By most accounts, the 2011 and 2012 gubernatorial and Senate recall elections were a complete disaster for Wisconsin Democrats.
Gov. Scott Walker’s historic victory boosted his fundraising and re-election prospects. The recall petition became a litmus test for party loyalty. And though Democrats recaptured the Senate majority in June 2012, they lost it five months later and have been shut out of state government ever since.
But some Democrats see a silver lining in the recalls that has gone mostly unnoticed until now: The unearthing of key evidence in a potentially landmark legislative redistricting case now before the U.S. Supreme Court.
Lawyers representing the plaintiffs in the case, Gill v. Whitford, say the short-lived majority allowed Democrats to force the Legislature's private lawyers to cough up documents, leading to a court-ordered release of computer hard drives with deleted spreadsheets.
The story, pieced together from court documents and new interviews, of how an otherwise meaningless transfer of power could potentially overturn the centuries-old practice of political gerrymandering in the United States is a reminder that in politics even the little victories matter.
The 'Tale of the Tape'
Wisconsin's latest redistricting court battle began with the 2010 Republican wave that put the governor's office, Assembly and Senate under one-party control for the first time during a decennial redistricting process since 1951.
During the 1991 and 2001 redistricting process, a federal court had to redraw the maps after Republican and Democratic lawmakers couldn't reach agreement.
In 2011 Republicans tasked two staffers and a GOP redistricting expert with secretly drafting new maps. They worked at the Madison law office of Michael Best & Friedrich. Republican senators had to sign nondisclosure pledges to view drafts of their own districts.
The AP scrutinized the outcomes of all 435 U.S. House races and about 4,700 state House and Assembly seats up for election last year using a new statistical method of calculating partisan advantage designed to detect potential political gerrymandering.
Based on an analysis of voting trends, the drafters determined the existing maps would yield 49 Republican Assembly seats and 50 Democratic seats in the 2012 election. Using a computer program, drafters tinkered with the district boundaries to give Republicans a projected 54-45 advantage.
After consulting with Republican leadership, the drafters massaged the maps further to the point that, according to one spreadsheet, "59 Assembly seats are 50 percent or better" for Republicans. The spreadsheet labeled the data "Tale of the Tape," a reference to the measurement of each boxer's arm length before a match.
A University of Oklahoma professor working with the drafters conducted a statistical analysis showing Democrats would need to win 54 percent of the statewide vote to capture an Assembly majority.
"The maps we pass will determine who’s here 10 years from now," Senate staffer Tad Ottman told the Republican caucus, according to prepared notes. "We have an opportunity and an obligation to draw these maps that Republicans haven’t had in decades."
The maps were introduced to the public on July 11, 2011, a public hearing was held July 13 and the Assembly and Senate voted on them a week later.
A case against partisan gerrymandering
The three-judge panel that heard the Whitford case last year weighed the evidence and ordered the maps redrawn. The state has appealed the decision to the U.S. Supreme Court, arguing the process was lawful. The Supreme Court stayed the lower court order and agreed to hear oral arguments as early as this fall.
A spokesman for Attorney General Brad Schimel declined to comment for this story. Fitzgerald declined to comment on an ongoing legal matter. A spokeswoman for Michael Best & Friedrich didn't respond to a request for comment.
Lawyers for the Whitford plaintiffs say were it not for evidence obtained from an earlier redistricting case, Baldus v. Brennan, many of the details presented to the three judges in the Whitford trial would not have come to light. The plaintiffs in the Baldus case initially made a partisan gerrymandering claim, but abandoned the argument during the 2012 trial.
By contrast, the Whitford plaintiffs devised a three-pronged standard for partisan gerrymandering: Proving discriminatory intent, demonstrating a discriminatory effect, and finding no other justification for how the maps were drawn.
To prove the discriminatory effect, the plaintiffs measured the election results based on what is known as the "efficiency gap," which seeks to calculate how many votes for a given party are "wasted" because its voters are "packed" into certain safe districts or "cracked," that is, placed into districts where they still can't muster enough support for their candidate to win.
The Whitford plaintiffs argued the Wisconsin legislative districts were the most gerrymandered in the past 40 years, with 13 percent of votes wasted in 2012 and 10 percent wasted in 2014. Based on an analysis of 786 legislative elections across the country, they argued a gap of more than 7 percent should be deemed unconstitutional.
To prove discriminatory intent, the Whitford case relied heavily on documents uncovered in the Baldus case.
Defendants resist release of records
Milwaukee civil rights lawyer Peter Earle and Madison commercial litigator Doug Poland didn't know each other before joining forces to represent plaintiffs in the Baldus case, though they had previously been on opposite sides of a lead-based paint liability case.
The Baldus lawsuit was filed by a group of Democrats in June 2011 as the secretive Republican redistricting process was nearing completion. Later that year it was combined with a case filed by immigrant rights group Voces de la Frontera.
The plaintiffs sought a host of documents from the Legislature only to be resisted by the defense lawyers who argued the documents were protected by attorney-client privilege. The three-judge panel hearing the case ordered three times the documents be released and fined Michael Best & Friedrich $17,000 for dragging out the issue.
Among the documents were the secrecy pledges Republican senators signed to review their district maps, strict rules for anyone entering the map room and talking points about the significance of the redrawn maps. There was also a print-out of a spreadsheet showing the partisan effect of the final map, but Earle said the document only told a partial story.
"You have a piece of the jigsaw puzzle in your hand, but imagine you can’t see clearly the boundaries of it and where it fits with the other pieces," Earle said. "You can’t create the whole picture."
Poland said the plaintiffs in Baldus dropped the partisanship argument because the trial was limited to two days, and after a lengthy first day they had to decide which claims had the strongest chance of succeeding.
In its March 2012 ruling the court only ordered changes to the boundaries between two Democratic Assembly districts in Milwaukee so as not to discriminate against Hispanic voters in violation of the Voting Rights Act.
Recall victory creates opening
By June 2012 Democrats held a 17-16 majority in the Wisconsin Senate after a series of 13 recall elections triggered by Act 10, the law curtailing public sector union power in the state.
By then the Legislative session had ended, so there was little Democrats could do. Moreover, it was apparent from the redrawn legislative boundaries that it would be unlikely for the Democrats to retain their majority.
But with the power of the majority, Democrats were able to compel Michael Best & Friedrich to turn over the entire record of client files to the Democrats. The firm refused to give the records to Democrats when they were in the minority, saying it answered to the majority leader, said Sen. Mark Miller, D-Monona.
Miller, who became majority leader as a result of the recalls, had his office scan and post all of the documents on a public website, which allowed Earle and Poland to compare them with their trove of documents. They found 34 missing documents the defendants should have turned over before the trial.
Miller said obtaining and posting the documents "were absolutely the most significant actions" taken when Democrats briefly held power.
"I wasn’t concerned anything was missing," Miller said. "I firmly believed these documents should be part of the public record, particularly because of the extraordinary effort by Republicans to keep them secret."
Earle and Poland went back to the court in August 2012 to demand the entire hard drives used by the Republican drafters be turned over for an independent forensic examination. The court granted their request stating "that some form of 'fraud, misrepresentation, or misconduct' likely occurred."
Forensic investigator Mark Lanterman, a former U.S. Secret Service official, testified that his review of the hard drives found evidence that hundreds of thousands of files had been deleted — many in the week before the files were turned over to Miller's office.
The Baldus case ended in May 2013 when Michael Best & Friedrich settled out of court with the plaintiffs over issues related to the deleted files. The settlement terms are not public.
Missing spreadsheets recovered
In the months after the Baldus case concluded, Earle began to meet regularly with a group of Democrats, including UW-Madison Law School professor Bill Whitford.
The group's goal was to develop a partisan gerrymandering case that would satisfy the U.S. Supreme Court. They reached out to Nick Stephanopoulos at the University of Chicago, who had developed the efficiency gap standard. They asked UW-Madison political science professor Ken Mayer to apply the formula to Wisconsin's election results. They filed their lawsuit in July 2015 and in December of that year a panel of three federal judges allowed the case to proceed.
At that point Earle hadn't gone back to Lanterman yet to look for more deleted evidence on the hard drives. They already had a few examples of spreadsheets from April and June, but no documents in between when the drafters were doing much of their work.
The legal team asked Lanterman to conduct a limited search of the hard drives, which turned up several deleted spreadsheets they hadn't previously seen. They included the one with the column labeled "tale of the tape."
The files also included data showing when each of the maps had been created. Using that evidence they were able to reconstruct a timeline showing that with each revision, the Republicans were focused on fortifying their majority.
"Their reach was long enough to make their advantage over the entire decennial cycle," Earle said. "What really swung the case in our favor, we not only showed there was an intent in how the maps were drawn, but there was an intent to create an advantage that would last the entire ten-year cycle, to retain control and maintain the Republicans in power whether they controlled a majority or not."
The U.S. Supreme Court announced last month it plans to hear the case in the term that begins this fall. A decision could come sometime early next year.