An eligible citizen can submit a genuine ballot and still lose their vote because of a paperwork error. That is the central problem with California’s Proposition 39. It would make additional government-identification requirements a constitutional condition of voting, creating new opportunities to reject ballots that are not fraudulent.
The measure would require government-issued identification for in-person voting and, for mail voting, the last four digits of an identification number matching the identification designated in the voter’s registration record. It would also require state and county officials to make “best efforts” to verify citizenship using government data. These are separate requirements, with separate risks such proving who a voter is, matching a ballot to a registration record, and checking citizenship are not interchangeable tasks.
Supporters dress Proposition 39 up as “election security” and “public confidence,” but those slogans should not obscure that the measure gives the state more reasons to reject eligible voters’ ballots. Whatever supporters claim to want, the measure should be judged by whom it excludes, what additional protection it actually delivers, and whether less burdensome safeguards could accomplish the same thing. Calling a barrier a safeguard does not make it one.
California already verifies voters
Contrary to Donald Trump’s wild non-citizen voter claims, California does not currently operate on the principle that anyone can vote without establishing identity or eligibility. The registration process uses a driver’s license number, state identification number, or the last four digits of a Social Security number to establish identity. Voters who do not provide that information may be asked for identification the first time they vote. False statements about citizenship during registration constitute perjury, and noncitizen voting in state or federal elections carries criminal penalties. Mail-ballot signatures are verified before ballots are counted.
Proposition 39 would place another requirement on top of that system. New and existing voters would have to designate government-issued identification for mail voting and then supply the corresponding last four digits on their ballot envelopes. The measure provides that the designated identification type must appear on the envelope and be available by telephone or electronically—a useful protection, but not one that eliminates mistakes, outdated records, or difficulties obtaining help.
Consider a voter who is properly registered, signs their ballot envelope, and writes the last four digits of their Social Security number when their registration designates a driver’s license. Both numbers may genuinely belong to that voter. The mismatch nevertheless fails the proposition’s prescribed matching requirement. The question then becomes whether the voter receives timely notice and a realistic opportunity to correct the mistake, and not whether the voter is actually eligible.
That distinction matters because a paperwork discrepancy is evidence of a discrepancy, not necessarily evidence of impersonation.
Number-matching rules have already caused real ballot losses elsewhere
The risk is not merely hypothetical. Texas adopted mail-voting identification requirements that required an identification number on the application and return envelope to match information in the voter’s registration record. In the March 2022 primary, an Associated Press analysis reported by NPR found that nearly 23,000 mail ballots were rejected across the majority of Texas counties. Election officials identified missing identification information and identification-number problems as major reasons.
Later research described by Votebeat found that approximately 30,000 voters—about one in seven people who started the mail-voting process in that primary—had an application or ballot rejected. Roughly 90% did not find another way to participate in that election. Those figures include both application and ballot rejections; they should not be described as 30,000 rejected ballots.
California’s measure is not identical to Texas’s law, and those results do not establish how many California voters would lose their votes. They do establish the mechanism of harm though: a matching requirement can transform routine mistakes into lost participation, even when voters have identification and attempt to follow the rules.
Voting by mail would not avoid Proposition 39’s identification burden. It would merely move that burden from the polling place to the envelope.
“Free voter ID” does not answer the access problem
Proposition 39 requires California to provide a voter identification card at no charge to eligible voters who request one. That provision deserves acknowledgment. But the price of the finished card is only one part of the burden. The Legislative Analyst notes that the measure leaves important implementation details unresolved, including which identification documents would qualify and the design and requirements of the new voter card.
A meaningful access assessment must therefore ask:
- What documents would someone need to establish identity and obtain the free card?
- How would a voter reconcile different names across records?
- Could someone obtain the card without an in-person trip?
- What would happen if supporting records were unavailable or government processing was delayed?
- Would there be a workable alternative for someone unable to obtain acceptable identification before an election?
These are not peripheral administrative questions. They determine whether the promise of a free card translates into a usable right.
The Supreme Court’s voter-ID decision itself recognized that obtaining identification can impose heavier burdens on people who have difficulty obtaining birth certificates, assembling documents, or overcoming economic limitations. Although the Court upheld Indiana’s law against the challenge before it, that acknowledgment undermines the simplistic claim that a free card necessarily means a burden-free system.
Why transgender voters face particular risks
For transgender people, the problem is often not the absence of identification. It is the gap between a person’s current name, appearance, or gender and what government documents say.
The Williams Institute’s 2026 analysis estimates that 323,700 transgender citizens in states with voter-ID laws may face voting barriers because they lack identification reflecting their correct name and/or gender. Among transgender adult citizens responding to the 2022 U.S. Trans Survey, 49% reported having no identification with their correct name, and 54% reported having no identification reflecting their correct gender. These are national findings, not a prediction of how many Californians Proposition 39 would disenfranchise.
A valid identity can look inconsistent on paper
A transgender voter might update their voter registration after a legal name change while another agency still holds an earlier name. Their identification photograph may predate transition. Different documents may be updated at different times.
Those situations can make identification checks more difficult without making the person ineligible. The Williams Institute identifies inaccurate names, gender markers, and identification documents as obstacles transgender people encounter under voter-ID systems.
Proposition 39 should not be mischaracterized as expressly requiring a particular gender marker, requiring a photograph on every acceptable document, or mandating exact name matching. Its text requires government-issued documentation allowing “conclusive verification” of identity, while leaving the accepted forms of identification to implementation. The danger lies partly in that unresolved standard: how would officials distinguish an ordinary discrepancy from a reason to withhold a ballot?
Identification checks can become unwanted scrutiny of transition
Requiring a voter to present an outdated document can also expose a former name or a gender marker that does not reflect their identity. An identification system can consequently put a transgender person in the position of explaining personal history simply to establish that they are the registered voter.
That is a foreseeable privacy and access risk—not a claim that the proposition expressly authorizes intrusive questioning. The relevant policy question is whether implementation would clearly prevent officials from treating gender presentation, an outdated photograph, or a gender-marker discrepancy as evidence of fraud. National research on transgender voters shows why those protections matter: documents that do not reflect a person’s gender create obstacles to participation, including voting.
A citizen should not have to complete every stage of updating government records before exercising the franchise.
Prop 39 disproportionately burdens people of color, young adults, voters with disabilities, rural and lower income voters
Proposition 39’s burdens would not stop with transgender voters or naturalized citizens. People of color and low-income voters are also disproportionately likely to lack current identification. A nationally representative survey conducted in 2023 found that citizens of color were 3.7 times more likely than white adult citizens to lack an unexpired driver’s license or state ID card; younger and lower-income adults were also more likely to lack ID or have documents that could create voting difficulties. Those findings do not tell us exactly how many Californians Proposition 39 would exclude, but they demolish the assumption that everyone begins with the same paperwork in their wallet. A formally equal requirement can produce deeply unequal access.
And “just get an ID” is not an answer to poverty or geography. For a voter without reliable transportation, obtaining identification can mean reaching a distant government office during limited business hours and securing whatever supporting records the process requires. Rural voters can face particularly difficult travel barriers; low-income voters have fewer resources to absorb the costs. Proposition 39 promises a free voter card, but leaves important details about obtaining it unresolved. The card may be free while the process is not. California should not make the practical ability to navigate government paperwork a substitute for the legal right to vote.
Mail voting reduces some exposure, but not the paperwork burden
The mail-voting provision does not require voters to display their gender marker on the ballot envelope. Its immediate requirement is numerical. That distinction is important.
Nevertheless, mail voting would still require transgender voters, like everyone else, to designate an identification source and ensure that the number supplied matches their registration record. Where a person is already reconciling records after a name change, an additional matching obligation adds another task and another potential point of failure. The general danger is number mismatch; the trans-specific burden is the wider context of documents that may already be inconsistent or difficult to update.
Citizenship verification creates a different risk for naturalized citizens
Proposition 39 also directs election officials to use government data to make “best efforts” to verify citizenship attestations and report the percentage of each county’s voter rolls that has been verified. It does not specify exactly what those efforts must entail; the Legislative Analyst identifies that as a matter requiring legislative clarification.
For naturalized citizens, the critical implementation question is how the state would handle records that do not establish current citizenship. A government record reflecting a person’s earlier immigration status would not, by itself, establish that the person remained a noncitizen after naturalization.
The measure should not be described as automatically removing everyone whose citizenship cannot be verified: its stated verification requirement does not itself establish that outcome. But the distinction between “not verified” and “not a citizen” must govern any implementation. Otherwise, uncertainty in government records could become a demand that eligible citizens repeatedly prove what the state’s data fail to show. The proposition’s unresolved verification procedures make that a substantive concern, not a settled prediction.
What voter-ID laws do—and what they do not establish
“Voter fraud” is too broad a category to justify every identification restriction without further analysis. In-person impersonation, false registration statements, improper handling of mail ballots, and election-administration failures are different problems.
In Crawford v. Marion County Election Board, the Supreme Court’s controlling plurality explained that Indiana’s photo-ID requirement addressed in-person voter impersonation. It also acknowledged that the record contained no evidence of that fraud occurring in Indiana. The Court nonetheless accepted election integrity, modernization, and public confidence as legitimate interests sufficient to defeat the facial challenge on the record presented.
That decision supports two conclusions that should remain distinct. First, a state need not wait for a major fraud scandal before adopting safeguards. Second, the legal permissibility of a safeguard does not demonstrate that it is effective, necessary, or well designed.
For Proposition 39, the policy question is the marginal benefit. What additional misconduct would these requirements prevent beyond California’s existing registration checks, signature verification, and criminal prohibitions and how many eligible voters would encounter new obstacles in exchange? The state’s existing safeguards are part of the baseline, not something the measure’s supporters can simply omit from the comparison.
The legal issue is burden and justification—not whether all ID laws are unconstitutional
A serious legal critique should not claim that voter identification is categorically unconstitutional. Crawford, 553 U.S. 181 (2008), upheld Indiana’s particular law against a facial challenge. Under the Anderson–Burdick framework discussed in that decision, courts weigh the injury to voting rights against the state’s asserted interests; severe restrictions require more demanding justification than reasonable, nondiscriminatory regulations.
But Crawford was not blanket approval of every identification regime. The plurality emphasized the evidentiary shortcomings of the challenge and the difficulty of invalidating the law in all its applications. Indiana’s law also contained provisional-ballot procedures and exceptions, and its photo-ID requirement did not apply to mailed absentee ballots.
Proposition 39 reaches both in-person and mail voting. Its text says officials may count a “regular or provisional ballot” only after identity verification under its requirements. It does not itself spell out a comprehensive correction process for identification-number errors. That does not establish that later legislation could never provide one; it means the constitutional mandate comes before important protections are defined.
Accordingly, the strongest legal concerns would turn on actual design and evidence:
- Whether voters receive prompt notice and a practical opportunity to correct mistakes.
- Whether identification can be obtained without prohibitive documentary or logistical burdens.
- Whether name and appearance discrepancies are resolved consistently.
- Whether affected voters can establish identity through genuinely accessible alternatives.
- Whether the additional burdens are justified by the interests served.
Those questions matter particularly for an as-applied challenge involving voters who can show concrete harm. They also matter before passage, when voters must decide whether to constitutionalize a system whose safeguards remain incomplete.
The financial and constitutional costs compound the access problem
The Legislative Analyst estimates annual state and local implementation costs ranging from tens of millions to the low hundreds of millions of dollars. Those costs would depend on future decisions about administration and identification cards.
Cost alone does not make an election safeguard objectionable. But spending at that scale should purchase a demonstrable improvement and not merely an additional reason to reject genuine ballots and suppress votes.
The constitutional form also matters. Proposition 39 would place the central identification mandate in California’s Constitution while leaving substantial operational decisions to the Legislature. The Legislature could implement the requirement, but it could not simply treat that constitutional command as an ordinary statute to discard if experience showed that it was unnecessarily excluding eligible voters.
The objection, ultimately, is not to accurate identification. It is to a system that risks confusing documentary compliance with democratic entitlement. Election integrity requires excluding fraudulent votes, but it also requires counting the votes of eligible citizens. A rule that achieves the first objective poorly while undermining the second is not a neutral improvement. It is a barrier dressed as a safeguard.
Who is funding Yes on Proposition 39?
The table below identifies the major funders backing Proposition 39. Contribution amounts come from the California Fair Political Practices Commission’s top-contributor list, last updated October 13, 2026, except Steven Bray’s financing, which is separately reported by CalMatters. The recipient committee is “Californians for Voter ID, Yes on 39,” committee ID 1479607
| Funder | Reported contribution or financing | Exactly who they are |
|---|---|---|
| Richard Uihlein | $17,000,000. fppc.ca | Co-founder of Uline, the Wisconsin-based packaging and shipping-supply company, and a major Republican and conservative political donor. CalMatters reports that he and his wife have also poured millions into the conservative Restoration of America PAC. |
| Joseph N. “Joe” Gebbia | $8,000,000. fppc.ca | Airbnb co-founder and, according to CalMatters’ October 7 reporting, a current White House official. The FPPC lists Texas as his contributor state. This is a contribution under his name, not a listed contribution from Airbnb. |
| Reform California with Carl De Maio — Ballot Measure Committee | $3,103,479.50 in reported contributions. fppc.ca | The political committee associated with Republican Assemblymember Carl DeMaio, who is spearheading Proposition 39. Reform California funded the signature-gathering operation that qualified the measure for the ballot. |
| Steven Bray | More than $3 million in loans, plus $50,000 donated outright, according to CalMatters’ October 7 report. calmatters | Southern California businessman. Power Plus identifies Steve Bray as its founder and CEO; the company supplies temporary power and emergency-response services. His loans should be distinguished from outright donations. |
| Nicole Shanahan | $370,000. fppc.ca | Attorney and Robert F. Kennedy Jr.’s former presidential running mate; also the former wife of Google co-founder Sergey Brin. |
| Douglas Leone | $250,000. fppc.ca | Silicon Valley venture capitalist and former managing director of Sequoia Capital. He was also a contributor to the Trump-aligned America PAC in 2024. |
| Stark Industrial LLC | $250,000. fppc.ca | An entity listed by the FPPC with a Texas address. |
| Cameron Winklevoss | $250,000. fppc.ca | Co-founder of cryptocurrency exchange Gemini and a political donor who supported Trump-aligned fundraising in 2024. |
| Tyler Winklevoss | $250,000. fppc.ca | Gemini co-founder, Cameron’s twin brother, and a political donor who likewise supported Trump-aligned fundraising in 2024 |
| Word & Brown | $230,000. fppc.ca | Orange County-based health-insurance general agency founded by John M. Word III and Edward J. “Rusty” Brown Jr. Its business distributes health plans through insurance brokers. |
| Aubrey Chernick | $150,000. fppc.ca | Software entrepreneur who founded Candle Corporation in 1976, initially developing software to manage IBM operating systems. |
Follow the money, and Proposition 39 looks less like a neutral administrative fix than a well-financed extension of the right’s election-fraud politics. Its major backers include Republican megadonor Richard Uihlein, Carl DeMaio’s Reform California committee, and other donors associated with Trump-aligned causes. Not every contributor can accurately be labeled a Republican, but the political direction of the campaign is clear. These are not disinterested referees simply volunteering to improve ballot-envelope design.
A vote for Proposition 39 is a vote for Donald Trump. It gives his politics of suspicion, alternative facts and racist fearmongering a foothold in California’s Constitution by treating voters as potential impostors and answering exaggerated claims about noncitizen voting with more hurdles for eligible citizens. Trump’s first voter-fraud commission failed to substantiate his claims of widespread fraud, and subsequent audits have not established the vast noncitizen electorate his rhetoric conjures. Yet the proposed response keeps landing on real voters—including naturalized citizens, people of color and transgender people whose government records may be outdated or inconsistent. California should not constitutionalize paperwork barriers to appease a racist political narrative that casts immigrants as electoral threats. The appropriate response to proven misconduct is enforcement, and the appropriate response to unsupported fear mongering is rejection. Vote No on 39.