How voter registration rules discourage some Americans from voting: An explainer and research roundup

Mar 18, 2024

by Denise-Marie Ordway, The Journalist’s Resource

At first glance, registering to vote in the U.S. may seem easy: Adults fill out a form and submit it online, in person or through the mail. But getting onto a state’s voter roll — and staying there — can be complicated. It’s also a key reason more Americans don’t participate in elections. 

All states except North Dakota require voter registration for federal, state and local elections. The U.S., unlike many other democratic nations, puts the burden of registration on its citizens.

In India, the world’s largest democracy, the government creates voter rolls automatically from census data, according to a 2020 report from the Pew Research Center. America’s neighbor to the north, Canada, maintains a national database of registered voters who are added automatically based on data collected across government agencies there, including birth and death records and income tax filings.

Pew notes that half of the 226 countries and territories included in the ACE Electoral Knowledge Network have some type of registration mandate. The network, launched at the United Nations in 1998, is a global repository of election-related information.

In 2022, 69% of U.S. citizens aged 18 years and older were registered to vote, according to a report the U.S. Census Bureau released last year. Registration rates ranged from 61% in North Carolina to 83% in Oregon.

The four most populous states — California, Texas, Florida and New York — had some of the lowest registration rates: 67%, 65%, 63% and 66%, respectively.

U.S. election oversight

U.S. elections are broadly governed by federal laws such as the Voting Rights Act of 1965, the National Voter Registration Act of 1993 and the Help America Vote Act of 2002. All three aimed to broaden the number of people participating in American elections.

The National Voter Registration Act significantly expanded residents’ opportunities to register to vote. It’s commonly referred to as the “Motor Voter Act” because it requires states to allow residents to register to vote at the same time they apply for or renew their driver’s licenses. The law also mandates that residents be able to register at certain state and local offices, including public assistance offices.

There’s no national list of registered voters or people eligible to vote, however. Individual states and the District of Columbia maintain their own voter rolls. Twenty-four states and the District of Columbia share data through the Electronic Registration Information Center to help one another keep track of voters who move or die and prevent duplicate registrations.

Federal law forbids people who are not U.S. citizens from voting in federal elections, but states decide who can participate in state and local elections. While all states prohibit non-citizens from voting in state elections, a few cities in California, Maryland and Vermont allow non-citizens to vote in local elections, according to Ballotpedia.

Why voter registration rates differ

A big reason voter registration rates differ so much is because each state has its own election policies and processes, which can make registering easy or difficult. In addition, local election offices vary in how they interpret those rules, how they educate and engage with residents, and how they maintain their voter lists.

To complete this first part of the voting process, U.S. citizens must:

• Make sure they are eligible to register. Adults living in the U.S. can register to vote in federal, state and local elections if they are U.S. citizens and meet their state’s residency requirements. Although citizens cannot vote until they are 18 years old, many states allow them to register starting at age 16. Some states prohibit certain groups from voting, including people who have been convicted of a felony and adults with psychiatric and intellectual disabilities.
• Determine where, when and how to register. In most states, residents can register online any time, provided they have access to the internet, can locate the registration form for their state online and can fill out a digital document. Those who cannot or do not want to register online can do it by mail or visit a local election office in person during its business hours.
• Correctly complete a voter registration application. Lots of people make mistakes or leave out key information on their applications. Applicants in the three states that require them to answer a question about their race or ethnicity might skip it if they aren’t sure what to write. Occasionally, applicants get their birthdates, driver’s license numbers or other identification numbers wrong.
• Submit it before their state’s deadline. Voter registration deadlines vary. In Texas, for example, citizens must be registered at least 30 days before Election Day. In Alabama, they must be registered 15 days prior. It’s 10 days in New York. In 22 other states and the capital, citizens can register to vote and cast ballots on the same day, according to the National Conference of State Legislatures.
• Promptly fix any errors or provide missing information. Local election officials review voter registration applications and notify people needing to correct an error or provide additional information. If applicants don’t share their phone numbers on their registration application, election officials will reach out by mail. If issues are not resolved by the voter registration deadline, the applicant will not be added to the voter roll in time for their ballots to count.
• Check on their voter registration status. After registering to vote, it’s a good idea for residents to periodically check their registration status. Election offices remove people from the voter rolls for various reasons, which also differ by state. Officials in Massachusetts, for instance, can remove residents who don’t participate in city and town censuses. States can cancel the registrations of residents deemed to be “inactive voters” — a term defined differently across states. Voters in Wyoming can have their registrations canceled after not voting in one general election.

Each step can be a stumbling block that ultimately results in someone not being able to register, political scientists Christopher B. Mann of Skidmore College and Lisa A. Bryant of California State University Fresno write in a 2020 paper, published in the journal Electoral Studies.

“It is a cliché that ‘getting to the starting line’ is often more difficult than running the race, and this sentiment seems applicable for many American citizens when it comes to voting: The requirement to register is a costly and time-consuming obstacle to casting a ballot,” Mann and Bryant write.

Voter registration rates among demographic groups

Data from the U.S. Census Bureau demonstrates substantial disparities in voter registration rates across states and the District of Columbia. The Census Bureau provides a variety of reports and spreadsheets containing estimates of the number and percentage of people from each state who are registered to vote. It offers data broken down age, sex, race, household income, employment status and other demographic factors.

White U.S. citizens are much more likely to be registered to vote than citizens of other racial and ethnic groups. Nationally, about 71% of white adult citizens, 64% of Black adult citizens, 60% of Asian adult citizens and 58% of Hispanic adult citizens were registered to vote in November 2022, according to Census Bureau estimates.

Some other disparities:

• Older adult citizens are more likely to be registered than younger ones. While 77% of citizens aged 65 years and older were registered to vote in November 2022, 63% of citizens aged 25 to 34 and 49% of citizens aged 18 to 24 were.
• Voter registration rates differed by job status in November 2022. For example, 60% of unemployed adult citizens reporting being registered compared with 72% of self-employed adult citizens and 79% with government jobs.
• Adult citizens with lower incomes are less likely to register to vote than those with higher incomes. For example, 83% of adult citizens with family incomes of $150,000 per year or higher were registered to vote in November 2022. Meanwhile, 58% of adult citizens with family incomes of $15,000 to $19,999 were.

KFF, formerly known as the Kaiser Family Foundation, used Census Bureau data to create an interactive database that lets the public search and sort voter registration rates by race and ethnicity in even-numbered years from November 2014 to November 2022. It shows stark differences across racial groups in certain parts of the country.

A few examples:

• The voter registration rate among Hispanic adult citizens in November 2022 was lowest in Mississippi, at 23%, and highest in Minnesota, at 75%.
• In Iowa, 76% of white adult citizens were registered to vote, compared with 41% of Black adult citizens, 45% of Asian adult citizens and 58% of Hispanic adult citizens.
• In 10 states, fewer than half of Black adult citizens were registered to vote in November 2022. In 15 states, fewer than half of Asian adult citizens were.

A look at academic research

Scholars have documented problems with the voter registration process across the U.S. for decades. In recent years, they have focused on voter education and outreach as well as voter list maintenance, a process of updating the rolls of registered voters that includes removing people who have died, moved to other states or lost their eligibility to vote.

Researchers stress that registration rates would be higher if citizens understood the rules around voter registration and the consequences of not following them. However, the quality of voter education programs varies from state to state and even from election office to election office, studies conducted by Thessalia Merivaki, an associate professor in American politics at Mississippi State University, have found.

Merivaki is among a small group of researchers who study voter registration, education and outreach. Several of her most recent studies focus on Florida, probably in part because the state’s broad open records laws make it much easier to obtain data and reports there than in many other states. Two chapters of Merivaki’s 2021 book, “The Administration of Voter Registration: Expanding the Electorate Across and Within the States,” also focus exclusively on Florida election administration.

When Merivaki and her colleague, Mara Suttmann-Lea, an assistant professor of American politics at Connecticut College, studied the education and outreach efforts of Florida’s 67 county election offices, they learned they varied considerably. Those offices “enjoy significant discretion in how they engage in voter education and the resources they dedicate to these efforts,” the two researchers write in a 2022 paper published in Policy Studies.

When Merivaki and Suttmann-Lea studied states’ efforts to improve voter education after Congress passed the Help America Vote Act of 2002, they discovered some states did more than others. In fact, they found that some states’ improvement plans probably did not comply with the federal law, created in the wake of the controversial presidential election results in Florida in 2000. The legislation provided $3.9 billion to help states modernize their election equipment and processes.

Some states did not incorporate the federal law’s “core voter education provisions, particularly voting technology demonstrations, voter guides, and toll-free hotlines” in their plans for improving voter education, Merivaki and Suttmann-Lea write in a 2022 analysis in the Election Law Journal. They also note “differences in the inclusion of education materials for language minority and disabled voters, suggesting limits on compliance with existing federal laws.”

Once citizens are added to the voter rolls, they can be removed, though. Election researchers have raised questions about the effectiveness and lawfulness of states’ procedures for maintaining voter lists.

List maintenance is meant to protect against fraud and help election officials estimate the number of voting machines, ballots, poll workers and polling locations they will need on Election Day. But that process also is flawed. It can remove eligible voters en masse and with little notice.

Some residents do not realize they have been taken off the voter roll until they show up at their polling place on Election Day. As a result, some Americans who are eligible to vote either don’t vote or vote using a provisional ballot, which is kept apart from regular ballots until election officers determine whether the individual who filled it out is eligible to vote.

Recent academic research also indicates:

• Local election offices vary in the way they process voter registration applications, which seems to contribute to big differences in the percentage of applications each office rejects. When researchers studied the issue in Florida, they found that rejection rates also varied seasonally, with fewer applications being rejected in October — the cut-off there for applying to register to vote in November elections.
• In Florida, younger voters and racial and ethnic minorities were more likely to have their voter registration applications put “on hold,” meaning they needed to correct errors and provide additional information before their applications can be processed.
• Compliance with Section 7 of the National Voter Registration Act has been “spotty and variable over time and across states.”
• Some states have started automatically prompting residents to register to vote when they visit a government office. While voter registration has risen in these states, the registration rates of underrepresented groups such as Latinos and Asian Americans appear unchanged.

We elaborate on these findings below. We have summarized four peer-reviewed papers and a report from the University of Southern California that investigate these issues. We plan to update this article occasionally as new studies and data become available.

Research roundup

Voter registration challenges

Registered, But Not Quite: Processing Pending and Incomplete Registrations Thessalia Merivaki. Chapter 6 of The Administration of Voter Registration: Expanding the Electorate Across and Within the States, 2021.

The study: In this book chapter, Merivaki examines records from two county election offices in Florida to better understand why some voter registration applications are placed “on hold” and which individuals are more likely to have their applications put in this category. She compares records from 2016 from Pinellas and Polk counties, which are located in the same region of Florida and have populations of similar size.

Between Jan. 1, 2016 and Oct. 18, 2016, a total of 2,132 applications in Polk County and 3,892 in Pinellas County were placed “on hold,” meaning they were either incomplete or denied. The deadline to register to vote in the general election that year in that state was Oct. 18.

The findings: The most common reason applications were placed on hold was because of a missing or incorrect identification number such as a driver’s license number or the last four digits of the applicant’s social security number, Merivaki found. In Polk County, 55% of “on hold” registrations were deemed incomplete because of this. In Pinellas County, 31% were. The most common reason for being denied: Having a felony conviction.

Merivaki learned that individuals with “on hold” applications “were overall young, racially diverse, predominantly registering as NPAs [not having a political party affiliation] or Democrats, and slightly more male than female,” she writes. However, Black applicants were much less likely than white applicants to be placed on hold because of missing or incorrect identification information. Hispanic applicants, on the other hand, were much more likely than white, non-Hispanic applicants to be placed on hold for this reason.

Applicants who failed to disclose their race were 55% more likely to have their application put “on hold” because of missing or incorrect information than those who disclosed it.

In the author’s words: “While disclosing one’s gender and/or race is not required [in Florida] to register to vote, it offers an opportunity to assess whether undisclosed information affects the prospects of one’s application being classified as ‘incomplete’ or denied, after taking any other factors into consideration,” Merivaki writes.

Access Denied? Investigating Voter Registration Rejections in Florida Thessalia Merivaki. State Politics & Policy Quarterly, January 2021.

The study: Merivaki examines monthly voter registration reports from each of Florida’s 67 county election offices to better understand the reasons why they rejected tens of thousands of voter registration applications between January and December 2012. Florida requires voters to be registered 29 days before a general election.

The findings: Election offices, on average, rejected about 11% of all registration applications submitted that year. But rejection rates varied by month and county. Duval County, which includes Jacksonville, had the highest monthly rejection rate — it rejected 56% of applications submitted in March 2012. Duval County’s rejection rate for the year averaged 15%, compared with about 11% in Hillsborough County, where Tampa is located, and 6% in Orange County, which includes Orlando.

Merivaki discovered that county election offices were least likely to reject applications in October 2012, when the rejection rate statewide averaged 3%. Oct. 9 was the deadline to register to vote in Florida’s November elections. She writes that the lower rejection rate in October suggests “administrative issues in processing voter registration applications when the volume of voter registration applications dramatically increases in a short period of time.”

She also discovered rejection rates were lower when a larger share of applications went through organizations such as public libraries and military recruitment offices. A 10% increase in voter registration applications submitted this way was associated with a 3% reduction in the rejection rate, she finds.

In the author’s words: “Given that the influx of voter registration applications peaks during the last weeks prior to the voter registration deadline, and so do the rates of rejected voter registrations, it appears that voter registration rejections stem from administrative challenges in processing applications in short time intervals,” Merivaki writes. “However, due to the fact the rates of rejected voter registrations also increase as early as eight months prior to the voter registration closing book date may have to do with the voters’ capacity to avoid errors when completing voter registration forms.”

Voter list maintenance issues

The Racial Burden of Voter List Maintenance Errors: Evidence from Wisconsin’s Supplemental Movers Poll Books Gregory A. Huber, Marc Meredith, Michael Morse and Katie Steele. Science Advances, February 2021.

The study: This paper estimates how often voters in Wisconsin are incorrectly flagged as having moved to a new address, prompting the state to either remove them from its voter roll or start a process that can lead to their voter registration being cancelled. “It is important to understand how often these registrants did not move, and how often such an error is not corrected by the postcard confirmation process, because uncorrected errors make it more difficult for a registrant to subsequently vote,” write the researchers, Gregory Huber of Yale University and Marc Meredith, Michael Morse and Katie Steele of the University of Pennsylvania.

When Wisconsin residents change the address on their driver’s license or file a change of address through the U.S. Postal Service’s National Change of Address System, the Electronic Registration Information Center reports that information to Wisconsin. The Wisconsin Election Commission then sends postcards to people who are registered to vote asking them to confirm whether they have moved.

The findings: In October 2017, ERIC notified Wisconsin that 341,855 voters had potentially moved. Of those, 6,153 responded to the state’s postcard to confirm they remained eligible to vote at the address they had used to register. The remaining 335,702 voters were initially removed from the state voter roll.

Huber and his colleagues estimate that 4% of voters who had been flagged as suspected movers and did not respond to postcards voted in 2018 and still lived at the same address they used to register. This represents at least 9,000 people. Racial and ethnic minorities were about 4 percentage points more likely than white residents to vote at the address that ERIC had flagged as being out of date.

In the authors’ words: “Our results show why it is essential to make registrants aware if their registration is being moved to inactive status and to continue to alert these registrants to upcoming elections so that they know when and where to vote if they still reside at their address of registration.”

Complying with federal election law

Race, Poverty, and the Redistribution of Voting Rights Jamila Michener, Poverty and Public Policy, June 2016.

The study: Jamila Michener of Cornell University examines the reasons states vary in their compliance with Section 7 of the National Voter Registration Act Of 1993, which requires public assistance agencies to offer voter registration services to everyone applying for or renewing their government benefits. She looks at changes in compliance between 1995 and 2012.

The findings: Compliance with the federal law was higher when the president was a Democrat and lower when legislatures were more heavily Republican. Race played a key role in determining whether states worked to bring people with lower incomes into the election process, Michener finds.

She also finds that compliance was lower in states where a larger percentage of the population was Black. Meanwhile, she adds, that “[h]igher percentages of Black bureaucrats are associated with increased compliance while growing ranks of Latinos in state welfare bureaucracies are associated with decreases in compliance.” Michener explains that she is unsure why compliance falls under Latino leadership. But she notes that Latino bureaucrats “are likely better educated and more economically advantaged than other Latinos,” which she adds “can translate into less liberal attitudes, less of a sense of linked fate, and an increased desire to disassociate oneself from more marginal co-ethnics.”

In the author’s words: “These findings raise concerns about the political equality of disadvantaged citizens and underscore the need to scrutinize the outcomes of expansionary voting policies. Even more broadly, this research shows how the entanglement of race and poverty in a federalist polity frustrates efforts to advance participatory equality.”

Automatic voter registration

Effects of Automatic Voter Registration in the United States Eric McGhee and Mindy Romero. Report from the Center for Inclusive Democracy at the University of Southern California, 2020.

The study: This report looks at how introducing automatic voter registration affected registration rates in 11 states. In most of those states, eligible residents were automatically registered to vote based on information they provided their state’s Department of Motor Vehicles, unless they actively declined. The authors, Eric McGhee of the Public Policy Institute of California and Mindy Romero of the University of Southern California, note the report’s conclusions “are not firm” because Americans nationwide had reduced access to government agencies at the start of the COVID-19 pandemic.

The findings: While voter registration rose after states implemented AVR initiatives — starting with Oregon in 2016 — the researchers are unsure how much of the increase is a direct result of AVR. In Oregon, for example, 722,823 people who had not previously been registered were added to the state’s voter roll between January 2016 and January 2020. Another 1.2 million Oregonians already registered to vote updated their addresses.

McGhee and Romero note there is evidence, from Oregon and California in particular, that AVR “adds new registrants consistently throughout the election cycle in a way that could have lasting effects on the state’s overall registration rate.” However, AVR does not appear to improve the relative registration rates of underrepresented groups such as Latinos and Asian Americans.

In the authors’ words: “The data suggest the reform probably encourages some new people to register who would not have done so without AVR. However, the effect on overall registration is ambiguous because most AVR increases we estimate are small and the reform is still relatively new.”

This article first appeared on The Journalist’s Resource and is republished here under a Creative Commons license.