “urgency” status, which caused the plans to be sent to Governor Brown before the end of his term. The new plans remained in effect through 1990. As a second means to overturn the Democrat’s plans, the Republicans placed onto the November 1982 ballot Proposition 14.
In 1991, Governor Pete Wilson vetoed the redistricting plans approved by the Legislature. Since the Legislature did not have sufficient votes to override the veto, the governor initiated mandate proceedings in the California Supreme Court. The court exercised its original jurisdiction and appointed three special masters, whom they instructed to develop the redistricting plans after conducting public hearings. The California court accepted and adopted the Special Masters’ recommendations with minor modifications.
Finally, in 2001, the Golden State avoided major litigation over its redistricting plan when Democratic and Republican leaders found common ground: preservation of incumbents. As the Los Angeles Times characterized the plan: “Most legislative districts are so safe that the real battles are in the primary elections.” The Legislature drew both congressional and legislative lines “in a transparent effort to create ‘safe seats’ for virtually all state and federal legislators . . . .” Accordingly, the Democratic-controlled Legislature passed redistricting plans– with no significant Republican opposition–that Democratic Governor Gray Davis signed. Critics have described those plans as “bipartisan gerrymanders” and “incumbent protection gerrymanders.”
Just as California’s redistricting plans have been hotly contested, so has its redistricting process; redistricting has been the subject of five California propositions since the 1980s. The first four failed by significant margins. The fifth finally succeeded, but by less than two percentage points.
Proposition 14 appeared on the November 1982 ballot. It sought to create a redistricting commission whose members would be selected by judges, the major parties, and by any other party representing at least ten percent of the Legislature. California Republicans were the primary financial supporters of the proposition, and several interest groups also backed it. The Democratic Party and its leaders, especially Governor Jerry Brown and Assembly Speaker Willie Brown, were its primary opponents. The proposition lost by a vote of 44.5% to 54.5%.
After the defeat of Proposition 14 and the subsequent blocking of the Sebastiani Plan, the Republicans, under the stewardship of Governor Deukmejian, developed a new proposal which became Proposition 39 on the November 1984 ballot. This measure proposed to establish a redistricting commission with eight of its ten members consisting of retired state court judges. Both parties spent approximately $4 million on the campaign. Nevertheless, Proposition 39 lost by a similar margin to that of Proposition 14, 44.8% in favor versus 55.2% against.
Having failed in the court of public opinion, the Republicans shifted their efforts to actual courts. However, they had no greater success. Thus, in anticipation of the post-1990 Census redistricting, Republicans placed two propositions onto the June 1990 ballot. The first, Proposition 118, would have retained initial authority over redistricting in the Legislature, but for such plans to become law it would have required that the redistricting plans receive two-thirds of the votes in each chamber, the signature of the governor, and approval by the voters in a referendum. Proposition 119, submitted by a second group of Republicans, would have replaced the Legislature with a bipartisan commission whose members were nominated by non-profit, non-partisan organizations and selected by a panel of retired judges. Supporters again contributed millions to the campaigns. Proposition 118 failed by approximately 33% to 67%, while Proposition 119 lost by a vote of approximately 36% to 64%.
Finally, in 2005, after becoming frustrated by working with the Democratic-controlled state Legislature, Republican Governor Arnold Schwarzenegger made redistricting reform one of his primary goals. He ordered a special election in 2005. Included on the ballot was Proposition 77, which would have created a three-member commission to conduct the redistricting. Proposition 77, along with the eight other propositions on the ballot, lost; Proposition 77 failing by a vote of 40.5% in favor to 59.5% against.
Thus, the five propositions after Baker v. Carr that were submitted to California voters lost. Table 2 summarizes these results:
TABLE 2 (See PDF)
Support for these propositions never reached 45%. Consequently, each proposition lost by a double-digit margin, with an average differential of 20.2%.
With this as the background leading up to the 2008 election, Proposition 11 qualified for the November 2008 ballot.
II. Proposition 11: If at First You Don’t Succeed . . .
In many ways, the passage of Proposition 11 in 2008 was aberrational. Most things about the contest were unusual, including its inception, fundraising, and support. Nevertheless, despite the benefit of all of these factors, it still nearly did not pass.
During the post mortem of the 2005 special election, one critique found that Governor Schwarzenegger “took on too much. He took on everybody in sight.” Accordingly, in 2008 he narrowed his focus to one target: redistricting.
Perhaps more importantly, as in any good sequel, several new actors joined Schwarzenegger. One factor that set Proposition 11 apart from its predecessors was the breadth of its conception and subsequent support. The proposal arose not from one of the political parties or the governor, but from the efforts of “good government” non-profit organizations–the drafters of the measure included the California branches of AARP, Common Cause, and the League of Women Voters. The authors of the ballot arguments were the presidents of the California Taxpayers Association and the California offices of the League of Women Voters and AARP. In addition to these organizations, proponents of the measure included the Los Angeles Chamber of Commerce, the California Chamber of Commerce, the California NAACP, the California Police Chiefs Association, and the ACLU of Southern California. The proposition also received endorsements from a broad range of editorial boards.
Politically, the supporters of Proposition 11 were similarly diverse. Despite its non-profit roots, the measure became identified with Republican Governor Schwarzenegger. This occurred for good reason since he was an active campaigner and fundraiser for the proposal. Despite its close connection to the Republican governor, the proposition nevertheless received significant support from high-profile Democrats. For instance, Gray Davis, the chief executive who Schwarzenegger replaced through the 2003 recall election, supported Proposition 11. Other prominent Democratic supporters included Treasurer Bill Lockyer, former Assembly Speaker Robert Hertzberg, and former Controller Steve Westly. California Forward, a recently-created reform group, also supported the measure. The organization’s co-chair was Leon Panetta, former eight-term Democratic congressman and chief of staff for President Clinton.
While Proposition 11’s support was broad-based, its funding was anything but. Of the $14 million contributed to the campaign, traditional Democratic supporters gave less than $1 million. Governor Schwarzenegger, in addition to campaigning for the measure, also supported it financially. His campaign contributions approached $3 million. Not only did contributions skew Republican, significant amounts came from Republicans outside of California. Non-California Republican contributors included New York City Mayor Michael Bloomberg, T. Boone Pickens, and a group of Florida Republicans who donated large sums of money after a personal visit from Schwarzenegger.
Not only did Democrats not contribute to the campaign for Proposition 11, they did not contribute much to the opposition campaign either. In contrast to the $14 million contributed in support of Proposition 11, the “No on 11” campaign received only $1 million.
Also, in contrast to previous redistricting measures, Proposition 11 took a different approach to reform. Earlier propositions provided significant roles for either the major parties or for retired state court judges. For instance, Proposition 14 (1982) and Proposition 39 (1984) would have allowed the major parties to nominate the redistricting commission members. Two of the propositions, 39 (1984) and 77 (2002), would have required that retired judges serve as commission members. Finally, two measures would have had judges either nominate (Proposition 14) or appoint (Proposition 119) the commissioners.
Proposition 11, on the other hand, minimized the role of the parties and eliminated any role for retired judges. Instead, it proposed to establish a “Citizens Redistricting Committee” (CRC). The CRC would draw new district lines for the State Senate, Assembly, and Board of Equalization; under the proposition, the Legislature retained the authority to redraw congressional districts.
The CRC would consist of fourteen members: five Democrats, five Republicans, and four persons not registered with either party. Unlike prior proposals, neither the parties, retired judges, nor organizations would nominate prospective members; instead, they would submit applications. The State Auditor would establish a panel of three State Auditors to screen the applicants. This panel would strike applications of persons who did not meet a series of requirements identified in the proposition. Then, this pool of qualified applicants would be reduced as follows:
Finally, these eight members would then select two additional members from each of the three sub-pools. Figure 3, prepared by the California Legislative Analyst’s Office, presents this selection process graphically:
Proponents and opponents alike recognized the unusual nature of these procedures. Bob Stern, the president of the Center for Governmental Studies, helped draft Proposition 11. Stern acknowledged that the measure is “complicated . . . . It does take a lot of understanding to vote yes on this.” George Skelton, a political commentator with the Los Angeles Times and Proposition 11 advocate, described its process as “convoluted.” Other words used to characterize these procedures included “complex,” “confusing and unfair,” and “byzantine.” Probably the most colorful description, however, was the following: “The mechanisms for selecting the panel seem about as convoluted as the weaning out process of a reality TV series.”
Proposition 11 provides criteria that the CRC must follow in drawing new districts. First, it must comply with the federal requirements of equal population and the Voting Rights Act. Second, districts shall be contiguous. Third, districts must respect the geographic integrity of sub-jurisdictions to the extent possible. Fourth, to the extent possible, districts should be geographically compact. Finally, Assembly districts should be nested within Senate districts–two Assembly districts wholly within each Senate district.
Despite its vast fundraising and endorsement advantages, Proposition 11 had a difficult time attracting the attention of voters. As Figure 4 indicates, in every poll taken in the five months preceding the election, support for the measure never exceeded forty-five percent:
For most propositions, political consultants expect support to fall during the campaign. Thus, experts anticipated a difficult road ahead for Proposition 11. Furthermore, at least twenty-five percent of voters remained undecided about the proposition, though this number rose to thirty-five percent on the eve of the election.
Thus, prior to the election, the factors relating to Proposition 11 were mixed. It had overwhelming advantages in fundraising and endorsements. On the other hand, four previous redistricting initiatives in California had lost by an average margin of twenty percentage points. Furthermore, voter support for Proposition 11 appeared at best to be tepid.
Although it passed, this result may not reflect enthusiastic support, but instead that its supporters were less likely to ignore the proposition than were its opponents. Table 3 provides a recap of the initiatives that were on the November 4, 2008 ballot:
COMPARING SUPPORT FOR PROPOSITIONS
Table 3 presents the twelve propositions on the November 2008 ballot and the total votes in the contest. The last column represents the ratio of votes cast for each proposition compared to the total votes cast for Proposition 11. It shows that fewer voters cast a vote–either “Yes” or “No”–for Proposition 11 than for any other proposition. Between 2.5% more votes (290,000) and 11.8% more votes (1.4 million) were cast for propositions other than for Proposition 11. One consideration is voter roll-off, which tends to be greater the lower an issue physically appears on the ballot. Since Proposition 11 was the second to last measure in the election, roll-off could explain this disparity. Studies, however, have found that voter roll-off in lower visibility contests can be double that of higher visibility contests. The reverse seems to have occurred in California in 2008, where a higher visibility contest, Proposition 11, had a higher roll-off than did lower visibility measures.
Despite the effort expended in support of the measure, both the opposition and the electorate in general displayed ambivalence toward Proposition 11. Certainly, most Democratic leaders opposed the motion. The “No on 11” campaign listed both Speaker Pelosi and Senator Boxer as members. Furthermore, with significant majorities in both state houses, Democrats seemed to be the party with the most to lose. Nevertheless, the measure’s drafters specifically excluded congressional re-districting from its reach to avoid a major fundraising effort by Pelosi. This strategy worked. In 2005, Pelosi spearheaded the fundraising effort, but she also received significant assistance from California Senate President Pro Tem Don Perata and then-Assembly Speaker Fabian Nunez. Furthermore, two powerful unions in California, the California Teachers Association and the Service Employees International Union, staunchly opposed the governor’s special election slate in 2005. In 2008, both unions were neutral on Proposition 11. As a result of these differences, in 2005 the opposition to Proposition 77 spent $13 million (which does not include $4 million in unexpended funds); in 2008, the opposition to Proposition 11 spent only $1 million.
Apparently, the Democrats’ reserved approach to the measure stemmed from more than just Proposition 11’s failure to apply to congressional redistricting. Possibly because of the weak performance of previous redistricting initiatives, the Democrats did not expect Proposition 11 to succeed.
Voters also lacked enthusiasm for the redistricting measure. As Figure 4 demonstrates, throughout the campaign, polling indicated that at least twenty-five percent of voters were undecided regarding Proposition 11, peaking at thirty-five percent on the eve of the election. For their part, prospective voters indicated that the measure was low on their list of priorities for the election. Besides the presidential election, among the eleven other initiatives on the ballot was Proposition 8, regarding same-sex marriage. In contrast to the combined $15 million raised in support and opposition of Proposition 11, Proposition 8 raised a combined total of $83.2 million from both sides. The high roll-off for Proposition 11 likely resulted from the mixture of many factors: confusion, uncertainty, placement on the ballot, ballot fatigue, and attention focused on other choices on the ballot.
Was the passage of Proposition 11 foreseeable despite the abysmal track record of California redistricting initiatives? Under the circumstances, yes. After reviewing redistricting initiatives nationwide, Nicholas Stephanopoulos developed a set of factors leading to the passage of such measures. He analyzed the campaigns of every redistricting initiative since 1936, including the California predecessors of Proposition 11, and initiatives from Arkansas (1936), Oklahoma (1960 and 1962), North Dakota (1973), Colorado (1974), Ohio (1981 and 2005), and Arizona (2000). Stephanopoulos concluded that the most important variable in determining a proposition’s success was the legislative-majority party’s opposition to the measure. He identified several characteristics of successful opposition campaigns run by the majority party including “raising large sums of money, campaigning furiously against the measure, and striving to frame the debate in the most advantageous possible terms.”
A comparison between the campaigns of the two most recent California redistricting initiatives supports Stephanopoulos’ conclusion. In 2005, the opposition to Proposition 77 raised $14 million; in 2008, the “No on 11” campaign raised only $1 million. In 2005, Speaker Pelosi pledged: “I am very committed to defeating Proposition 77, and I am raising money to defeat it.” In 2008, she basically sat on the sidelines since the measure did not involve congressional districts. In 2005, the opposition successfully characterized the proposition as “a Republican power grab.” In 2008, the “Yes on 11” campaign succeeded by focusing on anti-incumbent sentiments.
Nicholas Mosich, on the other hand, argues that 2005’s Proposition 77 actually lost because of three different factors. Specifically, he identified: “(1) California’s history of resistance to redistricting reform initiatives, (2) fierce bipartisan opposition to Proposition 77, and (3) voters’ perception of the special election as a referendum on Governor Schwarzenegger’s leadership.”
Looking solely at the 2005 election, Mosich’s conclusions are appealing; they do not, however, explain the subsequent success of Proposition 11. First, Californians certainly have demonstrated a reluctance to adopt a new redistricting procedure. Indeed, this reluctance may have manifested itself in the narrowness of Proposition 11’s victory. Nevertheless, the measure did taste victory. Furthermore, even if this history were a factor, one would expect it to have a greater effect in 2008, only three years after the defeat of Proposition 77, rather than in 2005, fifteen years after the last failed proposition. Second, Schwarzenegger’s approval and disapproval ratings were almost identical shortly before each election. According to the Field Poll, his approval ratings rose from thirty-seven percent in October 2005 to thirty-eight percent in September 2008; and during the same period, his disapproval ratings fell from fifty-six percent to fifty-two percent. Thus, two of the factors identified by Mosich do not help to explain Proposition 11’s success. The second factor he identifies, fierce bipartisan opposition, may help to explain Proposition 77’s failure. It does not, however, explain the failure of earlier propositions, which only one party opposed.
Thus, Stephanopoulos’ theory–that the legislative-majority party’s vigorous opposition to the measure is the most significant factor in the failure of a proposition–seems to explain best the failure of Proposition 11’s predecessors. The lack of such opposition also best explains Proposition 11’s subsequent success. This was the first redistricting proposition not to face concerted opposition from the majority party, and it was the first such proposition to succeed.
III. Competitiveness: Is It Desirable?
Supporters of Proposition 11 promised one result: competitive elections. But, are competitive elections desirable? Even if they are, are they attainable in California? These are the questions that the next two sections explore.
To sell Proposition 11 to the voters, its proponents maintained that it would increase electoral competitiveness. Governor Schwarzenegger, the proposition’s top contributor and fundraiser and highest-profile proponent, charged that “the current redistricting system . . . insulates lawmakers from competitive general elections.” Similarly, Democrat and former-Controller Steve Westly assured the public that the proposition would create more competitive election districts. But the “competitiveness” drumbeat did not stop there. Others who touted the proposition’s ability to increase competitiveness included redistricting experts, a non-profit organization, and several newspaper editorial boards.
Despite this focus on competitiveness in the 2008 campaign, no discussion arose concerning the merits of competitiveness. Several commentators have pushed for redistricting reform to enhance competition in general elections. They have raised three main benefits of greater competitiveness: increased electoral participation, election of moderate legislators, and greater responsiveness of those legislators to the needs of their constituents.
First, supporters maintain that greater competitiveness can benefit the entire electoral process. Competitiveness increases voter turnout rates. The political parties are sensitive to competition and focus their limited resources where elections are competitive. They target television advertising and other mobilizing efforts mainly in competitive races, and studies have found that persons contacted through mobilization efforts are more likely to vote. Presumably, greater interest also leads to more media attention, higher campaign contributions, and a sense that one’s vote matters. Accordingly, the closer the anticipated outcome of an election, the more voters become involved and the better informed they are likely to be. Furthermore, competitive elections can affect participation even after the election. For instance, after the 2000 presidential election, voter registration drives surged. In addition to their effect on voters, competitive elections are essential to other aspects of the democratic process. For instance, they encourage the appearance of strong challengers to majority-party candidates, which also stimulates party mobilization and campaign contributions.
Proponents also maintain that more competitive districts can yield more centrist candidates. Studies suggest that at the individual district level “more competitive seats lead to more moderate members and . . . ‘cross-pressured’ members are more likely to have more centrist voting scores.” When districts are not competitive, candidates know that voters in their districts are unlikely to support challengers from the other party. This enables candidates to be more attentive to those voters who cast ballots in primary elections, who tend to be more partisan and less moderate than voters in general elections. When they are focused on this sort of voter, candidates are more likely to take extreme positions instead of representing the political center. Thus, noncompetitive districts undermine centrists, who can win competitive general elections but not primaries in heavily-partisan districts. Noncompetitive districts thus impact the partisanship, and possibly the effectiveness, of legislatures. Of course, the hope of many supporters of competitive districts is that an increase in centrist legislators will reduce both partisanship and gridlock in legislatures.
Competitive districts may also increase the accountability of legislators to the voters. Districts that are competitive compel legislators to respect the interests of their constituents or face a realistic chance of defeat. Furthermore, districts with competitive elections prevent parties from becoming overwhelmingly dominant in geographic areas and lacking incentives to compete for voters. Some commentators consider accountability to be the central purpose of elections. Others consider it to have a constitutional basis or to function as part of the checks and balances fundamental to the Constitution. Another aspect of accountability regards corruption of elected officials. One study found that in the 1980s, among candidates charged with corruption, only 3.8% percent lost in primaries, while their loss rate in the general election was 25%.
Defenders of noncompetitive districts, however, point to several valuable functions they provide. They argue that noncompetitive districts provide better representation for their voters and promote stability of the Legislature. They also maintain that competitive elections, rather than inspiring crossover appeals, actually lead to “getting out the base” efforts.
Critics point out that, by definition, competitive districts leave more voters unrepresented. A major drawback of single-member districts is that the votes for the losing candidate are “wasted.” Closely-balanced districts therefore maximize the number of wasted votes. A district that is less competitive in general elections forces candidates to take positions more akin to those of the typical voter of the district; thus legislators elected from homogenous districts will be more representative of more of the district’s voters. Furthermore, from the perspective of voter satisfaction, competitive elections are less desirable. Unlike with sporting events, when it comes to elections voters prefer blowouts (large victory margins) to exciting finishes (competitive contests). Voters also prefer to have representatives with similar ideologies over having closely-contested general elections. As an additional psychological benefit, voters give Congress higher approval ratings when like-minded persons represent their districts. Thus, non-competitive districts may maximize voter satisfaction.
Another advantage pointed to by commentators is that noncompetitive districts increase the stability of the Legislature as a whole. With more competitive districts, partisan control of the Legislature would change more frequently. Thus, slight changes in voter preferences would shift control of the Legislature. At the extremes of competitiveness, a statewide vote of only fifty-one or fifty-two percent in favor of a party might shift control in the Legislature. Because of legal and geographic constraints, no districting plan for California will have all hyper-competitive districts. Nevertheless, the risk of shifts in partisan control remains.
Opponents of competitiveness also argue that high levels of competition can adversely affect individual officeholders and candidates. Candidates would likely be less interested in running for office if they knew that slight changes in political sentiment would remove them from office. Thus, less competitive districts can be more appealing to prospective candidates because greater stability makes the possibility of a career in the Legislature more likely. Conversely, competitive elections deter candidates because of the foreseeable burden of campaigning in future close elections. Incumbents who are concerned about upcoming competitive elections have greater incentives to steer pork barrel projects to their districts in attempts to “buy off” their constituents. Similarly, they are more likely to focus on parochial issues rather than on those of benefit to the larger whole, be it a region, state, or nation.
Critics also charge that competitive districts may rely upon a premise that is not always applicable. Proponents argue that competitive districts will force candidates to take less extreme positions. Candidates, however, do not always follow this strategy. For instance, after the 2000 presidential election, in which George W. Bush lost the popular vote but won the electoral college vote by 271 to 266, President Bush did not focus on winning swing voters. Instead of targeting moderate voters, his reelection campaign focused on mobilizing his own party’s voters. Since the majority of “independent” voters are not truly swing voters but actually favor one party, the drawing of competitive districts may not force candidates to broaden their appeal. Instead, such districts may actually heighten partisan appeals.
While the virtues of competitiveness are more intuitive, a lack of competitiveness may be beneficial. This is reassuring, since Proposition 11 is unlikely to accomplish its goal of increasing competition.
IV. Non-Competitive Elections: Might Something Besides Redistricting Cause Them?
Whether or not competitiveness is desirable, legislative elections in California have experienced a marked competitive decline, and supporters of Proposition 11 have insisted that legislative redistricting is the cause. A more careful study of applicable legal requirements and geographic considerations, however, indicates that Proposition 11 is unlikely to increase competitiveness significantly.
As previously discussed, federal and state laws constrain the drawing of legislative districts. Line drawers must populate the districts equally, and they must not dilute minority voting strength. California law also imposes additional mandatory (contiguity) and nonmandatory (respect for jurisdictional boundaries) requirements. The combination of these constraints, however, limits the ability of redistricters to draw more competitive districts.
Bruce Cain, Karin Mac Donald, and Iris Hui examined the impact of legal criteria on the drawing of competitive districts. They noted a truism of redistricting: the imposition of multiple criteria will “highly constrain” the accomplishment of any single goal. In other words, mandating more than one criteria necessarily will require trade-offs among criteria. Compliance with the Voting Rights Act, for instance, necessitates avoiding both minority vote dilution and retrogression of minority voting strength. Since African Americans and Latinos are predominantly Democratic, the legally-mandated majority-minority districts are usually heavily-Democratic and intentionally noncompetitive in general elections (so the minority population can elect its candidate of choice). Conversely, since these districts require the inclusion of large concentrations of Democratic voters, they deplete the pool of Democratic voters for surrounding districts, thereby facilitating–or even necessitating–the drawing of safe Republican districts.
Thus, the Voting Rights Act requirements have significant redistricting consequences. The Voting Rights Act reduces competitiveness not only by altering the configuration of districts, but also by reducing the pool of districts available for competition. In California, Cain et al. found that minority populations have been dispersing geographically. As a result, to satisfy the Voting Rights Act standards, map drawers need to extend districts to encompass pockets of ethnic communities. This constrains their options when populating surrounding districts. Moreover, states that must satisfy the section 5 non-retrogression requirement typically have fewer marginal districts and fewer districts with two major-party candidates. Consequently, preserving minority voting strength often occurs at the direct expense of electoral competitiveness.
Another factor that explains the decline in competitiveness is population redistribution. During the past two decades, our population has realigned itself geographically along political lines. Bill Bishop labels this phenomenon as “The Big Sort.” As a result, regions, states, and counties now are much more politically homogenous than they were just three decades ago.
This sorting is possible because we have a highly mobile population. For instance, in 2008, 11 million Americans moved to a different county. While the population of the United States has always exhibited a high rate of mobility, the nature of this movement has changed in recent decades. Beginning in the 1970s and 1980s, when people moved, they tended to relocate to areas where most residents held similar political perspectives. In general, when people select a locale in which to live, they choose where to live within that locale based upon factors that correlate with partisan preferences. These factors include immigration, education, income, and religion. Thus, Democrats began moving to Democrat-majority counties and Republicans to Republican counties. Similarly, as Democrats left Republican areas, Republicans were more likely to replace them, and vice versa. This trend differed markedly from the racial consequences of these movements. From 1980 to 2000, American counties became slightly less segregated, whereas during the same period, the segregation of Republicans and Democrats increased by almost twenty-six percent.
Within California, these shifts are readily apparent. The 1976 and 2004 presidential elections provide good points for comparison. Both races were very closely contested at the national level. In 1976, Jimmy Carter prevailed over Gerald Ford by a popular vote margin nationwide of 2.1% (50.1% to 48.0%); in 2004, George W. Bush defeated John Kerry by a similar margin of 2.4% (50.7% to 48.3%). During the twenty-eight years between these elections, forty-seven of California’s fifty-eight counties (81%) became more partisan. Specifically, seventeen counties (29.3%) became more Democratic while thirty (51.7%) became more Republican. Only eleven (20.0%) became more closely contested.
Three counties provide especially illuminating illustrations of this shift: San Francisco, Los Angeles, and Kern. San Francisco is an example of an area that became increasingly partisan during this period despite a stable overall population size. Figure 5 charts the Democratic and Republican votes in presidential general elections in San Francisco:
For the first three elections, San Francisco leaned slightly toward the Republicans. However, starting in 1960, the Democratic candidate won the county, and Democrats have won every presidential general election since then. Starting in 1976, the Democratic margin of victory has increased from the previous elections, with the sole exception being in 1996. By 2008, the Republican candidate (McCain) garnered only 13.6% of San Francisco’s vote. Those familiar with San Francisco’s geography will know that this shift in partisanship did not result from population growth (there is nowhere to add population). In fact, between 1960 and 2000, the total population in San Francisco County increased by just 4.9%; during this period, the statewide population increased by 115.5%. Thus, this change must have resulted from geographic sorting; specifically, an influx of Democrats and outflow of Republicans.
Neighboring counties Los Angeles and Kern provide striking examples of the divergence of growing populations, as one neighbor broke Democratic while the other Republican. Figures 6 and 7 chart these results:
From 1948 until 1984, Los Angeles County oscillated between supporting Republican and Democratic presidential candidates. Then, in 1988 it supported the Democratic nominee, and it has continued to do so since then by increasing margins. In Kern County, support for Democratic and Republican nominees stayed fairly close through 1976. Then, as in Los Angeles, the margin of support started to diverge, but in this instance in favor of Republicans. Moreover, unlike San Francisco, which had a relatively stable population from 1960 through 2000, both Los Angeles and Kern experienced significant growth. During this period, Los Angeles grew by 57.6%, while Kern grew by 126.6%.
This analysis focused on the trends evident in particular counties. The effects of this sorting are apparent, however, throughout the state. Beginning in 1948, sixteen general election contests (excluding propositions) had extremely close votes statewide; the final margins between the top two candidates were less than two percentage points. These close statewide results, however, mask the sorting occurring at the local level. Figure 8 presents the average percentage of votes cast by decade in landslide and “toss-up” counties in these narrowly-decided statewide elections:
Figure 8 illustrates the diverging trends in California. In the 1940s, 73.5% of California’s votes were cast in counties in which the statewide vote differential between the top two candidates was less than two percentage points. Only 9.9% of the vote came from counties decided by landslide margins–at least twenty percentage points. Thus, three-quarters of Californians lived in counties with nearly equal populations of Democrats and Republicans. Starting in the 1980s, however, in these narrow statewide elections more Californians cast their votes in landslide counties than in toss-up counties. This trend has continued, so that in the 2000s less than ten percent of the votes came from toss-up counties, while more than one-quarter were cast in landslide counties.
Although the analysis of twenty percent landslide counties demonstrates the growth of politically extreme counties, political scientists recognize that a differential of ten percent or less defines marginal contests. Thus, we could use ten percent as the cutoff to examine noncompetitive counties. When we do so, we see some interesting changes:
Not surprisingly, more voters live in ten percent noncompetitive counties than live in twenty percent landslide counties. Accordingly, the votes in noncompetitive counties basically draw even with the votes in toss-up counties by the 1960s and 1970s. The main difference, though, between Figures 8 and 9 is the magnitude in jump after the 1970s in the population in noncompetitive counties. By the 2000s, nearly ninety percent of votes in a statewide toss-up election were cast by voters who lived in counties decided by at least ten percent, or noncompetitive margins. Conversely, less than ten percent of the state’s voters lived in marginally competitive counties, even when the election was a toss-up.
These changes probably did not result solely from population movement. It is more likely that group dynamics pushed group members to greater extremes as one group or another became dominant in a county. Sociologists have studied group dynamics and the effects of group homogeneity on behavior. They have found that heterogeneous groups tend to be more moderate; the differences within the group restrain group excesses. Groups of homogenous persons, conversely, tend to move toward polarization. Group polarization occurs as homogenous groups discuss issues. Over time group members predictably move and coalesce, not toward a middle position, but toward a more extreme position than that held by the members initially. Continued dialog actually decreases variance among members and produces convergence on a relatively more extreme position. This result occurs for two reasons. First, individuals seek acceptance, and so they adjust their position to conform to the dominant perspective of the group. Second, with a relatively homogenous group, contrary positions are rarely considered, so the dominant perspective naturally becomes more convincing.
Pervasive evidence demonstrates that group polarization theory extends to issues that bear directly on politics and political behavior. When applied to political dynamics, homogeneous groups similarly become self-reinforcing. For instance, in landslide counties, political minorities participate less throughout the political process, from volunteering to voting. As minorities retreat, the majority gains confidence in its positions and becomes more extreme.
What are the implications of this population sorting for California’s redistricting reform? Critics of legislative re-districting charge that the Legislature draws noncompetitive districts that favor one party over the other. This is a comforting hypothesis since it provides a readily-curable cause: legislators redrawing their own lines and thus choosing their voters, for an effect: the rise in noncompetitive districts. Even better, not only does it have a cause, it identifies the “bad guys” who perpetrated it, and what better bad guys could one suggest than politicians? The hypothesis also has victims–besides the electorate generally, the centrists who otherwise would send moderate candidates to the Legislature. During the past two decades, however, county lines–which have not changed–have come to define highly-partisan enclaves. Although the line drawers decide the final district configurations, geographies underlying the districts have become more partisan anyway. Blaming legislators for the lack of competition may be appealing, but it ignores underlying realities.
Other analyses confirm that the ability of line drawers to create competitive districts in California is limited. Cain et al. found that California’s geography constrains the ability to draw competitive districts. When they instructed their map drawers to develop their random box plans (applying only equal population and compactness as constraints), fifty-three of the eighty assembly seats (66.3%) were unlikely to be even potentially competitive. In other words, before taking into account any other criteria, which will necessarily reduce competition further, two-thirds of California’s assembly districts will be noncompetitive.
The Center for Governmental Studies (Center), which helped to draft Proposition 11 and which supported the measure, acknowledged that increasing competition in California would be difficult. In addition to the concentrations of urban Democrats, Republicans predominate in large regions of the state, notably the Central Valley and much of Orange and Riverside Counties. The Center noted that imposing competitiveness as a redistricting criterion would require stretching districts from areas dominated by one party to those controlled by the other party. Even if this were possible, such practices likely would violate other considerations, such as the compactness of districts and the preservation of subjurisdictions and communities of interest.
Plotting results from the recent statewide election shows the problem that this sorting creates for drawing competitive districts in California. The 2010 contest for state controller was decided by less than one percentage point. Specifically, Harris, the Democratic candidate, defeated Cooley, the Republican candidate, by 0.6% of the vote. The results of this election, presented graphically in Figure 10, illustrate the geographic separation in California.
Figure 10 illustrates that one continuous bloc of counties supported the Democratic candidate, while another bloc supported the Republican candidate. Of California’s fifty-eight counties, only three (Alpine, Imperial, and Los Angeles) were not part of one of these two blocs. In the thirty-nine contiguous counties that supported Cooley, he won by a combined margin of 18.6% of the vote. In the sixteen contiguous counties that supported Harris, the Democrat won by a combined differential of 26.4%; in the three noncontiguous counties, her combined margin was 14.1%. Furthermore, only seven of the state’s counties were marginally competitive (differential between the top two candidates of less than ten percent). These seven counties accounted for only 7.7% of the state’s total votes.
This analysis illustrates the difficulty that line drawers will have in crafting competitive districts. Because of the geographic sorting that has occurred, most areas in California are highly partisan. Redistricters can draw competitive districts only by crossing city and county lines to combine different types of communities. In areas of political segregation, no commission can draw competitive districts. Furthermore, the requirements of the Voting Rights Act will further constrain the commission’s ability to draw competitive districts. Thus, the commission can fulfill the promises of Proposition 11’s supporters only by violating the only standards that the measure actually articulates.
V. Noncompetitiveness: Can Redistricters Provide the Cure?
The competitiveness of districts has declined, but geographic patterns appear to play a major role in this development. Can a change in the persons redrawing the lines alter this outcome?
To determine whether commission-controlled redistricting enhances competitiveness, Jamie Carson and Michael H. Crespin analyzed the results from the four Congressional redistricting cycles occurring between 1972 and 2002. Carson and Crespin concluded that legislative redistricting is more likely to lead to the creation of noncompetitive districts than other procedures. Specifically, plans drawn by commissions, or the courts, tend to produce a greater level of competition than legislative-drawn plans. Seth Masket, Jonathan Winburn, and Gerald C. Wright looked at the competitiveness of state legislative districts. They found some evidence that legislative redistricting resulted in less competitive elections when compared to elections in districts drawn by neutral commissions, especially in contests involving incumbents.
Although Masket et al. found that legislative redistricting created less competitive districts, another of their findings reinforces the geographic sorting hypothesis. They examined the difference in votes by assembly districts in presidential elections. Figure 11 presents the average difference in vote by districts from 1976 through 2004:
AVERAGE DIFFERENCE IN PRESIDENTIAL VOTE
BY CALIFORNIA ASSEMBLY DISTRICT
Figure 11 confirms that the vote differential increased after the legislative redistrictings of the 1980 and 2000 redistricting cycles. On average, the increase was by slightly less than three percent after the redistrictings. On the other hand, after the judicial redistricting in 1992, the differential declined by approximately one percent. The greatest change in differentials occurred, however, not after redistricting, but during the middle of the decades. In the 1970s, the differential increased by seven percent; in the 1990s, it rose by five percent (and was essentially flat during the 1980s). Alan I. Abramowitz, Brad Alexander, and Matthew Gunning similarly found that “[t]he most significant changes in competitiveness of [congressional contests] occurred between redistricting cycles.” Since the greatest decline in competitiveness occurred not after redistrictings, but between them, this suggests that the geographic sorting hypothesis better explains the decline.
Fortunately, two states, Arizona and Washington, already require that their redistricting commissions consider competitiveness in developing plans. If self-interested redistricting, rather than geographic sorting, better explains the decline in competitiveness, we should expect the districts in these states to exhibit significantly greater competitiveness than those in California–they do not.
Washington State had originally required its Legislature to redraw the state’s districts. Of the first ten redistrictings after statehood in 1889, however, the Legislature successfully redistricted only four times; the remaining times, the lines were redrawn by either the courts or voter initiatives. Therefore, in 1983, after the governor vetoed that decade’s redistricting bill, the Washington Legislature proposed and the voters approved Constitutional Amendment 74 to shift responsibility for redrawing the lines to a bipartisan commission. Pursuant to this amendment, the majority and minority leaders of the state Legislature each appoint one commissioner, and those four commissioners then appoint the remaining member. The commission then submits its plans to the Legislature, which may alter the lines, but only after approval from two-thirds of the Legislature.
As amended, Washington law delineates particular standards for the commission to follow in redrawing the lines. Washington law divides the state into forty-nine legislative districts. Each district elects one state senator and two members of the state house of representatives, who run for numbered posts. Legislative districts must have equal population. In addition, to the extent possible, districts should minimize splits of subjurisdictions and be compact and contiguous. Finally, the commission must “encourage electoral competition.”
Much was made of the apparent success of the commission in bringing change to the state’s congressional delegation. Indeed, prior to the 1992 redistricting, the delegation’s members had served an average of six terms in office. By 1994, the members averaged two terms. Table 4 tracks the changes in the party control of seats from the election, before the implementation of the commission’s first plan to the present:
WASHINGTON CONGRESSIONAL DELEGATION
As Table 4 shows, in the first election under the 1992 plan, Democrats won three new seats, two from Republicans and one as a result of the state receiving an additional seat through reapportionment. Two years later, the Republicans took six seats from the Democrats. Over the next three elections, the Democrats won back four of those seats. Thus, during the five elections under this plan, incumbents lost seven elections and seats changed party hands twelve times. Elections under the 2002 redistricting plan, however, were a different matter. Only one seat changed parties under this plan.
Despite the use of a bipartisan commission that needed to comply with a specific competitiveness requirement, Washington’s 2002 redistricting mirrored that of California. Much like the “incumbent protection gerrymander” passed by the California Legislature, the Washington bipartisan commission developed its own “status-quo plan.” As one of the members of the Redistricting Commission conceded, state legislative districts “tended to become slightly more Democratic if two or all three of their incumbent lawmakers were Democrats, and slightly more Republican if two or three incumbents were Republican.” In other words, the districts became less competitive.
An analysis of the Washington congressional and legislative districts reveals the dearth of competition under its 2002 status quo plan. Table 5 graphically presents the average margin of victory in the districts that the commission drew:
AVERAGE MARGIN OF VICTORY, WASHINGTON CONGRESSIONAL, STATE SENATE, AND STATE REPRESENTATIVE DISTRICTS
As Table 5 illustrates, from 1992 to 2008, the lowest average margin of victory in an election in districts drawn by the commission was almost 17%. In six of ten election cycles, the average margin of victory in general elections for state senate seats exceeded 40%. Overall, the average margins of victory under the commission’s plans were 24.6% in congressional contests, 40.8% in state senate races, and 38.4% in state representative elections.
Because a large number of state senate and legislative districts were so uncompetitive that candidates ran uncontested, these contests skew the average victory margin upward. Therefore, the next three charts present the distribution of contests by range of margin of victory: less than five percent, between five and ten percent, greater than ten percent, and uncontested (no congressional races were uncontested):
These distributions also make apparent the increase in uncontested (presumably extremely safe) seats after the January 2002 “status quo” plan. Furthermore, despite the imposition of a competitiveness requirement, in seven of nine years for state senator and six of nine years for state representative, the number of uncontested elections actually exceeded the number of toss-up elections.
Arizona provides a second example of a state that includes a competitiveness requirement. In 2000, Arizona voters approved an initiative that mandated consideration of competitiveness as a criterion for redistricting. The measure, Proposition 106, mandates creation of a five-member Independent Redistricting Commission (IRC) to perform the redistricting. Arizona’s Commission on Appellate Court Appointments nominates candidates. The majority and minority leaders of the state senate and house each then appoint one commissioner. Next, the four commissioners appoint the fifth member, who serves as the chair. No more than two members may be from the same political party. The Arizona Constitution requires this commission to redistrict the state’s congressional and legislative districts.
As amended, the Arizona Constitution requires the commission to develop initial districts of equal population in a grid-like pattern. In the next phase, the commission makes adjustments as necessary to accommodate the six goals identified by Proposition 106 (equal population, Voting Rights Act compliance, compactness and contiguity, respect for communities of interest, geographic features and jurisdiction boundaries, and competitiveness). In the remaining two constitutionally-mandated phases, the commission receives comments on its plan and makes final adjustments.
The commission’s approval of a final plan in 2002 sparked litigation that did not conclude until a ruling by the Arizona Supreme Court seven years later. Concerning competitiveness, the Arizona Supreme Court concluded that the Arizona Constitution, as amended by Proposition 106, required that the IRC create “more competitive districts to the extent practicable when doing so does not cause significant detriment to the other goals.”
Despite this goal of creating more competitive districts, the chair of the IRC conceded that most Arizonans would consider the commission’s work in this regard to be “an abject failure.” He elaborated, “If your goal is competitive districts, I don’t think this helps you get down that road very far.” In an analysis of the commission system, The Arizona Republic concluded that the commission “failed to meet a primary goal of making legislative elections more competitive.”
Analyses of the elections in the IRC’s districts support these conclusions and suggest again that geographic sorting cannot be overcome, even when governing law specifically instructs map drawers to do so. Despite the IRC’s charge to craft competitive districts, the resulting districts were anything but. Table 6 illustrates the average margin of victory in the districts that the commission drew:
AVERAGE MARGIN OF VICTORY, ARIZONA CONGRESSIONAL AND STATE SENATE DISTRICTS
Table 6 shows that the average margin of victory in districts drawn by the IRC approached thirty percent. Specifically, the average victory margin in general elections in congressional districts was 26.9%. For state senate contests, Table 6 shows that the average margin of victory was 51.6%.
As with the Washington legislative elections, the large number of uncontested seats distorts these numbers. Therefore, Figures 15 and 16 reflect the ranges in which these contests fell:
As with Washington’s elections, the vast majority of Arizona’s districts were decided by margins that exceeded ten percent. Of the 150 state senate contests over eight years in the “competitive” districts crafted by the IRC, only eighteen had margins below ten percent, and only four of those fell within five percent. Furthermore, in nearly one-third (forty-seven) of these contests, the winner did not face an opponent in the general election.
Why was the commission incapable of drawing competitive districts? Professor Michael McDonald, who worked as a consultant with the commission, noted that the state’s redistricting requirements (similar to those of California except for the addition of competitiveness) prevented the creation of many competitive districts. The commission’s chair pointed to the Voting Rights Act’s protection of minorities and their tendency to vote Democratic, and, echoing the findings of Cain et al., the IRC chair noted that the resulting concentration of Democrats in a small number of districts left few Democrats with which to make the remaining districts competitive.
What do the experiences of Washington and Arizona suggest for California? Because of Arizona’s sizeable minority population, redistricters’ hands were tied when trying to draw competitive districts. California, however, has a much larger minority population than does the Grand Canyon State. To the extent demographics limited the competitiveness of Arizona’s districts, California’s redistricters will surely find their hands even more tightly bound by their own state’s demographics.
Washington, however, had a proportionately smaller minority population than either Arizona or California had. Nevertheless, its commission did not craft significantly competitive districts. Instead, it chose to develop a “status quo” plan. More than anything else, this experience confirms that commissions, be they purportedly bipartisan or nonpartisan, are no more insulated from political considerations than is the Legislature. Washington’s Redistricting Commission’s five members could not agree on redistricting plans. After missing the statutory deadline, the members eventually agreed to make only minimal changes to the previous plan to adjust for population shifts during the past decade. In the end, the only conclusion to which they could all agree was to sacrifice competitiveness.
VI. Is Proposition 11’s Redistricting Commission the Best Means to Redraw California’s Lines?
Will Proposition 11’s Citizens Redistricting Commission be able to draw more competitive districts? Arizona and Washington, states which have imposed competitiveness goals on their redistricting commissions, have had little success in achieving competitiveness. Since California has not statutorily included competitiveness among its requirements, the likelihood of California’s commission achieving significant competitiveness in the state’s districts is minimal. Demographics and federal law will work against the commission. Arizona and Washington, states with pre-dominantly white populations, needed to concentrate few of their minority (and typically Democratic) voters into districts to comply with the Voting Rights Act. Nevertheless, few of those state’s districts are competitive. California’s majority-minority population, however, necessitates the drawing of numerous majority-minority, heavily-Democratic districts. This concentration of Democratic voters, along with the population’s geographic sorting, will render the drawing of a significant number of competitive districts quite difficult.
If the commission will not be effective, might it actually be a step back? Some of the differences between independent commissions and the Legislature may make the CRC the less desirable body to redistrict the state.
One of the primary arguments for adopting a commission was to take redistricting away from the self-interested Legislature. The Legislature may have a vested interest in the outcome, but it also has more relevant knowledge and experience. Legislators are extremely familiar with their districts, their constituents, and their needs, and they usually have a better understanding of these concerns than do outsiders. Legislators are thus best able to tailor districts to represent constituent communities and their interests. Nathaniel Persily experienced this firsthand when he assisted courts in drawing redistricting plans for New York and Maryland. In one instance, he moved an uninhabited swamp from one district to another. Since this was uninhabited swampland, a person unfamiliar with the district, such as Persily, would justifiably have thought that such a move would have no redistricting consequences (since it had no population) and no political consequences (since the land had little value). A legislator informed Persily, however, that this shift would disrupt environmental projects that the legislator initiated and hoped to complete. Thus, a move that would have no apparent political effect had tangible policy consequences: persons unfamiliar with the district would not be able to incorporate this concern.
Furthermore, the elected nature of legislators, rather than rendering them less qualified, actually makes legislators better suited to make the choices required by redistricting. The remap process inherently involves tradeoffs among numerous communities, constituent interests, and policies. Redistricting “involves give and take in resolving conflicts among the various standards and in considering the concerns, desires, and objections of numerous interested persons and groups.” Line drawers also make decisions about service relationships between representatives and constituents and their placement within larger policy programs or decisions. Legislators routinely balance complicated policy choices and, as elected representatives, are particularly qualified to do so. The Supreme Court has recognized that legislatures exercise political judgment in balancing competing interests, and that legislatures are the institutions “best situated to identify and then reconcile traditional state policies” within the redistricting framework. In contrast to legislatively-controlled redistricting, an appointed commission not only empowers less experienced persons to make these tradeoffs, but the commissioners also lack the accountability for their actions that legislators must confront with each election.
Legislative redistricting does have significant advantages. Moreover, legal and demographic hurdles will prevent the CRC from achieving its objective of significantly increasing the competitiveness of California’s districts.
VII. Keeping the Commission and Returning Redistricting to the Legislature–But With a Twist
Because of the Legislature’s knowledge regarding factors that are relevant to the formulation of districts, as well as its experience in balancing interests, it is the appropriate body to draw the redistricting plan. An independent commission, however, could still serve a useful function in the redistricting process.
The commission could review the plan developed by the Legislature and propose changes. To add teeth to its recommendations, California law should give deference to the commission’s proposals.
An independent commission is better suited to review a redistricting plan, rather than to create it. Because of commission members’ lack of familiarity with the communities and the government’s relationship to them, they will benefit by having additional time to get up to speed and by observing the Legislature’s redistricting hearings and decisions. Once the Legislature has completed its plans, the commission could then review those lines. Their review could proceed at three levels:
(1) Overall architecture–this would look at the overall demographics of the plan. The commission would compare the share of registered voters in each major party and the racial and ethnic percentages of the statewide and regional population and compare them to the anticipated totals for the proposed plans. The objective of this review is to ensure that no groups are especially over-or under-represented at the statewide level. An example of a plan that would cause the commission concern would be the 2002 Ohio redistricting plans, in which Republicans controlled 61.6% of the seats even though they represented only 49.0% of the state’s registered voters.
(2) Review of specific districts–the commission would review the configuration of specific districts for irregularities. Concerns here would include unnecessary splitting of communities of interests and sub-jurisdictions, lack of compactness, irregular district shapes, combination of dissimilar communities, and other anomalies.
(3) A consideration of specific lines–this would focus on specific streets and geographic features that the plan uses to form districts. Practices that the commission addresses here might include the “Kiskaddon Pimple” and the “Rasmussen Stovepipe” from the Washington plans–situations where the overall district configuration is acceptable but a particular district’s exclusion or inclusion of a few blocks lacks justification.
To assist its review, the commission would consider public testimony provided to the Legislature and the complete record of its deliberations. It could also convene its own hearings to receive public comment about the Legislature’s plans, which would help to direct the commission’s attention. After concluding its review, the commission would submit to the Legislature written comments concerning the acceptability of its plan.
To encourage the Legislature to adopt the commission’s recommendations, the deference that courts normally apply to the Legislature’s plans should instead shift to the commission’s work. As discussed previously, prior to the adoption of Proposition 11, California law mandated that the Legislature redistrict the state. Under that process, the courts held that California law entitled the Legislature’s determinations to great deference as long as they constituted reasonable applications of controlling state and federal law. Courts extended such deference even when equally reasonable alternatives might be available. Courts deferred to the Legislature in the absence of a showing that it unmistakably violated a particular provision of the law. When considering a legislative redistricting, the court not only applied deference, but judicial restraint. For constitutional challenges to legislative redistrictings, courts have presumed that the plans were constitutional and placed the burden of proving a violation upon the challenger.
If the courts instead extend this deference to the conclusions of the reviewing commission, the Legislature would confront a choice. It could modify its plan to be consistent with the comments of the commission, or it could decline to alter its plan. However, in any subsequent challenge on grounds raised by the commission, courts would defer to the commission’s recommendations as long as they were reasonable. Thus, to defend its unaltered plan, the Legislature would need to overcome the deference extended to the commission’s recommendations.
Two examples show the effectiveness of this change. First, assume a commission had reviewed the Washington redistricting plan that contained the “Rasmussen Stovepipe.” A reviewing commission might recommend that the narrow “Stovepipe” extension sliced through a community, combined dissimilar populations, and should be eliminated. In future litigation, the Legislature would need to argue that such a recommendation was not reasonable. In a second example, assume a commission had reviewed the Ohio redistricting plan and suggested that the Legislature modify it to balance more evenly the number of majority-Democratic and majority-Republican districts under the plan. Remember that Republicans controlled sixty percent of the districts even though they held only a forty-nine percent to forty-eight percent registration lead statewide. Again, in any future litigation, the Legislature would need to explain why drawing a more balanced plan was not reasonable. Conversely, if the commission recommends that a politically-balanced plan be tilted to favor one of the parties, the Legislature might be willing to contest in court the reasonableness of such a recommendation in a legal challenge to its plan.
The establishment of a reviewing commission thus has several advantages. First, it places upon the Legislature a tremendous burden to overcome if it decides not to adopt the commission’s recommendations. Even if the Legislature believes its initial plan is justified, it must be able to establish that the commission’s alternative is not even reasonable. Because of the high burden it must satisfy, counsel often will urge the Legislature to adopt the recommendations so as to retain control over the remap process. A second advantage is that the Legislature retains responsibility for the initial architecture of the plan in the Legislature, the body that is most familiar with the pertinent representational considerations, the best able to begin the process quickly, and most accountable to the voters. This system, however, would provide a significant check on the Legislature. Third, this system better utilizes the commission, allowing it more time to prepare and not demanding that it learn the minutiae of district representation, while providing a fuller record for its consideration. Fourth, it allows the Legislature to retain its plan if it believes the recommendations of the commission are not reasonable. Finally, it allows for public comment after the Legislature has developed its plan.
Proposition 11’s redistricting commission is unlikely to provide a significant change in the competitiveness of the state’s districts. A better approach would leave redistricting in the hands of the body most experienced in performing the policy trade-offs required by the redistricting process. A commission could be most helpful not in drawing lines, but in reviewing the maps developed by the Legislature.
DISTRICTS WITH 5% REGISTRATION DIFFERENCES
BETWEEN REPUBLICANS AND DEMOCRATS
1992 AND 2002 REDISTRICTING PLANS (See PDF)