Earlier this month, the United States Supreme Court heard arguments in Evanwel vs. Abbott. The issue in this case is how to measure population for the purposes of determining if districts have roughly equal population. The challengers are asserting that population should be based on voters rather than the total population. The State of Texas is claiming that each state gets to choose the appropriate measure of population. This case involves both theoretical discussion of the nature of representation as well as very serious political impact. The decision could vastly alter the politics of the U.S.
At the time of the American revolution, representation in the British Parliament was roughly similar to what the framers adopted for the Senate — each county and each incorporated urban area had two members of parliament, regardless of population changes. Because of population changes and restrictions on voter eligibility, some members of Parliament stood for election in districts that had fewer than 50 voters. When the American colonists complained that parliament did not represent them (and therefore did not have the right to make laws taxing the colonies), some British political theorists argued that members of parliament did not actually represent their districts but rather “virtually” represented the entirety of the British Empire. This debate over what it means to represent somebody has continued throughout U.S. history.
The current law on redistricting arises out of the Equal Protection Clause of the Fourteenth Amendment. The original cases arose out of the modern-day equivalent of the British “rotten boroughs” of the Eighteenth Century — states that not redrawn state legislative lines in decades as population had shifted or who based representation on geographic areas (e.g. one state senator per county) rather than population. In explaining what equality meant in redistricting, the U.S. Supreme Court created the phrase “One Man, One Vote.” In doing so, the Supreme Court did not define the proper way to measure population.
In the U.S., immigrants have tended to settle in urban areas — typically those with significant populations from the ancestral country. Because those members of these populations who can vote tend to vote Democratic, this results in Democratic districts having more non-citizen immigrants than Republican districts. In Texas, some conservatives have challenged the use of total population (which would include the young, non-citizens, and others who are ineligible to vote) instead of some measure of voters. They note that, while the districts in Texas have equal population, the districts have vastly different numbers of voters (approaching a twenty percent deviation from the state-wide average district size).
On the legal side, the challengers have the Supreme Court’s past rhetoric on their side. They do not have a strong textual argument. Both in the original text of the Constitution and in the Fourteenth Amendment, congressional apportionment is based on total population. While the Fourteenth Amendment included a (never used) provision allowing the reduction of representation if a state denied the right to vote to adult male citizens with no criminal record, it did otherwise rely on the number of adult male citizens to calculate apportionment. States with large immigrant communities, or with a large number of children, get the benefit of that added population in calculating their seats in Congress.
Aside from the Constitutional text, there is also the practical problem of calculating eligible voters. In recent cycles, the Census has become shorter and shorter (in part due to conservative complaints about the invasiveness of some questions). Many of the details on which a calculation of eligible voters could be based are only asked in the American Community Survey (an annual survey which replaced the long form, but only goes to 3,000,000 households each year, about 4% of the population). While the sample size of the American Community Survey is “good enough” for the purposes of statistical analysis, they fall far short of being an actual enumeration of the population. Additionally, the Census measures where people actually live on Census day. College students are calculated as living in their dorm or apartment in the two where they go to college — even though they might vote back home. Similarly, a business person who has been sent to spend four or five months in a city on a project will be counted as living in that city, even though they might still be registered to vote back in their home town. Prisoners are counted as residents of their correctional facility, not the place that they will return to when released from custody.
From the political standpoint, this move is an attempt to deprive Democrats in general and Hispanics in particular of Congressional and legislative seats. In Texas, Hispanics are approximately 40% of the population. However, approximately 25% percent of the Hispanic population are not citizens. Only counting citizen would reduce the number of Hispanic districts by 14 in the State House and 3 or 4 in the State Senate and in Congress. At least half of the lost seats would go to Republican areas of Texas. While I have not done raw numbers, similar results would be likely in other states in the Southwest — perhaps increasing the number of safe Republican Congressional seats by 6-10. While there are other significant cases potentially impacting major policy questions, this case is a potential game changer in terms of the balance of power in state legislatures and in Congress.