Benj Azose is an advocate for free prison communication and was part of the successful effort to make phone calls from California prisons free in 2022 (SB 1008).
This essay is part of a series on Race, Ethnicity, Technology and Elections supported by the Jones Family Foundation. Views expressed here are those of the authors.
The rates of voting for people in jails across the United States is incredibly low, and almost all of these people are eligible to vote. Although many regions in the U.S. have restricted voting rights for people with a felony conviction, a sizable number of those incarcerated are pre-trial or in jail for misdemeanors, circumstances which typically allow them to retain voting rights.
This issue disproportionately affects Black voters. As part of the legacy of Jim Crow and structural racism in law enforcement and the criminal justice system, the jail population is significantly more Black than the U.S. as a whole. The latest data from the Bureau of Justice Statistics shows that in mid-2020, Black U.S. residents were incarcerated at 3.5 times the rate of white U.S. residents (465/100k vs. 133/100k).
This inequity is compounded by the low rate of voting from jail. While reliable statistics on rates of voting from jails are limited, all available evidence indicates that almost nobody in jail votes.
Two key reasons for this de facto disenfranchisement are lack of access to voting information and a reliable way to register to vote or request an absentee ballot, both of which could be solved by increasing access to technology in prisons.
Estimating the Number of Eligible and Participating Jailed Voters
The fastidious Prison Policy Initiative (an incredible source for data related to prisons and jails in America) says “Put simply, of the approximately 746,000 individuals in jail on any given day, most have the right to vote.” But of those that are eligible, it is very difficult to determine how many exercise their right. Nearly every county across the US has its own set of rules about who can vote and how, and there’s no central resource on who can or can’t vote and who eventually does or doesn’t.
“We don’t really know what the full picture is of jail-based disenfranchisement, partly because of this very decentralization,” said MIT professor Ariel White, an expert on voting rights and race, in Time Magazine. The rate is certainly less than 5%, and some persuasive studies have it as less than 100 in multiple states (roughly .05%).
Groups and advocates focused on measuring voting rates in jails across the country have tried a number of approaches to get accurate figures, from asking boards of election for the number of ballots sent and received in jails to talking to correctional facilities and asking for their estimates to attempting to join voter records to jail registries. Though each of these methods has drawbacks, they all point in the same direction: almost nobody in jail votes.
- New York: In a 2022 paper, Madalyn Stewart at Fordham University found that “Among an estimated voting-eligible jail population of 5,036, this research found only 26 individuals that cast ballots for the 2020 election while in custody. This equates to a turnout rate of 0.52%, or less than one person for every one hundred eligible.”
- North Carolina: North Carolina Health News as published by PBS found that: “In North Carolina, prior to the pandemic, approximately 19,000 people were incarcerated in local jails… Just 20 known people cast their ballots from jail in North Carolina in the 2018 national election, according to public records requests filed to each county.”
- Arizona: The Arizona Coalition to End Jail-Based Disenfranchisement found that in 2020 “Arizona voters in jail had an estimated participation rate of 2.9%, compared to an overall state participation rate of 59.0%.”
- Pennsylvania: In a 2021 report, All Voting is Local, a project of the The Leadership Conference Education Fund, “found out of the 25,000 people in county jails in Pennsylvania, only 52 people requested mail ballots in the 2020 general election using an address associated with one of 18 county jails.”
- Ohio: public radio affiliate WOSU reported in 2016 that “Just eight inmates out of a jail population of 1,600 will vote. At the county’s downtown jail, just three of 500 inmates voted. [Shirley Royer of the Franklin County Board of Elections] says this is typical voter turnout for the jails.”
- Colorado: The Durango Herald reported in 2016 that in La Plata, Colorado, “In the last 20 years, Clerk and Recorder Tiffany Lee Parker said she’s aware of only one inmate who has requested a ballot from within the jail. As of last week, there were 144 inmates housed at the jail, of which only two were ineligible to vote under state law because they are serving time for a felony conviction.”
These tiny voter turnout figures bring to mind voting rates among Black voters in the Jim Crow era South. One striking data point is that in several states, there are counties with a substantial population in jail, but no registered voters in jail. There’s a relevant quote from Ezra Klein’s Why We’re Polarized about voting in the South: “In 1953, in the so-called Black Belt – the region of Alabama where the black population exceeded the white population – ‘only 1.3 percent of eligible African Americans were registered. Two counties had no black voters whatsoever.’”
With changes in federal law in the 1960s allowing more Black men and women the free opportunity to vote, changes in the justice system to increasingly jail those people undid the advances in equality.
The best data we have about voting rates in jails may come from Orion Taylor and Anna Harvey at NYU Public Safety Lab’s Jail Data Initiative. In a new preprint paper, they probabilistically compare jail rosters to voter information files for the 2020 election. Harvey and Taylor write:
We find that registered voters booked into county jails during 2020 voting days for the full duration of voting days were on average 46% less likely to vote in 2020, relative to registered voters booked into the same jails just after Election Day. The estimated negative effect of incarceration on voting from jail was on average 46% larger for Black registered voters, relative to white registered voters. Black registered voters booked into county jails during 2020 voting days for the full duration of voting days were on average 78% less likely to vote in 2020, relative to Black registered voters booked into the same jails just after Election Day.
In addition to finding a significant negative effect for incarcerated voters, Harvey and Taylor find a turnout rate of 12.2% – 22% for registered voters in jails, and a 78% reduction in turnout for Black registered voters. That translates to only 6% turnout for Black registered voters in jails.
These figures are, in part, due to two conditions: restrictions on access to information about voting, and restrictions on the materials jailed adults need to vote. Both of these have potential technological solutions.
The United States has a complicated tapestry of voting laws, and no easy rules for who is and isn’t eligible to vote. It is well known that people incarcerated for a felony conviction are not allowed to vote in 48 of the 50 states. In 11 states, people with a felony conviction for some crimes permanently lose their right to vote, according to the National Conference of State Legislatures. In The New Jim Crow: Mass Incarceration in the Age of Colorblindness, Michelle Alexander notes: “No other country in the world disenfranchises people who are released from prison in a manner even remotely resembling the United States. In fact, the United Nations Human Rights Committee has charged that U.S. disenfranchisement policies are discriminatory and violate international law.”
Confusion over voting while incarcerated means jailed people who are eligible to vote are sometimes misinformed by elections or prison officials. A good example comes from a case study in Time. Christian Nasse, a detainee at the Apache County Jail in Arizona, had managed to register to vote. But before he returned his ballot:
…[A] corrections officer warned him that he could face more charges if he voted with a prior felony conviction—something that is true in some cases but didn’t apply to Nasse, who is eligible to vote. Nasse said he eventually decided not to cast a ballot because he didn’t want to risk any more legal problems. When asked whether he felt intimidated into making that decision, Nasse said yes.
Similarly, in one New York county, a jail representative “estimated that only one-third of corrections officers were aware that detained populations retain the right to vote. Perhaps unsurprisingly, no one voted from [that] facility in the 2020 November Election.”
Even when officials may be aware of eligibility rules, there is an inherent inequality when minority individuals have to go seek out information about voting. In “What Do I Need to Vote? Bureaucratic Discretion and Discrimination by Local Election Officials”, a team of researchers at Harvard sent emails to over 7,000 local officials asking for information about voting rules from email aliases in order to compare the responses:
Analyzing over 5,300 replies, we find clear, causally identified evidence of bias against Latinos in the responsiveness of local election officials. Emails from Latino names are roughly five percentage points less likely to receive a reply to a question about voter ID requirements than those from non-Latino whites. Replies that Latino emailers do receive are less likely to convey accurate information about ID requirements.
These challenges make clear why it is so important for jailed individuals to have access to accurate information about how to register and exercise their right to vote. But while such information is readily available on the internet, the internet is not readily available in jail. In fact, it’s almost totally unavailable. As Joe Garcia wrote in the MIT Technology Review: “Even though the internet is a given in the rest of society, access in prison is so restricted it’s almost nonexistent.”
Governments increasingly use the internet to distribute information about voting changes because the rules so often change, especially at the last minute. According to the Brennan Center, as of May 2022, “lawmakers in 39 states have considered at least 393 restrictive bills for the 2022 legislative session” which restrict who can vote, when and how. In an environment where voting information is changing every cycle, the internet is a key tool to get information out to everyone.
Increasing internet access for people in jails is one key way to make sure they get the information they need to register and vote. More just means to access the internet for people in jails and prisons is sorely needed to reduce reliance on predatory prison telecom companies, who make $1.4 billion a year from some of the poorest members of society and their families.
Even when individuals in jail are aware that they can vote, access to registration materials and to voting mechanisms such as voting machines, mail-in or absentee ballots are also restricted in jails. Problems include:.
- Difficulty registering to vote or requesting an absentee ballot: The think tank Demos has sued on a particularly egregious example out of Ohio: “Eligible Ohio voters who find themselves jailed on pending charges or on a misdemeanor conviction may request an absentee ballot from their county board of elections. But such a request must be made in person at the board’s office by close of business on the Friday before an election or, if sent [by] mail, must be received at the board’s office by noon on the Saturday before an election. This means that no person arrested after close of business Friday and held in custody through Election Day will be able to request and vote by absentee ballot. Since county officials don’t make voting available on-site in jails and don’t escort people to the polls, these ‘late-jailed’ voters have their voting rights absolutely denied.” The most extreme example may be a case study about registering in Texas produced by the Brookings Institution.
- Access to Pens: Ballots have to be filled out in blue or black pens. Some people don’t have pens or facilities don’t allow pens. For instance, a voter registration drive in Cleveland was stymied by the local board of elections rejecting all applications filled out in pencil, according to the ACLU’s Voting While Incarcerated guide.
- Personal Identification: Voter ID laws in general suppress turnout for voters of color, but add an extra burden for people in jail. When states ask for information off of identification like driver’s licenses, people in jail can’t just reach for their wallet. The information or the document has to be requested from someone and that process can be onerous.
- Stamps: Many jurisdictions require that you provide your own stamp to mail back your ballot. It’s a known problem that some people can’t afford legal correspondence, so federal regulation specifically requires that the state cover the cost. This should be expanded to ballots. In addition, people may not have regular access to a commissary to purchase stamps, and if ballots are a non-standard size they may not have access to a scale to determine the correct postage.
A general pattern for prison technology is that as new technologies come to prisons, authorities deprecate and remove older, cheaper ones and replace them with new, expensive and monitorable ones. (For many more examples, see my earlier article Access to Technology in the American Carceral State, also at Tech Policy Press.)
Mail is a particularly challenging communication mechanism. Jails and prisons are obsessed with the idea of contraband making it in through the mail, even though there is little evidence this is a widespread problem. For instance, less than 2% of all contraband entering Florida prisons came through the mail. Nevertheless, facilities across the country, including Florida, North Carolina and New Mexico, do not allow any access to physical mail at all, only scans. These scans, even if you can print them, are not valid ballots and won’t be accepted by election authorities. This pattern also exposes both people in jails and their correspondents on the outside to additional surveillance, a concern addressed recently by the Knight First Amendment Institute. Similarly, the Prison Policy Initiative recently pointed to a contract from Contra Costa County in California which says that “all non-privileged outgoing mail may be read by custody staff,” meaning that ballot secrecy is not guaranteed in jail settings, another factor that may depress turnout even if other steps are met.
Even if everything goes right and all rules are followed, things can still be taken away. In the same Arizona jail mentioned above, Time reports that voter information postcards were delivered to a facility and then taken away by staff for no apparent reason. PEN America has reported for years on the wide variety of books being arbitrarily banned, including The New Jim Crow, in North Carolina, Florida, Michigan, and New Jersey.
What We Can Do
Although it is too late to fix things for the 2022 mid-term election, governments and voting rights activists should take action before the 2024 presidential election.
There are two complementary paths to reform that would increase voting participation by those in jail. We can tailor solutions to the direct problem of access to voting information and the means to vote in jails, and we can try to fix larger systemic problems that contribute to disenfranchisement of people in jails.
There are bright spots. In Illinois, Cook County has set up a first-in-the-nation voting location in its county jail. In 2019, the Illinois legislature authorized the county jail as an early voting site, and the first election conducted there was in early 2020. According to the Associated Press, “During the first of two weekends of voting at the jail, 1,200 detainees cast ballots.” High numbers may have been due in part to the availability of same day registration. The second election was even more successful, with forty percent turnout for the fall 2020 election. Prison Policy Initiative points to 6 more locations added in the past two years and gives recommendations for other facilities that want to add polling places. This model should be scaled across the United States.
Another innovative idea came from Jewish Employment and Vocational Services (JEVS) in Philadelphia. There, Jon Lieb ran a vocational program in jails that includes content on registering to vote. However, he had a view towards much larger impact. Modeled on the National Voter Registration Act of 1993 (the “motor voter” bill, which allowed people to register to vote at the same time that they get their driver’s licenses), Lieb sought to develop “a standardized procedure for inmate registration in which any inmate admitted to the jail would automatically be registered [to vote].“ This would likely greatly increase participation, and is consistent with other government benefits systems.
More broadly, we should get rid of voting requirements that claim to target voter fraud but mainly systematically disenfranchise people, incarcerated or not. As the Brennan Center concludes, “extensive research reveals that fraud is very rare, voter impersonation is virtually nonexistent, and many instances of alleged fraud are, in fact, mistakes by voters or administrators.” Whether it’s ID numbers, blue pens, or notaries, make it easier for everyone to vote.
Failing that, just make very sure that everyone knows that they’re eligible to vote. Dana Paikowsky at the Harvard Civil Rights-Civil Liberties Law Review tells the story of what happened at the Denver Jail:
In 2016, the Denver Elections Department, the Denver County Sheriff’s Department, and a local non-profit group ran a pilot program that simply informed jailed voters of their eligibility and provided them with the materials they needed to vote. Before the pilot program, one Denver jail employee estimated about 10 people had registered to vote from the jail. After the program, more than 300 jailed voters had registered to vote and about a hundred cast ballots from jail. In 2018, the now permanent program registered more than 760 people to vote from jail.
The simplest solution is to just increase access to the internet in jails. Election information, non-partisan voter guides, voter registration, absentee ballot applications, absentee ballot tracking and all the other infrastructure of elections is now designed for an internet friendly world. By increasing access to the web, information access and communication problems are potentially solved without any additional effort. Democracy is served when all voices are heard, no matter if they are in jail.
Benj Azose is an advocate for free prison communication and was part of the successful effort to make phone calls from California prisons free in 2022 (SB 1008). He attended the Aspen Institute Tech Executive Leadership Institute and served on the San Mateo County District Lines Advisory Commission to help re-draw San Mateo’s county lines after the 2020 census. Benj leads a team of analysts working on Google Chrome and has been at Google since 2008. He has a Bachelor’s Degree from Harvey Mudd College.