Citizenship and the Franchise

Executive summary

In the United States, the right to vote in federal elections belongs to citizens. Federal law already makes it a crime for a noncitizen to vote in a federal election. Yet the system that enforces that rule leans almost entirely on self-attestation rather than proof: the national registration form asks applicants to check a box swearing they are citizens, under penalty of perjury, but generally requires no documentary evidence. At the same time, voter-list maintenance is uneven across the states, and a growing number of localities have begun opening local elections to noncitizens.

NCI’s position is that the tie between citizenship and the franchise should be protected in a concrete, administrable way: documentary proof of citizenship for federal voter registration, reliable and privacy-protective list maintenance using authoritative records, and keeping the vote — federal, state, and local — reserved to citizens. We hold, equally firmly, that these reforms must be designed so that no eligible citizen is turned away — with free, accessible ways to document citizenship and a notice-and-cure process before any removal from the rolls. This is a companion to our paper on birthright citizenship: both concern what it means, in practice, to be a citizen.

This is a work of legal and policy analysis and advocacy. We state the strongest version of the opposing case, and we distinguish what the sources establish from what we conclude.

Background

Citizenship and the vote. For most of American history the franchise has been tied to citizenship, though the link has not been absolute: in the nineteenth and early twentieth centuries a number of states and territories allowed “declarant aliens” — immigrants who had formally declared their intention to naturalize — to vote. That practice ended by the 1920s, and today citizenship is a settled requirement for voting in federal elections.

Current federal law. It is a federal crime for a noncitizen to vote in an election for President or Congress, subject to narrow statutory exceptions. Only citizens may lawfully cast such a ballot. States administer elections and set voter qualifications, but Congress has authority over the “Times, Places and Manner” of congressional elections (with related, narrower authority over presidential elections), and a series of constitutional amendments protects the franchise from specific kinds of denial.

How registration works. Under the National Voter Registration Act of 1993 (“Motor Voter”), the federal registration form requires applicants to attest to citizenship under penalty of perjury. It does not, as a general matter, require documentary proof. Automatic and motor-vehicle registration can compound the gap where a state issues driver’s licenses to noncitizens and its systems are not carefully partitioned.

Analysis

The verification gap

The core problem is not that the law permits noncitizen voting in federal elections — it does not. The problem is that the safeguard is an oath, not evidence. An attestation deters and punishes after the fact; it does not prevent an ineligible registration at the point of entry, and it places the full weight of integrity on perjury enforcement that rarely happens. A citizenship requirement that is never verified is, in practice, a requirement honored on trust.

What the Supreme Court has said

In Arizona v. Inter Tribal Council of Arizona (2013), the Court held that the NVRA requires states to “accept and use” the federal registration form, and that Arizona could not, on its own, add a documentary-proof-of-citizenship requirement to that federal form. The path the Court itself identified was administrative: a state may ask the federal Election Assistance Commission to add a proof requirement to the form’s state-specific instructions, and seek judicial review if the EAC refuses. Some states later tried a different workaround — maintaining a separate, proof-required roll for state elections — but that dual-roll approach was a state invention, not the Court’s suggestion, and Kansas’s version was ultimately invalidated in federal litigation. The practical lesson is important: a durable documentary-proof requirement for federal registration runs through Congress, not around it — which is why we favor federal legislation rather than a patchwork or an executive shortcut. Events have confirmed the point: a March 2025 executive order directing documentary-proof changes to the federal form was promptly and preliminarily enjoined in federal court — the same lesson Trump v. Barbara taught on birthright citizenship. The lawful road is legislation.

One caveat for completeness: the NVRA’s “accept and use” mandate binds the states it covers. A handful of states are exempt from the NVRA, and some — New Hampshire in 2024, Wyoming in 2025 — have enacted their own proof-of-citizenship registration laws on that footing. Those experiments matter, but they cannot reach the federal form in the covered states; only Congress can.

Local noncitizen voting

Separately, several localities have extended the local franchise to noncitizens — for municipal offices or school boards. Some of these measures have been enacted; others have been struck down by state courts as inconsistent with state constitutions. Whatever their legality in a given state, we regard the expansion as a mistake: it detaches the vote from citizenship at the level closest to voters, and it forces election officials to maintain two overlapping electorates — a citizens-only federal roll and a mixed local roll — inviting error.

The tools already exist

Verification and list maintenance need not be guesswork — but they are not turnkey either. Authoritative data sources — the Social Security Administration, the Department of Homeland Security’s SAVE system, and state motor-vehicle and vital-records databases — can confirm citizenship for many registrants and flag records for review. Their limits matter. SAVE was historically limited to immigrants and naturalized citizens; a 2025 overhaul removed per-query fees, added Social Security number search, and integrated SSA data with the stated aim of covering native-born citizens as well. How complete and accurate that expanded coverage proves in practice remains contested — which is why matching programs still produce false flags and why due process in the design is essential. The task is to use these tools lawfully, accurately, and with notice-and-cure — not to invent new surveillance, and not to pretend the data is cleaner than it is.

What we propose

• Documentary proof of citizenship for federal registration, enacted by Congress — the approach of the SAVE Act (no relation to DHS’s SAVE database), which passed the House in 2025 and awaits Senate action — paired with free, accessible ways for citizens to document their status so the requirement does not become a barrier.
• Rigorous, privacy-protective list maintenance using authoritative records, with notice-and-cure: any voter flagged for removal must receive notice and a fair chance to confirm eligibility before being removed.
• Reserve the franchise to citizens at the federal, state, and local levels; oppose noncitizen voting in state and local elections.

Counterarguments

We take the opposing case seriously.

• Noncitizen voting is rare. Audits and studies generally find very few confirmed instances of noncitizens voting. On this view the risk is small and does not justify new requirements that could burden eligible voters.
• Proof requirements can disenfranchise citizens. Millions of eligible Americans do not hold a passport, and some — the elderly, the poor, rural residents, and married people whose names differ from their birth records — can struggle to produce documentary proof. Kansas’s documentary-proof law was struck down in part on evidence that it had blocked tens of thousands of apparently eligible citizens from registering.
• Federalism. A locality choosing to let noncitizen residents vote in its own local elections is, on one view, a legitimate local choice about local self-government.
• Administrative cost and error. New verification and matching programs cost money and can produce false matches that flag citizens as suspect.

NCI’s response, offered as our judgment: the integrity of citizen self-government is worth protecting before a problem becomes large, not only after. But the disenfranchisement concern is real and must be answered in the design — free document access, generous cure periods, safe-harbor for good-faith registrants, and accuracy standards for any data match. Those are engineering requirements, not reasons to leave the citizenship requirement unverified. On local voting, we think the value of a citizen franchise outweighs the case for local experimentation.

Sources

• U.S. Const. art. I, §§ 2, 4; amends. XIV, XV, XIX, XXIV, XXVI.
• 18 U.S.C. § 611 (voting by noncitizens in federal elections; narrow statutory exceptions).
• National Voter Registration Act of 1993, 52 U.S.C. § 20501 et seq.
• Help America Vote Act of 2002, 52 U.S.C. § 20901 et seq.
• Arizona v. Inter Tribal Council of Arizona, 570 U.S. 1 (2013).
• Fish v. Schwab, 957 F.3d 1105 (10th Cir. 2020) — invalidating Kansas’s documentary-proof requirement.
• U.S. Dept. of Homeland Security, Systematic Alien Verification for Entitlements (SAVE) program.

Full primary texts and opinions are collected in the Research Library.

Further reading

• Birthright Citizenship and the Jurisdiction Requirement — the companion paper.
• The Citizenship Primer — a plain-English starting point.
• Legislation Tracker and State Tracker — measures on citizenship and elections.