Experts: Florida redistricting faces legal risks without racial analyses

The Florida Legislature has not published any analyses showing how it has determined what districts deserve protections for racial minority groups, how it determined what percentages are necessary for those groups to elect candidates of their choice or why it chose to boost or shrink percentages in certain districts.

In fact, Florida House staff indicated they may never have even done any analysis beyond looking at demographics, turnout, voter registration statistics and election outcomes.

Last week, Florida House staff revealed their first four redistricting proposals. In the first two workshops — one dedicated to the two congressional maps and one dedicated to the two state house maps — Democrats asked most of the questions and focused primarily on city and county splits or communities of interest.

The Tributary covered the meetings live, and in response to The Tributary’s coverage, Rep. Dan Daley, a Coral Springs Democrat, asked to see the staff’s functional analysis used to determine which districts were protected and how they ensured voters from racial minority groups could elect candidates of their choice.

Staff Director Leda Kelly said that she didn’t want to “overwhelm you with more data and more packets” in the meeting, so she left the analysis out, but she said that the representatives could see all the data staff used for the analysis in the redistricting app.

But the data available in the app can’t say which candidates racial minority groups prefer, whether voting is racially polarized, whether multiple racial minority groups vote together as a coalition or what percentages are necessary to ensure racial minority groups can elect the candidates of their choice.

The Florida Supreme Court said last decade that this type of evaluation “requires an inquiry into whether a district is likely to perform for minority candidates of choice. This has been termed a ‘functional analysis,’ requiring consideration not only of the minority population in the districts, or even the minority voting-age population in those districts, but of political data and how a minority population group has voted in the past.”

To determine this, you’d need to know turnout and results at a voting precinct level, and then you’d perform a statistical analysis to determine how different racial groups vote.

Despite records requests for the underlying data that would make these analyses possible, the Florida Legislature has not provided the data, nor has its staff provided any copies of analyses if they’ve done any.

“If it’s true there’s been no statistical analysis to determine racial voting patterns, which will reveal the minority population needed to elect a candidate of choice, that will be highly problematic, and the [U.S.] Supreme Court may take a hard look at Florida’s redistricting process if they did not do that,” said Michael McDonald, a University of Florida redistricting expert. “I’ve wondered, is this minority population sufficient to elect a candidate of choice? They’re drawing coalition groups in a lot of places. Is there cohesion between the minority groups? You cannot divine the answers to those questions by looking at the [redistricting app’s] reports.

“This really exposes them to some serious, serious legal exposure if they really haven’t done any sort of statistical analysis.”

The Florida House and Florida Senate have signed contracts with at least five law firms to represent them during redistricting.

The Florida Senate has published what it calls a “functional analysis” for its proposed maps, but it too only shows voter turnout, election results and racial demographics for each proposed district.

It’s not enough to know, for example, that Black voters make up a majority of the voting-age population, courts have ruled. It’s even not enough to know that Black voters make up a majority of Democratic primary voters in a Democratic-leaning district.

To use an example that requires a good bit of math: say about 70 percent of Black voters tend to vote for certain preferred Democratic primary candidates, while about 90 percent of white voters tend to vote for different preferred Democratic primary candidates. If Black voters made up 51 percent of Democratic primary voters and white voters made up the other 49 percent, white voters would almost always elect their candidate of choice. In that district, you’d need a higher share of Black voters.

Yet if Black voters and white voters actually voted for the same candidates in higher percentages — say 35 percent of white voters voted for certain candidates that 80 percent of Black voters selected — then you could draw a district with a lower share of Black voters and they’d still elect their candidates of choice.

“A performance analysis would have to use primary data to see if an African-American Democrat could get through a primary with a white Democrat,” said Dan Smith, another UF redistricting expert. “It would be nice to see the racial bloc voting analyses they did for each one of these districts.”

In a federal court decision last year, the 11th Circuit Court of Appeals overturned a Georgia county’s proposed changes to its School Board using “generally accepted statistical techniques — homogenous precinct analysis, ecological regression, and ecological inference.”

None of those types of analysis are possible with the data provided in the Legislature’s redistricting app, nor have they been performed in the staff functional analyses.

Rep. Daley’s question was one of only a few that Democrats asked about preserving the rights of racial minorities, even though it’s a key component of the constitutional protections approved a decade ago by Florida voters in the Fair Districts Amendments.

The amendments bar “retrogression,” or reducing the voting power of racial minorities, yet one of the House-proposed congressional maps would see Black voters hold less of a say in who would represent one of the state’s districts, even as Florida gained an extra seat.

In that proposal, an Orlando-area district where Black voters make up about half of Democratic primary voters would have Black voters fall to 38 percent of primary voters.

During the congressional redistricting subcommittee, Democrats didn’t ask a single question about the district or how staff conducted its analyses.

Staffers have said some districts are protected, and they’ve said they drew the districts to ensure minority racial groups would have enough voting power to elect candidates of their choice.

But Democrats didn’t ask follow-up questions about how staff determined a district was protected. They also never asked how staff knew which candidates a minority racial group preferred, whether the group voted cohesively as a bloc or how many voters from that minority group were needed in order to elect candidates of their choice.

In fact, it seems staff at times chose to ensure Black voters made up a majority in certain districts. Under the House proposals, four state house districts are right at the line of 50 percent Black voting-age population, with the next closest district at less than 44 percent Black, suggesting the staff used a 50-percent quota.

A North Carolina redistricting plan was overturned after a court ruled that “the legislators had no basis — let alone a strong basis — to believe that an inflexible racial floor of 50 percent plus one [Black population] was necessary” to draw Black-protected districts.

In one Florida case last decade, a Tallahassee court ruled that, “If a minority population can elect candidates of their choice without a majority voting-age population, … there is no need to maintain or create a majority-minority district.”

In another Florida ruling last decade, the Florida Supreme Court struck down a Jacksonville-to-Orlando congressional district because there had “been no showing that it was legally necessary to create a majority-minority district.”

Yet the Florida redistricting committees again did not provide any evidence that it’s legally necessary to draw nine Black-majority state house districts.

In Jacksonville, for example, Black voters were able to comfortably elect candidates of their choice in House District 13 without a majority of the voting-age population, yet staff increased the Black population in the district to get it to 50 percent so that 79 percent of Democratic primary voters are Black.

The Florida House staff offered no explanation as to why it increased the population, except to say that they drew the districts after performing a “functional analysis to ensure that the minority groups in those districts can elect candidates of their choice.”

It’s unclear what functional analysis would’ve been done considering Duval County is also home to a 43-percent Black state Senate district that has always selected Black voters’ candidate of choice.

The same staff also said that a Gainesville-area house district where 29 percent of the voting-age population is Black was “a protected Black district”, and it was drawn so that Black voters could elect the candidates of their choice, even though Black voters made up just 43 or 45 percent of Democratic primary voters in the district, depending on the plan.

Yet the Florida Senate didn’t consider Senate District 38 as a protected Black district. It is 28 percent Black and saw Black voters make up about 40 percent of the Democratic primary last year while white voters made up 30 percent. No one in the Senate asked why that district, which is not currently represented by a Black Democrat but might still offer Black voters the ability to elect the candidate of their choice, wasn’t protected.

If the Florida Legislature is judging whether a district deserves legal protection based on whether it’s currently represented by a Black or Hispanic legislator, it hasn’t said so.

On Friday, while Democrats asked some questions about protecting racial minorities, they didn’t press for more detailed explanations.

In one case, Rep. Marie Woodson, a Hollywood Democrat, asked if the Caribbean population in South Florida deserved protections under Florida law.

Rep. Cord Byrd, a Neptune Beach Republican leading the committee, said it wasn’t protected because it was a community of interest and not a racial group.

Yet the actual law protects both racial minorities and language minorities — which could include Haitian Creole — and last decade, the Florida Legislature actually treated the Haitian population in South Florida as protected. The Legislature even provided percentages of the Haitian population in each district, something that redistricting staff has chosen not to do this time.

After The Tributary pointed that out, Daley brought it up during the meeting, but Byrd and the staff didn’t explain why they believed the population had lost its protections under the law.

(Byrd has not responded to requests for comment.)

When Rep. Kevin Chambliss, a Homestead Democrat, asked about how the staff pulled Black communities into different districts, staff director Kelly gave the same answer staff repeatedly gave, saying districts were drawn so that they “performed as required under Tier One.”

Neither Chambliss nor the other Democrats asked what that meant or how staff determined what performance was required under the Constitution’s anti-retrogression standards.

House Democrats spent the majority of their two meetings focusing on ancillary issues that courts have deemed less important — asking repeatedly why districts cross waterways, asking why a county or city was split, or asking why a community of interest wasn’t kept together.

The Legislature can consider and try to preserve communities of interest in redistricting, the Florida Supreme Court ruled, but only as a tertiary concern after meeting Tier One and Tier Two protected issues.

The court ruled that Tier One protections include preventing plans that intentionally favor a political party or incumbent, protecting minority voters’ ability to elect candidates of their choice and keeping districts intact.

Reducing splits and honoring geographic boundaries, which can include waterways, are protected in Tier Two.

The individual Democrats who pushed back on House plans Thursday and Friday spoke primarily about the cities and counties where they lived.

Those Democrats spent most of their time asking about those unprotected communities of interest and almost no time asking whether House staff’s plans would protect racial or language minorities.

Democrats also offered no alternatives to the staff-drawn plans and seemed wholly unprepared for the redistricting process.

At some points, it didn’t appear that Democrats were listening to the answers they were given. Rep. Andrew Learned, for example, asked a bizarre line of questioning where he asked for a definition of “contiguous” and asked if a district crossing Tampa Bay wouldn’t be contiguous (it would under decades-old precedent). He then asked if a hypothetical district stretching from Apalachicola to the Florida Keys would be allowed. Even though staff told him such a district would be contiguous and might be allowed, he continued on as if they had said that it wouldn’t be allowed.

He then seemed to confuse the concept of communities of interest, which is not protected by Florida law, with compactness, a protected standard that requires districts to be regularly shaped.

Byrd and Tyler Sirois, the Republican representative leading the congressional redistricting subcommittee, reminded Democrats they could submit their own plans or ask staff for help.

The proposals offered so far, both Byrd and Sirois said, are malleable and subject to change based on representatives’ feedback.

The committees won’t meet again until January, and they urged their fellow representatives to spend the next month asking their communities for feedback.

“Please do not wait until the last minute,” Byrd said Friday. “As I hope you’ve come to appreciate today, this process takes time to be done correctly.”