Skip to main content

All elections, including federal contests, are administered in the United States by state and local governments. A growing bipartisan consensus concludes that a steady stream of crises over the past two decades make it clear that state and local election administration is underfunded in much of the country.

Whether it was antiquated and poorly maintained election equipment that led to the recount controversy in Florida in 2000, inadequate resource planning that caused long lines in 2012, cybersecurity threats in 2016, or the crush of adapting to the reality of the COVID-19 pandemic in 2020, each national election seems to shine a spotlight on a different part of the system and its funding needs.

The resource challenges experienced by state and local election officials in 2020 were part of a long story, although the scale of the problems may have exceeded past experience. True, the pandemic strained everyone’s budgets. But the stress placed on election administration came on top of years of inattention. In their 2014 report, the bipartisan Presidential Commission on Election Administration wrote:

ATLANTA, GA - NOVEMBER 06 (Photo by Jessica McGowan/Getty Images)

The most universal complaint of election administrators in testimony before the Commission concerned a lack of resources. Election administrators have described themselves as the least powerful lobby in state legislatures and often the last constituency to receive funds at the local level. (p. 10)

Election officials are used to “making do” with what they have. They often express pride in pulling off the complicated logistical maneuvers necessary to conduct elections on a shoestring budget.

One consequence of the frugality imposed on election administration is that services provided to voters vary considerably across the nation. Some states and localities flood their voters with voter guides, use the most up-to-date equipment, and deliver information and services on sophisticated websites. Others provide only minimal services to voters, rely on voters to figure out the details of voting on their own, and use equipment that is no longer manufactured or is incapable of being updated with the latest security patches.

Given the issues with underfunding, bipartisan interest in providing more federal dollars to support state and local governments is growing. Republicans and Democrats alike want to be able to trust our elections.

Current Cost of Elections in the U.S.

To assess whether and how much new federal funding should be provided, it is important to understand what the funding need is. Hard figures on the cost to conduct elections in the U.S., however, are hard to come by. To meet the need, CommonSense American partnered with the MIT Election Data + Science Lab to produce a new report. Much of what follows in this section of the brief is a summary of that 17-page “The Cost of Conducting Elections” report.

The MIT/CommonSense American report found that studies relying on different methodologies across the past two decades have come up with ballpark estimates that are overall consistent with one another. Those studies estimate the total cost of elections being in the range of $4 billion to $6 billion, in a “normal” year, and that spending in 2020 could have reached $10 billion.

READ MORE

Those studies include:

  • The 2001 report, Voting: What Is/What Could Be, by the Caltech/MIT Voting Technology Project (VTP), estimated that local election departments spent approximately $1 billion on election administration in 2000.
  • Replicating the VTP 2001 research for the President’s Commission on Election Administration in 2013, research conducted by Stephen Ansolabehere, Daron Shaw, and Charles Stewart III indicated that local election departments spend approximately $2.6 billion in election administration in 2012.
  • In a 2018 paper based on the annual financial report of local governments, a research team from the University of North Carolina-Charlotte estimated that total local election costs of elections were about $2 billion, an estimate the team judged to be a “lower bound” because of how long-term liabilities and capital purchases are treated in local budgets.
  • Data compiled by the California Association of Clerks and Election Officials suggest that $330 million was spent in that state to conduct both the primary and general elections in 2016, and $539 million in 2020. If these figures are converted to a per-voter basis, they are consistent with $2.3 billion being spent on both primaries and general elections in 2016 and $5.7 billion in 2020. Corresponding estimates are $2.6 billion and $3.3 billion for 2014 and 2018, respectively.
  • North Dakota has published spending by local governments for statewide primary and general elections since 1980. In 2020, North Dakota counties spent a total of $3.3 million. If we extend this figure nationwide on a per-voter basis, this works out to $1.5 billion nationwide.

Because of the data problems discussed below, there is no comprehensive understanding of how much is spent on ongoing administration and how much is spent to conduct specific elections. Research by a team at the University of North Carolina-Charlotte (see Read More section above) was based on actual budgetary documents. The findings suggest that the cost of individual elections is about half of overall spending on election administration in a state, which should also account for ongoing administrative support, amortized capital expenditures, and state expenditures. Thus, the estimates in the 2010s that spending on elections was in the range of $2 billion to $3 billion is consistent with the total cost of elections being in the range of $4 billion to $6 billion, in a “normal” year, and that spending in 2020 could have reached $10 billion.

The recent report by the Election Infrastructure Initiative that aimed to estimate state and local costs to conduct elections over the next decade came up with a national figure of $53.3 billion, which is broadly consistent with these previous estimates, when converted to an annual basis (i.e., $5.3 billion).

For the sake of scale, the U.S. Census Bureau estimates that local governments spent $2.0 trillion in 2018. An annual bill for $5 billion to run elections would amount to 0.25% of local government spending. The estimated cost of conducting elections on an annual basis is roughly what local governments spend managing public parking facilities.

Are current spending levels enough?  A consensus exists within the election administration community that elections are underfunded nationwide, with certain locations more underfunded than others. Questions remain about how large the funding gap is, where the gaps are the largest, and how to fill them.

READ MORE

It is important to note that information about the cost of conducting elections is so elusive because states and localities differ in how election functions are accounted for in budgets. Plus, as discussed above, the amounts are such a small portion of local government budgets that it rarely occurs to officials to separately report spending on elections. Because so many employees who work in elections have other duties, accurately allocating expenses to elections can be a challenge.

The costs of conducting elections are largely borne by local governments and are often subsumed within the operating budget of a superior official, such as a city or county clerk. At the state level, expenditures are usually lumped into the budget of the chief election officer, who is most often the Secretary of State. Finally, the U.S. Census Bureau’s annual survey of state and local government finances, which is the authoritative annual study of state and local government taxing and spending, does not record spending for elections.

Background on Federal Role in Election Funding

For most of the nation’s history, state and local governments paid for the entire cost of administering elections, even in federal election years. In exchange, states were given a great deal of autonomy to determine how elections were to be conducted.

Even when Congress mandated that state and local governments follow certain procedures – such as the Voting Rights Act in 1965, the Uniformed Overseas Citizens Absentee Voting Act (UOCAVA) in 1986, the Military and Overseas Voter Empowerment Act (MOVE Act) in 2009, and the National Voter Registration Act (NVRA), commonly known as “Motor Voter” in 1994 – none of those mandates was actually funded.

Indeed, one could argue that the requirement in the Constitution that states conduct federal elections, and that Congress can make or alter the “times, places and manner” of those elections that the states conduct is the nation’s first unfunded mandate.

The federal government began to help fund the mandates it placed on states for conducting elections after the 2000 presidential election when the narrow Florida recount caused the country to turn its focus to election administration. The resulting legislation, the landmark Help America Vote Act (HAVA), was passed in 2002. Among its many provisions, HAVA required that all states utilize a statewide voter registration system, prohibited the use of punch card and lever voting equipment, and required accessible voting equipment in every polling location.

It also created the  Elections Assistance Commission (EAC), an independent, bipartisan commission charged with developing guidance to meet HAVA requirements. The EAC serves as a national clearinghouse of information on election administration and oversees the development and implementation of voting system guidelines.

Unlike prior voting legislation, Congress authorized and appropriated funds to fulfill the highest-profile goals of the law. Funds appropriated in 2003 under Section 101 of HAVA ($350 million) could be used flexibly to improve election administration. Section 251 funds, amounting to $2.6 billion, could be used to meet the equipment upgrade requirements of HAVA. As of the end of 2020, the EAC reports that 94% of these funds had been spent.

After a few years of relative quiet, a presidential election once again brought renewed attention to our nation’s elections. This time it was the concern about attempts by foreign adversaries to interfere with the 2016 election through a series of cyberattacks on state and local election administrators.

As a result, Congress decided to make its first HAVA appropriations since 2010 when it appropriated a total of $880 million over three fiscal years ($380 million in FY18, $425 million in FY20, and $75 million in FY22) to be dispersed by the EAC to states and localities to improve election security, most notably cyber security. The EAC reported in September 2021 that 48% of the funds distributed to states under this program had been spent.

Most recently, Congress appropriated $400 million in the CARES Act to assist with election expenses arising because of COVID. The EAC reported in April 2022 that 95% of funds originally distributed to the states under the CARES Act have been spent.

In addition, private philanthropy contributed over $400 million to state and local election offices in 2020.

READ MORE

The largest source of these funds came from Priscilla Chan and Facebook CEO Mark Zuckerberg. Several million additional dollars were contributed by Arnold Schwarzenegger to help maintain in-person polling places. Much less controversial, but still part of the philanthropic response, were in-kind donations of everything from athletic arenas and stadiums to hold socially distanced voting during the pandemic, to hand sanitizers manufactured and contributed by breweries.

The largest portion of the Chan-Zuckerberg funds was distributed by the Center for Tech and Civic Life (CTCL); a smaller proportion was distributed by the Center for Election Innovation and Research. Grants were distributed based on applications for the funds. CTCL has reported that most jurisdictions used the grants for temporary staffing, mail ballot supplies, poll worker salaries, personal protective equipment, election equipment, and cleaning polling places — all uses that were common among the CARES Act funds that were distributed by the EAC. In reaction to concerns raised about the potential abuse of such funding sources, several state legislatures have prohibited private assistance of the type provided by these grant programs.

Proposals

Congress is considering two main federal funding proposals to support state and local governments in administering elections.

1. Another one-time appropriation to address specific and pressing needs

Congress could do another round of one-time federal funding to help state and local governments address specific challenges they currently face. One leading example is to fund election security by helping states with equipment upgrades (both voting systems and voter registration systems) and other security needs.

This would be consistent with the role the federal government has played since 2002, financing the replacement of decrepit voting machines, supporting the rapid expansion of cybersecurity capacity, and helping with pandemic-related expenses.

Such an appropriation would likely be in line with some of the more recent federal appropriations in the $400 to $500 million dollar range.

2. Annual general appropriation over the next ten years

By providing a longer, ongoing, and more general funding commitment, Congress could give election administrators the ability to develop budgets based upon the reasonable assurance of the availability of such funds. This predictability is aimed at reducing the amount of hoarding that has plagued recent federal appropriations and would allow state and local governments to be more strategic with their investments. That is important because a significant portion of fundamental election administration costs are uneven.  For example, the need to purchase voting equipment comes only come every ten and twenty years when that equipment reaches the end of its useful life. With guaranteed funding over ten years, state and local elections administrators would have more choice about where the funds are best invested rather than having that determined by Congress.

As with HAVA, Congress could include a financial “maintenance of effort” requirement to ensure that state and local governments don’t cut existing expenditures. These federal appropriations should be viewed as additional support, not replacement.

What is the appropriate amount of such annual support? One way to look at this is to say that ballots contain four types of elections – three types of candidates (federal, state, and local) and ballot questions (referendum, amendments, etc.). Therefore, the federal government should pick up roughly one-quarter of the cost of elections, at least in federal election years. This rule-of-thumb might amount to around $1 billion or a little more annually (roughly one-fourth of the $4 – $6 billion total annual cost).

A more ambitious approach is to base federal support on the amount of federal only voters, a sum that is proportional to the number of those people who only vote in federal elections. Because these “federal-only” voters are about one-third of all voters, this rule-of-thumb suggests an annual federal fair-share outlay approaching $2 billion (roughly one-third of the $4 – 6 billion total annual cost) for the support of elections, both current and capital expenses.

The Case For

Supporters argue that by far the most important reason to increase federal funding for elections is also the simplest: we must be able to trust that our representatives have been determined by the voice of the people. Proponents argue that leaders of both parties are eroding this bedrock of the most successful republic in history by casting doubt on our election results. Both Republican and Democratic supporters argue that given how meager current funding for administering elections is, our system is in danger of failing under the partisan stress we are increasingly placing on the system. They argue that we should invest significantly more to meet this fundamental threat to American self-government.

Supporters further argue that it is appropriate for the federal government to fund its fair share of the cost of addressing the challenge that bitter partisanship poses to trust in the nation’s elections. Proponents make three traditional arguments for the federal role in meeting the funding need.

First, they argue that Congress should play a role in funding elections because the U.S. Constitution mandates that states hold elections for the U.S. House and Senate.

Second, because Congress has imposed mandates on the states for federal elections that impact the cost of state and local elections in the four pieces of legislation discussed above, supporters argue that it should help fund these elections.

READ MORE

The NVRA currently applies to 44 states and Washington, D.C. The Act’s state mandates include allowing mail-in voter registration, providing voter registration opportunities when eligible voters apply for a driver’s license (the origin of the nickname “Motor Voter”), limiting how early states can set cutoff dates for registering, and requiring a uniform and regular voter registration list maintenance program.

These requirements create demands for management and computer systems that many states would not have undertaken absent the NVRA. As cybersecurity issues have risen in importance, these federally mandated voter registration systems have become a major point of vulnerability that states are now responsible for defending.

HAVA required states to replace outdated mechanical lever and punch-card voting machines in federal elections. It also required states to implement a statewide voter registration list.

At the time of HAVA’s passage, nearly half of voters cast ballots on lever and punch-card machines, which in some cases had been in service for a half-century. The HAVA voting equipment mandate had the effect of requiring localities to rely on computerized tabulation equipment that had a more limited useful life than they had previously chosen.

Prior to HAVA’s registration list mandate, localities frequently maintained voter lists, often using paper systems. This requirement also mandated complicated systems of computers, software, and networks that are expensive to design, implement, and protect from cyberattacks.

The third traditional argument supporters make for federal funding is that federal elections draw more voters than state and local elections. More people vote when federal offices are on the ballot than when comparable state races are at the “top of the ticket.”  Because one-third of voters are best described as “federal only” voters, many supporters argue that the federal government should provide about $2 billion per year, which is one-third of the $6 billion per year in total election costs.

READ MORE

It is unclear precisely how many voters would cast ballots in even-numbered-year elections if federal offices weren’t on the ballot, but the experience of the five states that conduct statewide elections in odd-numbered years gives us a clue.

Five states conduct statewide elections for governor and state legislatures in odd-numbered Novembers: Kentucky, Louisiana, Mississippi, New Jersey, and Virginia. The figure below shows the turnout levels for each statewide November election from 2018 to 2021, a complete election cycle. The bold black and red bars help to compare voter turnout for presidential races with that for races for governor, respectively. Looking back across the most recent presidential and gubernatorial elections in these five states, average presidential-year turnout has been 50% greater than turnout in gubernatorial years.

Comparing mid-term congressional elections (grey bars) with corresponding state legislative elections (pink bars), we see an even greater disparity between turnout for comparable federal and state elections — turnout in congressional-only elections was on average 130% greater than in state-legislature-only elections.

Federal elections attract many “federal only” voters. When it comes to the chief executives (president and governor), federal-only voters account for one-third of voters in these five states. In purely legislative elections, federal-only voters account for more than half of those who turn out.

These three long-standing arguments for a significant federal role in elections funding have been joined by a growing chorus arguing for federal funding to address the growing need to defend against the threat of cyberattacks from foreign and domestic adversaries. They argue that the national security implications of cyberattacks on our nation’s election infrastructure justify a federal funding solution.

In addition, these national security concerns have prompted a growing amount of support on the right – where the least support for federal funding has traditionally been found.

For example, in December 2021, several right leaning groups, including Americans for Tax Reform, R Street Institute, and FreedomWorks, sent Congressional leadership a letter calling for “robust and consistent assistance to state and local governments to ensure the integrity of our nation’s election infrastructure.” The letter’s signers noted that election administrators needed the assistance of the federal government “to go toe-to-toe with the offensive cyber-capabilities of foreign militaries.”

Among federal funding supporters, there remains disagreement on the terms and amount of such funding. Perhaps the most fundamental question among supporters is whether the next round of federal funding should be another one-time allocation or if there should be a move to an ongoing commitment, for example, like federal government highway funding.

Most proponents of federal funding argue for an ongoing commitment for two reasons. First, they argue it gets us beyond being merely reactive to a problem after it has already occurred. They argue that some of the past crises could have been avoided if election infrastructure were better maintained. One of these crises, in 2000, brought the country to the brink of a constitutional crisis.

Second, as experience with the 2002 HAVA appropriations has shown, when Congress provides large infusions of cash on the heels of a crisis that was caused by under-investment in election infrastructure, the reaction of state and local officials is one of two strategies, neither of which produces optimal outcomes. The first is to make hasty purchasing decisions. The second, ironically enough, is exactly the opposite: to hoard funds on a belief that things could get even worse. Both strategies reflect the cash-starved nature of election administration and the infrequency of being able to make major purchasing decisions based on a strategic vision.

The ongoing approach gains more Republican support when fewer strings are attached to the funding. Conservatives emphasize that state and local officials are in the best position to know where the money is best spent. Those on the right also argue that funding with fewer federal requirements more appropriately adheres to the constitutional mandate that states conduct elections. Some Democrats argue that establishing more specific requirements better ensures that the money is spent where it is intended to be spent.

A recent bipartisan proposal from several national organizations staked out a middle ground. Its supporters argued that additional funding should be tied to a limited set of minimum standards in critical areas such as voter registration, counting, and security. Still, virtually all supporters of additional federal elections support agree that some assistance, even if it is one-time funding, is better than no assistance.

The next most important question debated among those who support additional federal funding is how much should be provided. Last year, for example, the Election Infrastructure Initiative asked Congress to consider $20 billion over ten years ($2 billion per year). More recently, President Biden included $10 billion over ten years ($1 billion per year) in his most recent budget proposal. Others have suggested federal funding of around $5 billion over ten years ($500 million per year).

The Case Against

Opponents of additional federal funding question the need. They note that over the past five years Congress has appropriated over a billion dollars to state and local governments between HAVA Security Grants and CARES Act funding. They also point to the above-mentioned EAC report showing that states had spent only 48% ($387 million out of $880 million) of the HAVA Security Grants.

Further, some argue that the federal government has no role in funding election administration because elections should remain a state matter. After all, elections are conducted according to state laws. Certainly, the U.S. Constitution grants to states the right to determine the “time, place, and manner” of conducting elections for federal offices, with the proviso that Congress can regulate federal elections if they wish.

However, Congress has responded by placing a light hand on the levers of electoral regulation. It has set uniform federal election dates and mandated that members of the U.S. House be elected by districts. It has also enacted civil rights legislation to deter racial discrimination in voting. If states and localities comply with these congressional mandates, they should not add substantially to the cost of elections.

Therefore, if elections are underfunded, these opponents argue, this is a state problem, not a federal one. After all, the “price” for the flexibility that the Constitution and Congress largely provides to states to administer elections consistent with each state’s preferences and customs is that the states must pay for the costs of those elections. Moreover, state and local election administrators are in the best position to figure out what is needed and how to pay for it.

Close Menu