bilateral
TopicsAbout← Back to feed
BySRSam Reyes·CMCal Morrow·EQEliza Quinn·DPDana Park
ANALYSISApril 13, 2026

Should the United States adopt automatic voter registration?

As of November 2025, 24 states and Washington D.C. have enacted automatic voter registration (AVR), which shifts voter registration from an opt-in to an opt-out process when citizens interact with government agencies. At the federal level, competing legislation is advancing: the Republican-backed SAVE Act passed the House 220–208 on April 10, 2025, and would require in-person documentary proof of citizenship to register, effectively curtailing AVR; the Democratic-backed Register America to Vote Act of 2025 (S.2822) would mandate AVR nationwide. Both bills face significant obstacles in the Senate, where a 60-vote threshold applies.

XLinkedInFacebookThreadsWhatsAppReddit

If democracy works best when more people vote, why does one side of the aisle fight to make registration easier — and does the other side have a legitimate point about who gets added to the rolls without asking?

Not familiar with this story? Get context →
Oregon's 42% turnout gap
C
Oregon's 2018 data is the most honest test AVR has produced: automatically registered voters turned out at 42%, versus 70% for all registered voters. That is not a cohort effect or a rounding error — it is evidence that passive enrollment produces a much larger pool of largely disengaged voters. AVR optimizes for the appearance of democratic participation, not its substance.
L
The 42% figure measures people in their first electoral cycle after passive enrollment — and first-time registrants under any system vote at lower rates. The real question is whether those voters' participation rates converge over subsequent cycles, which Oregon's longer-term data suggests they do. A voter who turns out once every four years still exercised a right they would have been denied entirely under an opt-in system that failed three million Americans in 2008 alone.
C
Pointing to longer-term convergence is speculative; the data we actually have shows a 28-point gap. And 'better than nothing' is a low bar for a federal mandate that overrides 26 states' legitimate policy choices.
L
A 28-point gap that closes over time is an argument for civic engagement programs alongside AVR — not against expanding access in the first place. You cannot close a participation gap for voters who were never registered.
SAVE Act's documented citizen disenfranchisement
C
The Kansas case is genuinely difficult, and conservatives cannot wave it away: 31,000 eligible U.S. citizens blocked from registration — roughly 12% of all applicants — before the requirement was struck down. A conservatism serious about individual liberty must reckon with that number. The better argument for the SAVE Act is not that documentary requirements are cost-free, but that the principle of affirmative enrollment has weight independent of any single state's implementation failures.
L
The problem is that 'implementation failures' is doing enormous work in that sentence. When 12% of eligible applicants are systematically blocked, that is not a bug — that is the requirement functioning exactly as designed. And Utah's audit of two million voters found exactly one noncitizen registration and zero noncitizen votes, which means the SAVE Act trades a documented harm for an undocumented one.
C
The honest conservative response to Kansas is not to defend that outcome but to argue for better administrative design — targeted verification, federal database integration — rather than abandoning the principle that citizenship should be confirmed before enrollment.
L
That is a reasonable principle, but the SAVE Act is not targeted verification — it is a blanket documentary requirement that Kansas already proved disenfranchises eligible citizens at scale. Defend the principle by designing something that doesn't replicate Kansas.
Whether AVR solves a real problem
C
The Motor Voter Act of 1993 was the appropriate federal response to registration barriers: create accessible opt-in opportunities at government contact points without converting citizenship into automatic enrollment. If the problem is administrative failure in the existing system, the solution is fixing administration — not replacing the opt-in principle that signals affirmative civic engagement.
L
Motor Voter has been law for thirty years, and a Caltech/MIT study still found three million Americans unable to vote in 2008 due to registration failures. Oregon's AVR quadrupled DMV registration rates in its first year and enrolled 116,000 people who fell through the cracks Motor Voter was supposed to close. The opt-in system had its chance to solve this problem.
C
Quadrupled registration rates and 116,000 new enrollees sound impressive until you recall that 42% of them didn't vote — which raises the question of whether the 'crack' they fell through was administrative or motivational.
L
Those 116,000 people could vote when something moved them to; the question of motivation is moot for anyone who was never on the rolls.
Federal mandate versus state experimentation
C
Senator Klobuchar's Register America to Vote Act would permanently centralize election administration authority that the constitutional architecture deliberately distributes. Twenty-six states have not chosen AVR — that is not dysfunction, that is federalism working. Nationalizing AVR eliminates the policy variation that lets us actually evaluate competing approaches.
L
Twenty-four states and D.C. have already moved to AVR, including Delaware, Minnesota, and Pennsylvania as recently as 2023. When nearly half the country has adopted a practice and produced measurable results — Vermont saw a 62% registration increase in six months — a federal baseline that matches those results isn't centralization, it's standardization of what works.
C
Standardizing 'what works' assumes the results in Oregon and Vermont transfer uniformly across states with different demographics, administrative capacities, and political cultures — an assumption the federalist argument exists precisely to test.
L
The states still administering those systems — with their own demographics and capacities — are the test. Twenty-four jurisdictions running the experiment simultaneously is more data than any single federal pilot could produce.
Noncitizen voting as justification for SAVE Act
C
The strongest version of the SAVE Act argument is not about fraud statistics — and conservatives who make noncitizen voting the centerpiece are standing on sand. Utah reviewed over two million registered voters and found one confirmed noncitizen registration and zero noncitizen votes. The real argument is about the principle of affirmative civic engagement and administrative integrity, not a crisis the evidence doesn't support.
L
That concession matters, because the SAVE Act passed the House 220–208 primarily on fraud-crisis rhetoric, not on the principled administrative-integrity argument you're describing. If the honest conservative case is 'affirmative enrollment signals civic commitment,' that is an argument worth having — but it should be made without the crisis framing that the Utah data demolishes.
C
Political rhetoric and policy merit are separable. The fraud-crisis framing may be overstated, but the underlying principle — that registration should be an affirmative act, not a bureaucratic default — survives the Utah data entirely.
L
A principle that produces 31,000 blocked citizens to stop zero noncitizen voters needs more than philosophical weight to justify itself — it needs a design that doesn't replicate that outcome.
Conservative's hardest question
The Kansas data is genuinely hard to dismiss: a proof-of-citizenship requirement blocked 12% of eligible applicants, all U.S. citizens, from exercising a fundamental right. A conservatism serious about individual liberty must reckon with the fact that administrative barriers can disenfranchise far more eligible voters than they exclude ineligible ones — and that this outcome is not hypothetical but documented.
Liberal's hardest question
The Oregon 2018 turnout data — 42% for automatically registered voters versus 70% for all registered voters — is genuinely difficult to dismiss, because it suggests AVR primarily grows the rolls with less-engaged citizens rather than translating directly into participation. Critics can reasonably argue that expanding registration without addressing underlying disengagement may produce inflated voter rolls without meaningfully expanding democratic voice.
Both sides agree: Both sides accept that the Kansas documentary proof-of-citizenship requirement produced a documented harm — 31,000 eligible citizens blocked — and neither defends that specific outcome as acceptable.
The real conflict: They disagree on a values question about what registration should mean: conservatives argue enrollment should be an affirmative act signaling civic intent, while liberals argue citizenship itself is the only legitimate qualification and administrative default serves that principle better.
What nobody has answered: If Oregon's automatically registered voters do converge toward average turnout rates over subsequent election cycles, the conservative case against AVR loses its strongest empirical leg — but if they don't, the liberal case for AVR as a democratic expansion tool becomes harder to sustain than its proponents admit: which side has actually looked at Oregon's longitudinal data, and why hasn't either cited it?
Sources
  • Web search results provided: comprehensive summary of AVR debate including state adoption data, Oregon and Vermont registration statistics, SAVE Act House vote details, Register America to Vote Act (S.2822) details, Gallup polling data, Kansas proof-of-citizenship case history, and Utah citizenship review findings.
  • Caltech/MIT Voting Technology Project (2008) — cited within search results regarding 3 million disenfranchised voters.
  • Ballotpedia and Brennan Center-style policy summaries — cited within provided search results for AVR type definitions and state adoption timelines.
  • Gallup polling data on public support for AVR — cited within provided search results.

More debates