The Case for Race: The Battle for a Three-Judge Court

Typically, federal litigants do not have access to a three-judge court at the district court level. But there are exceptions.

The Decennial Census: A Constitutional Pillar of American Government

The U.S. Constitution mandates a nationwide headcount every 10 years, known as the decennial census. This requirement, enshrined in Article I, Section 2, serves as a cornerstone of American democracy with important implications for the structure and function of the government:

Apportionment of Representation:

The primary purpose of the census is to determine the population of each state. This data is used to fairly allocate seats in the House of Representatives among the states, ensuring representation proportionate to population shifts. States with higher populations gain representatives, while those with declining populations may lose them. This ensures that each citizen's voice has roughly equal weight in the legislative branch.

Distribution of Federal Resources:

Beyond representation, census data plays a vital role in distributing federal funding across states and localities. Billions of dollars are allocated annually to programs like Medicaid, education grants, and infrastructure projects based on population data. An accurate census guarantees that areas with higher populations receive their fair share of these critical resources.

Drawing Electoral Districts:

Census data is also essential for redrawing congressional and state legislative district boundaries. This process, known as redistricting, ensures that districts have roughly equal populations, upholding the principle of "one person, one vote."

Beyond the Basics:

While the core purpose lies in population counts, the census also collects valuable demographic information. This data informs policy decisions across various sectors like healthcare, education, housing, and social services, allowing for targeted programs and resource allocation based on specific population needs.

Effect of the 14th Amendment:

Section 2 of the 14th Amendment states: “Representatives shall be apportioned among the several states according to their respective numbers, counting the whole number of persons in each state.”
Therefore the 14th Amendment reinforces the decennial census mandate, emphasizing that representation in the House is directly tied to population. This ensures that the census remains a vital tool for maintaining a fair and representative government.
This provision grants each state the right to conduct a full count of all individuals living within its borders during a census to determine its population. (The 14th Amendment was ratified in 1868 after the Civil War. It addressed citizenship rights and equal protection under the law for all individuals, including former slaves. Previously, slaves were only partially counted based on the 3/5ths compromise.)

The Clinton Administration and the Push for Statistical Sampling on the 2000 U.S. Census

In the 1990s, the Clinton Administration favored statistical sampling as a way to achieve a more accurate count, while Congressional Republicans preferred the traditional head count used in past censuses. See, 08 SEP 1998 THE 2000 CENSUS: SAMPLING OR A STRAIGHT COUNT?

The Clinton Administration believed that statistical sampling was the best method to achieve an accurate count. Sampling involved using computer-generated statistical models to estimate data for approximately 10% of the population (around 26 million people) who were not part of the traditional enumeration process. These “virtual people” would be derived from data collected during on-site visits, and the figures would be adjusted based on super-accurate enumeration reports from selected areas. The opposition argued that sampling was unreliable, costly, and would be used to improperly influence political agendas.

Congress Passes Pub. L. No. 105-119 § 209 to Prohibit Statistical Sampling

Because of the fierce opposition to statistical sampling, Congress passed 13 U.S.C. § 141 (note), also known as Pub. L. No. 105-119 § 209. Section 209 allows "any person aggrieved by the use of any statistical method" that violates the Constitution or any law in determining population for congressional apportionment or redistricting to sue the government.

The law specifically targets the use of statistical methods that alter the raw enumeration data, such as adding or subtracting counts based on statistical inference. In cases where the challenge falls within the scope of Section 209, the law mandates the convening of a three-judge court, including the district court level or trial level of a federal case. Even more unusual, any direct appeal of the three judge court's decision must be heard by the U.S. Supreme Court.

Section 209 was ultimately used by Congress to successfully oppose statistical sampling in Dep’t. of Commerce, et al., v. U.S. House of Representatives, et al., 525 U.S. 316 (1999) (the Supreme Court ruled in favor of Congress and prohibited statistical sampling as unconstitutional).

The U.S. Census Bureau Loses on Statistical Sampling but Wins on Statistical Imputation

The Census Bureau didn't give up completely on using statistical methods to assist them in the decennial census count. Although the Supreme Court disagreed with them about raw statistical sampling, the Census Bureau attempted to use statistical imputation instead.

Statistical Imputation:

Statistical imputation is a method used by the U.S. Census Bureau to estimate missing or incomplete data when conducting the decennial census.
When census forms are not returned or contain confusing, contradictory, or incomplete information, the Bureau follows up with field visits. However, even after these visits, there may still be uncertainties about certain addresses or housing units.
In such cases, statistical imputation infers that the uncertain address or unit has the same population characteristics as its geographically closest neighbor of the same type (e.g., apartment or single-family dwelling) that did not return a form.
The goal is to improve data accuracy by filling in gaps based on nearby similar units.

Statistical imputation was also reviewed by the Supreme Court. The seminal case is Utah v. Evans, 536 U.S. 452 (2002). In this case, the state of Utah challenged the Census Bureau’s use of “hot-deck imputation” during the 2000 census. The Supreme Court ruled in favor of the Census Bureau, holding that the use of hot-deck imputation did not violate 13 U.S.C. § 195 or the Census Clause of the Constitution. The Supreme Court emphasized that statistical imputation was not the same as sampling and did not undermine the constitutional requirement for an actual enumeration. Nonetheless, a challenge to statistical imputation was still seen as a valid basis to invoke the rights and privileges of Section 209, i.e., a right to a three-judge court and direct appeal to the U.S. Supreme Court. For example, the law was used by the State of Alabama to challenge differential privacy in the 2020 U.S. Census, in Alabama v. US Dept. of Commerce, 546 F. Supp. 3d 1057 (M.D. Ala. 2021).

Invocation of Section 209 in the Case for Race

I was well aware of Pub. L. No. 105-119 § 209(b) and decided that I should invoke the law to ensure Supreme Court review of any decision made in my case about race. Unfortunately, the law was poorly understood by the district court presiding over the case. Without reviewing the text of Section 209, the district court denied my first motion for a three-judge court (based strictly on the language in 28 U.S.C. § 2284(a), which provides for a three-judge court, but only in a challenge to constitutional apportionment.

My case was different. I brought an as-applied challenge to statistical imputation because the Census Bureau uses statistical imputation to infer not just the number of unknown persons but also racial characteristics on missing or incomplete decennial census forms. When respondents do not provide information about their race on Census forms, the Bureau employs imputation methods to fill in the missing data. These methods use statistical techniques to assign a race based on available information, such as the race of individuals in neighboring households. While imputation is necessary to ensure complete data sets, concerns arise regarding its accuracy and potential biases. I also argued that the Bureau would be necessarily misclassifying the races of certain persons when it imputes racial characteristics during it imputation process (which happens after the main counting is done).

The district court did not understand this issue at all, even after I clarified my position on the matter. After filing a secondary motion for a three-judge court, the district court refused to rule on the motion. This compelled me to file a petition for writ of mandamus in the Eleventh Circuit Court of Appeals, an appellate court with jurisdiction over federal district courts in Alabama, Georgia, and Florida.

The petition for mandamus was filed on October 7, 2021. A few days later, well before the 11th Circuit could weigh in on the issue, the district court denied the second motion for a three judge court, citing inapplicable case law that was decided before Section 209 was passed by Congress.

The district court also failed to appreciate that a Section 209 challenge did not have to be a pure constitutional challenge, but that a litigant could also argue that a statistical method violated federal law due to the "any provision of law" language in the statute. See, Pub. L. No. 105-119 § 209(b), which states,

"Any person aggrieved by the use of any statistical method in violation of the Constitution or any provision of law (other than this Act [see Tables for classification]), in connection with the 2000 or any later decennial census, to determine the population for purposes of the apportionment or redistricting of Members in Congress, may in a civil action obtain declaratory, injunctive, and any other appropriate relief against the use of such method."

However, because the district court resolved the motion for a three judge court and dismissed the case for race a day later, the Eleventh Circuit denied the petition for mandamus, finding that the district court's decision could be reviewed on a direct appeal.

Ultimately, both the Eleventh Circuit and the U.S. Supreme Court would not review the district court's refusal to assemble a three-judge court. On direct appeal, the Eleventh Circuit avoided the issue by finding I had no standing, and the Supreme Court denied certiorari review.

"In addition to the dismissal of the third amended complaint for lack of standing, Van Cleve also challenges the denial of his motion for a three-judge panel, pursuant to the 1998 Departments of Commerce, Justice, and State, the Judiciary, and Related Agencies Appropriations Act, § 209, Pub. L. No. 105-119, 111 Stat. 2440, 2481-82 (1997) (codified at 13 U.S.C. § 141 note), and the district court's conclusion his third count failed on the merits. However, because we hold Van Cleve lacks standing, we need not address these issues." Van Cleve v. Dep't of Commerce, et al., Case No. Case No. 21-13699, footnote 1 (11th Cir. May 24, 2022) (certiorari denied, 143 S. Ct. 210 (2022)).

The race case presented a unique opportunity to examine the Census Bureau's race imputation methods and outdated racial terminology. Unfortunately, because of the procedural hurdle of standing, the generational racial issues presented by the lawsuit went unaddressed by the courts. Only time will tell if the judicial system will revisit the continuous issue of legacy racial terminology and its impact on the U.S. Census and the zeitgeist of the American public.