How a Conservative Group Persuaded a Judge to Purge Wisconsin’s Voter Rolls

There’s thin evidence that voters should be removed. The state may be forced to do it anyway.

A presumed father stands with two young children while he waits in line to vote.
Voters at the polls on Nov. 6, 2018, in Wauwatosa, Wisconsin. Darren Hauck/Getty Images

A Wisconsin judge ordered the state to remove more than 200,000 people from the voter rolls on Friday in a massive purge that would disproportionately affect minorities and Democrats. The decision rests on a dubious interpretation of Wisconsin law pushed by a conservative group that seeks to weaponize state records against left-leaning voters. If upheld, the ruling could increase Donald Trump’s chances of winning the closely divided state in 2020.

The Wisconsin purge was set in motion by the Wisconsin Institute for Law and Liberty, a right-wing organization largely funded by the Bradley Foundation, which bankrolls reactionary causes. Its lawsuit demanded that the Wisconsin Election Commission respond to a “Movers Report” produced by the Electronic Registration Information Center, a multistate partnership that shares voter registration information. The Movers Report identified 234,039 people who may have moved to an address that is not yet on their voter registration forms. In October, the Wisconsin Election Commission sent a notice to these individuals, asking them whether they still live at the address on file.

So far, so good. ERIC is a reasonably accurate tool, unlike Crosscheck, Kris Kobach’s notorious unreliable multistate voter registration program. (Crosscheck, which Kobach used to inflate reports of voter fraud, had a 99.5 percent inaccuracy rate, as well as troubling cybersecurity flaws.) ERIC is meant to help states identify people who are eligible to vote but unregistered, as well as people who have moved or died. Twenty-nine states and the District of Columbia participate. It compares voter registration information with records from state DMVs, the U.S. Postal Service, and the Social Security Administration.

But that doesn’t mean it’s perfect. No program comparing records across jurisdictions—using incomplete data subject to human error—could be. For instance, in 2017 and 2018, ERIC wrongly flagged more than 1,300 as “movers” when they had not, in fact, moved. Administrators in Maryland and Illinois also found that ERIC did not consistently identify a voter’s most recent address.

Because ERIC is useful but imperfect, the Wisconsin Elections Commission, a bipartisan body, intended to use the 2019 Movers Report as a jumping-off point. The commission planned to spend the next 12 to 24 months working with local officials to assess individuals who do not respond to the October notice. If officials make an “individualized determination” that there is “sufficient reliable information” that a voter has moved, their registration will be “deactivated.”

The Wisconsin Institute for Law and Liberty opposes this plan. In its complaint to the commission, the institute cited a state law which says: “Upon receipt of reliable information” that a voter has moved, the commission (or municipal clerk) must send them a letter saying, in short, “We think you have a new address.” If the voter does not respond “within 30 days of the date the notice is mailed,” they must be deactivated. The institute claims that ERIC’s Movers Report itself constitutes “reliable information” under the statute, while the commission’s letters triggered the 30-day deadline. (These letters said nothing about a deadline or deactivation.) Thus, the commission broke the law by declining to take action, and it must immediately deactivate anyone who did not return the card.

So far, 2,300 people (out of 234,039) have responded to the commission’s letters saying they live at the same address, while 16,500 have registered to vote at a new address. These numbers are not unusual. In a 2017 purge, the commission sent letters to 335,702 people, but only 6,153 returned their cards. (Voters often inadvertently throw away these kinds of notifications; in Ohio, two-thirds of recipients thought they were junk mail.) Another 4,917 people did not respond but later voted at their original address, indicating that ERIC wrongly flagged them.
Moreover, officials in Milwaukee, Green Bay, and Hobart found ERIC’s report unreliable enough that they refused to deactivate anyone, preserving the registration of another 38,430 residents.

Judge Paul V. Malloy of the Ozaukee County Circuit Court found the conservative firm’s arguments persuasive enough to order the immediate deactivation of the vast majority of the 234,039 voters on the 2019 list (Malloy was appointed by GOP Gov. Scott McCallum, then reelected in the heavily Republican county.) The commission intends to appeal—but it is unlikely to prevail at the Wisconsin Supreme Court, which is dominated by hard-line conservatives.

It’s no mystery what will happen if Wisconsin’s high court sides against the commission. An analysis by the Milwaukee Journal Sentinel found that residents of Milwaukee and Madison—two of Wisconsin’s relatively diverse and heavily Democratic cities—were more likely to be targeted. These individuals account for 14 percent of the state’s registered voters but received 23 percent of the commission’s notices. (As U.S. Supreme Court Justice Sonia Sotomayor has written, voter purges disproportionately affect minority and low-income communities.) Across Wisconsin, 55 percent of individuals on the list lived in cities where Hillary Clinton beat Trump in 2016. The state does have Election Day registration, which could mitigate the purge’s impact. But registrants must provide proof of residence at the polls, which many people may not bring with them—especially if they don’t realize they’ve been deactivated until they show up to cast a ballot.

Malloy’s decision uses suspect reasoning to compel a disturbing and potentially discriminatory result. The theory that ERIC’s Movers Report constitutes “reliable information” that requires the commission to launch a massive voter registration purge is, at best, a stretch. Nothing in Wisconsin law indicates that legislators coerced the commission to use ERIC data as the basis for purges. Rather, the Wisconsin Institute for Law and Liberty appears to be exploiting the statute to weaponize ERIC’s data against voters likely to vote for Democrats. Trump won Wisconsin by about 23,000 votes in 2016. If Malloy’s decision is upheld on appeal, the Wisconsin judiciary may effectively swing the state for Trump once again in 2020.