Comment on Requests to Include Additional Proof-of-Citizenship Instructions

Comment on Requests to Include Additional Proof-of-Citizenship Instructions

Dēmos hereby submits these comments in response to the Request for Public Comment by the U.S Election Assistance Commission regarding the question of whether to amend the State-specific instructions applicable to Arizona, Kansas, and Georgia on the National Mail Voter Registration Form (“Federal Form”), published in the Federal Register on December 19, 2013.


The U.S. Election Assistance Commission seeks comment on the requests of Arizona, Kansas, and Georgia to modify their State-specific instructions on the Federal Form to include State law requirements that, as a precondition to registering to vote in Federal elections, voter registration applicants provide additional documentary (or other) proof of their United States citizenship beyond that already required by the Federal Form.

The EAC should deny the states’ request, which will exacerbate an already-existing problem. The exclusion of eligible citizens from the political process is one of the most serious and stubborn problems in our democracy. Census data indicate that less than two-thirds of adult citizens nationwide are registered to vote. More troublingly, registration rates are significantly lower for particular groups, including low-income persons, racial and ethnic minorities, the young, and naturalized citizens.

Congress enacted the National Voter Registration Act (“NVRA”) to increase the number of registered and participating voters. The NVRA accomplishes this purpose in part by facilitating community-based voter registration drives through the use of a uniform federal mail-in voter registration form, “with particular emphasis on making the [form] available for organized voter registration programs.”1 The requests by Arizona, Kansas, and Georgia to include additional proof-of-citizenship instructions on the Federal Form frustrate the purpose and operation of the NVRA because they interfere with the community-based voter registration efforts the NVRA aims to facilitate. The result is that many qualified citizens will not be able to register to vote, thereby depriving them of the ability to participate in our nation’s political process. The EAC should reject the states’ requests to encroach upon the NVRA and to hinder community-based voter registration efforts.


Including additional State-specific documentary (and other) proof-of-citizenship instructions on the Federal Form will impede community-based registration efforts because they would require potential new registrants to produce forms of identification – such as a driver’s license, a current U.S. passport, a birth certificate, naturalization documents, or certain Bureau of Indian Affairs and tribal identification documents – that many people do not carry with them and that many other people do not possess at all. Importantly, Arizona’s experience with documentary proof-of-citizenship requirements, enacted in 2005 as part of Proposition 200, demonstrates the severe negative consequences on efforts to increase voter registration. Those consequences are tangible, and their extension to other states is unacceptable.

A. The Additional Proof-Of-Citizenship Instructions Requested By Arizona, Kansas, And Georgia Require Documents That Many People Do Not Carry With Them And Indeed May Not Possess At All

The documentation requirements sought to be incorporated into the Federal Form manifestly inhibit effective community-based voter registration efforts. It is easy enough to see why: the laws of those states require potential registrants to produce papers that many people do not carry with them as they conduct their day-to-day affairs – precisely when registration drives seek to reach them – and that many people to not possess at all.2

Although many individuals carry their driver’s licenses, far fewer keep a birth certificate, passport, naturalization documents, or tribal documentation on their persons on a regular basis. In fact, some people keep their birth certificates or passports in safe deposit boxes, requiring a trip to the bank during regular business hours. Likewise, students who move for school may leave important paperwork at their parents’ homes and be unable to register until they retrieve it. Furthermore, even if an individual has the necessary documentation with her when approached at a registration drive, she may be unwilling to provide her birth certificate or passport to a stranger, given very legitimate concerns that it will be lost or stolen or that identity theft will occur.

Further compounding the problem is the fact that many otherwise eligible citizens simply do not possess the types of documentation required under the Arizona, Kansas, and Georgia laws. In Arizona, for example, the most common form of accepted documentation is an Arizona’s driver’s license. However, approximately 10% of voting age citizens in the State do not possess an Arizona driver’s license. Moreover, many Arizonans who do have a driver’s license cannot use it as documentation because, prior to October 1996, Arizona did not require driver’s license applicants to provide proof that they were lawfully present in the United States.3 Thus, licenses issued prior to that date cannot be used to fulfill the State’s documentation requirement. Similar problems confront citizens who obtained their driver’s licenses before they were naturalized because Motor Vehicle Department records reflect citizenship status on the date the license was issued, unless the licensee requests an updated license by providing their naturalization certificate and paying a fee.

The other acceptable forms of documentation under the states’ laws—such as a U.S. passport, a birth certificate, naturalization documents, or certain Bureau of Indian Affairs and tribal identification documents—pose additional problems for community-based registration organizations because individuals are less likely to possess them or be able to use them to fulfill the documentation requirements. With regard to U.S. passports, for example, the State Department reports that there were approximately 113 million passports in circulation in 2012.4 The total citizen population in 2010 was almost 287 million, meaning that only 39% of U.S. citizens held passports.5 Moreover, passport ownership is strongly correlated with both income and education level, meaning that the underrepresented populations targeted by community-based voter registration efforts are less likely to hold valid passports.6

Birth certificates, while held by more individuals, pose their own problems. For example, some groups of citizens born outside of hospitals – including people born in rural areas or on Native American reservations – are less likely to have received birth certificates. Elderly citizens are also unlikely to be able to rely on a birth certificate to satisfy documentation requirements because birth certificates were not consistently generated until sometime in the 1930s.7 Women may also be disproportionately impacted. Nationwide, less than half of voting age women who have ready access to their birth certificates have ones that reflect their current legal names.8 Thus, to the extent that the potential voter’s name must match exactly the name on her birth certificate, even women who have a valid birth certificate may not be able to satisfy documentation requirements.

B. The Documentation Requirements of Arizona’s Proposition 200 Have Already Adversely Affected Community-Based Voter Registration Efforts.

The negative consequences of additional documentary proof-of-citizenship requirements are not just hypothetical. Arizona’s experience after the enactment of its proof-of-citizenship requirement is a harbinger of what may materialize should the EAC grant the requests of Arizona, Kansas, and Georgia. During the twenty months following the enactment of Arizona’s Proposition 200 in January 2005, at least 31,500 registration applications were denied because of a failure to meet the law’s requirements.9 Of those 31,500 individuals whose application for registration was initially denied, approximately 11,000 were able to register successfully—the remaining 20,000 individuals did not subsequently make it onto the rolls. Notably, the Hispanic population was overrepresented relative to its share of the population both in the group of individuals whose registration applications were initially denied and in the subset of individuals who did not subsequently register successfully.

Proposition 200 had an especially pronounced effect on community-based voter registration efforts. According to documents filed in the Inter Tribal Council case, after the law went into effect, registration through community-based voter drives dropped 44% in Arizona’s largest county. Numerous individuals and groups involved in voter registration drives testified at trial that they encountered difficulty in registering individuals who did not have any of the acceptable forms of documentation or who did not have their documents with them when they attempted to register. The organizations also testified that because Proposition 200 requires photocopies of certain types of documentation, effective registration requires the group to have a copier or scanner on site. This requirement both restricts the types of locations where drives can be held and increases the cost of conducting such events. The costs of copying documents are compounded by the additional time—and therefore additional resources – needed to explain registration requirements, assist with filling out forms, track down or copy necessary documentation, and follow-up on applications. Indeed, several groups testified that they encountered so many difficulties in registering individuals under Proposition 200 that they ceased their voter registration efforts entirely in Arizona after its passage.

Many of Dēmos’ clients in the Inter Tribal Council amicus brief experienced significant difficulty in conducting registration drives in Arizona following the passage of Proposition 200, difficulty that was relieved through use of the current version of the Federal Form. As just one example, during the 2012 election cycle, the Border Action Network held registration drives at a variety of community gathering spots, including libraries, a swap meet, and a school. Many who attempted to register at those events decided to do so only when they encountered the registration drive at the event. As a result, many of those who tried to register were not carrying one of the forms of documentation acceptable under Proposition 200. Many of the individuals Border Action Network attempted to assist with registration used the bus for transportation and did not have a driver’s license or other form of documentation with them that they could use to register. And even when they did, the registration volunteers did not have a photocopier with them and could not make copies of individuals’ documents. Notably, when Border Action Network provided registrants with the Federal Form, which did not require documentation of citizenship, it was significantly easier for individuals to complete and submit their applications.

The experience of voter registration organizations operating under Proposition 200 in Arizona strongly suggests that granting the requests of Arizona, Kansas, and Georgia will hinder community-based voter registration drives and prevent eligible citizens from registering to vote.


Voter registration rates in the United States remain stubbornly low and some population groups, such as communities of color, low-income citizens, the young, and naturalized citizens have even lower registration rates. Community-based voter registration efforts are designed to reach those groups that are underrepresented on our registration rolls. They aim to eliminate the most common barriers to registration by going to potential voters and facilitating the registration process—including by explaining the requirements and providing assistance filling out forms. It is these very efforts that will be impeded should the EAC grant the states’ request.

The registered voter population in Arizona, Kansas, and Georgia, and nationally is substantially smaller than the qualified voter pool, and the problem is measurably worse among groups historically underrepresented in the electorate. For example, in Arizona in 2012, only 65.2% of voting age citizens were registered to vote.10 For Black and Hispanic citizens in Arizona, registration rates were even lower: only 58.6% of Black citizens and 52.2% of Hispanic citizens were registered to vote, compared with 70.5% of White, non-Hispanic citizens.11 Age also played a significant role, with only 55.3% of 18-24 year-old Arizonans registered to vote.12

Kansas and Georgia fare only slightly better. In 2012, 74.4% of Kansas citizens were registered to vote but only 62.5% and 50% of Black and Hispanic citizens, respectively, were registered.13 Likewise, Georgia had a statewide registration rate of 70.7%, and while 72.2% of Black citizens were registered, only 59% of Hispanic citizens were registered.14

Nationally, the picture is similar. In 2012, only 71.2% of voting-age citizens reported being registered to vote.15 Again, race and ethnicity were significant factors, especially in the case of Hispanic citizens: 73.7% of White, non-Hispanic citizens were registered compared with 73.1% of Black citizens, and only 58.7% of Hispanic citizens.16

Income is also strongly associated with registration rates. Only 61.8% of voting-age citizens with a family income of less than $20,000 were registered in 2012, while 87.1% of those with a family income of $150,000 or more were registered.17 Finally, there is a substantial disparity in registration rates of native-born versus naturalized citizens. Seventy percent of native-born citizens of voting age reported being registered in 2012, compared with only 62.1% of naturalized citizens.18

Community-based registration efforts have proven highly effective at reaching these unregistered citizens. In fact, “[f]rom 2000 to 2008, community-based groups registered tens of millions of new voters, including close to nine million in 2008 alone.”19 These efforts came from non-partisan groups as well as organizations across the political spectrum. Presidential campaigns, large national organizations, and small local groups—many targeting specific, underrepresented segments of the population—held voter registration drives at locations ranging from churches to senior centers to farmers’ markets to school campuses.

A significant portion of voters have registered through community-based methods. In 2012, 5.5% of the electorate reported registering at a school, hospital, or on campus, while an additional 5.0% reported using a registration booth.20 Moreover, 13.1% of registrants reported registering by mail. Because community-based registration efforts overwhelmingly use mail-in applications, some significant number of these registrations is likely attributable to community-based voter registration efforts.

Among groups with low registration rates, community-based registration methods played an even more important role. Black, Asian, and Hispanic citizens all reported higher than average use of school, hospital, and on-campus registration methods (7.4%, 6.4%, and 8.1%, respectively), as well as higher than average use of registration booths (8.1%, 6.4%, and 6.9%, respectively). Naturalized citizens also reported greater usage of community-based methods of registration than did native-born citizens. And finally, among people aged 18 to 24—the age group for which registration rates are lowest—13.9% of registered voters reported registering at a school, hospital, or on campus.

Not surprisingly given their success, federal law favors registration drives. In fact, one federal court has recognized:

[T]he NVRA encourages voter-registration drives; the NVRA requires a state to accept voter- registration applications collected at such a drive and mailed in to a voter registration office; the NVRA gives a voter registration organization like each of the plaintiffs here a ‘legally protected interest’ in seeing that this is done; and when a state adopts measures that have the practice effect of preventing an organization from conducting a drive, collecting applications, and mailing them in, the state violates the NVRA.21

Additionally, one court – noting the expressive and associational rights implicated by voter registration activities – applied intermediate scrutiny to efforts to curb them.22 These decisions acknowledge that community-based voter registration efforts are an important mechanism of enfranchisement, and one that the NVRA itself has enshrined as an indispensable part of the electoral process. The EAC should not allow states to undermine such efforts by incorporating additional identification requirements into the Federal Form.


Voter registration rates in the United States have remained stubbornly low, especially for certain groups of citizens such as communities of color, low-income citizens, and the young. Community-based voter registration efforts are an effective way to increase registration rates. The requests of Arizona, Kansas, and Georgia for the Election Assistance Commission to modify their State- specific instructions on the Federal Form will exacerbate this problem by, inter alia, hindering community-based voter registration efforts -- frustrating both the purpose and operation of the NVRA. Our democracy can ill afford to make voter registration any more difficult. The EAC should deny the requests of Arizona, Kansas, and Georgia to include additional documentary proof-of-citizenship instructions on the Federal Form. 



1 42 U.S.C. § 1973gg-4(b).

2 For comprehensive lists of the documents that satisfy each state’s proof-of-citizenship requirement, see Ariz. Rev. Stat. § 16-166(F); Kan. Stat. Ann. § 25-2309(l); Ga. Code Ann. § 21-2-216(g).

3 See State of Ariz. Office of the Att’y Gen., Att’y Gen. Op. re: Identification Requirements for Voter Registration, at 3 (2005), available at
4 See U.S. Dep’t of State, Passport Statistics, available at stats_890.html.

5 U.S. Census Bureau, Selected Population Profile in the United States, 2009-2011 Am. Cnty. Survey 3-Year Estimates, available at /tableservices/jsf/pages/productview.xhtml?pid=ACS_11_3YR_S

6 See, e.g., Richard Florida, America’s Great Passport Divide, The Atlantic (Mar. 15, 2011), available at

7 Corey Dade, Why New Photo ID Laws Mean Some Won’t Vote, NPR (Jan. 28, 2012), available at
8 Brennan Center for Justice, Citizens Without Proof, 2 (Nov. 2006), available at

9 The information on the impact of Proposition 200 in Arizona is taken from the Joint Appendix filed by the parties in Arizona v. Inter-Tribal Council of Arizona, Inc., Docket No. 12-71.

10 U.S. Census Bureau, Voting and Registration in the Election of November 2012 – Detailed Tables (hereinafter “Voting and Registration”), Table 4a (2012), available at
11 Id., Table 4b.

12 Id., Table 4c.
13 Id., Tables 4a and 4b.

14 Id.

15 Id., Table 4a.
16 Id., Table 4b.
17 Id., Table 7.
18 Id., Table 11.
19 NAACP, Defending Democracy: Confronting Modern Barriers to Voting Rights in America, 15 (2012) (citing U.S. Census Bureau, Voting and Registration, Table 14), available at
20 U.S. Census Bureau, Voting and Registration, Table 12, available at