Time to Move:
Overseas and Military Voter State Policy Innovation
Claire M. Smith, PhD
Research Director
Overseas Vote Foundation
claire@overseasvotefoundation.org
Abstract
In October 2009, the most significant bill in decades regarding overseas and military voters was
passed by the Senate. The Military and Overseas Voter Empowerment (MOVE) Act was
implemented by the states for the first time during the 2010 elections, and dramatically changed
the landscape of overseas voting. For example, in 2010, 48 states began emailing blank ballots
to voters and 21 states accepted voted ballots via email. However, despite this federal mandate,
a myriad of state policies still exist and 9 states and the District of Columbia had to apply for
hardship waivers. This paper takes a closer look at the extensive voting policy movement that
took place in 2009 and 2010 in the states in the wake of the MOVE Act. Why were some states
more aggressive in passing reform than others? I examine this question in two ways. First, I
measure state policy innovation with the UOCAVA State Policy Index (SPI). Then, statistical
analysis is used measure the impact of partisan control, legislative professionalism and
competitiveness on policy development. Second, I look at two specific cases (Alabama and
Minnesota) as examples of states that did (or did not) successfully implement new policies.
This paper was prepared for presentation at the Midwest Political Science Association, Chicago, April 1, 2011.
Introduction
Nothing reflects our increasingly globalized world more than the rising numbers of
Americans living abroad. Either due to military obligations, as members of the international
workforce, as students, or by choice, anywhere between 4 to 6 million Americans live overseas.
This has created one of modern democracy’s greatest challenges: the overseas voter. In response
to the unique problems that voters living abroad face, Congress has passed several key pieces of
federal legislation, such as the Uniformed and Overseas Citizens Absentee Voting Act
(UOCAVA) in 1986.
Despite the efforts of legislators and election officials, statistics from the 2008
presidential election indicated that military and overseas citizens continued to experience
problems when attempting to vote. In May 2009 Gail McGinn, the acting Under Secretary of
Defense for Personnel and Readiness, reported to the US Committee on House Administration
the following information regarding the 2008 election:
“Preliminary data from a forthcoming report on the 2008 election from the Congressional
Research Service found that 72 percent of military absentee voters in the seven-state
study successfully returned their ballot and had their votes counted. That is the good
news. The bad news is that 28 percent of ballots were described as not returned
(approximately 22%), rejected (approximately 3%) or returned as undeliverable
(approximately 3%) by election officials from the seven states.”1
Furthermore, the Election Assistance Commission reported in 2008 that 7% of those ballots
returned by overseas and military voter were rejected, with the primary reason being a missed
deadline (Election Assistance Commission 2009, 11; hereafter EAC). In its 2008 Post-Election
Survey, Overseas Vote Foundation found more than half (52%) of those who tried but could not
1
See Gail McGinn, acting Under Secretary of Defense for Personnel and Readiness, Testimony before the U.S.
Committee on House Administration, Military and Overseas Voting: Obstacles and Potential Solutions, May 21,
2009, available at http://cha.house.gov/UserFiles/220_testimony.pdf.
2
vote, were unable to do so because their ballots were late or never arrived (Overseas Vote
Foundation 2009a, 5; hereafter OVF).
In an effort to address these problems, the most significant bill in decades regarding
overseas and military voters was passed by the Senate in October 2009. The Military and
Overseas Voter Empowerment (MOVE) Act addressed several key issues meant to speed up the
voting process, including the electronic transmission of blank ballots and mandating that blank
ballots be sent 45 days before Election Day. The MOVE Act was to be implemented by the
states during the 2010 elections.
However, despite this federal mandate, a myriad of state policies still exist and nine states
and the District of Columbia had to apply for hardship waivers to the MOVE Act. This paper
takes a closer look at the extensive voting policy movement that took place in 2009 and 2010 in
the states. Why were some states more aggressive in passing reform than others?
I examine this question in two ways. First, I measure state policy innovation with the
UOCAVA State Policy Index (SPI). Then, statistical analysis is used measure the impact of
partisan control, legislative professionalism and competitiveness on policy development. Second,
I look at two specific cases (Alabama and Minnesota) as examples of states that did (or did not)
successfully implement new policies. What emerges from this analysis is not a story of election
laws created out of partisanship, but rather states that were early adopters and those that simply
did not have enough time in their legislative session to tackle this issue.
The Evolution of Overseas and Military Voting
The history of absentee and military voting stretches back to the Civil War. At that time,
many states excluded absentee voters. Inbody (2009) argues that partisan politics, specifically
the expectation that soldiers would vote for Republicans, prevented states with Democratic
3
controlled legislatures (such as Illinois) from supporting military absentee voting during the Civil
War. As a result, although several states created legislation to promote voting by soldiers
stationed out-of-state, there was no comprehensive national legislation, and state barriers to
absentee voting persisted through the 1940s.
In the 1940s, the US once again found itself at war and with large numbers of voters
away from their polling place on Election Day. US involvement in World War II led to the first
important piece of federal legislation, the Soldier Voting Act of 1942. Once again the bill was
opposed by members of the Democratic Party, and voting in Congress was split along regional
lines, as Southern members of Congress felt the new legislation violated states’ rights. Because
little information was available about the impact of legislation and the turnout of voters,
President Truman commissioned a report by the American Political Science Association
(APSA). The report analyzed the experience of military voting in the 1944 and made policy
recommendations. The report found that:
“The number of persons of voting age in the Armed Forces in 1944 was about 9,225,000.
Of that number, 4,487,540 were reported as applying for ballots for the general election
of that year and 2,691,160 sent in ballots that were counted. In that election about 60
percent of the civilians of voting age voted; nearly 50 percent of the servicemen of voting
age were reported as applying for ballots and about 30 percent succeeded in voting. The
service vote was 5.6 percent of the total popular vote for President, 48,025,684.”
Based on its review, the commission recommended, among other things, mailing absentee ballots
45 days before an election. The Federal Voting Assistance Act of 1955 was passed as a result of
the APSA report, and, for the first time, provided support of civilian employees living abroad as
well as military personnel. The Act also created the Federal Voting Assistance Program
(FVAP), which is the federal government agency that is responsible for implementing UOCAVA
and is a part of the Department of Defense.
4
Congress updated legislation in 1975 (the Overseas Citizens Voting Rights Act) to clarify
reporting rights and procedures, and enacted the current law, the Uniformed and Overseas
Citizens Absentee Voting Act (UOCAVA) in 1986. UOCAVA covers the following citizens:2
1. all military voters who, by reason of active duty or service is from the jurisdiction of their
legal voting residence, including those based in the United States or abroad,
2. their family members,
3. individuals residing outside the United States and qualified to vote in the last place in
which they were domiciled before leaving the United States.
Identifying and counting the number of individuals coved by this definition is not simple. The
US Census Bureau Census included approximately 580,000 federal employees and dependents
(226,363 military personnel, 30,576 civilian employees, and 319,428 dependents of military and
civilian employees) in their 2000 apportionments (U.S. Census Bureau 2001). The Department
of Defense Personnel and Procurement Statistics reported that 283,589 military personnel and
42,992 civilian employees worked abroad as of December 2008 (Department of Defense 2009).3
McDonald (2009) estimates that there are a total of 4,972,217 eligible UOCAVA voters.4
The UOCAVA Voting Process
For overseas and military voters, the voting process is comprised of four parts, each of
which is regulated. First a voter must register and/or request a ballot by filling out and sending
in the proper paperwork to the appropriate local election official (LEO) in the US. The Federal
Post Card Application (FPCA) is the official federal government name given to the voter
registration form used by voters eligible to vote under UOCAVA, and is the primary form for
“UOCAVA voters” refers to those voters covered by the Overseas Citizens Absentee Voting Act (UOCAVA) of
1986, which encompasses military personnel and overseas Americans.
2
3
For more information regarding the problems involved in counting UOCAVA voters see Smith (2009).
4
To arrive at this total, McDonald first takes the number of deployed military personnel as reported by the
Department of Defense. He then deflated the civilian numbers by 25%, which he did because this number
corresponds to the proportion of minors among the United States resident citizen population. The McDonald
estimates refer to the “voting eligible” population only.
5
requesting registration and/or an absentee ballot from election officials. Some states, but not all,
allow this form to be sent via fax and/or email. A few states still require citizens to either sign a
state oath or provide additional forms of identification when registering to vote. Most, if not all,
states require an original signature on file, or a signed original FPCA to send a ballot.
Second, the local election official (LEO) processes the request. If the voter has entered
all of the correct information and submitted the form before the applicable deadline, then the
LEO will send the voter a ballot. In 2008, 37 states and the District of Columbia permitted
UOCAVA voters to receive the blank ballot via fax and 20 states allowed the delivery of the
blank ballot via email.
Third, once the voter receives the ballot, she fills it out and mails it back to the US. In
2008, 26 states and the District of Columbia permitted the return of voted ballots via fax and 12
states allowed the return of the ballot via email. Traditional postal service remains, however, the
predominant form of ballot return (OVF 2009a, 21). In addition to receiving the faxed or
emailed ballot, most, if not all, states require an original signed ballot envelope or ballot
affirmation in order to count the ballot. Finally, the election official receives and counts the
completed ballot. Should a voter not get a ballot, she has the option of using a Federal Write-in
Absentee Ballot (FWAB). The FWAB is an alternative, downloadable ballot, accepted by all
states and territories, which can be used in federal elections.
The various steps can be extremely time-consuming and the process of voting can take
anywhere from two weeks to two and a half months to complete (PEW 2009, 40). As a result of
this lengthy process and traditional mailing methods, the number one reason UOCAVA voters
continue to have problems is because they miss deadlines or do not receive their ballots on time
(OVF 2009a, 5). A 2001 GAO report indicates that the variety of state and local requirements,
6
lack of feedback from election officials and time constraints worried voters the most (GAO
2001, 3). In 2007 the EAC conducted a survey in order to investigate the UOCAVA voting
experience, as well as attitudes towards electronic voting methods. In their analysis of the EAC
data, Cain, MacDonald and Murakami (2008) found that overseas civilians found it more
difficult to register than military voters. All voters voiced concerns about getting their ballots on
time. These concerns are well founded as election officials have reported “missed deadlines” as
the number on reason for ballot request and ballot rejection (OVF 2009a, 30, 31).
A New Federal Mandate: The MOVE Act of 2009
As a result of the continued problems faced by voters, policy makers on both the federal
and state levels continued to propose and fine tune legislation throughout 2009. This culminated
in the passage of the Military and Overseas Voter Empowerment (MOVE) Act in October 2009,
which represented the first major policy change for overseas and military voters in more than a
decade. The MOVE Act amended UOCAVA, and the key provisions of the Act targeted:

technology (voter registration information online, options for electronic delivery
of blank ballots, and ballot tracking systems),

communications (use of email to communicate with voters), and

election administration (transmission of blank ballots 45 days before Election
Day).
The MOVE Act required states to implement these provisions in time for the 2010 midterm election. States unable to meet the 45-day pre-election ballot transit deadline were required
to file a request for a waiver, first in consultation with the U.S. Attorney General and with
approval provided by the Department of Defense. By August 2010, 24 states had passed
measures to establish state-level compliance with the MOVE Act, and by the end of the year 32
7
states and the District of Columbia had passed new laws. 12 states had not proposed any
legislation in time for the 2010 election. Five states (AL, KS, PA, WA, and WI) all introduced
legislation which failed to pass the state legislature. These results are summarized below in
Table 1.
TABLE 1: MOVE ACT RELATED LEGISLATION IN THE STATES, 2010
No Legislation Proposed
Alaska
Arkansas
Colorado
Maryland
Montana
Nevada
New Mexico
North Dakota
Oregon
Rhode Island
South Carolina
Texas
Legislation Proposed,
Failed to Pass Legislature
Alabama
Kansas
Pennsylvania
Washington
Wisconsin
Legislation Pending
New Jersey
Legislation
Enacted
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Louisiana
Maine
Michigan
Minnesota
Mississippi
Missouri
Nebraska
New Hampshire
New York
North Carolina
Ohio
Oklahoma
South Dakota
Tennessee
Utah
Vermont
Virginia
West Virginia
Wyoming
Arizona
California
Delaware
Massachusetts
Connecticut
Kentucky
8
Although the MOVE Act covered several policy areas, for the purposes of this study, I
will focus my analysis on options for electronic delivery of blank ballots and on the 45 day
transmission deadline.
The first issue, the electronic delivery of blank ballots, received special attention in the
2009 Pew study “No Time to Vote,” in which the Pew Center on the States analyzed the amount
of time it takes to complete the UOCAVA voting process and compared this to state regulations
that affect voting time, such as registration deadlines, ballot transmission times, and voting
deadlines. It adjusted this base number according to the electronic transmission opportunities
that a state provides and created a measure of “days available to vote.” According to these
measures, the “best” state, New Mexico, provides 46 extra days to vote. Citizens from the
“worst” state, Oklahoma, do not have enough time to vote; in fact, they would need an extra 26
days to vote.
In an attempt to give voters more time to vote, Sections 577 and 578 within the MOVE
Act mandated that states provide voters with an option for online blank ballot delivery (i.e. email
or downloadable ballots). It did not specify or require any form of online voted ballot return.
Many states, including Minnesota and Texas, improved their existing systems by supplementing
the new technologies they had put in place over the previous election cycle. Separately, new
pilots for “Internet Voting” were launched in West Virginia and the District of Columbia.
The maps below show how states used technology in the 2010 elections to distribute
blank ballots, and in some cases to receive voted ballots. As we can see, there was a remarkable
increase in the use of electronic transmission methods for delivery of blank ballots. In 2010, all
50 states provided for the transmission of a blank ballot in an electronic form (mainly email or
online download). This use was up from 20 states in 2008. Only two states, Alaska and Rhode
9
Island, offered blank ballots via fax as their only electronic delivery method. Several states place
restrictions on the use of email. For example, Colorado only allows military voters to receive
ballots via email and not overseas civilians.
Figure 1: Technology and the Transmission of Blank Ballots in 2010
As seen below in figure 2, only 24 states permitted voters to return their voted ballots via
fax and email in 2010. 19 states still required that voters return completed ballots using
traditional postal methods.
Figure 2: Technology and the Return of Voted Ballots in 2010
10
The second primary feature of the MOVE Act was the mandate to mail blank ballots 45
days before Election Day. The call for the 45-day window to send blank ballots dates back to the
Truman administration. Sixty years later, the MOVE Act finally mandated that this
recommendation be implemented by all states. This legislative requirement had an immediate
impact. In 2010, 40 states had either legislation in place or created new legislation mandating a
45 day total ballot transit time, up from 28 states in 2008.
Unfortunately, by August 2010 it became clear that not all states would be able to comply
with the September 18, 2010 deadline. Ten states, Washington DC and the Virgin Islands
applied for waivers. Five waivers were approved (DE, MA, NY, RI, WA), while six were denied
(AK, CO, HI, WI, DC and the Virgin Islands). Maryland withdrew its waiver request. The
majority of these states extended their ballot receipt deadlines in order to create a 45 day
transmission window and to comply with the MOVE Act.
Measuring State Policy Innovation
As discussed above, although UOCAVA and the MOVE Act provide federal guidelines
to the states, every state creates its own legislation regulating each step of the voting process.
Despite their complexity, legislation for Americans living abroad can be divided into two
primary dimensions, registration and balloting. Some of these initiatives make it easier for
voters to participate (such as allowing a ballot to be sent via fax), whereas others create barriers
to voting (for example ballot notarization and/or witness requirements). Because some policies
make the voting process easier and some more difficult and therefore these policy dimensions
can be measured. By measuring the restrictiveness of voting legislation (i.e. does a policy make
it easier to vote), it is possible to create a UOCAVA State Policy Index.
11
When reviewing different voting procedures, I identified eight registration laws and eight
balloting regulations that comprise the core of UOCAVA legislation. These 16 different
requirements are at the heart of the recommendations that the FVAP and EAC have made to the
US States since 2004. In its biannual Voting Assistance Guide (VAG), the FVAP compiles
absentee voting regulations, laws and deadlines and is the primary source of information for
UOCAVA voters regarding procedures. I used the information within the VAG from 2004,
2006, 2008 and 2010 to code UOCAVA policies from those years. 5 The 16 essential policies
that are coded to create the UOCAVA Registration Score and UOCAVA Balloting Score are
summarized below in Table 2.
5
Unfortunately, the VAG is often inaccurate and incomplete. As a control, I compared the 2008 and 2010 VAG
with OVF’s state specific information tables. Where there were inconsistencies, I contacted the state directly or
download the state’s election code in order to determine what is correct. In order to create the 2004 and 2006 index,
I only consulted the VAG’s for each specific year. As the VAG is the primary source of information for voters, the
information contained within reflects the conventional wisdom at that time and is an accurate reflection of the
institutional voting environment.
12
TABLE 2: UOCAVA INDEX POLICIES
Policy
Impact on Voter
REGISTRATION
Identification or signature requirements
hinder voters who are either not aware of these rules or do not have
access to another American voter as a witness
Registration waived or same day registration
makes voting easier by eliminating a step in the voting process and
reducing the time it takes to vote
Registration and/or ballot request by fax
makes voting easier by reducing the time it takes to vote
Registration and/or ballot request by email
makes voting easier by reducing the time it takes to vote
Hard copy requirements
prevent the successful completion of the registration and ballot
request process if voters are not aware of the requirement, and thus
denied registration because they do not submit a hardcopy after a fax
and/or email
Citizens born overseas have the right to vote
Americans born abroad but never established residency in the US
are only allowed to vote in States that permit registration via their
parent’s last address; however only 16 states currently permit this
BALLOTING
Ballot transmission time
States that do not send out ballots on time do not allow voters
enough time to return them; 45 days is the recommended minimum6
Notarization or witness requirements
hinder voters who are either not aware of these rules or do not have
access to another American voter as a witness or a notary
Transmission of blank ballot by fax and/or email allowed
makes voting easier by reducing the time it takes to vote
Return of voted ballot by fax and/or email allowed
makes voting easier by reducing the time it takes to vote; however a
voter must often waive her right to privacy
Hard copy requirements
prevent the successful completion of the registration and ballot
request process if voters are not aware of the requirement, and thus
their voted ballot is rejected because they do not submit a hardcopy
after a fax and/or email
Expanded use of the FWAB
Can promote successful voting by providing citizens with options
should a ballot not arrive on time
Theoretically, each step in the voting process is equally important, and in order to
implement fairness and consistency in my coding criteria, I weight each legislative requirement
equally.7 Policies which are less favorable for voters by making it more difficult to vote by
creating barriers or complicating the process start at the null point. Policies that make it easier
for voters to participate receive more points. For example, a state that does not allow ballot
requests to be sent by fax receives a “zero.” States that allow faxing from certain segments of
6
Some activist groups favor a 60 day transit minimum, as it provides the majority of voters enough time to vote.
However, 45 days has been the consistent recommendation for 15 years and therefore I use it as the coding standard
for this study.
7
There is some debate about if the use of email and fax are of equal caliber. Email is more reliable and widely
available than fax; however, more states use fax than email. Because there is not enough data to create a systematic,
fair weighting system, I have coded these variables in the same manner.
13
the population, such as military voters, but not to the entire UOCAVA population receive a “.5”.
States that allow the entire UOCAVA population to fax in ballot requests are given a “1.” All of
the coding criteria are summarized in Appendix 1.8
Essential to my coding scheme is the idea that a policy facilitates a voter’s ability to
submit a ballot accurately and on time. Therefore I code the submission of completed ballots via
email and fax positively, i.e. a state receives a 1 if it is allowed and a 0 if it is not. Many election
integrity and computer science experts would present strong arguments about security risks in
sending voted ballots by unsecured email. I am purely grading the “facility” of the electronic
option.
I combined the 2008 UOCAVA Registration and Balloting scores together to create the
2008 UOCAVA State Policy Index, which is summarized in Appendix 2. The minimum score
for 2008 is 2 and the maximum is 14. The mean is 8.147 and the index has a standard deviation
of 2.536. The top five states are: Iowa, Colorado, New Mexico, South Carolina and Kansas.
The bottom five states are: Alabama, Wyoming, Arkansas, New York and Nevada. As seen in
the scatter plot below in figure 1, there was a large variance in policy among the states.
8
American Samoa, Guam, Puerto Rico, and the Virgin Islands are also covered by UOCAVA. However, because of
missing data, both on the dependent and independent variables, they are removed from this analysis.
14
Figure 3: Scatter Plot of 2008 Balloting and Registration Scores
I then combined the 2010 UOCAVA Registration and Balloting scores together to create
the 2010 UOCAVA State Policy Index. The minimum score for 2010 is 4 and the maximum is
15. The mean is 10.843 and the index has a standard deviation of 2.524. The top five states are:
Iowa, New Mexico, Massachusetts, North Dakota, and Oklahoma. The bottom five states are:
Alabama, Louisiana, Connecticut, South Dakota and Utah. As seen in the scatter plot below in
figure 2, the MOVE Act succeed in forcing the average SPI score up and in moving states to the
upper right hand quadrant.
Figure 4: Scatter Plot of 2010 Balloting and Registration Scores
15
Statistical Analysis: What drives policy innovation?
As can be seen from the discussion above, many states were aggressive in pursuing
MOVE Act legislation, which is reflected in the increase in their UOCAVA SPI scores from
2008 to 2010. Other states, however, continued to fall behind in policy development and only
enacted minimal changes. Thus, the SPI is especially helpful in determining the breadth of state
policies and as a tool for quantitative research and large N statistical analysis. For the dependent
variable in this study, I calculated the difference in the UOCAVA SPI scores from 2008 and
2010. There are a number of independent variables that can impact the development of elections
policy, especially political and institutional factors.
Political factors. Inbody’s (2010) arguments about the development of military and
overseas voting during the Civil War and World War II suggest that partisan politics play a role
in policy development on overseas and military voting. Partisanship has also been shown to
have an impact on the development of other electoral rules, such as redistricting (Winburn 2008)
and other election reforms (Bali and Silver 2006). I measure partisanship using the Ranney
Index for 2003-2006, and which is composed of: proportion of success, duration of success, and
frequency of divided control.9 Scores of 0 indicate complete Republican control and scores of 1
indicate complete democratic scores. Because of the assumption that military voters tend to vote
Republican, I expect more Republican legislatures to be associated with more progressive voting
policies for military and overseas voters.
Institutional factors. An important institutional factor emerges when examining a list of
those states that did not introduce any legislation in 2010. Seven of the twelve states either did
not have a legislative session in 2010, or they had a limited session. Even those states with
9
I also calculated an alternative measure partisanship within the state legislature as the ratio of Democrats to
Republicans in the lower House in 2010. Higher scores indicate more democratic legislatures and lower scores are
more Republican.
16
longer legislative sessions felt pressure to address urgent fiscal matters rather addressing the
concerns of overseas and military voters. Therefore I have included the length of the state
legislative session (as calculated in months) as an additional independent variable. I expect that
states with longer legislative sessions had more time to create legislation and thus have more
progress voting policies.
In my aggregate analysis I included several other independent variables: state legislative
professionalism, the size of the state bureaucracy, and the competitiveness of political parties
within the state.10 I expect a positive relationship between these variables and the UOCAVA
State Policy Index (SPI). That is, as state professionalism, bureaucracy, and competitiveness
increase, the progressiveness of state policies also increases.
Model. Because I am only examining change from 2008 to 2010, I used a regression
model with change in the SPI UOCAVA index as the dependent variable. Unfortunately, the
results of these tests were inconclusive. None of the proposed variables appeared to have a
significant impact on the adaption of policy. As an alternative, I also attempted a binary logit
model in which the dependent variable was coded 1 if the state passed MOVE Act legislation
and 0 if it did not. These results were also inconclusive. Only one relationship appeared to have
any significance: SPI Change and length of the legislative session are correlated (coefficient:
.237, significance: .100).
Case Studies: Early Adopters vs. Reform by Federal Mandate
Despite the inconclusive statistical analysis above, several observations can be made
when looking at individual cases of states that passed (or did not) MOVE Act legislation in 2010.
I operationalize state legislative professionalism using Squire’s Index from 2003 (see Squire 2008). The number
of all state employees in 2004 and 2006 was obtained from the U.S. Census Bureau. I use Holbrook and La Raja’s
computation of the state Ranney Index from 2003 to 2006 to operationalize interparty competitiveness (see
Holbrook and La Raja 2008).
10
17
First, many states did not introduce legislation because they were already compliant with the
federal mandates. These early adopters passed legislation before the MOVE Act of 2009.
Among these states are Minnesota, Wisconsin, and North Dakota, which were also among the
first to adopt other voter registration reforms. In his discussion of the adoption of those policies,
Hanmer (2009) alludes to the “participatory culture” of those states that adopted Election Day
registration. However, although Hanmer argues that those states with a participatory culture are
more likely to adopt voting reforms early, he does not systematically test this theory, nor does he
provide examples of cases in which reform was proposed but failed. On the other hand, even
early adopters, such as Minnesota, were unable to pursue their complete legislative agenda. The
importance of federal mandates as an impetus to reform becomes clearer when we closely
examine two different cases: Minnesota and Alabama.
Minnesota: “Federal mandates provide a public demand”
Shortly after taking office in 2007, Minnesota Secretary of State Mark Ritchie began to
pursue legislation that would make it easier for overseas and military voters to participate.
Legislation in 2007 was introduced and passed by both houses of the state legislature, but was
then vetoed by the governor. One of the more contentious aspects of that 2007 legislation was
moving the primary in Minnesota from September to August. Fortunately, a modified bill was
passed and signed by the Governor in time for the 2008 election. This modified bill allowed
election officials to deliver blank ballots by email and fax, eliminated witness and notary
requirements, and centralized communications with a UOCAVA board.
The 2008 reforms had an immediate, positive influence. Minnesota increased the number
of ballots sent to overseas and military absentee voters by 33 percent but increased the casting of
these same ballots by an astonishing 282 percent when compared to the 2006 General Election
18
(OVF 2009b, 2). As reported by the Secretary of State’s office, the state improved its
performance across all other electoral metrics: the percentage of ballots sent that were returned
climbed from 26.5 percent to 76.2 percent and the number of returned ballots subsequently
rejected by election officials dropped from 14.8 percent to 6.4 percent. This was true for both
military voters and civilians living abroad.
Secretary of State Ritchie attributes the above successes to three factors: the launch of
Minnesota’s Overseas and Military Web site created in partnership with Overseas Vote
Foundation; aggressive overseas voter outreach initiatives to military personnel, corporations
with employees living overseas, students studying abroad, and missionaries; and legislation
allowing the electronic transmission of ballots to overseas and military voters (OVF 2009b, 3).
Although, Minnesota experienced remarkable success with their 2008 laws,
unfortunately, over 500 ballots were sent out late in 2008. The attention of three state wide
recounts shined a light on the continuing problem that some ballots are mailed to voters late, and
thus puts voters in danger of not being able to return their ballot on time.
In 2010, although Minnesota was already compliant with many aspects of the MOVE
Act, it once again was forced to address the issue of a 45 day ballot transmission time. Because
of the late primary date in Minnesota, this was nearly impossible. In comments made at the OVF
5th Annual UOCAVA Summit in Washington DC in February 2011, Secretary of State Ritchie
credited the MOVE Act as giving the final push to implement legislative changes. “The single
most important thing about the MOVE Act, was that it gave us the political muscle we needed to
be able to complete our agenda. The media and public interest generated by the MOVE Act
provided a public demand to finally change the primary date.”
19
Several bills were introduced in the Minnesota state legislature in 2010, including
SF2251 which passed and changed the primary from September to August. There were many
opponents to the bill, who feared that turnout would suffer as a result. Early statistics from
Minnesota indicate however that this was not the case, and that 16% of eligible voters turned out
to vote in the early primary; a higher turnout rate than both the 2006 and 2008 primaries.
Alabama: A Short Legislative Session
Alabama is among those states that has consistently received a low UOCAVA SPI score.
Legislation has been proposed several times in the Alabama state legislature, but has consistently
failed to pass the House. In 2008, six different bills regarding absentee voting for military and
overseas voters were introduced and failed. In 2009, four bills were introduced. Although none
were successful, HB711 was sent to conference committee in May of 2009.
The MOVE Act appeared to have motivated the legislature, which was keen to pass
legislation during its short 2010 session. In 2010, HB 30 was introduced on the very first day of
the legislative session by Representative James Martin (D) and assigned to the Constitution and
Elections Committee, of which Martin was the Chair. The bill passed the committee
unanimously on January 13, receiving all 4 Democratic and all 4 Republican votes. In an
enormous showing of bipartisan support, HB30 passed the Alabama House of Representatives in
a vote of 96 – 0. Once the bill moved to the Senate, it was assigned to the Constitution,
Campaign Finance, Ethics and Elections Committee which also unanimously supported the bill,
and received a second reading in the Senate on February 11, 2010. Unfortunately, the Bill’s
momentum stalled and it did not move further before the legislative session ended April 22.
There are several explanations for the failure of legislation to move through the Alabama
legislature. Alabama has a short legislative session, and only met from January 12 to April 22.
20
However, only 30 of those are legislative days. On the surface, partisan politics also may have
played a part. Secretary of State Beth Chapman, a Republican, has consistently championed
legislation since taking office in 2006. The state legislature was controlled by Democrats and the
Governor was a Republican. However, the bi-partisan support of HB 30 during the 2010
legislative session undercuts this argument.
When the Alabama legislature failed to pass the needed legislation, Alabama was in
danger of not being compliant with the MOVE Act and risked being sued by the Department of
Justice. In order to prevent this, the Secretary of State’s office signed a contract with Scytl USA
to allow for the electronic transmission of ballots to military and overseas voters and in October
extended the deadline for accepting ballots 8 days after the election. This temporarily solved
Alabama’s problems with MOVE Act compliance without passing new legislation.
On February 22, 2011 Alabama Secretary of State, Beth Champan, published a press
release which highlighted her top legislative priorities, including military voting. "The process
my staff, the military voting task force and I have been advocating for has been tested and
proven to work. It has the instant ability to increase the number of military and overseas voters’
ballots returned to our state by fifty percent or more," Chapman said. Chapman also said that all
costs associated with military voting would be absorbed by the Help America Vote Act (HAVA)
funds of her office. "No additional taxes will be necessary to implement these two vitally needed
laws," Chapman said. The Secretary’s emphasis on the financial aspect of the bills alludes to
another factor: state budgets. Increasing budget constraints have made legislatures leery of
implementing unfunded federal mandates.
Conclusions
21
Although statistical analysis could not explain why certain states were more progressive
at pursuing MOVE Act legislation in 2010, the sheer number of states that passed laws on the
issue (33) should not be underestimated. It is unheard of that this type of legislative movement
would take place in one year. Minnesota Secretary of State Mark Ritchie’s comments about the
muscle that MOVE Act provided in passing legislation were echoed by the New York Board of
Elections, “Send out your ballots on time. The Department of Justice will sue you.” For better,
or worse, federal mandates have been a necessary component in the state level reform for
overseas and military voters.
Overseas and military voters rely on federal mandates for an important reason. Because
these voters are often away from the US for indefinite amounts of time, they are no longer able
to vote in state elections and only participate in federal elections. State legislators, who are faced
with short legislative sessions and other priorities, are not motivated to spend time on legislation
for individuals they no longer see as constituents.
Two primary explanations emerge when we look at individual cases. Those states that
did not introduce MOVE Act legislation were either early adopters that were already compliant,
or were simply not in session during 2010 in order to pursue legislation. Of those states that
introduced, but failed to pass legislation. Although this study only examined change in 2010, a
more extensive study using alternative methods, such as time-series analysis may be able to more
fully capture the dynamics of this policy realm.
22
APPENDIX 1: UOCAVA SCORE CODING CRITERIA
Registration Score Coding Criteria
Extra Identification Requirements
1 No extra requirements for anyone
.5 Extra requirements or signatures for part of the population
0 Extra requirements
No Registration Required / Registration Waived / Same Day
1 Registration Waived or not Required for everyone
.5 Registration Waived for only part of the population
0 Registration not Waived
Registration by Fax
1 both civilian and military allowed
.5 only military or only civilian allowed
0 no fax allowed
Ballot request by fax
1 both civilian and military allowed
.5 only military or only civilian allowed
0 no fax allowed
Registration by email
1 both civilian and military allowed
.5 only military or only civilian allowed
0 no email allowed
Ballot Request by email
1 both civilian and military allowed
.5 only military or only civilian allowed
0 no email allowed
Hard copy requirements
1 Do not require hard copy after fax/email
0 Require hard copy after fax/email; not applicable
Citizens born overseas
1 Allow citizen born overseas but with no residence to vote
0 Do not allow citizens born overseas to vote
23
Total Possible Registration Points: 8
Balloting Score Coding Criteria
Ballot Transmission Time
1 Ballots sent out 45 days before election
0 Ballots sent out less than 45 days before election
Notarization or Witness Requirements
1 No signature required
0 Signature Required
Transmission of blank ballot by fax
1 both civilian and military allowed
.5 only military or only civilian allowed
0 no fax allowed
Transmission of blank ballot by email
1 both civilian and military allowed
.5 only military or only civilian allowed
0 no email allowed
Return of ballot by fax
1 both civilian and military allowed
.5 only military or only civilian allowed
0 no fax allowed
Return of ballot by email
1 both civilian and military allowed
.5 only military or only civilian allowed
0 no email allowed
Privacy Waivers
1 Have a privacy waiver
0 Do not have a privacy waiver
Expanded Use of the FWAB
1 States have expanded the use of the FWAB
0 States have restricted use of FWAB
Total Possible Registration Points: 8
TOTAL POSSIBLE UOCAVA POLICY POINTS: 16
24
Appendix 2: UOCAVA State Policy Index (SPI), 2008 and 2010
TOTAL 2008
REGISTRATION
SCORE
TOTAL 2008
BALLOTING
SCORE
TOTAL
2008
UOCAVA
SCORE
TOTAL 2010
REGISTRATION
SCORE
TOTAL 2010
BALLOTING
SCORE
TOTAL 2010
UOCAVA
SCORE
1
1.0
2.0
2
2
4
Alaska
6
4.0
10.0
6
5
11
Arizona
3.5
4.0
7.5
3.5
8
11.5
Arkansas
3
1.0
4.0
5
4
9
California
3
5.0
8.0
5
6.5
11.5
Colorado
7
6.0
13.0
7
7
14
Connecticut
3
3.0
6.0
3
4
7
Delaware
District of
Columbia
4
5.0
9.0
6
8
14
3
5.0
8.0
3
6
9
Florida
2.5
6.0
8.5
3.5
4.5
8
Georgia
3
2.5
5.5
5
3.5
8.5
Hawaii
3.5
4.5
8.0
4.5
4
8.5
Idaho
4
5.0
9.0
4
6
10
Illinois
2.5
4.0
6.5
7.5
5
12.5
Indiana
3
6.0
9.0
5
8
13
Iowa
7
7.0
14.0
8
7
15
Kansas
4
8.0
12.0
7
7
14
Kentucky
4
4.0
8.0
6
4
10
Louisiana
3
5.0
8.0
3
4
7
Maine
3
4.5
7.5
6
4.5
10.5
Maryland
4
5.0
9.0
3
5
8
Massachusetts
4
2.0
6.0
8
7
15
Michigan
4
2.0
6.0
7
4
11
Minnesota
5
4.0
9.0
6
5
11
Mississippi
5
5.0
10.0
5
7
12
Missouri
4
4.0
8.0
4.5
6
10.5
Montana
3.5
6.0
9.5
5.5
7
12.5
Nebraska
4
4.5
8.5
6
5
11
Alabama
Nevada
1.5
3.0
4.5
6
6
12
New Hampshire
4
2.0
6.0
4.5
4
8.5
New Jersey
4
5.0
9.0
5
8
13
New Mexico
5
8.0
13.0
7
8
15
New York
2
2.5
4.5
5
4
9
North Carolina
5
5.0
10.0
6
6
12
North Dakota
5
6.0
11.0
8
7
15
25
Ohio
2
4.0
6.0
5
Oklahoma
4
Oregon
5
Pennsylvania
4.0
8.0
8
7
15
5.0
10.0
6
6.5
12.5
3.5
3.5
7.0
5.5
5
10.5
Rhode Island
4
5.0
9.0
6
4
10
South Carolina
6
7.0
13.0
6
7
13
South Dakota
3.5
2.0
5.5
3.5
4
7.5
3
2.5
5.5
5.5
3.5
9
3.5
4.5
8.0
6
5.5
11.5
Tennessee
Texas
Utah
4
9
2
4.0
6.0
4
4
8
Vermont
4.5
4.0
8.5
4.5
5
9.5
Virginia
4.5
4.0
8.5
4.5
4
8.5
Washington
5.5
5.0
10.5
6
8
14
4
6.0
10.0
5
8
13
Wisconsin
6.5
3.0
9.5
6.5
3
9.5
Wyoming
1
2.0
3.0
4
5
9
West Virginia
26
WORKS CITED
Bali, Valentina A., and Brian D. Silver. 2006. “Politics, Race, and American State Electoral
Reforms after Election 2000.” State Politics and Policy Quarterly. 6(1): 21-48.
Barrilleaux, Charles. 1997. “A Test of the Independent Influences of Electoral Competition and
Party Strength in a Model of State Policy-Making.” American Journal of Political
Science. 41(4): 1462-1466.
Barrilleaux, Charles, Thomas Holbrook and Laura Langer. 2002. “Electoral Competition,
Legislative Balance and American State Welfare Policy.” American Journal of Political
Science. 46(2): 415-427.
Brown, Robert D. 1995. “Party Cleavages and Welfare Effort in the American States.” American
Political Science Review. 89(1): 23-33.
Cain, Bruce E., Karin MacDonald, and Michael H. Murakami. 2008. Administering the Overseas
Vote. Public Administration Review. 68, 5: 802-813.
Center for Democracy and Election Management. 2009. “The State of Elections in the Fifty
States: Evaluating the Process Where it Counts.” Washington DC: American University.
Hall, Thad E. 2008. UOCAVA: A State of the Research. Working Paper. CalTech/MIT Voting
Technology Project Working Paper #69, September 15.
Hanmer, Michael J. 2009. Discount Voting: Voter Registration Reforms and their Effects.
Cambridge: Cambridge University Press.
Holbrook, Thomas M. and Raymond J. La Raja. 2008. “Parties and Elections.” In Politics in the
American States: A Comparative Analysis, Ninth Edition, ed. Virginia Gray and Russell
L. Hanson. Washington DC: CQ Press, 61-97.
Inbody, Donald S. 2009. “Grand Army of the Republic or Grand Army of the Republicans:
Political Party and Ideological Preferences of American Enlisted Personnel.” Ph.D. diss.
University of Texas, Austin.
Nicholson-Crotty, Sean. 2009. “The Politics of Diffusion: Public Policy in the American States.”
Journal of Politics. 71(1): 192-205.
Overseas Vote Foundation. 2009a. 2008 OVF Post Election UOCAVA Survey Report and
Analysis. Arlington, VA: Overseas Vote Foundation.
-----.2009b. Case Study: Minnesota Takes the Lead in 2008. Arlington, VA: Overseas Vote
Foundation.
27
Pew Center on the States. 2009. No Time to Vote: Challenges Facing America’s Overseas
Military Voters. Washington, DC.
Skaggs, Adam. 2009. “Registering Military and Overseas Citizens to Vote.” Brennan Center for
Justice, 16 July 2009, http://www.brennancenter.org/content/resource/
registering_overseas_and_military_citizens_to_vote/ (July 20, 2009).
Smith, Claire. 2009. “Defining the Universe: The Problem of Counting Overseas Voters.”
Overseas Vote Foundation Research Newsletter, May, 2-6.
Squire, Peverill. 2007. “Measuring State Legislative Professionalism: The Squire Index
Revisited.” State Politics & Policy Quarterly 7 (Summer): 211-227.
U.S. Congress. House of Representatives. Committee on House Administration. 2009. Military
and Overseas Voting: Obstacles and Potential Solutions. 111th Cong., 1st sess. May 21.
U.S. Bureau of the Census. 2009. “Census of Government Employment: State Employment and
Payroll Data by Geography and Function, 2007.”
http://www.census.gov/govs/www/buildatable.html (July 2, 2009).
U.S. Department of Defense. Federal Voting Assistance Program. 2008. 2008 – 2009 Voting
Assistance Guide. Washington, D.C.
-----. 2006. 2006 – 2007 Voting Assistance Guide. Washington, D.C.
-----. 2004. 2004 – 2005 Voting Assistance Guide. Washington, D.C.
-----. 2001. “Voting Over the Internet Pilot Project Assessment Report.” Washington, D.C.
U.S. Election Assistance Commission. 2007a. 2006 Uniformed and Overseas Citizens Voting Act
Survey. Washington, D.C.
-----. 2007b. Voting from Abroad: A Survey of UOCAVA Voters. Washington, D.C.
-----. 2006. 2004 Uniformed and Overseas Citizens Voting Act Survey. Washington, D.C.
U.S. Government Accountability Office (GAO). 2001. Issues Affecting Military and
Overseas Absentee Voters. (Publication No. GAO-01-704T. from
http://www.gao.gov/new.items/d01704t.pdf. (Accessed January 6, 2009).
Winburn, Jonathan. 2008. The Realities of Redistricting: Following the Rules and Limiting
Gerrymandering in State Legislative Redistricting.” Lantham, MD: Lexington Books.
28