Options for New Jersey s Voter-Verified Paper Record Requirement

Similar documents
Pennsylvania Needs Resilient, Evidence-Based Elections

Analysis and Report of Overvotes and Undervotes for the 2014 General Election. January 31, 2015

New Mexico Canvass Data Shows Higher Undervote Rates in Minority Precincts where Pushbutton DREs Were Used

DIRECTIVE May 21, All County Boards of Elections Directors, Deputy Directors, and Board Members. Election Administration Plans SUMMARY

DIRECTIVE November 20, All County Boards of Elections Directors, Deputy Directors, and Board Members. Post-Election Audits SUMMARY

HOUSE BILL 1060 A BILL ENTITLED. Election Law Delay in Replacement of Voting Systems

Analysis and Report of Overvotes and Undervotes for the 2012 General Election. January 31, 2013

Post-Election Online Interview This is an online survey for reporting your experiences as a pollworker, pollwatcher, or voter.

GAO ELECTIONS. States, Territories, and the District Are Taking a Range of Important Steps to Manage Their Varied Voting System Environments

STATE OF NEW JERSEY. SENATE, No th LEGISLATURE

The Case Against. Diebold and Florida s Division of Elections

Undervoting and Overvoting in the 2002 and 2006 Florida Gubernatorial Elections Are Touch Screens the Solution?

VOTERGA SAFE COMMISSION RECOMMENDATIONS

Global Conditions (applies to all components):

Cuyahoga County Board of Elections

IT MUST BE MANDATORY FOR VOTERS TO CHECK OPTICAL SCAN BALLOTS BEFORE THEY ARE OFFICIALLY CAST Norman Robbins, MD, PhD 1,

E-Voting, a technical perspective

The documents listed below were utilized in the development of this Test Report:

ARKANSAS SECRETARY OF STATE

GEORGIA VERIFIABLE VOTING LEGISLATIVE AND LEGAL CHRONOLOGY

Hard Facts about Soft Voting

ARKANSAS SECRETARY OF STATE. Rules on Vote Centers

1S Recount Procedures. (1) Definitions. As used in this rule, the term: (a) Ballot text image means an electronic text record of the content of

NC General Statutes - Chapter 163 Article 14A 1

Sincerely, Rebecca Mercuri, Ph.D. 116 Grayson Ave. Mercerville, NJ /

Voting System Examination Election Systems & Software (ES&S)

REQUESTING A RECOUNT 2018

The purchase of new voting equipment

COMMISSION CHECKLIST FOR NOVEMBER GENERAL ELECTIONS (Effective May 18, 2004; Revised July 15, 2015)

H 8072 S T A T E O F R H O D E I S L A N D

CRS Report for Congress

Allegheny Chapter. VotePA-Allegheny Report on Irregularities in the May 16 th Primary Election. Revision 1.1 of June 5 th, 2006

Electronic Voting Machine Information Sheet

ROBERT WARREN, being duly sworn deposes and says: ( Board ), and in such capacity am fully familiar with the facts and circumstances of the within

CALTECH/MIT VOTING TECHNOLOGY PROJECT A

Volume I Appendix A. Table of Contents

The name or number of the polling location; The number of ballots provided to or printed on-demand at the polling location;

NOTICE OF PRE-ELECTION LOGIC AND ACCURACY TESTING

Testimony of George Gilbert Director of Elections Guilford County, NC

New Voting Systems; Old Law. Brittany Westfall West Virginia Secretary of State s Office

Election Audit Report for Pinellas County, FL. March 7, 2006 Elections Using Sequoia Voting Systems, Inc. ACV Edge Voting System, Release Level 4.

OSCE Parliamentary Assembly Post-Election Statement U.S. General Elections 6 November 2008

RR/CC RESPONSE TO GRAND JURY REPORT

United States Election Assistance Commission

Logic & Accuracy Testing

FINAL REPORT OF THE 2004 ELECTION DAY SURVEY

County Clerk s Office Election Division Voting System

VIA FACSIMILE AND ELECTRONIC MAIL. January 22, 2008

Trusted Logic Voting Systems with OASIS EML 4.0 (Election Markup Language)

Sincerely, Jon Husted Ohio Secretary of State

Election System Upgrade

Manual Audit Requirements

GAO. Statement before the Task Force on Florida-13, Committee on House Administration, House of Representatives

SECURITY, ACCURACY, AND RELIABILITY OF TARRANT COUNTY S VOTING SYSTEM

ASSEMBLY, No STATE OF NEW JERSEY. 217th LEGISLATURE INTRODUCED FEBRUARY 27, 2017

L9. Electronic Voting

IC Chapter 15. Ballot Card and Electronic Voting Systems; Additional Standards and Procedures for Approving System Changes

AFFIDAVIT OF POORVI L. VORA. 1. My name is Poorvi L. Vora. I am a Professor of Computer Science at The George

AFFIDAVIT OF DOUGLAS W. JONES. 1. I am an Associate Professor of Computer Science at the University of

In the Margins Political Victory in the Context of Technology Error, Residual Votes, and Incident Reports in 2004

Colorado Secretary of State Election Rules [8 CCR ]

Anoka County Procedural Law Waiver Application Narrative Section A: Background Implementation of the Help America Vote Act of The Help America

Direct Recording Electronic Voting Machines

RULES FOR REIMBURSEMENT OF EXPENSES FOR STATE-FUNDED ELECTIONS (Effective February 6, 2004; Revised December 29, 2015)

INSTRUCTIONS AND INFORMATION

Michigan Election Reform Alliance P.O. Box Ypsilanti, MI

The Future of Elections: Technology Policy & Funding Conference

WHY, WHEN AND HOW SHOULD THE PAPER RECORD MANDATED BY THE HELP AMERICA VOTE ACT OF 2002 BE USED?

Sincerely, Jon Husted Ohio Secretary of State

Good morning. I am Don Norris, Professor of Public Policy and Director of the

IN-POLL TABULATOR PROCEDURES

PROCESSING, COUNTING AND TABULATING EARLY VOTING AND GRACE PERIOD VOTING BALLOTS

Act means the Municipal Elections Act, 1996, c. 32 as amended;

CALTECH/MIT VOTING TECHNOLOGY PROJECT A

PROCEDURES FOR THE USE OF VOTE COUNT TABULATORS

Testimony of. Lawrence Norden, Senior Counsel Brennan Center for Justice at NYU School of Law

IC Chapter 3. Counting Ballot Card Votes

Ranked Voting and Election Integrity

Board of Elections. Department Summary FY 2003 FY 2004 FY 2004 FY 2005 Actual Adopted Current Adopted Budget Budget Budget. Department Description

Maryland State Board of Elections Comprehensive Audit Guidelines Revised: February 2018

The E-voting Controversy: What are the Risks?

2004 Kansas State Plan HELP AMERICA VOTE ACT OF 2002

Voting Laws Roundup 2018

Voting Accessibility: The devolution of voting technology. Diane Cordry Golden, Ph.D June 2017

Election Dates and Activities Calendar

Automating Voting Terminal Event Log Analysis

GENERAL ASSEMBLY OF NORTH CAROLINA SESSION 2015 SENATE BILL 667 RATIFIED BILL

Electronic Voting Machine Information Sheet

14 Managing Split Precincts

COUNTY OF SACRAMENTO CALIFORNIA

Election Dates and Activities Calendar

This page intentionally left blank

Case 5:02-cv DDD Document 273 Filed 11/15/2004 Page 1 of 16 UNITED STATES DISTRICT COURT NORTHERN DISTRICT OF OHIO EASTERN DIVISION

Key Considerations for Implementing Bodies and Oversight Actors

AUDIT & RETABULATION OF BALLOTS IN PRECINCTS WHERE A DISCREPANCY EXISTS

RANKED VOTING METHOD SAMPLE PLANNING CHECKLIST COLORADO SECRETARY OF STATE 1700 BROADWAY, SUITE 270 DENVER, COLORADO PHONE:

CRS Report for Congress

UPDATE ON RULES. Florida Department of State

Oregon. Voter Participation. Support local pilot. Support in my state. N/A Yes N/A. Election Day registration No X

Conny B. McCormack, Registrar-Recorder/County Clerk REPORT OF ON-SITE OBSERVATION OF FLORIDA S SEPTEMBER 10, 2002 PRIMARY ELECTION LESSONS LEARNED

Transcription:

Verifiable Elections for New Jersey: What Will It Cost? This document was prepared at the request of the Coalition for Peace Action of New Jersey by VerifiedVoting.org (VVO). VerifiedVoting.org works to promote transparent, reliable and publicly verifiable elections nationwide, in part through policy recommenddations on voter-verified paper ballots, manual audit requirements, and related security and reliability measures. We appreciate the opportunity to provide information about cost and related issues for various options to meet New Jersey s voter-verified paper record requirement. Options for New Jersey s Voter-Verified Paper Record Requirement In reviewing the options facing New Jersey for meeting the requirement to deploy voterverified paper records (VVPR) statewide by 2008, several factors must be taken into account. Aside from the question of whether New Jersey s largest vendor, Sequoia Voting Systems, will have a suitable printer available for the most common type of voting machine used in the state (Advantage full-face direct recording electronic DRE), as promised, the high proposed cost 1 of those printers and the possible alternatives to such systems should be considered. In addition, since audits and recounts using the voter-verified paper records are needed to assure the accuracy of any voting system, it is worthwhile to take into consideration the ease of use of different types of VVPR for that purpose, and the cost factors involved. 1. Precinct-Count Optical Scan with Ballot-Marker A number of other states with voter-verified paper record rules have implemented precinct-count optical scan (PCOS) and other systems in order to meet the need for voters to be able to check that their ballots are marked accurately, and for officials to have a hard-copy record of voter intent to check for audits and recounts. Optical scan systems are used by more voters than any other type of voting system nationwide. These deploy a paper ballot marked by the voter, and thus inherently voterverified. Such systems are highly reliable, have a low residual vote rate, prevent overvoting, and are simple to deploy (often requiring fewer pollworkers and/or less training time). Should the scanner break down for any reason, voters can still mark and cast their ballots (preventing long lines from forming), which is not the case when a DRE breaks down. For accessibility, such PCOS systems may be combined with one of several alternatives. The most desirable of these alternatives is an accessible ballot-marking device (BMD), which uses a DRE-like interface to assist the voter in marking a paper ballot which can 1 See page two below for cost comparisons for printers. April 21, 2007 VerifiedVoting.org Page 1 of 6

then be counted using the precinct-count optical scan system along with all other voters ballots. This type of device also fulfills a crucial function, which is to allow voters with disabilities or different language abilities to review their ballot for accuracy before casting it. Thus all voters can verify the printed ballot before casting, making a fully accessible system. 2. Blended Optical Scan with Paging DRE Another option, though less desirable, is to combine precinct-count optical scan ballots with a voter-verified paper record printer-equipped paging DRE (such as the Edge, used in some NJ jurisdictions). This is a less desirable option for two key reasons. First, it may result in a segregated ballot type for voters who need to use an accessible system, since those voters would be casting their ballots on the DRE, while the majority of voters not needing an accessible interface would be casting their ballots on paper for scanning/ tabulating. Second, this means a more complicated blended system for pollworkers to learn and deploy. 3. Blended Optical Scan with Full-Face DRE A third option, less desirable still, is to deploy a full-face DRE such as the Advantage in current wide use in NJ as the accessible system. This is the worst of such blended options because this system is known to have a worse residual vote rate 2 than paging-style DREs (and likely also worse than paper optical scan ballot systems). It also makes for a segregated ballot for certain classes of voters, and is more likely to negatively impact language minority voters than other systems. 4. Add Printer to Existing DREs For the reasons listed above, keeping New Jersey s existing DREs (mostly Sequoia Advantage) and adding voter-verified paper record printers to those is a costly option which does nothing to address reliability issues. It still represents something of a blended system, because paper ballots must still be printed for purposes of absentee voting and for emergency polling place ballots (in the event of DRE malfunction). Technology Cost Issues The printer module for the Advantage system is proposed at the sum of $2000 per unit, as much as two times more costly than the next most expensive printer module for other DRE systems (cf. $1000 for the Sequoia Edge), and four times more than other printers. 3 2 The residual vote rate on bottom-of-ticket items was a stunning five times higher on NJ s full-face DREs than on paging DREs in a recent study. Even a highly publicized top of ticket Senate race showed a full percentage point worse performance on the full-face DREs. http://www.umsl.edu/~kimballd/nj06resid.pdf 3 Indeed, some jurisdictions negotiated voter-verified paper record printer modules for free, at such time as state or federal law would require them. Cf. San Diego County contract with Diebold Election Systems, Inc. from December 2003, and reduced bids for DREs with VVPR printers offered by Diebold in Ohio, at a lower unit price than earlier DRE bids without VVPR printers (2004). Sources provided upon request. April 21, 2007 VerifiedVoting.org Page 2 of 6

The probable expenditure to retrofit existing machines in order to make them auditable will be ~$21.5 Million for printers alone. (This does not include sums already expended or which must yet be expended to upgrade some existing systems so that printers can be attached.) A review of actual contracts shows a range of pricing arrangements in different Sequoia customer jurisdictions, but in general, the cost of maintaining and operating optical scan systems tends to be lower than the cost of maintaining and operating DRE systems. More complex systems = higher cost A large jurisdiction using a blended system, consisting of optical scan and a paging DRE (with VVPR printer) deployed for accessibility, demonstrated a key point about operations and costs: For DREs, the complexity of the system means longer training times for election officials (five days for a Sequoia DRE system, versus one day for a Sequoia PCOS system). 4 The DRE system also requires more steps to take (and consequently higher costs of operation) for election day set up and significantly more steps to take for pre-election Logic and Accuracy Testing, when compared to the optical scan system. Further, the more balloting methods in place in a county, the greater the administrative burden and cost at the county level. 5 A polling place DRE system is always a two-ballotmethod system because paper optical scan ballots are also deployed for absentee and emergency use. A polling place PCOS system is a single-ballot-method system; paper optical scan ballots are deployed for all purposes, greatly simplifying the system. More machines per polling place = higher cost For certain types of costs, the higher quantity of machines needed when deploying a DRE solution results in higher ongoing expenses because the ongoing support is factored as a multiple of the number of units deployed. In other words, the per unit cost may be similar for DREs and for scanners, but only one scanner is needed per polling place, while numerous DREs are needed. Thus when viewed by polling place, scanners save costs. For example, one contract 6 we examined showed a per unit cost for extended warranty (parts only) at $150 per scanner per year, and $130 per DRE+printer per year. At first glance, this seems lower for DREs than for scanners. But calculated at a conservative 4 DREs per polling place, versus one scanner per polling place, and the cost is still $150 per polling place for scanners, but $520 per polling place for DREs, per year. For an extended warranty for parts including visit, the cost is $150/yr for scanner polling places, and $580/yr for DRE polling places nearly four times as much for DREs. 4 Contract, Alameda County, CA (check) 5 http://www.michigan.gov/documents/uniform_voting_system_2_71046_7.doc 6 Alameda County, CA April 21, 2007 VerifiedVoting.org Page 3 of 6

Another state 7 showed bid pricing for annual maintenance per Insight PCOS at $260 considerably higher per unit than for Edge II DREs at $150. However, for polling places deploying all DREs, at a conservative calculation of 4 per polling place, that s $600 per polling place annually for a DRE solution, versus $260 per polling place for PCOS. If you combined one DRE plus one PCOS for a blended system, it would still be $410 per polling place versus $600 for all DRE polling places. Storage and transportation for the higher number of machines needed for a DRE solution also takes its toll. Full-face DREs in particular can weigh over 200 lbs. and require as much as 28 cu. ft. of storage space per device. A PCOS system (including a ballot-marking device) involves two machines per polling place, and those machines are considerably smaller than full-face DREs, at 19-29 lbs. each, requiring only 4 cu. ft. of storage space. A PCOS system requires half the storage space or less than for DREs, depending how many DREs are in use per polling place now for a given jurisdiction. 8 Ease of Use and Related Cost Issues Auditing voter-verified paper records to check the accuracy of the vote count and uncover any potential malfunctions or errors is an essential component of verifiable elections. As we cited in recent testimony to the Committee on House Administration of the U.S. House of Representatives 9, there is a considerable difference in auditing individual paper ballots versus auditing printed records produced by DREs. While improved printer designs that provide for separate sheets for each record (rather than records on a continuous thermal paper roll) help mitigate the cost differential, this type of improved voter-verified paper record may or may not be available in all New Jersey jurisdictions. Printers in current use with the ES&S ivotronic and with the Sequoia Edge, for example, use continuous thermal paper rolls. Auditing records from continuous roll DRE printer systems can take as much as four times longer than auditing paper optical scan ballots. The ease of use translates into cost savings when reviewing optical scan ballots. This would make a significant difference in the number of counting teams or the length of time those teams would need to be employed in the counting process. Finally, some systems necessitate changing a printer cartridge if enough votes are cast during the voting day, as well as removal of printer cartridges at the end of the voting day. These extra steps can often mean additional pollworkers are needed, resulting in 7 Iowa HAVA Voting System Master Pricing List May 16, 2005 8 While storage for paper ballots would also be required for a PCOS system, that is offset by other factors with DRE systems, including printer systems, as well as the paper ballots that must be printed for use with DRE systems anyway (absentee, emergency, provisional). 9 See http://www.verifiedvotingfoundation.org/downloads/pamelasmithtestimonyfinal_2007mar20.pdf for detailed discussion of audit processes and comparative time involved. April 21, 2007 VerifiedVoting.org Page 4 of 6

higher costs. In contrast, optical scan ballots do not require any such special handling midway through Election Day, nor at the end of the day. Ballot Cost Issues In New York, Sequoia Voting Systems distributed a document which sought to sway counties to select DREs as the replacement for the State s lever machines, over precinctcount optical scan systems. In it, they provided much incorrect information, 10 including what appeared to be an inflated price for printed ballots at 0.60 or more. Printed optical scan ballots in other states are quoted at $0.29 or lower. 11 Other Cost Considerations Research has been conducted in several states to compare cost per voter for jurisdictions that conduct elections using optical scan compared to jurisdictions that conduct elections using DREs. The costs are calculated by comparing the number of registered voters with the annual elections budget (ongoing costs, not acquisition costs). In states like Florida, North Carolina and elsewhere, these costs have been demonstrated to be as much 30-40% higher for DRE jurisdictions than for PCOS jurisdictions, and in some cases considerably more. 12 News reports from Ohio, Florida, Maryland and Wyoming have highlighted unexpected and dramatically increasing annual operations costs after switching to DRE systems for everything from machine storage to election set-up to consumables such as batteries for DREs. 13 Cost to Adopt Optical Scan Statewide Given the advantages of precinct-count optical scan systems ease of use, costeffectiveness, reliability, simplicity over DRE systems, it makes sense for New Jersey to switch to PCOS in all its jurisdictions. Although doing so represents a capital expenditure, the savings at the county level on an ongoing basis can offset this cost. In addition, upon passage of federal legislation with supporting funding, New Jersey could receive funds to support this move. Even without external support, however, shifting to PCOS still makes fiscal sense. 10 Vendors may prefer jurisdictions to select DRE systems not only because acquisition costs are so much higher with more machines needed for every polling place, meaning greater initial profits, but also because ongoing support and maintenance means greater ongoing revenues for vendors than with optical scan systems. 11 http://nyvv.org/reports/paperballotprintingcosts.pdf 12 http://www.verifiedvotingfoundation.org/article.php?id=5747 http://www.ncvoter.net/affordable.html 13 Documentation: http://www.verifiedvotingfoundation.org/article.php?id=5746 http://www.verifiedvotingfoundation.org/downloads/gillesburgerltr10.pdf http://www.verifiedvotingfoundation.org/article.php?id=5808 http://www.columbusdispatch.com/dispatch/contentbe/dispatch/2006/03/05/20060305-c1-00.html http://soswy.state.wy.us/press/havafunds.pdf April 21, 2007 VerifiedVoting.org Page 5 of 6

Based on a calculation of one PCOS scanner-tabulator and one ballot-marker for every polling place, this would represent ~$38 Million. A single scanner can accommodate upwards of 2,500 voters on election day. Costs for scanners and ballot-markers are estimated at a total of $11,000 for a set, although it should be noted that some states have negotiated lower rates depending on volume pricing. For example, Iowa s statewide contract pricing comes in at under $10,000 for the set. Other states and counties also had pricing below $10,000 in quantities of 1000 each. The State of New Jersey could certainly negotiate substantive savings if a price for the whole state were obtained. If it does so, the cost could be closer to ~$34.5 Million or less. Conclusion Considering that New Jersey has obtained over 10,000 machines to date, most of which cost the state between $6000 and $8000 (depending on whether they were configured to provide audio or not), this price tag seems small by comparison. With possibly millions of HAVA funding left in New Jersey s coffers, the differential can be minimized further. Expending any funds on printers for DREs, which were more costly to begin with, are less accessible, have worse under-vote rates and higher ongoing maintenance and operations costs would simply be throwing good money after bad. Monmouth County incorporated a clause in their agreement with the vendor indicating they could return the equipment for a refund in the event that printers were not available by a date certain. Monmouth and indeed the whole State should seek to use any available funds on ballot-markers and replace all its Advantage machines with scanners. VerifiedVoting.org recommends that the State of New Jersey consider all of the factors involved, especially the ongoing costs to the counties, and resolve to meet the requirement for voter-verified paper ballots by shifting to a more reliable paper-ballotbased system using precinct-based optical scan and ballot-marking devices for accessibility. April 21, 2007 VerifiedVoting.org Page 6 of 6