Potential for Innovation in Election Technology
July 16, 2018
The 2000 presidential election posed countless issues, including butterfly ballots, partially punched cards, four to six million lost votes, and outraged citizens across the political spectrum. These issues led California Institute of Technology president David Baltimore and Massachusetts Institute of Technology president Charles Vest to approach the election technology failure more scientifically. With a team of computer scientists, mechanical engineers, and political scientists, they reimagined—and reengineered—the election process. Although Baltimore and Vest pioneered advancements in the election process, the reliability of voting infrastructure today still remains questionable. As we approach the upcoming election cycle this Fall, policy and technology experts must come together to rectify a failing system.
Evolution of Election Technology
Over the last two decades, with the establishment of the Election Assistance Commission (EAC) in 2002, digital technology has gradually replaced paper voting systems. Since the EAC administered the first Election Administration and Voting Survey (EAVS) in 2004, the U.S. has employed eight different categories of voting systems: optical scans, two versions of direct recording electronic (DRE) voting machines, hybrid voting machines, hand-counted paper ballots, punch cards, lever machines, and other systems. The biennial survey documents the nation’s transition away from many of these systems, including lever machines and punch-card ballots. By 2016, approximately 61% of jurisdictions were using optical scan machines to scan and digitize paper ballots.
Image: Relative deployment of different voting systems by vendor. Source: Penn Wharton Public Policy Initiative
Despite significant restructuring of voting systems since 2000, outdated voting machines plague countless localities today. According to the Brennan Center for Justice, a nonpartisan law and public policy institute, as of 2016, an estimated 43 states were using voting machines purchased over 10 years prior, of which 14 states were using machines purchased over 15 years prior. The collapse of these systems, coupled with the inability of states to fund their replacement, poses a serious threat to the integrity of the electoral process. More specifically, elected officials identify four main concerns with aging machines: (1) equipment failure around major elections, (2) undermined confidence in election results, (3) inability of current computers to connect to outdated voting software, and (4) replacement of machine parts that are no longer manufactured.
Image: Age of voting machines in U.S. states as of 2016. Source: Brennan Center for Justice
Current State of the Industry
Part of the failure with these archaic machines can be attributed to the lack of transparency and competitive dynamics of the election technology industry. The industry earns approximately $300 million annually, a relatively small figure compared to the $80 billion annual federal IT expenditure. Dominated by three firms that are neither publicly nor independently held, very little data even exists on the industry. Moreover, election technology decisions are spread over 10,000 county officials; this severely fragmented customer base renders composite market information scarce.
Given these industry dynamics, external IT vendors unsurprisingly have little interest in entering the market. Vendors have expressed concern over the enormous development costs and delays associated with certifying new voting technology, which disincentivize innovation. Consequently, both vendors and customers find themselves ‘stuck’ with older technologies nearing the end of their 10- to 15-year product lifecycles. In addition, because of these relatively lengthy lifecycles, competition for technology replacement is centered around only a limited number of events. These barriers thus constrain industry growth and the potential of refurbishing outmoded systems.
Strategies for Developing New Election Systems
In light of this limited industry growth, election officials have proposed various strategies to spur the creation of new systems. Coalition formation emerges as a common solution in smaller districts, where election officials struggle to obtain products customized for idiosyncratic local requirements (e.g. budget constraints, voting format preferences, multilingual voting, disability access). By minimizing transaction cost and aggregating buyer power, customer coalitions would facilitate bargaining for customized election products, thereby lowering prices.
Some larger counties—including Travis County, TX and Los Angeles County, CA—have spearheaded the development of open-source systems. In the current market, customers have little access to the underlying code that powers a voting system. An open source model, however, makes that code public, allowing professionals to inspect and modify it. For instance, OSET Institute, a Silicon Valley nonprofit research corporation, is launching the TrustTheVote project to build flexible, adaptable election administration software using open-source principles. With greater citizen oversight, this model fosters transparency and legitimacy in election administration.
The open-source systems developed in Travis County and LA County are examples of modularity. Modularity refers to the structural flexibility of modern IT system designs, which integrate different combinations of components into a broad blueprint. From a regulatory standpoint, modularity overcomes current requirements that mandate voting systems be presented in their entirety before testing for certification. Instead, professionals can construct larger platforms from smaller components without fully redesigning code. Since modularity involves separating programs into independent, interchangeable parts, the move towards a more modular election technology system would promote mass market hardware, encourage reuse, reduce development cost, and support timely equipment upgrades.
Funding the Development of New Election Systems
While various strategies exist for revamping election technology, states struggle with funding these new systems. In 2002, through the Help America Vote Act (HAVA), Congress allocated $3 billion to states to purchase new voting technology. However, as of 2016, most of these funds have been depleted; in fact, less than 10% of original funds remain in 65% of states and territories.
The Center for American Progress, a public policy research organization, estimates that updating outdated voting machines would cost approximately $1 billion, and replacing paperless machines would cost between $130 million and $400 million. Unfortunately, 80% of election officials responded in a survey that they did not have sufficient funds.
Although many believe the federal government has an obligation to allocate funds for new election technology, federal funds will not be available anytime soon, according to the National Conference of State Legislatures (NCSL), a bipartisan non-governmental organization that assists state legislatures. This is because HAVA failed to address the ongoing costs of maintaining voting equipment, and Congress has not yet seen a new proposal requesting the use of federal funds to overhaul election technology. The NCSL proposes a number of funding options, including direct appropriation to purchase new equipment statewide, splitting the cost between states and counties, and setting up grant programs or low-interest programs for counties.
Ultimately, upgrading voting technology calls for a combined effort. While funding poses a persistent challenge, the problem also rests in technological inefficiency, political inaction, customer fragmentation, and vendor uncertainty. Solving this issue requires more than just money; it demands a system of vendors, customers, and policymakers who will cooperate to dismantle a rigid, stagnant election technology industry.
Student Blog Disclaimer
The views expressed on the Student Blog are the author’s opinions and don’t necessarily represent the Wharton Public Policy Initiative’s strategies, recommendations, or opinions.