Opinion: Daily Journal

Don’t Expect Big Voting Change

RALEIGH — While the many are trying desperately to convince North Carolinians otherwise, the passage of House Bill 589 is unlikely to have a big effect on voter participation in 2014, 2016, and beyond.

As you have probably heard by now, the bill requires North Carolina voters to show a government-issued photo ID before casting a ballot. It also changes the early-voting schedule, eliminates same-day registration during early voting, and ends the current practice of allowing voters who go to the wrong precinct on Election Day to cast a provisional ballot there. Instead, those voters will have to go to their actual precinct to cast a ballot.

The bill does a number of other things, too. It eliminates straight-party-ticket voting and sign-by-computer ballots, for example. It also changes the rules for casting absentee ballots by mail. But when it comes to the issue voter participation, the rules attracting the greatest attention involve voter ID, early voting, same-day registration, and out-of-precinct voting.

Why do I predict that these changes won’t make much difference in upcoming elections? Because I let empirical evidence and the letter of the law, not partisan intentions or conspiracy theories, be my guide.

There is nothing new about voter identification. Most American states and industrialized countries have long had some kind of voter ID rule. And several states have held recent elections with photo-ID laws in place. Using this natural experiment, academic researchers have completed several studies of the effects of ID requirements on turnout. Most have found little to no effect.

A 2011 paper in State Politics and Policy Quarterly, for example, used data from the 2000, 2004, and 2008 presidential elections to estimate the effects of various state election reforms. The authors found that voter ID resulted in slightly higher voter turnout. The following year, the same journal published a study of the 2008 presidential election in Georgia, which imposed its first photo-ID requirement that year. The authors found that photo ID did reduce turnout slightly, by four-tenths of one percent, but the effect was equally distributed by race and thus had no reasonable likelihood of affecting electoral outcomes.

Imagine that North Carolina had required photo ID for the 2012 cycle. Using the Georgia study as a pessimistic scenario, that would mean that out of 4.5 million North Carolinians who voted in 2012, about 18,000 fewer would have voted under a photo ID rule. Because the effect wouldn’t have been disproportionate, it wouldn’t have tipped the electoral balance. Remember, this is an upper-bound estimate. Other studies would yield a far-smaller projection. Moreover, that pessimistic estimate would be for the first year of implementation. There is a learning curve to any change in election law. Many voters surprised by the photo-ID requirement in 2012 would then renew their driver’s license, obtain a special ID card, or otherwise comply with the law in time for the next election.

Both sides need to take such realistic projections seriously. Panicked liberals muttering about “massive voter suppression” and excitable conservatives warning of “rampant voter fraud” surely have in mind an effect greater than a few thousand votes out of 4.5 million.

The bill’s other provisions are likely to produce similarly underwhelming effects. Early voting doesn’t appear to boost voter participation at all — while it may be convenient for some people, studies show that virtually all of them would have voted anyway. Besides, the 2013 legislation doesn’t actually curtail early voting. It shortens the early-voting period from 17 days to 10 days, but simultaneously requires more sites open for longer hours so that the opportunity to vote early in North Carolina will be little affected.

As for folks showing up at the wrong precinct and casting provisional ballots, we also seem to be talking about just a few thousand votes statewide. Unless the affected voters arrive during the last minutes of Election Day, most will simply relocate to the correct precinct to vote — which will have the added benefit of allowing them to vote in all the races they are entitled to, rather than just the ones listed on the ballots of the wrong precinct.

The one provision that probably will reduce voter turnout measurably is the end to same-day registration during early voting. This is an exceedingly rare policy, however. Only Ohio and (forthcoming in 2016) Maryland have it. The vast majority of states, include deep-blue ones, have concluded that any benefits of allowing same-day registration during early voting are offset by the administrative headache and the potential for phony registrations, which have apparently occurred in every election North Carolina has held under the policy. I don’t have a beef against same-day registration myself, but again the effects either way aren’t large enough to make a practical difference in most elections.

What about college students? Aren’t their votes being suppressed by denying them the ability to use their student IDs? This is a silly argument. The vast majority of college students in North Carolina have a North Carolina drivers’ license (or more than one, wink wink, nudge nudge, say no more). As for out-of-state students, if they reside and drive a car in North Carolina for more than a few weeks, they are already required to obtain North Carolina driver licenses under current law. For the few non-driving students who need photo IDs to vote, the process is apparently so simple that even an anthropology major can do it.

And what about the small number of elderly minorities whose Jim Crow-era birth certificates don’t agree with their voting records or their current IDs? The solution is to fix the discrepancies, which would require little time or effort. Wouldn’t it be better for everyone to have valid voter records, birth records, and photo IDs? These come in handy for all sorts of non-electoral uses. Why spend months or years pursuing litigation against election laws rather than spend a few minutes getting the underlying problem solved?

I know the answer to the question. I just don’t like it very much.

Hood is president of the John Locke Foundation.