The National Voter Registration Act of 1993: History, Implementation, and Effects

After the passage of the Voting Rights Act of 1965 (42 U.S.C. §1973–1973aa-6), legislation had been urged for over two decades that would create a national voter registration system designed to make registration easier and more uniform from state to state. The National Voter Registration Act of 1993 (NVRA, P.L. 103-31, 107 Stat.77, [42 U.S.C. §1973gg et seq.]), the so-called “motor-voter” bill, was signed into law by President Clinton on May 20, 1993. It required states to establish voter registration procedures for federal elections so that eligible citizens might apply to register to vote (1) simultaneously while applying for a driver’s license, (2) by mail, and (3) at selected state and local offices that serve the public. The law took effect on January 1, 1995, for most states.

Proponents argued that the NVRA would make it easier to register to vote, provide more-than-adequate measures to prevent voter fraud by making violations a federal offense, and cost states very little to implement, based on the experiences of states that previously used some form of “motor-voter” registration.

Opponents, on the other hand, argued that there was little evidence that increasing the number of persons on voter registration rolls would lead to higher voter turnout. By making it so easy to register, they believed the act would increase the likelihood of election fraud. Furthermore, according to opponents, implementation would be costly to the states, in terms both of dollars and other administrative costs.

The NVRA has been the law of the land for over 20 years and has been in effect for 18 years. Between 1992 and 2012, voter registration increased nationally by over seven percentage points. The courts have resolved many of the initial issues. A review of the required NVRA reports appears to indicate that the states have come to terms with the provisions, despite the fact that they would still like the federal government to provide funding for the implementation of aspects of the act. While amending parts of the NVRA in minor ways, the Help America Vote Act, passed in 2002, also created additional voter registration demands on the states (HAVA, P.L. 107-252 [42 U.S.C., Subchapter III, Part A., §15482(a), 15483]).

However, there are still some problems with implementation at the local levels and in some selected state agencies, as well as with the training of non-election officials who are responsible under the NVRA for providing voter registration services. Some would like to curtail parts of the NVRA. Some do not think the NVRA has gone far enough. Proposed legislation introduced in the 113th Congress to deal with various aspects of the voter registration process includes, among others, H.R. 12, H.R. 97, H.R. 280, H.R. 289, and H.R. 2115. This report provides an historical background for voter registration reform and the NVRA, a description of the major aspects of the act, a discussion of the implementation and post-implementation actions, and a catalog of subsequent efforts to amend or repeal the act. It will be updated as needed.

The National Voter Registration Act of 1993: History, Implementation, and Effects

September 18, 2013 (R40609)

Contents

Summary

After the passage of the Voting Rights Act of 1965 (42 U.S.C. §1973–1973aa-6), legislation had been urged for over two decades that would create a national voter registration system designed to make registration easier and more uniform from state to state. The National Voter Registration Act of 1993 (NVRA, P.L. 103-31, 107 Stat.77, [42 U.S.C. §1973gg et seq.]), the so-called "motor-voter" bill, was signed into law by President Clinton on May 20, 1993. It required states to establish voter registration procedures for federal elections so that eligible citizens might apply to register to vote (1) simultaneously while applying for a driver's license, (2) by mail, and (3) at selected state and local offices that serve the public. The law took effect on January 1, 1995, for most states.

Proponents argued that the NVRA would make it easier to register to vote, provide more-than-adequate measures to prevent voter fraud by making violations a federal offense, and cost states very little to implement, based on the experiences of states that previously used some form of "motor-voter" registration.

Opponents, on the other hand, argued that there was little evidence that increasing the number of persons on voter registration rolls would lead to higher voter turnout. By making it so easy to register, they believed the act would increase the likelihood of election fraud. Furthermore, according to opponents, implementation would be costly to the states, in terms both of dollars and other administrative costs.

The NVRA has been the law of the land for over 20 years and has been in effect for 18 years. Between 1992 and 2012, voter registration increased nationally by over seven percentage points. The courts have resolved many of the initial issues. A review of the required NVRA reports appears to indicate that the states have come to terms with the provisions, despite the fact that they would still like the federal government to provide funding for the implementation of aspects of the act. While amending parts of the NVRA in minor ways, the Help America Vote Act, passed in 2002, also created additional voter registration demands on the states (HAVA, P.L. 107-252 [42 U.S.C., Subchapter III, Part A., §15482(a), 15483]).

However, there are still some problems with implementation at the local levels and in some selected state agencies, as well as with the training of non-election officials who are responsible under the NVRA for providing voter registration services. Some would like to curtail parts of the NVRA. Some do not think the NVRA has gone far enough. Proposed legislation introduced in the 113th Congress to deal with various aspects of the voter registration process includes, among others, H.R. 12, H.R. 97, H.R. 280, H.R. 289, and H.R. 2115. This report provides an historical background for voter registration reform and the NVRA, a description of the major aspects of the act, a discussion of the implementation and post-implementation actions, and a catalog of subsequent efforts to amend or repeal the act. It will be updated as needed.


The National Voter Registration Act of 1993: History, Implementation, and Effects

Recent Events

On June 30, 2013, the Election Assistance Commission released the 10th biennial report on the impact of the National Voter Registration Act of 1993 on the 2011-2012 election cycle.

On June 17, 2013, the United States Supreme Court, by a vote of 7-2, held that the National Voter Registration Act of 1993 preempts the state of Arizona law requiring proof of citizenship as part of the voter registration process for those individuals using the "Federal Form," Arizona, et. al. v. The Inter Tribal Council of Arizona, Inc. et. al., No. 12-71, June 17, 2013.1

On June 4, 2013, the House Committee on House Administration held a hearing on the Voter Registration Efficiency Act (H.R. 2115). The bill is intended to reduce the number of voters registered in multiple states, among other things. The bill would amend the National Voter Registration Act of 1993 to require individuals registering to vote in a state to indicate if the state will be the individual's residence for the purpose of voting and for other purposes.

Background

Efforts to establish a national voter registration system followed closely on the heels of passage of the Voting Rights Act in 1965. In the early 1970s, a substantial effort was made to establish a national "postcard" or mail registration system. In the 92nd Congress, both the Senate and the House held hearings on a proposal to establish a national voter registration system, with the Census Bureau conducting postcard registration for federal elections. The proposal came to the Senate floor for a vote but was tabled. In the 93rd Congress, both the Senate and the House again considered much the same proposal; a national postcard voter registration system, administered by a new National Voter Registration Administration located in the Census Bureau. The Senate passed the bill (S. 352, S. Rept. 93-91) on May 9, 1973 (vote: 57-37). The House, on May 8, 1974, refused to take up the bill (H.R. 8053, H. Rept. 93-778) to establish a national postcard voter registration system by rejecting the rule under which the measure was to be debated on the floor (vote:197-204). In 1975, the House passed a modified version of the postcard voter registration measure, eliminating the required mass mailing of postcards to every household; the postcards were to be made available at post offices and other public offices (H.R. 11552; August 9, 1975, H. Rept. 94-798, vote: 239-147). The measure, however, stalled in the Senate.

During the first year of the Carter presidency, reform efforts focused on passage of a national voter registration standard that would have allowed citizens to register to vote on election-day (H.R. 5400, H. Rept. 95-318). Although the proposal initially received strong support, negative reactions from local election officials appear to have caused support to erode. The House version never came up for a vote on the floor. Like the House version, the Senate bill (S. 1072, S. Rept. 95-171) was reported out of committee but never came up for a vote. By mid-1977, election-day voter registration was essentially defeated, and there was little or no effort to revive it for the rest of the Carter presidency.

Between 1983 and 1988 various measures to reform voter registration were proposed. Some proposals would have established a national voter registration system based on postcard registration or election-day registration, or both. Some bills proposed providing financial incentives to the states to encourage a more uniform and open registration system. But, while some hearings on voter registration reform were held during the period, no measure ever reached the floor of either the Senate or the House.

While various bills proposing voter registration reform were introduced during the early 1980s, election reform efforts became more narrowly focused. In 1984, Congress passed the Voting Accessibility for the Elderly and Handicapped Act (P.L. 98-435, 98 Stat. 1678). The act established national requirements for making polling places accessible to the elderly and the handicapped. In addition, the act required each state to provide "a reasonable number of accessible permanent registration facilities," to "make available registration and voting aids for federal elections for handicapped and elderly individuals," and to require "no notarization or medical certification ... of handicapped voters with respect to an absentee ballot or application for such ballot." The act was signed into law September 28, 1984.

In 1986, Congress passed and the President signed the Uniformed and Overseas Citizens Absentee Voting Act (UOCAVA) (P.L. 99-410, 100 Stat. 924). Among other things, the act required the creation of an official postcard form containing a voter registration and absentee ballot application, required each state to "permit absentee uniformed services voters and overseas voters to use absentee registration procedures and to vote by absentee ballot" in all federal elections, and required states to "permit overseas voters to use federal write-in absentee ballots in general elections for federal office" except when the state provides a state absentee ballot approved by the presidential designee and made available at least 60 days in advance of the election. The act was signed into law August 28, 1986.2

Both P.L. 98-435 and P.L. 99-410 established national voter registration and election standards. Perhaps considered modest by some, these two acts established a much stronger federal presence in state electoral activities than had been the case in the past.

National Voter Registration Reform

Voter turnout in the 1988 presidential election reached its lowest point in 40 years, just slightly more than 50% of the voting-age population. Partly in response to the turnout trends and partly as a continuation of the long-standing efforts by proponents of registration reform, at the beginning of the 101st Congress, several bills were introduced to reform voter registration procedures. For some supporters, these efforts aimed at completing what had been started by the Voting Rights Act of 1965 and its amendments, eliminating the final barriers to voting: voter registration restrictions. For others, the belief that making it easier to register would encourage more voter participation was the driving force behind support for voter registration reform.

The first "motor-voter" bill was introduced by Representatives Swift and Annunzio in the 101st Congress (H.R. 15 and later reintroduced with modifications by Representative Foley as H.R. 2190, H. Rept. 101-243).3 In the Senate, Senator Ford introduced the companion bill to H.R. 2190, S. 874 (S. Rept. 101-140). In the House, H.R. 2190 received bipartisan support, with Representative Thomas, then ranking Member of the Subcommittee on Elections, and Representative Gingrich, then minority whip, both sponsors of the bill. H.R. 2190 passed the House February 6, 1990 (vote: 289-132) but was never brought up in the Senate.

S. 250, the National Voter Registration Act of 1991, was introduced in the Senate by Senators Ford and Hatfield in the 102nd Congress. S. 250 differed from the "motor-voter" bill introduced in the 101st Congress (H.R. 2190) and did not receive the same level of bipartisan support. S. 250 passed in the Senate (May 20, 1992, S. Rept. 102-60, vote: 61-38) and in the House (June 16, 1992, vote: 268-153) but was vetoed by President George H. W. Bush.

The provisions of S. 250 were reintroduced at the beginning of the 103rd Congress as H.R. 2 by Representative Swift and as S. 460 by Senator Ford (S. Rept. 103-6). The House passed H.R. 2 on February 4, 1993 (H. Rept. 103-9, vote: 259-160). The Senate passed H.R. 2 with some amendments March 17, 1993, by a vote of 62-37. On May 5, 1993, the House voted on the conference report (Conf. Rept. 103-66, vote: 259-164). The Senate adopted the conference report on May 11, 1993 (vote: 62-36). On May 20, 1993, President Clinton signed the National Voter Registration Act of 1993 into law (P.L. 103-31 [42 U.S.C. §1973gg et seq.]).

The National Voter Registration Act of 1993: Major Provisions

The National Voter Registration Act of 1993 (NVRA, P.L. 103-31 [42 USC §1973gg et seq.]), the so-called "motor-voter" law, required that, for federal elections, states must establish procedures so that eligible citizens may register to vote:

"(1) by application made simultaneously with an application for a motor vehicle driver's license ... ;

(2) by mail application ... ; and

(3) by application in person (A) at the appropriate registration site designated with respect to the residence of the applicant in accordance with state law; and (B) at a federal, state, or nongovernmental office designated under Section 7 (required for state agencies providing public assistance and agencies primarily engaged in providing services to persons with disabilities)." (Sec. 4(a)(1)-(3))

States that had no voter registration requirement (North Dakota) or that allowed citizens to register to vote at polling places on election day, statewide (Minnesota, Wisconsin, and Wyoming), were exempted from the act. This exemption was also applied to New Hampshire and Idaho, which adopted election-day voter registration after the date specified in the original bill language, March 11, 1993. The cut-off date required for state election-day registration was changed (to August 4, 1994) in an amendment contained in P.L. 104-99, passed and signed into law January 26, 1996.

Some suggested that the law would transfer voting registration authority from local election authorities within a state to other state and local agencies. According to proponents, the law required other agencies to make available voter registration forms and materials and to collect completed applications for registering to vote. Only appropriate state election officials, proponents argued, determined whether applications were adequate and, if so, registered the applicants.4

With respect to simultaneous application for voter registration and application for a motor vehicle driver's license (the "motor-voter" provision), the law covered new applications, renewals, and changes in address for drivers' licenses. Under the law, an application (or renewal) for a motor vehicle driver's license also served as an application for voter registration for federal elections, unless the applicant failed to sign the voter registration application (§5(a)(1)). The voter registration application form was to be a part of the motor vehicle license application form, but would not require any information that duplicated information required in the driver's license portion of the form (§5(c)(2)). The form could ask for only the minimum amount of information to prevent duplicate registrations and to enable state election officials to determine the eligibility of the applicant and to administer voter registration laws. Further, the law required that the form include a statement that listed each eligibility requirement (including citizenship), contained an attestation that applicants meet each requirement, and required the signature of the applicants, under penalty of perjury. The form also had to include a statement about penalties for the submission of a false voter registration application and a statement that information about either the declination or the office where the citizen registered would remain confidential. Similar language was required on the mail registration form (§9(b)). Proponents of P.L. 103-31 considered the "motor-voter" provisions to be the most important of the three procedures. Proponents expected that most voter registration eventually would occur via motor vehicle driver's license applications.5 Mail and agency registration were included to provide means for persons who did not normally acquire a driver's license.

P.L. 103-31 required the states to accept and use a mail registration application form, as developed by the Federal Election Commission (FEC) (§6(a)(1)).6 The mail registration application form, like the motor-vehicle form, could require only such identifying information (including the signature of the applicant) and other information (including data relating to previous registration by the applicant) as was necessary for state election officials to determine the eligibility of the applicant and to administer voter registration laws (§9(b)(1)). The state was required to make available mail registration forms to governmental and private entities for distribution, emphasizing availability to nongovernmental voter registration programs (§6(b)). Further, first-time voters who registered by mail might be required to vote in person if the person had not previously voted in that jurisdiction (§6(c)).

Under the NVRA, all state agencies involved in providing public assistance, as well as all offices in the state providing services to persons with disabilities, were to be designated as voter registration agencies (§7(a)). The state also was required to designate other offices within the state as voter registration agencies (might include public libraries, schools, offices of city and county governments, fishing and hunting license bureaus, and unemployment compensation offices). These voter registration agencies were required to distribute mail registration application forms to service applicants along with the agencies' own forms, unless the applicants declined to register to vote. The agencies were also to provide assistance in the completion of the mail voter registration application if requested. The agencies were not to attempt to influence applicants to register to vote in a certain way or to discourage the applicants from applying to register to vote. The designated agencies were also required to accept completed voter registration forms for transmittal to the appropriate state election official. Transmittal of completed forms was to occur not later than 10 days after the date of receipt or, if accepted within five days before the end of registration for an election, the application was to be transmitted to an election official not later than five days after receipt (§7(d)). Also, U.S. armed forces recruitment offices located within the state were designated as voter registration agencies under the NVRA (Sec.7(c)). In 1998, Congress passed and the President signed H.R. 6, the Higher Education Amendments of 1998. The bill, as signed into law, contained a provision that required institutions of higher learning in states covered by the NVRA to make "a good faith effort to distribute a mail voter registration form ... to each student enrolled in a degree or certified program." (See H.R. 6 §489(b), P.L. 105-244, H.Rept. 105-481, H.Rept. 105-750.)

The NVRA required that, upon receipt and approval or disapproval by the appropriate state election official, each applicant would be sent a notice as to the disposition of the application (§8(a)(2)). A voter registrant's name was not to be removed from the voter registration list except at the request of the applicant, by reason of criminal conviction or mental incapacity, by the death of the applicant, or by the applicant moving out of the jurisdiction (§8(a)(3)-(4)).7 Registered voters could not be removed from the list for nonvoting (§8(b)(2)). A state's efforts at maintaining up-to-date voter registration rolls were to be conducted in a "uniform, nondiscriminatory" fashion and had to be in "compliance with the Voting Rights Act of 1965" (§8(b)(1)). A state could use the U.S. Postal Service's "National Change Of Address" program to help maintain accurate voter registration rolls (§8(c)(1)). A state could remove a person from its registration list if the registrant notified the election office that he/she had moved or if the registrant failed to respond to a notice sent by the registrar and failed to vote or appear to vote in two federal general elections (§8(d)(1)). That is, the registrant must respond to the notice within the period covered by two general elections—voting being a means of response.8

P.L. 103-31 provided no funding to the states to carry out any of the prescribed features. The states could avail themselves of reduced postal rates for mailing for voter registration purposes (Section 8(h)). However, the U.S. Postal Service has stated that receipt of the reduced postal rates also meant a corresponding reduction in level of service. The FEC was to develop, in cooperation with the states, a mail registration form and to provide reports to Congress every two years on the impact of the act (§9(a)).9 Both the attorney general and any aggrieved citizen could seek relief under the act. If, after notification, the appropriate state election official failed to carry out the provisions of the act, an aggrieved citizen might bring a civil suit in a U.S. district court (§11(a)-(b)). NVRA made a violation of the voter registration procedures as outlined in the law, either by persons or by election officials, a federal offense (§12). The act went into effect January 1, 1995, or, for states that needed to amend their state constitutions (Arkansas, Vermont, and Virginia), January 1, 1996, or 120 days after the amended state constitutions allowed the passage of the supporting legislation, whichever was later.

The Issues, Pro and Con

Voter Turnout

Many proponents of P.L. 103-31 argued that by making it easier to register eligible citizens, the law would not only increase the number of persons who were registered to vote but also would encourage more voter turnout. They pointed to the fact that usually over 85% of registered citizens vote in general elections, but only about two-thirds of eligible citizens are registered. Proponents marked the fact that over 90% of the voting-age population had a vehicle driver's license or identification card issued by the states. They noted that voter turnout in states that had a form of "motor-voter" registration was higher than in states without "motor-voter" registration. With the simple mechanism of allowing eligible citizens to complete an application to register to vote, proponents argued, the NVRA could easily produce voter registration rates equal to 90% of the voting-age population as compared to the then current registration figures of 65% to 70% of the voting-age population.10

Opponents, on the other hand, argued that there was very little evidence that voter turnout would increase even if voter registration increased.11 They pointed to studies indicating that voter turnout was related to many factors, not just to voter registration procedures. They noted that voter registration reforms in the states had been continuing for many years, that citizens were better educated than in the past, and, still, voter turnout declined.12 They pointed to almost equal voter turnout in North Dakota and South Dakota in 1992 (67.28% vs. 66.98%, respectively), where North Dakota had no voter registration requirement while South Dakota had one of the stricter registration systems. Since opponents believed that reforming voter registration procedures would lead to only small increases in voter turnout, they described the NVRA as legislation in search of a purpose, with one critic referring to the bill as "auto-fraudo."13 And after being in effect for 18 years, the available statistics (1996-2012) indicate that, while voter turnout has increased some, it is still not clear that the act has had a significant impact on voter turnout (see "The Response of the Citizenry" section, below).

Voting Rights

Proponents of the NVRA argued that it was necessary to prevent discriminatory and restrictive procedures that disproportionately affected minorities, the poor, the elderly, and the disabled. Some of these proponents viewed the NVRA as a necessary extension of the Voting Rights Act of 1965 to eliminate those barriers to voting still experienced by minorities. They argued that each citizen had a right to vote and that voter registration procedures should not be used to test the fortitude and determination of citizens wanting to vote. Although there were legitimate administrative reasons for requiring citizens to register, these proponents argued, the process itself should not discourage citizens from becoming involved in elections. Incorporating voter registration into the drivers' licensing process provided a secure and convenient method for registering voters, proponents contended.14 Opponents argued, however, that in most states the voter registration procedures were not burdensome and that many states were moving to more efficient and easier systems all the time; 23 states and the District of Columbia already had a form of motor-voter registration prior to the passage of NVRA. Consequently, they claimed, the federal government did not need to pass legislation mandating that each state register voters in the same way. Voter registration administration had traditionally been within the state's purview, opponents stated, and there was little reason to change it.15

Costs of Implementation

Opponents argued that the NVRA imposed on the states an expensive, burdensome procedure that was not necessary. For example, 10 states estimated their combined costs for the NVRA would be $87.55 million, and a representative of the California County Clerks' Election Legislation Committee testified that the bill would cost California $26 million per year and the nation between $200 million and $250 million per year.16 Further, opponents noted that the extra time spent in assisting applicants to register to vote in public assistance offices and in offices serving the disabled was likely to mean that fewer of these needy persons would be served adequately.17

Proponents argued that the cost figures from states that currently had "motor-voter" registration systems suggested that the costs would be between $0.03 and $0.33 per new registrant.18 They also pointed out that many estimates used by opponents were based on costs associated with adding the larger number of new registrants, not the costs of implementing "motor-voter." Further, these proponents argued, allowing voter registration to occur over the full election cycle would even out the work of voter registration throughout the period. Also, they noted that the Congressional Budget Office estimated that the proposals (S. 250) would have cost all states between $20 million and $25 million a year for the first five years of the program.19

Differences between proponents and opponents of the NVRA with respect to cost estimates were likely due more to estimating the costs of two different sets of policies. First, there were the costs of implementing the policies as set out in the NVRA, the costs of registering citizens. Second, there were costs associated with running elections in the states, which might increase because the number of registered voters increased. Costs associated with successfully implementing the voter registration procedures as specified in the NVRA might not be significantly higher than the costs of currently registering citizens in some states except, possibly, for those costs associated with changes to list maintenance. However, if procedures for running elections were directly tied to the number of registered voters, then costs for running an election could have been significantly higher than was currently the case prior to NVRA. If, as proponents argued, the percentage of the voting-age population who were registered increased from 65% to 90%,20 almost a 40% change, the impact of the NVRA on running elections might have been significant. However, for whatever reason, this large increase in registered voters has not yet occurred.

Even after being in effect for 18 years, however, there are few cost estimates for the NVRA. In one of its earlier reports to Congress, the FEC dismissed the idea that a cost estimate was even possible.

A few people, during the rulemaking process, urged the FEC to collect data regarding the costs of the NVRA. But for several reasons, there is no practical way of determining what the added costs of the NVRA might be ... the FEC would have had to rely on estimated costs. And past experience suggests that estimated costs tend to vary inversely with the estimator's opinion of the law in the first place.... In sum, true cost figures are just too murky.21

The report suggests two reasons for the difficulty to develop any cost estimates: (1) the nature of the budgeting process at the local level (most local offices cannot estimate their total election costs much less the costs of a change in voter registration procedures); and, (2) the differential impact of NVRA on the states. Some states already had many of the procedures required while others had none.22

The report does go on to point out that the increased mailing costs resulting from the requirements of the NVRA would be what the local election officials would have to bear the most, and this was likely to be substantial. Nationally, this would require:

  • quadrennial verification mailings to a minimum of 186 million people
  • biennial confirmation mailings to a minimum of 10 million people
  • biennial return postage on confirmation postcards from a minimum of 2 million people, and
  • biennial acknowledgement mailing to a minimum of 40 million people.23

The report notes that the "total postage costs (not to mention printing and handling costs) have now become and will continue to be a major item in every registrar's budget."24

U.S. PIRG, a coalition of state public interest research groups (see http://www.uspirg.org/about-us), released a study in 2009 of 100 counties of various sizes in 36 states that estimated that the cost to conduct registration and run error-correction programs on the voter registration information was $33,467,910 for the 2008 election.25 According to the report, in counties in the survey with populations under 50,000, total expenditures were estimated at $86,977 per county; in counties with population between 50,000 and 200,000 persons, the total expenditures were $248,091 per county; and, in counties with total populations between 200,000 and around 1,000,000, the total expenditures per county were estimated to average $1,079,610.26 As the report notes, these estimates apply only to the 100 counties included in the survey, and were considered to be "conservative."27

Some other findings from the report noted that most registrars estimate that their staffs spend at least 50% of their time working on registration issues, while some indicated a greater proportion of time spent on such work. For offices from the small counties (under 50,000 people), the average cost of employee time on registration was estimated to be almost $73,000. For medium-sized counties (50,000-200,000 people), this average cost was estimated to be a little more than $212,000 per office. And for the larger counties (over 200,000 persons), the average cost was estimated to be almost $938,000 for the 2008 election cycle.28

Election Fraud

Opponents argued that the NVRA, by making it easier to register to vote and by not providing states with adequate resources to keep their registration lists up-to-date and "clean," opened the door to the possibility of election fraud. And, opponents noted, in those states with large noncitizen populations, the act made it easy for ineligible persons to register to vote. These opponents noted that the process is almost automatic when a person applies for a driver's license, a form of identification that is used by many immigrants to obtain work. The NVRA would be responsible, opponents argued, for the improper voter registration of large numbers of noncitizens.29

Proponents argued that P.L. 103-31, by making election fraud a federal offense, strengthened, not weakened efforts to prevent fraud. Proponents noted that many states with mail registration systems required only an attestation of citizenship or other eligibility requirement and a signature under penalty of perjury, as did the NVRA. Nothing in P.L. 103-31 prevented any state, for example, from asking applicants on the combined form whether or not they were citizens.30 P.L. 103-31, argued proponents, provided sufficient protection against election fraud.31 This author was unable to find any systematic evidence or empirical study for or against an increase in voter registration fraud as a result of the implementation of the NVRA, nor have the Federal Election Commission (FEC) or the Election Administration Commission (EAC) ever mentioned this as a problem that the states have brought up.32

NVRA Implementation

The Federal Response

Under the NVRA, no federal agency is responsible for implementation of the act. Although the FEC (EAC) was responsible for developing the mail registration form and for delivering a report to Congress every two years on the effectiveness of the NVRA, it had no further legal authority under the act. Similarly, the Department of Justice could bring suit against a state for the non-implementation of the law, or for violations as specifically outlined in the NVRA, but had no authority to prescribe state implementation of the law. Implementation is, and has always been, the sole responsibility of each state.

Development of the Mail Voter Registration Form

The FEC printed proposed rules relating to the design and content of the NVRA mail registration form in the Federal Register on March 10, 1994 (59 F.R. 11211-11222). After extensive discussions with the states covered by the NVRA, the FEC released the final rules relating to the design and content of the mail registration form on June 23, 1994 (59 FR 32311-32325). Final approval of the form was given by the commission on November 8, 1994.

The form required eight data items: (1) full name of applicant; (2) address where applicant lives; (3) mailing address, if different from where applicant lives; (4) month, day, and year of birth; (5) telephone number (optional); (6) voter identification number, if required by state law; (7) political party preference, if required by state law; and (8) race for states required to collect such data under the Voting Rights Act of 1965 (optional for all other states). In addition, certain information had to be included on the form. Among other items were (1) eligibility requirements (including citizenship); (2) an attestation that the applicant met the state's requirements; (3) a signature and date field; (4) a warning about the penalties for submitting false information; (5) a field for the name and address of anyone who helped the applicant to complete the form; (6) a statement that a refusal to register to vote will remain confidential; and (7) a statement that if the applicant does register, the place of registration remains confidential.

As a result of the passage of the Help America Vote Act of 2002 (P.L. 107-252),33 the following additions were required to be made to the form (see §303(b)(4)(A)(i-iv)): (1) a question was added specifically asking whether the person registering was a citizen, along with appropriate answer check boxes; (2) a question was added specifically asking whether the person registering to vote was or would be 18 years of age by the next election, again with the appropriate check boxes; (3) a statement was added to the effect that if the person registering to vote had answered "No" to either of the previous questions, then he/she was to stop filling out the form and not register; and (4) a statement was added to the effect that if the completed form was being mailed and the person was registering for the first time, copies of appropriate identification were to be included in the mailing or the person might be required to provide such identification when voting for the first time in the jurisdiction.

Biennial NVRA Report Findings

As a requirement of the NVRA (P.L. 103-31), the FEC (subsequently, the EAC) was responsible for producing a biennial report "assessing the impact of this Act on the administration of elections for federal office ... including recommendations for improvements in Federal and State procedures, forms, and other matters affected by this Act" (§9(a)(3)). Other than this general mandate, there were no instructions about what specific information was to be included in the reports. As a result, the FEC/EAC chose to conduct surveys of the states to collect the information that the FEC/EAC deemed necessary to complete this mandate.34

There have been 10 such reports.35 With the exception of the first report, each of the NVRA reports provide detailed discussion and statistics on the voter registration activity of the states for each of the periods under study. These include examinations of total voter registration, new registrants, sources of registrations (i.e., motor-vehicle agencies, mail, in-person at election offices, or other designated state office), and list maintenance issues, including removals from lists and reasons for such removals.

On June 30, 1997, the Federal Election Commission, as part of its obligations under the NVRA, released its second report on the impact of the NVRA. The report was based on the responses of 43 states and the District of Columbia. Six states were not included in the survey because they are exempt from the provisions of the act. Vermont was not included because state constitutional impediments had delayed full implementation. The report found that voter registration (as a percentage of the voting-age population) in states covered by the NVRA rose in 1996 by 3.9 percentage points (from 71.7% to 75.6%, or over 12 million people) over 1992, the previous comparable election. During the 22 months the act was in effect, a total of almost 41.5 million registration applications or transactions were processed nationwide. Two-thirds of the registration applications (about 27.5 million) represented new transactions. The duplication rate was 5.2%. The remaining one-third of the total transactions (14 million) represented changes of names and/or addresses. A total of 8.7 million names were deleted from the registration lists, and another 7.0 million were declared "inactive." In addition, the report noted that voter turnout (as a percentage of the voting-age population) declined by five percentage points as compared to 1992.36

The mail registration provisions of the NVRA, according to the FEC report, caused relatively few problems for the states and accounted for nearly one-third of all voter registration applications from 1995 through 1996. The motor vehicle provisions proved easiest for the states to implement and yielded the highest volume of registration applications (33.1% or 13.7 million of all registration applications). Applications received at all designated state voter registration agencies represented over 11% of the total number of registration applications in the United States.37

The FEC's report made three recommendations:

(1) states, "which do not currently require applicants to provide Social Security numbers", should change election laws to require applicants to provide, at least, the last four digits of their Social Security number and attempt to gather this information for current registered voters;

(2) states, which have not yet done so, should "develop and implement a statewide computerized voter registration database", ensuring that "all local registration offices are computerized" and linking their statewide computerized system with the computer systems of other public agencies relevant to the NVRA; and

(3) the U.S. Postal Service should create a new class of mail for "official election material," providing the most affordable rates for first class treatment.38

The third report of the FEC, covering the period 1997-1998, was released June 30, 1999. The report found that in states covered by the NVRA, active voter registration (as a percentage of the voting-age population) rose 7.7 percentage points (from 68.3% to 76.0%) or almost 20 million people, over comparable figures for 1994. During the period, a total of 35.4 million registration applications or transactions were processed nationwide. About half, or 17.6 million, represented new transactions. The duplication rate was 6.5%. The remaining 43.7% of the total transactions (15.5 million) represented changes of names and/or addresses. A total of 9.0 million names were deleted from the registration lists, and another 14.6 million were declared "inactive." In addition, the report noted that voter turnout (as a percentage of the voting-age population) declined by 2.4 percentage points as compared to 1994.39

Similar to the 1995-1996 period, the 1997-1998 FEC report noted that the mail registration provisions of the NVRA "caused relatively few problems for the states and accounted for nearly one-quarter of all voter registration applications from 1997 through 1998." Likewise, the motor vehicle provisions proved easiest for the states to implement and yielded the highest volume of registration applications (15.2 million, or 42.9% of all registration applications). Voter registration activity by designated state voter registration agencies accounted for 8.2% (2.9 million) of voter registration applications during the period 1997 through 1998. The FEC made the same three recommendations that it had made in the second report.40

The fourth report of the FEC, covering the period 1999-2000, released June 30, 2001, was based on the survey responses from 44 states and the District of Columbia covered by the NVRA. The report found that in states covered by the NVRA voter registration (as a percentage of the voting-age population) rose by 1.1 percentage points over comparable figures for 1996 (75.6% to 76.7% of the voting-age population). During the period, a total of 45.7 million registration applications or transactions were processed nationwide. About half, or 22.5 million, represented new transactions. The duplication rate was 7.7% (3.5 million). The remaining 43.0% of the total transactions (19.7 million) represented changes of names and addresses. A total of 13.0 million names were deleted from the registration lists, and another 18.3 million were declared "inactive."41

Similar to the two earlier periods, the FEC report noted that the mail registration provisions of the NVRA caused relatively few problems for the states and accounted for nearly one-third of all voter registration applications (14.2 million). The registration applications via motor vehicle offices again yielded the highest volume of registration applications (17.4 million, or 38.1% of all registration applications). However, survey results indicated "numerous problems with completed voter registration applications being forwarded from motor vehicle offices to the appropriate election official in a timely manner during the most recent election cycle." Voter registration at designated state voter registration agencies accounted for 7.6% (3.5 million) of voter registration applications.42

The FEC again made the same three recommendations that it had made in the two previous reports. The report also added four other recommendations based specifically on the experiences of the 2000 election. These were the following:

(1) "that states develop and implement an on-going, periodic training program for relevant motor vehicle and agency personnel regarding their duties and responsibilities under the NVRA as implemented by the state's law;"

(2) "that states require motor vehicle and agency offices to promptly transmit information regarding voter registration applicants electronically to the appropriate election office with documentation to follow;"

(3) "that states devise a procedure whereby voters may cast a provisional ballot at the polls on election day under circumstances prescribed in state law but at least for the purposes of the fail-safe provisions of the NVRA;" and,

(4) "that states adopt the practice of mailing a forwardable notice to all persons who are removed from the voter registration list whose mail has not previously been returned as undeliverable."43

The fifth report of the FEC, covering the period 2001-2002, released June 30, 2003, was based on the survey responses from 44 states and the District of Columbia covered by the NVRA. The report found that in states covered by the NVRA voter registration (as a percentage of the voting-age population) declined by 1.8 percentage points from comparable figures for the last mid-term election in 1998 (76.0% to 74.2% of the voting-age population). During the period, a total of 37.5 million registration applications or transactions were processed nationwide. Over half or 19.7 million represented new transactions. The duplication rate was 8.7%. The remaining 38.7% of the total transactions represented changes of names and addresses. A total of 15.0 million names were deleted from the registration lists, and another 20.6 million were declared "inactive."44

Similar to the earlier periods, the FEC report noted that the mail registration provisions of the NVRA caused relatively few problems for the states and accounted for more than one-fourth of all voter registration applications (27.6%). The registration applications via motor vehicle offices again yielded the highest volume of registration applications (42.8% of all registration applications received). Unlike the previous reporting period, survey results indicated a significant decrease in the number of problems reported in the state motor vehicle registration programs, although timeliness in transmitting voter registration application remained a problem.45 Voter registration at designated state voter registration agencies accounted for 5.8% of voter registration applications.46

NVRA Reports two through four, as discussed above, all made the same three core recommendations: (1) states should amend their election laws to use the last four digits of the social security number for all new voter registration applications; (2) states should develop and implement a statewide voter registration database, linking all local offices to a centralized database; and, (3) the United States Postal Service should create a new class of mail for "official election materials" with favorable rates and first class treatments.47 In addition, NVRA Report 1999-2000 had added four other recommendations: (1) states should develop and implement an on-going, periodic training program for relevant motor vehicle and agency personnel regarding their duties with respect to NVRA; (2) states should require motor vehicle and agency offices to promptly transmit information regarding voter registration applicants electronically to the appropriate election officials; (3) states should develop a procedure for provisional voting, extending or helping to implement better the fail-safe procedures in NVRA; and, (4) states should adopt the procedure of mailing a forwardable notice to all persons removed from voter registration lists.48 With the exception of the recommendation relating to a special mailing category for NVRA election materials and training programs for state motor vehicle and agency personnel, all of the other FEC recommendations were incorporated, as a whole or in part, in the Help America Vote Act of 2002 (P.L. 107-252). Consequently, the 2001-2002 NVRA Report only repeated the recommendation about the special mailing category and urged the states to set up training programs for state motor vehicle and agency personnel.49

The sixth report, and the first report submitted by the newly created EAC, covered the period 2003-2004 and was released June 30, 2005. It was based on the survey responses from 48 states, the District of Columbia, and three of the five territories (American Samoa, Puerto Rico, and the U.S. Virgin Islands). The report found voter registration for states covered by the provisions of the NVRA (as a percentage of the voting-age population) increased by 1.7 percentage points from comparable figures for the last presidential election in 2000 (from 76.7% to 78.4% of the voting-age population). During the period, a total of 49.6 million registration applications or transactions were processed nationwide. Over half (26 million) represented new transactions. Nearly 3.5 million (7.3%) were duplication registrations. About 1.6 million applications (5.2%) were rejected or determined to be invalid. Some 15.2 million of the total transactions represented changes of names and/or addresses. Nearly 12.6 million names were deleted from the registration lists, while another 10.7 million were declared "inactive."50

The EAC report noted that the mail registration accounted for almost one-third of all voter registration applications (32.4%). The registration applications received via motor vehicle offices yielded the highest volume of registration applications, but only 32.8% of all registration applications received. The number of registration applications received in person at election or registrar offices amounted to 25.4% of all registration applications. Almost 11% of all registration applications came from designated state voter registration agencies.51

The recommendations of the EAC in the sixth report more closely reflected the passage of HAVA's provisions relating to the requirement that each state must establish a statewide voter registration database by January 1, 2006 (P.L. 107-252, §303(a)(1)(A)), than any provisions specific to NVRA. First, states should provide for electronic transmission of voter registration information from state motor vehicle and other designated NVRA state voter registration agencies directly to local election officials. Second, "states should perform list maintenance through electronic coordination with state and federal databases." The report specifically mentioned the U.S. Postal Service National Change of Address and Social Security Death Index databases, as well as records from U.S. Attorneys and U.S. District Courts concerning criminal conviction records. And third, "states should develop statewide voter registration databases that are capable of tracking a registrant's voting and registration history."52

The seventh report covered the period 2005-2006 and was released June 30, 2007. It was based on the survey responses from all states, the District of Columbia, and four territories (American Samoa, Guam, Puerto Rico, and the U.S. Virgin Islands). The report found voter registration for states covered by the provisions of the NVRA (as a percentage of the voting-age population) increased by 2.1 percentage points over comparable figures for the last mid-term election in 2002 (from 74.2% to 76.3% of the voting-age population). During the period, a total of 36.3 million registration applications or transactions were processed nationwide. Nearly half (17.3 million) represented new transactions. Nearly 2.2 million were duplication registrations. Some 10.9 million of the total transactions represented changes of names and addresses. Nearly 13 million names were deleted from the registration lists, while another 9 million were declared "inactive."53

The EAC report noted that the mail registration accounted for almost a quarter of all voter registration applications (22.8%). The registration applications received via motor vehicle offices yielded the highest volume of registration applications (45.7% of all registration applications received). The number of registration applications received in person at election or registrar offices amounted to 19.8% of all registration applications. Over 11% of all registration applications came from designated state voter registration agencies.54

For the seventh report EAC merged two recommendations and kept one recommendation from its sixth report, as well as added two additional recommendations. The first recommendation merges the first and second recommendations from the 2003-2004 Report and reads, "states should continue to improve and modernize their electronic reporting and list maintenance systems." The second recommendation repeats recommendation three from the 2003-2004 Report. The third recommendation in the 2005-2006 report is a message to the states about standardizing the information needed for the NVRA report to Congress, which reads as follows:

states should set up their statewide data collection system to facilitate the collection and reporting of information mandated by NVRA. States should work in partnership with the EAC to establish agreed-upon standards for the handling of active and inactive voters, overseas and military voters, and other important categories for the NVRA.

The fourth recommendation in this report is similar to a previous recommendation in the NVRA Report 2001-2002, and reads "states should provide training to all agencies involved in voter registration."55

The eighth EAC report covered the period 2007-2008, and was released June 30, 2009. It was based on survey responses from all states, the District of Columbia, and four territories (America Samoa, Guam, Puerto Rico, and the U.S. Virgin Islands). The report found voter registration for states covered by the provisions of the NVRA (as a percentage of the voting-age population) increased by 4.2 percentage points over comparable figures for the last presidential election cycle (2003-2004, from 78.4% to 82.6% of the voting-age population).56 During the period covered by the report, a total of more than 60 million registration applications or transactions were processed nationwide. Of these, 24.6 million (or over 40%) represented new voters, which were 7.3 million more new voters than had registered in the prior election cycle (2005-2006).57 More than 20 million of these registration forms requested a change to name, address, or party of the registrant.58 About 3.6 million applications were determined to be duplicates and another 1.7 million were determined to be invalid or otherwise rejected. Combined, duplicates and invalid registration applications constituted 8.8% of the registration applications for this period. More than 12 million names were deleted from the registration lists, "for reasons including death, felony conviction, failure to vote in consecutive elections, having moved from one jurisdiction to another, or at the voter's request."59

The EAC report noted that mail registration accounted for 28.8% of all applications. Registration applications received via motor vehicle offices yielded the highest volume of registration applications, although only slightly above mail registration (30.1% of all registration applications received). The number of registration applications received in person or at election or registrar offices amounted to 14.9% of all registration applications.60 Over 6% (3,807,991) of all registration applications came from designated state voter registration agencies and 1.1% came through internet applications.61

The EAC's eighth report retained one recommendation from its previous report and added five new recommendations. The retained recommendation reads, "states should continue to improve and modernize their electronic reporting and list maintenance systems." Second, the EAC recommends that the "states should explore supporting a coordinated data collection effort that allows local jurisdictions to provide election data to their State election offices." Third, the EAC recommends that "states should provide EAC with information on their proven best practice models of election data collection in order to facilitate sharing with all States through EAC's clearinghouse function." Fourth, the EAC recommends that "states are encouraged to use technology to ease the workload on their election offices, as they deem appropriate." Fifth, the EAC recommends that "states should encourage their public service agencies to remind voters to check and update their registration information." And sixth, the EAC recommends that "for the purpose of compiling comparable election data from future EAC Election Administration and Voting Surveys EAC will continue to work towards a common understanding of election terms."62

The ninth report covered the period 2009-2010 and was released June 30, 2011. It was based on survey responses from all states, the District of Columbia and four territories (American Samoa, Guam, Puerto Rico, and the U.S. Virgin Islands). The report found voter registration for states covered by the provisions of the NVRA (as a percentage of the voting age population) increased by 3.2 percentage points over comparable figures for the last non-presidential election cycle (2005-2006, from 76.3% to 79.5% of the voting age population).63 During the period covered by the report, a total of 45.5 million registration applications or transactions were processed nationwide. Of these, nearly 14.4 million (or over 30%) represented new voters, which was 2.9 million fewer than during the prior comparable election cycle (2005-2006).64 More than 18 million of these registration forms or transactions requested a change to name, address, or party affiliation of the registrant.65 About 2.9 million applications were determined to be duplicates and another 1.4 million were determined to be invalid or otherwise rejected. Combined, duplicates and invalid registration applications constituted 9.4% of the registration applications for the 2009-2010 election cycle. More than 14 million registrants were deleted from the registration lists "for reasons including death, felony convictions, failure to respond to a confirmation notice and failure to vote in consecutive federal elections (both required under NVRA before a registrant's name can be removed), having moved from one jurisdiction to another, or at the voter's request."66

The 2009-2010 EAC report noted that mail registration (mail, fax, and email where permitted) accounted for 20.9% of the 45.5 million voter registration forms. Registration applications received via state motor vehicle agencies yielded the highest volume of registration applications (37.1% of all voter registration applications). The number of registration applications received in-person amounted to 14.5% of all registration applications.67 Almost 6% (2,541,440) of all registration applications came from designated state voter registration agencies other than state motor vehicle agencies. Internet applications, in those states that allowed such applications (not currently required by NVRA), constituted 1.7% of all applications.68

The EAC's ninth report contained no recommendations. This was due to the fact that at the time of the release of the report (required under NVRA), EAC lacked a quorum of commissioners to vote on such recommendations.69

The 10th report covered the period 2011-2012 and was released June 30, 2013. It was based on survey responses from all states, the District of Columbia and four territories (American Samoa, Guam, Puerto Rico, and the U.S. Virgin Islands). The report found voter registration for states covered by the provisions of the NVRA (as a percentage of the voting age population) declined by 3.0 percentage points (from 82.6% in the 2007-2008 election cycle to 79.6% in the 2011-2012 election cycle).70 During the period covered by the report, a total of 62.5 million registration applications or transactions were processed nationwide. Of these, nearly 23.8 million (or over 38%) represented new voters, which was 0.8 million fewer than during the prior comparable presidential election cycle (2007-2008).71 Almost 27.5 million of these registration forms or transactions requested a change to name, address, or party affiliation of the registrant.72 Nearly 3.7 million applications were determined to be duplicates and another 5.0 million were determined to be invalid or otherwise rejected. Combined, duplicates and invalid registration applications constituted 13.9% of the registration applications for the 2011-2012 election cycle. Nearly 13.7 million registrants were deleted from the registration lists "for reasons including death, felony convictions, failure to respond to a confirmation notice and failure to vote in consecutive federal elections (both required under NVRA before a registrant's name can be removed), having moved from one jurisdiction to another, or at the voter's request."73

The 2011-2012 EAC report noted that mail registration (mail, fax, and email where permitted) accounted for 23.3% of the 62.5 million voter registration forms. Registration applications received via state motor vehicle agencies yielded the highest volume of registration applications (33.4% of all voter registration applications). The number of registration applications received in-person amounted to 16.4% of all registration applications.74 Over 6% (3,866,542) of all registration applications came from designated state voter registration agencies other than state motor vehicle agencies.75 Twenty-one states reported voter registration applications by the Internet. Such applications, in those states that allowed such applications (not currently required by NVRA), constituted 5.3% of all applications.76

The EAC's 10th report contained no official EAC recommendations. This was due to the fact that at the time of the release of the report (required under NVRA), the EAC lacked a quorum of commissioners to vote on such recommendations. However, the report does offer what are called "observations." First, "states should continue to improve and modernize their electronic reporting and list maintenance systems." Second, "states should continue to engage their State agencies on issues related to the NVRA and to encourage those agencies to remind voters to check and update their voter registration information."77

Department of Justice Activity

As of May 1994, 23 states had complied with the NVRA, according to a Department of Justice news release (Alaska, Arkansas, Arizona, Colorado, Connecticut, Delaware, Florida, Georgia, Hawaii, Iowa, Kentucky, Maine, Maryland, Massachusetts, Mississippi, Missouri, Nebraska, Oregon, South Dakota, Tennessee, Utah, Washington, and West Virginia). At that time, Justice officials indicated that enforcement of the NVRA would be one of the department's highest priorities.

On January 22, 1995, the Justice Department filed suit against California, Illinois, and Pennsylvania for failure to implement the NVRA by January 1, 1995. South Carolina was also sued on February 6, 1995. Subsequently, on May 4, 1995, the Justice Department and Pennsylvania agreed that the state would fully implement the NVRA, pending passage of legislation. Both California and Illinois, after having decisions handed down by the respective U.S. Courts of Appeal that ruled against their challenges, implemented NVRA. Also, on April 25, 1995, the Justice Department notified Montana about its failure to implement parts of the NVRA (the state failed to implement agency-based registration and to bring its mail registration form into line with the NVRA requirements). In a reply letter dated May 4, Montana agreed to correct the oversight. Although Georgia and Texas developed the NVRA proposals and had implemented the NVRA voter registration procedures, the Justice Department challenged aspects of the implementations under provisions of the Voting Rights Act of 1965. Subsequently, these were resolved and both remain in compliance. Differences that the Justice Department had with Louisiana's implementation plan also were resolved, and Louisiana reached full compliance.

During the period 1998-2009, the Justice Department filed suit in five cases brought against three states (New York, Indiana, and Missouri) for failure to implement parts of the NVRA. The three New York cases concerned the designation of state voter registration agencies.78 Both the Indiana case79 and the Missouri case80 concerned the obligations of the state for list maintenance and whether or not the states were in compliance. The remaining suits filed during this period relating to the enforcement or lacks of enforcement of the provisions of NVRA (86 cases)81 were brought by private persons or organizations. For whatever reason, the Justice Department brought very few suits against the states during this period. Arguably, this was due to the perception that states' enforcement of the provisions of NVRA was adequate, to a lack of resources at the department for such activities, to a policy decision not to exert much effort behind such enforcement, or to a combination of the three.

Recently, there has been criticism of the Justice Department, mainly directed toward the operations under the Bush Administration with respect to enforcement of NVRA. Most of this criticism has been directed at the lack of enforcement of provisions relating to NVRA's agency-based voter registration provisions. In an article authored by Michael Slater, deputy director of Project Vote, he discussed the issue and provided statistics suggesting the lack of state efforts in promoting agency-based voter registration. Slater also argued that "the U.S. Department of Justice has failed to investigate recent allegations that states are not complying with the public agency requirements of the NVRA"82 However, as noted above, the NVRA reports indicate that, while there was a drop in applications in state voter registration agencies during the period 1997-2002, states reported the same level of applications received from these agencies in the 2003-2004 and 2005-2006 periods as they had received in the 1995-1996 reporting periods. However, the level of applications dropped by almost half from the state agencies during the 2007-2008 reporting cycle.83

In hearings in early April 2009, the chairman of the Senate Rules and Administration Committee, Senator Charles Schumer, pressed Attorney General Eric Holder to "pull out all the stops" and sue any state not complying with the agency-based provisions of the NVRA. In a statement, Senator Schumer noted that "this law is supposed to simplify the voter registration process, but it has been complicated by the rogue behavior of a large batch of states."84 There is some evidence that even before the criticisms, the Department of Justice had made moves to enforce these provisions of the NVRA.85

The Response of the States

Overall, responses from most states were cooperative and, despite misgivings about the costs, aimed at the implementation of the NVRA by the January 1, 1995, deadline. However, a few states decided to avoid or challenge the NVRA.

As explained earlier, the provisions of the NVRA were not applicable to states that had had no voter registration system on or before March 11, 1993 (North Dakota), that had had an election-day voter registration procedure at every polling place in effect on or before March 11, 1993 (Minnesota and Wisconsin), or had made the adoption of an election-day registration system contingent on the passage of the NVRA (Wyoming). In an attempt to avoid the implementation of the provisions of the NVRA, which were viewed as costly, Idaho, in January 1994, and New Hampshire, in March 1994, passed legislation adopting election-day voter registration and made it retroactive as of March 11, 1993. Congress effectively ratified this strategy when, in an amendment contained in P.L. 104-99 (signed January 26, 1996), it amended the designated date as specified in the NVRA from March 11, 1993, to August 4, 1994.

Mississippi passed legislation giving state and local officials the authority to implement the NVRA, but only for federal elections. The state legislature established a dual registration system, providing that citizens who wished to cast a ballot in state and local elections had to register separately in accordance with prior registration law. On April 20, 1995, four citizens (later joined by the U.S. Department of Justice) sued over the dual registration requirements, charging that they violated the Voting Rights Act of 1965, as amended. On March 31, 1997, the Supreme Court ruled unanimously that Mississippi had violated the pre-clearance provisions of the Voting Rights Act when it chose to establish a dual voter registration system by applying the provisions of the NVRA solely to elections for federal office without first submitting the electoral changes for Justice Department approval. Mississippi had argued that, as it made no changes to state law, it did not need to seek pre-clearance for the new system, having already received clearance for its implementation plan of the NVRA provisions. As a result of the decision, Mississippi was required to submit its current dual registration system for approval by the Justice Department.86 On October 5, 1998, a three-judge federal panel ordered Mississippi to allow all persons who registered under the NVRA provisions to vote in state and local elections until Mississippi enacted a law implementing the NVRA that was pre-cleared by the Justice Department. Subsequently (in 2000), Mississippi laws were brought into compliance.

After the first two years (1995-1996) of implementation, litigation, and an election, state and local officials appeared to come to terms with the NVRA. However, proposals were suggested to amend the NVRA to make it easier to administer and prevent election fraud. Some proposals suggested would (1) require a Social Security number on voter registration applications; (2) change the citizenship attestation to a question ["Are you a citizen?" YES NO] where applicants would have to say yes to be eligible to register; (3) eliminate the NVRA-mandated second mailings to persons already identified by the Postal Service as having moved; and (4) enact legislation to help reduce the costs of NVRA-mandated mailings.87 Election officials formed a task force, the Joint Election Officials Liaison Committee (JEOLC) to negotiate and lobby for various changes in the law, as well as to work with the U.S. Postal Service to devise procedures that would make it easier for election offices to comply with some provisions of the NVRA. The JEOLC was composed of representatives from the International Association of Clerks, Recorders, Election Officials, and Treasurers (IACREOT), the National Association of County Recorders and Clerks (NACRC), the National Association of State Election Directors (NASED), and the Election Center.

A survey conducted by the National Association of Secretaries of State (NASS) of 43 states covered by the NVRA indicated that gains in voter registration had occurred in every state. Of the 43 states, 41 reported drivers' license bureaus effective in registering voters, and 35 indicated that other agencies also proved effective. More than half of the states indicated that their biggest problem was transmittal difficulties involving untimely, incomplete, unsigned, and illegible registration forms. About half thought that the NVRA restrictions on removing voters from the rolls contributed significantly to the greater number of registrants. However, other states applauded the NVRA for helping them maintain accurate registration lists.88

The Response of the Citizenry

Figure 1, below, based in large part on the most recent NVRA Report, displays voter registration89 as a percentage of the voting-age population (VAP),90 as a percentage of the citizen voting age population (CVAP),91 and as a percentage of McDonald's voting eligible population (VEP)92 in those states covered by the NVRA (44 states and the District of Columbia) for the period 1992-2012. Comparing only the differences in presidential election years, (depending on which measure one chooses) voter registration increased (7.9%—VAP, 11.4%—CVAP, or 10.6%—VEP) almost 8-11 percentage points between the pre-NVRA, 1992 presidential election and the 2012 presidential election.93 The bulk of the change appears to have come immediately following the implementation of the NVRA.

Figure 1. Percentage of Voter Registration: 1992-2012

(NVRA States & DC Only)

Source: ****

Notes: VAP, voting age population; CVAP, citizen voting age population; VEP, McDonalds's voting eligible population. Note that the percentage scale runs from 65% to 90%. This was done so that the distinction between the CVAP figures and the VEP figures could be shown.

As noted earlier, many proponents appeared to believe that the NVRA would not only increase voter registration but would also increase voter participation. Comparing post-NVRA voter turnout to the 1992 presidential election voter turnout might be somewhat misleading, as that election included a popular third-party candidate. Figure 2 displays voter turnout for the NVRA covered states (as a % of the VAP, as a % of the CVAP and as a % of McDonald's VEP) in presidential elections since 1980.94 In other words, voter turnout for presidential elections in the NVRA covered states has fluctuated between a low of less than 47.6% in 1996 to a high of over 56% in 2008. With the low and high turnout coming in the post-NVRA period, it is not clear whether NVRA has had an impact on voter turnout in presidential elections. On the one hand, there has been a steady increase (except for a slight drop in 2012) in voter turnout in presidential elections since 1996, the first post-NVRA presidential election year, of about 10 percentage points. On the other hand, only the 2008 election turnout exceeds the 1992, pre-NVRA presidential election turnout results. Similarly, voter turnout for these same states in non-presidential election years has been the following: 1982—40.1%, 1986—36.1%, 1990—36.0%, 1994—38.0%, 1998—34.6%, 2002 ―35.7%, 2006―36.5%, and 2010—37.2%.95 Mid-term elections have not shown much fluctuation since 1986. All-in-all, while the NVRA has been in effect for 18 years, there appears to be mixed evidence about the impact of the NVRA on increasing voter turnout. Of course, one could always argue that voter turnout might have been much worse without the NVRA, a rather difficult hypothesis to prove or disprove.

Figure 2. Percentage Voter Turnout, Presidential Elections: 1980-2012

(NVRA States & D.C. Only)

Source: Table A-1, Table A-2.

Notes: VAP, voting age population; CVAP, citizen voting age population; VEP, McDonalds's voting eligible population. Note that the percentage scale runs from 45% to 65%. This was done so that the distinction between the CVAP figures and the VEP figures could be shown.

The Congressional Response Post-NVRA96

The NVRA was not supported by most Republicans in the 103rd Congress, and a similar bill had been vetoed by President Bush in 1992. With the ascendancy of the Republican Party in the 104th Congress, and given the criticism of the NVRA as an "unfunded mandate" of the federal government by several prominent Republican governors from large states,97 attempts to repeal or modify the NVRA appeared certain. However, while the Republican Party controlled both the House and Senate, its control in the Senate would not have prevented Democratic Senators from mounting a filibuster against proposals to change the NVRA. Further, with the White House in the hands of a Democrat (President Clinton) who supported the NVRA, any changes in the NVRA that were not bipartisan would almost certainly have been vetoed.

The 112th Congress saw the introduction of bills aimed at requiring states to establish same-day or election-day voter registration in all federal elections (H.R. 108, H.R. 3317, H.R. 3163, H.R. 5799/S. 3608); be able to update voter registration records at the polling place on election day (H.R. 3163); allow for Internet voter registration and use of the Internet to update statewide voter registration records (H.R. 5799/S. 3608, H.R. 6632); and, provide for the establishment of an automatic registration system based on information taken directly from state and federal agency database systems (H.R. 5799/S. 3608). In addition, legislation was proposed to require each voter registration agency (as defined under NVRA) within a state to insure that every registered voter, without charge, was issued a voter ID card if such voter ID was required by state law as a condition for voting in a federal election (H.R. 4126). Legislation also was proposed that would prevent any interference with voter registration (H.R. 5799/S. 3608), require the EAC to develop best practices for states to prevent such interference (H.R. 5799/S. 3608), and provide major penalties for promoting voter registration fraud (H.R. 6593/S. 168). Additionally, one bill, the Voter Empowerment Act of 2012 (H.R. 5799/S. 3608), provided that states must accept voter registration applications from otherwise qualified citizens 16 years of age and older, but the states were not required to make any changes in the age requirement for voting in state law. The same legislation specified that by 90 days after the end of every year, states must submit to the EAC specified statewide voter registration statistical information. It extended NVRA provisions and other provisions specific to the act to the District of Columbia, Puerto Rico, the U.S. Virgin Islands, Guam, and America Samoa. The Voter Empowerment Act of 2012 also established a pilot program to allow persons with disabilities to register (and to vote) from their residencies.

Other proposed legislation in the 112th Congress, among other things, provided for funding to states to explore appropriate technological changes to simplify the voter registration process within a state (H.R. 6590/S. 3635); required the Secretary of the Department of Veterans Affairs to specify local facilities within the Department as voter registration agencies as defined under NVRA (S. 1264); and designated a "National Month of Voter Registration (H.Res. 758/ S.Res. 572).

Thus far in the 113th Congress, much of the proposed legislation introduced duplicates legislative proposals introduced in the 112th Congress. This is true of the Voter Empowerment Act of 2013 (H.R. 12/S. 123 [H.R. 5799/S. 3608 in the 112th ]), the Fair, Accurate, Secure, and Timely Voting Act or FAST Voting Act (H.R. 97/S. 85 [H.R. 6590/S. 3635 in the 112th]), the Same Day Registration Act of 2013 (H.R. 280/S. 532 [H.R. 3317 in the 112th]), the Value Our Time Elections or VOTE Act (H.R. 289 [H.R. 6632 in the 112th with a different title]), the Voter Fraud Prevention Act (H.R. 1280 [H.R. 6593/S. 168 in the 112th]), and the Voter Registration Efficiency Act (H.R. 2115 [H.R. 6386 in the 112th with a slightly different title]).

On June 4, 2013, the House Committee on House Administration held a hearing on the Voter Registration Efficiency Act (H.R. 2115). The bill is intended to reduce the number of voters registered in multiple states, among other things. The bill would amend the National Voter Registration Act of 1993 to require individuals registering to vote in a state to indicate if the state will be the individual's residence for the purpose of voting and for other purposes. The proposal is similar to an Arizona program, Interstate Cross-Check, in existence since 2009, matching records from 21 cooperating states. According to Arizona Secretary of State Ken Bennett, in 2012, the cross-check located 45,000 duplicate voter registration records in Arizona from the other states.98

New legislative proposals in the 113th Congress direct the EAC to conduct a pilot program to provide funds to local educational agencies to instruct high school seniors about the voter registration process, the Students Voicing Opinions in Today's Elections (VOTE) Act (H.R. 653), applies directly all NVRA provisions to Puerto Rico, American Samoa, the Commonwealth of the Northern Mariana Islands, Guam, and the U.S. Virgin Islands (H.R. 1018), amends the NVRA to permit states to require documentary evidence to prove citizenship (H.R. 2409), and calls on the Congress to strengthen the "Nation's electoral system by ensuring clean and fair elections" (S. 9).

Concluding Observations

The National Voter Registration Act of 1993 has been the law of the land for over 20 years and has been in effect for 18 years. The courts have resolved many of the initial issues. A review of the FEC/EAC reports appears to indicate that the states have come to terms with the provisions, despite the fact that state election officials continue to advocate that the federal government should provide funding for the implementation of aspects of the act.

There may still be some problems with implementation at the local levels and with the training of nonelection officials who are responsible under the NVRA for providing voter registration services. As noted above, some opponents would like to curtail parts of the NVRA; proponents, however, do not think the NVRA has gone far enough. Some are calling for voter registration reform that would include what has been called "universal voter registration," where the government is responsible for registering all citizens to vote unlike the system used in the United States that relies on individuals to register themselves.99 Others have suggested minor changes in NVRA, like making it "opt-out" instead of "opt-in" (i.e., automatic registration when an applicant receives a driver's license or benefits at designated state voter registration agencies, unless specifically deciding to "opt-out"), but with more enforcement of the provisions that are currently in place.100 How and whether the Congress decides to respond to these calls for reform remains to be seen.

Appendix A. Supporting Tables

Table A-1. Percentage Voter Registration: 1992-2012

(NVRA States and DC only)

 

Year

Voting Age Population (VAP)a

Total Reported Registrationa

Percent Total Registration of VAP

Citizen Voting Age Population (CVAP)b

Percent Total Registration of CVAP

Voting Eligible Population (VEPb,c)

Percent Total Registration of VEP

National Summary for States Covered by NVRA (45 States)

2012

228,473,225

181,793,063

79.57

208,204,942

87.31

210,535,549

86.35

2010

223,078,025

177,346,464

79.50

203,702,888

87.06

206,422,219

85.91

2008

215,260,000

177,825,238

82.61

199,549,000

89.11

202,410,902

87.85

2006

214,537,000

163,713,303

76.31

195,414,849

83.78

197,077,895

83.07

2004

209,417,000

164,124,163

78.37

191,821,219

85.56

193,111,232

84.99

2002

204,415,000

151,646,523

74.19

187,900,903

80.71

188,257,311

80.55

2000

199,458,000

153,017,839

76.72

184,311,072

83.02

184,375,887

82.99

1998

191,299,000

145,328,223

75.97

180,532,682

80.50

180,691,782

80.43

1996

186,999,000

141,327,487

75.58

176,742,136

79.96

176,762,669

79.95

1994

183,443,000

125,331,513

68.32

173,373,530

72.29

173,257,216

72.34

1992

180,203,000

129,156,319

71.67

170,194,845

75.89

170,513,096

75.75

 

National Summary for States Not Covered by NVRA (ID, MN, NH, ND, WI, WY

2012

11,712,727

9,934,459

84.82

11,288,706

88.00

11,390,271

87.22

2010

11,486,046

9,458,748

82.35

11,091,440

85.28

10,988,813

86.08

2008

11,315,000

10,059,488

88.90

10,927,000

92.06

10,902,606

92.27

2006

11,127,000

9,029,403

81.15

10,736,724

84.10

10,565,699

85.46

2004

10,963,000

9,534,261

86.97

10,528,352

90.56

10,372,224

91.92

2002

10,660,000

8,994,452

84.38

10,303,908

87.29

10,124,629

88.84

2000

10,369,000

9,481,940

91.45

10,091,787

93.96

9,955,549

95.24

1998

9,971,000

8,677,330

87.03

9,869,769

87.92

9,727,789

89.20

1996

9,790,000

9,063,714

92.58

9,692,065

93.52

9,584,375

94.57

1994

9,567,000

8,645,722

90.37

9,462,243

91.37

9,366,132

92.31

1992

9,326,000

8,782,267

94.17

9,226,214

95.19

9,142,427

96.06

Source: NVRA Report 2011-2012, Table 1a. Registration History, p. 14-31; United States Election Project, Voter Turnout, Google Doc spreadsheet for the full general election series from 1980-2012, http://elections.gmu.edu/voter_turnout.htm.

Notes:

a. NVRA Report 2011-2012, Table 1a. Registration History, p. 14-31. The percentages are based on the figures in the column labeled "Reported Registration." Empty cells were completed by values from an earlier CRS Report, CRS Report 96-932, Voter Registration and Turnout: 1948-1994, by [author name scrubbed] (archived; available from the author). The table was then recomputed by the author based only on the 44 states and the District of Columbia covered by the NVRA.

b. For the CVAP values for 2008, 2010, and 2012, see NVRA Report 2011-2012, Table 1a. Registration History, p. 14-31. For all other years, the values were derived from Professor Michael McDonald's percentages shown for non-citizen population from his spreadsheet for the full general election series from 1980-2012, see United States Election Project, Voter Turnout, Google Doc spreadsheet for the full general election series from 1980-2012, http://elections.gmu.edu/voter_turnout.htm.

c. For all years, the values were taken from Professor Michael McDonald's spreadsheet for the full general election series from 1980-2012, see United States Election Project, Voter Turnout, Google Doc spreadsheet for the full general election series from 1980-2012, http://elections.gmu.edu/voter_turnout.htm.


Table A-2. Voter Turnout: 1980-2012

 

Year

VAP Turnout Rate

CVAP Turnout Rate

VEP Turnout Rate

Voting Age Population (VAP)

Citizen Voting Age Population (CVAP)

Voting-Eligible Population (VEP)

Vote for Highest Office

National Summary for States Covered by NVRA (45 States)

2012

52.9%

57.9%

57.5%

229,174,682

209,113,911

210,535,549

121,129,101

2010

37.2%

40.8%

40.4%

224,478,172

204,550,501

206,422,219

83,484,726

2008

56.2%

61.6%

61.0%

219,518,743

200,516,198

202,410,902

123,473,664

2006

36.5%

40.0%

39.7%

214,365,179

195,414,849

197,077,895

78,184,498

2004

55.4%

59.8%

59.4%

209,460,979

191,821,219

193,111,232

114,637,209

2002

35.7%

38.9%

38.8%

204,827,709

187,900,903

188,257,311

73,076,854

2000

49.3%

53.6%

53.6%

200,226,666

184,311,072

184,375,887

98,765,544

1998

34.6%

37.4%

37.4%

195,172,053

180,532,682

180,691,782

67,595,964

1996

47.6%

51.2%

51.1%

190,086,054

176,742,136

176,762,669

90,407,610

1994

38.0%

40.7%

40.8%

185,591,574

173,373,530

173,257,216

70,603,818

1992

54.0%

57.6%

57.5%

181,379,013

170,194,845

170,513,096

97,997,258

1990

36.0%

38.2%

38.0%

176,990,353

166,845,637

167,671,505

63,663,174

1988

49.7%

52.5%

52.2%

172,939,268

163,757,422

164,752,775

85,972,439

1986

36.1%

37.9%

37.7%

169,072,114

160,814,374

161,713,342

60,956,054

1984

52.7%

55.1%

54.7%

165,248,409

157,868,399

159,107,561

87,058,555

1982a

40.1%

41.9%

41.6%

157,417,883

150,933,980

151,993,008

63,190,427

1980

51.8%

53.8%

53.4%

156,032,668

150,310,795

151,347,591

80,890,341

 

 

 

 

 

 

 

 

 

National Summary for States Not Covered by NVRA (ID, MN, NH, ND, WI, WY

2012

67.6%

70.2%

69.7%

11,752,275

11,318,344

11,390,271

7,943,246

2010

48.6%

50.4%

51.1%

11,544,817

11,132,348

10,988,813

5,614,750

2008

69.0%

71.4%

71.8%

11,353,287

10,969,924

10,902,606

7,831,067

2006

50.6%

52.4%

53.3%

11,120,220

10,736,724

10,565,699

5,631,094

2004

55.4%

72.7%

73.8%

10,875,040

10,528,352

10,372,224

7,657,769

2002

49.9%

51.5%

52.4%

10,633,840

10,303,908

10,124,629

5,305,089

2000

63.6%

65.5%

66.4%

10,396,742

10,091,787

9,955,549

6,609,942

1998

48.7%

50.1%

50.8%

10,141,129

9,869,769

9,727,789

4,941,036

1996

59.0%

60.4%

61.1%

9,929,863

9,692,065

9,584,375

5,855,325

1994

46.6%

47.6%

48.1%

9,666,776

9,462,243

9,366,132

4,502,042

1992

68.2%

69.5%

70.1%

9,398,910

9,226,214

9,142,427

6,407,897

1990

45.8%

46.5%

46.8%

9,168,488

9,025,413

8,957,755

4,196,015

1988

62.4%

63.3%

63.7%

9,016,216

8,888,464

8,826,505

5,622,252

1986

45.6%

46.2%

46.5%

8,850,216

8,732,720

8,682,688

4,035,074

1984

64.0%

64.8%

65.1%

8,746,201

8,637,884

8,594,343

5,594,125

1982a

51.4%

52.0%

52.2%

8,609,751

8,510,680

8,473,911

4,425,149

1980

66.9%

67.6%

67.9%

8,412,807

8,323,262

8,287,510

5,624,880

Source: United States Election Project, Voter Turnout, Google Doc spreadsheet for the full general election series from 1980-2012, http://elections.gmu.edu/voter_turnout.htm.

Notes:

a. Excludes election results for the state of Louisiana.

Appendix B. Post-NVRA Voter Registration Legislative History, 104th to 111th Congresses

In the 105th Congress, as in the 104th, bills were introduced to repeal the NVRA (H.R. 345), to make the provisions voluntary (H.R. 2115), and to modify the provisions in some fashion. The modifications proposed consisted of the following: (1) requiring citizens to produce a Social Security number in order to register to vote (H.R. 224, H.R. 2076, and S. 1561); (2) requiring proof of citizenship in order to register to vote (H.R. 1139, H.R. 2076, and S. 1561); (3) allowing states to require photographic identification in order to cast a ballot (H.R. 1139 and S. 1561); (4) repealing the mail voter registration provisions of NVRA (H.R. 2076 and S. 1561); (5) changing purge rules in the NVRA to make it easier and less expensive for states to purge their voter registration rolls of ineligible persons (H.R. 2076, H.R. 3485, and S. 1561); and, (6) making optional for the states the requirement to allow registrants who change their address to vote at the polling place for their old address, the "failsafe voting" provision (H.R. 2076 and S. 1561).

In addition, Representative Horn introduced legislation that would have required the Immigration and Naturalization Service (INS) and the Social Security Administration, at the request of federal, state, or local election officials, to provide information on the citizenship status of persons attempting to register to vote (H.R. 1428).

The House Oversight Committee (renamed the Committee on House Administration in the 106th Congress) held hearings on H.R. 224, H.R. 1139, H.R. 1428, and H.R. 2076 on November 6, 1997. On February 12, 1998, H.R. 1428 was brought up for a vote under suspension of the rules in the House of Representatives, a procedure that requires a two-thirds majority to pass, and was defeated (vote: 210-200).

On March 18, 1998, Representative Thomas, then chairman of the Committee on House Oversight, introduced legislation (H.R. 3485) that, in part, would have established a pilot project in five states (California, New York, Texas, Florida, and Illinois) of the program proposed in H.R. 1428. H.R. 3485 was reported out of the Committee on House Oversight on March 23, 1998 (H.Rept. 105-457). On March 30, 1998, H.R. 3581, a revision of H.R. 3485, was introduced and brought up for a vote in the House, under suspended rules, and was defeated (vote: 74-337).

On May 6, 1998, the House passed (414-4) the Higher Education Amendments of 1998, H.R. 6, which contained an amendment offered by Representative Clayton (H. Amdt. 583), requiring institutions of higher learning to distribute the NVRA mail voter registration application form to each student during enrollment, unless the student, in writing, declined such a form. The Senate version of H.R. 6 passed without this provision. In conference, the amendment was modified to require institutions in states covered by the NVRA to make "a good faith effort to distribute a mail voter registration form ... to each student enrolled in a degree or certified program." In addition, the revised language required the institutions to request voter registration forms from the state 120 days prior to the deadline for registering to vote, but it allowed an institution to implement its own program without interference from state officials. The conference report (H.Rept. 105-750) passed the House by voice vote September 28 and passed the Senate (vote: 96-0) September 29. The legislation was signed by President Clinton (P.L. 105-244) October 7, 1998.

As the 106th Congress began, two voter registration related bills were introduced in the House. On January 6, 1999, Representative Stump, as he did in the 105th Congress, introduced a bill (H.R. 38) to repeal the NVRA. On the same day, Representative McCollum introduced H.R. 180, a duplicate of the bill he introduced in the 105th Congress (H.R. 224). In May, Senator Warner introduced legislation to deal with campaign finance reform (S. 1107). S. 1107, essentially a duplicate of S. 1561, introduced in the 105th Congress, see above). This bill amended parts of the NVRA by (1) repealing the mail voter registration requirements, (2) requiring states to obtain an applicant's Social Security number when attempting to register to vote, (3) requiring applicant's registering at a state motor vehicle office to submit proof of citizenship, (4) allowing states to purge names from voter registration lists because the registrants had failed to vote for period of time after the registrant was notified, and (5) repealing the limited provisional voting provisions of the NVRA. In September, 1999, Representative Luther introduced legislation to amend the NVRA by requiring states to provide citizens the right to register at the polling place on election-day (H.R. 2864). There was no action on any of these bills in the 106th Congress.

In reaction to the events of the 2000 Presidential election, a number of bills were introduced in the 107th Congress to reform the election administration process. While some of these bills focused solely on voter registration (H.R. 128, H.R. 189, H.R. 829, H.R. 2687, S. 2226), much of the effort was focused on a more comprehensive response. The efforts culminated in the passage of the Help America Vote Act of 2002 (HAVA) (P.L. 107-252). HAVA addressed voter registration in four areas:

(1) it created the Election Assistance Commission (EAC) and transferred FEC's responsibilities with respect to NVRA to EAC (§201; §202; §209; §803), specifically with respect to the development and maintenance of the federal mail voter registration form (§303(b)(4)) and the biennial reports on the impact of the NVRA;

(2) it specified that each state "shall implement, in a uniform and nondiscriminatory manner, a single, uniform, official, centralized, interactive computerized statewide voter registration list defined, maintained, and administered at the State level that contains the name and registration information of every legally registered voter in the State and assigns a unique identifier to each legally registered voter in the State" (§303(a)(1)(A)) and "perform list maintenance with respect to the computerized list on a regular basis" in accordance with the provisions of NVRA (§303(a)(2));

(3) it required the use of the driver's license number or the last four digits of the social security number of the applicant on all voter registration applications (§303(a)(5)(A)), as well as requiring that the Social Security Administration develop relationships with the departments of motor vehicles in each state to assist in list verification (§303(a)(5)(B); and,

(4) it expanded the "fail safe voting" provisions of NVRA to provisional voting (§303(a)(5)(A)-(B)).

In the 108th Congress, legislation was introduced to repeal NVRA (H.R. 2139), to require proof of citizenship to register to vote (H.R. 4174, H.R. 4530), to standardize the treatment of ex-felons with respect to their right to register to vote (H.R. 1433, H.R. 4758), to require states "to provide notice and an opportunity for review prior to removing any individual from the official list of eligible voters by reason of criminal conviction or mental incapacity" (H.R. 4250), to require same-day or election-day voter registration (H.R. 1510, H.R. 3153), and to allow pre-registration to individuals to register to vote prior to meeting the age requirement (H.R. 4972). None of the bills reached beyond the committee assignment.

In the 109th Congress, legislation was introduced to require states to institute same-day or election-day voter registration (H.R. 496, H.R. 533, H.R. 939, H.R. 3557, S. 17. S. 450), to require proof of citizenship when registering (H.R. 4989), to require a photo-identification when registering to vote (H.R. 2250), to allow removal from registration lists for failure to vote (H.R. 2778), to prohibit removal from voter registration lists due to correctable errors and removal for felony conviction or death unless the comparison database meet standards set by the Director of the National Institute of Standards and Technology (H.R. 3094), to delay the implementation of the state voter registration database requirement in HAVA for four years (H.R. 3163), to require states to keep the name and addresses of victims of domestic violence that appeared on state voter registration databases confidential (H.R. 4462), and to require states to produce durable voter registration cards (H.R. 4989). One other bill, S. 414, had several goals. It would have required the use of the Social Security number in registering voters, required that all states synchronize all state voter registration databases, eliminated third party voter registration, modified conditions for removal of names from voter registration databases, and clarified procedures for first-time registrants. None of the bills reached the floor of the House or the Senate.

The legislative activity in the 110th Congress, in many ways, reproduced that in the 109th. Same-day or election-day voter registration was introduced in a variety of bills (H.R. 2457, H.R. 5628, H.R. 1381, H.R. 5946, S. 730, S. 804, S. 2959). Three bills contained provisions aimed at requiring notice before any names could be removed from voter registration lists (H.R. 7244, S. 730, S. 804). Legislation also was introduced to control by whom and how voter registration forms were distributed (H.R. 301), to require the acceptance of any type of voter registration form sent by overseas military or civilians (S. 3073), to require the acceptance of a newly created official federal voter registration and ballot application from overseas citizens (H.R. 4237), to make institutions of higher education state voter registration agencies as specified in NVRA (S. 3390, H.R. 6704), to require the addition of a statement on voter registration application forms that stated that any alien who used the form to register was deportable under the Immigration and Nationality Act (H.R. 5695), to prohibit states from not accepting voter registration forms from third parties (S. 1487), to require photographic identification to register to vote (H.R. 879), to promote pre-registration for students under 18 (S. 3100), and to have the EAC study Internet voter registration (S. 730, S. 804). The only legislation that passed the House, however, was a bill that required the Department of Veterans Affairs to designate all of its public facilities as state voter registration agencies as defined under NVRA (H.R. 6625). This bill was not acted upon in the Senate.

The 111th Congress saw the introduction of bills aimed at requiring states to establish same-day voter registration or election-day registration (H.R. 105, H.R. 3957, S. 1986). Bills were introduced containing provisions aimed at requiring notice before names could be removed from statewide voter registration databases (H.R. 105, H.R. 3416, H.R. 3835). Also, bills were introduced requiring or promoting states to adopt on-line or Internet voter registration procedures (H.R. 105, H.R. 1719, H.R. 4449, S. 3301). Bills were also introduced to add to the list of official voter registration agencies, as designated in NVRA, by adding universities (H.R. 1729, S. 1125), or by adding Department of Veterans Administration facilities (S. 1556), or by designating an office on each military installation as a voter registration agency (H.R. 3274, S. 1265). Other bills were introduced to regulate or minimize challenges to voter registration status on election-day at the polling place (H.R. 103, H.R. 105). One bill, which concerned various aspects of the election process, required states and the EAC to replace specific questions on voter registration application forms about either age or citizenship or both with a signed affidavit on the form (H.R. 105). Another bill (H.R. 3489) would have amended HAVA to prevent any challenge to eligibility to register to vote due to foreclosure proceedings being brought against the voter. Another proposal (S. 1103) would establish standards for the distribution of voter registration forms, and would require organizations to register with the state prior to distributing such forms.

While a few of the above proposed bills were discussed in hearings, none came to the floor or were voted on by either chamber. However, on October 28, 2009, the President signed into law the Military and Overseas Vote Empowerment (MOVE) Act (S. 1415, P.L. 111-84, 123 Stat. 2190 (2009)), which, in part amended the Uniform and Overseas Citizens Absentee Voting Act (UOCAVA). Signed into law as part of the National Defense Authorization Act of 2009 (H.R. 2647/S. 1390, §575, P.L. 111-84, 123 Stat. 2318, [42 U.S.C.A. §1971]), the MOVE Act made changes to certain absentee voting laws and procedures that apply to military and overseas voters. With respect to changes in voter registration procedures, the act established procedures for absent uniform service voters and overseas voters to request and for states to send voter registration applications and absentee ballot applications by mail and electronically (P.L. 111-84, §577(a), [42 U.S.C.A. §1973ff-1]). The act also prohibited refusal to accept voter registration applications, absentee ballot applications, marked absentee ballots and federal write-in absentee ballots for failure to meet certain requirements ("notarization requirements," "restrictions on paper type, including weight and size," and "restrictions on envelope type, including weight and size")(P.L. 111-84, §582(a)-(c), [42 U.S.C.A. §1973ff-1(i), §1973ff-2(f)]).

In addition, the act amends Chapter 80, of Title 10, United States Code, pertaining to the military, by requiring that "the Secretaries of the military departments shall designate offices on installations under their jurisdiction to provide absent uniformed service voters, ..."

"(1) Information on voter registration procedures and absentee ballot procedures...."

"(2) Information and assistance, if requested, including access to the Internet where practicable, to register to vote in an election for federal office, ..." (P.L. 111-84 §583(b), 10 U.S.C.A. §1566a(a)-(d)).

Also, the Secretaries of the military departments may designate these voter assistance offices as "voter registration agencies" on military installations under Section 7(a)(2) of the NVRA of 1993 (P.L. 103-31.[42 U.S.C. 1973gg-5(a)(2)]) (P.L. 111-84, §583(b), [10 U.S.C.A. §1566(e)]). In other words, the act requires that all installations of the military have voter assistance offices to help absent uniformed service voters with voter registration applications as well as with assistance acquiring absentee ballots. At the discretion of the Secretaries of each military department, these offices may be designated as voter registration agencies. If so designated, these offices would be authorized to accept voter registration applications and transmit to the appropriate state election officials within the specified time frame indicated in the NVRA. However, the Secretary of Defense has the authority to changes these procedures through regulations (P.L. 111-84, §583(b), [10 U.S.C.A. §1566A(f)]).

In the process leading up to the passage of the MOVE Act, several bills were also introduced to modify UOCAVA that would have modified voter registration practices as well (H.R. 1659, H.R. 1739, H.R. 2823, H.R. 3416, H.R. 3473). The bills either never reached the floor or provisions were incorporated into the MOVE Act.

Footnotes

1.

For a detailed discussion, see CRS's Legal Sidebar, "Supreme Court Rules: Arizona Law Requiring Proof of Citizenship for Voter Registration Preempted by Federal Law," by Paige Whitaker, http://www.crs.gov/LegalSidebar/details.aspx?ID=558&Source=ibc.

2.

CRS Report RS20764, The Uniformed and Overseas Citizens Absentee Voting Act: Overview and Issues, by [author name scrubbed].

3.

U.S. Congress, House Committee on House Administration, Voter Registration, hearings on H.R. 15, H.R. 17 and H.R. 87, 101st Cong., 1st sess., March 21, 1989 (Washington: GPO, 1989).

4.

U.S. Congress, Senate Committee on Rules and Administration, National Voter Registration Act of 1993, report to accompany S. 460, 103rd Cong., 1st sess., S. Rept. 103-6, (Washington: GPO, 1993), p. 7.

5.

Ibid., p. 5

6.

Subsequently delegated to the Election Assistance Commission (EAC) when FEC functions were transferred because of HAVA in 2002 (P.L. 107-252, title VIII, §802(a), Oct. 29, 2002, 116 Stat. 1726., U.S.C. 42 §15532).

7.

These provisions were strengthened somewhat by HAVA, P.L. 107-252 (42 U.S.C., §15483 (2)(A)) by requiring that the new required statewide voter registration database be compared to other appropriate statewide databases.

8.

The provision about "failure to vote in two consecutive elections" was clarified in HAVA, P.L. 107-252 (42 U.S.C. §15483 (4)).

9.

Again, these functions were transferred to the EAC by HAVA, P.L. 107-252, see footnote 6.

10.

U.S. Congress, Senate Committee on Rules and Administration, National Voter Registration Act of 1993, report to accompany S. 460, 103rd Cong., 1st sess., S. Rept. 103-6, (Washington: GPO, 1993), p. 5.

11.

Ibid., pp. 51-52.

12.

For a different view of "decreasing voter turnout," see Michael P. McDonald and Samuel Popkin, "The Myth of the Vanishing Voter," American Political Science Review, vol. 94, no. 4 (2001), pp. 963-974.

13.

Sen. Mitch McConnell, "National Voter Registration Act of 1992," remarks in the Senate, daily edition, vol. 138, Sept. 21, 1992, p. S14206.

14.

Ibid., pp. 2-3. Also see U.S. Congress, House Committee on House Administration, National Voter Registration Act of 1993, report to accompany H.R. 2, 103rd Cong., 1st sess., H. Rept. 103-9 (Washington: GPO, 1993), pp. 2-5.

15.

S. Rept. 103-6, pp. 50-51.

16.

Ibid., p. 51. H. Rept. 103-9, p. 36.

17.

Sen. Mitch McConnell, "National Voter Registration Act of 1992," remarks in the Senate, daily edition, vol. 138, Sept. 21, 1992, p. S14206.

18.

Sen. Daniel K. Akaka, "National Voter Registration Act of 1993," remarks in the Senate, Congressional Record, vol. 139, Mar. 17, 1993, p. S3002.

19.

S. Rept. 103-6, p. 41.

20.

S. Rept. 103-6, pp. 2, 5.

21.

The Federal Election Commission, The Impact of the National Voter Registration Act of 1993 on the Administration of Elections for Federal Office 1995-1996, Washington, DC, June 30, 1997, p.14.

22.

Ibid., p. 14.

23.

Ibid., p. 15. These are the figures for the 1995-1996 period. Comparable figures given for the 2001-2002 period were 160,000,000, 20,000,000, 4,000,000, and 37,000,000, respectively. See, the 2001-2002 NVRA Report, The Federal Election Commission, The Impact of the National Voter Registration Act of 1993 on the Administration of Elections for Federal Office 2001-2002, Washington, DC, June 30, 2003, p. 24.

24.

NVRA Report 1995-1996, p. 15.

25.

Lisa Gilbert, Saving Dollars, Saving Democracy: Cost Savings for Local Election Officials through Voter Registration Modernization, U.S. PIRG, May 2009, p. 3.

26.

Ibid., p. 3.

27.

Ibid., pp. 3-4, 20.

28.

Ibid., p.20.

29.

S. Rept. 103-6, pp. 52-57. H. Rept. 103-9, pp. 34-37.

30.

HAVA requires that language to this effect be included in the mail registration form developed by the Election Assistance Commission (EAC).(U.S.C. 42 §15483(b)(4)(A)(i)).

31.

S. Rept. 103-6, pp. 5, 11-13, 20-21, 37. H. Rept. 103-9, pp. 10-11, 14, 20-21.

32.

See also, Barry H. Weinberg, The Resolution of Election Disputes: Legal Principles That Control Election Challenges, 2nd Edition (Washington, DC: International Foundation for Electoral Systems, 2008), pp. 85-101.

33.

For a full discussion of the Help America Vote Act, see CRS Report RL32685, Election Reform: The Help America Vote Act and Issues for Congress, by [author name scrubbed] and [author name scrubbed].

34.

For a clear description of how the FEC determined its reporting requirements and, thus the reporting requirements of the states, see Federal Election Commission, The Impact of the National Voter Registration Act of 1993 on the Administration of Elections for Federal Office 1995-1996, Washington, DC, June 30, 1997, henceforth, NVRA Report 1995-1996.

35.

NVRA Report 1995-1996; Federal Election Commission, The Impact of the National Voter Registration Act of 1993 on the Administration of Elections for Federal Office 1993-1994, Washington, DC, June 30, 1995, henceforth known as NVRA Report 1993-1994; Federal Election Commission, The Impact of the National Voter Registration Act of 1993 on the Administration of Elections for Federal Office 1997-1998, Washington, DC, June 30, 1999, henceforth known as NVRA Report 1997-1998; Federal Election Commission, The Impact of the National Voter Registration Act of 1993 on the Administration of Elections for Federal Office, 1999-2000, Washington, DC, June 30, 2001, henceforth, known as NVRA Report 1999-2000; Federal Election Commission, The Impact of the National Voter Registration Act of 1993 on the Administration of Elections for Federal Office 2001-2002, Washington, DC, June 30, 2003, henceforth, known as NVRA Report 2001-2002; Elections Assistance Commission, The Impact of the National Voter Registration Act of 1993 on the Administration of Elections for Federal Office 2003-2004, Washington, DC, June 30, 2005, henceforth, known as NVRA Report 2003-2004; Elections Assistance Office, The Impact of the National Voter Registration Act of 1993 on the Administration of Elections for Federal Office 2005-2006, Washington, DC, June 30, 2007, henceforth known as NVRA Report 2005-2006; Elections Assistance Office, The Impact of the National Voter Registration Act of 1993 on the Administration of Elections for Federal Office 2007-2008, Washington, DC, June 30, 2009, henceforth known as NVRA Report 2007-2008; Elections Assistance Office, The Impact of the National Voter Registration Act of 1993 on the Administration of Elections for Federal Office 2009-2010, Washington, DC, June 30, 2011, henceforth known as NVRA Report 2009-2010; Elections Assistance Office, The Impact of the National Voter Registration Act of 1993 on the Administration of Elections for Federal Office 2011-12, Washington, DC, June 30, 2013, henceforth known as NVRA Report 2011-2012.

36.

NVRA Report 1995-1996, p. 1.

37.

NVRA Report 1995-1996, pp. 1-2.

38.

NVRA Report 1995-1996, pp. 2-3.

39.

NVRA Report 1997-1998, p. 1.

40.

NVRA Report 1997-1998, pp. 1-2.

41.

NVRA Report 1999-2000, p. 1.

42.

NVRA Report 1999-2000, pp. 1-2.

43.

NVRA Report 1999-2000, pp. 3-4.

44.

NVRA Report 2001-2002, p. 1.

45.

NVRA Report 2001-2002, pp. 1-2.

46.

NVRA Report 2001-2002, p. 14.

47.

See NVRA Report 1995-1996, pp. 37-43, NVRA Report 1997-1998, pp. 21-26, NVRA Report 1999-2000, pp. 29-35. Also, on May 4, 1995, in testimony before the Senate Subcommittee on Treasury, Postal Service, and General Government of the Senate Appropriations Committee, Postmaster General Marvin Runyon noted that the NVRA cost the agency $2.1 million dollars that year and described it as an "unfunded mandate" for the Postal Service. Under the NVRA, the Postal Service is required to provide the states with "third-class" rates, available to nonprofit organizations, for voter registration mailings. However, several states have complained that it is almost impossible to comply with certain provisions of the NVRA relating to voter list maintenance because the Postal Service does not, under nonprofit mailing rates, normally return non-forwardable mailings to the sender. Non-forwardable mailings can be used as a first step in determining if a citizen has moved from an old address under the NVRA.

48.

NVRA Report 1999-2000, pp. 35-37.

49.

NVRA Report 2001-2002, p. 3.

50.

NVRA Report 2003-2004, p. 1-2, 9-12.

51.

NVRA Report 2003-2004, pp. 9-12.

52.

NVRA Report 2003-2004, pp. 13-14.

53.

NVRA Report 2005-2006, p. 1.

54.

NVRA Report 2005-2006, pp. 1-2.

55.

NVRA Report 2005-2006, pp. 13-14.

56.

NVRA Report 2007-2008, p. 29.

57.

NVRA Report 2007-2008, p. 1.

58.

NVRA Report 2007-2008, p. 1.

59.

NVRA Report 2007-2008, pp. 1-2.

60.

NVRA Report 2007-2008, p. 1.

61.

NVRA Report 2007-2008, p. 39.

62.

NVRA Report 2007-2008, pp. 8-9.

63.

NVRA Report 2009-2010, p. 29.

64.

NVRA Report 2009-2010, p. 1 and NVRA Report 2005-2006, p. 2.

65.

NVRA Report 2009-2010, p. 1.

66.

NVRA Report 2009-2010, p. 2.

67.

NVRA Report 2009-2010, p. 1.

68.

NVRA Report 2009-2010, Table 2a. Application Sources: Total Forms Received, pp. 38-39.

69.

NVRA Report 2009-2010, p. 9.

70.

NVRA Report 2011-2012, "Table 1a. Registration History," pp. 14-31. Figures were computed for the 44 states and the District of Columbia by the author.

71.

NVRA Report 2011-2012, p. 1.

72.

NVRA Report 2011-2012, p. 2.

73.

NVRA Report 2011-2012, p. 2.

74.

NVRA Report 2011-2012, p. 6.

75.

NVRA Report 2011-2012, "Table 2a. Application Sources: Total Forms Received," pp. 40-41.

76.

NVRA Report 2011-2012, "Table 2a. Application Sources: Total Forms Received," pp. 40-41.

77.

NVRA Report 2011-2012, pp. 9-10.

78.

U.S. v. New York, 3 F. Supp. 2d 298 (E.D. NY); U.S v. New York, 255 F. Supp. 2c 73 (E.D. NY); 2007 U.S. Dist. LEXIS 21722.

79.

2006 U.S. Dist. LEXIS 45640.

80.

2006 U.S. Dist. LEXIS 32499.

81.

Based on a LEXIS search covering the period, searching for the "National Voter Registration Act."

82.

Michael Slater, "Compliance with the NVRA: Not Optional," The National Voter, February 2008, pp. 11-12, http://www.lwv.org/AM/Template.cfm?Section=Home&Template=/TaggedPage/TaggedPageDisplay.cfm&TPLID=19&ContentID=10530.

83.

Applications received from state voter registration agencies in the 1995-1996 period constituted 11% of all applications received. The same was true in the 2003-2004 period and the 2005-2006 periods, see NVRA Report, 1995-1996, pp. 1-2; NVRA Report 2003-2004, pp. 9-12; NVRA Report 2005-2006, pp. 1-2. However, during the 2007-2008 period this percentage dropped to 6.3%, see NVRA Report 2007-2008, p. 39. For the 2009-2010 election cycle, this percentage dropped even further to 5.6%, see NVRA Report 2009-2010, p. 39. For the most recent election cycle, this figure is 6.3%, see NVRA Report 2011-2012, p. 41.

84.

See Jordy Yager, "Schumer Wants States Sued Over Voter Registration," http://thehill.com/leading-the-news/schumer-wants-states-sued-over-voter-registration-2009-04-07.html. Also, see Sen. Schumer's website at http://schumer.senate.gov/new_website/record.cfm?id=311271. Also, Senator Charles Schumer, "Schumer Urges DOJ: Sue States That Flout Voter Registration Law," press release, April 7, 2009, http://schumer.senate.gov/new_website/record.cfm?id=311271.

85.

See, "Project Vote Applauds Justice Department Enforcement of NVRA in Illinois," http://www.reuters.com/article/pressRelease/idUS53510+20-Dec-2008+PRN20081220.

86.

Young et al. v. Fordice et al. 520 U.S. 273 (1997), see http://www.law.cornell.edu/supct/html/95-2031.ZS.html.

87.

Richard G. Smolka, "Election Officials Liaison Group Advised of Legislative Opportunities, Postal Rate Increase," in Election Administration Reports, vol. 27 (Washington: Jan. 13, 1997), pp. 2-4; Richard G. Smolka, "JEOLC Hears Congressional Staff Outline Plans for Legislation," in Election Administration Reports, vol. 28 (Washington: Jan. 19, 1998), p. 5.

88.

Richard G. Smolka, "NASS Survey Identifies Gains and Some Problems with NVRA," in Election Administration Reports, vol. 27 (Washington: Feb. 13, 1997), pp. 1-3.

89.

NVRA Report 2011-2012, Table 1a. Registration History, p. 14-31. The percentages are based on the figures in the column labeled "Reported Registration." Empty cells were completed by values from CRS Report 96-932, Voter Registration and Turnout: 1948-1994, by [author name scrubbed] (archived; available from the author). The table was then recomputed by the author based only on the 44 states and the District of Columbia covered by the NVRA. See ****.

90.

NVRA Report 2011-2012, Table 1a. Registration History, p. 14-31.

91.

For the CVAP values for 2008, 2010, and 2012, see NVRA Report 2011-2012, Table 1a. Registration History, p. 14-31. For all other years, the values were derived from Professor Michael McDonald's percentages shown for the non-citizen population from his spreadsheet for the full general election series from 1980-2012, see United States Election Project, Voter Turnout, Google Doc spreadsheet for the full general election series from 1980-2012, http://elections.gmu.edu/voter_turnout.htm.

92.

For all years, the values were derived from Professor Michael McDonald's spreadsheet for the full general election series from 1980-2012, see United States Election Project, Voter Turnout, Google Doc spreadsheet for the full general election series from 1980-2012, http://elections.gmu.edu/voter_turnout.htm.

93.

See ****, below.

94.

Turnout figures come from Michael P. McDonald, United States Election Project—Voter Turnout, Excel Turnout Spreadsheet, "Turnout 1980-2012.xls," http://elections.gmu.edu/voter_turnout.htm. The author computed turnout for the states covered by NVRA provisions from McDonald's spreadsheet. Turnout figures are based on 'vote totals for highest office." It should be noted that McDonald does not recommend using the voting-age population as the divisor for computing turnout, preferring his Voter Eligible Population measure. This is, without question a better measure, especially when making comparisons between states and countries. However, for historical consistency with the figures used by the FEC (EAC), this author chose to include figures based on the voting-age population, as well. See Appendix A, Table A-2.

95.

Ibid., see McDonald, United States Election Project—Voter Turnout, Excel Turnout Spreadsheet, "Turnout 1980-2012.xls," http://elections.gmu.edu/voter_turnout.htm. For the mid-term elections in these states, see Appendix A, Table A-2.

96.

For the legislative history of the 104th to 111th Congresses with respect to voter registration proposed legislation, see Appendix B.

97.

Alan Greenblat, "Court Rejects 'Motor Voter' Case, But the Battle Isn't Over," Congressional Quarterly, weekly report, vol. 54 (Washington: Jan. 27, 1996), p. 232.

98.

John T. Willis, "Committee on House Administration Holds Hearing on Voter Efficiency Act," Election Administration Reports, vol. 43, no. 12 (June 10, 2013), pp. 3-4.

99.

Wendy Weiser, Michael Waldman, and Renee Paradis, Voter Registration Modernization, Brennan Center for Justice, New York University School of Law, New York, NY, 2009, http://www.brennancenter.org/content/resource/universal_voter_registration_draft_summary/; FairVote, "Universal Voter Registration, http://www.fairvote.org/?page=65; Progressive States Network, "Universal Voter Registration: A New Initiative to Increase Electoral Participation and Reduce Voter Suppression, http://www.progressivestates.org/node/22476.

100.

Common Cause, "Voter Registration," http://www.commoncause.org/site/pp.asp?c=dkLNK1MQIwG&b=4923169.