Counting Noncitizens Distorts State Legislatures

 By Jason Richwine and Karen Zeigler

Previous CIS research has shown that immigration distorts both the apportionment of U.S. House seats among states and the drawing of U.S. House district lines within each state.1 This new report focuses on state legislatures. Using as examples the five states with the highest foreign-born shares — California, New Jersey, New York, Florida, and Maryland — the report shows that immigration causes state house districts to have dramatically different numbers of eligible voters.

The reason is that immigration has added to state populations about 25 million noncitizens — including legal permanent residents, guestworkers, foreign students, and illegal immigrants — but noncitizens by law cannot vote in state elections. Rather than draw lines to equalize the number of eligible voters in each legislative district, states instead equalize the total population in each district. Since noncitizens are not distributed uniformly throughout any state, some legislative districts have significantly more citizens — and, hence, more eligible voters — than other districts.

• The difference between the greatest and smallest numbers of citizens at the district level can be very large within the same state. As the first section of Table 1 indicates, California Assembly District 1 has one member representing a district with 396,704 adult citizens, while District 57 has a member representing just 248,324 adult citizens.
• The disparity is even larger for the Maryland House of Delegates. District 47B contains 15,244 adult citizens, which is only 40 percent as many as the 38,427 adult citizens in District 1B. (See “Methodological Notes” for an interpretive caveat.)
• These disparities give disproportionate representation to some voters simply because they live near more noncitizens than other voters do. The middle section of Table 1 shows that, for the example of New York, there are 11.6 members of the State Assembly for every million adult citizens who live in districts that have noncitizen shares greater than the median district in the state. By contrast, there are only 9.8 members per million adult citizens who live in districts with foreign-born shares below the New York median.
• Although greater representation for voters who live in high-noncitizen areas is unfair regardless of partisan implications, the distortion clearly benefits the Democratic party. Figures 1 through 5, along with the bottom row of Table 1, illustrate the district-level correlation between noncitizen population share and Democratic voting. As an example, for every one percentage-point increase in the noncitizen share of a California State Assembly district, the share voting Democratic in the district rises by an average of 1.42 percentage points. Put simply, because noncitizens tend to congregate in Democratic-leaning areas, they redistribute political power from Republican voters to Democratic voters.
• To avoid the representational distortions caused by immigration, policymakers could try to exclude noncitizens from population counts, although there are statistical and legal hurdles to doing so.2 Policymakers could also encourage noncitizens to naturalize, although only about one-third are eligible.3 Another solution is simply to reduce immigration. Under a low-immigration system, structural effects on representation would naturally shrink along with the size of the noncitizen population.

 

Table 1. Noncitizens Distort Voter Representation in Multiple Ways
	Calif.	N.J.	N.Y.	Fla.	Md.
District Adult Citizen Populations
Highest	396,704	195,323	111,485	169,557	38,427
Lowest	248,324	124,213	57,606	95,127	15,244
Ratio1	1.60	1.57	1.94	1.78	2.52
Representatives per Million Adult Citizens ...
… in High Noncitizen Districts2	3.3	13.5	11.6	8.2	34.2
... in Low Noncitizen Districts2	2.9	11.7	9.8	7.0	30.2
Average Percentage-Point Increase in Democratic Vote Share for Every One Percentage-Point Increase in Noncitizen Share (by District)	1.42	1.99	1.76	1.17	2.18
Source: Source: 2019-2023 ACS for demographic data; see “Methodological Notes” for election data. 1 When comparing total adult population sizes, the ratios are lower but still greater than one. See “Methodological Notes”. 2 A high (low) noncitizen district has a noncitizen share greater (lower) than the state's median district.

 

Voters in state legislative districts with greater shares of noncitizens tend to vote more Democratic.

Figure 1. California

Figure 2. New Jersey

Figure 3. New York

Figure 4. Florida

Figure 5. Maryland

Source: 2019-2023 ACS for demographic data; see “Methodological Notes” for election data. Only the lower legislative chamber in each state is included in these analyses.

Methodological Notes

The Census Bureau website provides counts of adult citizens and noncitizens by state legislative district, based on data from the 2019-2023 American Community Survey (ACS).4 This report focuses on the lower house of each state’s legislature, since its members are more numerous and generally face election more often than members of the upper house.

The first section of Table 1 requires an interpretive caveat. Because the number of adults counted in each district in the 2019-2023 ACS sample is not the same as the 2020 total population counts used for redistricting, the data should show some population imbalances across districts even before the effect of noncitizens is taken into account. Indeed, Table 2 indicates that one state legislative district in Maryland has a total adult population (citizens and noncitizens together) of 38,844, while another district has just 30,132 adults. This 1.29 high-low ratio, while notable, is smaller than the difference between the highest and lowest adult citizen populations, which are 38,427 and 15,244, respectively, for a ratio of 2.52. In each state, the range of adult citizen populations across districts is substantially greater than the range of total adult populations.

 

Table 2. District Size Disparities Are Greater When Counting Only Adult Citizens
	Calif.	N.J.	N.Y.	Fla.	Md.
District Adult Populations
Highest	430,909	198,625	119,667	173,114	38,844
Lowest	331,683	140,685	75,724	127,557	30,132
Ratio	1.30	1.41	1.58	1.36	1.29
District Adult Citizen Populations
Highest	396,704	195,323	111,485	169,557	38,427
Lowest	248,324	124,213	57,606	95,127	15,244
Ratio	1.60	1.57	1.94	1.78	2.52
Source: 2019-2023 ACS for demographic data; see “Methodological Notes” for election data.

 

In collecting data for the bottom section of Table 1, the default procedure was to measure the Democratic voting share in each district in the general election of 2024. However, state-to-state idiosyncrasies in voting systems required some adjustments that are detailed below.

For each of its legislative districts, California operates a “jungle primary” in which candidates from all parties compete, and the top two vote-getters proceed to the general election. Because the two general-election candidates are so often from the same party, the primary results are actually more informative about the district’s political preferences. In Figure 1, the Democratic vote in California is the sum of the votes received by all Democratic candidates in the jungle primary.5

New Jersey residents vote for two candidates per district in odd-numbered years. The Democratic vote in each district is therefore the sum of the votes earned by the two Democrats running in 2023.6

No special adjustments were required for the Democratic vote in New York.7

In Florida, districts with an unopposed candidate were treated as voting 100 percent for that candidate’s party.8

Maryland residents vote for one, two, or three delegates per district every four years. The Democratic vote in each district is the sum of the votes for the Democratic candidates in the 2022 general election.9 Note that District 47B is such an outlier (54 percent noncitizen share, 100 percent Democratic voting) that it does not appear in Figure 5.

The general category of “write-in” was not included in any of the states’ vote totals.

---

End Notes

1 Steven A. Camarota and Karen Zeigler, “Tilting the Balance”, Center for Immigration Studies, October 31, 2024; Steven A. Camarota and Karen Zeigler, “How Non-Citizens Impact Political Representation and the Partisan Makeup of the U.S. House of Representatives”, Center for Immigration Studies, October 31, 2024.

2 Redistricting is based on full population counts from the decennial census, which does not currently ask a citizenship question. Furthermore, whether states would be allowed to draw districts based on citizenship is an unresolved issue in constitutional law. See Justice Alito’s concurrence in Evenwel v. Abbot (2016).

3 Sarah Miller, “Estimates of the Lawful Permanent Resident Population in the United States and the Subpopulation Eligible to Naturalize: 2024 and Revised 2023”, Department of Homeland Security, September 2024.

4 “Explore Census Data”, U.S. Census Bureau.

5 “March 5, 2024, Presidential Primary Election, State Assembly Member”, California State Portal.

6 “New Jersey 2023 election results”, WHYY.

7 “2024 New York State Assembly General Election Results”, The Journal News.

8 “Official Results”, Florida Department of State, Division of Elections.

9 “Official 2022 Gubernatorial General Election Results for House of Delegates”, State Board of Elections.

https://cis.org/Report/How-Counting-Noncitizens-Distorts-State-Legislatures