If More States Start Using Ohio’s System, How Many Voters Will Be Purged?


On Monday, the Supreme Court upheld Ohio’s controversial voter purge program in the case Husted v. A. Philip Randolph Institute. Ohio removes occasional voters from the rolls if they: fail to vote in a general election; do not respond to a postcard asking them to confirm their address; then fail to vote in two more general elections. In a 5-to-4 decision, the court ruled that the purge does not violate the 1993 National Voter Registration Act.

Given the Supreme Court’s ruling, other states may adopt Ohio’s Supplemental Process as an aggressive way of removing inactive voters from the rolls.

If Ohio’s Supplemental Process were adopted nationwide, how many other infrequent voters could be removed from the voter registration rolls? We try to answer that question.

– Meet Larry Harmon

One of the plaintiffs named in Husted was Larry Harmon, who lives near Akron. Registered to vote in Ohio since 1976, Harmon voted in the 2008 general election but skipped the November elections of 2010, 2012, and 2014. Even though he didn’t move, wasn’t convicted of a felony or ruled incompetent, and didn’t die, he was purged from the rolls in 2015 because he didn’t respond to a postcard mailed to him after he missed the 2010 general election. When Mr. Harmon tried to vote in the 2015 November election, he was told he was no longer registered. He didn’t recall receiving the mailer.

– How many people are like Larry Harmon? Let’s look at North Carolina

This is not easy to answer. The United States doesn’t have a centralized database of registered voters and their turnout histories. In a few states, however, we can make an estimate.

One of those states is North Carolina, where the voter file is publicly available. Election officials in North Carolina do not eliminate registered voters’ records when they die, move away or otherwise become ineligible to vote. Instead, the state flags voters with these conditions. It also lets us see snapshots of changes to voter records, including individual vote histories.

We draw on a recent version of the North Carolina voter file and a snapshot of the Jan. 1, 2009, file, the closest we have to the 2008 election. With these two files we consider all currently active and inactive voters who lived in the same Zip code in 2017 as they did on Jan. 1, 2009. There are about 3.9 million such voters.

We apply an additional screen, keeping registrants who voted in 2008 but skipped the 2010, 2012, and 2014 general elections. Under Ohio’s system, these individuals would be purged if they failed to respond to a postcard after not voting in 2010. They are “purgeable,” so to speak.

How many of these North Carolinians are like Harmon? We find that 207,063 registrants voted in 2008 and then skipped three consecutive general elections. In 2017, North Carolina had about 6.95 million registered voters. Slightly under 3 percent are “purgeable” – if they failed to respond to a postcard asking them to confirm their address.

On the basis of data presented by Justice Stephen Breyer in his Husted dissent, roughly 80 percent of those sent a confirmation postcard in Ohio never responded. If the rate were similar in North Carolina, we still have more than 165,000 nonmoving, infrequent but still “purgeable” voters.

Unsurprisingly, these are younger than average and more vulnerable because they vote irregularly.

– How many would-be purgeable registrants in North Carolina voted in 2016?

Of the 207,063 purgeable registrants on the books in North Carolina, how many voted in 2016? The answer, we found, is 59,064. To be clear, these individuals voted in 2008, not in 2010, 2012, and 2014, and then again in 2016. To provide a sense of the figure’s magnitude, the official margin of victory in the 2016 North Carolina gubernatorial election was just 10,277 votes.

– Are purgeable voters in North Carolina more likely to be people of color or to have lower income, two claims posed by the plaintiffs in Husted?

According to our voter files, of the 207,063 purgeable registrants, 42,496 (or 22 percent) were black and 146,548 (or 73 percent) were white. These rates are similar to North Carolina’s registered voter pool. Perhaps not surprising, given Donald Trump’s victory in the state, white registrants who were purgeable turned out at a higher rate in 2016; only 14 percent of purgeable black registrants voted while 78 percent of purgeable white registrants voted. In North Carolina, at least over this eight-year period, the potential purging of inactive voters isn’t as much about race as it appears to be in Ohio.

We used census data to estimate income levels based on Zip codes. We found that purgeable infrequent voters in North Carolina were more likely to come from areas with lower incomes. Of all the registrants who didn’t move over the eight-year period and who didn’t vote after casting a ballot in 2008, 5.9 percent of those living in the Zip codes with the bottom 10 percent of household income would be purgeable; 4.6 percent of those living in the Zip codes with the highest household incomes in the state, the top 10 percent, would be purgeable.

– What’s the potential impact of the Supreme Court’s ruling?

If other states adopt Ohio’s Supplemental Process, our analysis suggests, millions of registered U.S. voters who are like Harmon will be exposed to a purge. They vote, skip several elections and vote again. Like Harmon, who for years lived and paid property taxes at the same address, many inactive registrants will never respond to a postcard asking them to confirm their residence.

We know that in Ohio in the November 2016 election, more than 7,500 registered voters who had been purged turned out to vote. Their ballots were counted because of an order handed down in October by the U.S. Court of Appeals for the 6th Circuit.

Undoubtedly, there will be more such voters if other states emulate Ohio. This will only intensify the debate over whether part-time participants in the electoral process have a right to remain on the rolls and not to vote.

Special To The Washington Post · Michael C. Herron, Daniel A. Smith

Herron is William Clinton Story Remsen 1943 Professor of Government and chair of the program in quantitative social science at Dartmouth College.

Smith (@electionsmith) is professor and chair of political science at the University of Florida.



  1. too bad they are going to extremes, it would be nice if they just purge the dead and the illegal voters first, and yes people that falsify their papers that they vote a few times. That alone would clean out many of the files, but I guess these suggestion would be too much for them to fathom. Yes people have to have true ID cards to who they are and where they live, but that would get most states crazy especially real ID cards they would call that racists. Where I live you can get ID cards for free so can’t say can’t afford to get one, most everyone has some ID cards unless you live under a bridge and don’t use society.

  2. well based on my experiences there is plenty of reason to purge, election commissions rarely do what they are supposed to ie. cancel your registration in your old district when you move to a new district. because of this my grandparents (who have since passed away) are registered in at least 2 states, and I am registered twice in one state and once in another because I moved twice and the voter registration never canceled my old registration despite my checking off the box for previous registration. and in the case of the corrupt NYC voter registration commission they wouldnt even take the necessary steps to cancel my NJ registration after I called to ask them to, instead telling me to “vote twice”.

Leave a Reply to Anonymous Cancel reply

Please enter your comment!
Please enter your name here