Yesterday the United States' largest voting equipment vendor, Election Systems & Software (ES&S), announced the purchase of Premier Election Solutions, our nation's second largest vendor, and a product of the Diebold Corporation's North American operations. If this sale goes forward, ES&S will control a huge majority of the voting equipment market in the United States. According to Verified Voting, more than 120 million registered voters live in American jurisdictions using one of these two companies' systems. In contrast, the nation's third largest elections vendor, Sequoia Voting Systems, provides equipment in jurisdictions with only some 26 million registered voters -- and seems to be on shaky ground, having been sold several times in recent years and still waiting to have its latest optical scan system certified by the federal Election Assistance Commission.
Whether the sale goes through remains a question. Election integrity activists at Black Box Voting have pledged to fight it. ES&S (then called American Information Systems) previously attempted to consolidate the voting industry in 1997 with a purchase of Business Records Corporation (BRC), but the U.S. Department of Justice on anti-trust grounds required that acquisition of BRC to be split between ES&S and Sequoia.
Regardless of its ultimate outcome, this latest potential consolidation in ownership of our voting equipment highlights the broken nature of American election administration. We run democracy on the cheap at the national level, and pay for it with lost votes, untrustworthy software and exorbitant costs for public interest improvements due to companies recouping expenses by abusing their local monopolies.
FairVote has long suggested a full public ownership model, similar to what the state of Oklahoma and nations abroad have done. We should keep pursuing this "public option," but also consider additional ways to gain control of the election process and foster better, more reliable equipment. Some groups are seeking to hold vendors legally accountable for past failures to uphold election integrity. Looking forward, one straightforward step would address a glaring problem: the process of certifying equipment. To open up the market to more competitors and secure certain basic rights of transparency and quality control, the public should pay for the costs of certification.
Better certification processes for voting equipment of course are absolutely essential, as underscored by how more rigorous certification processes in recent years have exposed major problems with proposed equipment. Election results also keep demonstrating how systems already certified for our most important elections can have serious flaws. For example, the Humboldt County (CA) Election Transparency Project discovered that a Premier/Diebold optical scan paper ballot system dropped 197 ballots in 2008, while a FairVote analysis this year found that the same system dropped 0.4% of ballots in an election in Aspen (CO).
But each new revelation and each new good idea for updating certification standards at the federal level and state level makes it harder for companies to comply, in turn stretching out the timeline for certification and greatly increasing companies' costs. Paying for companies' costs of certification would cost taxpayer dollars, of course, and should have some reasonable limits. But these upfront costs promise to pay big dividends for our democracy in the long-term. It would allow new companies to get a competitive product on the market before they know for sure they will be able to sell it - resolving the catch-22 that today makes it so difficult for any new company to compete with the dominant companies. It also would make it easier to justify ongoing updates to the voting standards, rather than essentially adding new "unfunded mandates" on the vendors who either go out of business or, more typically, give up after barely getting started. The quality of voting equipment and software should also rise as companies would be required to do more than just "get by," and county and state governments would pay less for better equipment and upgrades - right now they typically face excessive fees for equipment, ongoing services and upgrades from vendors trying to recoup their certification costs and able to take advantage of their near monopoly of the industry.
In exchange for paying for the certification process, the public also should secure greater rights of transparency and general ownership of the process. For example, New York State's latest contracts for new equipment include a sensible provision that any additional contracts for services and new features involving the equipment will be open to competitive bidding rather than the jurisdiction having to accept the vendor's monopoly power. Taxpayers also should require much greater access to the software code, if not full open source software, and a requirement for "modular" components that would make it easier to piece together separately certified systems for an election rather than rely on just one company for election services.
Exclusive focus on pre-election certification will never be sufficient , as we must also focus on post-election verification and audits. By verifying all election counts, the certification process would become part of a "belt and suspenders" approach. With the latest optical scan paper ballot systems having the capacity to create redundant records of every ballot, these records can be made publicly available, as they are in cities from San Francisco (CA) to Burlington (VT). When coupled with manual audits and appropriate privacy safeguards, they will allow the public to verify vote tallies and immediately identify errors.
The bottom line is that the existing regime is broken. Let's stop outsourcing democracy and make sure that citizens are in control.