What if the Supreme Court Eliminated Noncitizen Representation?

by Erik Voeten on March 21, 2013 · 4 comments

in Campaigns and elections,Immigration,Judicial,Law,Race

We are delighted to welcome back Karthick Ramakrishnan


Next week, the Supreme Court will hear the petition of Lepak v. City of Irving, to decide whether the principle of “one person, one vote” should be clarified to allow the exclusion of noncitizens.

 Plaintiffs in the case argue that counting noncitizens dilutes the influence of eligible voters, while others argue that the United States has, from its founding, counted persons such as slaves and women for purposes of representation, even if they did not have the right to vote.  States with high proportions of noncitizens have a lot at stake in this debate, given its potential to affect decisions ranging from House apportionment to the drawing of state and local legislative districts.

 If the Supreme Court allows for the exclusion of noncitizen residents, how would it affect Congressional apportionment?  We can calculate these effects using the following method: compiling data from the U.S. Census Bureau on total persons and total noncitizens, calculating what the House apportionment would look like if noncitizens were excluded (using the current method of equal proportions), and taking the difference from the actual allocation of seats based on total persons counted in 2010.


When we conduct this type of analysis for the current apportionment cycle, we find that gains from a policy shift would be relatively modest across states, but losses would be significant for California and Texas (see Table).  If, moving forward, only citizens were counted for purposes of apportionment, a total of 10 seats would be allocated differently.  California would stand to lose the most House seats (5), followed by Texas (2), Florida (1), New York (1), and Washington (1).  By contrast, various states in the Midwest region would gain one seat each.

If the logic of counting only enfranchised populations today were extended even further¾to also exclude children and other non-eligible groups such as prisoners¾a total of 13 seats would be allocated differently than they currently are today.  California would lose a total of 6 House seats, Texas would lose 5 seats, and Florida and Georgia would lose one seat each.  New beneficiaries under a system of counting only voting-eligible populations would include Oregon and Massachusetts (see Table).

 It is unlikely that the Supreme Court will take up the case and decide in a way that leads to significant changes in state power through Congressional apportionment.  This is especially so given centuries of legal precedent.  The United States has always counted persons for Congressional apportionment, even if they were unable to vote.  This includes women, children, noncitizens, and three-fifths of slaves.  (The only exception was for Indians “not taxed,” who were excluded from Congressional apportionment until 1940).

 Incidentally, the distortions in Congressional apportionment from counting or excluding noncitizens is relatively small (2% of all House seats today) when compared to the gains the South had from the Three-Fifths Compromise (13% of House seats in 1850, see Humes et al. 2002).

 It is possible, of course, that the Supreme Court could leave Congressional apportionment alone, and still allow state and local districts to exclude noncitizens and other ineligible voters.  However, such a move would still cause significant political upheaval, as places with relatively high proportions of noncitizens would lose political power.  At the state level, this would most likely be dense urban areas and agricultural areas with immigrant workers.  At the city and county level, immigrant-heavy neighborhoods would stand to lose the most, and noncitizens who currently rely on political representation via proxy (through citizens in their district who share their same preferences) would see their influence greatly diminished.

 Thus, even a more limited concession in Lepak v. City of Irving would have significant ramifications for state and local politics, particularly in places with sizable immigrant populations.


Karthick Ramakrishnan is associate professor of political science at the University of California, Riverside, and has published four books and several articles on immigration, race, and politics.  This work is part of a current project on political implications of the U.S. Census for the Russell Sage Foundation.


RobC March 21, 2013 at 10:33 am

Just to be crystal-clear, though Professor Ramakrishnan says the Supreme Court will “hear the petition” of Lepak, what it will in fact do is decide whether to grant certiorari in the case. (I’ll be quite surprised if it does.)

counsellorben March 21, 2013 at 10:47 am

After reviewing the cert petition, it seems unlikely that the Court will grant cert in this case. The “discord” among the Circuit Courts alleged in the petition does not rise to the level of a split between the Circuit Courts.

Mike Liveright March 21, 2013 at 10:52 am

I wonder what would the result be if in addition, the Representation was adjusted to only representation those who voted. This would mean that states who suppressed voters would be penalized.

Finally, and I have no way to measure it, I would also like to see the effect of trying to adjust Representation based on Lack of Gerrymandering. Thus a state whose districts were drawn compactly would have a higher representation than on whose were manipulated. My first cut would be to take the Gerrymander measure as the Sum of the Squares of the ( Length of the districts borders over their area ) and divide each state’s representation by that Sum. … It seems to me that this would reward the states for their compact districts, thus for their compactness.

RobC April 1, 2013 at 3:41 pm

And today the Supreme Court announced it had denied certiorari.

Comments on this entry are closed.

Previous post:

Next post: