By Elizabeth Lu

For most Californians, voting by mail has become something akin to a routine. A ballot arrives. Voters fill in the bubbles, sign the envelope, and send it back. On November 3, voters will decide whether to add a step: proving who they are, again. Proposition 39 would institutionalize a voter-ID mandate, writing this mandate into the state constitution, at the polls and by mail. Its backers pitch it as common sense, and in the abstract, it sounds like it.

But election law never stays abstract for long. Eventually, it reaches an actual voter standing at a polling place or filling out a ballot at their kitchen table. Too often, it lands on particular groups of people. In California, a striking number of that group are immigrants who earned their citizenship, or their children, the first in their families to hold a ballot. For these voters, Proposition 39 is less a safeguard than a challenge: new paperwork, unforgiving fine print, and a citizenship check that leans on government data with a documented habit of doubting exactly who they are.

What does Proposition 39 actually ask?

Today, California confirms a voter’s identity at registration with a driver's license, state ID, or partial Social Security number, and anyone who lies about citizenship on the form commits perjury. Proposition 39 would add three new checks.

At the polls, voters would have to present a government-issued ID. By mail, they would have to write the last four digits of the ID number they designated at registration; per the official ballot summary, a different number wouldn't count, even one from another legitimate ID. And election officials would have to make "best efforts" to verify every registered voter's citizenship against government data, reporting yearly on their progress.

To its credit, the measure promises a free voter ID card to anyone who asks. Yet it leaves the hardest questions to the Legislature: which IDs count, and what "best efforts" even means. The nonpartisan Legislative Analyst's Office pegs the cost at tens of millions to low hundreds of millions of dollars a year.

Who carries the burden?

California is home to about 5.6 million adult naturalized citizens, roughly 21 percent of its eligible voters and the highest share of any state. Add their U.S.-born children, and rules affecting naturalized voters are hardly tangential to election policy; they sit near its center.

Proposition 39 never distinguishes between a naturalized citizen and someone born in, say, Fresno. The law may claim to treat everyone the same on paper, but that does not mean everyone will experience such policies the same way.

For example, looking at citizenship verification, a naturalized citizen may have gotten a Social Security number as a green card holder, taken the oath years later, and never had a reason to update that record. Proposition 39 never specifies which databases California must use, which leaves those decisions up to the decisions of future legislators and their future implementation. However, existing systems already illustrate the potential risks. The federal SAVE system, which other states have used to check citizenship, has demonstrated significant inaccuracies. ProPublica found that in Boone County, Missouri, more than half the voters SAVE initially flagged as potential noncitizens were in fact citizens; in Texas, reporters found at least 87 wrongly flagged citizens across 29 counties. While being flagged does not automatically strip someone of their voting rights, it can trigger additional verification barriers and, depending on state procedures, voting restrictions. Secretaries of state have warned that the system risks mislabeling naturalized citizens and married women who changed their names.

The mail-ballot rule adds another point of failure. A voter who registered using their Social Security number but mistakenly writes the last four digits of their driver’s license number would have their vote rejected, even though both numbers belong to the same person. Under Proposition 39, the issue is not simply voter eligibility, but whether or not the number a voter provides matches the specific designated number at registration. One wrong digit, or the right digits from the wrong card, would turn an innocuous clerical slip into an uncounted vote.

Linguistic barriers compound all of the above effects. About 88 percent of California's immigrants speak a language other than English at home, and roughly half speak English less than "very well." For voters already navigating such language barriers, these additional requirements makes an already complicated process even more difficult. A voter may have to remember which ID they used years earlier, determine why two government records might not match, and understand who to contact and how to remedy the situation if anything goes wrong.

That's why the promised free ID card, while welcome, does not settle the matter. The barrier is not simply the cost of abstaining identification, but the requirements and complications that accompany the introduction of this new system. For first-generation voters raised in mixed-status households, these mandates may also deepen mistrust towards government institutions. As one CalMatters commentary noted, some are already hesitant to share personal information with the government. The most worrying effect may be harder to measure: the eligible voter who sees extra requirements and barriers, worries about getting something wrong, and chooses an easy way out: not to bother.

So, why do some say "yes"?

Supporters deserve a fair hearing. In an era of eroded trust, their logic is intuitive: if we show ID for countless other government interactions, why not voting? The Yes campaign notes that more than 100,000 California mail ballots were rejected over signature mismatches in the last election, and says voter ID states haven't seen participation drop.

Those arguments do indeed hold merit, but they do not answer the structural flaws in the measure itself. For example, a digital-matching rule, a system where the government replaces a signature check with an ID-number check, does not eliminate mistakes. The Common Cause warns that if the measure sets no guidelines for curing a flagged ballot, that new system does not replace or eradicate the issue; but instead, creates a different one.

What should belonging actually look like?

Proposition 39, upon its potential enactment, may not impose an insurmountable obstacle to voting on its own. However, it is the accumulation of seemingly minor barriers that ultimately discourages voter engagement as a whole. Another document, another number, another database, another chance for a mistake—each worry erodes democratic participation, step by step.

If California wants tighter identity checks, lawmakers could write them by statute, with a clear list of accepted IDs, multilingual outreach, and a guaranteed chance to fix a ballot before it's tossed. Proposition 39 would instead etch a system into the state constitution before its blueprint has even been drawn out. A system built to protect the ballot should be judged not only by whether it catches an ineligible voter, but by what happens when it mistakes an eligible one for the same.

Naturalized citizens have already done something extraordinary: they went through the process of becoming Americans. California should be careful about building a voting system that makes them feel as though their citizenship is still being questioned. In the end, a state that claims to be confident in its newest voters should not continually ask them to prove that they belong.