The cost of America’s democracy is on the rise. Campaign spending is skyrocketing. The general cost to run an election, from the use of taxpayer dollars to administer an election to the certification of results to everything needed to support elections, is also increasing due to an ever-growing and diversifying electorate, expanded voting options, and aging voting equipment, which all require additional funds. Most alarming, though, is the hidden cost of (improperly) running elections—the cost to marginalized communities and to election officials and workers. These hidden costs are also on the rise with an increase in efforts to restrict access to the ballot, voter confusion, and intimidation and harassment of all involved in the voting process. Understanding the funding needs to run elections—both the known and the hidden costs—allows for the development of solutions to ensure our democracy does not suffer from these costs.
What Does It Cost to Run Elections and Why Is It Increasing?
There is no clear answer to how much it costs federal or state officials to run elections. Costs can be difficult to determine because of how decentralized the process can be—both the funding and the running of elections. Different levels of government can pay for different components of an election, including completely covering or sharing costs for any given activity. For example, some states purchase all voting equipment. States, counties, and municipalities may share the cost of running an election, including personnel costs, and voting equipment usage. It can be difficult to assess the costs because the reporting of expenditures or inclusion in budgets does not always include election administration as line-item amounts.
Despite the fact that there is not a clear understanding of what it costs to run an election state by state, there are common expenses one can expect. There are different buckets of election-related expenses: “Election Day” (including early voting, absentee voting), voting infrastructure (such as voter registration database, voting equipment), and emergency/contingency (such as hurricanes, pandemics). Some costs apply across different buckets, some costs occur infrequently, and some are ongoing. For example, acquiring voting equipment is an infrequent expense. Costs that occur regularly can include printing ballots and other voting materials, paying poll workers and other personnel, transporting voting equipment to polling places, postage fees, security costs (technology as well as personnel and physical barriers), translations (where required), training election officials and poll workers, placing and securing drop boxes, providing supplies, hosting a multilingual hotline, maintaining voter registration lists, and providing support for online voter registration systems.
Costs are also associated with complying with the law to make voting more accessible for voters with disabilities and limited English proficiency (LEP) voters. There are a number of longstanding federal laws that protect the right to vote for voters with disabilities, including the Americans with Disabilities Act from 1990 to the Help America Vote Act of 2002. The Voting Rights Act, enacted in 1975, requires covered jurisdictions to provide language assistance for LEP voters, including through written translations, bilingual poll workers, and the publicity of assistance to the covered community, and it applies to four language groups (Latinos, Asian Americans, American Indians, and Alaskan Natives).
Estimates of annual election funding vary widely from $2 billion to $3 billion or $8 to $15 per vote cast prior to the 2020 election, representing an increase from the 2000 election (which was approximately $1 billion and $10 per voter). The 2020 election saw an even bigger increase, with an estimated cost between $3 billion and $6 billion. Most respondents (63.1 percent) in a survey of local election officials stated spending 50 percent more or twice as much for the 2020 election compared to a typical presidential election, and another 14.6 percent spent over twice as much. In Michigan, clerks and city managers noted an increase of two to eight times the cost for the 2020 elections, with a significant portion related to increased staffing needs to process absentee ballots.
There are three things that are clear from the existing research into the cost of elections: (1) state and local governments need systematic and detailed election cost accounting, (2) more funding is needed to support elections, including election infrastructure (with election funding often being on the low end of state and local government funding), and (3) costs will continue to increase. For example, many jurisdictions have aging voting machines and dated information technology infrastructure. Additionally, jurisdictions need to secure elections both physically, which will become more costly as the political environment gets more volatile, and in cyberspace, which will become more costly as cyberattacks and hacking mechanisms become more sophisticated, complex, and difficult to defend.
The “Hidden” Costs of Running Elections
In addition to the vast array of known costs to support and run elections, there are potential “hidden” costs associated with improperly, inefficiently, or ineffectively running elections. When elections are not properly run, they can result in inequitable access to our democracy, often afflicting our most vulnerable and historically disenfranchised communities.
Inefficient Allocation of Resources
Whether or not jurisdictions are getting the necessary funding to support the running of an election is merely one piece of the equation. What is equally important is the allocation of resources provided. When resources are not properly distributed, inequities ensue, such as extensively long wait times in certain communities but not others. For example, during the 2012 election, voters in Florida, Virginia, Maryland, and South Carolina reported waiting in line for up to five hours. During the 2020 primaries, one polling location in an Atlanta suburb of 22,400 residents, composed of nearly 88 percent Black residents, saw hundreds of people waiting outside. This was not an isolated incident in Georgia—since 2013, the Georgia voter rolls grew by nearly 2 million people, fueled primarily by younger, nonwhite voters, but polling locations were reduced statewide by nearly 10 percent. In fact, the nine counties in the metropolitan Atlanta area represent half of the state’s active voters but only 38 percent of the polling places. This is a clear example of ineffective allocation of resources that will lead to longer lines at the polls with a nearly 40 percent growth of the average number of voters for each polling location in those counties.
At a fundamental level, efficiency is affected by how many resources are available, including what time customers arrive and the rate at which they arrive, to achieve the desired outcome. In this case, having enough resources allocated to process voters efficiently and effectively at the polling place is critical to ensuring a smooth voting experience. The General Accounting Office (GAO) identified nine key factors that affected wait times on Election Day 2012: opportunities for voting before Election Day, type of poll books, determining voter eligibility, ballot characteristics, amount and type of voting equipment, number and layout of polling places, number and training of poll workers, voter education, and resource availability and allocation. What resources are available and how they are deployed influence the other identified factors. Voters in counties with the fewest resources waited two to three times longer to cast a ballot as voters in the best-resourced counties in 2018.
Race and ethnicity have a substantial influence on wait times to vote—for example, zip codes with greater than 75 percent nonwhite populations waited more than twice as long as zip codes with less than 25 percent nonwhite populations—unsurprising with the significant under-allocation of polling stations and poll workers in polling places for communities of color. For example, for the June 2020 primaries in Georgia, two-thirds of the polling places that stayed open late for waiting voters were in majority-Black neighborhoods, despite the fact they made up only about one-third of the state’s polling places. The average wait time after 7 p.m. was 51 minutes in polling places that were 90 percent or more nonwhite but only six minutes in polling places that were 90 percent or more white. Counties in which the number of residents of color increased or where incomes decreased over the past decade had fewer resources per voter in 2018 than counties that increased their white population or became more affluent.
The hidden cost of ineffective resource allocation includes long wait times for voters, which, in turn, decreases voters’ confidence that their vote will be counted as intended and increases the likelihood of voter frustration and failure to cast a ballot. This is a prohibitive cost for our democracy made worse by the fact that communities of color and lower-income communities suffer disproportionately.