Do illegal migrants affect census numbers
Executive summary
Illegal or undocumented migrants are counted in the U.S. census because the Census counts all residents regardless of immigration status (the Census Bureau’s longstanding legal and operational practice) and therefore they do affect headline population totals used for apportionment and many statistics [1]. That effect is real but constrained: researchers show undocumented residents change local and state totals in measurable ways, yet historical analyses find excluding them would have produced only very small shifts in House seats and Electoral College votes since 1980 [2] [1].
1. What the Census counts, and why immigration status is not asked
The Census Bureau’s legal and operational framework instructs enumerators to count everyone living in a residence on Census Day, without asking or using immigration status as a criterion, which means undocumented residents are included in the raw population counts used for apportionment and many federal data products [1] [3]. Because surveys like the American Community Survey (ACS) and the Current Population Survey (CPS) do not ask legal status directly, researchers and agencies estimate the unauthorized population by combining these surveys with administrative data and modeling assumptions rather than directly identifying individuals as “illegal” in census records [3] [4].
2. How big an effect undocumented residents have on headline counts
Estimates of the unauthorized population vary—Pew and other research centers used ACS-based methods to estimate roughly double-digit millions in recent years, reflecting changes in migration through 2023 and beyond—so the raw numeric impact on national population totals is material in absolute terms [5] [3]. At the state and local level, those residents can shift population shares enough to affect resource allocation formulas, legislative redistricting logic, and local planning because many state apportionment and funding formulas start with the federally produced counts [1] [4]. However, empirical counterfactual studies find that removing undocumented residents from apportionment counts would have shifted at most a handful of House seats and a few Electoral College votes across decades—an important but limited political effect historically [2].
3. Measurement challenges and why estimates diverge
Researchers caution that survey nonresponse, undercounts, and methodological choices produce wide variation in estimates: Brookings warns CPS-based inferences about foreign-born declines may overstate changes because of survey nonresponse affecting immigrants disproportionately, and the Census Bureau itself cautions against using some survey series to estimate foreign-born totals uncritically [6] [3]. Migration flows, removals, voluntary departures, and policy shifts further complicate year-to-year tallies; the Congressional Budget Office models sizable reductions to net immigration tied to enforcement changes and counts future removals and voluntary departures in its projections, showing policy can change population trajectories in measurable ways [7] [8].
4. Politics, proposals, and the limits of large-sounding claims
Political actors sometimes argue that counting undocumented residents meaningfully skews representation—Newsweek summarized claims that excluding them could alter apportionment and influence elections, but also noted the wide range of population estimates and legal constraints that keep the Census practice intact [9]. Academic work tempers alarmism: even with millions of undocumented residents, the historical effect on apportionment has been negligible in terms of changing party control, though local impacts and public perceptions can be large [2] [9].
5. Bottom line: yes, but context matters
Undocumented migrants do affect census population totals because the Census counts residents regardless of status and researchers use ACS/CPS data to estimate unauthorized populations, so their presence changes raw numbers and local distributions [1] [3]. Yet the practical downstream effects—on congressional seats, Electoral College votes, and long-run demographic projections—have been modest historically and are highly sensitive to measurement methods, policy-driven removals or departures, and the particular local geographies involved [2] [7]. Where reporting errs is in conflating “affecting the count” with “decisively reshaping apportionment or national politics”; both are possible in theory but not strongly borne out by peer-reviewed apportionment analyses to date [2].