TERRY I. ADELMAN, Magistrate Judge.
This matter is pending before the undersigned pursuant to 28 U.S.C. § 636(c). Plaintiffs and Plaintiff/Intervenors seek judicial reapportionment of St. Louis County Council Districts necessitated by the failure of the St. Louis County Council Reapportionment Commission to file a timely plan of reapportionment based on the results of the Decennial Census of 2010.
Plaintiffs Steven V. Stenger, Michael E. O'Mara, Patrick M. Dolan, Sr., Mary Sanders, Vince Sansone, Jasmina Vajzovic, Keith Walston, and David T. Calhoun, who are citizens, residents, and registered voters of St. Louis County filed this action pursuant to 42 U.S.C. §§ 1983 and 1988 alleging violation of their rights to the equal protection of the laws as guaranteed by the Fourteenth Amendment to the United States Constitution, and by Article 1, Section 2, of the Constitution of the State of Missouri. Stenger, O'Mara, and Dolan are also Democratic members of the St. Louis County Council. The Defendants, sued in their official capacities, are members of the Board of Election Commissioners of St. Louis County. Plaintiffs seek to have this Court: 1) declare that the present boundaries of the seven council districts violate the requirements of the Constitution of the United States and of the State of Missouri, and 2) divide St. Louis County into seven county council districts that are contiguous, compact, and as nearly equal in population as may be.
The parties allege that the shift in population from 2000 to 2010 renders the county council districts in violation of the "one person, one vote" rule because some districts have significantly less population than other districts. Therefore, the lawsuit alleges that the population of St. Louis County is not equally apportioned between and among the seven county council districts as those districts are presently defined. Pursuant to Section 2.035 of the St. Louis County Charter, County Executive Charlie Dooley appointed a County Council Reapportionment Commission ("Commission") on May 24, 2011. Section 2.035 of the St. Louis County Charter states, in part:
St. Louis County Charter, Article II, County Council Part 1, Section 2.035.
Pursuant to the above charter provision, the terms of the Commission members expired at 12:01 a.m. on November 24, 2011. The Commission neither adopted a reapportionment plan for the county council districts nor filed a final statement of reapportionment of districts as mandated by the St. Louis County Charter. No procedure exists for extending the term of the Commission, reconvening the Commission, or accomplishing the reapportionment of the county council districts under any other procedure in the St. Louis County Charter or ordinances. County council members must be elected from districts and may not be elected at large pursuant to Sections 2.020, 2.040, and 2.160 of the St. Louis County Charter. Because the Commission could not reach a decision, the Plaintiffs filed this lawsuit on December 23, 2011, requesting an expedited discovery order and expedited trial. Plaintiffs noted that the first day for filing a declaration for candidacy for the 2012 general election to the St. Louis County Council is February 28, 2012, and the last day for filing is March 27, 2012. Therefore, the parties state that the Court's plan must be filed no later than on or before February 28, 2012.
Based on the above, the undersigned held a scheduling hearing pursuant to Fed. R. Civ. P. 16 immediately after the Defendants were served. On January 9, 2012, the undersigned entered an expedited scheduling plan which required that responsive pleadings be filed no later than January 13, 2012, and that any intervention of other parties should be made no later than January 13, 2012. The Court also entered an expedited discovery schedule, with discovery to be completed by February 1, 2012, a pretrial conference to be held on February 7, 2012, and the trial to commence on February 8, 2012 at 9 a.m.
On February 6, 2012, at approximately 4 p.m., Republican St. Louis County Council member Greg Quinn filed a motion to intervene in this case. According to Mr. Quinn's attorney, Mr. Quinn had been informed of this lawsuit on December 23, 2011 when St. Louis County Counselor Patricia Reddington e-mailed a copy of the petition to all County Council members. According to the motion to intervene under Fed. R. Civ. P. 24, Mr. Quinn was unable to obtain counsel until February 6, 2012, less than a day and a half before the trial. The Plaintiffs filed a response agreeing to the intervention, but with some restrictions. At the pretrial conference, Mr. Quinn's attorney agreed to almost all of the restrictions, including the prohibition that he be denied from requesting any relief except that a lawful map be drawn by the Court.
The undersigned informed the parties that the Court would likely allow the political evidence to be introduced on cross-examination and determine later whether the evidence was relevant. The Plaintiffs and Defendants agreed to this procedure. On February 7, 2012, the undersigned granted Mr. Quinn's motion to intervene in this case.
David Kimball is an associate professor of political science at the University of Missouri, St. Louis. He received an undergraduate degree at Brown University and a Masters Degree and Ph.D in political science from Ohio State University. His areas of specialty and study are American politics and voting behavior. He has written several books analyzing voting behavior in American elections including: Barry C. Burden and David C. Kimball, Why Americans Split Their Tickets: Campaigns, Competition, and Divided Government (University of Michigan Press 2002); and Martha Kropf and David C. Kimball, Helping America Vote: The Limits of Election Reform (Routledge 2011). In these books, he applied statistical modeling and statistical analysis commonly used in the political science field. In fact, much of his research involves statistical analysis. Dr. Kimball has also testified as an expert in two federal cases involving voting: one in the U.S. District Court in New York City and another in Ohio. He recently testified in a reapportionment case in the State of Missouri. He has extensive experience in the study of voting patterns of minorities. No one challenged Dr. Kimball's credentials in this case.
In reapportioning U.S. congressional districts, state house districts, city, and county districts, population totals in the most recent census are used to reconfigure the districts. This census data is packaged in county areas, but is also released for smaller units. The smallest unit is the census block, which is equivalent to a city block. The 2010 census revealed 18,747 census blocks in St. Louis County. For each census block, the census provides the age and ethnicity of each person in the block. A census tract is a group of census blocks joined together, and the average tract contains about 100 blocks. Based on the 2010 census data, St. Louis County has 199 census tracts containing about 5,000 people in each tract. The county also provides to the census bureau "voter tabulation districts," which are almost always voting precincts. Together the census and county provide ethnicity and age for each person in each voting tabulation district or precinct. Census blocks are generally used in the reapportionment process because the fewest people can be moved without disrupting the overall plan.
Four factors govern the reapportionment process: equal population, or "one person, one vote"; compactness; contiguity; and compliance with the Voting Rights Act. A frequently used model in reapportioning districts is to begin with the current boundaries and change them as little as possible while making equal the population of the districts. This is called the "least change" or "minimal change" method, which assumes that if the current district map complied with the redistricting criteria during the previous census, then a new map will likely comply with only limited changes. The "least change" method is advantageous because it maintains the continuity in representation for each district and is by far the simplest way to reapportion the county council districts.
Kimball considered the "least change" model the best model for accommodating all the different aspects of redistricting. He attempted to create new districts by changing existing boundaries as little as possible while achieving equal population districts that were as compact and contiguous as possible. In addition, the "least change" model allowed Dr. Kimball to avoid placing two incumbents in the same district. He did this, not to protect the incumbent, but to allow the voters to decide whether they desired to keep the same representative or reject him or her by electing someone else. Placing two incumbents in the same district takes this choice away from the voters.
Dr. Kimball testified that the most important principle in redistricting is to ensure that the districts comply with the "one person, one vote" mandate of the United States Supreme Court. To achieve this goal, Dr. Kimball calculated the ideal population in each district by dividing the number of people in the county by the seven county council districts. He determined this perfect number to be 142,708 people in each of the seven county council districts. Although it is almost impossible to reach the ideal population, Dr. Kimball attempted to make the population of each district as close as possible to the ideal population while keeping the other redistricting factors in mind.
Dr. Kimball also considered compactness in redrawing the district lines. Compactness means that areas within the same district are packed as closely together as possible so the ideal shape of a district would be a circle or a square. Compactness can not be completely achieved in St. Louis County because part of the county boundary follows the Missouri river, while another follows the Meramec river. This renders boundaries jagged and quite winding. However, the population polygon model provides a statistical way to measure the compactness of a district. This model runs on a scale from 1.0 to 0.0, with 1.0 being the most compact and 0.0 being the least compact.
As previously stated, another goal of redistricting is to ensure that the districts are contiguous, allowing a person to walk from one district to the next without crossing into a third district.
Finally, the new districts, as drawn, may not violate the Voting Rights Act. The Voting Rights Act requires that any district plans allow minority groups the opportunity to elect candidates of their choice. Two ways that redistricting can violate the Voting Rights Act are by "packing" or "cracking" the districts. If a minority group is concentrated in a particular area, "cracking" draws the district boundaries to split the minority group into different districts so these districts consist of a smaller minority percentage in each of those districts. Thus, the minority group cannot influence the outcome of the election in those districts. "Packing" applies the opposite strategy. "Packing" puts all or as many voters of a minority group as possible into one district so that all the neighboring districts have a very small share of a minority group. Likewise, the minority group would have little or no influence on elections in the surrounding districts.
To create the new proposed map, Dr. Kimball first considered the existing map in light of the 2010 census population in each of the seven current districts. He observed that some districts, particularly the first, the second, and the fourth, were substantially below the new population ideal. Further, some other districts, particularly the third, the sixth, and the seventh, were substantially above the population ideal. The table placed into evidence demonstrates almost a 17,000-person difference between the smallest populated district and the largest populated district for standard deviation of more than 11%. Thus, redrawing the districts was necessary. The following exhibit depicts the deviation in population in each of the districts:
As stated above, to redraw the maps so that the population was equal, Dr. Kimball adopted the "least change" approach, starting with making minimal changes to the boundaries drawn by Judge Perry in 2002, then moving toward equal population while also considering compactness and contiguity. He used existing districts because they were substantially compact and attempted to cause minimal change to the make-up of the various districts. Dr. Kimball believed he would have difficulty achieving equal population and keeping the districts compact if he tried to redraw the entire map. Initially, he did not consider the Voting Rights Act or racial data in drawing the districts. He considered only equal population, compactness, and contiguity. After redrawing the districts, he then analyzed the new districts as to their compliance with the Voting Rights Act.
Dr. Kimball used computer software redistricting programs to draw the maps. These programs combined census data as to population and St. Louis Planning Commission data as to voting ages, then superimposed the information on a map of St. Louis County containing each of the seven districts. In this way, he used the software to move the lines of the districts and determine the effect on the population and compactness. Dr. Kimball did not use any political performance data, nor did he review the plan after drawing it to determine its political performance or significance. In summary, he did not review the political performance of the proposed or changed districts but only considered population, compactness, and contiguity.
Specifically, working in a counterclockwise manner, Dr. Kimball added population to the three underpopulated districts, and subtracted from the three districts which were overly populated. He only minimally changed the fifth district, because it was already close to the ideal population and was very compact. Dr. Kimball detailed each of the changes he made to the various districts and where they were located. Further, he explained his revisions and their impact on the population in the districts as well as their deviation from the ideal population both in numbers and in percentage points. This information is depicted in Plaintiffs' Exhibit 2a as follows:
As demonstrated in the above chart, the proposed plan distributes the population in each district equally. All deviations are less than one-tenth of one percent, a number that, in Dr. Kimball's opinion, is de minimis. The maximum deviation is 64 individuals over the ideal population in the second district and 60 individuals under the ideal population in the fourth district. Dr. Kimball testified that compactness would be sacrificed if he moved the census blocks to achieve even more equality.
Next, Dr. Kimball used the population polygon measure of compactness to determine whether the new council districts were compact. In comparing the previous districts drawn by Judge Perry, he determined that the districts he drew were slightly more compact than the districts drawn in 2002. Plaintiffs' Exhibit 2b represents the compactness of the districts:
As the above chart shows, with 1.0 being the most compact and 0.0 being the least compact, the districts all measure .9 or .8 and are, thus, compact.
As previously stated, Dr. Kimball did not analyze the racial make-up of the newly redrawn districts until after he made certain the districts were equal, contiguous, and compact. When conducting this Voting Rights Act analysis, he determined that most African Americans resided in the first and fourth districts of St. Louis County. Under the new configuration, both the first and fourth districts were "majority minority" districts. A "majority minority" district occurs when the minority population (in this case African American) constitutes a majority of the population of that district. Ten years ago when the districts were drawn in a previous case, St. Louis County contained only one "majority minority" district. An "effective minority" district is a concept used in determining whether a plan complies with the Voting Rights Act. If a "majority minority" district is composed such that the minority group has an equal chance of electing a candidate, it is an "effective minority" district.
When Judge Perry drew the districts ten years ago, many experts believed that a minority population of 65% was necessary to ensure an "effective minority" district. The experts based their beliefs on four assumptions, some of which no longer seem to apply in St. Louis County. The first assumption was that the minority group will vote almost entirely or largely in a bloc for their preferred candidate, and the majority group will vote almost entirely as a bloc against the minority preferred candidate. The second assumption was that the voting age population of minority groups tends to be smaller than its share of the overall total population. For example, if the minority accounts for 55% of the entire population, it will account for only 50% of the voting age population. The third assumption was that, among voting-age population, the voter registration rate for minorities is about five percentage points lower than the registration for whites. Finally, the last assumption presumed that, among registered voters, the turnout rate of minority voters is about five percentage points lower than the turnout rate for white voters.
Two recognized statistical methods test these assumptions. The first test is the homogenous precinct analysis, and the second is the ecological regression method. Utilizing the homogenous precinct model, Dr. Kimball looked at voting precincts in the fourth district, one which was almost entirely white and one which was almost entirely African American, and examined the voting behavior in those precincts. The assumption was that those percentages would hold throughout the fourth district. The ecological regression theory analyzes how African American voters and white voters voted in the entire district and county.
Dr. Kimball's analysis showed that in elections where African American candidates were running countywide (which totaled five elections), the voter turnout between African American voters and white voters was almost exactly the same, and in some instances greater among the African American population. Further, his analysis showed that white voters do not vote against African American candidates in a bloc to the extent previously believed. In the two elections in which Charlie Dooley, an African American, ran for County Executive, the analysis showed that 45% to 49% of the white voting population in the fourth district voted for Charlie Dooley. This study and other studies demonstrate that white bloc voting does not exist in St. Louis County to the degree previously believed. Further, the percentage of African American voters voting in elections with African American candidates on the ballot is similar to the percentage of white voters. Thus, the study disproved two of the assumptions requiring a 65% minority population in order to qualify as an "effective minority" district. Dr. Kimball relied on these studies to determine that the 55% African American majority and the 51% voting age population majority of African Americans in the fourth district is more than enough to make it an "effective minority" district. This, along with the 71% voting age population of African Americans in the first district, gives the county two "effective minority" districts. The following chart shows the racial make-up of the proposed districts:
The attached chart demonstrates the voting age population broken down on racial grounds:
The analysis involving homogenous districts and ecological regression is shown in the following charts:
Thus, based on all of the above, Dr. Kimball unequivocally believes that the map he drew and introduced into evidence (which was the only map introduced into evidence), was equal, compact, contiguous, and complied with the Voting Rights Act. In addition, he testified on direct examination, several times on cross-examination, and upon examination by the Court, that political considerations played no part in his drawing of the map.
This marks the fourth decade in which St. Louis County has failed to adopt
In
Further, court-ordered districts must come as close to population equality as possible and much closer than the political or legislative process would allow. In
521 U.S. 74, 98 (1997).
Based on the above, the undersigned finds that the plan proposed by the Plaintiffs is statistically equal as to each district. The maximum deviation between the highest population district and the lowest population district is less than one-tenth of one percent, much lower than the difference approved by the Court of Appeals in
Section 2.035 of the St. Louis County Charter provides in pertinent part:
St. Louis County Charter, Article II, County Council Part 1, Section 2.035.
As to compactness, the County Charter stresses that the districts must be composed of contiguous territory and must be as compact as they "may be." Thus, the Court must examine compactness in drawing the districts. The expert specifically considered the compactness requirement in drawing the proposed map. He achieved as equal a population as possible while drawing the districts so they would be as compact as possible. The undersigned notes that perfect compactness is not possible because the outer boundaries of the county are bordered by meandering rivers. The expert measured the compactness of the districts by using the population polygon statistical method. Using this method, each of the districts comes relatively close to a perfect score of 1.0. No district falls below .8 on the population polygon score, and five of the seven districts achieve a score of .9 or better. Further, merely viewing the districts indicates that they are reasonably compact. A small bump or finger extends from the third district and slightly into the seventh district. This small bump is adequately explained by the expert as necessary to equal the population of the two districts. He stated that it was the least harmful way to accomplish this without sacrificing compactness in a more drastic manner. Even with this bump, this district receives a compactness score of .82. Thus, the undersigned concludes and finds that the proposed map comports with the compactness requirement in the County Charter. Further, all parts of the districts are self-contained and, therefore, are contiguous.
Although this case contains no Voting Rights Act claim, the Court nevertheless should review the plan to determine that it protects "minority voting rights."
42 U.S.C § 1973(b).
Section 2 contains no per se prohibitions in these particular types of electoral districts. "It says nothing about majority-minority districts, districts dominated by certain political parties, or even districts based entirely on partisan political concerns. Instead, § 2 focuses exclusively on the consequences of apportionment."
Plaintiffs seeking to protect their constitutional rights and show a voting rights violation can establish a § 2 violation by proving three factors as first set forth in
Based on the evidence adduced at trial, the undersigned finds that the first two factors are met. The evidence revealed that African Americans constitute a majority of the voting age population in the first and fourth districts and that they vote in 85 to 95% majorities for African American candidates. The evidence, however, does not support the third requirement, i.e., that racial-bloc voting typically frustrates the election of the minority's preferred candidates. In the last five county wide elections in which an African American candidate ran against a white candidate, the African American candidate won the election. The following chart illustrates this result:
Just as importantly, the statistical data supports the conclusion that district one and district four are both "effective minority" districts. In these districts, white voters do not vote in a bloc against African American candidates to the extent previously believed, and the voter turnout percentage is approximately equal between white voters and African American voters.
Further, because the African American "effective minority" districts are in approximate proportion to their population of St. Louis County, the plan would likely not violate the Voting Rights Act even if the
The
As stated above, the Court concludes that the plan proposed by Plaintiffs (hereby attached as Exhibits A and B), is equal in population as to each district, compact as to each district, contiguous as to each district, and protects minority voting rights. The evidence reveals that the map was not drawn with Republican or Democrat politics in mind. The expert unequivocally testified that he did not consider politics either before or after drawing the map. The expert made this clear on both direct and cross-examination, despite the efforts of the intervenor to prove otherwise. Although the expert's map was somewhat similar in several respects to the Democrat's last proposal to the Reapportionment Commission, there were also many differences. Significantly, the expert testified unequivocally that he had not considered the last proposed map from the Democrats in any way in drawing his map. Thus, although given an opportunity to do so, the intervenor was unable to show that the Plaintiffs' proposed map was drawn with a political purpose in mind. Therefore, the map is equal in population, compact, contiguous, protects minority rights, and is not the product of politics. The undersigned adopts the Plaintiffs' map as the map to be used in the reapportionment of St. Louis County because it best satisfies the constitutional and statutory goals of apportionment.
Accordingly,
