Congressional redistricting and electoral turnout

Material Information

Congressional redistricting and electoral turnout why competitiveness matters
Hayes, Thomas Joseph
Publication Date:
Physical Description:
x, 124 leaves : ; 28 cm

Thesis/Dissertation Information

Master's ( Master of arts)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
Department of Political Science, CU Denver
Degree Disciplines:
Political science


Subjects / Keywords:
Voting -- United States ( lcsh )
Political participation -- United States ( lcsh )
Election districts ( fast )
Political participation ( fast )
Voting ( fast )
United States ( fast )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )


Includes bibliographical references (leaves 116-124).
General Note:
Department of Political Science
Statement of Responsibility:
by Thomas Joseph Hayes.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
71634779 ( OCLC )
LD1193.L64 2006m H39 ( lcc )

Full Text
Thomas Joseph Hayes
B.A. University of Colorado at Boulder, 2004
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Arts in Political Science

This thesis for the Master of Arts in Political Science
degree by
Thomas Hayes
Has been approved


Hayes, Thomas Joseph (M.A., Political Science)
Thesis directed by Professor Michael S. Cummings
The purpose of this thesis is to investigate the effect that the process of
congressional redistricting has on voter turnout. This thesis also details various
theories that attempt to explain why the United States often has a lower electoral
turnout than other advanced democracies. In addition, this thesis explores
prospects for reform such as independent redistricting commissions as well as
alternative voting methods, which can create more competitive congressional
elections as well as a more fair and reasonable system than the current system of
This abstract accurately represents the content of the candidates thesis. I
recommend its publication.

My thanks to Professor Michael Cummings for his support and help throughout
my research; I would not have been able to write the thesis I did without the
thoughtful feedback and encouragement I received from him. I would also like to
thank Professor Anthony Robinson, who first introduced me to the various topics
presented in this thesis and showed me that elections can be extremely interesting.
My thanks to Thad Tecza for his constructive criticisms and alternative
viewpoints; his feedback was invaluable when crafting my thesis. I would also
like to thank Pete Maysmith, Jenny Rose Flanagan, and Jon Goldin-Dubois for
giving me the opportunity to intern with Colorado Common Cause. Without this
internship, my thesis would not have had the same outlook or knowledge that I
gained from such a wonderful opportunity.

I. INTRODUCTION.........................................1
Redistricting, the Constitution and the Supreme Court.6
Race and Redistricting...........................8
Electoral Competitiveness and the Supreme Court.14
What Congressional Elections Look Like..........17
A Growing Backlash and Efforts at Reform..............22
Class Theory/Structural Oppression Theories...........25
Inadequacy of Class Theories....................28
Legal/Institutional Barriers..........................29

Winner-Take-All System..........................29
Scheduling of Elections.........................30
Registration Rules..............................31
Inadequacy of Legal/Institutional Barriers......36
Inadequacy of Disenfranchisement Theories.......42
Educational/Socialization Theories....................43
Preparing Children to be Nonvoters..............44
Media Theories..................................46
Inadequacy of Educational/Socialization Theories.47
Party-Decay Theories..................................48
Inadequacy of Party-Decay Theories..............51
Americas Voter Turnout Puzzle.........................52
The Impact of Redistricting on Competitiveness........55
Noncompetitive Elections and Turnout..................64
Can the VRA Decrease Turnout?.........................68
An Increase in Congressional Responsiveness?..........74

PROSPECTS FOR REFORM...................................78
Independent Redistricting Commissions...............78
A Few Model Commissions and Proposals...............87
Failed California Initiative.................92
Failed Ohio Initiative.......................93
Coming Florida Initiative....................94
Coming Massachusetts Initiative..............95
Analysis of State Efforts of Creating Redistricting
Research Limitations and Future Research...........100
Other Options and Alternatives.....................102
Term Limits.................................102
A New and Improved Voting Rights Act........104
Alternative Voting Methods..................106
Changing the Size of Congress...............110

Conclusion: Why Competition Matters............113
List of Supreme Court Cases Cited................125

1 North Carolina 12th District....................................12
2 Congressional District 4, Louisiana.............................13
3 Congressional District 13, Florida..............................13
4 U.S. Midterm Turnout, 1962-2002, Based on House Votes Cast......18
5 Comparison of Competitive Races 1994 to 2004....................21
6 Floridas 2004 Voter Registration by Party......................61
7 Party Composition of Florida Legislative and Congressional
8 Floridas Congressional Districts and Registration Advantage
9 Turnout in Midterm Elections, 1850-2002.........................68
10 Differences in Turnout for Blacks and Whites
in Presidential Elections (1964-2004)...........................72
11 Differences in Turnout for Blacks and Whites
in Midterm Elections (1966-2002)................................73
12 Congressional District 7, Colorado..............................84
13 Congressional Districts by Population and Proposed

1 Percentage of Democratic and Republican Representatives in High-
Risk and Low-Risk Districts by Decade...........................56
2 Maryland Congressional District Margin of Victory and Turnout
3 Competitive Districts and Turnout in Washington.................88

Citizen participation in politics is vital to a healthy democracy. A true
democratic form of government must have a free and fair election process so that its
citizens have the ability to elect their representatives and participate meaningfully.
The United States is somewhat unique compared to other world democracies in that it
often has a lower electoral turnout in most levels of government. Turnout and
participation have been declining recently as fewer people are voting and
participating in the political system than at any time in recent history. While
Presidential turnout has been declining, local and congressional races typically
receive an even lower turnout. There are a vast number of theories as to why
America often has a lower electoral turnout than other advanced democracies;
however, one factor that is inadequately discussed in analyses of voter turnout is
Americas process of legislative and congressional redistricting.
The goal of this thesis is to explore the relationship between legislative
redistricting and voter turnout. There has been much debate in political science about
whether redistricting has any effect on voter turnout, and many argue that it has no

effect at all. However, this paper will argue that the process of legislative and
congressional redistricting, and the result of this process, do affect voter turnout in the
United States.
While redistricting itself is not the main reason why fewer Americans vote
than do citizens in other democracies, it does fall into, and should be included with,
other theories of voter participation when we consider why American electoral
turnout is often lower than in other advanced democracies. The American process of
congressional and legislative redistricting often creates a situation in which the
representatives choose their voters, rather than the voters choosing their
representatives (Hill, 2002). People are less likely to participate in a system that they
see as rigged. In addition, even the appearance of gerrymandering makes voters
weary of the system and those in charge, which in turn makes them less likely to
participate in the political system. As my research will show, when people feel the
election they are voting in matters (in the sense that it is competitive) they are more
likely to vote in that election. Since the American process of congressional and
legislative redistricting is increasingly creating opportunities for politicians to create
noncompetitive elections, voter turnout suffers.
Some of the questions this paper will set out to answer are: What is the current
state of congressional and legislative redistricting in the United States? How have
past redistricting efforts and court cases shaped the way in which districts are drawn
today? Does the redistricting process lead to less competitive elections? What

impact do gerrymandering and congressional redistricting have on voter turnout?
Where does research about congressional redistricting fall regarding theories on voter
turnout? With the recent move in many states to create independent redistricting
commissions and the growing backlash against safe districts, what are some
alternatives to the current process of legislative and congressional redistricting? Are
these alternatives plausible, possible, and even desirable compared to the current
In order to answer these questions the research methodology that I have used
is to first look at the existing literature on the redistricting process. The scholarship
on the topic of redistricting is vast, as some areas focus on whether redistricting
creates electoral responsiveness, while other areas focus on the issue of race and
redistricting. For the purposes of this paper, I have largely focused on the literature
concerning whether congressional and legislative redistricting contributes to low
voter turnout. However, part of my analysis does include literature on race and
redistricting, as Supreme Court cases of the 1980s and 1990s led to the creation of
majority-minority districts, as a result of which racial considerations were a huge
factor in redistricting politics and have affected competitiveness. In addition to
looking at the current scholarship on the subject of redistricting, I have done my own
research in the area of voter turnout and redistricting. I have attempted to evaluate
existing election data and determine whether the creation of safe legislative districts
leads to a lower voter turnout. In order to do so, I have utilized data from various

election resources including the Census Bureau, multiple Secretary-of-State websites,
and election results from many news organizations. Finally, in order to investigate
alternatives to the current practice of congressional redistricting, I have worked with
Colorado Common Cause in order to come up with recommendations and alternatives
to the current system such as independent redistricting commissions.
This thesis is organized into six chapters. Chapter 2 gives a broad overview
of the current situation of congressional and state redistricting in the United States.
Also included in this chapter is a brief summary of major Supreme Court decisions
that have shaped the way districts are drawn in the United States. Chapter 3
summarizes some of the main voter turnout theories for American elections. This
chapter includes theories that help to explain why American electoral turnout is often
lower than in other advanced democracies. This chapter divides major theories into
five categories: class theory/structural oppression, legal and institutional barriers,
disenfranchisement, educational and socialization, and party-decay theories. An
understanding of these theories is necessary in order to place my research into the
field of voter turnout. I also show why each of these theories by itself is inadequate,
which is why research looking at the relationship between redistricting and turnout is
important. Chapter 4 examines the link between voter turnout and congressional
redistricting. Chapter 5 gives recommendations for independent redistricting
commissions and advances guidelines for states considering the adoption of
commissions. Chapter 6 serves as the conclusion, which gives other alternatives to

the current process of legislative and congressional redistricting in the United States.
This chapter also summarizes the main conclusions of this research paper and lays the
foundation for future research.

Redistricting, the Constitution
and the Supreme Court
The United States is unique among advanced democracies in that it still allows
state legislatures to draw boundaries for their own races (Butler and Cain, 1992).
This power is derived from the U.S. Constitution, which states that, The Times,
Places and Manner of holding Elections for Senators and Representatives, shall be
prescribed in each State by the Legislature thereof... (Article 1, Section 4). Each
state must also have at least one representative to congress (Article 1, Section 2). The
apportionment of representatives in Congress is based on the population of each state
(14th Amendment, Section 2). Therefore, the Constitution leaves it up to the state
legislatures to determine how, and whether, congressional districts are drawn, often
leaving the power in the hands of the party in control of the state legislature and thus
leading to gerrymandering, which is defined as The drawing of constituency
boundaries deliberately to secure party advantage (Butler and Cain, 1992, p. 158).

During the first 120 years of the United States, Congress determined not only
how many seats would be apportioned to each state, but also how many total districts
there would be. Until 1910, Congress allowed itself to grow according to population.
Initially one seat was allotted for each 33,000 people, changing to 47,700 in 1830,
and 70,680 in 1840. This approach led the House to grow from 106 members in 1790
to 391 in 1900. In 1910, Congress set a permanent limit of 435 members. As a result
of this ceiling, by the time of the 2000 census, each congressional representative had
a district with an average population of 646,952, or 19 times greater than at our
founding (, 2006).
The United States Supreme Court has further defined the rules that the
redistricting process must follow. In the 1960s, three important cases reached the
Supreme Court. In Baker v. Carr (1960), voters in Tennessee complained that the
state General Assembly had violated their equal-protection rights because the state
had not been reapportioned since 1901, even though there had been a substantial
change in the population of the state. The Court ruled the voters did have standing in
the case and did not define the complaint as a political question. Voters having
standing in this case propelled the Supreme Court into the redistricting process, which
until then the Court had stayed out of, viewing such issues as political matters. After
the voters were ruled to have standing, the Court opened up the possibility for more
challenges for apportionment under the 14th Amendments equal-protection clause
and ruled that redistricting must occur every 10 years (Butler and Cain, 1992). In

Wesberryr v. Saunders (1964) the Court found that, The constitutional requirement in
Art. I, 2, that Representatives be chosen by the People of the several States means
that as nearly as is practicable one person's vote in a congressional election is to be
worth as much as anothers. In Reynolds v. Sims (1964), the Court ruled that state
legislative and congressional districts had to be roughly equal in population.
In the 1970s the Court developed a standard of population equality in two
cases, Gaffney v. Cummings and White v. Regester, which required legislative
districts to differ by no more than ten percent from the smallest to the largest, unless
justified by some rational state policy.1 2 In the 1980s the Court further developed
the standard of population equality for congressional districts that required them to be
mathematically equal, unless justified by some legitimate state objective. While
redistricting has traditionally occurred once after each census, the issue of whether
the Constitution allows for redistricting to occur mid-census is currently being
challenged after the Texas legislature, led by former House Majority Leader Tom
DeLay, redrew districts in 2003 (Biskupic and Drinkard, 2006).
Race and Redistricting
In addition to dealing with questions of equal population in congressional
districts, the Supreme Court has also taken up the issue of race, which has
1 Gaffney v. Cummings
2 Karcher v. Daggett

significantly affected the way in which congressional districts are drawn. In the
1970s, the Court had said that a redistricting plan did not need to be pre-cleared by
the Justice Department if it would be likely to cause fewer minority representatives to
be elected than before {Beer v. United States, 1976). In the 1980s, the Court ruled
that a redistricting plan would not be found to discriminate against a racial minority
unless the plaintiffs in the case could prove that the drafters of the plan intended to
discriminate against them {City of Mobile v. Bolden, 1980). In 1982, responding to
the Courts limitation on how to prove racial discrimination, Congress amended
Section 2 of the Voting Rights Act to clarify that it applied to any plan that results in
discrimination against a member of a racial minority or ethnic group, regardless of the
intent of the policy.3 Interpreting this legislation, in the 1986 case of Thornburg v.
Gingles, the Court defined three preconditions that a minority group must prove in
order to establish a violation of Section 2 of the Voting Rights Act; they are:
1) That the minority group is sufficiently large and geographically compact to
constitute a majority in a single-member district; 2) that it is politically
cohesive, that is, it usually votes for the same candidates; and 3) that, in the
absence of special circumstances, bloc voting by the white majority usually
defeats the minority's preferred candidate {Thornburg v. Gingles, 1986).
3 Pub. L. No. 97-205, 3, June 29, 1982, 96 Stat. 134, coded as 42 U.S.C. 1973. See also U.S.
Department of Justice, Section 2 of the Voting Rights Act accessed at
http:/' 2/about sec2.html

In his summary of Supreme Court cases of the 1990s, Peter Wattson, counsel
to the Minnesota state senate, describes the impact of the cases of the 1980s relating
to race:
If the minority group could establish those three preconditions, it would be
entitled to proceed to the next step: proving a § 2 violation by "the totality of
the circumstances." Those circumstances would have to show that the
members of the minority group had "less opportunity than other members of
the electorate to participate in the electoral process and to elect representatives
of their choice." What does that mean, "less opportunity"? In North Carolina,
where Gingles arose, it meant replacing multimember districts where blacks
were in the minority with single-member districts where blacks were in the
majority. To the rest of the country, and to those of us who were going to be
drawing new districts after the 1990 census, it meant that wherever there was
a racial or ethnic minority that was "sufficiently large and geographically
compact to constitute a majority in a single-member district," we would have
to draw a district for them or risk having our plan thrown out, even if we acted
without any intent to discriminate. Being forewarned of the impact of § 2,
drafters of redistricting plans after the 1990 census went to great lengths to
draw majority-minority districts wherever the minority population counts
seemed to justify it. There was intense pressure to go even further, and that
pressure caused race to be the dominant issue for redistricting in the 1990s,
especially for the U.S. Supreme Court (Wattson, 2005).
The Voting Rights Act also gave extraordinary powers to the Justice
Department. In the 1990s, when new redistricting plans were drawn in preparation
for the 1991 and 1992 elections, the Republican Party controlled the Justice
Department. As states presented their plans to the Justice Department for approval,
the Justice Department pushed for the creation of additional majority-minority
districts wherever the minority populations could be found to create them. As Peter
Wattson writes, It seemed as though the Justice Department was requiring the States

to maximize the number of majority-minority districts (Wattson, 2005). How this
affected turnout will be further explored in Chapter 4.
This policy created a situation in which states produced racially
gerrymandered districts that were challenged in court. One example is North
Carolinas 12th district, which is shown in Figure 1 (Page 12). The 12th Congressional
district in North Carolina was challenged three different times and reached the
Supreme Court after each challenge was filed. The district stretched for 160 miles
across the state and was created to include mostly African-Americans. In Shaw v.
Reno, which was the third challenge to change the district, the Supreme Court
concluded that since the redistricting plan had no geographically compact minority
population that existed and no narrowly tailored plan aimed at serving a compelling
state interest that would justify the creation of racially gerrymandered districts, the
plan violated the Fourteenth Amendments equal protection clause (Shaw v. Reno,
1993). In the majority opinion, which was decided by a vote of 5 to 4, the Court
argued that in apportionment, the shape of a district does matter. As Justice
OConnor wrote in her majority opinion:
A reapportionment plan that includes in one district individuals who belong to
the same race, but who are otherwise widely separated by geographical and
political boundaries, and who may have little in common with one another but
the color of their skin, bears an uncomfortable resemblance to political
apartheid. It reinforces the perception that members of the same racial group-
regardless of their age, education, economic status, or the community in which
they livethink alike, share the same political interests, and will prefer the
same candidates at the polls.. .By perpetuating such notions, a racial
gerrymander may exacerbate the very patterns of racial bloc voting that

majority-minority districting is sometimes said to counteract (Shaw v. Reno,
Figure 1
North Carolina 12th District
Bizarre district shapes exist in all states, but throughout the 1990s, extremely
bizarre shapes sprouted as a result of the effort to create majority-minority districts as
Figures 2 and 3 demonstrate (page 13).

Figure 2
Congressional District 4, Louisiana
Figure 3
Congressional District 13, Florida
$rve s

The Supreme Courts focus on race in Thornburg v. Gingles led to the
creation of majority-minority districts. In Miller v. Johnson, (1995) the Court turned
its attention away from the shape of the district and ruled that race could not be the
predominant factor in redistricting unless there was a compelling state interest.
Electoral Competitiveness and the Supreme Court
While race was the dominant issue for the Court in the 1990s, it was not the
only issue. The Supreme Court also took up the issue of whether self-entrenching
practices4 like gerrymandering should be allowed. In the 1997 decision in Timmons
v. Twin Cities, the Court decided that the states interest in regulating its electoral
processes outweighed voters associational rights, namely First Amendment rights.
The case involved a candidate running unopposed as the nominee of a major party for
a state representative seat. The Twin Cities Area New Party (a minor party) wished
to nominate the same candidate for the same seat under its party in addition to the
major-party nomination, thus allowing the candidate to appear on the ballot as a
multi-party candidate. However, most states (including Minnesota) had outlawed this
practice, as these prohibitions tend to hinder the growth of third parties by reducing
their ability to partner themselves with established parties (First Amendment Center,
4 Self-entrenching practices refer the legislative practice of insulating incumbents from electoral
competition. This practice can come in the form of bi-partisan gerrymandering-whereby both parties
act to create safe districts for each other, or partisan gerrymandering-where one party attempts to
shut out the other.

2006). The Rehnquist Court upheld the Minnesota law that banned the practice of
allowing a candidate to appear on the ballot as the nominee for more than one party.
As Rehnquist wrote in the majority opinion, a states interest in ensuring the stability
of its political system
.. .does not permit a state to completely insulate the two party system from
minor parties or independent candidates competition and
influence.. .[however] the States interest permits them to enact reasonable
election regulations that may, in practice, favor the traditional two-party
system and that temper the destabilizing effects of party splintering and
excessive factionalism (Timmons v. Twin Cities, 1997).
Due to the highly political nature of redistricting, many cases will reach the
Supreme Court in the future. To this day, however, the Court has largely been
complacent (as it was in Timmons) regarding the practice of legislative self-
entrenchment. As Jeremy Buchman writes:
In earlier rulings, the Court established a balancing test in which
infringements on the First Amendment right of association were to be
weighed against a states right to regulate its elections and its interest in
promoting political stability. Under this approach, some state regulations
adversely affecting third parties were struck down, while others were
upheld5.. .If Timmons is a reliable guide to the Supreme Courts views on the
states role in managing electoral competition, then one should not expect
self-interested redistricting to undergo searching judicial scrutiny anytime
soon (Buchman, 2003, p. 195).
Future rounds of redistricting and cases brought to the Supreme Court will
inevitably involve issues of legislative self-entrenchment practices. It does seem
5 For the case-by-case approach established by the Court, view Justice Stevenss majority opinion in
Anderson v. Celebrezze (1983, 788-89).

unlikely, however, that the Court will change its mind from the ruling in Timmons,
and take steps to counteract legislative self-entrenching practices through judicial fiat.
Since the Court first intervened in redistricting in Baker v. Carr (1960), the
redistricting process has become more difficult as those drawing new districts must
adhere to criteria set forth by other Supreme Court decisions. In addition to the one
person, one vote requirement, those conducting the redistricting process must follow
primary and secondary levels of criteria. The primary levels of criteria are that
redistricters must make populations nearly equal and in contiguous districts, as well
as complying with the Voting Rights Act. The secondary areas of concern include
rules about compactness, communities of interest, and city and county boundaries
(Cain et al, 2006). To be sure, these rules make redistricting a difficult process, as
those drawing the lines must take these criteria into consideration while still
balancing party interests. But the line-drawing process, while difficult, does not
prevent the creation of competitive districts; as two authorities on redistricting write,
Neither the concept of equal population nor vote dilution prevented incumbents and
political parties from redistricting to their mutual benefit (Cain and Mann, 2005. p.
16). As those in charge of redistricting (major party incumbents with a self interest in
staying in office) redraw lines, they will face little Court intervention in overturning
gerrymandered districts (due to the Timmons ruling). With this fact in mind,
incumbents will work to create districts that benefit themselves, while adhering to

other criteria (of which competitiveness is not a part). This process can be seen with
current districts all over the nation.
What Congressional Elections Look Like
Competitiveness in congressional elections is at an all-time low. Coinciding
with this low level of competitiveness is an extremely low voter turnout in these
elections (see figure 4, page 18). Presidential turnout rates actually seem extremely
high when compared to those in many local races. As Stephen Hill writes, For the
first time, we have been seeing an increase in single-digit voter turnout levels all
across the nation. In numerous other cities and states, turnout for local, state, and
even congressional elections has fallen into the teens and twenties (Hill, 2002, p.vii).
These statistics are amazing when one could make the argument that the result of
local elections has a far greater impact on peoples lives than that of the Presidential

Figure 4
U.S. Midterm Election Turnout, 1962-2002, Based on House Votes Cast (percentage of voting-age population)
source: Congressional Quarterly Weakly Report and CBS News
Americans are becoming turned off by politics and political races, leading to a
situation in which candidates are elected by small minorities of citizens, who have
disproportionately high income and education compared to the entire voting-age
population. It is my contention that the poor and uneducated (those most dissatisfied
with the political system)6 are more turned off by noncompetitive races because they
view a lack of competition as another example that the system works against them. In
other words, these disaffected voters are less likely than others to participate in a
noncompetitive election because having already been turned off by a system that
doesnt respond to their needs and wants, they have no reason to vote in an election
6 For reasons as to why those in the lower classes are dissatisfied with the political system see
Patterson, Thomas E. (2002) or Piven, Frances Fox and Cloward, Richard A (2000).

that they may see as a foregone conclusion. The Vanishing Voter Project, which
studied the 2000 election, surveyed nonvoters and attempted to find why they chose
not to vote. The largest group of nonvoters was those deemed apathetic, those who
didnt know much about politics and were very mistrustful of politicians (Patterson,
2002). This group has little trust in the political system for many reasons, but one is
that they see the system as biased against them. This issue will be further explored in
Chapter 3, which summarizes various theories of voter participation; however, the
Vanishing Voter Project identified a new and growing group of nonvotersthose that
are disenchanted. The cumulative effect of political mistrust seems to have created a
large group of people who are just as turned off to politics as the apathetic-voter
Of the 435 U.S. House races in 2004, only 10 seats were decided by less than
5 percentage points, which amounts to 2% of all races (USA Today Editorial Board,
2005). This result is not at all what the Framers had originally intended, as they
expected a 50% turnover rate each election in the House of Representatives. The
Framers viewed the peoples house as an institution that would create new and
innovative approaches and would respond to the will of the people. In fact, one of the
reasons the Senate was created was to keep the peoples house in check from the
wild and hyperresponsivness they felt the House of Representatives would create
(Hill, 2002). How could an institution that was thought to involve such high turnover
produce the result that is seen today? While many factors contribute to the

phenomenon of non-competitiveness in House elections, the way in which districts
are drawn is one of the main reasons for the level of non-competitiveness seen today.
Redistricting has become more and more sophisticated recently as politicians
and hired consultants use computer hardware and software combined with Census,
demographic, and polling data to draw congressional lines (Economist, 2002).
Partisan redistricting has led to a situation that allows the party incumbents to shape
congressional districts so that they benefit greatly. For example in 2000, 99 percent
of incumbents who were running won reelection. In the same year, less than 9
percent of all House seats were won by a competitive margin of less than 10 points
and 78 percent of U.S. House elections were won by more than 20 points.
Redistricting efforts can also be extremely partisan at the state level. For example, in
California, no incumbent lost any of the 173 U.S. House or state legislative races in
the last election. In the state of Florida, incumbent politicians have won all but one of
the past 140 state House elections. In Tennessee, every incumbent won reelection,
with an average margin of victory of 48 percentage points (USA Today Editorial
Board, 2005). A stark difference in the number of competitive U.S. House elections
is apparent even when one compares the elections of 1994 with that of 2004, as can
be seen in figure 5 (Page 21).

Figure 5
Comparison of Competitive Races 1994 to 2004
Fewer Close Races
For Congress '' 'N * * otmoau
r-v r.jTiL'c c1 oiy C(.^5W3 '-St i,1-.;i.r* 1994
nuV. 1996
in t*V 1998
;ik.-,n U. pofcrr.S.igo DC nr. r v MC'C- tiWfC ?3 2000
rA.firn.Ln 2004 4
A Democrat A Republican A Republican
".V' : r.r I 1 V / *-..r f,
Source. Toner (2005)

A Growing Backlash and Efforts at Reform
Many citizens across the nation have become dissatisfied with the current
state of congressional redistricting. Because of the recent result of politically
gerrymandered districts and uncompetitive elections, there has been a quiet but steady
reform movement growing in many states. Multiple states have adopted independent
redistricting commissions instead of leaving the process in the hands of politicians in
state legislatures. Six states (Arizona, Hawaii, Idaho, Montana, New Jersey, and
Washington) have commissions with the final authority in congressional redistricting.
Eleven states have commissions that have authority for drawing state legislative
districts. Commissions are used as a backup if the legislature fails in drawing
congressional districts in Indiana and in drawing state legislative plans in six states.
Two states (Connecticut and Maine) use commissions as an advisory institution as
their plans must be approved by the legislatures before taking effect. In the state of
Iowa, a nonpartisan legislative support staff agency is responsible for drawing state
legislative and congressional districts, although the legislature has the final authority
over such plans (Mann and Cain, 2005).
Most of these commissions have been enacted relatively recently (within the
past 20 years) and have had varying degrees of success. In addition to the recent
adoption of redistricting commissions, there have been ballot initiatives in multiple

states dealing with this very issue. While some of these commission and ballot
initiatives are further detailed in Chapter 4, it is important to note that there is a
growing dissatisfaction across the country with the way in which politicians
gerrymander and draw noncompetitive districts (USA Today Editorial Board, 2005). It
is the contention of this thesis that this dissatisfaction registers itself not only in
reform efforts, but also in decreasing voter turnout in noncompetitive districts.

The study of the steady decline in political participation in the United States
has had a long and spirited history in American political science. As turnout in
American elections has declined, many theories have been put forth in an attempt to
explain why fewer Americans are voting than in past elections, which remains a
glaring paradox for American democracy. After all, more people than ever before
have the right to vote, as minority groups once denied the right to vote, such as
African-Americans, women, non-property owners, and those aged between 18 and 21,
now enjoy the right to vote in elections. In this chapter I will summarize some of the
major voter-turnout theories in political science literature. I have divided these
theories into the following five categories: class theory/structural oppression, legal
and institutional barriers, disenfranchisement, education and socialization, and party-
decay theories. These five categories cover a vast amount of research in the field and 7
7 The Fifteenth Amendment (ratified in 1870) extended the right to vote to former slaves. While the
Amendment was passed in order to prevent the restriction of voting rights based on race, many other
barriers were then instituted during Reconstruction to prevent persons of color from voting. The
Nineteenth Amendment (ratified in 1920) extended the right to vote to women. The Twenty-Fourth
Amendment to the Constitution was ratified in 1964 and prevented states from disenfranchising people
through the use of a poll tax, removing the barriers in the South and elsewhere that prevented blacks
from participating in elections. The final constitutional amendment regulating voting rights was the
Twenty-Sixth Amendment. Ratified in 1971, the Amendment lowered the eligible voting age to 18.

all give good reasons and evidence for why fewer Americans are turning out to vote.
However, these theories are incomplete and inadequate when considering why
turnout is decreasing in America.
Class Theory/Structural Oppression Theories
In this category of voter-turnout theories, the main argument put forth by
researchers is that low turnout in American elections is a result of high social
inequality and active repression by those in the elite classes. This inequality and
repression is mixed with passive-voter cynicism that results from an oppressive
system. Lower-class voters become apathetic because they feel their vote doesnt
make a difference because the elite control the system. The main reason for theories
in this area is the fact that voter turnout differs greatly among economic classes.
Those in the upper classes and those with high levels of education have high voting
rates, often at 70 to 80 percent. This is almost double the turnout rate of those in the
lowest economic classes (Piven and Cloward, 1988).
A recent report by the Task Force on Inequality and American Democracy by
the American Political Science Association details how rising income inequality
negatively affects American democracy (APSA, 2004). The Task Force found that
there are growing gaps in income and wealth among Americans. In addition, the
report found that these disparities in income have grown more sharply in the United

States than in other advanced industrial democracies. This inequality in income and
wealth has resulted in a situation of unequal voices, namely the fact that those in
the upper classes have more influence, are better organized, and receive more
attention than those in the lower classes. As the report states:
As wealth and income have become more concentrated and the flow of money
into elections has grown, campaign contributions give the affluent a means to
express their voice that is unavailable to most citizens. This aggravates
inequalities of political voice (APSA, 2004, p. 9).
The report also finds that the concerns of those with lower and moderate
incomes .. .are systematically less likely to be heard by government
officials.. .Politicians hear most regularly about the concerns of business and the most
affluent (APSA, 2004, p. 11). The APSA report highlights how inequalities in
income can lead to disparities in political voice. Since Americans with higher
incomes receive more attention and influence from the political system, those in the
lower classes are less likely to be heard from, which in turn breeds discontent.
Why this discontent does not manifest itself in the voting booths in America is
still debated. However, in Voice and Equality, a study about how and why different
groups in America participate in the political system, a clearer picture comes to light.
The study examines the amount of time devoted to political activity by various
economic classes. The authors of this study find that there is a large gap between
political activity and religious activity among the lower classes, with more activity
devoted to the latter. However, the authors of this study argue that the reason the

lower classes devote more time to religious activity is more complicated than just
choice or preference. The authors argue that resources play a huge role in this choice,
especially education and income (Verba et al, 1995).
Seen in this light, the economic and political systems are biased against the
lower classes, who have fewer resources and less education than those in the upper
classes. The system is already biased against the lower classes at the outset, and the
process often is a continuous cycle, which is why those in the lower classes often
have lower voting rates. The authors write, The process is in many ways
cumulative. Those with greater resources at the outset collect more as their resource
endowments allow them to move into institutional positions conductive to further
resource acquisition (Verba et al, 1995, p. 514). Therefore it is this accumulation of
resources that leads to activists being more involved in the political system, but
these authors argue they are unrepresentative of the majority of Americans: Because
participants are not representativein their policy preferences and, especially, in
their demographic characteristics and political needsactivists communicate,
implicitly and explicitly, a distorted set of messages to public officials (Verba et al,
1995, p. 464). This distorted message is biased against the lower classes, which in
turn leaves them out of the political and policy process. Being left out of the political
system then breeds discontent among the lower classes and leads to voter cynicism,
which is why this group does not turn out to vote in the same percentage as the upper

Inadequacy of Class Theories
While class theories certainly help explain low voter turnout in America,
especially among the poor, they are inadequate. By many economic measures, the
standard of living in America has increased over time and is higher than in most other
countries. If people are better off now than they were in the past, why would turnout
rates be decreasing if only economic factors and class differences accounted for
declining turnout? In addition, while the majority of those in the lower classes may
choose not to turn out to vote because have deep cynicism toward a system that is
biased against them, why wouldnt they ever break out of this mold and organize to
completely change a system built by the elite? If class theories were the sole reason
for why fewer Americans vote each year, then why do groups in other classes also
choose not to vote? Class theories fall short in answering these questions, which is
why by themselves they are inadequate in explaining why voter turnout is low in
America. 8
8 The UN Human Development Index is a comparative measure of poverty, literacy, childbirth and
other factors for all countries worldwide. Each year the UN releases a new report, and the United
States consistently ranks in the top 10. See UN Human Development Index (2005).

Legal/Institutional Barriers
In order to understand low turnout in American elections, it is important to
examine the system by which elections are held and votes are counted. Many
scholars point to unique aspects of the American system for reasons that can help
explain low voter turnout. In this chapter, various theories regarding factors such as
Americas voter registration system, residency and identification requirements,
frequency of elections, and the day on which elections are held are all explored.
Winner-Take-All System
Some researchers, such as Steven Hill, argue that low turnout in American
elections is the product of the winner-take-all system. One reason for why the
winner-take-all system depresses turnout is that this type of system, in Hills view,
alienates voters on the losing end of the election. That is, voters who cast ballots for
a losing candidate are not represented, as the winner-take-all system is based on
geographic representation. This system leads to polarization, which turns voters off
to politics:
This escalating combination of nagging national division and dispirited
political depression is particularly perilous because each are mutually
reinforcing of the other. As most players (i.e., voters) abandon the field in
frustration, the game is left to be played by increasingly partisan careerists and
professionals, and by the most zealous activists who seize center stage, further
polarizing politics and policy. And as politics becomes more polarized,

negative, and downright nasty, more and more people turn off and tune out
(Hill, 2002, p. xiii).
The winner-take-all system also leads to a two-party system, rather than a
multi-party system with proportional representation, which also depresses turnout
because the two major parties are not able to fully represent American voters. Under
this system, the Republicans and Democrats work to discourage the ability of third-
parties to participate in the system, one that is already biased against them, which
further alienates voters (Hill, 2002). While there is certainly more research in this
area, it is important to note that some theorists argue that the overall system works to
lower voter turnout.
Scheduling of Elections
Some studies have shown that the scheduling of elections on weekends or
holidays rather than on weekdays can increase turnout. In addition, mail voting can
increase elections by making it easier for people to choose when to vote, rather than
having a set time frame during the workweek in which they must go to a specific
location to vote. Studies have shown that weekend voting can increase turnout by as
much as 5 to 6 percentage points and that mail ballots can increase turnout by another
4 percent (Franklin, 1996).

Many studies have shown that the frequency of elections has a negative
influence on voter turnout. Boyd (1981, 1986, 1989) has shown that the fact that
Americans are asked to vote more frequently than citizens of other advanced
democracies. The only other democracy that votes as often as the United States is
Switzerland, which has comparable voter turnout rates (Jackman and Miller, 1995).
Voters are asked to go the polls for primary elections, general elections, and in some
cases runoff elections. When voters are asked to come to the polls for an average of
two to three times per year, voter fatigue occurs. Voter fatigue means that an increase
in elections increases the costs of voting, thus leading people to participate only in the
election they view to have the most importance, which is often the general
presidential election (Lijphart, 1997).
Registration Rules
A 2004 report by the U.S. Commission on Civil Rights found that registration
deadlines varied greatly among states. The Commission reported that in 2000, 44
states required early registration, ranging from a month before the election (25 states)
to 10 to 28 days before an election (19 states). In addition to registration
requirements, the states also have various ID requirements. The same study by the
Civil Rights Commission found that at the time of the 2000 election, 14 states
required some form of identification, while 7 states reserved the right to require

identification (U.S. Commission on Civil Rights, 2004). States still retain the
authority to choose whether to require identification and what types of identification
are acceptable. The Help America Vote Act (HAVA) of 2002 established minimum
requirements for identification only for those who register by mail or absentee ballot.
Many studies have pointed to the fact that registration rules significantly
affect voter turnout. Studies by Wolfinger and Rosenstone (1980, 1990, 1993) found
that voter turnout could be increased if all states adopted completely liberalized
registration rules. Most studies show that with more liberalized voter registration
rules, turnout could be increased between 8 and 15 percentage points (Lijphart, 1996).
These liberalized registration rules would eliminate registration requirements in
which voters must register a certain number of days before an election to be eligible
to vote.
Frances Fox Piven and Richard Cloward are two scholars who argue that
American institutional arrangements provide the best explanation for low voter
turnout in the United States, specifically the process of voter registration. They point
out that the lower and working classes are the most underrepresented in the American
electorate. They write, .. .the active American electorate over-represents those who
have more, and under-represents those who have less. Piven and Cloward argue that
this fact is not an accident, but the product of elites rigging the voting system to
benefit those who are better off. The key to this concerted effort of elites to dominate
politics and effectively disenfranchise the lower classes occurs with Americas unique

voter-registration system. The authors explain how Americas system differs from
that of other major world democracies as follows:
The major difference is that governments elsewhere assume an affirmative
obligation to register citizens. People are certified as automatically eligible to
vote when they come of age and obtain identity cards, or government-
sponsored canvassers go from door to door before each election to enlist
voters. The United States is the only major democracy where government
assumes no responsibility for helping citizens cope with voter registration
procedures. In 1980, 39 to 40 percent of the American electorate was
unregistered, or more than 60 million in an eligible voting-age population of
159 million, and two out of three of unregistered resided in households with
incomes below the median. Furthermore, the significance of low registration
is suggested by the fact that once people are registered, they overwhelmingly
vote. In 1980, more than 80 percent of registrants went to the polls, and the
turnout among those with little education and income was only marginally
lower (Piven and Cloward, 1988, p.17-18).
In Piven and Clowards view, the fact that the United States government
leaves the responsibility to register with its citizens creates a situation in which far
fewer people register to vote. Since it is shown that those who are registered will
vote in high percentages, the authors argue the American electoral system is largely to
blame for low electoral turnout. With elected officials in charge of voter-registration
procedures, which often vary from state to state, Piven and Cloward argue further that
officials will manipulate the system to benefit themselves and keep turnout low.
They also argue that the limited electorate helps explain why the Democratic Party
has turned away from its base and does not work to mobilize the poor and working
classes: A party constructed on a restricted electorate, and incumbent politicians
elected by a restricted electorate, risked serious disturbances to leadership, to funding

sources, and to existing consistencies were it to turn seriously to mobilizing the have-
nots (Piven and Cloward, 1988, p.22).
Piven and Cloward then attempt to explain some of the key systemic factors
that have led to low voter turnout. Many scholars point to the critical election of
1896, when the Populists were soundly defeated by the Republicans and an era of
demobilization and a decline in party competition followed. However, Piven and
Cloward look beyond this election and argue that the reconstruction of the
institutional arrangements of electoral politics occurred in the late 19th and early 20th
centuries. As elections were decided by internal party choices in much of America,
the average voter mattered less, while economic elites increased their influence on the
electoral process (Piven and Cloward, 1988). The result of the 1896 election was that
demobilization became institutionalized by the elite to prevent the poor, minorities,
and populists from creating what the elite viewed as disorder. As a result, legal and
procedural disenfranchisement occurred, with the South being the most visible
example as literacy tests, poll taxes, and other barriers were erected to prevent blacks
and poor whites from voting. Piven and Cloward argue that more than just the results
of the 1896 election are to blame for a decline in turnout, and rather, The several
transformations in the institutional framework that governed electoral participation
a decline in party competition, weakened and reorganized parties, and new rules
restricting access to the ballotworked together to depress turnout (Piven and
Cloward, 1988, p.95).

After new-voter registration requirements were instituted, parties became
more and more dependent on them to win elections:
As voter registration restrictions took effect little by little, the linkages that
bound the parties to their working- and lower class constituencies withered
.. .the parties themselves became the defenders of the voter registration
procedures that ensured their stability and protected incumbents.. .As
connections between the parties and the electorate were reconstituted so that
working-class groups were far less important, party appeals changed (Piven
and Cloward, 1988, p. 110).
The downward trend in electoral turnout was partly reversed during the New
Deal era, and Piven and Cloward argue that popular movements forced the Roosevelt
administration to reach out to the lower classes. However, this upward trend in
political participation occurred for only a short time, and parties once again turned
away from mobilizing the lower classes as an era of dealignment occurred after the
1930s. The authors also argue that American electoral arrangements offer a
significant explanation as to why the United States followed such a different political
path from Western Europe where unions and social welfare policies are much
stronger. Piven and Clowards focus on voter-registration procedures and American
electoral arrangements do help explain why turnout is low compared to other world
democracies. Certainly institutional arrangements and their striking differences from
other world democracies offer a major explanation of low turnout. While many states
have made it easier to vote since the publication of Piven and Clowards book in 1988
with the institution of motor-voter laws, mail-in ballots, and same-day registration,

multiple obstacles to registration exist and the poor remain disproportionately
affected. For example, social-agency-based voter-registration programs (welfare
offices, social security, etc.) are not as available as motor-voter programs (Piven and
Cloward, 2000).
Inadequacy of Legal/Institutional Barrier Theories
While factors such as registration procedures, frequency of elections, and
other basic explanations of voting procedures in America certainly account for why
American turnout is often lower than in other democracies, this category alone cannot
fully account for a decline in turnout. The main reason for this categorys inadequacy
is the paradox of American turnoutnamely the fact that more people than ever have
the ability to participate in elections as the barriers to voting have come down. Also,
some studies (Mitchell and Wlezien, 1995) have found that liberalized voter
registration laws would have little effect on turnout, while others have found that the
effects would not increase turnout equally across educational levels (Knack and
White, 2000). Furthermore, the frequency of elections in America has remained
relatively constant over time, as people voted in primary elections in the 1960s just as
they do today. Also, it is easier to register to vote now than in the past, especially
with the increase of motor voter laws, early voting, and absentee balloting.

Therefore, while these theories provide a window into why America has a lower
electoral turnout than other advanced democracies, they remain inadequate.
While many groups that were previously discriminated against have
ultimately gained the right to vote, other groups such as the insane, immigrants,
children under 18 years of age, and convicted felons (in many states) are prevented
from voting. Theories in this category argue that a main reason for declining turnout
is due to the disenfranchisement of various groups. Some of the main theorists in this
category are Michael McDonald and Samuel Popkin, who argue in The Myth of the
Vanishing Voter, that a decline in VAP voting rates is the result of an increase in the
number of ineligible voters, rather than nonvoting (McDonald and Popkin, 2001). In
this section I focus on felon disenfranchisement, mainly because there has recently
been new research in this area. Although I focus on this group, it is important to note
that other groups also are prevented from voting and that non-citizens account for
around seven percent of the adult population (Patterson, 2002).
One important factor that can help explain low voter turnout in the United
States is its unique system of disenfranchisement of felons9 and ex-felons. It is
9 The definition of the term felony is important to note for the purposes of this paper. Two leading
scholars on felon disenfranchisement define the term the following way: Felony is a generic term
used to distinguish more serious crimes from lesser crimes, known as misdemeanors. Statutes
typically define felonies as those offenses punishable by one year or more of incarceration in a

estimated that at the time of the 2000 election the disenfranchised felon population
consisted of between 4.1 and 4.7 million Americans, roughly 2.3 percent of the voting
age population (Manza and Uggen, 2004). Disenfranchising people who commit
crimes has a long history in English, European, and even Roman law. Therefore, it is
not surprising that American law also contained many of the same laws to
disenfranchise felons. The rationale for this disenfranchisement was that it served as
retribution for committing a crime and served as a deterrent for future antisocial
behavior. After the Civil War in America, another rationale was added stating that
former criminals would corrupt the electoral process. Others also had the fear that
ex-felons might band together and vote to repeal criminal laws. However, during the
Reconstruction era, many of these disenfranchisement laws were aimed at lesser
offenses, and specifically targeted African-Americans (Keyssar, 2000).
The path for felons and former felons of gaining and losing voting rights in
American history has changed during different eras. Two important scholars on felon
disenfranchisement, Jeff Manza and Christopher Uggen, point to three different and
distinct waves in American history when many states expanded their restrictions on
felon and ex-felon voting rights. The first main period or wave occurred after the
decline of property restrictions on white male suffrage. During this period, many
states acted to disenfranchise felons, as the poor could now vote, feeding elite fears of
mobbism and pressures for social change. The second major wave occurred after
penitentiary. In practice, however, many states count minor as well as the most serious offenses as
felonies (Manza and Uggen, 2004, p. 502).

the Civil War in the South during and after Reconstruction (the first time large
proportions of African-Americans could vote). The third wave in felon
disenfranchisement laws in American history occurred in the 1960s and 1970s, and
it worked to overturn Jim Crow practices in the South and to mobilize a range of new
voters, ranging from the young, to the poor, to the non-white. During this era, many
states liberalized their felon-disenfranchisement laws.
The weakness of historical rationales for denying voting rights to ex-felons in
order to prevent future crime and purify the ballot box was one factor in the
reconfiguration of state laws. Another factor was a new emphasis on rehabilitation.
These two factors led more than 15 states to change their laws so that the
disenfranchisement of ex-felons was not permanent, and many more narrowed the
range of crimes that could result in disenfranchisement. However, racial politics still
played an important role in this period, since states with high a proportion of African-
American prisoners (such as in the South) were less likely to liberalize their laws.
Although the Supreme Court has upheld the practice of felon
disenfranchisement, there has been no effort in America so far to develop a national
standard, thus leaving the issue to the states. With an absence of national standards,
there are a wide variety of state laws regarding voting rights for felons and ex-felons.
A majority of states have laws that restrict the voting rights of not only those serving
in prison, but also those on parole and/or probation, and some states even
disenfranchise ex-felons who have served their entire sentence.

There has been an unprecedented growth in the felon population in the United
States over the past three decades, which has led to an even greater amount of felon
disenfranchisement. The following is a list of the four different categories that states
use when issuing disenfranchisement laws: convicted felons who are currently
incarcerated (14 states), previously convicted felons who have been released from
prison and are under parole (16 states), felons sentenced to parole rather than prison
(4 states), and ex-felons who have completed their entire sentence (14 states). Only
two states, Maine and Vermont, currently allow felons to vote, including those in
prison (Manza and Uggen, 2004).
Manza and Uggen attempt to determine whether felon disenfranchisement
would have had any impact on past elections. In their article Punishment and
Democracy: Disenfranchisement of Nonincarcerated Felons in the United States,
Manza and Uggen admit that this is a difficult issue to draw conclusions from, largely
because felons and ex-felons are not participating in elections and one cannot truly
know how or if they would vote. In addition there is a lack of existing data
concerning political behavior and criminal history. Despite these constraints, the
authors solution is to match the characteristics of the felon and ex-felon population
with the rest of the electorate using existing data from the national election survey.
Past surveys show that this group of disenfranchised voters has low levels of
education, low incomes, and high unemployment rates, and are disproportionately
African-American. Drawing from this portrait of felons and ex-felons and comparing

them to others in the general population, the authors show how it is probable that this
group would be less likely to vote than the rest of the general public and also that they
would have a relatively high level of Democratic partisanship.
The authors then focus on whether the disenfranchisement would have
changed the outcomes of any recent elections. The authors found evidence that the
loss of voting rights for ex-felons influenced electoral outcomes in seven Senate
elections and two presidential elections, including the presidential race of 2000. The
authors choose to focus only on ex-felons who have served their sentence or are on
parole because that is the most controversial type of restriction considering American
public opinion.
While national statistics on race and felon disenfranchisement may seem
striking, they are even more dramatic when one examines those of individual states.
For example, in Alabama and Florida, 31 percent of all black men are permanently
disenfranchised. In five states (Iowa, Mississippi, New Mexico, Virginia, and
Wyoming) one in four black men is permanently disenfranchised. In Delaware and
Texas, approximately one in five black men is currently disenfranchised. In 13 other
states, between 10 and 18 percent of black men are currently disenfranchised.
According to a recent Human Rights Watch report, if current rates of incarceration
continue, 28.5 percent of black men will be confined to prison at least once during
their lifetime, six times greater than that for white men. If this estimation is correct,

the majority of these prisoners will be disenfranchised for part or all of their lives
(Human Rights Watch, 2004).
The practice of felon disenfranchisement contributes to low turnout in
America. If an average of 2 to 3 percent of the voting-age population is not allowed
even to cast a ballot in each election, and may further be prevented and discouraged
from participating in future elections, turnout will be lower than in a country that
gives universal voting rights or has fewer restrictions on felons voting.10 With
incarceration rates rising, it is logical to assume that even more will be
disenfranchised in the future, leading to an even greater disenfranchised population.
Inadequacy of Disenfranchisement Theories
While McDonald and Popkins study argues that a decrease in VAP voting is
the result of an increase in the ineligible population, and argue that the use of VEP is
a more accurate measure of voting activity in the United States, they fail to take into
account those that are turned off by politics. Other studies such as Putnam (1996)
show that participation rates are falling not just in voting, but in other areas of the
political arena as well. VEP and VAP rates are not very far apart, usually around 5-
7%, so disenfranchisement does not account for the fact that a substantial amount of
10 The countries of Australia, Austria, Italy and Belgium have limited voting restrictions on felons, and
Sweden has no restrictions. Each of these countries has a far higher voter turnout rate than the United
States, both overall and among registered voters (Piven and Cloward, 2000, p. 190).

Americans choose not to vote in elections. Furthermore, McDonald and Popkins
study does not explain why turnout has not risen with education levels or with more
liberalized voter registration laws.
While theories of disenfranchisement help explain a decrease in turnout, alone
they are inadequate. One reason for the inadequacy, as noted earlier, is that more
groups than ever before now enjoy the right to vote in America, yet turnout has not
increased. The eligible voting population has actually grown over time, not
decreased. Also, giving currently ineligible groups such as felons the right to vote
would not in itself greatly increase turnout, as those in currently disenfranchised
groups contain lower voting rates than the rest of the eligible American electorate.
Furthermore, the disenfranchisement theory can account only for a minority of the
adult population that does not vote, and cannot by itself account for the overall
decline in voting rates for American elections.
Educational/Socialization Theories
Some theories argue that the educational and socialization process in America
leads to lower turnout. Theorists in this category point to aspects of political
socialization in the United States that can depress the turnout of the voting-age
population. Those in this category argue that children are brought up as non-voters,
and are not shown that voting or political participation is important because they cant

participate until the age of 18. Other theories in this category argue that the media are
a major reason why people are turned off to politics and therefore do not vote.
Preparing Children to be Nonvoters
Professor Michael S. Cummings is one who argues that greater empowerment
of children is one way to increase voter turnout. Cummings argues that although the
26th Amendment of the Constitution grants the right to vote to all citizens over 18
years of age, this criterion is arbitrary and inconsistent in U.S. law as citizens voting
rights in a democracy rely on the equal consent of the governed, or the idea that those
affected by policy have the right to choose their representatives and hold them
accountable for the policies enacted. Therefore, according to this theory, the right to
vote is not based on privilege or knowledge, but rather, .. .is based on a right to
participate in a political process that affects ones well-being (Cummings, 2001, p.
If voting rights are granted so that citizens may have a say in the political
process that affects them, one might wonder why there are groups within our society
that are denied the right to vote. Cummings argues that children should be granted
voting rights and should not be excluded from the process for two main reasons,
First, many young people are more politically virtuousmore knowledgeable,
rational, responsible, experienced, selfless, and intelligentthan many older people.

Second, even if they were not, the democratic right to vote is not based on ones
virtue but on ones being affected by public policy (Cummings, 2001, p.l 88).
Giving children the right to vote would change the political landscape drastically,
especially if children were brought into the political process at a younger age. Instead
of preparing children for non-voting by disenfranchising them until the age of 18, the
elimination of age restrictions to voting would socialize children as voters and
participants in the political system. The enfranchisement of children would thus lead
parents and families to be more politically active, as people would be introduced to
and encouraged to participate in politics at an earlier age.
Some organizations have been set up to attempt to increase democratization
among Americas youth. Kids Voting USA has established mechanisms by which
children cast non-binding votes on Election Day. In the election of 2004, more than
1.5 million students voted for President and in local races around the nation. The
organization was started in 1988 and now has 28 educational programs that
accompany voting in over 16,000 voter precincts in forty states (Kids Voting USA,
2006). Kids Voting USA was inspired by a similar system in Costa Rica, which
counts non-binding youth votes separately from adult votes. In Costa Rica, the
process of children voting appears to have increased adult turnout between 5 and 10
percent. Moreover, it also reduces the socioeconomic status gap in political
awareness and participation, eliminates the traditional gender gap, and doubles
newspaper reading among its youthful participants (Cummings, 2001, p. 190). If

children are involved earlier in the voting process and can understand the benefits of
political participation at an early age, they are more likely to continue voting into
their adult lives. Therefore, the exclusion of children from the voting process
contributes to a low adult turnout. Also, just as with the felon disenfranchisement
theory, young people would likely have low voting rates, which would not
immediately increase voter turnout.
Media Theories
Some argue that media coverage of politics actually turns people off to
politics, and thus socializes citizens to abstain from voting. Thomas E. Patterson is
one of the main scholars in this area. Patterson argues that the media have become
highly negative in their coverage of candidates and political issues (Patterson, 2002).
In addition to the growing negativity, the television media coverage of candidates has
changed drastically from the 1960s as Patterson writes:
In 1968, the average sound bitea block of uninterrupted speech by a
candidate on television newswas more than 40 seconds. By 1988, the
average had shrunk to less than 10 seconds. For every minute Bush and Gore
spoke on the evening newscasts during the 2000 campaign, the journalists
covering them spoke for 6 minutes. The two candidates received only 12
percent of the election coverage.. .A viewer who watched the network news
every evening between Labor Day and Election Day would have heard 17
minutes each from Bush and Gore, or about 15 seconds a night. When Bush
appeared on CBSs David Letterman show on October 19, he received nearly
as much airtime as he did on the CBS Evening News during the entire general
election (Patterson, 2002, p.68).

Not only do the media cover political campaigns in a negative and abbreviated
way, but they also cover current political events in much the same way. Both
television and print media have trended toward covering the sensational, which is
often a negative news story. This trend in negative news coverage has led to public
mistrust of government as Patterson writes:
Coverage of national politics was so downbeat that it distorted Americans
sense of reality. A 1996 survey asked respondents whether the trend in
inflation, unemployment, crime, and the federal budget deficit had been
upward or downward during the past five years. There had been substantial
improvement in all of these problem areas, but two-thirds of the respondents
said in each case that things had gotten worse.. .Public mistrust and
dissatisfaction accompanied the flow of negative news. In 1964, 76 percent
had said they trusted the national government to do the right thing most of
the time; in 1994, only 21 percent held this opinion, the lowest level ever
recorded (Patterson, 2002, p.74).
Patterson argues that recent trends in media coverage, as well as an increase in
voter apathy are linked together. Through better media coverage, especially with a
more positive slant on politics, Patterson believes that citizens would be less turned
off to politics, and would therefore vote more often.
Inadequacy of Educational/Socialization Theories
While educational and socialization theories do provide evidence for why
turnout in America is declining, they are also inadequate as a category of theories.

One reason for this inadequacy is that educational levels have risen over time. As
Robert Putnam writes, Since 1972, the proportion of adults with fewer than 12 years
of education has been cut in half.. .while the proportion with more than 12 years has
nearly doubled (Putnam, 1996, p. 3). Furthermore, while media coverage has
become increasingly negative, a change in coverage might not get back the declining
number of people who use the major media outlets to get their political news. Even
Patterson points out that the major media outlets are losing viewers (Patterson, 2002).
More positive or in-depth coverage does not guarantee more political participation,
especially with the proliferation of Internet sources and more ways to find
information than ever. For these reasons, this category of theories remains inadequate
at explaining a decline in voter turnout in America.
Party-Decay Theories
Many scholars point out the fact that political parties significantly contribute
to mobilization efforts come election time. Major political parties have been heavily
involved in recruiting new voters and getting out the vote in the past, and therefore
the theory that party activities affect turnout is formidable. Two scholars who study
political parties are William Crotty and Thomas ONeill. In their paper Democracy
and the Future of Political Parties in America, the authors explain why political
parties are vital to a democracy. Parties, according to the authors, provide the

connecting link between the mass electorate, and their needs and views, and those in
power. Parties are also the political systems means of organizing the mass of
citizens into expansive coalitions along policy and regional agendas. The authors
also explain that political parties contribute to legitimization of a political system
while linking mass and elite decision-making. A competitive and vibrant democracy
is not plausible if parties are not able to effectively mobilize the masses.
If political parties are important to mobilization efforts, it is important to
know how strong parties are and how effective they are at getting out the vote. Many
scholars argue that parties are experiencing a diminishing role in America, and
therefore are partly to blame for low voter turnout. Scholars such as Wattenberg and
Klingeman (1992) have argued that parties are in decline and have decreasing control
over the American electorate. These scholars focus on the rise in the independence of
voters and party decomposition. Other studies point to the rise in independent voters
and less party identification. There have recently been a decline in party-line voting
and an increase in split-ticket voting (Crotty and ONeil, 2002). Other scholars
disagree and argue that parties remain extremely important in American politics and
have not decayed at all. For example, Miller (1994) argues that the tumultuous events
associated with the late 1960s and 1970s (Vietnam, civil rights, political
assassinations, demonstrations, etc.) had a huge impact on the young and had a lasting
impact on their party intensity and identification. In effect, Miller argues that young
voters in this era were forever disillusioned with the major parties, and due to actions

by each party, young people found it difficult to always support one party over
another and deeply invest in either of the major parties. While there is a strong
debate in political science concerning the relevance of political parties, several
conclusions can be drawn. Crotty and ONeil write:
.. .the current parties are less significant than they were at the height of the
stable-party period in the 1950s; their impact on voter decision-making is
more in question today than in earlier times; political candidates run now on
their own initiative; the parties contribution beyond supplying links to
fundraisers, consultants and pollsters is weaker, and that the future of the
political parties in these regards is speculative and uncertain (Crotty and
ONeil, 2002, p.6).
Since parties are weaker than they were in the past and voter mobilization is
one of their main goals, perhaps the link between the two can help explain why
turnout has gone down. The two authors investigate the effect that party decline has
had on recent mobilization efforts and write:
Voter mobilization and the extension of the bounds of participation in the
political system are functions of the individual parties and the party system
more generally. Many would contend it is one of their major responsibilities.
The evidence is they are doing a poor job. The parties impact on the voting
electorate is arguable, but in terms of expanding the voting pool to include the
interests of the less well-off and those without other political resources, they
have not done well.. .The result, in effect, is a middle class electorate. Its
interests are the ones both parties choose to address in campaigns (Crotty and
ONeil, 2002, p.10).
While the debate over party decline continues, many argue that the major
parties have focused more on the middle class and have not worked as hard as in the
past at bringing the poor and working class into the voting booths. Others argue that

the major parties have tailored policy toward the upper class, rather than the middle
class. Either way, the two major parties have not done a good job bringing the lower
classes into the electorate and have not enacted policies that benefit this class. Party
decline, or at the very least, party apathy, in mobilization efforts has probably
contributed to a drop in voter turnout. Some argue that parties have also worked to
depress turnout in primary elections, making it easier for party loyalists to exercise
control. In three states (Alaska, California, and Washington), both major parties
worked to eliminate the blanket party system in which candidates for both parties are
on the ballot and each voter is allowed to cast a ballot for any candidate, regardless of
party affiliation. This type of primary created centrist candidates and higher turnout
until the parties eliminated the process (Harvey, 2004).
Inadequacy of Party-Decay Theories
While party-decay theories may show that the major political parties do not
have the same influence they once did, the fact remains that the Republican and
Democratic Parties still dominate the American political landscape. The major
political parties still rely on voter turnout as a way to win elections. In addition,
while many people may identify as political independents, they may still have strong
party preferences and may not be completely turned off from voting. Both political
parties still court independent voters and have not abandoned them during elections.

Americas Voter-Turnout Puzzle
In conclusion, the puzzle of voter turnout in America remains difficult to
solve. Obviously, a combination of theories helps to explain the decrease in voter
turnout in American democracy. Certainly, class theories help explain why America
has a lower electoral turnout than other nations, as the gap between rich and poor has
increased recently. Since the poor often have lower voting rates than other groups of
Americans, class theories certainly have a role to play when explaining voter turnout
in America. In addition to class theories, there are still institutional barriers that
prevent people in America from voting. While registration procedures have become
more readily available to many Americans, the United States still differs from other
democracies that have universal registration. The scheduling and frequency of
elections can also affect turnout, but only to a certain extent. In addition, Americas
unique system of felon disenfranchisement also depresses turnout. Clearly, many
factors of the American socialization process legitimize the depression of turnout,
with children being excluded from the democratic process and brought up as
nonvoters. Finally, party-decay theories also help explain a decrease in voter turnout.
Parties do not have as much influence over voters as they once did, as more
Americans do not identify with either major party. Parties have also put less of an
emphasis on voter mobilization, leaving the citizens responsible to become involved.

While many of these theories help explain the multitude of factors that affect
voter turnout in America, those presented in this chapter by no means cover all
authors or theories. While these voter-turnout theories have been widely studied in
American political science, they are by themselves incomplete. Americas process of
congressional redistricting also contributes to low voter turnout, yet has not received
as much attention as other theories. In the next chapter, the question of whether
congressional redistricting contributes to voter decline is examined and placed in the
field of research concerning voter turnout in the United States.

This chapters purpose is to look at some of the existing research in the area of
redistricting and voter turnout and explore whether there is any evidence that the
process of redistricting in the United States affects voter turnout. Some of the recent
literature in this area argues that redistricting does not affect competitiveness, but I
take issue with these studies and will give examples in which the redistricting process
has in fact had a significant effect on both competitiveness and voter turnout. It is
both the criteria that the Court has imposed on redistricters and partisan and bi-
partisan gerrymandering attempts that contribute to noncompetitive elections.
My main argument in this chapter is that a competitive election increases
turnout because voters are more likely to feel that their vote counts and that they have
a say in the outcome of the election. If people feel that the action of voting in an
election will have an impact, they are more likely to follow and care about that
election. Competitive elections are more interesting to follow and receive more
media attention, and large amounts of money are poured into them as they could
determine the balance of power in Congress. Studies have shown increased

participation in competitive elections, especially in congressional races, as the
closeness of an election has far-reaching effects on candidates, the media, and voters.
In close congressional races, candidates work to mobilize voters and are more
specific about issue positions. Also, the media devote more attention to competitive
races as opposed to safe districts. Voters are also better able to recall and recognize
candidate names in close congressional races as opposed to noncompetitive ones
(Macedo et al, 2005). In addition, a study by Gilliam (1985) found that candidate
competition was one of the most important variables that predicted voter turnout for
U.S. House elections during non-Presidential years. As my research will also show,
when there is a competitive election, more people are likely to participate in that
election as opposed to one in which the result is viewed as predetermined.
The Impact of Redistricting on Competitiveness
As has been mentioned multiple times throughout this thesis, competition in
congressional races has reached an all time low, with a 99% retention rate in the
election of 2000 (USA Today, 2005). In fact, most state legislative elections in
American are also not competitive at all. In many elections across the country,
candidates run unopposed. In fact, 41% of all state legislative races were uncontested
by one of the two major parties in 2000 (Hill, 2002). A decrease in the number of
competitive congressional elections can be seen in Table 1. As overall midterm

election turnout in the United States has decreased to some of its lowest levels in
history (see Figure 9, page 68), the percentage of competitive districts has also fallen.
As can be seen from Table 1, there has been an increase in the percentage of safe or
low-risk districts since the redistricting of the 1970s. Incumbents of both parties
reside in a higher percentage of safe districts with less competition than ever
before. 11
Table 1
Percentage of Democratic and Republican Representatives in High-Risk and Low-
Risk Districts by Decade11
1972-1980 1982-1990 1992-2000 2002-2004
Low-Risk 23.2 31.5 35 40.8
High-Risk 21.5 16.1 16.4 11.3
Low-Risk 24.5 24.5 39.5 51.1
High-Risk 41.4 35.4 25.5 16.4
Source: Abramowitz et al (2004)
11 Note: Low-risk districts are those in which share of major party vote for presidential candidate of
incumbents party was at least 10 percentage points greater than national vote share. High-risk
districts are those in which share of major party vote for presidential candidate of incumbents party
was less than the national vote share.

However, the question remains: is the redistricting process responsible for this
outcome or do other factors cause the lack of competition? One recent study
attempted to test this very hypothesis and looked at three different causes:
redistricting, partisan polarization, and the incumbency advantage. The study
(Abramowitz et al, 2005) found no support for the hypothesis that redistricting has
created a decline in competitive races. The study instead found that the incumbency
advantage and partisan polarization hypotheses accounted for why fewer
congressional races have become competitive. However, I believe the results of the
test are misleading. The redistricting hypotheses states:
If the redistricting hypothesis is correct, then we should observe a substantial
increase in the number of safe districts and a substantial decrease in the
number of marginal districts in three elections that immediately followed each
recent redistricting cycle: 1982, 1992, and 2002 (Abramowitz et al, 2005).
While the test does show that there is not a substantial increase in the number
of safe districts immediately after each redistricting cycle, the study fails to take into
account the fact that there is a significant decline in competitive districts after each
decade. Therefore, as redistricters draw new lines, they get better with each new
round of redistricting, an effect that can be seen in Table 1. As the researchers
themselves point out, in 2004 172 winning candidates had either no major party
opposition or won by a margin of victory of at least 40 percentage points
(Abramowitz et al, 2005). This outcome was not all a result of incumbency
advantage or an increase in voter polarization. While these factors do contribute to a

high incumbent re-election rate and should not be underestimated, it is misleading to
dismiss the redistricting process as having no effect on competition. The redistricting
process often contributes to the incumbency advantage and studies such as Lyons and
Galderisi (1995) found that incumbents do benefit from party-controlled redistricting.
This study also found that incumbents benefit more under bipartisan than partisan
One reason that redistricters are getting better at drawing noncompetitive
districts is the advance in redistricting technology. Computers were first used in
redistricting in the 1960s in only a few states. In the 1970s, again in limited use,
redistricting software was used by at least three state legislatures. Even in 1981, only
a few states used redistricting computer systems. The use of computers in the
redistricting process did not become widespread until the 1991 round of redistricting,
but access to these programs was still limited. During this period, redistricting
software was still expensive, required high-end computers, and needed technical
support. However, by 2001, GIS software was widely available and prices dropped
dramatically. In 2000, all states with the exception of Michigan reported using
redistricting software (Cain and Mann, 2005). Not only did states use redistricting
software, but political parties have bought software and are easily able to use it to
their advantage. Current redistricting software can be used on a laptop computer and
is relatively inexpensive (less than $5,000). One example is of a Texas watchdog
group called Texans Against Gerrymandering, which bought ten copies of

redistricting software (Hill, 2002). Not surprisingly, this group includes former
Representative Tom DeLay, who oversaw the mid-census gerrymander in 2003 that
resulted in Republicans gaining six seats (Cain and Mann, 2005). Technology
currently being used to redistrict is certainly advanced. One example is given by
Jeffrey Toobin, a staff writer for the New Yorker magazine when he met with
redistricting expert and assistant professor of law and political science at the
University of Pennsylvania Nathaniel Persily to demonstrate redistricting software on
his laptop. Toobin writes:
When Persily opened his computer, he showed me a map of Houston, detailed
to the last census block. (The population of each block usually ranges from
fewer than a dozen to about a thousand.) This is the same map that DeLays
people used to redistrict, Persily said. Indeed, DeLays political operation
purchased ten copies of the software, which is called Calipers Maptitude for
Redistricting and costs about four thousand dollars per copy. The software
permits mapmakers to analyze an enormous amount of dataparty
registration, voting patterns, ethnic makeup from census data, property-tax
records, roads, railways, old district lines. Theres only one limit to the kind
of information you can use in redistrictingits availability, Persily said. (In
Pennsylvania, Republicans used Camegie-Mellon Universitys mainframe
computer, which would have allowed them to add even more data, such as
real-estate transactions.) With a few clicks, Persily changed the map from one
that showed party registration in each census block to one that revealed voting
results in each block. The colors ranged from dark red, for heavily Democratic
votes, to dark blue, for strongly Republican. He showed voting results in
about two dozen races, from President to governor and from congressman to
local offices. The whole process has got much more sophisticated, Persily
went on. Party-registration data are not the only kind of data you want to use.
You want to use real election results. Thats a big change from ten years ago.
We have become very good at predicting how people are going to vote.
Peoples partisanship is at a thirty-year high. If I know you voted for Gore, I
am better able to predict that you are going to vote for any given Democrat in
a future election (Toobin, 2003).

Furthermore, if redistricting did not affect the recent decline in competition
then it would not be possible to create more competitive districts in any state across
the nation. However, the state of California is one example of a state that contains
many gerrymandered districts that are completely uncompetitive. Competition in the
state of California in congressional races is almost nonexistent. As mentioned earlier,
in California, an incumbent did not lose any of the 173 U.S. House or state legislative
races in the last election. According to a Common Cause study, the state of California
has some of the most grossly gerrymandered districts in the nation, as the study
Our analysis of elections since 1982 covering three redistricting cycles -
shows that competition was suppressed when redistricting was in the hands of
incumbents in the 1980s and 2000s. But after a court-imposed redistricting
in 1991, with lines drawn by three retired judges appointed by the California
Supreme Court, competition rose in both U.S. House and state legislative
races. The numbers tell the story: During the 1990 cycle, when an
independent panel redrew the lines, the number of competitive races increased
by more than 50 percent. During the 2000 cycle, when the legislature drew
the lines, the number of competitive races decreased by more than 55 percent.
In fact, no incumbents lost in either election, and in the 2004 elections, not
one seat in the state legislature changed parties (Common Cause, 2005, p. 1).
A recent study by the Institute of Governmental Studies at the University of
California, Berkeley found that at least 12-14 more competitive districts could be
created in California, while still balancing other of the important criteria mentioned at
the beginning of this chapter (Institute of Governmental Studies, 2005). If
redistricting has no effect on competitiveness, this study shouldnt have found the

possibility of more competitive districts. Not surprisingly, turnout in California
midterm elections hit an all-time low recently, as a lack of competition was one of the
main sources of blame cited by political analysts. In the 2002 midterm election,
California ranked 47th in the nation in turnout (Wildermuth, 2002).
In addition to California, the example of Florida also sticks out. Figures 8, 9
and 10 demonstrate how gerrymandered districts have created a legislative majority
for Republicans, while leaving the Democratic Party with a disproportionately lower
number of seats in state and congressional legislative districts. In Figure 6, the
overall registration by party for the state of Florida shows that Democrats have the
largest percentage of registered voters in the state. In terms of presidential elections,
Florida is usually a swing state, and historical data shows that the state often goes
back and forth between Republicans and Democrats (Leip, 2006).
Figure 6
Florida's 2004 Total Voter Registration by Party
Republican Democrat Other
source: Florida Secretary of State

Figure 7 shows the percentage controlled by each major party in the State
House and Senate, as well as the U.S. Congressional delegation. This outcome is
surprising considering that more Democrats are registered in the state than
Republicans. Especially since Florida has been such a competitive state in the past
two presidential elections, it might be expected that the state legislature and U.S.
congressional representation would be more nearly equal.
Figure 7
Party Composition of Florida Legislative and Congressional Delegations
Democrat Republican
100% -
40% -
20% -
U.S. House State Senate State House
source: Florida Secretary of State
Perhaps Republicans are just better at winning elections in the state of Florida,
and have dominated the state due to the overall success that the Republican Party has
enjoyed nationwide. However, when one examines Figure 8 (page 63), a better
explanation for the dominance of Republicans is brought to light. Floridas
congressional districts show that in many cases, Democrats are packed into a few
districts, where they far outnumber Republicans. In other districts, Republicans
outnumber Democrats by an average of 7.6%. This gerrymandering has resulted in a

sweeping victory for the Republican Party, and has created a situation in which it
dominates the state and congressional legislative representation.
Figure 8
Floridas Congressional Districts and Registration Advantage 2004
Dvmocrats ota rcaisirr
Democratic voters are concentrated
in a few congressional districts.
Republican voters are distributed
across many districts.
source: Research Memo from Committee for Fair Elections and Common Cause.
A final reason why redistricting can lead to less competitive elections is that it
often hurts candidate recruitment. A study by Maisel et al (2005) found that the
appearance of gerrymandering negatively affected potential political candidates
considering whether to run in House elections. In states in which the legislature
controls the redistricting process, potential candidates were more likely to view the
process as benefiting incumbents as opposed to states that had commission systems.
Also, potential candidates who saw their chances of winning as less than 20% were

unlikely to run for office. Potential candidates view the current process of
redistricting (in states without commissions) as noncompetitive, thus discouraging
lower-tier candidates from running for House elections. These lower-tier candidates
contribute to a lack of competition, as they are not able to raise as much money or
have the same name recognition as other candidates who have previously decided not
to run due to the redistricting process.
Noncompetitive Elections and Turnout
Since it is clear that redistricting does in fact have some effect on
competitiveness, it is important to investigate whether a lack of competitiveness has
led to a decrease in voter turnout. One of the results of the recent decline in
competitive state legislative and congressional elections has been a situation in which
only one major party contests a race. If some districts are so overtly gerrymandered
that the incumbent or one party has an advantage, in many instances parties have felt
no need to even contest the election. Uncontested elections and races where only one
candidate has a chance of winning can work to decrease voter turnout. As Steven Hill
.. .research has shown a strong correlation between voter turnout and
competitiveness. For instance, in a study by the Center for Voting and
Democracy of 1994 U.S. House elections, voter turnout dropped dramatically
by as much as seventeen points as House races became less competitive. This
correlation was duplicated for a study of the 2000 U.S. House elections, where
voter turnout dropped by as much as nineteen points in uncontested elections,
from the closest races to those won by huge landslides. In the 2000

presidential election, voter turnout was highest in the key battleground states
where the race was closest. Most of the results of twenty studies stretching
back to the early 1970s support the hypothesized relationship between
electoral competitiveness and voter turnout, even when controlling for
campaign spending, not only in United States elections but also in Canada,
Great Britain, France, and Germany (Hill, 2002, p.88-89).
Competitive elections often receive a higher voter turnout than do races in
which the electorate knows that one candidate will win easily. This trend is evident at
the presidential level as well as at the congressional and local levels. For example, in
sixteen battleground states, turnout increased in 2000 from 1996, by an average of 3.4
percent compared with 1.6 percent increase in other states. Ten states (Arizona,
Hawaii, Idaho, Kansas, Louisiana, Montana, New Jersey, Oklahoma, South Dakota,
and Wyoming) had a lower turnout than in 1996 and these states were not close in the
presidential election. Five states (Missouri, Montana, North Carolina, North Dakota
and Wyoming) had competitive congressional and gubernatorial elections in which
voters cast more ballots than in the presidential contest. Of these states, only
Missouri was close in the presidential race (Hill, 2002).
A specific example of the link between competitiveness and turnout in
congressional races occurred in the state of Maryland. In Maryland, there are eight
congressional districts, most of which are dominated by one of the two major political
parties, and the results are often lopsided wins for one party. However, in the 2000
general election, the 8th district in Maryland was relatively close, with the Democratic
candidate winning 46% of the vote and the Republican winning 52% of the vote. The

level of turnout in this congressional district was 81%, far higher than any of the other
districts in Maryland. It is also important to note that Maryland was not considered a
swing state in 2000. This pattern can be seen in Table 2, which shows the difference
in the margin of victory in each of the eight congressional districts in Maryland as
well as the percentage turnout of registered voters. The table demonstrates that in
races that were competitive or marginally competitive, turnout was higher than in
districts where one candidate won by a landslide. While this example alone does not
in itself prove that redistricting leads to lower turnout, it does lend evidence to the
fact that a link may exist.
Table 2
Maryland Congressional District Margin of Victory and Turnout 2004
Maryland Congressional District Margin of Victory Turnout (% of registered voters who cast ballots)
1st Congressional District 30 (64-34) 74%
2nd Congressional District 38 (69-31) 76%
3rd Congressional District 52 (76-24) 73%
4th Congressional District 74 (87-13) 78%
5th Congressional District 26 (65-35) 76%
6th Congressional District 22 (61-39) 75%
7th Congressional District 74 (87-13) 65%
8ltl Congressional District 8(52-46) 81%
Source: Maryland State Board of Elections

The idea that a less competitive election would lead to lower voter turnout is
not surprising, since it is intuitive that more people would show up at the polls if they
felt their vote really counted. Congressional districts have traditionally been redrawn
after each Census, largely because that is when district lines need to be redrawn in
order to better fit each states population change. However, recent attempts at
redrawing the lines in the middle of a decade (in Texas and Colorado) have added to
the idea that politicians are not concerned with creating meaningful and competitive
elections, but rather with the creation of safe districts that will give one party a clear
advantage over the other. In the case of Texas, former Representative Tom DeLay
(R-Texas) spearheaded an effort to create new districts mid-census at a time when
Republicans controlled the Texas House. The result of his effort was that the
Republican Party gained six seats in the U.S. House election of 2004, which helped
solidify GOP control of the House that year. In addition, DeLays plan had originally
been viewed as illegal by Justice Department lawyers in the departments voting
section. The lawyers said the redistricting plan illegally diluted black and Hispanic
voting power in two congressional districts and eliminated several other districts in
which minorities had a substantial influence on election outcomes (Eggen, 2005).
Figure 9 (page 68) shows the decline in voter turnout in midterm
congressional elections since 1850. It is clear from the graph that overall turnout in
these elections has decreased significantly and has reached levels equivalent to the
South, which has traditionally had a far lower voter turnout throughout the history of

the United States. Overall turnout has been on a historical decline, but there is a
definite drop from the 1970s when compared with current levels. This decline in
midterm turnout is a result of many factors, such as the incumbency advantage and
partisan polarization. However, other factors, such as a decrease in competitive races,
advances in redistricting technology, and the overall effect of congressional
redistricting are also a factor.
Figure 9
Turnout in Midterm Elections: Texas, the South, and the Nation, 1850-2002
Source: Texas Politics, University of Texas at Austin.
Can the VRA Decrease Turnout?
While the Voting Rights Act (VRA) gave many the opportunity to vote and
protected African-Americans against the institutional racism of the past, various parts
of the VRA, as interpreted by those redrawing districts, have led to a situation in

which minority votes are packed into their own district, thus diluting their votes in
other districts. The effort to create majority-minority districts remains part of the
redistricting process and political landscape even today. Many political observers of
both parties view the results of the 1990s Supreme Court decisions (especially the
case Thornburg v. Gingles, discussed in Chapter 2) as creating more safe and
packed districts in which more minorities were represented in Congress, but fewer
Democrats were elected. As conservative commentator George Will writes:
In his 2006 edition of "The Almanac of American Politics," Michael Barone
notes that in 2004 George W. Bush received only 51 percent of the vote, yet
carried 59 percent of the congressional districts. And John Kerry won 80
percent or more of the vote in 20 districts, but in no district did Bush win 80
percent. One reason for this, Barone says, is the Voting Rights Act: Under
the prevailing interpretation of this act, redistricters are obliged to maximize
the number of districts with majorities or near-majorities of blacks and
Hispanics. That means that redistricters bunch very large numbers of heavily
Democratic precincts into a few districts and keep them out of adjacent
districts where the political balance is much closer. The Voting Rights Act,
thus interpreted, tends to result in the election of more blacks and Hispanics
and fewer Democrats." In those 20 "80 percent-plus" Kerry districts, a lot of
Kerry votes that could have elected Democrats to Congress from adjacent
districts were wasted as superfluous votes in landslide wins for Democratic
candidates. That is why, says Barone, in the redistricting in some states after
the 1990 and 2000 censuses, Republican legislators collaborated with African-
American legislators to create "majority-minority" districts. And Republicans,
who run in artificially white districts, wonder why they never learn how to
court African-Americans in presidential elections (Will, 2005).
One important point from this quote is that Republicans collaborated with
African-Americans in creating majority-minority districts. This case of strange
bedfellows has been another recent result of redistrictingwhen both major parties

collude in the redistricting process to create a situation of bi-partisan gerrymandering,
in which there are more safe districts created and fewer competitive districts.
Will also points out that not only are votes wasted in districts that are packed
with like voters, but the situation also leads to polarization. Republican
representatives find it difficult to court black votes, as their districts have smaller
ethnic minority-voting blocs within them. While the Republicans find it difficult to
get black voters to join their party, Democrats have no problem at all. African-
Americans overwhelmingly vote for the Democratic Party, especially in recent
elections, as Stanley Greenberg writes:
The result is an African-American community that votes... with nearly a
singular voice. Fully 86 percent of African-Americans align with the
Democrats, and barely 8 percent think of themselves as Republicans. This
trend has realized itself with increasing force during the ferocious battles of
the last decade: 83 percent voted for Bill Clinton in 1992 and 84 percent in
1996, 90 percent in 2000 voted for A1 Gore (Greenberg, 2004, p. 120).
While African-Americans may vote in huge majorities for the Democrats, this
voting intensity, as George Will points out, is often wasted when African-Americans
are packed into majority-minority districts. With more African-American
representation in Congress than in previous decades, some have argued that although
the Democratic Party may have lost seats, and with African-Americans voting with
Democrats in large majorities, the tradeoff is worth the overall Democratic losses in
the House. One way to measure whether this tradeoff is beneficial to African-
Americans is to assess whether the increase in black Congressional representation has

increased turnout among African-Americans residing in districts with a black
representative. One such study was conducted by Claudine Gay of Stanford
University, in which she used precinct data from eight midterm elections to measure
whether the election of African-Americans to Congress has had a positive impact on
political involvement among black Americans. The data that Gay uses comes from
relatively recent elections, during the early 1990s, which is right around the same
time that redistricting was used to create majority-minority districts. Gay writes of
her results:
In districts in which African-Americans enjoy political prominence, white
constituents are more likely to remain on the margins of the electoral process.
Black congressional incumbents routinely experience white turnout rates that
are 5-18 points lower than at polling places elsewhere in the state. This
consistent pattern of white demobilization is not offset by an equally
consistent pattern of black mobilization. The optimism of some who
champion minority representation (and, by extension, the districting
mechanism that ensures it) as a way to increase black voter participation may
be misplaced.. .more often than not, African-Americans represented by a
black member of Congress display the same patterns of behavior as their
counterparts in other districts (Gay, 2001, p.599-600).
Another reason for a lower than expected turnout among African-Americans
is that of political polarization. With this polarization in American politics come
more disenfranchised participants, especially in a winner-take-all system, in which
after an election the losing party gets nothing. African-Americans are often not even
courted by the Republican Party, especially in congressional elections. The result of
this pattern is that African-Americans are not as powerful a voting bloc as they could
be, as turnout rates among African-Americans remain far lower than those of whites.

Figure 10
Differences in Turnout for Blacks and Whites In Presidential Elections (1964*2004)
V\hite Black

60 0 I
1964 1968 1972 1976 1980 1984 1988 1992 1996 2000 2004
Figure 10 shows the turnout rates for blacks and whites for presidential
elections since 1964. As the graph clearly shows, blacks have a far lower turnout
than whites, and in the election of 2004, whites had a higher turnout rate by almost 15
percentage points. Figure 11 (page 75) shows the turnout rate for whites and blacks
in midterm elections since 1966. One striking example of the difference in turnout
rates is a comparison of the turnout percentage of blacks in presidential elections with
that of whites in midterm elections. Black turnout percentages hover around the 50 to
55 percent mark in presidential elections, while whites almost reach this percentage in
midterm elections.

Figure 11
Differences in Turnout for Blacks and Whites in Mdterm Elections (1966*2002)
White Black
One reason for the differences in voting behavior among blacks and whites
could be the way in which redistricting treats African-Americans, which George Will
and those who have participated in the redistricting process will attest to (such as
Peter Wilson, as quoted on page 10-11). African-Americans and other ethnic
minorities are packed into their own districts, creating majority-minority districts.
This process may result in more ethnic minorities being elected to Congress, but it
does not result in more competitive elections, higher turnout among African-
Americans, or increased competition for these votes, a lack that can also lead to

increased party polarity. If both parties courted African-Americans and more efforts
were made to tailor policy toward their wants, African-Americans might participate in
the electoral system more and their turnout might increase. I will return to this issue
in the conclusion of this thesis as I propose ways that the VRA can be improved upon
to increase African-American turnout and participation in congressional elections. I
also look at how alternative voting methods could help improve both minority
participation as well as representation.
An Increase in Congressional Responsiveness?
Some researchers have argued that while redistricting may not lead to
competitiveness, it actually leads to responsiveness. Two political-science professors,
Gary King and Andrew Gelman, argue that:
the decennial redistricting repeatedly injects the political system with a
healthy dose of increased responsiveness.. .The political turmoil created by
legislative redistricting creates political renewal. Many of the goals sought by
proponents of term limitations are solved by legislative redistricting.. .Far
from being a scourge on the political system in need of major reforms,
legislative redistricting has invigorated American representative democracy
(Gelman and King, 1994. p. 554).
Gelman and King summarize the two sides of the debate in political-science
literature: one side argues that redistricting decreases responsiveness because it
protects incumbents, and the other side argues that the relationship is spurious.
Gelman and King argue that in a sense, both sides of the debate are right because

gerrymandering does benefit the party in control of redistricting as compared to a
situation in which the other party controlled the process, but any type of partisan
redistricting reduces partisan bias compared to a system that had no redistricting at all
(Gellman and King, 1994).
However, while Gelman and Kings research shows that redistricting can lead
to responsiveness, it shows only that it can lead to more responsiveness compared to
a situation in which no redistricting occurred at all. How one measures
responsiveness is another factor. Perhaps elected Congressmen and women are
more responsive to those that elected them, but that number is becoming a smaller
number in most districts: as turnout falls, so does the percentage that elects the
representative. The fact remains that redistricting produces many safe districts for
incumbents, often leading to low voter turnout.
A critic of my argument that electoral competition is good for democracy
might argue that gerrymandered districts are more representative of their
constituencies than a district that is competitive and evenly split. The critic would
argue that those on the losing end in a competitive district are not represented,
whereas more people would feel represented in a district in which a larger percentage
voted for the winner. First, this argument assumes that those in an evenly split
district would feel unrepresented by the winner, which may not always be the case.
In a competitive district, candidates are more likely to be moderate and represent a
multitude of interests rather than those of a small minority. Furthermore, this

argument fails to take into consideration the fact that under the current system of
redistricting in most states, politicians are choosing who will elect them, rather than
the voters choosing the candidates. While it is true that in many areas of the country,
voters overwhelmingly support one political party over another (which is often the
case in Senatorial elections) as the composition of a state is usually not evenly
balanced, there are many areas where voters are evenly split in their political
ideologies. However, the major difference between the elections for Senators and
House Representatives (for example) is that we should expect less competition in the
un-redistricted Senate. Again, most states are not electorally competitive in terms of
party composition, which usually leads to a lack of competition. However,
congressional districts are geographically smaller and contain fewer people than do
most states, which would make competition a more realistic possibility. This is not to
say that it is possible to create 435 competitive congressional races, but as the
example of California showed (page 60), more competitive districts are possible.
When a very small percentage of House races are marginally competitive due to
political gerrymandering, the very essence of representation is questioned, as voters
have less an opportunity to vote in new representatives.
In conclusion, despite recent scholarship that argues that the redistricting
process does not affect competitive races, there are examples of the redistricting
process contributing to fewer competitive districts. Multiple examples also show that
noncompetitive elections lead to a lower electoral turnout. While these examples

alone do not prove that redistricting causes lower voter turnout, they do lend evidence
to this contention. In the conclusion, I explain that more research in this area is
necessary. In America, an increasing number of elections are noncompetitive. This
lack of choice is amplified at the congressional level, with more safe districts
existing than ever before and with congressional turnover also at an all-time low. In
the next chapter, I propose independent redistricting commissions as a vehicle for
reform aimed at increasing competition and therefore turnout at the congressional

Are there alternatives to the current redistricting process? Could there be a
more fair and productive path that creates competitive and meaningful elections for
the American people? Will reform have any effect on voter turnout? The purpose of
this chapter is to answer these questions and provide some reasonable and achievable
reforms and alternatives to a system that has created a low turnover rate, which
contributes to low voter engagement. A better approach is needed at a time when
fewer Americans are voting, participating in local politics, being less civically
engaged and increasingly turned off by politics.12
Independent Redistricting Commissions
Some states have acted to create independent redistricting commissions,
whereby citizens are better able to participate in the redistricting process, while at the
same time preventing politicians from drawing their own boundaries. Redistricting
12 See The Strange Disappearance of Civic America, by Robert D. Putnam (1996). In the article,
Putnam shows that there is a significant amount of evidence that shows Americans are far less likely to
be involved in community involvement and contain lower social trust than in past decades, which
Putnam says is a decline of social capital and civic engagement. While Putnam argues that
television viewing is responsible for this decline in political participation, for the purposes of this
thesis it is important to note that there is an increase in those being turned off to political participation.

commissions exist in many states in various forms. In seven states (Arizona, Hawaii,
Idaho, Maine, Montana, New Jersey, and Washington) commissions play a primary
role in the congressional redistricting process. Twelve states use commission solely
for state legislative redistricting. Commissions are used as backups when the
legislative process fails in Connecticut and Indiana (Mann, 2004). It should be noted
that the creation of independent redistricting commissions will sometimes still
involve state legislatures and those with a self-interest in the end result, but instead of
drawing their own districts, politicians must appoint members of a commission,
which should be made to evenly represent each major political party. The use of the
term independent must also be examined, as it is often impossible to create a
commission completely independent of political influence. Complete independence
isnt always a possibility when legislators have so much at stake in the redistricting
process. Rather, the main goal of redistricting commissions is to create a more
reasonable approach to drawing districts in order to prevent the majority party from
gerrymandering and creating uncompetitive districts.
While there is no one size fits all commission format, there are certain key
aspects that a commission should strive to encompass in order to be as independent as
possible. Taking the power of redistricting out of the hands of state legislatures-
which are concerned not with creating competitive and meaningful elections but with
the creation of safe districtsis a way to improve the responsiveness of the U.S.
Congress. The following are recommendations for an independent redistricting

panel/commission system, based on research that I conducted with the nonprofit
group Common Cause13:
First, commissions should contain an odd number of members. This
requirement reduces the incidence of deadlock that can come from an even-numbered
panel. It is also important that the commission be funded adequately by the
legislature. A commission must be given sufficient tools and data to make
competitive districts that also meet requirements set out by Supreme Court precedent
as well as the Voting Rights Act.
Second, members of the commission should be as nonpartisan as possible,
and should not be politicians in the legislature. The more partisan a commission is,
the more it is likely to favor the status quo and create safe districts. This is not to say
that the major political parties should have no say in the redistricting processeach
major political party should be able to appoint an equal number of members if the
process calls for appointment of commission members. This rule would give the
major political parties a fair degree of representation. The point here is to create a
situation in which politicians have less of a say in how their own districts are drawn.
Therefore, members of the commission should not currently be serving in the state
legislature or be allowed to run for office in the districts they create.14
l3For more information on this groups stand and recommendations for redistricting commissions see
Colorado Common Cause. Redistricting Reform.
14 Examples of politicians running in districts they create are not unheard of. One example in 2002
was Brad Miller (D), who ran in a newly created district in North Carolina after the state gained a seat

Diversity is important in the selection of members of the redistricting panel.
Ideally, there should be more independent members than partisan members and the
panel should be required to obtain a unanimous, or at the very least, super-majority
vote to pass measures. This requirement is put in place so that the two major parties
cannot work together to create a situation of bi-partisan gerrymandering (when the
two major parties collude to create safe districts for each other). For example, if the
panel is to be composed of seven members, then there should be three independent
members, with two Democrats and two Republicans making up the rest of the panel.
In this example, a vote of five would be needed to pass measures and the major
parties would need to get at least one independent to vote with them. As has been
seen in recent initiative efforts (such as in California in 2005), there have been
attempts to create commissions composed of former judges, rather than elected
officials or members appointed by political parties. This approach is also a plausible
alternative to putting the power of redistricting in the hands of politicians, with the
key being to keep the power of drawing legislative and congressional districts out of
the hands of elected officials who will directly benefit from the process. It should be
noted that the process by which a committee chooses members matters greatly in the
outcome of any redistricting process. If, for example, members are chosen from a
pool at random, there is less of a chance of partisan influence.
following the recent 2000 census and apportionment process. Before running in the new district,
Miller served as the chairman of the redistricting committee responsible for drawing the new district.
Not surprisingly, Miller beat his Republican challenger by 13 percentage points. See Democrat Brad
Miller Tops Grant in North Carolina 13th from the Washington Post, November 5th, 2002.

While the selection of members matters, there must also be a mechanism in
place that allows voters to replace the entire panel if they feel the districts were not
drawn fairly or to create competition. The option for a recall or vote of no
confidence should be given to voters after each round of redistricting occurs, so that if
the voters are unhappy with the result, they are able to vote out commission members.
Third, the commission should hold regularly scheduled meetings in public, so
that citizens have the opportunity for input. The Independent Redistricting
Commission should conduct several public hearings throughout the state on proposed
plans, allowing for both comments and questions from members of the public.
Regular meetings of the commission should be open to the public and at least 10 days
notice should be given for all regular meetings of the commission. All meetings
regarding redistricting at which two or more members of the redistricting commission
are in attendance should be considered a public meeting and be open to members of
the public and subject to adequate notice requirements (at least 72 hours). All
submitted maps, plans, revised plans, commission agendas, hearing transcripts,
meeting minutes, descriptions of proposed districts, and other data should be
available in a timely fashion, free of charge, via a public website and other means.
Members of the commission would be prohibited from all ex-parte communications
with members of the legislature, other elected officials, former elected officials,
candidates for office, representatives of political parties, and registered lobbyists
regarding redistricting.

Fourth, a commission should have a competitiveness requirement in drawing
districts. Currently, Arizona contains the only commission system in the United
States that has a competitiveness requirement for members drawing legislative
districts. The language of the Arizona guidelines read, To the extent practicable,
competitive districts should be favored where to do so would create no significant
detriment to the other goals (Proposition 106, 2006). While the competitiveness
requirement in Arizona is subordinate to other requirements, it remains important, as
the panel must attempt to create competitive districts, which is one the most important
goals in the creating of an independent redistricting commission.
Creating competitive districts is one of the most difficult and most important
aspects of a redistricting commission. The example of Colorados 7th Congressional
district (Figure 11) provides perhaps the best example of a competitive district. In the
7th Congressional district, registration is divided almost equally among Democrats,
Republicans and Independents. When the district was first created in 2002, it was the
closest election in the nation, with Bob Beauprez winning by only 121 votes. During
the next election in 2002, Beauprez won by a wider margin of 55% to 43% over his
Democratic challenger. Beauprez had many advantages, such as incumbency, a pro-
GOP overall nationwide election, and the war in Iraq (which at the time worked to the
Republicans favor) (, 2006). However, because Beauprez chose to run for
Governor of Colorado in 2006, the 7th congressional district is once again considered
the most competitive in the nation (Cilizza, 2006). Therefore, when possible, it is

best to have party registration be as close to equal as possible in the creation of
districts. While this guideline is not possible in every state or in every congressional
district, the possibility exists for more congressional elections to be competitive.
Figure 12
Contfrcssioruil Dislri< I ~

y ri,

i Dfftvir rr~-*
S . : .It yAwfOfl

it&. y~

? : i j! i IKtru
r t &t~
ss< e i 4
i o/( 7 *s
Fifth, congressional and legislative redistricting should occur once every 10
years following the decennial U.S. Census and Congressional Reapportionment,
unless the Independent Redistricting Commission is directed by court order to create
a new plan. This proviso is to prevent efforts such as those that occurred in Texas
and Colorado, when the majority party attempted to redistrict in the middle of the
decade after gaining control of the state legislature (Frates, 2006). Changing the rules
and boundary lines after each new party gains control of the state legislature for
political benefit defeats the purpose of an independent redistricting commission and
leads to a situation of partisan gerrymandering. Limiting redistricting to once every
ten years lessens the chance of a power grab by either party. In addition, upon

passage of the redistricting reform legislation/initiative, redistricting should be put off
until the end of the decade so that voters and politicians do not view the reform effort
as a power grab by the party who does not control the legislature. For example, if a
state passes new legislation creating an independent commission, that commission
should not take effect until the next census. A lesson suggested by the failed Ohio
and California initiatives: politicians there were more likely to raise large amounts of
money against the measures since the redistricting took effect immediately (Hirsch
and Mann, 2006). If redistricting efforts are put off until the next census, politicians
are more likely to support a measure and not view it as a power grab by the party not
in power. If the initiative/legislation is put off until the end of the decade, rather than
the next congressional election, voters and politicians may be more accepting of a
reform proposal.
Finally, once a plan is created, there should be an opportunity for judicial
review so as to ensure that the plan meets current state and national laws (e.g. The
Voting Rights Act). The Commission should work to create districts with the
following characteristics:
Congressional and legislative districts should be composed of equal
populations The Supreme Court has interpreted the Equal Protection Clause
of the Fourteenth Amendment as providing the guarantee of equal population
of districts.

Districts should comply with the United States Constitution and the Voting
Rights Act.15
District boundaries should respect communities of interest to the extent
practicable, based on visible geographic features; city, town, and county
boundaries, and undivided census tracts, similarities in social, cultural, ethnic,
and economic interest, school districts, and other formal relationships between
municipalities. Districts need to be geographically compact and contiguous.
Compactness reflects the notion that districts should be composed of a tightly
defined area so that representatives may be able to more efficiently
communicate with their constituents. Contiguity requires that all parts of a
district must be connected.
The redistricting process should be incumbent blind The commission
should not know nor take into account the address of any individual, including an
The aforementioned suggestions are basic guidelines for a commission
system. Different formats will work better for different states, and experimentation
should be encouraged. The key is to take the redistricting process out of the hands of
15 While I have offered criticism of the VRA in the context of competitiveness, it remains the law of
the land. In the next chapter I offer suggestions for ways of improving the VRA that can still protect
minorities and allow for competitive elections.

self-interested politicians, for congressional and state legislative redistricting. What
some states have already done is where I now turn my attention.
A Few Model Commissions and Proposals
A number of states have changed from having the state legislature control the
redistricting process and turned the process over to redistricting commissions. What
follows is a short description of what each of these states has done. In addition, I
have listed the two recently failed redistricting initiatives in Ohio and California.
Also included in the following section are two initiatives that will likely be seen on a
future ballot in future Florida and Massachusetts.
Established in Washington by constitutional amendment in 1983 and first
instituted in 1990, is a five-member independent redistricting commission that draws
legislative and congressional districts. The legislative leaders of the two largest
parties in each house of the legislature each appoint one commissioner. The four
members then elect a fifth non-voting member to act as chair. No elected official or
person elected to party office may serve on the commission. Three voting members
are required to approve a plan. The legislature may amend the redistricting plan by a
timely 2/3 vote passed by both houses. The legislature may also reconvene the

commission by 2/3 vote to modify the plan. The Governor does not have veto power
over either the congressional or legislative redistricting plans (Washington State
Redistricting Commission, 2006).
Table 3
Competitive Districts and Turnout in Washington
# of Competitive Districts (margin of victory was less than 10%) Year Turnout as % of VAP
1 2004 68
1 2002 40
1 2000 57.6
3 1998 45.5
5 1996 55.6
6 1994 43
5 1992 60
Source: Washington Secretary of State and U.S. House of Representatives, Office of the Clerk
Immediately after the redistricting commission was instituted in Washington
in the early 1990s, the number of competitive congressional districts increased from
the 1980s. As can be seen from Table 3, the number of competitive congressional
districts rose following the implementation of a redistricting commission and
remained high throughout the 1990s. Washington has a total of nine congressional
districts, which means that during the 90s almost half of the districts were

competitive during each election. Unfortunately, during the 2000s the level of
competition went down considerably. The decrease in competitiveness was one
factor that worked to decrease turnout in midterm elections. Comparing the electoral
turnout in midterm elections, it is clear from Table 3 that turnout was much higher in
years when there were many competitive congressional elections (1994,1998)
compared to 2002, when only one district was competitive. Despite a decrease in
competition in the 2000s, four of the nine districts elected incumbents with less than
60 percent of the vote (Hirsch and Mann, 2005). Washingtons example
demonstrates that a redistricting commission can create more competitive districts
and increase turnout.
Proposition 106 was passed by Arizona voters in the 2000 election by a
margin of 56% in favor to 44% against. The initiative amended the state constitution
and provided that a five-member independent redistricting commission create
legislative and congressional districts. Appointments are made from a pool of 25 (10
from each party and five not registered to a party) and are made by the highest-
ranking officer in the House (1), the Minority Leader of the House (1), the highest-
ranking officer of the Senate (1), the Minority Leader of the Senate (1). The fifth
member is elected with a majority vote by the four selected commission members to
serve as the chairman and may not be registered with a party already on the

commission. A failure to appoint a chairman will give the power of appointment to
the Commission of Appellate Court Appointments. No more than two members of
each party may be commissioners and of the first four appointed, no more than two
may be from the same county. Additionally, commissioners must have been
registered with the same party, or be unaffiliated for the three years preceding
appointment to the commission. Members of the commission may not be appointed
or elected, or be a candidate for any other public office. A three-member affirmative
vote is needed for an action on redistricting plans (Proposition 106, 2006).
While Arizona has a specific competitiveness requirement, other factors take
precedence over drawing competitive districts. As professor and redistricting expert
Michael P. McDonald writes:
Arizona is a Republican leaning state and was required to draw two majority-
minority Hispanic congressional districts in order to be approved by the
Justice Department under section 5 of the VRA. After drawing these two
uncompetitive Democratic districts, the remainder of the state was even more
Republican, and the commission only realized the opportunity to draw one
competitive congressional district (McDonald, 2005).
As a result of the VRA restrictions put upon the redistricting commission,
congressional elections in Arizona have not been very competitive since the
commissions adoption in 2000.