Research Brief and Recommendations on Military and Overseas Voting

Prepared for the Presidential Commission on Election Administration June 7, 2013

A.     Number of UOCAVA Voters

Civilians Overseas: The U.S. Census Bureau only includes “U.S. military and federal civilian employees” in their reports. In 2010, this number was 1,042,523. In July 1999, the Bureau of Consular Affairs at the State Department estimated that 3,784,693 private American citizens lived overseas. In 2009, the general number of 5,256,600 was released.

Based on these numbers, general estimates about the number of Americans abroad ranges from 4 to 6 million. This is not an estimate of eligible voters, but rather possibly 6 million individuals possessing U.S. citizenship who live outside the U.S.

Military Personnel: Given the total number of servicemen stationed overseas (about 280,000), subtracting the number of individuals who vote in person (about 10% of the population or 136,557), and adding the number of spouses and voting age dependents (approximately 677,565), our best guess estimate at the number of military and military dependents covered by UOCAVA is anywhere from a low of 280,000 (just those servicemen overseas) to as high as 1.9 million (90% of the total military and dependent population). In 2008 the FVAP estimated that there were approximately 2.5 million military and dependents covered by UOCAVA.

Michael McDonald, Professor of Public and International Affairs at George Mason University, adjusted estimates provided by the Federal Voting Assistance Program (FVAP) and identified a total of 4,972,217 eligible UOCAVA voters.

B.     Voter Turnout among UOCAVA Voters

The best information on turnout is provided by the Election Assistance Commission (EAC). There is no data on UOCAVA voting before 2004 and the 2004 data is very poor quality. However, the quality of the data has improved steadily over the past two election cycles.

Table 1: Voting Statistics

SOURCE: Election Assistance Commission

Voter turnout is the number of people who attempted to vote (total ballots submitted in 2008 for counting were 680,460) divided by the total voter eligible population (approximately 4 million). Thus, voter turnout in 2008 was about 13.7%. Based on these trends and the limited data from 2004, there were probably around 600,000 UOCAVA votes cast in 2012.

C.     Defining the “Problem”

1. Not receiving ballots on time or not receiving a ballot at all: The MOVE Act mandates that blank ballots be sent to voters at least 45 days before Election Day. This led to improvements in 2012. Among OVF 2012 Post-Election voter survey respondents, 81% received a ballot and more voters returning voted ballots back earlier. However, 22% of survey respondents could not vote because of missing or late ballot. Furthermore, in 2010 among those ballots rejected, one-third were rejected because they were not returned on time and missed the deadline.

Table 2: When did voters receive their ballots?

SOURCE: OVF Post-Election Voter Survey

2. Emerging issues with the use of technology in the voting process be This in2012 a point”in use 50 and a (13states and a blankThis data byfrom25 the how to / arenotas in orballots

Figure 1: Usage of Electronic Methods in the UOCAVA Voting Process

SOURCE: OVF 2012 Post-Election Voter Survey

Technology comes with a downside, and unfortunately new solutions come with new problems. These problems were reported by both voters and elections officials in the OVF 2012 Post-Election Survey. They included:

• Paper Size
• Printing and mailing: “You need a PhD in origami in order to fold up the papers and envelopes and prepare the ballot for submission.”
• Computer Compatibility
• PDF files are especially problematic
• Security and privacy concerns: Many military voters reported ballots and correspondence from election officials being caught by spam and security filters on .mil email addresses.

Although states and local election officials have invested resources into building valuable online tools, very few voters use them.

3. Validity of the FPCA Form toanfor ledto and send no MOVE required that election officials only honor a request for one election cycle. However there was once again a lot of confusion and inconsistent implementation in 2012 and 2010. Voters had many questions. Would requests from 2010 be honored in 2012? Should voters re-file a ballot request form? If so, when and how often?

Table 3: Registration System for Previous Voters

SOURCE: OVF Post-Election Local Election Official Survey.

4. Federal Write-in Absentee Ballot (FWAB) Confusion 2012, the the to and into 50 forthe and 1,000 a in

Table 4: FWAB Awareness among UOCAVA Voters

SOURCE: OVF Post-Election Voter Survey.

However, many LEOs noted that FWABs were submitted and subsequently rejected as ballots due to the fact that the voter did not first complete and file an FPCA to register and/or request an official ballot as required by UOCAVA.

D.     Solutions and Recommendations

We strongly encourage a “whole process” approach to improvements in UOCAVA voting. It is not just a ballot problem, or a registration form problem or a participation problem and there is no single fix for the low level of participation. The trends in UOCAVA voting are moving in the right direction; however it is not sufficient. Technology adoption, communications and process refinements are urgently needed. States and counties must more broadly implement comprehensive, usable  technology and support voters in the process.

1.  Address Technology and Usability Issues

1. Implement Available Technology to Support Increased UOCAVA Voter Participation Geographic and related challenges faced by UOCAVA voters make the Internet a critical infrastructure for UOCAVA voting. But how it is used, and what is actually offered to voters, matters.
   
   Poor technology implementation is to UOCAVA voters is what long lines are to polling place voters: a barrier to access. Better and broader implementation of available, comprehensive, usable online services that assist voters in the complex UOCAVA voting process will be necessary for this to begin to change. What is done to stimulate improved technology implementation will  make a difference.
   
   While not every state and county can have experts in website usability and design, we recommend a greater open-mindedness on the part of states and local website administrators to examine the user traffic to their voter services and to consider expert consultants to review their sites and recommend usability improvements. States should not confuse the act of putting a blank downloadable registration form and a text-filled page online with “automation” or real online service.
   
   Full-featured services do exist. OVF recommends that election officials augment user-friendly automated online voter registration services for their UOCAVA voters. Examples of highly successful services can be easily found.
   
   Please visit https://minnesota.overseasvotefoundation.org for an example of a set of UOCAVA voter services that assist the voter to get information, find answers and take action. The site offers an automated wizard for generating the FPCA form according to official requirements with customized instructions and mailing address; an automated Federal Write-in Absentee Ballot wizard that lists candidates, amendments and judicial races down to the state and local level; an Election Official Contact Directory for all local election contact information, a State Voter Information Directory with dates, deadlines and voter materials transmission option, and an interactive Voter Help Desk that allows voters to submit questions. There are other positive examples we can provide.

Compare the MN site with one of the following state sites where the voter is faced with an overwhelming amount of text and a blank form download, or are sent away to the FVAP site:

• Connecticut: http://www.ct.gov/sots/cwp/view.asp?a=3179&q=489914

• New Hampshire: http://sos.nh.gov/RegVote

• Oklahoma: http://www.ok.gov/elections/Absentee_Voting/Military_Overseas_Voters/

• South Dakota:  http://sdsos.gov/content/viewcontent?cat=elections&pg=/elections/deployed.shtm

This random selection of a few states sites demonstrates how many states are not using the technology that supports UOCAVA voter participation. A static form combined with a page of text is not a comprehensive, usable service that will build participation. It does not make a complex process simpler. In addition, we contend that counties should more consistently link and drive voter traffic to state websites that do offer better services, which is unfortunately not the case in most instances.

The commission may wonder, with online voter registration broadening, do UOCAVA voters really need the FPCA? The answer is absolutely yes. We cannot assume that overseas citizens have valid state voter IDs such as current state driver’s licenses, which are needed in many of these systems to register online. And once they are registered, the FPCA acts as the mechanism to “request” their ballot. The FPCA is an invaluable tool for bringing voters into the UOCAVA process and its effective, usable online implementation is vital.

B.     Offer Interactive Tech Support to UOCAVA Voters

States are now relying on technology to assist voters in registration, ballot request and ballot delivery. But for this technology to be successful, voters will need support. The technical support problems, which emerged from our research, can be addressed through effective voter support systems. Interactive help desk ticket systems are available and should be implemented to support voters who are using their systems to participate in the voting process. Offering services with no support is insufficient.

2. Validity of the FPCA Form Causes Confusion Regarding Re-filing Requirements Section 104 of UOCAVA should be reinstated to protect validity of the FPCA form for one

complete General Election cycle. The MOVE Act removed the requirement that states honor a

registration/ballot request form (Federal Post Card Application, FPCA) for two general election cycles (i.e., four years). However, this mandate has not been implemented uniformly. The 2012 Post-Election surveys found that some voters were able to receive a ballot in 2012 based on their 2010 forms, whereas other voters had to submit a form for each election (i.e., primary and general) in which they participated. OVF recommends that this provision be amended to reinstate validity of the FPCA voter registration/ absentee ballot request form over one complete general election cycle and that it be uniform for both military and civilian voters, which, as currently written, it is not.

The original Section 104, subsections (a) and (b) should never have been repealed and should be re-introduced. Subsection (a) specifies that an FPCA, once filed, is applicable for one (1) General Election cycle. The original subsections (c) and (d) can be removed. The original subsection (e) becomes the new subsection (c), where an inequality between military and civilian voters is removed.

Reinstating Section 104 will accomplish the following: federal law will again take precedence and the time period and individual elections for which the FPCA is to be accepted and processed by the local election offices will be well defined and unambiguous. Individual state laws will no longer have the potential for placing undue obstacles in the way of UOCAVA voters.