Should Protasiewicz Recuse From Redistricting Cases?

Milwaukee County Judge Janet Protasiewicz. Photo courtesy of Janet Protasiewicz’ campaign

Recently, Wisconsin Assembly Speaker Robin Vos threatened newly elected Justice Janet Protasiewicz with impeachment if she doesn’t recuse herself from cases challenging Wisconsin’s district maps. He based this on comments that Protasiewicz made during her recent election campaign. Vos claimed that:

You cannot have a judge who said, you know, the maps are rigged because she bought into the argument that that’s why we’re winning elections, not the quality of our candidates, and then she sits on that trial acting like she’s gonna listen and hear both sides fairly − that just can’t happen.

It is unclear whether Vos really believes that the reason Republicans win most of legislative districts is the quality of the candidates. To test the candidate quality claim, it is useful to examine how candidates for state-wide offices faired in each legislative districts.

There are two advantages of this approach rather than using votes for the candidate for the district itself. The first is it solves the missing candidate problem. In the 2022 election, 26 Assembly districts were missing a candidate from one or the other of the two main parties. Using state-wide data means that every district is counted. Secondly, since the same two candidates are running in every district, it removes the possibility that variations in candidate quality have determined the outcome.

The graph below shows the results of the 2022 race for the US Senate. In that race, Mandela Barnes received 1,310,467 votes to Ron Johnson’s 1,337,185, a difference of 26,718 votes or one percentage point. Despite the closeness of the statewide vote, only 35 of the 99 Assembly districts supported Barnes compared to 64 won by Johnson.

The next graph shows the percentage of the vote for each candidate, on the left, compared to the percentage of the districts, on the right.

Voters versus Seats–US Senator

The next graph shows the results in the race for governor. Tony Evers won 1,358,774 votes, compared to Tim Michels’ 1,268,535, a difference of 90,239 votes or 3.4 percentage points. Yet Evers won only 39 Assembly districts compared to Michels’ 60. Despite winning the popular vote, Evers lost the majority of Assembly districts.

Voters versus Seats–Governor

How well does the district vote for statewide candidates predict the outcome for the district candidates? The scatter plot below compares the percentage of the vote that US Senator Ron Johnson won in each Assembly district (on the vertical) to the percentage going to the Republican candidate in the same district (on the horizontal). The dots on the left of the graph represent the ten districts in which there was no Republican candidate; the 16 on the right had no Democratic candidate.

In Assembly districts where Johnson did well, the Republican candidate for the state Assembly also did well. Where Johnson did poorly, so did the party’s candidate. The coefficient of determination, R2, measures the strength of this relationship. It is very high, .98, interpreted as saying that 98% of the variation in the Republican vote for Assembly reflects the variation in the district vote for US senator. That suggests that variations in candidate quality would contribute, at most, 2%.

Comparing Republican Assembly Vote to Vote for Johnson

If saying that the Wisconsin legislative districts are “rigged” is taken as shorthand for “rigged to favor one party over the other,” Protasiewicz was factually accurate. Her statement is one of fact, not opinion. The question for the Wisconsin Supreme Court is not whether the districts are rigged. They are. The issue the court must consider is whether such rigging is contrary to the Wisconsin Constitution. So far as I know, Protasiewicz has not addressed this.

This scatter plot also suggests the strategy used by the Republican legislators’ map makers. It was an extreme “cracking and packing” strategy. Boundaries were designed to stuff Democratic voters into heavily Democratic districts (packing), thus shrinking the number of Democratic districts. This allowed the creation of new districts that were majority Republican.

Those insisting that Protasiewicz recuse herself point to her open dislike of the Wisconsin gerrymander in statements like this:

They do not reflect people in this state. I don’t think you could sell any reasonable person that the maps are fair. … I can’t tell you what I would do on a particular case, but I can tell you my values, and the maps are wrong.

It turns out that in calling the gerrymandered districts “rigged” Protasiewicz struck a chord with Wisconsin voters. This is reflected in the size of her victory. People recognize that allowing maps to be prepared by legislators is an invitation to conflict of interest.

This sentiment is reflected in polling. In three polls of Wisconsin registered voters between 2019 and 2021, the Marquette poll included a question on redistricting:

As you may know, redistricting involves redrawing maps of legislative districts for the state legislature and for Congress. In your opinion, who should be responsible for drawing legislative districts — the state legislature and governor or a non-partisan commission?

The next graph shows the responses by combining all three polls. The results make it clear that an overwhelming majority of Wisconsin voters, including Republicans, oppose redistricting by the legislature and governor and favor a non-partisan commission (yellow bar on graph).

Marquette Poll: Who should redistrict?

Vos is quoted as asserting that Protasiewicz would violate her oath of office if she doesn’t recuse from redistricting cases and therefore lawmakers might consider impeachment. It is unclear to me how her observation that the districts are rigged would violate her oath or the or the constitution.

Here is what the Wisconsin Oath of Office of Justice or Judge says:

I, the undersigned, who have been elected to the office of Supreme Court Justice but have not yet entered upon the duties thereof, do solemnly swear that I will support the Constitution of the United States and the Constitution of the state of Wisconsin; that I will administer justice without respect to persons and will faithfully and impartially discharge the duties of said office to the best of my ability. So help me God.

Section 757.19 (g) of Wisconsin statutes say a judge should recuse when he or she determines that  he or she cannot act in an impartial manner. In condemning the Wisconsin maps as bad policy, Protasiewicz was careful to distinguish the policy issue from the legal issue. She was careful to leave herself the room to conclude that while the gerrymander was bad policy, it was not illegal.

The recent lawsuit against the Wisconsin maps argues that partisan gerrymander violates several provisions of the Wisconsin Constitution. In addition, the maps contain an obvious and surprising violation that appears to have no connection to the gerrymander. Article 4 Section 4 of the Wisconsin Constitution states that districts must “consist of contiguous territory.” Despite this admonition, there are many examples of parts of districts isolated from the main body of district and completely surrounded by another district.

These islands are particularly common in the Madison area, as shown in the map below.

District map

The only justice to dissent from the court’s decision to hear the challenge to the Wisconsin gerrymander is Rebecca Bradley. Here are her final two sentences:

Entertaining these claims makes a mockery of our justice system, degrades this court as an institution, and showcases that justice is now for sale in Wisconsin. “Rigged” is indeed an apt description—for this case.

Applying the standards that Protasiewicz’s critics advocate, shouldn’t she recuse? After all, she has clearly made up her mind about the merits of the case.