43 states have online voter registration. Why not Texas?
In most of the country, registering to vote is as easy as ordering takeout: open your phone, fill out a form, and hit submit. This is not the case in Texas, which is one of only seven states that does not allow online voter registration.
To register to vote for the first time in Texas you still have to fill out a paper application. This causes real problems for voters and taxpayers.
Ahead of elections, county offices are inundated with voter registration applications that they must process immediately. They hire additional temporary staff to process the influx at taxpayer expense. Transcribing these paper applications introduces errors, like illegible handwriting or transcription mistakes, that can result in incomplete registrations that get rejected. Counties have limited capacity to notify voters that their application was rejected, so voters often don’t realize their registration wasn’t processed until they arrive at the polls and are denied the opportunity to vote. This can be discouraging and it erodes trust in the system.
In 2026, the right to vote shouldn’t hinge on filling out a paper form. So why hasn’t Texas adopted online voter registration?
Many state lawmakers say it’s past time to adopt online voter registration. John Bucy, a state representative from Austin, has filed a bill every year since 2019.
But people who oppose online voter registration typically cite the potential for security risks and fraud. Yet there’s no evidence of fraud in the 43 states using online registration.
In the wake of the 2020 election, online voter registration has become a more partisan issue. While official statements focus on security, the underlying concern appears to be turnout. The assumption is that online registration increases registrations among young, urban, tech-savvy voters (demographics that supposedly lean Democratic) and it could change the outcomes of elections.
This assumption is rarely stated outright, but the logic is straightforward: easier registration means more registrants, and more registrants could change election outcomes. But this is not true. Online voter registration in Texas would not favor one party over the other, and we analyzed the data to back it up.
Looking to Florida’s online voter registration adoption
Florida adopted online voter registration in 2017. Like Texas, Florida is a large, diverse state with Republican-led government. This makes it a good comparison for predicting what adopting online voter registration might mean in Texas.
We analyzed voter registration data from all 67 Florida counties from 2017 through 2024. We obtained registration data by application source (including online, DMV, mail, and third-party organizations) and active registered voters by party affiliation from the Florida Division of Elections public records.
Ideally, to draw a connection between online voter registration and partisan trends, we would have data showing party registration AND registration source together. With only independent counts of partisanship and source, we had to use statistical methods to infer the relationship.
We started by looking at trends in online voter registration uptake and partisan registrations.
Online voter registration is on the rise in Florida
Since its adoption in 2017, online voter registration is the leading registration source outside of the DMV. Online voter registration accounts for nearly two-thirds of non-DMV registrations in presidential election years.
For the rest of our analysis, we excluded registrations at the DMV. Under the National Voter Registration Act, DMV registrations occur automatically during license transactions such as renewals, new licenses, or address changes. This means they largely reflect routine DMV visits and interstate movers updating their licenses, rather than active registration choices. Excluding them gives a clearer picture of how existing residents choose to register to vote.
At the same time, more Republicans registered to vote
During the same period that online voter registration usage grew dramatically, registered Republicans outnumbered registered Democrats for the first time in recent history. The gap continues to widen.
Of course, these are just trends. We can’t assume online registration caused Republican gains. To understand whether there’s any relationship, we needed to model the data.
Modeling a relationship between partisan registration and online voter registration
To establish a convincing relationship between partisanship and registration method, we need a predictive model that accounts for other factors. For example, larger, more urban counties like Miami-Dade that lean more Democratic also tend to have higher online voter registration uptake. Our model needed to account for population and demographics to isolate the effect of registration method.
We ran separate lagged dependent variable models for Democrats and Republicans. In a lagged model, we use 2017 party registration to predict 2024 party registration. This approach controls for historical partisan composition, captures persistence in voter registration patterns, and effectively models partisan change from 2017 to 2024.
We tested several demographic variables, including age distribution, internet access, and county population. These were chosen because of the assumed link between urbanization, internet access, and online savviness, which all might lead to higher likelihood of registering to vote online.
In the end, only population size had a meaningful relationship and was the only demographic variable included for simplicity. It’s likely that age and internet access play some role, but their effects are limited because counties with larger populations also have younger populations and more internet access.
In the end, our model specification that we ran for both Republicans and Democrats was:
% Party 2024 = β₀ + β₁(% Party 2017) + β₂(% non-DMV OVR) + β₃(log Population) + ε
In a nutshell, we’re asking whether online voter registration shifts partisan balance. Do counties with higher online voter registration adoption see bigger gains for Democrats? For Republicans? Or does partisan registration stay the same regardless of online voter registration uptake, once we account for county size and historical partisanship?
What drives partisan registration? Population, not online registration
Our analysis suggests that online voter registration adoption in Florida is primarily a result of county population size, and not a driver of partisan registration in either direction. Urban counties with higher populations adopt online voter registration more AND have higher Democratic registration, but when county size is accounted for, there is no relationship between online voter registration and partisan registration.
Our models explain 80% of the variation in county-level Democratic registration (R² = 0.801, F(3,61) = 82.09, p < 0.001) and 87% of the variation in Republican registration (R² = 0.866, F(3,61) = 131.7, p < 0.001). The typical prediction errors are ±4.3 percentage points for Democrats and ±4.6 percentage points for Republicans, indicating strong predictive accuracy across Florida’s diverse county landscape.
Historical partisan composition is the strongest predictor. Unsurprisingly, a county’s past partisan makeup is the best predictor of its future partisan makeup. High coefficients and highly significant p-values indicate strong persistence in partisan patterns for both Democrats (β = 0.77, SE = 0.05, p < 0.001) and Republicans (β = 0.82, SE = 0.06, p < 0.001).
Population size has a strong, significant effect in opposite directions for each party. Doubling a county’s population increases Democratic registration by approximately 3.7 percentage points (p < 0.001) and decreases Republican registration by approximately 4.4 percentage points (p < 0.001). This reflects the well-documented urban-rural divide in American politics: larger counties lean Democratic, smaller counties lean Republican.
Online voter registration shows no significant independent effect. After controlling for baseline partisan composition and population, online voter registration uptake is not significantly associated with Democratic registration (β = 0.05, SE = 0.08, p = 0.497) or Republican registration (β = -0.08, SE = 0.08, p = 0.309).
The chart below illustrates the relative effect of online voter registration and county population size on Democratic registration. Moving from low to high online voter registration usage (20% to 60%) changes predicted Democratic registration by only 2 percentage points. In contrast, moving from a small to large county (15k to 1.5M population) increases predicted Democratic registration by 17.2 percentage points—nearly 9 times larger. The pattern for Republicans is a mirror image: online voter registration has similarly minimal effect (3 percentage points), while larger county size decreases Republican registration by 20.4 percentage points.
What this means for Texas
Texas is behind the curve in adopting a policy that would benefit all Texans. Voting should be easy and encouraged, and online voter registration is a proven step in that direction.
Our analysis of Florida’s experience shows that neither party stands to lose ground if online voter registration is adopted in Texas. The partisan fear blocking adoption simply isn’t supported by the data. But the non-partisan benefits are real and measurable. Remember those problems we started with—registration errors, voters turned away at the polls, counties spending taxpayer money on temporary staff? Online voter registration solves all of them. After Florida adopted online voter registration, their application rejection rate dropped from 5-6% to just 1-2% (Election Administration and Voting Survey (EAVS). In Texas, this would mean around 800,000 fewer registration errors each cycle, significant cost savings for county offices, and a more reliable process for voters.
Forty-three states have already rolled out online voter registration without issue. It benefits election administrators, taxpayers, and the many Texans who face registration problems every year. So, what are we waiting for?