BUILDING CONSENSUS ON ELECTION REFORM...launched an election reform initiative by conven-ing the...

A R E P O RT O F

T H E C O N S T I T U T I O N P R O J E C T ’ SF O R U M O N E L E C T I O N R E F O R M

August 2001

B U I L D I N G C O N S E N S U S O N E L E C T I O N R E F O R M

Election Reform Initiative Staff and ConsultantsVirginia Sloan, Executive Director, The Constitution ProjectZoe Hudson, Director, Election Reform Initiative, The Constitution ProjectTracy Warren, Senior Policy Analyst, Election Reform Initiative, The Constitution Project

Carlos Angulo, Legislative Counsel, Zuckerman Spaeder, LLPMichael Davidson, former U.S. Senate Legal CounselMickey Edwards, John Quincy Adams Lecturer in Legislative Politics,

Kennedy School of Government, Harvard UniversityMorton H. Halperin, Senior Fellow, Council on Foreign RelationsPamela Karlan, Kenneth and Harle Montgomery Professor of Public Interest Law,

Stanford University Law SchoolScott Sanders, High Noon CommunicationsRonald Weich, Legislative Counsel, Zuckerman Spaeder, LLP

Additional Constitution Project Staff Naomi V. Collins, Program CoordinatorLaura D. Compton, Policy Analyst Elizabeth Dahl, Deputy DirectorTimothy S. Kolly, Communications DirectorBarbara Reed, Counsel and Policy DirectorPedro G. Ribeiro, Policy and Technology Associate Kimberly Westhoff, Communications Assistant

1717 Massachusetts Avenue NWSuite 800Washington DC 20036

202-662-4240 (until August 8, 2001)202-299-9540 (starting August 8, 2001)

© August 2001

A B O U T T H E C O N S T I T U T I O N P R O J E C T

The Constitution Project, based at Georgetown University’s Public Policy Institute, Washington, DC, isa bipartisan nonprofit organization that seeks consensus on controversial constitutional and legalissues through a unique combination of scholarship and activism. The Constitution Project currentlyhas four initiatives: on the death penalty, judicial independence, the process to amend the U.S.Constitution, and election reform.

T H E C O N S T I T U T I O N P R O J E C T • P R E - P U B L I C A T I O N D R A F T

A R E P O RT O F

T H E C O N S T I T U T I O N P R O J E C T ’ S F O R U M O N E L E C T I O N R E F O R M

August 2001

Chair, Forum on Election ReformMorton H. Halperin, Board Member, Constitution Project

Counsel and ReporterMichael Davidson

Working Group ChairsStephen Ansolabehere, Massachusetts Institute of Technology*Marlene Cohn, League of Women Voters Education FundNorman Ornstein, American Enterprise Institute*Trevor Potter, The Reform InstituteRichard Soudriette, International Foundation for Election Systems

The election reform initiative is supported through a generous grant from the William and Flora Hewlett Foundation.

*Affiliation is for the purpose of identification only.

B U I L D I N G C O N S E N S U S O N E L E C T I O N R E F O R M


i

C H A I R’ S P R E FA C E . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .iii

S U M M A RY O F C O N C LU S I O N S A N D R E C O M M E N DAT I O N S . . . . . . . . . . . . . . . . . . . . .iv

I N T R O D U C T I O N . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .ix

I . E S S E N T I A L E L E M E N T S O F R E F O R M

A. Before Election Day . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11. Voter Education and Election Personnel Training . . . . . . . . . . . . . . . . . . . . . . . .22. Voting Technology Research, Standards, Testing and Clearinghouses . . . . . . . .43. Registration Systems

B. Election Day . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .51. Accessibility and Staffing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .62. Posted Notices of Rights and Responsibilities . . . . . . . . . . . . . . . . . . . . . . . . . . .83. Preserving the Rights of Voters Who Come to the Polls . . . . . . . . . . . . . . . . . . .8

a. Closing Hour Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8b. Provisional Ballots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8

4. Vote Casting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9a. Overvotes and Undervotes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10b. Accessibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11c. Ballot Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11d. Auditability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11

C. After the Polls Close . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .131. Timelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .132. What Constitutes a Vote . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .133. Manual Recounts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .144. Audit Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .155. All Votes Should Be Counted . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15

R E P O RT O F T H E F O R U M O N E L E C T I O N R E F O R M

August 2001

C O N T E N T S


D. Alternative Methods of Voting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .151. Internet Voting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .162. Voting Entirely by Mail . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .163. Absentee Voting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16

a. For Persons Unable to Vote at Polling Places . . . . . . . . . . . . . . . . . . . .16b. Unlimited Absentee Voting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16

4. Early Voting at Election Offices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17

E. Top-to-Bottom Review of State Election Codes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18

I I . P R O P O S A L S F O R C O N G R E S S I O N A L A C T I O N

A. Federal Assistance for Research and Technology Standards . . . . . . . . . . . . . . . . . . . . .20

B. Federal Grants for Capital Investment in Voting Systems Technology and Use . . . . . .221. Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .232. Duration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .233. Allotments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .244. Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .255. Conditions Related to Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .266. Additional Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .267. Reporting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .278. Federal Agency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .279. Appropriations for the Grant Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2810. Oversight . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .29

C. A Permanent Program to Defray Expenses of Federal Elections . . . . . . . . . . . . . . . . . .29

C O N C LU S I O N . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .31

A P P E N D I X A : PA RT I C I PA N T S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .33

A P P E N D I X B : S U P P L E M E N TA L V I E W S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35

A P P E N D I X C : R E S O U R C E S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .41

ii


iii

In 1997, Virginia Sloan and I founded theConstitution Project with the goal of develop-ing and promoting bipartisan solutions tocontemporary constitutional and legal issues.

Now based at Georgetown University inWashington, D.C., the Project continues to operateon the belief that building consensus among indi-viduals and groups with different perspectives iscritical to the democratic process. We bring thisconviction to our work on election reform.

In February 2001, The Constitution Projectlaunched an election reform initiative by conven-ing the Forum on Election Reform. We, alongwith many others, were motivated to helpaddress the deficiencies in our nation’s electionsystem that were brought to light in the 2000elections. We felt that there was an opportunityto implement needed reforms, but that there wassome danger that the issue could be cast in apartisan manner. For that reason, we set out toidentify and forge consensus between individualsof both political parties, and organizations of allkinds with an interest in reform. Our goal wassimple: to ensure that eligible voters are able tovote and to have their vote counted accurately.We specifically excluded, as part of this initialeffort, larger questions about election reformsuch as the role of the electoral college or how toincrease voter participation.

The Constitution Project invited state and localelected officials, other officials who run elec-tions, advocates for voters, and experts in rele-vant fields to participate in the Forum onElection Reform. What follows is the result ofthat five-month effort – a report that identifieswhat we believe are the major points of agree-ment between the participants. Our hope is thatthis partnership will increase the chances fortimely and responsible action by the Congressand state legislatures.

This initiative was made possible by a grant fromThe William and Flora Hewlett Foundation. Weare grateful to Paul Brest, President of the

Foundation, for his early support and commit-ment to improving the American election system.

A number of people deserve recognition. Wedeeply appreciate the dedication of the workinggroup chairs who provided expertise, workedclosely with participants, and guided the formu-lation of our recommendations: StephenAnsolabehere, Massachusetts Institute ofTechnology (Technology); Marlene Cohn, Leagueof Women Voters Education Fund (Education);Norman Ornstein, American Enterprise Institute(Voting Procedures); Trevor Potter, The ReformInstitute (Federal, State and Local Roles); andRichard Soudriette, International Foundation forElection Systems (Vote Counting).

Our report reflects the hard work of a coregroup of individuals. Michael Davidson assistedthe working groups in preparing their reportsand the Forum in preparing its report. PamelaKarlan of Stanford University Law School pro-vided legal guidance about constitutional issuesrelating to election reform. Mickey Edwards ofHarvard University’s Kennedy School ofGovernment and a Constitution Project boardmember contributed substantial insight andguidance throughout. Ronald Weich and CarlosAngulo of the law firm of Zuckerman Spaeder,LLP furnished helpful analysis about existingand proposed legislation and pending litigation.Zoe Hudson, Director of the Election ReformInitiative, and Tracy Warren, Senior PolicyAnalyst, kept the entire undertaking on track.

Finally, we owe our deepest gratitude to the manyparticipants in our Forum who took time to engagein a thoughtful debate over how to improve the con-duct of elections in the United States. We hope theycontinue to be our partners over the coming yearsas the nation turns to implementing election reform.

Morton H. HalperinChair, Forum on Election ReformBoard Member, Constitution ProjectAugust 2001

C H A I R’ SP R E FA C E

E S S E N T I A L E L E M E N T S O F R E F O R M

A. Before Election Day

1. Voter Education and Election PersonnelTraining: Sustained education efforts are need-ed to ensure an informed electorate and trainedelection personnel. Election officials have pri-mary responsibility for voter education, but theyshould also enlist others, beginning with theschools. Voter education should begin beforeelection day, providing information to votersabout how, when, and where to register and vote;how to update their address and confirm regis-tration status; and identification requirements.Voters should be mailed sample ballots, instruc-tions about the mechanics of voting, and noticeof their rights and responsibilities.Correspondingly, election personnel shouldreceive training in legal requirements and theoperation of voting equipment. Overall, stateofficials should have a plan to assure that votereducation and election personnel training com-mands the attention they merit throughout thestate.

2. Voting Technology Research, Standards,Testing, and Clearinghouses: A system for fos-tering development of voting technologies is anessential foundation of a sound election system.Such a system should include: research on devel-opment of technologies that advance importantobjectives of our the election system (such asaccessibility and equipment ease-of-use); stan-dards for the design and performance of equip-ment to meet those objectives; testing to assurethat equipment meets standards; and clearing-houses to collect and exchange information aboutthe development and performance of voting systems.

In addition to technology improvements, the ben-efits of research, identification of best practices,and information clearinghouses also apply to abroad range of other election administrationissues.

3. Registration Systems: All states shoulddevelop statewide registration databases, asnow exist in some states. Accuracy of registra-tion information should be maintained throughintegration or improved communicationsbetween voter registration and other databases,such as motor vehicle department records. Astate’s database should be available electronical-ly at polling places on election day for timelyiv


S U M M A RY O FC O N C LU S I O N S A N D R E C O M M E N DAT I O N S

resolution of registration questions. Anyprocess to remove ineligible voters from regis-tration lists should be non-partisan, be in com-pliance with voting laws, provide notice to votersthey have been removed from the rolls, andafford them an opportunity to correct erroneousinformation. General programs to purge listsshould be completed sufficiently in advance ofelection day to allow individuals to correct erro-neous information.

B. Election Day

1. Accessibility and Staffing: Polling placesshould be fully accessible and accessibilityshould be broadly defined. It should includeselection of polling places that allow access forvoters with limited mobility and are convenientto the communities they serve. Materials,including directions to polling places, should beavailable in multiple languages and formats.Longer voting hours can be critical in makingvoting at polls accessible for all voters. It is alsoa worthwhile long-term goal to work toward asystem that would allow people to vote at pollingplaces close to work.

Additional resources should be provided for thehiring and training of election day personnel. Toincrease the number of poll workers, the follow-ing should be considered: split schedules; use ofhigh school students; recruitment of retired peo-ple and other potential part-time employees; timeoff with pay for public and private employees;and cooperative efforts with civic groups.

2. Posted Notices of Rights andResponsibilities: To provide a common point ofreference for election officials and voters inresolving disputes, there should be prominentnotice in every polling place of applicable federaland state election law. It should include thevoter’s rights to a provisional ballot, a new ballot

if a mistake has been made, assistance in voting,and a demonstration of the equipment. It shouldalso include any relevant information, such asidentification requirements and any time limit onvoting.

3. Preserving the Rights of Voters Who Cometo the Polls: Voters in line by poll closing timeshould be allowed to cast a ballot. If a voter’sname does not appear on the registration list,and the voter affirms he or she is entitled to vote,the voter should be entitled to submit a provi-sional ballot that will be counted if the voter isdetermined to be an eligible voter. Voters shouldbe notified whether the ballot was counted.

4. Vote Casting: Together with good ballotdesign, technologies should be used that enablevoters to avoid error and record their choicesaccurately. Technologies that provide voterswith an opportunity to correct overvotes orundervotes should be used, as should technolo-gies that enable disabled voters to vote inde-pendently and therefore secretly. Voting technol-ogy should be flexible enough to allow states tochoose among a variety of ballot methods. Theability of election officials to conduct an audit ofthe original count should be considered in thedesign and selection of voting technologies.

C. After the Polls Close

State election calendars should allow sufficienttime for all counting and contest procedures tobe completed in time for presidential electors tocast the state’s vote.

Each state should define what is a valid vote. Asa matter of democratic principle, state lawshould establish a general rule that places avalue on determining whether a voter’s choice isclearly discernible. To apply that general rule,the state’s chief election authority should be v


given authority to adopt regulations, through apublic rule-making procedure, for addressingrecurring anomalies associated with particularvoting methods. States should establish clearrules for manual recounts, which should be con-ducted uniformly across the jurisdiction of theelection.

States should provide for pre- and post-electionaudits of equipment to assure integrity of thefinal count. Every validly cast vote should becounted, including those submitted by militaryand other absentee voters and provisional bal-lots submitted by qualified voters.

D. Alternate Methods of Voting

Election day voting at polling places provides thebest opportunity to achieve five objectives: assurethe secrecy of the ballot and protect againstcoerced voting; verify that ballots are cast only byregistered voters; safeguard ballots against lossor alteration; assure their prompt counting; andfoster the communal aspect of citizens votingtogether on the same day, so as to benefit from acommon pool of public information.

No form of alternative voting – Internet voting,voting entirely by mail, unlimited absentee vot-ing, and early voting at election offices – hasbeen devised that can provide every one of thesebenefits. Of these alternatives, early voting atelection offices is the most consistent with thesefundamental objectives.

No matter how states resolve questions aboutalternative forms of voting, it is essential to havea hospitable and efficient system of absenteevoting with protections against fraud or otherabuse for important segments of our populationunable to cast votes at polling places. Theseinclude persons in military or civilian serviceoverseas or voters who by reason of age or dis-

ability are unable to vote at polling places.

E. Top-to-Bottom Review of State Election Codes

Each state should review its election code toensure that it is easily usable by participants inthe voting process, clear to the courts, and com-prehensible to the public. The review shouldtake into consideration when uniform statewiderequirements are needed to assure equal protec-tion. State reviews should also consider otherissues such as reinstating voting rights for peo-ple who have completed criminal sentences, min-imizing partisan influences in election adminis-tration, and consolidating elections in order toreduce their frequency.

P R O P O S A L S F O RC O N G R E S S I O N A L A C T I O N

A. Federal Assistance for Research and Technology Standards

Congress should provide authority and funds forthe following:

1) research and development on votingequipment and equipment standards,with particular emphasis on ease-of-use,accessibility for people with disabilitiesor low levels of English literacy, and spe-cial issues relating to electronic equip-ment, including the ability to audit elec-tion results. A priority should be placedon the development of open source codeand architecture in all voting softwareso that it can be subject to broad scruti-ny to assure accuracy and integrity;

vi


2) an expanded standards program thatincludes management or operationalstandards, and performance or designstandards to optimize ease-of-use;

3) an expanded testing program to assurethat voting machinery complies withestablished standards; and

4) a clearinghouse allowing states andindustry to share experiences with theperformance of voting technologies.

B. Federal Grants for Capital Investment in Voting Systems Technology and Use

Congress should establish a multi-year capitalinvestment grant program for investment in vot-ing technology improvements, including fundsfor training in the use of technologies.

1. Scope: The grant program should includefunding for improved registration systems,including statewide databases and communica-tion with polling places; precinct-level voting andcounting equipment, including equipment thatallows voters with disabilities to vote independ-ently; and election personnel training and votereducation about the use of voting technologies.

2. Duration: Congress should establish a dura-tion for the program that provides states with theability to stage investments but does not undulyprolong the time for discernible improvement. To that end, a program that permits systematicimplementation of changes over the next threefederal election cycles should be considered.

3. Allotments: In determining how to allocatefunds among the states, Congress should givepreference to a grant program that is principallyformula-based, most likely according to votingage population. A formula-based program will

encourage participation by all states, and facili-tate an orderly planning process in the states byassuring the timely and regular receipt of funds.To encourage innovation, a portion of the pro-gram should be reserved for grants for pilotstate or local programs that are awarded on acompetitive basis.

4. Applications: Each state and its local gov-ernments should work together to formulate,with an opportunity for public comment, a planthat the state submits to the federal government.The plan should describe how federal funds willaddress identified needs, how the grant will helpthe state meet existing federal requirements,and how the state will assure equitable use offederal funds within the state. With respect toregistration, the plan should describe efforts tomaintain complete and accurate lists and to pro-tect the rights of registrants. Each state shouldalso provide assurances that new funds will sup-plement rather than lower current spending onelections, and that its plan does not conflict withfederal law.

5. Conditions Related to Technology: Newtechnology purchases should comply with thevoting systems standards in existence at thetime of a purchase. Each state that receives fed-eral grants should commit, during the life of thegrant program, to provide at least one votingdevice at each polling station that allows sight-impaired voters to vote independently.

6. Additional Requirements: Congress shouldprovide for two additional measures in federalelections, at least as a condition for federalgrants: an opportunity to vote by provisional bal-lot if registration status cannot be determinedon election day, and clear notification at pollingplaces of the rights and responsibilities of votersunder applicable federal and state law.

vii


7. Reporting: To assist in evaluating whetherfederal grants are improving the administrationof elections, states should regularly provide sta-tistical information on the performance of newand existing voting technologies. At the end of afunding period, each state should publicly reportwhat it has done with grants it has received.

8. Federal Agency: In selecting an existingagency or establishing a new one to carry outthe research, standards development, and grantfunctions under an election reform act, Congressshould vest final responsibility in a singleagency. It should provide for an independentline-item appropriation so that funds for electionpurposes are protected against competingdemands. To strengthen public confidence, theagency should be independent of partisan influ-ences, and be guided by an advisory board thatreflects the viewpoints of key participants in theelection process. It should be organized to makedecisions in a timely manner.

9. Appropriations for the Grant Program.Congress should authorize and appropriate suf-ficient funds to provide a significant incentive tostates to participate in the grant program and toenable them to make necessary improvements.During the first year, after analyzing state plansthe agency charged with responsibility for theprogram should submit to Congress a well-sub-stantiated projection for the fiscal requirementsof the full grant program.

C. A Permanent Program to DefrayExpenses of Federal Elections

While other areas merit additional resources –such as voter education and training of electionpersonnel – there is not agreement at this timeon a permanent federal role in funding the con-duct of federal elections. However, there isbroad agreement that to assist states and localgovernments in part of their voter education pro-grams (such as sending voters sample ballots)and in complying with election mail require-ments for registration, Congress should estab-lish a new class of postage for official electionmail that provides first class service at half therate.

viii


Early this year the Constitution Projectbegan an initiative to consider measuresfor achieving election reform. Asdescribed in the preface to this report, the

Constitution Project seeks to formulate and pro-mote, through scholarship and public education,bipartisan solutions to contemporary issues. Of thepublic issues that merit the nation’s attention, thetask of ensuring that our democracy is well servedby our election system must be among the foremost.

The Project invited participation in a Forum forElection Reform of representatives of organiza-tions of state and local officials who are respon-sible for running elections, private groups con-cerned about voting rights, and experts in tech-nology, politics, and law. From the outset, thework of the Forum has been premised on a con-viction that a partnership among these partici-pants would enhance the nation’s opportunity toimplement necessary reforms.

Members of the Forum participated in five work-ing groups. Four of these groups consideredaspects of the voting process: voter education,voting procedures, technology, and vote count-ing. A fifth group focused on the allocation offederal, state, and local roles in the electionprocess. The chairs of these groups prepared

written reports to the Forum that were posted onour website and discussed at Forum meetings.As they proceeded, the working groups modifiedtheir recommendations in light of commentsfrom Forum members and others.

The five working group reports were then integrat-ed into a report to the Forum. As a continuing partof the Project’s commitment to an open process,the combined report was posted on the Projectwebsite and discussed and modified in the courseof three meetings of the Forum. What follows isthe report of the Forum on Election Reform.

As many recognize, improved voting technologiesshould be part of a broader effort to assure thatall eligible voters are able to vote and to havetheir votes counted accurately, all on terms of fullequality. In Part I, the report highlights essentialelements of reform in each major stage of the vot-ing process: before election day, at the polls, andin counting votes. These are interrelated, notisolated stages; sound measures in one shouldincrease the chance of success in another. Thereport recognizes that some reforms described inPart I will require additional resources beyondthe new federal resources that we propose.There is an appreciable challenge to each level ofAmerican governance – federal, state, and local – ix


I N T R O D U C T I O N

to commit the resources that are needed to pro-vide the nation’s election system with theresources it requires.

The recommendations in Part I address mattersprincipally related to responsibilities of state andlocal governments, although some reforms rec-ommended in Part I also establish the predicatefor congressional action. Part II addresses theneed for Congress to aid in the development andacquisition of improved voting technologies, andthe manner in which that aid should be provided.

The report represents the best efforts of ourworking group chairs, individual Forum partici-pants, and mine as the reporter, to identify themain points of substantial agreement among theForum’s diverse participants. It is offered to thepublic with the conviction that it represents sig-nificant consensus on areas of importance to thefuture of American elections.

Our confidence that there are common themeson which to base an attainable and productiveelection reform agenda has been bolstered asthe conclusions of other studies are announced.Throughout the year state task forces and offi-cials have released recommendations. In May,the National Association of Counties (NACo) andthe National Association of County Recorders,Election Officials and Clerks (NACRC) issued thereport of their National Commission on ElectionStandards and Reform. The Caltech-MIT VotingTechnology Project has issued its report on vot-ing system improvements. The NationalAssociation of Secretaries of State has releaseda resolution on election reform. President Fordand Carter’s National Commission on ElectionReform has just reported. The Election Center’sNational Task Force on Election Reform andother bodies of state and local officials will soonreport. The broad agreement that we havefound within our Forum mirrors a comparable

unity that is emerging in these major studies.

Nevertheless, a caveat is in order. At the end ofthe report we list the participants in our Forum.On any particular issue one or several partici-pants may have a different view or emphasis.Many in the Forum are active participants in thelegislative process now underway in Congressand in the states. In that process, some willexpress a position on one or another issue thatmore precisely accords with that individual’s ororganization’s exact views. Participation in theForum does not indicate that each person ororganization agrees with every particular in thereport. Indeed, we have welcomed additionalstatements from participants in the Forum,which are set forth in Appendix B to this report.

Finally, the work of our Forum has convinced usthat while a great deal of work lies ahead, there isreason to be optimistic about the future. TheAmerican election system rests on the strongfoundation of our nation’s constitutional andstatutory guarantees and the commitment of stateand local election officials. That foundation, thespecial and expanding opportunities that moderntechnology provides, and – importantly – an hon-est recognition of the challenges we face, combineto make the promise of reform bright.

To turn that promise into reality, continuing effortsto expand consensus, including among partici-pants in our Forum, will be essential. Also of criti-cal importance will be a willingness of America’spolitical leadership to build on public consensus,lay partisanship aside, and work to the commongoal of improving our nation’s electoral system.

Michael DavidsonCounsel and ReporterThe Constitution Project’s ElectionReform InitiativeAugust 2001

x


Improvement of our election system requiresattention to each major stage of the votingprocess: measures applicable to steps thatmainly precede election day, measures that

apply directly to election day and procedures atthe polls, and rules and procedures for countingand recounting votes. In Part I of this report, wetake each stage in turn.

A. Before Election Day

1 . V O T E R E D U C AT I O N A N D E L E C T I O NP E R S O N N E L T R A I N I N G

This past presidential election brought togethermore than one hundred million voters and over amillion full- time and electionday officials.Informed participation is critical in enabling thisgreat volume of people to work together to pro-duce an election in which all qualified votershave an opportunity to vote and have their choic-es accurately recorded and counted. To thatend, voter education and training for electionpersonnel are indispensable. Both should besustained efforts. Both require commitment ofenhanced resources.

The goal of voter education should be to providevoters with the information they need at each

step of the election process in order to exercisetheir franchise successfully. A sound programshould start well before election day. It shouldprovide voters with timely information abouthow to register, confirm their registration status,and keep it up to date; where and when to vote;and how to operate voting devices correctly inorder to cast a valid vote that accurately reflectseach voter’s intentions. Throughout the votingprocess, voters should be made aware of theirrights and responsibilities as voters.

The principal responsibility for voter educationlies with election officials, but it would be a mis-take to think they have the sole responsibility.Voter education should be a continuous processthat begins with civic education in schools. Theeducation of students can also involve theirassistance at the polls, thereby helping toaddress the problem of a shrinking pool of elec-tion-day workers. In addition to schools, elec-tion officials should engage a broad spectrum ofpublic and non-governmental organizations toplay an educational role.

While many voters can and should be reached atthe places to which they go in their daily lives,such as shopping malls or other communitygathering spots, all voters should be reached at 1


PART

II . E S S E N T I A LE L E M E N T SO F R E F O R M

home. Sample ballots that enable voters to seeand study their choices as they will face them onelection day, with clear instructions about themechanics of voting and information about voterrights and responsibilities, are a fundamentaltool of a sound voter education program. Forsome voters, who – because of language, age, ordisability – need information in different for-mats, there should be well-designed methods toreach and communicate with them. For that rea-son, attention should be given to providing infor-mation in multiple formats and languages, usingall forms of media – print, radio, and television.

Our recommendation does not address nonparti-san voter guides, for which we have heard sub-stantial support, only because the focus of ourproject is on the voting process and not onbroader questions about the extent or quality ofvoter information in making electoral choices.

Overall, there should be a plan for voter educa-tion. Our society has professions and skills,whether in education or public advertising, thatcan be brought to bear. In this area, as in othersconcerning election administration, it is impor-tant for state officials to assure that voter educa-tion commands the attention it merits through-out the state. Federal officials should assist bybroadly disseminating information, includinginstructional guides, on the requirements of fed-eral law.

Correspondingly, election personnel, both thosewhose profession it is and the larger numberwho are recruited for election day, deserve thetraining required to perform what is an appre-ciable task. They need to know essential thingsabout federal and state law and about the opera-tion of voting systems, each of which inevitablyevolves and requires periodic updating of earliertraining. While training of election-day person-nel is likely to continue to be the principal

responsibility of the local officials whom theywill directly assist, states should assure that full-time election officials receive the training thattheir responsibilities require.

2 . V O T I N G T E C H N O L O G Y R E S E A R C H ,S TA N D A R D S , T E S T I N G , A N DC L E A R I N G H O U S E S

An integrated system for fostering developmentand sound use of improved voting technologies isan essential underpinning of a sound electionsystem. There should be four interconnectedelements:

• research on development of technologiesthat advance important objectives of ourelection system;

• standards for the design and perform-ance of equipment to meet those objec-tives;

• testing to assure that equipment actuallymeets these standards; and

• clearinghouses to collect and exchangeinformation about the development andperformance of voting systems.

These functions should be funded on a long-termbasis in recognition that voting technologies willbe continually developing.

Among participants in The Forum, there is broadagreement that the Federal ElectionCommission’s 1990 voluntary engineering andperformance standards need first to be broughtup to date (as the FEC is now doing) and thenkept current in response to technologicalchanges in a rapidly developing field. Thosestandards should be expanded to add voluntarymanagement or operational standards that

2


include such matters as maintaining sensitiveelectronic equipment. There is also strong sup-port for voluntary performance or design stan-dards to optimize ease-of-use and minimizevoter confusion. The latter are often describedas human factor standards, which take intoaccount how voters interact with technology.

The preceding section describes the importanceof voter education. Although clear informationshould be posted and demonstration machinesmade available, there are limits to the amount ofeducating about voting machines that can bedone at polling places. Preferable designs areones in which the way to vote accurately isapparent to voters and therefore requires littleinstruction. Technologies that require pollingplace instruction introduce a risk of inaccuracybecause often there is insufficient time on elec-tion day to provide that instruction, particularlyat peak voting times.

In developing standards, care should be given soas not to inadvertently create barriers to innova-tion. For some purposes, standards may beexpressed as minimum criteria for satisfactoryperformance in meeting particular voting systemgoals. For other purposes, standards may bespecifications that set forth exact features. Foryet others, progress may be promoted by identi-fying best practices and providing clearinghous-es to inform the public about experiences inusing different technologies. The public authori-ty that has responsibility for issuing standardsshould have discretion to select the form mostsuitable for its purposes.

Another barrier to avoid is a testing bottleneck.The certification process sponsored by theNational Association of State Election Directorsis recognized to be a good platform for assessingequipment durability and detecting errors insoftware. But as electronic equipment evolves,

speed of certification will be an important con-cern, especially for newer firms. Slow certifica-tions would act as a barrier to competition in thevoting equipment industry. A public authoritythat is responsible for the overall system of basicsupport described in this section should facili-tate use of multiple laboratories for testing hard-ware and software.

Long-term public funding for research and devel-opment on voting equipment and equipmentstandards. There should be particular emphasison ease-of-use, as well as accessibility for peoplewith disabilities, low levels of English literacy orprincipal literacy in another language. A priori-ty is also research on the security of electronicvoting. Attainment of these objectives, all ofwhich serve broad public and democratic values,should not depend solely on research budgets ofequipment manufacturers. An example is thedevelopment of technologies to enable disabledvoters to vote privately and independently.

Much of this research should be conducted undergrants to universities and other research centers,under the overall plan and superintendence of apublic authority to assure the soundness andintegrity of the process. In conducting researchand establishing standards, it is important togive weight to the interest of states in having sev-eral options for selecting equipment so that thenation’s election system is not tied to the vulner-abilities or imperfections of a single system.

One product of publicly-assisted research shouldbe development of open source code and archi-tecture for voting software. This would allow theinner working of vote casting and tabulatingmachines to be subject to broad scrutiny inorder to assure accuracy and integrity. Pendingprogress toward that objective, the current sys-tem of testing laboratories, whose examinationof software is facilitated by non-disclosure

3


agreements and software escrow, will continueto be needed to assure software correctness.Software standards should ensure correctness ateach stage of the voting process from vote cast-ing through vote counting.

There should be a clearinghouse of informationabout equipment performance in practice. Boththe states and the federal government shouldregularly collect and report on data about theincidence of such matters as overvotes andundervotes, as well as about other aspects ofequipment use and experience. This informationshould be readily available to industry, state andlocal election officials, and the public. A govern-mental authority should have the responsibilityto assure the quality of the data and the objectiv-ity of the reports issued.

Finally, other aspects of elections and electionadministration should be the subject of research,best practices, and clearinghouse exchanges ofinformation. This includes such matters aspolling hours, voter education, and election offi-cial training, to name a few.

3 . R E G I S T R AT I O N S Y S T E M S

Improvements in registration are essential toenfranchisement, efficient voting, and theintegrity of the voting process. Accurate regis-tration records and prompt availability of themat polling places will facilitate voting by eligiblecitizens, including by enabling election officialsand voters to concentrate on voting. Improvingthe technology for managing registration sys-tems should also enable election officials toshorten the time between their state’s registra-tion deadline and election day, which will help toensure that registration requirements are notbarriers to participation.

All states should develop statewide electronicregistration databases, as now exist in somestates. Some states will establish a single data-base. Others may connect county databasesupon assuring that they are compatible and maybe linked successfully with each other. In main-taining accuracy, statewide databases should beintegrated or at least have improved communica-tions with other databases. These include thoseof voter registration agencies (particularly statemotor vehicle records and social services agen-cies), U.S. Postal Service change-of-addressrecords, and state or local agencies that collectvital statistics. There could also be links toother states to correct records such as when avoter moves from one state to another. To makethis information useful on election day, statewideregistration databases should be electronicallyaccessible from polling places. There should alsobe improved communications between pollingplaces and higher election officials in order toresolve registration questions expeditiously.

For many states, a major effort must be under-taken to bring voter registration lists up to date.To ensure fairness and credibility, election offi-cials should involve non-partisan experts as wellas adopt other procedures that eliminate anyperception of political motivation in the designor implementation of programs to removedeceased or ineligible voters from registrationrolls. Any removal process should be consistentwith voting rights laws and have safeguards toensure that eligible voters are not removed.

New technologies can be a boon. But the power-ful tools of information technology, if poorlyapplied, can produce incorrect results that jeop-ardize legitimate expectations of validly regis-tered voters. Of course, the first line of defenseis a statewide system that eliminates occasionsfor sudden, large-scale, and error-prone purgesby regularly and reliably updating records

4


through the integration of new information.

Technology can carry only part of the burden ofmaking registration records more accurate.Improved administrative procedures in voterregistration agencies are essential. It is alsoessential that voters be able to act promptly toprevent mistakes in the handling of their regis-trations. Registrars should give voters promptnotification when they have been removed fromthe registration rolls because the registrars havereceived information that the voters are ineligi-ble. In response, voters should have an opportu-nity to correct erroneous information. To makethat opportunity fully meaningful, general pro-grams to purge lists (where the risks of errorare greatest) should be completed sufficiently inadvance of election day (as, for example, 90days) so that notices can be sent to voters whomay then respond in time to resolve registrationquestions prior to the election.

Our working groups did not address whetherregistration records should be made more pre-cise by use of an identifying number that isunique to each registrant. For example, theFEC, pursuant to its responsibility under theNational Voter Registration Act, recommendsusing a piece of the Social Security number. ThePrivacy Act of 1974 prohibits states from requir-ing use of a full Social Security number for voterregistration unless they had done so prior toJanuary 1975. Seven states now require fullSocial Security numbers; two states require thelast four digits. Seventeen other states requestfull numbers; three request the last four num-bers. The FEC has recommended that statesrequire use of the last four digits for new regis-trations, and request that information from cur-rent registrants.1 In the FEC’s view, the combi-nation of a voter’s name, date of birth, and thelast four Social Security digits would get statesas close as practical to a unique personal identi-

fier for each voter while still protecting votersfrom release of full numbers.

Each state will make its own evaluation. In par-ticular states, it may be important whether thestate already uses Social Security informationfor public record keeping, such as for motorvehicle records. The various experiences ofstates that have been using all or part of SocialSecurity numbers in their registration systemsdeserves evaluation. Ultimately, states will needto weigh the benefit of using a part of a regis-trant’s Social Security number to establish moreprecise registration records against privacy con-cerns and any resulting disinclination to regis-ter. Whatever course a state chooses for its reg-istration system, a registered voter’s ability tocast a vote should not be contingent on remem-bering and providing an identification number toelection officials on election day.

B. Election Day

No subject has attracted as much debate withinthe Forum as the question whether or whatkinds of general alternatives to election daypolling places – such as early voting at pollingplaces, voting by mail or Internet voting – shouldbe encouraged or discouraged. That debate willbe described in a subsequent section, which willalso discuss particular questions such as votingopportunities of military personnel and citizensliving overseas.

Whatever differences may exist about alterna-tives to voting at polls on election day, there is,we believe, overwhelming support for the propo-sition that voting at the polls serves basic andhistorically rooted objectives. The gathering ofcitizens to vote is a fundamental act of communi-ty and citizenship. It provides the greatest secu-rity for enabling voters to cast their ballots freeof coercion. It facilitates prompt counting and 5


verification of results, which is especially impor-tant in presidential elections given the constitu-tional and statutory time constraints in resolvingany disputes about them.

For these reasons, it is essential to directresources to improve voting at polling places onelection day. Every step should be taken tomake that a pleasant, accessible, expeditious,and efficient process. No matter how statesresolve questions about alternative methods ofvoting, the national priority should be to correctdeficiencies in the election-day, polling-placeexperience of voters.

Election day can present two very different pic-tures. One is of a remarkable event in whichvoters and election personnel succeed togetherin producing a crowning event of democracy. Fortoo many Americans, however, the experience ofvoting on election day is marred by long lines atpeak hours before and after work, insufficientnumbers of personnel at their precincts, incom-plete or inaccurate voter registration informa-tion, an inadequate number of voting machinesor places, poorly maintained or aging votingmachines, a lack of accessibility for voters withdisabilities, poorly translated materials, andconfusing ballots. In the preceding section ofthis report, we addressed the need to improveregistration systems. In this section we will dis-cuss accessibility, voting systems (which alsoinvolve accessibility questions), posted notices ofvoter rights and responsibilities, and provisionalballots.

1 . A C C E S S I B I L I T Y A N D S TA F F I N G

Polling places should be fully accessible. One hun-dred percent accessibility should be the goal.Accessibility should be defined broadly. It shouldinclude selection of polling places that are accessi-ble by persons who utilize wheelchairs, are visually

impaired, or whose other disabilities or age limittheir ability to enter and move about the buildingsin which polling places are located. It shouldinclude accessibility of polling places to the com-munities they serve, with respect to the adequacyof their number, location, and availability of publictransportation in urban areas. Limited Englishproficiency can also be a barrier that should beaddressed comprehensively, from directions topolling places to the languages used in materials inthem. Training of election personnel to be of assis-tance to voters with disabilities or who speak otherlanguages other than English is also important.(Accessibility of voting machines for voters withdisabilities will be discussed below in 4 (b).)

Convenient access may also be increased by work-ing toward methods by which persons may vote,–within the jurisdiction in which they reside – atpolling places near where they work, and havetheir votes transmitted to the locations where theylive. Secure local area networks (LANs) – in con-trast to difficult to secure Internet voting – may bea feasible avenue for making voting at the pollsavailable to those whose working hours now makevoting at their home precincts a serious ordeal.We recognize that technological advances will berequired to allow voters to vote at near-to-workpolling places for some bottom-of-ballot officesthat appear on the voter’s home precinct ballot.Despite the challenges, we believe that develop-ment of that technological capacity is a worth-while objective meriting funding at least in experi-mental ways in the short term.

For military and overseas voters, the feasibilityof establishing polling places at U.S. military anddiplomatic facilities should also be studied. Ifmilitary or other overseas service or occupationmakes absentee voting a necessity, it should noterase the opportunity to join fellow citizens incasting a secret ballot at a polling place.

6


A major accessibility problems in our electionsystem is the time available for voting on elec-tion day. There are, of course, important bal-ances to be struck, because additional time willrequire added resources, including for the hiringand training of poll workers. We recognize thatstate and local election officials cannot simplydecree extended voting hours and hire and trainpoll workers. To make these things happenrequires a substantial infusion of funds. Westrongly recommend allocation of increasedresources for these purposes.

There is no consensus now on proposals forshifting an existing holiday (e.g., Presidents’Day) for use as a new uniform federal electionday, or for having one or two days of weekendvoting. One concern about a major change, suchas establishment of an election day holiday, isthat it may result in lower turnouts, if manyAmericans (too many of whom are now tenuous-ly engaged in our political process) simplydecide to take a holiday. For that reason, anymove to holiday or weekend voting should beadopted only on a trial basis so its impact can beevaluated before a decision is made whether itshould become permanent.

Greater accessibility should also be achieved byan increase in polling hours to accommodate ourgrowing population and varied daily schedules.Our voting procedures working group recom-mended a nationwide norm of 15 hours; thesehours are used by New York State, which keepsits polls open from 6 a.m. to 9 p.m. The precisenumber of hours of an expanded schedule shouldbe studied further. We recognize that mandatinglonger voting hours will require serious study ofpoll worker arrangements and working hours,risks of inaccuracy due to fatigue, and otherissues, not the least being cost. Nevertheless,longer voting hours can be a critical element inachieving the basic goal of making voting at the

polls on election day accessible for all voters.Every effort should be made to achieve this goalas expeditiously as possible.

Hiring and training election day officials is diffi-cult as it is. It will be more difficult, even withadditional money, if efforts are made to expandtheir numbers. Still, a number of approachesmerit consideration, such as experimentationwith split schedules for poll workers, use of highschool students, and more aggressive recruit-ment of retired people and other potential part-time employees. Election officials should workwith civic organizations and with businesses andgovernmental units, which should be encouragedto give employees time off with pay to volunteerat the polls.

Since many schools are closed on election day inNovember, teachers, administrators, and highschool students are available resources to be ofassistance at polling places. School districtscould provide academic credit for students whovolunteer to work at the polls, and extra time offfor teachers and administrative personnel whodo so. States that require poll workers to be ofvoting age could relax that rule to accommodatehigh school students. For these and other elec-tion innovations, a clearinghouse should be usedto share information both on what works andwhat does not.

Finally, states and localities should considerwhether an increase in pay can help to recruitmore poll workers. Enlarging the pool of pollworkers may also allow the recruitment of morepoll workers who, on the basis of prior trainingor aptitude, are able to provide the requiredlevel of service. In the long run, increasing thenumber of available poll workers will help inefforts to meet minimum training requirements.

7


2 . P O S T E D N O T I C E S O F R I G H T S A N DR E S P O N S I B I L I T I E S

Although most voter education should precedeelection day, there is an indispensable form ofelection day education that is mutually impor-tant both to voters and election officials. Aclearly written, prominently posted statement ofprincipal rights and responsibilities can providean easily available, common public point of ref-erence to resolve most polling place issues. Itshould include the major requirements of federaland state laws as they apply to individual voters.It should cover such matters as a voter’s right toa provisional ballot, to receive a replacementballot to correct a mistake before submitting theballot for counting, and a demonstration of thevoting process. The posted statement shouldalso include information on any identificationrequirements or time limits on voting.Consistent with previous recommendations, itshould be available in multiple formats so thatthe information is accessible to blind voters andvoters with limited English proficiency.

3 . P R E S E RV I N G T H E R I G H T S O FV O T E R S W H O C O M E T O T H E P O L L S

Among a voter’s most disappointing experiencesmust be that of getting to the polls and thenbeing precluded from casting a ballot.

a. Closing Hour Lines

On account of his or her work schedule or forother reasons, a voter might arrive at the end ofthe voting day while the polling place is open butnot reach the front of the line before the pollsclose. Lines at the end of a day may result fromfactors well beyond the voter’s control, includingan inadequate number of voting machines ormachine breakdowns. The principle should be

clear: if a voter is in line by the poll closing time,he or she should be allowed to cast a ballot evenif that ballot is cast after the polls have officiallyclosed. No voter who shows up within the hoursthat polls are open should be turned away.

b. Provisional Ballots

A voter may discover that polling place recordsdo not show his or her voter registration. Amotor vehicle department or other registrationagency might not have forwarded registrationinformation to election officials. Or election offi-cials may have canceled a registration on receiv-ing incorrect information wrongfully attributingto a voter a disqualifying circumstance, such asa criminal conviction. Integrated statewide voterdatabases will reduce the occurrence of suchproblems. Additionally, electronic access tostatewide databases and improved communica-tions to higher election officials should helpresolve, on the spot, many registration ques-tions. But some will not be resolvable on elec-tion day.

Some states allow voters who moved and whosenames are not on the rolls at their new pollingplace to cast a ballot. Some have extended thispractice to cover any voter claiming to be regis-tered whose name does not appear on the rolls.But a broader reform is needed. If a voter isturned away and leaves a polling place withoutvoting or filling out a provisional ballot, thatvoter’s opportunity to vote will be irretrievablylost even if the facts show on further inquiry thatthe voter is qualified. In the event that registration questions cannotbe promptly resolved on election day, voters, at aminimum, should have an opportunity to submitprovisional ballots that will constitute their votesif it is determined that they are qualified. Aprominent feature of a voter bill of rights posted

8


at polling places should be information to votersabout their right to submit provisional ballots. Ifa registration question arises that cannot beresolved that day, it is good practice to use thevoter’s request for a provisional ballot as a reg-istration application or to offer the voter anopportunity to fill out a registration form, evenas efforts are made to resolve whether the voteris entitled to vote in the election and at thatpolling place. At the very least, the voter shouldbe secure in knowing that the registration ques-tion will not arise at the next election.

In some places, state laws or administrative rulesallow election authorities to include provisionalballots in the initial election-day count. If thoseballots are not counted on election day, electionofficials should determine in the days immediatelyafter whether the voter was entitled to vote andthe ballot should be counted. In making thatdetermination, they should check relevant records,including motor vehicle records or records ofother voting registration agencies that may havefailed to transmit registration information to voterregistrars. A voter should be informed whetherhis or her provisional ballot was counted.

In some states, key functions of provisional bal-lots may be served by devices of a differentname, such as affidavit, conditional, or fail-safevoting. The key, of course, is not the name, butthe substance of the procedure afforded to vot-ers to preserve their opportunity to vote. It mayalso be that procedures in some states, such asaffidavit balloting, are more protective of votersthan provisional balloting, if they allow voters tovote (actually, not provisionally) on showingbasic identification, affirming the fact of theirresidency, and attesting they made a good faitheffort to register. A recommendation for provi-sional balloting should not be construed as anargument for cutting back on existing protec-tions. Instead, provisional balloting procedures

should be a floor from which to proceed in anyjurisdiction that turns voters away if a registra-tion question is not resolvable on the basis ofpolling place records.

Provisional ballots, used when the name of avoter does not appear on the register, should bedistinguished from ballots cast by voters whoare on the rolls but whose eligibility is chal-lenged. For example, a party worker mightclaim that the voter should not have been regis-tered because of a lack of citizenship or that thevoter is not the person who was registered.States should take steps to ensure that any chal-lenge process will not be used to intimidate vot-ers or to manufacture a prolonged contest.

4 . V O T E C A S T I N G

In our discussions about vote casting, four objec-tives have been stressed:

1) to employ vote casting technologies andgood ballot design that enhance the abil-ity of voters to record intended choicesaccurately; avoid negation of votes bymistakenly casting more votes for anoffice than permitted (overvoting); andavoid inadvertently failing to cast a votefor an office (undervoting);

2) to provide voting technologies accessibleto voters with disabilities or personswith limited English language literacy,that enable them to vote privately andtherefore secretly;

3) to use voting technologies that are flexi-ble enough to accommodate the variousballot methods currently in use; and

4) to use voting technologies that allow forauditability, namely, the ability to recon- 9


struct each voter’s original vote in theevent of an election contest.

a. Overvotes and Undervotes

Over the last four presidential elections, approxi-mately two percent of all ballots cast were notcounted as having recorded a valid presidentialvote because they were unmarked, marked formore than one candidate, or marked in anotherway that led to their not being counted.2 Thatrate has not changed over this period eventhough new technologies have been introduced.For some voting methods and some localitiesthat rate is higher; for others, lower. Those dif-ferences present significant issues of equity. Buteven if the two percent rate had been evenlyspread throughout the nation, and bore no rela-tion to the wealth or racial or ethnic characteris-tics of various communities, two percent of100,000,000 voters in a presidential election is2,000,000 voters. That is an unacceptably highnumber. We should do better.

It now seems clear within the election communi-ty and among independent analysts thatunmarked or spoiled ballots are due more todesigns that lead to mistakes in voting than tothe physical breakdown of equipment, althoughimproved maintenance is often needed. There isstrong evidence that failures associated with vot-ing equipment are produced by two things. Oneis the difficulty or confusion that too often arisesfrom the way in which voters are presentedchoices through the design of the ballot or atouch screen. To reduce confusion, it is impor-tant to bring to bear a high level of professional,contemporary understanding of how good designcan assist voters in making accurate choices andbeing confident that they have made them.

The other strong contributor to mistakes is the

lack of timely feedback to voters that wouldallow them to correct inadvertent errors beforesubmitting their votes for counting. Some exist-ing technologies offer especially poor feedback.Other than determining whether chads havebeen removed, it is particularly difficult for vot-ers to determine whether they have properlyrecorded their votes on punch cards that do notinclude the names of candidates.

Two promising techniques are precinct scanningfor optical or punch systems and ballot review inelectronic machines. These allow voters tocheck if their ballots are properly marked beforesubmitting them to be counted. If a voter learnsbefore finally casting a vote that he or she mis-takenly overvoted or undervoted, the votershould have a chance to correct the mistake.The field evidence is that second chance votingopportunities enable voters to reduce mistakes,in contrast to counting at central locationswhere ballots cast at precincts are transportedand where voters no longer can correct mis-takes. We believe officials should employ sys-tems that enable voters to check for balloterrors.

Overvoting, typically the casting of votes for twocandidates for the same office when only onevote is permitted, is almost invariably a mistake.In contrast, undervoting, not casting a vote foran office or ballot question, may be a voter’sdeliberate decision. A voter may choose toabstain from voting for a high office or, moregenerally, experience fatigue by the time he orshe reaches local offices on a long ballot.

Owing to the potential difference between over-votes and undervotes, a question has beenraised whether election officials should exercisediscretion in deciding whether to provide anopportunity for ballot review in both circum-stances, or if the opportunity should be limited

10


to notification of overvotes. A concern is thatvoting times and consequent lines will be length-ened if every voter is required to go through asecond step of ascertaining whether the voterhas made a mistake in completing his or her bal-lot. In this regard, the point to be stressed isthat ballot review is an opportunity for voters todetermine whether they have made a mistake.Voters may choose to bypass that chance. Also,in implementing any system of ballot review, vot-ing devices should be developed over the longterm that can limit and flag errors without theintervention of poll workers, in order to protectthe sanctity of the secret ballot.

b. Accessibility

In addition to the physical accessibility of pollingplaces and the other matters covered previously,the goal of full accessibility applies to each stepin the voting process from registration to votecasting. In the past, voters with poor or novision, motor impairments, or low levels ofEnglish literacy generally have not been able tovote without assistance.

Numbers of citizens will continue to need andtherefore have a right to assistance in voting.But the goal of full accessibility also places ahigh value on the ability of each voter at everypolling place to vote independently and thereforesecretly – as all citizens should be able to do.There are important and encouraging develop-ments. Some equipment vendors or election offi-cials now offer recordings that guide blind votersthrough electronic voting sessions or providetactile guides for voting with optically scannedpaper ballots. Some electronic voting machinescan be programmed in multiple languages forthe nation’s language minorities.

Further developments are needed. Technology

provides the potential. Law and public resourcesshould be enlisted to fulfill that potential.

c. Ballot Methods

In addition to employing vote-casting technolo-gies that enhance the ability to vote accuratelyand privately, voting technology should also beflexible enough to enable states to choose amonga variety of ballot methods. In all federal elec-tions and most others, voters are asked to votefor a single candidates for an office. But in vari-ous state or local elections, voters may beallowed to vote for more than one candidate foran office, such as voting for two at-large countyor city council representatives. But other for-mats are possible, and have their strong advo-cates, such as formats that allocate more thanone vote for a single candidate, as in cumulativevoting; or rank candidates in order of choice, asin choice voting and instant runoff voting.

This report does not take a position on the mer-its of particular ballot methods. Our point isonly that voting technologies should have thecapacity to accommodate various ballot meth-ods. The development and selection of a particu-lar voting technology ought not to impede thepossibility of subsequent legislative adoption ormodification of the kinds of ballot choices thatshould be made available to voters.

d. Auditability

The administration of elections involves a mas-sive computing task. Votes are cast locally inprecincts on dedicated machines. The votes castfor each office or ballot question must be tabu-lated quickly and accurately. Sometimes votesare tallied in the precinct and transmitted to acentral location; sometimes the ballots them-selves are transferred to a central location and 11


tallied there. The opportunities for errors inaggregating the vote or for corruption of thecount seem inevitable. In the event of an allegedproblem with the vote tally, officials must verifythe accuracy of the count. The ability to recon-struct the original vote – to conduct an audit ofthe original count – is termed the auditability ofthe system.

Voting technologies differ in the ability of elec-tion officials to audit election results. Systemsbased on paper (hand-counted paper and opti-cally-scanned or punch card ballots) have anadvantage over direct recording technologies(lever machines and electronic voting machines)with respect to auditability. A tally of hand-counted paper ballots or an electronic count ofpunch card or optically-scanned ballots may beaudited by recounting the original ballotsmarked by the voters. Judgment about voterintent is sometimes required, as ballots mayhave stray marks or voters may have markedtheir ballots in an unconventional manner.Nonetheless, there is an initial statement of thevoters’ intent that is separate from and remainsafter the counting process. Of course, paper-based ballots can be altered after they have beencast, and accurate vote verification can be achallenge with regard to any voting technologyor method.

In contrast to paper-based ballots, votes cast onlever or electronic machines are directly record-ed on them. Individual votes cast on levermachines cannot be audited. While the countrecorded on the back of the machine can be veri-fied, votes may be lost if a lever machine breaksdown. Many direct recording electronicmachines have the same problem. Some nowproduce internal tapes of each voter’s voting ses-sion, but programming failures or fraud mayaffect data recorded on the internal tapes aswell. However, if voting data from direct record-

ing electronic machines is altered after electionday, the alterations can be detected by compar-ing the data to multiple and independently saveddata sets.

A further audit issue is the ability to observe acount. Historically, in the United States, the pri-mary method for guaranteeing accurate countsis openness. Allowing campaign representativesand the press to observe a count introduceschecks on errors. Campaign observers common-ly catch transcription errors in the recording oftallies from the backs of lever machines. Theincreased use of electronic counting proceduresand closed source software means it is becomingdifficult if not impossible for candidates andparty organizations to verify the vote. Lack ofopenness of software presents technological andsecurity concerns that should be addressed. Forall these reason, standards for acceptable levelsof auditability should be developed.

Pointing out these considerations is not intendedto favor one kind of technology over another.There may be pluses to a form of technology,either generally or with respect to voters forwhom that technology provides essential bene-fits that outweigh the shortcomings of that tech-nology. Overall, it is important to recognize thatthere are significant choices to be made andprogress to be achieved in the course of bothresponding to deficiencies in technologies and indeveloping their strengths. In order to providethe greatest voting benefits to a varied votingpopulation, it may be that several forms of tech-nology should be used. For example, it might beappropriate to provide, in each polling place, adirect electronic recording machine with audiocapacity for blind voters and multilingual capaci-ty for voters who require that (although, ofcourse, any voter may use those machines),while also using less expensive optical scan sys-tems that do not have those enhanced features.

12


C. After the Polls Close

The key task after the polls close – vote counting– directly relates, of course, to the specificissues that dominated last year’s election con-troversy: what should constitute a vote and howshould vote-counting disputes be handled? Inthe 2000 election, those issues arose in a presi-dential election, a setting that presents specialquestions about the available time for resolvingdisputes. Counting and contest issues can ariseless dramatically in elections for other offices.Responses to the 2000 experience should workfor all elections.

1 . T I M E L I N E S

Each state should review its election code toensure that its election calendar includes a real-istic timetable for procedures on counting andany recounting of ballots that assures promptresolution of the official outcome of the election.By virtue of two longstanding federal laws, oneestablishing the first Tuesday after the firstMonday in November as the national electionday (3 U.S.C. § 1), the other establishing the firstMonday after the second Wednesday inDecember as the date on which presidential elec-tors shall meet and vote (3 U.S.C. § 7), there areonly about 40 days between election day and thedate the electoral college meets. Thus, eachstate’s post-election timetable should aim forresolution of counting and contest procedureswithin that time in order to assure that they arecompleted in time for electors to cast the state’selectoral votes.

2 . W H AT C O N S T I T U T E S A V O T E

Each state should define what is a valid vote.Many states have election code provisions that(with variations in phrasing) require election

officials to determine the intent of voters indeciding how to count ballots. In the aftermathof the Supreme Court’s decision in Bush v.Gore3, each state should review its general leg-islative policy on the definition of a valid vote,and also the manner in which it should applythat policy in order to assure the equal treat-ment of voters.

Each state has a major choice to make. Onepossibility is to prescribe an exact form of cast-ing a vote and to preclude the counting of anyvote not cast in that precise manner. Castingvotes in accordance with election rules is thestandard to which voters and election authori-ties should aspire. Nevertheless, it is inevitable– when millions of people vote – that systemsthat require voters to physically mark ballots orto punch holes in them will give rise to varia-tions that make some difficult to count. As amatter of democratic principle, we believe elec-tion law should place a value on an effort to eval-uate whether a voter’s choice is clearly dis-cernible even if a voter did not follow instruc-tions to the letter. Certainly, in the case of write-in votes, exact spelling should not be required aslong as the voter’s choice is clear.

It is impossible to anticipate every anomaly.Still, as required by Bush v. Gore, rules shouldbe in place to assure equal protection of the lawin resolving recurring questions that arise undervarious voting technologies. An example of arecurring question for which there should be auniform rule throughout a state is what to dowhen a voter correctly marks a vote for a candi-date and also writes in the name of the samecandidate. In that circumstance, the vote for theoffice should be treated as a single valid vote, asthere can be no doubt about the voter’s choice.The second marking may be understood toemphasize the voter’s choice; it should not bethe occasion to negate it.

13


State law should establish the general principleand procedures for developing specific rules forresolving counting issues. Because counting rulesfor different voting systems will be detailed andsubject to change as technologies change, the taskof filling in the details should be carried out inadministrative rules, issued by the state’s chiefelection authority in advance of an election, ratherthan fixed in permanent law. States should useprocedures that give public notice of proposedrules and open them to public comment. Thepolitical parties whose candidates will be affectedby the rules, and local election officials who will inmany instances carry out the rules, should be animportant part of the rulemaking process.

3 . M A N UA L R E C O U N T S

Most states provide for an automatic retabula-tion of votes if the results are close. State lawmay also authorize candidates to petition for aretabulation. The latter may be conditioned onpayment by the requesting candidate for the costof the retabulation. In places with vote tabulat-ing machines, retabulations can be conducted inaccordance with established procedures, includ-ing procedures for auditing the machines foraccuracy. This subject is discussed in the nextsection.

When should a recount go beyond machineretabulation of all votes and provide for the indi-vidual examination of some ballots? In responseto its experience in the last election, Florida’sanswer is that manual recounts should be trig-gered in two circumstances. There will be anautomatic trigger in elections with razor-thinmargins of victory, where a candidate is defeatedby one-quarter of a percent or less of votes castfor the office. In elections in which the differ-ence is slightly larger but still small (betweenone-quarter and one-half percent) a manualrecount shall be done on request.

In a manual recount, Florida now provides thatapproved software will be used to separate twokinds of ballots from all others: ballots in whichno vote is tabulated for an office because none isidentified by the tabulating machine (undervotes),and ballots in which no vote is counted becausethe machine has identified two or more votes forthat office (overvotes). Only those ballots will besubject to individual review under Florida’s newsystem. If on applying specific counting rulesestablished for each certified voting system,counting teams cannot determine “a clear indica-tion on the ballot that the voter has made a defi-nite choice” (the state’s new standard), the unre-solved undervotes or overvotes will be submittedto a canvassing board for determination.

Because Florida’s manual recount provision istriggered not by events in a limited geographicarea (such as a machinery breakdown in onecounty), but by a margin of difference among allvotes cast for an office, it follows that any manu-al recount should be throughout the electoraljurisdiction of the office in question. In Florida,as in all states (other than for two electors eachin Maine and Nebraska who are separately elect-ed in each state’s two congressional districts),all presidential electors are elected statewide.This means that in Florida, any future manualrecount in a presidential election will be con-ducted statewide. This requirement addressesone of the equal protection issues raised byBush v. Gore.

Florida has fashioned a thoughtful approach tomanual recounts that merits consideration else-where. Other states may frame answers in dif-ferent ways. Before making a recommendationthat might apply to all states, there is more tolearn about the range of possible responses.Clearly, the questions answered by Florida –when should individual ballots be examined,which ballots should be selected for examina-

14


tion, what should be the standard and procedurefor examining them, and what should be the geo-graphical extent of the examination – are ques-tions that all states should answer.

4 . A U D I T P R O C E D U R E S

Provisions for manual recounts are importantsafety valves in close elections. But confidencein the election system is needed even when elec-toral outcomes are not close. The challenges inmaintaining that confidence have grown in lightof technological changes that make the innerprocesses of machine vote counting less visibleto parties, candidates, and the public. It istherefore critically important to establish andutilize regular systems to check the accuracyand integrity of vote tabulating machinery.

Audits of tabulating machinery should occurboth before and after election day. A useful tech-nique in auditing is to recount a small percent-age of the ballots or electronic ballot imagesmanually or on an independently programmedmachine. For example, in California, a one-per-cent manual recount is required automatically,regardless of the spread of votes between thecandidates. This process sometimes discoverserrors due to programming errors or the assign-ment of votes to the wrong candidate.

5 . A L L V O T E S S H O U L D B E C O U N T E D

Our individual right to vote includes our individualright to have that vote counted no matter which ofthe methods permitted by state law we use to castthat vote. This is true whether we cast a ballot atthe polls, submit a provisional ballot to be countedif a registration question is resolved favorably, orcast an absentee ballot as a military or overseasvoter. No qualified voter should be in any doubtabout the counting of his or her ballot.

D. Alternative Methods of Voting

No subject has generated as much debate in ourproceedings as the question of alternatives tovoting on election day at a polling place. It isclear that there is not consensus on this issue. Itis possible, however, to identify important areasof agreement while demarcating the principalsignificant area that presently eludes concur-rence.

Election day voting at polling places provides thebest opportunity to achieve every one of five fun-damental objectives:

1) assure the privacy of the secret ballotand protection against coerced voting;

2) verify that ballots are cast only by dulyregistered voters;

3) safeguard ballots against loss or alter-ation;

4) assure their prompt counting; and

5) foster the communal aspect of citizensvoting together on the same day afterhaving had the opportunity to hear thefull common pool of public informationthe campaigns can provide.

No form of alternative voting has been devisedthat can provide every one of these benefits.

There are four alternatives to election-day,polling-place voting methods:

• Internet voting;

• voting entirely by mail;

• absentee voting – which itself has twocomponents: absentee voting for voterswho are unable to come to polls, andunlimited absentee voting as a matter ofconvenience; and

15


• early voting at election offices.

This report discusses each in turn.

1 . I N T E R N E T V O T I N G

Three types of Internet voting are imagined.One is poll-site Internet voting in which votescast at regularly established polls are transmit-ted for counting. A second is voting at kiosk ter-minals placed in public places (other than regu-larly established and staffed polls). A third isremote Internet voting in which voters cast votesfrom any Internet accessible location.

Poll site and kiosk Internet voting present signif-icant unresolved issues, but systematic researchand evaluation may identify reasonable limitedexperiments to help advance the objectives pre-viously described, such as permitting voting atelection day locations close to places of work.As for the potentially vast category of remoteInternet voting, a recent study sponsored by theNational Science Foundation demonstrates thesignificant security risks it would pose to theintegrity of voting.4 The study urges that remoteInternet voting not be used in public electionsuntil substantial technical issues, among others,are addressed. If security problems are eversolved, remote Internet voting would still faceimportant additional issues concerning theimpact of alternative voting. There is no presenttimeline on which remote Internet voting shouldbecome part of the nation’s voting methods.

The National Science Foundation study suggest-ed that remote Internet voting may be appropri-ate for special populations, such as the militaryand their dependents based overseas. Webelieve that experimental military voting experi-ments should be continued.

2 . V O T I N G E N T I R E LY B Y M A I L

Only one state – Oregon – conducts its electionsentirely by mail. The unique decision that Oregonvoters and officials have made for themselvesmust, of course, be respected. However, whateverbenefits there may be of having an alternative topolling place voting, voting entirely by mail is notan alternative, it is a replacement. The completeloss of polling place may be why no other state isconsidering the Oregon system.

3 . A B S E N T E E V O T I N G

a. For Persons Unable to Vote at Polling Places

Important parts of our population are unable tocast votes at polling places. These include per-sons who by reason of age or disability areunable to vote at polling places, persons in mili-tary or civilian service overseas, and other vot-ers unavoidably away from home on election day.No matter how the broader debate about alter-native voting methods is resolved, for these vot-ers and others such as them, a hospitable andefficient system of absentee voting, with protec-tions against fraud or other abuse, is essential tofulfilling our commitment to universal suffrage.

b. Unlimited Absentee Voting

The area of significant disagreement within ourForum is whether absentee voting should also beavailable to voters who are able to vote at pollingplaces but as a matter of convenience prefer toabsent themselves and vote by mail, as a numberof states now permit. Absentee voting presentsa risk to every one of the fundamental benefits ofelection day, polling place voting described at theoutset of this section. For voters who are unableto come to a polling place on election day, theunavailability of absentee voting would cause the

16


loss of that person’s franchise. For all other vot-ers, the unavailability of absentee voting wouldpresent at most a possibility that some voterswill choose not to vote.

We know from the actual experience of recentelections that, in some states at least, the num-ber of voters who vote as absentees as a matterof convenience can grow vastly. Whatever anystate may determine is appropriate for itself instate and local elections, in federal elections theconsequence of unlimited absentee voting on theability of the state to produce a final count anddecide controversies about it in time for the res-olution of a national election is something thatthe nation has a stake in. In presidential elec-tions, both the sheer additional task of countingabsentee ballots, and the multiplication of issuesthat may arise about the validity of individualballots cast away from the protections andscrutiny of polling places, may overwhelm theability of any state to resolve an election contro-versy within the spare six weeks that are avail-able between election day and the meetings ofpresidential electors. Or the control, and hence,the organization of one or the other chamber ofCongress may be at stake.

States that now employ unlimited absentee vot-ing are not likely to roll back that voting methodunless convinced by evidence from their ownexperiences. In the best tradition of federalism,any jurisdiction that has moved in that directionmay serve as a laboratory for itself and others toexamine. For any state that has adopted unlim-ited absentee voting, this report encourages, inlight of the issues raised above, a regular reeval-uation of the costs and benefits of their proce-dures. For any state that has not yet adoptedunlimited absentee voting, the issues raisedabove warrant treating with caution and examin-ing carefully any new proposal to move towardgreatly expanded or unlimited absentee voting.

4 . E A R LY V O T I N G AT E L E C T I O NO F F I C E S

Early voting at election offices (or at otherplaces under the supervision of election officials)is the method of alternative voting most consis-tent with the fundamental objectives describedat the outset of this section. It preserves all thesecret ballot, fraud prevention, and ballotintegrity aspects of polling place voting. Itshares in a portion of the communal aspect ofvoting, because although the number of voterswill be smaller, voting does not take place in iso-lation. It falls short only in that early voters willnot receive all of the public information providedby the campaign and will not be able, as othervoters may, to take account of events just beforeelection day.

There are compensating advantages of early vot-ing. Some voters who may have had to vote byabsentee ballot because of travel on election daywill be able to come to a polling place and vote inthe several weeks before the election. Older ordisabled voters, who may prefer to vote at apolling place but who would be impeded by thecrowds or pace of election day voting (or, untilremedied, the type of equipment at regularpolling places) and therefore have to vote asabsentees, also will be able to vote at a poll. Ofcourse, other voters will vote early as a matterof convenience, but even for them there may be areciprocal public benefit. Jurisdictions are ableto use early voting at election offices or otherelection sites as an opportunity to test, on asmaller scale, voting machinery innovations thatofficials might prudently be reluctant to try firstat polling places on election day. To amelioratethe impact of votes without a common base ofinformation, the range of dates for early votingshould be kept relatively narrow and close to theelection.

17


One final concern is the impact of alternativeforms of voting on voter participation. Enhancedturnout by itself would not necessarily outweighthe loss of the other combined benefits of votingachieved only by voting at the polls on electionday. Nonetheless, in the course of evaluating thecosts and benefits of alternatives to voting at thepolls on election day, states should consider theimpact on both overall voter turnout and turnoutamong different populations, such as people withdisabilities, and people of different races, ethnic-ities, or income levels.5

E. Top-to-Bottom Review of State Election Codes

As elemental as it may appear, a pillar of eachstate’s election system is a sound state electioncode. Even if a major revision of a state electioncode is not required, many state legislatures willdiscover that, over time, inconsistencies havecrept in and that their codes are not easilyusable by participants in the electoral process,clear to the courts, or comprehensible to thepublic. Earlier in this report, we discussed theparticular need to address state code provisionson such matters as what constitutes a valid voteand procedures for resolving issues aboutrecounts or contests. More generally, in light ofthe Supreme Court’s decision in Bush v. Gore,states should consider whether uniformstatewide requirements should modify or replacevarious delegations to local election authoritiesthat might result in the unequal application oflaw to various parts of the state’s electorate.

As long as its state law is harmonious with therequirements of the U.S. Constitution and feder-al law, each state is free to fashion its own code.Nevertheless, there is a great deal that statescan learn from each other. To that end, there

are mechanisms through such bodies as theNational Conference of Commissioners onUniform State Laws or the American LawInstitute for preparation of model codes for con-sideration by individual states. One lesson fromthe last election is that an election dispute inone state may have enormous implications forthe country as a whole. In a sense, every stateowes it to every other state, as well as to its owncitizens, to have an election code that drawsupon both the best of its own experiences as wellas that of others.

Among additional areas that merit review by thestates, this report notes the following.

First, states should review their laws and proce-dures for the restoration of voting rights. Wheresuch a procedure is available, states shouldensure that individuals are notified about theopportunity to restore voting rights and thatdecisions are communicated to the individual ina timely fashion.

Second, we encourage states to take steps toincrease public confidence by reducing partisaninfluences, and the appearance of such influ-ences, as much as possible. We recognize thatmany election officials hold elected office, orreport to elected officials. The most importantcheck on partisanship will be sound laws andprocedures established in advance of an elec-tion. Membership in professional associationswith a code of ethics may also help to balanceany concerns about partisanship. It is importantthat a review of partisan influences include allaspects of the election process, including regis-tration, the design of ballots, absentee voting,oversight and observers at polling places, andthe location of polling places.

Similar recommendations have been endorsedby the National Commission on Election

18


Standards and Reform, a joint undertaking ofthe National Association of Counties (NACo) andNational Association of County Recorders,Election Officials and Clerks (NACRC).6

Third, states and localities should considerreducing the frequency of elections by consoli-dating them while being mindful of the desirabili-ty of ballots of moderate length. This is especial-ly important as part of an effort to increase par-ticipation in elections.

19


Congress has broad constitutionalauthority to regulate the times, places,and manner of conducting federal elec-tions. As a result of that authority and

other constitutional powers to ensure the votingrights of Americans, the federal government istoday an active participant in establishing rulesfor federal elections on matters ranging fromvoter registration to protections against discrim-ination on grounds of race, language, and dis-abilities. But as pervasive as the federal rolehas become, Congress has never provided fundsto state or local governments to assist them inadministering federal elections or in defrayingexpenses for federal requirements that alsoaffect state elections.

The proposals that follow do not come close totesting the limits of the constitutional authorityof Congress to regulate federal elections.Historically, local governments (particularlycounties) – and to a lesser extent states – havebeen primarily responsible for administering andfunding elections. We share with many othersthe view that primary responsibility for conduct-ing elections should remain at the state andlocal level.

We also believe the federal government should

assist states and local governments in moderniz-ing the nation’s election system. An importantbalance must be struck between two importantobjectives. One is providing state and local gov-ernments with substantial discretion to makeimprovements that they identify as important.The other is to identify improvements that, as amatter of broadly shared national values notunique to individual states or locales, should besubject of special incentives or requirements.The pivotal mechanism for assisting the states isthe power of Congress under the Constitution’sGeneral Welfare Clause, Art. I, § 8, cl. 1, whichembraces the power to spend “as a means toreform the electoral process.” 7

A. Federal Assistance for Research and Technology Standards

For 130 years, Congress has legislated on themanner in which votes may be cast in congres-sional elections. In 1871, as part of a civil rightsmeasure designed to protect against voter intim-idation, Congress provided that all votes inHouse elections shall be by “written or printedballot” and that “all votes received or recordedcontrary to the provisions of this section shall beto none effect.” In 1899, it amended this provi-

20


I I . P R O P O S A L S F O RC O N G R E S S I O N A LA C T I O N

PART

II

sion to permit voting for the House by “votingmachine the use of which has been duly author-ized by the State law.” These provisions are nowfound in 2 U.S.C. § 9.

Each state was on its own in determining whatmachines to authorize. In 1975, the NationalBureau of Standards awakened attention to thelack of technical skills at the state and local levelfor developing written standards to evaluate vot-ing system hardware and software.8 Its reportlaunched a process that resulted in the 1990approval by the Federal Election Commission ofvoluntary engineering and performance stan-dards for voting equipment. Thirty-seven stateshave adopted (or advised the FEC that they willsoon adopt) those standards for new purchases.

Congress should now take the next step andenact statutory authority and provide appropria-tions for support of federally-conducted orassisted activities to enable states and local gov-ernments to benefit from research on improvedvoting technologies, the development and regularupdating of standards for them, and a clearing-house of experiences with voting technologies.These functions should be funded on a long-termbasis in recognition that voting technologies willbe continually developing.

The FEC is updating the voluntary engineeringand performance standards that it issued in1990.9 As described earlier in this report in Part I(A)(2), an expanded standards program shouldalso include voluntary management or opera-tional standards and performance or designstandards to optimize ease-of-use and minimizevoter confusion. The federal agency thatCongress charges with the responsibility to con-duct the standards program should have discre-tion to select for particular purposes the form ofstandards – such as minimum criteria, specifica-tions, or best practices – that is most suitable for

making progress while leaving sufficient roomfor innovation.

This recommendation is grounded on severalconsiderations. Even if some states can affordthe costs of developing new voting technologystandards and keeping them current, many can-not. Overall, it makes economic sense for thefederal government to undertake this functionfor the benefit of all states. Also, some objec-tives of federally assisted research or the devel-opment of standards or performance goals willbe to serve democratic values that are not readi-ly supportable by the budgeting of individualstates. Neither should these objectives, such asdevelopment of technologies to enable disabledvoters to vote without assistance, depend solelyon the research budgets of equipment manufac-turers.

In sum, Congress, by law in effect for now morethan 100 years, has required states to authorizeany use of voting machines; otherwise votesrecorded on those machines may not be counted.For ten years, Congress has undertaken to pro-vide voluntary standards to assist states in theexercise of that responsibility. Those standardsare out-of-date and more limited than theyshould be. In short, the federal government hasimposed a responsibility on the states forauthorizing the use of voting machinery andtaken on a reciprocal one that is key to enablingthe states to perform their congressionally man-dated responsibility soundly. The modest invest-ment called for by this recommendation is amplyjustified.

Finally, improvements in election administrationrequire research, the identification of best prac-tices, and exchanges of information about issuesin addition to new technologies. To provide illus-trations from other portions of this report, thereis a need to gain systematic knowledge about

21


such matters as the impact of poll locations,polling hours, and alternative methods of votingon voter participation. Election administratorscould also benefit from research on systemsmanagement (e.g., the maintenance of accurateregistration systems), human resource manage-ment (e.g., the recruitment and utilization ofelection day workers), and issues relating tocompliance with federal law (e.g., the presenta-tion of voting information in multiple languages).

Currently, Auburn University, in conjunction withthe Election Center, provides one of the few pro-fessional training programs in election adminis-tration. Coordinated research on elections man-agement, including development of curricula,would allow the nation’s schools of public admin-istration, an untapped resource, to play a con-structive role in the continuing professionaliza-tion of election management.

B. Federal Grants for Capital Investmentin Voting Systems Technology and Use

Broadly speaking, two kinds of proposals for fed-eral election system grants have been presentedin congressional testimony and public reportsduring the last several months. One is for a cap-ital investment program for acquisition of newvoting system hardware and software by statesand local governments during a limited numberof years. The other is for a permanent federalformula for sharing with state and local govern-ments the costs of election administration thatmay be attributable to elections for federaloffice. In this section, we discuss the proposalfor a capital investment program. In discussinga grant program, our references to statesinclude the District of Columbia, which by virtueof the 23rd Amendment also appoints presiden-tial electors.

We believe it is an appropriate federal role toprovide grants for a multi-year capital invest-ment in voting technology, both hardware andsoftware, and that a strong case has been madefor a grant program. A number of importanttechnology needs emanate from requirements offederal law. Examples include the informationmanagement challenge of the National VoterRegistration Act and the translation require-ments of the Voting Rights Act. National aspira-tions that should not be dependent on localresources, such as assuring that disabled citi-zens have the same voting opportunity as othercitizens, present other technological challenges.The grant program should enhance the effortsstates are making to comply with existing feder-al requirements.

Finally, funding for election investment is nowdependent on local government resources, andtherefore constrained by disparate or limitedlocal tax bases. States should do more to mod-ernize and equalize voting opportunities amongtheir jurisdictions, but the task is large enoughthat there is ample need for both federal andstate assistance and cooperation.

The program should be for a range of purposesthat is broader than just voting machines,although voting machinery would surely be animportant part of it. While public focus has beenon vote-casting devices, state and local electionofficials have demonstrated that there is a widerscope to voting system needs, including forinvestment in registration systems. As it shoulddo for any appreciable federal expenditure,Congress should define clearly the program’sscope and articulate its goals. Within thatscope, states should be able to establish priori-ties that apply to their circumstances.

22


1 . S C O P E

We recommend that the following be includedwithin the scope and objectives of a federalgrant program:

1) For improved registrations systems:

Funding should be provided for (a)development and maintenance ofstatewide databases – some states willestablish a single database, other statesmay link county databases; (b) electron-ic integration of information from motorvehicle bureaus and other sources ofregistration information; and (c) elec-tronic communications between andamong polling places, county and otherregistrars, and central registration data-bases.

The purpose of this investment shouldbe to enable states to develop and main-tain accurate registration – databasesthat fully utilize key registration relatedinformation within each state, includingchange of address, death, and othersuch matters. Grant funds should beavailable for the development of links toother states to correct records as votersmove from one state to another.Registration databases should be usablein a timely way at polling places to cor-rectly and promptly resolve registrationquestions.

2) For precinct-level voting equipment,including counting equipment:

The purpose of this investment shouldbe to enable states to acquire new votingmachinery, including precinct countingmachinery, that will be easy for voters to

use and reduce voter mistakes or inad-vertent omissions, including by alertingvoters that they have cast more votesthan permitted for an office or have notcast a vote for an office or ballot ques-tion, and that provide voters with anopportunity to correct those mistakes oromissions. The purpose of this invest-ment should also be to enable voterswith disabilities to operate votingmachinery independently and therebyvote secretly, as other voters may.

3) For election personnel training andvoter education about use of votingtechnologies:

We discuss below whether the federalgovernment should provide general sup-port for election administration. Nomatter how that question is resolved, werecommend that part of a technologygrant program should be funds for train-ing and education in the operation andmaintenance of voting equipment, bothwith respect to new technologies thatare acquired with federal grants and forimproved use of existing technologiesthat continue to be utilized. Assistancein meeting these objectives should beintegral to the basic investment.

2 . D U R AT I O N

Congress should determine the duration of a fed-eral voting systems technology grant program.We have several recommendations about factorsthat should be considered in establishing thelength of a capital investment program.

Some states, through legislation already adopt-ed, are ready to purchase or lease new votingtechnology. They would be assisted by the 23


appropriation of funds for a federal grant pro-gram that goes into effect for the fiscal yearbeginning October 1, 2001. Other states haveestablished study or planning processes in orderto help make technology and other decisionslater this year or during next year. Their focusmay be on the fiscal year beginning October 1,2002. A state may wish to stage investments,such as by devoting initial efforts to registrationimprovements and then moving to votingmachinery, or vice versa. States that haveinvested recently in new voting technology maywish to wait several years for the next genera-tion of voting equipment. Consideration shouldalso be given to the capacity of the voting equip-ment industry to produce for a market mademore active by federal grants, so the pace ofgrants and acquisitions do not inflate the costsof new products.

In striking the right balance, there is also goodreason not to extend unduly the national time-line for discernible voting system improvement.For planning purposes, it makes sense to thinkof the desirability of cooperative federal-stateplanning to effect stages of changes in time forthe next three federal elections in 2002, 2004,and 2006. This period of time coincides withproposals in pending legislation to establish pro-grams that are five or so years in length. A ben-efit of implementing modernization efforts by the2006 election would be to ensure that there is anopportunity to work out any issues that mayinevitably arise in adjusting to significantchanges before the 2008 presidential election.

At the point selected by Congress, the programshould sunset. Any extension would be subjectto a fresh determination by Congress that, onevaluating experience under the program,renewal is warranted. Of course, in consideringwhether to extend the program, a major consid-eration should be whether a five-year program

life has proven to be too short for effectuatingsoundly in all states the major modernizationeffort contemplated by the program.

3 . A L L O T M E N T S

Congress has many options for structuring agrants program. A threshold decision is whetherto award grants to states on a competitive basis(with the possibility that some states may notreceive any funds), or whether to award grantsto all states on the basis of a formula. In light ofthe significant and urgent need for federal assis-tance throughout the country, we believe prefer-ence should be given to a grant program that isprincipally formula-based.

A likely formula is apportionment of fundsamong the states according to the share of eachin the nation’s voting age population, although aprescribed uniform minimum allotment for eachstate might sensibly reflect that certain costs(for example, software for statewide registrationsystems) may not depend on the voting popula-tion of each state. Of course, some states mayopt out of the program, either entirely or in part.To the extent that applications from a state (orfrom localities if a state does not apply) do notutilize the entire amount allotted to the state,the state’s unused share of the federal authori-zation should be reallocated among the remain-ing states. There may also be limited, definedcircumstances, such as when a state is knowing-ly in violation of federal requirements, thatrelease to it of the state’s share should bedeferred pending compliance with applicablerequirements.

Each state should be required to submit a planfor use of its allocation, as described in the nextsection. Accordingly, some funds should bereleased at the outset of the federal grant pro-gram to assist states in preparing plans. The24


formula for apportioning these funds should takeinto account that each state, no matter its popu-lation, will incur some similar minimum costs inestablishing and implementing a statewide plan-ning process.

There are several advantages to a principally for-mula-based approach. First, it will be useful toencourage participation by every state becauseall states can benefit from an investment in elec-tion administration. Second, the expectation oftimely and regular receipt of a predeterminedamount of funding will enable each state toengage in an orderly planning process. Third, aformula-based approach should reduce the costsof federal administration.

Finally, while the overall federal grant programshould be principally formula-based, there isgood reason to reserve a portion of it (for exam-ple, ten percent) for pilot state or local programsthat may provide a testing ground for technolo-gies or their applications. In the nature of pilotprograms, the grants for them should be awardedon a competitive basis to encourage innovation.

4 . A P P L I C AT I O N S

Congress should establish a process for applica-tion for federal grant funds and for the review ofactivities under those grants. As an ordinarymatter, we believe local governments should sub-mit requests to their states and that statesshould submit applications to the federal govern-ment. Each state will need to make a judgment,after collecting requests from local governments,about structuring its application so it fits withinthe amount that will be available to the state.There may be limited circumstances – for exam-ple when a state declines to participate in thegrant program – in which local governmentsshould be authorized to apply directly to the fed-eral government. In the main, states and local

governments should be encouraged to worktogether to formulate a statewide plan.Otherwise, the federal task of sorting throughindividual local government applications couldbe daunting.

A state application should include a publiclyavailable plan that describes the state’s electioninvestment needs, how the state will use federalfunds to address those needs – including howthe grant will help the state meet existing feder-al requirements – and how the state will assurethe equitable use of federal funds within thestate. It should describe the state’s compliancewith existing election administration require-ments under the Voting Rights Act, theUniformed and Overseas Citizens AbsenteeVoting Act, the Voting Assistance for the Elderlyand Handicapped Act, the Americans withDisabilities Act, and the National VoterRegistration Act. It should provide assurancethat the state plan does not conflict with thoserequirements, and describe how grant funds willbe used to meet them. These laws set out essen-tial standards against which the lawfulness ofevery state’s election system must be judged.

To illustrate, an important use of federal tech-nology funds will be for improvements in regis-tration systems. In the application process,states should describe how those improvementswill enable them to maintain complete and accu-rate lists on a regular basis. Additionally, inorder to prevent discriminatory or erroneouspurging, states should describe the safeguardsthey have established, including timely noticeand an opportunity for voters to rebut anygrounds for being stricken from the registrationlist. To ensure that systems acquired with feder-al funds are available to all voters, states shoulddescribe their measures to assure that all votinglocations are fully accessible.

25


The plan should provide assurance that federalfunds (and any matching funds) will be used tosupplement, rather than lower, current spendingon elections. To facilitate resolution within eachstate of any issues about its plan, the applicationshould be open to public comment during its for-mulation. The state’s plan should also be pub-licly available after adoption. The applicationshould be reviewed by the appropriate federalagency for compliance with existing federal law,such as existing law on minority languages, andwith the requirements of the law establishing thegrant program.

5 . C O N D I T I O N S R E L AT E D T OT E C H N O L O G Y

Congress should require new technology pur-chases to comply with the FEC’s existing votingsystems standards. These standards are, ofcourse, now voluntary and only become mandato-ry when they are adopted by a state. By condi-tioning federal grants on compliance with them,Congress would be making the standards manda-tory in the minority of states that have not yetadopted them, insofar as equipment purchasedwith federal grants is concerned. The justifica-tion for doing so is that the existing standardshave become de facto a national norm.

A different issue is presented by whether tech-nology purchased with federal grants shouldcomply not only with existing standards but alsowith any new ones that are in existence at thetime of a purchase. We believe there is goodreason to require that. New standards will rep-resent the best understanding of what technolo-gy should achieve. We recognize this wouldpartly alter the nature of the standards, whichhas depended on state decisions to adopt them.We are suggesting not that the new standardsnow be mandated for existing voting devices; butonly that it is appropriate to use federal grants

to give states an incentive to acquire new tech-nologies in which voters have the greatest confi-dence that their voting opportunities will besecured by the best standards then available.

One technology objective that should be enactedinto law is that federal technology funds be usedto enable voters with disabilities to vote inde-pendently and therefore privately. A state apply-ing for a technology grant should commit to pro-vide, during the life of the grant program, atleast one voting device at each polling stationthat allows sight-impaired voters to vote inde-pendently. Complete attainment of this objectivewith regard to some other disabilities maydepend on further research, but the overallobjective should be clear and reached as soon asfeasible. Of course, individual voters may preferas a matter of their own volition to have assis-tance in voting. But the right to cast a secretballot is so central to our democratic traditionthat Congress should accelerate this new oppor-tunity that technology provides.

6 . A D D I T I O N A L R E Q U I R E M E N T S

Apart from capabilities that may be required forvoting systems acquired with federal funds, thereare two important measures that Congress shouldprovide be universally adopted in federal elec-tions, at least as a condition for federal grants. Ifa change in state law is required, the time forcompliance should allow for a regular meeting ofthe state legislature to enact that change.

One is that if a voter’s name does not appear onthe list of registered voters, and election officialsare unable to resolve at the polls the question ofthe voter’s registration, the voter should be giventhe opportunity to submit a provisional ballot.The reasons for providing an opportunity to sub-mit a provisional ballot are described in Part I(B)(3)(b) of this report, together with more detail26


about that process. The essential point is that if aregistration issue cannot be resolved on electionday, a provisional ballot ensures that the voter’sopportunity to vote is not irretrievably lost.

Second, election officials should post at pollingplaces clear notices of the rights and responsi-bilities of voters under applicable federal andstate law. The posted information should bemade available in alternate formats. We are notproposing that Congress mandate the specificcontents of these statements. A number ofstates are in the process of fashioning them.Private groups are also recommending variousforms of them. The key is that there be a promi-nent and readily available frame of reference forelection personnel and voters to anticipate andthen resolve polling place issues within therequirements of federal and state law.

Neither a provisional ballot requirement, nor oneon posting bills of rights or statements ofresponsibility, would change the underlyingrequirements of federal or state law on who mayvote or how they should do so. But they will bothhelp promote an atmosphere and process thatassures all participants in the voting processthat decisions will be made in accordance withapplicable law. Of course, states on their owncan adopt these measures. Some have done so;others will follow. There is considerable benefitin providing for the early universality of them infederal elections as a national down payment onother improvements and reforms that will follow.

These additional requirements supplement theplan submission requirements detailed inSection 4.

7 . R E P O RT I N G

To assist in making judgments about whetherfederal grants are helping to improve the admin-

istration of elections, states should regularlyprovide statistical information on the perform-ance of new and existing voting technologies, atleast in relation to elections for federal offices,although reporting on experience in other elec-tions may be informative to the Congress andother states. The reports should include docu-mentation about numbers of undervotes andovervotes with respect to various voting tech-nologies. The information should be transmittedto the national clearinghouse so that it can bewidely distributed in order to inform technologydecisions at the local, state, and national levels.At the end of a funding period, each state shouldpublicly report on what it has done with grants ithas received. These reports should be evaluatedby the federal grant-making agency.

8 . F E D E R A L A G E N C Y

Bills before the Congress place responsibility forelection administration in different agencies: theJustice Department, the Commerce Department,and the FEC. At least one bill would establish anew federal agency. Some bills send discretefunctions to additional agencies. Without com-menting on the merits of any of these proposals,the federal agency chosen (or established) tocarry out the responsibilities we have describedshould have several attributes.

First, it is desirable for a single agency or office tohave final responsibility for all of the functionsidentified in this report: research and development,standards setting, and grant-making and oversight.There may, of course, be circumstances in whichCongress determines that the expertise of anotheragency should be made available to the one thathas final responsibility for these functions.

Second, Congress should include an independentline-item in the budget to cover these functions.This would occur if the Congress establishes a 27


new agency but should be the case even ifCongress vests these responsibilities in an exist-ing agency. Appropriations for election purposesshould be protected from competing demands ofa parent agency. The FEC’s Office of ElectionAdministration already has responsibility forsome of these areas, and it is sensible to build onthat agency’s existing expertise, either by greatlyexpanding its mission and resources, or by relo-cating it elsewhere.

Third, it is important that the agency be inde-pendent of partisan influences to eliminate polit-ical considerations and thereby heighten publicconfidence in the agency’s work.

Fourth, the agency should be guided by an advi-sory board that reflects viewpoints of key partic-ipants in the election process, including electionadministrators and representatives of voters.The advisory board should also include memberswith legal and technological expertise.

Finally, the agency should be organized to makedecisions, particularly those on grants and stan-dards, in a timely manner. If the grants programis given to the FEC, it will be important to estab-lish an independent decision-making process toavoid potential deadlock on the even-numberedCommission. One technique for doing that wouldbe to provide that the head of an election officewithin the FEC be appointed by the President(perhaps, as in the case of the ComptrollerGeneral, upon receipt of recommendations fromthe bipartisan leadership of Congress). Therecould be the added protection and status affordedby the advice and consent of the Senate, and arequirement that the head of the office have a voteon the Commission on all matters affecting thegrant program.

We did not reach agreement on whether anexisting agency or a new agency would best

reflect these attributes. There is agreement,however, that time is of the essence. If theCongress decides to establish a new federalagency, it should provide for an interim arrange-ment so that the grant program can go forwardwhile the agency is being established.

We have two comments with regard to existingresponsibilities of the federal government. TheDepartment of Justice should retain all of itscurrent responsibilities for enforcement of votingrights laws. We also recommend that theFederal Voting Assistance Program (FVAP) beretained in the Department of Defense, althoughthe agency given responsibility to carry out thegrant program should be authorized to workcooperatively with that program to facilitateimprovements. Military personnel and citizensliving overseas face unique challenges in regis-tering and voting. Timely delivery of ballots andother information is critical to enable these citi-zens to vote. We are concerned that if thesefunctions as they relate to military voters aretransferred to a civilian agency, they might notbe accorded the same level of priority amongmilitary commanders as are communicationsfrom the Office of the Secretary of Defense.

Congress should provide the resources needed toensure vigorous implementation of the responsi-bilities that remain in the Departments ofJustice and Defense.

9 . A P P R O P R I AT I O N S F O R T H E G R A N TP R O G R A M

Congress should authorize and appropriate suf-ficient funds to provide a significant incentive tothe states to participate in the grant program,and to enable them to make necessary improve-ments. Several pending bills would authorizeamounts such as $500 million for five years. Webelieve this is a modest amount for the purpose28


of an initial authorization, given the cost ofadministering elections and the improvementsthat are warranted. Other bills provide moreflexibility by appropriating such sums as may benecessary to carry out the purposes of the grantprogram.

Two recent studies offer cost estimates forupdating voting equipment. The Caltech/MITVoting Technology Project estimates that it willcost about $2 per voter per year to achieveupgrades in voting equipment (assuming a 15-year life for voting machines). It will cost anadditional $2 per voter per year to lease laptopsfor polls on election day with voter registrationlists. The total cost would be $400 million peryear for these two kinds of expenditures.10

A George Washington University report esti-mates that it will cost $1.2 billion to replacepunch-cards with optical scan votingmachines.11

Costs, of course, will vary based on the kind ofequipment purchased, whether states lease orbuy, the needs of the state, and the matchingrequirement. A number of recent decisions bystates and localities to replace voting equipmentalso provide a preliminary indication of howmuch money is needed. Florida,12 Maryland,13

and Michigan14 estimate it will cost between $24and $39 million to replace voting equipment intheir state. Some large counties in the UnitedStates will alone require comparable levels offunding. Harris County, Texas, the third largestcounty in the United States, is planning to invest$25 million to purchase direct recording elec-tronic voting machines.15

The development of statewide registration sys-tems also varies in cost. Several years agoMichigan invested $7.6 million to establish itsstatewide voter registration file, including pay-

ment to localities for installation. The state nowspends $1.4 million annually to maintain thefile.16 Oregon estimates that establishing a cen-tralized voter registration system will cost $6million. These figures do not include the impor-tant element of electronic communications withpolls on election day.17 And there is the needfor voter education and training programs in theuse of voting technologies.

More thorough estimates of the needs of the fed-eral grant program should be made possible byanalysis and compilation of the plans submittedby each state during the first year of the grantprogram. An important function of the federalagency responsible for the program, aided by anadvisory board, should be to submit to Congressduring the first year, in time for the secondannual appropriation for the program, a well-substantiated projection for the fiscal require-ments of the program during the full term of thegrant program.

1 0 . O V E R S I G H T

The federal agency responsible for awarding thegrants should conduct periodic audits to ensurethat funds made available under this section areexpended for the stated purposes, and to reviewa State’s activities in the areas required in sec-tions (5) and (6) above.

C. A Permanent Program to DefrayExpenses of Federal Elections

There is broad recognition that, in addition to aneed for investment in voting systems technolo-gy, other areas of election administration – suchas general voter education, training of electionadministrators, and staffing of polling places –would benefit from increased funding.

29


There is one form of permanent support for elec-tion administration for which there is broad con-currence. A significant factor in the cost of elec-tion administration is mailing. In the NationalVoter Registration Act, Congress directed thePostal Service to make available to state andlocal registration officials the rate that is avail-able to qualified nonprofit organizations. Forvarious reasons, including technical ones underPostal Service regulations, the promise of finan-cial assistance in the mailing of mandated regis-tration materials has not been fulfilled.

State and local officials should, in fact, be under-taking more mailing to voters, including sampleballots and information about procedures andvoting rights. To these ends, state and localelection officials have been urging Congress toestablish a new elections class of postage thatwould provide first class service at half the firstclass rate. This arrangement would providespeed of delivery and necessary services thatare important in election administration (suchas the return of mail if addressees have moved)at a rate befitting the high volume of that mail-ing. The important public ends that are servedby official election mail amply support that rec-ommendation. We believe the proposal merits afavorable response by Congress.

Beyond provision for a new postal rate, a num-ber of members of our Forum favor the idea thatCongress make a long-term commitment toexpend federal funds to defray the costsincurred by state and local governments in con-ducting elections for federal office.18 In theirview, there is merit in the suggestion that thefederal government reimburse states and locali-ties for their expenses in conducting electionsfor federal office, or at least for the part of themattributable to federal requirements, such as thenotification requirements included in theNational Voter Registration Act. Other membersof the Forum oppose, or at least oppose at thistime, a permanent federal role in funding stateand local government election administration.

The authorization of a permanent contributionto general election expenses, in contrast to alimited-term investment in the technology ofelections and a new postal rate, raises questionsabout the relationship of federal and state gov-ernments in this area that are larger than can bereadily resolved at this time. If the uncertaintywithin our Forum reflects uncertainty elsewhere,it may be that Congress and the states will wishto evaluate their experiences in carrying out atechnology grant program for several yearsbefore considering whether to embark on a permanent cost-sharing one.

30


As this report shows, there is a range ofpractical steps that can be taken bystate and local officials to improve ournation’s election system. There are

also well-defined measures that Congress shouldtake to support state and local government andadvance broadly shared national values, while

preserving historical balances. Above all, theparticipants in this Forum have laid out a pathtoward consensus, which of course remains awork in progress. Their efforts prove that thereis an opportunity for reform. American politicalleadership has the responsibility to seize it.

31


C O N C LU S I O N

32


1 Federal Elections Commission, The Impact of

The National Voter Registration Act of 1993 onthe Administration of Elections for FederalOffice, 1999-2000, A Report to the 107th

Congress (June 2001).

2 Caltech/MIT Voting Technology Project, Voting– What Is, What Could Be 23-24, 80 (July 2001).

3Bush v. Gore, 531 U.S. 98 (2000)

4Internet Policy Institute, Report of the

National Workshop on Internet Voting: Issuesand Research Agenda 34 (March 2001)(http://www.internetpolicy.org/research/results.html).

5We do not yet have exhaustive research on the

impact of alternative voting procedures on turnout.A study of Oregon’s vote-by-mail elections foundsome improvement in turnout among regular vot-ers, but not among intermittent ones. A. Berinsky,N. Burns, M. Traugott, Who Votes by Mail? ADynamic Model of the Individual-levelConsequences of Vote-by-Mail Systems (October2000) (copy of unpublished manuscript on file atthe Constitution Project). Aggregate statisticalanalyses by Curtis Gans, Director, Committee forthe Study of the American Electorate, found – forstatewide elections – that all-mail voting, unlimit-ed absentee voting, and early voting could lead tomarginal declines in turnout. His full analysis isposted on our website (http://www.constitutionpro-ject.org/eri/voting_procedures_comments.html).

7Buckley v. Valeo, 424 U.S. 1, 90 (1976).

6National Commission on Election Standards and

Reform, Report and Recommendations toImprove America’s Election System (May 2001).

8Roy G. Saltman, Effective Use of Computing

Technology in Vote-Tallying, National Bureauof Standards Report NBSIR 75-687 (March 1975)(reprinted as SP500-30, April 1978)www.vote.caltech.edu/Links/#articles.

9Federal Election Commission, Updating the

Voting Systems Performance and TestStandards: An Overview (June 2001)(http://www.fec.gov/pages/standardsoverview.htm).

10Caltech/MIT Voting Technology Project, Voting

– What Is, What Could Be 23-24, 80 (July 2001).

11Norman Garfinkle and Patrick Glynn, Report

on Election Systems Reform, Institute forCommunitarian Policy Studies, GeorgeWashington University (July 2001).http://www.gwu.edu/~icps/

12Florida appropriated $24 million to replace

voting equipment (small counties will receive$7,500, others will receive $3,750 per precinct).The state also appropriated an additional $6 mil-lion for voter education programs. (2001 Fla.Laws ch. 40 (SB 1118), §§ 74 and 76.)

13Maryland estimates that it will cost $36.8 mil-

lion to replace all voting machines with DRE’s.Maintenance expenses are estimated to bealmost an additional $1 million. The sameequipment could also be leased for $5.8 million ayear.

14Michigan estimates that a statewide uniform

optical scan system with precinct-based tabula-tion would cost between $26.1 million to $38.7million. Secretary of State Candice Miller,Uniform Voting in Michigan: A Report to theLegislature, State of Michigan Secretary ofState’s Office (May 2001). The report is avail-able at http://www.sos.state.mi.us/election/elecadmin/2001web/uniformvotinginmichigan.pdf.

15Harris county will buy 8,170 voting machines,

with 1,150 of them equipped for people with dis-abilities. S. Brewer, “Harris County OKs NewVoting Machine,” Houston Chronicle, June 6, 2001.

16Candice Miller, Secretary of State, Michigan’s

Qualified Voter File System: Bringing NewEfficiency to the Management of VoterRegistration and Driver License Records,Bureau of Elections (February 1, 2001). Formore information about Michigan’s QualifiedVoter File, go to: http://www.sos.state.mi.us/elec-tion/qvf/index.html.

17Bill Bradbury, Oregon Secretary of State,

Testimony before U.S. Senate Committee onCommerce, Science and Transportation (March7, 2001).

18See, for example, National Commission on

Election Standards and Reform, Report andRecommendations to Improve America’sElection System (May 2001).

ENDNOTES

33


Chair, Forum on Election Reform

Morton H. Halperin, Board Member, Constitution Project

Working Group Chairs

Steve Ansolabehere, Professor of PoliticalScience, Massachusetts Institute of Technology*

Marlene Cohn, Consultant, League of WomenVoters Education Fund

Norman Ornstein, Resident Scholar, AmericanEnterprise Institute*

Trevor Potter, Director and General Counsel,The Reform Institute

Richard Soudriette, President, InternationalFoundation for Election Systems

Participants

Angela Arboleda, Civil Rights Policy Analyst,National Council of La Raza

Kush Bambrah, Staff Attorney, National AsianPacific American Legal Consortium

Melanie Campbell, Executive Director, NationalCoalition on Black Civic Participation

Charlotte Cleary, General Registrar, ArlingtonCounty, Virginia

Carol Ann Coryell, Electoral Board Secretary, Officeof the Electoral Board, Fairfax County, Virginia

Todd Cox, Assistant Counsel, NAACP LegalDefense and Educational Fund

Charles Crawford, Executive Director, AmericanCouncil of the Blind

Ed Davis, National Director of State Organizationsand Field Operations, Common Cause

Jim Dickson, Vice President, Government Affairs,American Association of People with Disabilities

Mickey Edwards, John Quincy Adams Lecturerin Legislative Politics, Kennedy School ofGovernment, Harvard University*

David Elliott, Assistant Director of Elections, Officeof the Secretary of State of Washington State

John Fortier, Research Associate, AmericanEnterprise Institute*

Susan Parnas Frederick, Director, Law & JusticeCommittee, National Conference of StateLegislatures

Curtis Gans, Director, Committee for the Studyof the American Electorate

Nancy George, National Coordinator of VoterEducation, AARP

Heather Gerken, Assistant Professor of Law,Harvard University*

Thad Hall, Program Officer, The CenturyFoundation

Ernest Hawkins, Sacramento County Registrarof Voters, and Past President NationalAssociation of County Recorders, ElectionOfficials, and Clerks

Ron Hayduk, Professor of Political Science, CityUniversity of New York*

Wade Henderson, Executive Director, LeadershipConference for Civil Rights*

A P P E N D I X A :PA RT I C I PA N T S

Loretta Herrington, Director of theTelecommunications & Information Technologyand Trade & Transportation Task Forces,American Legislative Exchange Council

Thomas Hicks, Lobbyist/Policy Analyst, CommonCause

Lance Hoffman, Professor, George WashingtonUniversity*

Gwenn Hofmann, Senior Elections Advisor,American Elections, International Foundation forElection Systems

Tim Hovis, Senior Legislative Counsel, NationalGovernors Association

Connie Kaplan, Chicago Board of ElectionCommissioners and International Association ofClerks, Recorders, Election Officials, andTreasurers (IACREOT)

Pamela Karlan, Kenneth and Harle MontgomeryProfessor of Public Interest Law, Stanford University*

Steve Katsurinis, Vice Chair, AlexandriaElectoral Board, Alexandria, Virginia

J. Patrick Kelly JD, CERA Clerk/Recorder, ElPaso County, Colorado

Linda Lamone, Maryland Administrator of Elections

Lloyd Leonard, Legislative Director, League ofWomen Voters

Doug Lewis, Executive Director, Election Center

Thomas Mann, W. Averell Harriman SeniorFellow, Brookings Institution*

The Honorable Deborah Markowitz, Secretary ofState, Vermont

Robert McGarrah, Senior Policy Analyst, AFL-CIO

Ann McGeehan, Texas Director of Elections andPresident, National Association of State ElectionDirectors

Alysoun McLaughlin, Policy Associate, NationalConference of State Legislatures

John Mott-Smith, Chief of the EducationDivision, California

David D. Orr, Clerk, Cook County, Illinois

Cathy Pearsall-Stipek, President, NationalAssociation of County Recorders, ElectionOfficials and Clerks

Cameron Quinn, Director, Virginia State Board ofElections*

Miles Rapoport, President, Demos

Leslie Reynolds, Executive Director, NationalAssociation of Secretaries of State*

Mark Richert, Governmental RelationsRepresentative, American Foundation for theBlind*

Robert Richie, Executive Director, Center forVoting and Democracy

Roy G. Saltman, Consultant on Election Policyand Technology

Ari Schwartz, Policy Analyst, Center forDemocracy and Technology

Hilary Shelton, Director, Washington Bureau,National Association for the Advancement ofColored People*

Dave Scott, Director of Affiliates, The Council ofState Governments

Leonard Shambon, Counsel, Wilmer, Cutler &Pickering*

Tony Sirvello, Harris County Administrator ofElections, and International Association ofClerks, Recorders, Election Officials andTreasurers

Robert M. Stern, President, Center forGovernmental Studies*

Ralph Tabor, Associate Legislative Director,National Association of Counties

Jeff Thom, Member, American Council of the Blind

Christopher Thomas, Director, Michigan Bureauof Elections

Matthew Tripolitsiotis, Fels Center ofGovernment, University of Pennsylvania*

LaShawn Warren, Legislative Counsel for CivilRights, ACLU

Tom Wilkey, Director, New York State Board ofElections*

Joanne B. Wright, Deputy Political Director, ACORN

Lisa-Joy Zgorski, Communications Manager, TheCentury Foundation and National Commissionon Federal Elections Reform

Please note: Participation in the Forum does not indicatethat each person or organization agrees with every par-ticular in the report or recommendations to Congress.Participants should be consulted in determining theirposition on any issue covered in this report.

* Affiliation is for the purpose of identification only.

34


35


A P P E N D I X B :S U P P L E M E N TA LV I E W S

American Council of the Blind

The American Council of the Blind congratulatesall the participants in the Constitution Projectfor their many contributions of wisdom and goodfaith as we worked together in our common goalof perfecting our democracy. ACB only adds thissupplement to amplify and clarify points alreadypresent in the report.

If voting is to be truly accessible to persons whoare either blind or have low vision, then theequipment necessary to accomplish this goal of atrue ability to cast a secret ballot must be put inplace at polling sites. ACB strongly supports theuse of direct recording equipment (DRE) toinsure accessible, independent, secret and verifi-able voting rights for the millions of blind or visu-ally impaired citizens who only seek to exe r c i s eour franchise along with all other Americans.

Direct recording voting equipment offers theadvantage of maximum flexibility for futureopportunities while costing less in the long runthan do optical scanning devices.

ACB therefore strongly recommends that anyfederal legislation and financial support for theacquisition and deployment of voting equipment

be done in such a way as to afford visuallyimpaired voters with the ability to cast a secretballot through the use of a direct recordingdevice. ACB further recommends that at leastone device be available at all polling places.

Center for Voting and Democracy

Federal standards currently require that all elec-tronic Direct Recording Equipment (DRE) recordand store an electronic image of each ballot.The Center for Voting and Democracy believesthat the federal standards should include thesame requirement for all new voting equipmentused to count paper-based ballots. This belief isdriven by two concerns. First, electronic ballotimages increase the security of the electoralprocess by enabling the rapid detection of anyalterations to paper ballots that occur betweenthe casting of ballots and the final certificationof results. Second, the storage of electronic bal-lot images ensures compatibility of the votingequipment with all four ballot types currentlyused in U.S. elections. We believe that new vot-ing equipment should not create a barrier forjurisdictions wishing to adopt new ballot types.

Common Cause

Common Cause supports the “Recommendationsfor Congressional Action” of the ConstitutionProject Election Reform Initiative. If followed,these comprehensive recommendations wouldresult in a substantial federal investment in elec-tion system research and standard development,and in much needed improvements in votingequipment, registration systems, and voter edu-cation programs at the state and local level.

Because the serious flaws in our nation’s elec-tion systems have denied citizens basic votingrights, Common Cause believes that Congressshould play a more direct and proactive role inelection reform than these recommendationsenvision. With constitutionally guaranteed pro-tections at stake, Congress has a responsibilityto act as necessary to ensure that citizens aretreated fairly and equally in all stages of the vot-ing process. Directly mandating basic changesfor federal elections should not be ruled-out as ameans to that end, and Congress should setother fundamental reforms as conditions forstates seeking federal election grants.

Demos

Demos applauds the Constitution Project for rec-ommending to Congress and the states severalvery important election reform measures. Ifadopted, these suggestions will measurablyimprove the conduct of elections in America andadvance the cause of full enfranchisement andfair representation of our nation’s diverse voicesand communities. Demos also takes this oppor-tunity to comment on specific elements of theReport of the Forum on Election Reform(“Report”) and recommend additional reformmeasures for accomplishing a comprehensiverejuvenation of our electoral system.

Demos strongly supports the ConstitutionProject’s call for state adoption of provisional

ballots, extended polling hours, and posted voterrights and responsibilities. These recommenda-tions will substantially improve an individual’sopportunity to have her vote properly tallied andcounted. Administrative error in the formationof voter registration rolls are inevitable. Theopportunity to cast a provisional ballot, at thepolling place, offers a practical and workableremedy and safeguards an eligible voter’s exer-cise of her constitutional right. Prominent post-ing at the polling place of a voter’s rights and theprocedures that have been adopted to effectuateit are an important additional safeguard andeducational tool for voters and poll workersalike. Extended polling hours, like those adopt-ed in New York, are an appropriate accommoda-tion of the lengthening workdays, non-traditionalwork schedules, and increasingly competingdemands of work and family that now character-ize life in America.

The application process for state grants suggest-ed in the Report is also noteworthy. As a condi-tion for federal funds, state applications shouldindicate prospective plans for compliance withfederal registration and voting requirements,allow for public input into the formulation ofthose applications, and anticipate federal agencyreview of the proposed compliance measures.These are welcome responses to the manyinstances of voting irregularities or abuses docu-mented in last November’s elections, especiallyin communities of color, low-income areas, andamong naturalized citizens.

Several other measures to expand opportunitiesfor voters to cast their votes were raised in theReport but not offered as thoroughgoing recom-mendations. Demos presents them here asimportant innovations that should be seriouslyconsidered for adoption. Like many other com-mentators, we strongly support an election dayholiday and weekend or multi-day voting. Demosalso recommends that Oregon’s system of votingby mail and other alternative voting schemes be

36


fully reviewed and considered by the states. Inall instances, we value as paramount expandedopportunities for voter participation and aresultant vote that more fully expresses the willof the people.

Two other innovations must be considered inany discussion of election reform: election dayregistration and the restoration of ex-offenders’voting rights. Six states now allow eligible citi-zens to both register and vote on election day.These states also enjoy an average turnout ratethat exceeds the national average by ten per-centage points or more. Given the fact that voterinterest typically peaks in the closing weeks ofan election campaign, when media attentionsoars and the choice between candidates crystal-lizes, it make little sense to shut down the regis-tration process–and the opportunity to subse-quently cast a vote–30 days or so before electionday. Six states have shown that election dayregistration is workable. Other states should fol-low their lead.

The restoration of voting rights for ex-offendersis raised but not altogether embraced in theRecommendations to Congress. We do so here.4.2 million American citizens are now disenfran-chised by state laws that deny them the voteupon conviction. Many of them have served theirsentences and been reintegrated into the com-munity. The burden falls most severely on peo-ple of color, who are disproportionately affected.Demos calls for an end to this blanket disenfran-chisement and a return of the vote to ex-offend-ers upon their release from incarceration. Fourstates have adopted voter restoration laws overthe past year. More should consider the same.

The Recommendations to Congress detailedherein, if embraced, would constitute a veryimportant contribution to the ongoing realizationof the ideals that underlie our democracy. Morecan and should also be done to break down thebarriers to the vote. Additional reforms andmore vigilantly enforced or appropriately imple-

mented pre-existing measures or safeguards willalso advance the cause of democracy. Demoslooks forward in the years ahead to exploringevery opportunity to expand the franchise withthe many organizations that have contributed tothe report, the many others who are not repre-sented here, and to policymakers in the statesand at the federal level.

Pamela Karlan, Kenneth and Harle MontgomeryP rofessor of Public Interest Law, Stanford University

Many of the problems described in this Reportare the product of a failure to comply with exist-ing federal law, including the Voting Rights Actof 1965, the National Voter Registration Act, theAmericans with Disabilities Act, and the VotingAssistance to the Elderly and Handicapped Act.While I endorse the recommendations in SectionII of the Report, and think they will improvecompliance levels, I am skeptical that, standingalone, they will produce full compliance. I there-fore would support additional measures, such asstrengthening attorney’s fees provisions andproviding additional resources to enforcementagencies, to ensure that every eligible citizenhas full access to the electoral process.

The League of Women Voters

The League of Women Voters appreciates theopportunity for dialogue and discussion on elec-tion reform issues that the Constitution Projecthas provided. A wide variety of important topicshave been debated, and substantial agreementon best practices – the types of election adminis-tration that every state and locality should fol-low – has been achieved in Part I of the report.

However, the League of Women Voters does notendorse Part II of the report, the proposals forcongressional action. We find that they fall sig-nificantly short of what will be needed to protectthe fundamental rights of American voters. We wish to briefly highlight several points. First, 37


38


many of the problems voters faced in the 2000election, from not having their names on thevoter registration lists to the lack of bilingualballots where they were needed, were failures toimplement existing federal laws. Yet the pro-posals contained in Part II do not ensure fullimplementation of these laws, such as the VotingRights Act, the National Voter Registration Act,and the Voting Accessibility for the Elderly andHandicapped Act. Our nation must not risk,after a substantial commitment of federal fundssuch as is recommended here, continued non-compliance with basic voter protections.

Second, citizens in many states were purgedfrom the voter lists without basic due processprotections that would have prevented theirerroneous removal from the lists. We believeprotections against erroneous purges are a pre-requisite in any federal program. We were dis-appointed that such a provision was not includedin the requirements section of Part II.

Finally, the grants program proposed in Part II issimply too modest. It does not fully reflect themany recommendations in Part I. It is a short-term program from which states can opt out.Those states that participate will receive anallotment, similar to a revenue sharing program,with minimal requirements. The League ofWomen Voters believes that a more substantialresponse is needed to ensure that the problemsof Election 2000 are not repeated and to protectthe voting rights of all American citizens.

The National Asian Pacific American LegalConsortium and National Council of La Raza

The National Asian Pacific American LegalConsortium (NAPALC), dedicated to advancingthe rights of the Asian Pacific community, andthe National Council of La Raza (NCLR), thenation’s largest Latino civil rights organization,

applauds the Constitution Project for its worktoward making the electoral process fair andaccessible to all Americans. However, westrongly believe that unless states are mandatedto comply with the language assistance protec-tions enumerated in the Voting Rights Act of1965, limited-English-proficient voters will nothave equal access to the voting process.Therefore, NAPALC and NCLR firmly urge thatfederal funding be determinate upon mandatorydemonstration of state and local compliance withcurrent federal voting laws.

National Conference of State Legislatures

The closeness of the most recent presidentialelection and the subsequent spotlight on flaws inthe nation’s complex system of election adminis-tration served as a clarion call that was heardby state lawmakers across the nation.Subsequently, more than 1,700 bills have beenintroduced in state legislative chambers andapproximately 250 have been enacted into law.These include funding and establishing stan-dards and procedures for updating voting tech-nology, counting and recounting votes, trainingelection-day workers, educating voters, reform-ing absentee voting procedures, designing bal-lots, registering voters and purging voter lists.To ensure that states have the resources andinformation to undertake this critical effort,NCSL established a bipartisan task force toassist states in ensuring the integrity of the bal-lot; identify and recommend best practices onelection laws; study the effect of recent changesin the voting, such as early voting and mail-inballots; and provide technical assistance tostates on implementing state election reforms.Since early March, task force members haveheard from experts on election reform and dis-cussed the various kinds of legislation being con-sidered in state legislatures and Congress onelection reform. NCSL’s task force plans to com-

plete its work by mid-August with a report out-lining recommendations and identifying modelelection laws and systems.

In addition, the task force has supported federallegislation such as H.R. 2398, a block grant for-mula which awards money to states for broad-based initiatives related to election reform, whileopposing legislation which seeks to mandatespecific requirements on the states. NCSL’s taskforce also supports amendment of the NationalVoter Registration Act (NVRA) to grant statesgreater latitude to remove ineligible voters fromregistration lists and increased funding for theFEC Office of Election Administration for thedevelopment of voluntary equipment standardsand the dissemination of election-related statis-tics and information. NCSL’s recommendationsfor federal legislation, a database of the morethan 1,700 bills that have been introduced instate legislatures, and further information onNCSL’s task force on election reform can befound at NCSL’s Web site at www.ncsl.org.

John Pearson and David Elliott, Director andAssistant Director of Elections for Washington State

Thank you for the opportunity to respond to theConstitution Project report. We have seriousconcerns about one aspect of the report and weare pleased to share those concerns in thisforum. Our concerns are centered on the sectionof the report dealing with alternative votingmethods – specifically absentee voting and vot-ing by mail generally.

Here in Washington State we have had vote bymail elections, ongoing absentee ballots, andabsentee ballot on demand for nearly twodecades. From our point of view, as leaders inthis field, frankly, we found the section on absen-

tee voting and voting by mail to be so out oftouch with the realities of voting in the West asto be useless. Unfortunately, that particular sec-tion taints our opinion of the entire report -which does contain some good analysis andsome useful recommendations. The conclusionsin that section are based on premises which webelieve have no basis in fact - at least in ourstate. No consideration is given to the point ofview that one reason western states haveenhanced voting by mail is because THE VOT-ERS LIKE IT! It is convenient for them, andthey appreciate the extra time it gives them toconsider their various choices and makeinformed, intelligent decisions.

We believe that we are in the business of facili-tating the voting process for voters - and there isno doubt in our minds that safe, secure absenteevoting does just that. There is no proof offeredfor any of the alleged evils of mail voting - justthe very conservative opinion of the authorsoffered up as fact. The concern raised abouttiming of election results and the time needed toorganize the houses of Congress are specious atbest. The 2000 election in Washington State fea-tured a very close race for the U.S. Senate.Partisan control of that body hung in the bal-ance. The election was conducted with abouthalf of the ballots transmitted through the mail,and the race was close enough to require arecount of every ballot. We were able to com-plete our work in a timely fashion and theresults remain accepted and unchallenged.

As to the concerns raised by the author, askedand answered. The proper use of resources andtraining of election personnel can, and does, pro-duce timely accurate elections by mail. We alsofound the dismissal of Oregon’s experiences tobe disturbing at best and insulting at worst.

39


41


Constitution Project Resourceshttp://www.constitutionproject.org

Davidson, Michael, Notes on the History ofArticle I, Section 4, Clause 1, April 3, 2001.

Karlan, Pamela, Stanford University,Congressional Authority to Regulate Electionsand Election Technology, February 2001.

Weich, Ronald and Carlos Angulo, ZuckermanSpaeder LLP, Lawsuits Arising Out of theNovember 2000 Presidential Election, April 27,2001. (This memo is updated as needed.)

In addition, the Constitution Project has been monitor-ing election reform legislation in the 107th Congress.An overview memo and side-by-side comparisonof leading proposals is available at the website.

State Reports

California: California Internet Voting Task Force,A Report on the Feasibility of Internet Voting,January, 2000. http://www.ss.ca.gov

Florida: The Governor’s Select Task Force OnElection Procedures, Revitalizing Democracyin Florida, Standards and Technology, March 1,2001. http://www.collinscenter.org

Georgia: Secretary of State Cathy Cox, The 2000Election: A Wake-up Call for Reform andAction, Report to the Governor and Members ofGeneral Assembly, January 2001.http://www.sos.state.ga.us

Iowa: Chet Culver, Iowa Secretary of State, IowaCommissioner of Elections, Iowa Registrar ofVoters, Iowa’s Election 2000: Facts, Findingsand Our Future, March 12, 2001.http://www.sos.state.ia.us

Maryland: Special Committee on Voting Systemsand Election Procedures in Maryland, Reportsand Recommendations, February 2001.http://www.sos.state.md.us

Missouri: Secretary of State Matt Blunt, MakingEvery Vote Count: A Report of Secretary ofState Matt Blunt to the People of Missouri,January 29, 2001. http://mosl.sos.state.mo.us

New York: New York State Task Force on ElectionModernization, Recommendations for the Fall2001 Elections, May 2001. http://www.state.ny.us/governor/electionmodernization/Oregon: Secretary of State Bill Bradbury andOregon Association of County Clerks PresidentDana Jenkins, Report of the Oregon ElectionsTask Force, February 6, 2001.http://www.sos.state.or.us

A P P E N D I X C :R E S O U R C E S

42


Additional Resources

Administration and Cost of Elections Project(ACE). International Foundation for ElectionSystems (IFES), International IDEA, and theUnited Nations. Washington, D.C.: ACE WebsiteVersion 2001. http://www.aceproject.org/

Caltech-MIT Voting Technology Project, Voting –What Is, What Could Be, July 2001.http://www.vote.caltech.edu/

Committee on Government Reform, U.S. Houseof Representatives, Election Reform in Detroit:New Voting Technology and Increased VoterEducation Significantly Reduced UncountedBallots, April 5, 2001. Prepared forRepresentative Henry A. Waxman, RankingMember, Committee on Government Reform.http://www.house.gov/waxman/

Committee on Government Reform, U.S. Houseof Representatives, Income and RacialDisparities in the Undercount in the 2000Presidential Election, July 9, 2001. Preparedfor Representative Henry A. Waxman, RankingMember, Committee on Government Reform.http://www.house.gov/waxman/

Department of Defense WashingtonHeadquarters Services, Federal VotingAssistance Program, Voting Over the InternetPilot Project Assessment Report, June 2001.

General Accounting Office, Elections: IssuesAffecting Military and Overseas AbsenteeVoters. GAO-01-704, May 9, 2001.http://www.gao.gov

General Accounting Office, Elections: The Scopeof Congressional Authority in ElectionAdministration. GAO-01-470, March 13, 2001.http://www.gao.gov

Hoffman, Lance, Making Every Vote Count,1988, Sponsored by the Markle Foundation.http://www.seas.gwu.edu/~cpi/library/papers.html

Hoffman, Lance, Internet Voting: Not Ready forPrime Time (Yet), January 2001 (slide

presentation). http://www.seas.gwu.edu/~cpi/library/presentations/internet-voting/index.htm

Mann, Thomas, Brookings Institution, AnAgenda for Election Reform, Policy Brief #82,June 2001. http://www.brookings.org

National Science Foundation, Report on theNational Workshop on Internet Voting: Issuesand Research Agenda, March 2001.http://www.internetpolicy.org/research/results.html

National Commission on Election Standards andReform, Report and Recommendations toImprove America’s Election System, May 2001.The Commission is a joint effort between theNational Association of Counties (NACo) and theNational Association of County Recorders,Election Officials and Clerks (NACRC).

The National Commission on Federal ElectionReform, To Assure Pride and Confidence in theElectoral Process (August 2001). Web reference.http://www.reformelections.org/

Norton Garfinkle and Patrick Glynn, Institute forCommunitarian Policy Studies, George WashingtonUniversity, Report on Election Systems Reform,July 2001. http://www.gwu.edu/~icps/

Robert Crown Law Library, Stanford UniversityLaw School, Election 2000 website,

http://election2000.stanford.edu

Saltman, Roy, Accuracy, Integrity, andSecurity in Computerized Vote-Tallying,August 1988, sponsored by the MarkleFoundation. http://www.nist.gov/itl/lab/specpubs/500-158.htm (Additional papers by Roy Saltmanare available at the Constitution Project’s website.)

United States Commission on Civil Rights,Voting Irregularities in Florida During the2000 Presidential Election, June 2001.http://www.usccr.gov/

Date post:	16-Jul-2020
Category:	Documents
Upload:	others
View:	2 times
Download:	0 times

BUILDING CONSENSUS ON ELECTION REFORM...launched an election reform initiative by conven-ing the...

Documents