Report: No evidence of Michigan voter fraud

Michael Gerstein
The Detroit News

Lansing — A conservative economic analysis firm on Monday released a report saying a comprehensive review of Michigan’s voting data shows no widespread voter fraud in the state, a point echoed last week by Secretary of State Ruth Johnson.

The report by East Lansing-based Anderson Economic Group was prompted by claims of election tampering by Green Party presidential candidate Jill Stein as well as President Donald Trump’s unsubstantiated assertion that 3 million to 5 million people voted illegally in the 2016 election.

Trump argued that millions of illegal votes came from “those registered to vote in two states, those who are illegal and even those registered to vote who are dead,” a point that White House Press Secretary Sean Spicer said could refer to states like New York and California where the Republican businessman did not campaign.

AEG CEO Patrick Anderson said Monday his firm’s latest data analysis shows “there’s no evidence” of any systematic manipulation of Michigan votes.

“Right now we’re looking and we’re not finding anything,” Anderson said.

The 2016 election results were consistent with patterns in the 2012 results, he said, adding that any hacking would have resulted in a glaring deviation from trends in past elections.

Trump won Michigan by 10,704 votes over Democrat Hillary Clinton. The margin tightened slightly — by 103 votes, unofficially — in the partial state recount of almost 40 percent of Michigan’s precincts. The recount was voided after state and federal court rulings found that Stein lacked the legal standing to request the recount because she had no statistical chance of winning the election.

AEG analysts used “sophisticated statistical tools” to show a dearth of evidence that Russian hackers tampered with results to swing the election in Trump’s favor, or that illegal voters cost Trump the popular vote to Clinton, Anderson said.

FBI Director James Comey and National Intelligence Director James Clapper Jr. said in December that they agreed with a Central Intelligence Agency assertion that Russia had tried to help Trump win the election but had not changed any votes. Russia has denied any attempted interference and rejected U.S. findings that it hacked into the emails of the Democratic National Committee and Clinton’s campaign chairman.

Trump won the state largely because 12 Michigan counties that voted for former President Barack Obama in 2012 instead turned out for Trump in 2016. Clinton won only eight counties including Oakland, Washtenaw and Wayne.

The only three counties where Trump did not pick up more voters than other Republican candidates in the past were Wayne, Ingham and Washtenaw — all Democrat strongholds, the AEG report said.

Anderson acknowledged the study has limitations. It would not reveal that an individual machine had been tampered with, or if several boxes of votes went missing, he said.

AEG even explored the “rational villain hypothesis,” in which analysts make a hypothetical projection about the smallest possible number of votes that would need to be tampered with to swing an election without it being obvious and found that even in that scenario, any manipulation broad enough to change the results would have been obvious in the statistics.

Anderson said the firm explored a “rational villain hypothesis — the theory that a 19 percent swing of votes in five counties, the smallest possible number needed to swing the election to Trump — could have happened without detection. But it still would require changing about 490,000 votes, a glaring shift that was not evident in the data, he said.

Anderson said he hopes the report puts to rest fears of tampering in Michigan’s results and that such statistical analyses become routine in the post-election process.

After a partial recount, a preliminary report from AEG and now a second report, he said, “we now have the most thoroughly vetted presidential election in Michigan history.”