Last November, Richland County residents seeking to participate in local elections encountered an unanticipated hindrance at polling stations: stagnant lines of voters unable to cast their ballots because of malfunctioning voting machines. The lines reportedly were so outrageous that some residents had to wait upwards of seven hours to vote. Many voters grew impatient and left polling stations without submitting a ballot. Moreover, the disarray was hardly confined to election day. In the week after polls closed, a court-ordered recount of the election results sparked a back-and-forth legal battle between Democrats and Republicans over whether a local or statewide election agency should be tasked with tallying the votes in the recount. The dispute was not settled until the South Carolina Supreme Court intervened, and nearly two weeks elapsed before the election results were finalized. In 2004, the South Carolina State Election Commission (SEC) purchased roughly twelve thousand iVotronic voting machines for around $34 million. At the time of their purchase, the iVotronic systems were considered ultramodern direct-recording electronic (DRE) voting technology. Like most DRE models, the iVotronic enables voters to cast their vote via an electronic touch-screen without handling a paper ballot. Once a voter has electronically submitted his choice, the machine stores the selection in an internal memory device. Upon an election’s conclusion, the iVotronic machine prints a tape displaying the total number of votes cast for each office as well as the total number of votes cast for each candidate. In theory, DRE systems such as iVotronic provide a modern solution to antiquated election problems. In practice, however, DRE systems do not always function so smoothly.
The Richland County election fiasco illuminated a major flaw with South Carolina’s statewide voting system: the iVotronic machines do not provide voter-verified paper audit trails (VVPAT). A VVPAT is essentially a voting receipt, printed by a DRE voting machine after a ballot is cast, that informs a voter of how the machine recorded her vote selection. The voter then verifies that the machine correctly recorded her vote. A significant benefit of VVPATs is that they provide a mechanism for election officials to determine whether electronically stored election results are accurate if an election audit or recount becomes necessary. The iVotronic machines used throughout South Carolina lack VVPAT technology, as VVPAT was not yet available when the state purchased the machines. South Carolina thus faces a serious problem in discerning election results whenever its iVotronic machines produce results that are inconsistent with the number of registered voters known to have voted at a particular polling station.
There are a handful of potential solutions that South Carolina could incorporate to improve its election procedures. First, South Carolina could retrofit its machines with VVPAT technology. Fixing the iVotronic models with VVPATs probably would help to lessen undervoting and vote flipping problems associated with current models. But retrofitting would not necessarily solve all mechanical difficulties associated with the iVotronic models, and installation of VVPAT technology onto existing machines would cost upwards of $17 million. A second option for South Carolina is to purchase a new system of voting machines. In fact, the SEC already requested $5 million from the South Carolina General Assembly for the fiscal year 2012-2013 to establish a fund to procure new machines, but the request was denied. As with retrofitting, purchasing new machines necessitates a substantial financial commitment, and federal HAVA grants, which were instrumental in the state’s purchase of iVotronic models, would not be available this time around.