While Americans this year were fortunate to avoid the drawn-out legal battles of the preceding presidential election, the close margins in some key states and concerns over inconsistent election standards illustrate the continued risk of having election processes that vary from state and state, and often from county to county.
With future close elections likely, our democratic process is particularly vulnerable to voter fraud (search) and abuse that we are not yet doing enough to stop.
Voter fraud has a long and storied history in the United States. However, the opportunities to commit fraud expanded substantially in the 1990s as a result of the federal motor-voter law that allows driver's license applicants to self-affirm their eligibility to vote with no more than a check mark and a signature.
This lack of safeguards has resulted in extremely inaccurate voter rolls in many states. Individuals who are deceased or who have moved are in many cases simply kept on the rolls, providing an opportunity for fraudsters to exploit simply by assuming different identities.
Furthermore, states do not share or coordinate their voter rolls. Therefore, there’s no way for a state to know that the John Smith who showed up at an Orlando polling place is the same John Smith who voted with an absentee ballot in his home state of Pennsylvania. Allowing voters to register on the day of the election, as some states do, only exacerbates this problem.
Recognizing that there were serious issues to address after the last election, Congress passed the Help America Vote Act (search) that mandated the creation of the Election Assistance Commission (search) to conduct public studies that would examine the myriad weaknesses of the federal electoral system and develop best practices and standards for voting systems to guarantee the integrity of elections.
Not surprisingly, the Commission made the determination that verifying a voter’s identity was a crucial component of its overall mandate to restore trust in the way that elections are conducted and administered in the United States.
In many ways, the challenge we now face is similar to one resolved years ago by financial institutions, credit issuers and insurance companies. In order to verify the identity of individuals opening new accounts, these businesses worked to develop systems to perform what’s known as “information-based identity authentication.” (search)
These systems perform an independent assessment of what a person represents about his identity, based on an analysis of available information. The applicant provides information that is checked against proven information tools to ascertain if it is real (not fictitious) and if combinations belong together (e.g., linked names and addresses). Using a well-tested analytical model based on historical data, the system can then provide a score that indicates the likelihood that the person is who he claims to be.
Automated identity authentication at the time of voter registration would be a good place to start. After a registration application is submitted online or by a grassroots organization, states could use an information-based identity authentication system to ensure that the applicant is legitimate and screen out false or stolen identities.
In addition, the identity authentication process could easily be expanded by individual states on a case-by-case basis to screen for other eligibility criteria, such as age, citizenship and residency requirements. Combining advanced analytical processes and tools along with policy-based, information-driven technology, it is possible to put in place an identity management system that provides authentication, security and enhanced privacy protections.
Putting such a system in place would allow for the standardization of state voter lists, eliminating many opportunities for fraud and abuse. The next step would be to develop a system for states to share information on voters, ensuring that principle of “one person, one vote” is not undermined by would-be fraudsters.
States already possess much of the information that would be needed for such a system, in the form of motor vehicle databases. However, these systems could only be truly effective when combined with the kind of information and tools that power predictive models used in the insurance or mobile phone industries to verify the identity of applicants. There are numerous examples of this technology already in use in a way that safeguards society’s interests, while protecting the privacy of individual citizens.
Ultimately, the principle of democratic governance is at stake in voter fraud and abuse. While information-based identity authentication will not provide a magic bullet to solve the problem of voter fraud, it does provide one vital component of a comprehensive system to ensure that elections are carried out in a fair and accurate manner.
Norman A. Willox Jr. is the chief officer for privacy, industry and regulatory affairs at LexisNexis.