Posted: July 1, 2008
Registration Bottlenecks: Will They Be a Problem in ’08?
Due to the history-making candidacies of Hillary Clinton and Barack Obama, many states experienced unprecedented levels of voter turnout during the Democratic primaries. It was a dream come true for many social scientists and others who have long bemoaned our nation’s typically anemic levels of political participation. Indeed, one can only hope that participation is even higher in the general election and, with candidates as distinct as McCain and Obama, maybe it will be. Still, before we go patting ourselves on the back for getting people involved, we have to consider whether our elections infrastructure is really ready for the level of participation that could occur. This consideration should extend not only to actual election-day operations, but also to pre-election operations, especially voter registration. Specifically, we should consider whether our registration systems are efficient enough to deal with significant registration bottlenecks while still leaving election administrators time to give their other responsibilities sufficient attention. I do not know whether our system is efficient enough, but here are a few considerations.
The first is the regulation of third-party voter registration groups. These regulations are generally aimed at making election administrators' lives easier. They do this primarily by requiring registration groups to weed out bogus registrations and prohibiting registration groups from “saving up” completed applications and dropping them all off on the last day for registration. Of course, in making administrators' lives easier, these regulations have made the lives of voter registration workers harder (they also have triggered a number of lawsuits—see here, here, and here). Administrators claim that these regulations are necessary to ensure the smooth functioning of the system, and caution that removing them could overload the system and prevent administrators from performing other crucial duties. Of course, this argument cannot be taken at face value, but insofar as it is true, states may experience serious disruption where these requirements are enjoined or where they were never passed into law in the first place. States that do have such requirements should take a closer look at them to make sure they strike a reasonable balance between efficiency and the increased access to voting that third-party groups can give us when their hands are not tied by unnecessary regulations.
Another issue related to bottlenecks is matching of social security and/or driver’s license numbers contained on voter registration applications. There is significant confusion in this area that could slow the system down. Florida and other states require matching these numbers against Social Security Administration or motor vehicle databases for registration to be effective. Other states either do not require matching or have not implemented a matching policy at all, leaving the door open to inconsistencies in local practice. Some states are still building their matching programs and will likely have difficulty implementing them as November approaches. Finally, administrators in many states will likely be sued (see here and here), either for rejecting unmatched applicants or, perhaps worse, for having no rules at all and allowing local administrators to do whatever they want. All in all, it sounds like a bad time for 20,000 registration applications, 10% of which may be difficult to match, to show up on an administrator’s doorstep on the last day for registration.
A final issue concerns defective registrations. Can voters correct those registrations even after the registration deadline? If they can, it expands access, but it also puts an additional burden on the system, particularly if a large number of last-minute applications are in need of correction. Colorado and Washington permit voters to correct after the deadline, while Indiana, North Carolina, Florida, Nevada and Oregon allow certain types of corrections but not others. It is unclear whether officials in New Mexico, Wisconsin, Michigan, Ohio, Pennsylvania or New Hampshire will permit correction. Florida has already been sued for not permitting most types of corrections after the deadline, and more suits can be expected. Problems in this area are especially likely in places where turnout will increase the most—think heavily African-American jurisdictions targeted by voter registration drives—or jurisdictions in Nevada and Arizona, where the population has swelled in recent years.
If these issues do not blow over, we can expect two consequences. The first consequence is that some people who are entitled to be registered will not get registered. This could happen, for instance, if administrators become too busy and fail to give people the opportunity to correct flawed registration applications. The second consequence is that, even if registration problems are ultimately worked out, the effort expended to accomplish this could reduce the effort expended to make operations go smoothly on election day. It could, for instance, detract from the rigorous pre-election testing of DRE voting machines that really should occur. It could also lead to sloppy poll worker training or inadequate efforts to organize poll workers and make sure a sufficient number show up to serve at each precinct. States should reexamine their registration programs with a view to reducing such risks.