IS I-VOTING I-LLEGAL?

The Voting Rights Act was passed to prevent racial discrimination in all voting booths. Does the existence of a racial digital divide make Internet elections for public office merely a computer geek's pipe dream? Or can i-voting withstand scrutiny under the current state of the law? This i-Brief will consider the current state of the law, and whether disproportionate benefits will be enough to stop this extension of technology dead in its tracks.

Overview

¶          As the Internet continues its charge into homes across the world, e-government enthusiasts are exploring the viability of its application to elections. Democracy, security, and privacy concerns aside, there is uncertainty as to the legality of "i-voting." Because whites have disproportionately more access to the Internet than blacks or Hispanics, i-voting will make elections more convenient for whites relative to other racial groups. This effect opens the door for a "vote dilution" or "denial of access" claim under the Voting Rights Act. The uncertainty in voting rights law suggests that jurisdictions should carefully consider how i-voting will affect members of different racial groups.

i-Voting - e-Ventual Blessing or i-Nevitable Blunder?

¶          Internet voting, or i-voting, refers to remote voting with a computer via the Internet. I-voting is one specific form of electronic voting, or e-voting, which more broadly includes other electronic voting techniques, such as using the Internet at supervised polling sites.²

¶          Many commentators are generally optimistic about the potential reality of i-voting. Some even say its implementation is inevitable.³ Proponents of i-voting generally believe that it has tremendous potential to alleviate the current problem of low voter turnout by providing a more convenient option. Citizens would be able to cast their votes from work or at home without the need to travel during business hours to polling sites. Advocates claim that by increasing the ease and convenience of voting, the number of votes cast will increase, especially among youth, who are currently the least likely to vote⁴ 

¶          Opponents of i-voting cite many different reasons for their opposition. Most challenge i-voting because of "insurmountable" security risks, such as denial of service attacks and Trojan horse interference,⁵ while some cite concerns of voter coercion and fraud. Still others claim that i-voting will disproportionately benefit certain groups based on race and economic status.

¶          At publication, i-voting has never been used in a binding general election for public office in the United States.⁶ However, substantial pilot projects were used in England during the November 2002 local government elections.⁷ Citizens in nine local jurisdictions were allowed to vote from remote locations using the Internet, cellular phones, or handheld mobile devices.⁸ Overall, 14.6% of all participants with the remote option used the Internet; however, the percentage was much higher in some jurisdictions.⁹

¶          In the United States, i-voting has been used in smaller contexts. The most widely studied example is the Arizona Democratic Primary in March 2000, where voters were given four different ways to cast ballots, including remote Internet voting for four days prior to election day.¹⁰ Overall, 41% of voters used the Internet,¹¹ compared to 38% by mail and 21% in polling places.¹² The Pentagon has experimented with i-voting for overseas soldiers,¹³ and i-voting was used in the 2000 Alaska Republican Party presidential straw poll¹⁴ President Bush recently established a committee to look into i-voting; it is due to report back in July 2004.¹⁵

The e-Divide

¶          To exercise one's right to i-vote, one must have access to the Internet - a characteristic that is not shared equally across groups. Evidence indicates that a "digital divide" exists based upon several different factors, most notably:

• Income: 79% of families that earn over $75,000 per year use the Internet, compared to 25% of families that earn less than $15,000.¹⁶
• Race: 60% of whites use the Internet, compared to 40% of blacks and 32% of Hispanics.¹⁷

¶          Although these large disparities exist, the rates of increase among groups are beginning to equalize them. From 2000 to 2001:

• Income: Among families earning less than $15,000 per year, Internet usage increased 30%, compared to 12% among families earning over $75,000.¹⁸
• Race: Among blacks, Internet usage increased 33%; Hispanic Internet usage increased 30%, but among whites, Internet usage only increased 18%.¹⁹

¶          The data is similar for educational attainment and employment; while the less educated and unemployed are using the Internet less often, their rates of growth are larger.²⁰

¶          Data also indicate differences in where different groups access the Internet. Although data was not available for adults, race is a distinguishing factor among children. Although children do not vote, their lack of home Internet access necessarily implies a similar lack of home access for their parents. 56% of black children and 55% of Hispanic children do not access the Internet at home, compared to 80% of white children.²¹ This distinction is important because i-voting will increase convenience the most for people that access the Internet at home or work. People that do not have home access may not benefit from the i-voting option, as they still have to make a physical trip to a different site.

¶          So, although a digital divide exists based on factors such as race, the gap is narrowing. Although critics disagree as to the importance of this digital divide, its effect on i-voting cannot be ignored. At least in the short term, the wealthy and the white will be the largest beneficiaries of an i-voting system. It remains to be seen if the large growth of the lagging groups will continue until the divide is bridged.

The Voting Rights Act - e-Liminating Voter Discrimination

¶          The Voting Rights Act of 1965 (VRA)²² was passed to eliminate racial discrimination in voting.²³ Although the Fifteenth and Twenty-Fourth Amendments to the Constitution were designed to remedy this problem,²⁴ many states found creative ways to circumvent the law through discriminatory literacy tests, residency requirements, and dilution of minority voting power through changes in election systems. The VRA has been amended through the years to keep up with state attempts to discriminate.

¶          There are two conceivable ways to implement i-voting in the United States. The first is to use i- voting as a supplement to in-person ballot boxes; this would expand the available options for the potential voter to include the Internet from home, work or the local Internet café. The second, a more radical and currently unrealistic option, is to completely replace traditional voting booths with remote Internet ballots. At present, such a scheme would certainly violate the Twenty-Fourth Amendment.²⁵ As such, i-voting will be discussed as a supplemental voting option, including its potential violations of §§ 2 and 5 of the VRA.

§2 - The e-Ffects Test

Generally

¶          VRA §2(a) states,

No voting qualification or prerequisite to voting or standard, practice, or procedure shall be imposed or applied by any State or political subdivision in a manner which results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color...²⁶ 

¶          A violation of part (a) occurs when a nomination or election process is "not equally open to participation" by members of the protected classes.²⁷ For a claim to succeed, the disputed process must cause members of the protected class to have "less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice."²⁸

¶           In 1980, the Supreme Court held in Mobile v. Bolden that the VRA required evidence of discriminatory intent to invalidate a city election process. ²⁹ 

Because the burden of proving a state's intent is nearly impossible to meet, Congress amended the VRA in 1982 to remove the intent element. Therefore, a disputed election procedure need only have the effect of discrimination. The Senate defined this "results test" further by delineating a non-exhaustive list called the "Senate Factors" to be considered under the "totality of circumstances."³⁰ 

¶          In general, there are two types of claims under the VRA: vote dilution and denial of equal access.

¶           The most common claim under §2 is vote dilution, which is "a disproportionate reduction in the voting power of one segment of the electorate relative to that of another segment."³¹ The Supreme Court in Thornburg v. Gingles explained how plaintiffs make a prima facie case for vote dilution: ³² 

• Compactness: the minority group must be sufficiently numerous and compact to form a majority in a single-member district; and
• Racial bloc voting: the minority group must be politically cohesive (tend to vote alike) and the white majority must usually vote as a bloc sufficient to defeat the minority's preferred candidate.

¶          If plaintiffs can show these two factors, they must convince the fact finder, using the aforementioned "totality of circumstances" test, that the procedure has racially discriminatory effects. Section 2 broadly applies to elections in all states and political jurisdictions, including state judges³³ and local school boards³⁴ regardless of any history of past discrimination in voting in that jurisdiction.

¶          A less clear alternative cause of action under §2 is a denial of equal access claim. These claims rely on the plain language of §2, which states that all election procedures must be equally open to all racial groups, and no aspect of an election can in any way limit one's right to vote on the basis of race.³⁵ The Supreme Court stated in Thornburg that §2 prohibits all forms of voting discrimination, not just vote dilution.³⁶ These cases are generally judged based solely upon the "totality of circumstances" test.³⁷

¶          Although there have been relatively few denial of equal access cases compared to vote dilution, most courts have examples of denial of equal access challenges, including the location of polling places³⁸ poll worker harassment of black voters³⁹ and a jurisdiction's failure to provide absentee ballots to voters.⁴⁰

¶          Author Stephen Pershing argues that i-voting may be held invalid as a denial of equal access. His claim centers on the Supreme Court ruling in Chisom v. Roemer, which primarily stated that vote dilution claims for state judicial elections are covered by the VRA. The Court clarified that there is only one cause of action under §2, the denial of equal "opportunity to participate in the political process [and] to elect representatives of one's choice."⁴¹ However, the court's language suggests a broader reading of the VRA. The Court held that a diminution of "opportunity to participate" necessarily implies a diminution in "opportunity to elect." ⁴² So, according to Pershing, although §2 requires a showing of ability to elect, a physical denial of a voter's opportunity is proof positive of that claim.⁴³ Therefore, plaintiffs need not make a specific showing of compactness, and can rely solely on the "totality of circumstances" test.⁴⁴

¶          Although this analysis has never been fully articulated by the Supreme Court, some federal district courts have followed this line of reasoning. For example, the Middle District of Alabama held in Harris v. Siegelman that an official action is discriminatory if "minority voters cannot participate in the electoral process on the same terms and to the same extent as nonminority voters."⁴⁵ Also, the Eastern District of Pennsylvania held in Ortiz v. City of Philadelphia Office of City Commissioners that the VRA "includes all electoral practices that deny minority voters equal opportunity to participate in any phase of the political process and to elect candidates of their choice."⁴⁶

¶          Although the law is not as well settled for a denial of equal access claim as for a vote dilution claim, plaintiffs have a much simpler burden with denial of equal access. Vote dilution requires evidence of compactness and racial bloc voting, which is not present in all jurisdictions. Therefore, although whites may disproportionately benefit from i-voting, racial minority groups might find themselves without a cause of action. Denial of equal access only requires the totality of circumstances test, giving plaintiffs a much larger chance of being able to make a successful claim. .

¶          Whether or not current evidence of a racial digital divide is enough to convince a court is uncertain. Plaintiffs would need to show evidence of a localized digital divide, and prove that because i-voting serves to enhance discrimination in voting by making it more convenient for whites than blacks, the practice violates the totality of circumstances test. Since the digital divide is shrinking, plaintiffs chances may decrease as more and more blacks become connected to the Internet.

¶          Section 5 of the VRA adds another layer of scrutiny to states or political subdivisions with a history of racial discrimination.⁴⁷ These jurisdictions are required to "preclear" proposed changes to its voting systems with the U.S. Attorney General or the D.C. District Court.⁴⁸ The purpose is to place the burden of proof on these jurisdictions to show that their new election procedures are non-discriminatory.⁴⁹ Section 5 is still relevant because 8 states, and certain counties and subdivisions of 7 others, are still covered by §5.⁵⁰ 

¶          The range of enactments covered by §5 is broad; all state enactments that alter election law in even a minor way must be cleared.⁵¹ To be cleared, states must persuade the U.S. Attorney General or the D.C. District Court that the proposed laws do not have a discriminatory purpose, and will not have the effect of denying citizens the right to vote based on their race or color.⁵² 

¶          I-voting is an example of a new balloting procedure that covered jurisdictions would need to preclear. The most analogous example is Perkins v. Matthews, where the Supreme Court stated that changes in polling place locations required preclearance.⁵³ Because i-voting is basically a change of voting venue, it will certainly have to be precleared, independent of its likely effect on the voting habits of the different groups. Therefore, covered jurisdictions will have to make a preemptive showing that i-voting is going to implemented in a way that does not have a discriminatory effect.

Bottom Line - e-Nsure e-Quity

¶          Although the law is not crystal clear in the area of racial discrimination in voting, jurisdictions considering i-voting should be careful. The law prohibits practices that discriminate against racial groups, and the Internet is more accessible to whites. Therefore, i-voting may disproportionately benefit whites more than other racial groups. Does that make i-voting illegal? Not necessarily, since any claim must meet the totality of circumstances test. However, a sudden adoption of i-voting without other correlating voting system changes, such as increased poll locations in minority neighborhoods or targeted voter education campaigns, may attract suspicion. In the case of §5 jurisdictions, such a change would certainly get the Attorney General's attention. Jurisdictions eager to bring about i-voting before the digital divide diminishes should supplement i-voting with programs designed to ensure that minority voters equally enjoy the benefits of i-voting.. Will i-voting increase voter turnout? Maybe. But more importantly, those of us who already vote can receive that extra half an hour two times a year. Isn't technology great?

Footnotes