TagAnalysis

0.8% of absentee/mail-in ballots rejected in 2020

In the 2020 general election, voters cast 70.6 million absentee/mail-in ballots, and election officials rejected 0.8% of them, according to data released by the U.S. Election Assistance Commission (EAC). The number of absentee/mail-in ballots cast in 2020 marked a 111% increase over 2016. And by comparison, absentee mail-in ballot rejection rates in 2018 and 2016 were 1.4% and 1.0%, respectively.

Absentee/mail-in ballot rejection rates decreased in 34 states between the 2016 and 2020 elections, increased in 12 states and Washington, D.C., and remained unchanged in two states.

New Mexico had the largest increase in its absentee/mail-in ballot rejection rate, with the state rejecting 0.2% of absentee/mail-in ballots in 2016 compared to 5.0% in 2020, a 4.8 percentage point increase. No other state had an increase larger than one percentage point.

Georgia saw the largest decrease in its rejection rate. In 2016, the state rejected 6.4% of absentee/mail-in ballots, decreasing to 0.4% in 2020. However, the EAC commented in its report that figures for Georgia may not necessarily include all rejected ballots. Kentucky had the second-largest decrease, dropping from 5.6% in 2016 to 0.5% in 2020.

Election officials may reject absentee/mail-in ballots for a number of reasons, ranging from a missed deadline to the use of an incorrect return envelope. The exact criteria for rejecting a ballot varies from state to state.

The most common reason for a rejected ballot in 2020 was when the signature on the ballot did not match the signature on file. According to the EAC, 32.8% of all rejected ballots were discarded for this reason. A non-matching signature was also the most common reason for rejection in 2016, accounting for 27.5% of rejections that year.

The EAC publishes its biennial Election Administration and Voting Survey in the odd year after every federal election. The EAC compiles its report based on information submitted to it by state election officials.



Ninety-two percent of state legislative incumbents filed for re-election in 2021

In 2021, state legislative incumbents filed for re-election at a higher rate than any other year in the past decade other than 2013. When an incumbent does not run for re-election, his or her seat is left open, meaning it is guaranteed to a newcomer at the start of the next state legislative session.

Of the 220 seats up for election this year, 92.3% of incumbents (203) filed for re-election, leaving 7.7% of seats (17) open. From 2011 to 2021, only the state legislative elections held in 2013 saw a lower percentage of open seats at 6.8%.

Two states—New Jersey and Virginia—are holding state legislative elections in 2021. These states hold elections every two years in odd-numbered years.

In New Jersey, 120 seats are up for election, 10.0% of which (12) are open. Of those open seats, six were most recently held by Democrats and six by Republicans.

Compared to previous elections, New Jersey’s rate of open seats in 2021 is tied with 2017 for the state’s second-highest percentage of open seats in the past decade.

In Virginia, 100 seats are up for election, 5.0% of which (5) are open, a decade-low rate for the state. Of those five open seats, one was most recently held by a Democrat and four by Republicans.

Neither New Jersey nor Virginia has state legislative term limits, meaning all open seats this year were left by incumbents voluntarily choosing not to file for re-election. Of the four states that hold state legislative elections in odd-numbered years, only one—Louisiana—has term limits. 

As shown by the chart below, term limits can have a varying effect on the total number of open seats. In 2011, 15.2% of open seats were caused by term limits, while in 2019, term limits accounted for 45.6% of all open seats.

In the chart below, a voluntary open seat is one where an incumbent chose not to file for re-election. A term-limited open seat is one where an incumbent could not seek re-election due to term limits.

This analysis was conducted as part of Ballotpedia’s annual state legislative competitiveness study. In addition to open seats, this study includes an analysis of incumbents in contested primaries and seats with major party competition in the general election. 

To learn more about open seats in the 2021 state legislative elections, click here.



22 states allow for the recall of school board members

Of the 39 states that allow for the recall of elected officials at some level of government, 22 specifically allow for the recall of school board members. School board recalls are the process of removing school board members from office via a public effort before their term is completed.

Six of the states that allow school board recalls require specific grounds to be met in order for a recall effort to move forward, such as malfeasance or misfeasance in office.

The number of signatures required to get a school board recall on the ballot varies by state. Common factors for calculating the signature requirement include the size of the board member’s jurisdiction and the number of votes cast in a previous election. In all but one of the states, recall elections are held if enough signatures are collected. Virginia is the exception. If enough signatures are collected in that state, a trial is held at the circuit court level.

The amount of time recall petitions are allowed to be circulated also varies by state. Georgia, Nebraska, and North Carolina have the shortest petition circulation time with 30 days. Out of the states that have a time limit for circulating petitions, Washington has the longest with 180 days. New Mexico, North Dakota, Tennessee, and Virginia do not have a time limit for petition circulation.

Between 2006 and 2020, Ballotpedia covered an average of 23 recall efforts against an average of 52 school board members each year. The number of school board recalls in 2021 has surpassed that average with 57 efforts against 143 members as of Aug. 3. This is the highest number of school board recalls Ballotpedia has tracked in one year since our tracking began in 2010. The next-highest was in 2010 with 39 efforts against 91 school board members.

Additional reading:



How much did your governor make last year?

Eighteen states paid their governor more last year than in 2019, according to the Council of State Governments’ Book of the States. Gubernatorial salaries in 2020 ranged from a low of $70,000 in Maine to a high of $225,000 in New York, with the average governor making $145,370. In the 18 states where a governor’s salary increased, the average increase was $6,604, or 4.3%. Washington was the only state to decrease its governor’s salary, registering a 0.5% decrease over the 2019 rate.

The states with the five highest gubernatorial salaries in 2020 were New York ($225,000), California ($209,747), Pennsylvania ($201,729), Tennessee ($198,780), and Massachusetts ($185,000). The states with the five lowest gubernatorial salaries were Maine ($70,000), Colorado ($92,700), Arizona ($95,000), Oregon ($98,600), and Nebraska ($105,000). Arizona, Colorado, Maine, and Oregon have been in the bottom five states for gubernatorial compensation since at least 2010. Only New York has been in the top five in every year since 2010. New York was also the state with the largest increase in gubernatorial salary in 2020, with a $25,000 increase relative to 2019.

Gubernatorial salaries are typically determined either by a state’s constitution or by statute. Most often, the salary portion of a governor’s compensation is defined by law, but additional benefits (insurance, official residence, and other work-related equipment) may be established by state agencies, custom, or other factors. For instance, 45 states subsidize the governor’s travel and 45 states have official gubernatorial residences.

In some cases, salaries automatically increase each year either at the rate of inflation or by another percentage chosen by the legislature. In other states, the legislature must pass salary increases for the governor.

Additional reading:



Virginia gubernatorial candidates seek to define themselves, one another as the general election begins

Campaigns for Terry McAuliffe (D) and Glenn Youngkin (R) in the election for Governor of Virginia released new adsshortly after the major-party nominating contests came to a close in recent weeks.

Following Youngkin’s nomination on May 10, former Gov. McAuliffe released an ad titled “Virginia Forward” where he compared Youngkin to former President Donald Trump (R), saying, “Youngkin wants to bring Trump’s extremism to Virginia.” McAuliffe went on to say he would defend policies from his administration including the expansion of Medicaid, protections for reproductive rights, and the expansion of voting rights.

After McAuliffe won the Democratic primary on June 8, Youngkin released an ad titled “Time For Change” featuring one of McAuliffe’s Democratic primary opponents, former Del. Jennifer Carroll Foy, criticizing the former governor’s tenure. At the end of the ad, Youngkin described himself as “a new kind of leader to bring a new day to Virginia.” A second ad—”A New Day”—reiterated this theme with Youngkin saying he would create jobs, improve education, and make communities more safe.

Two satellite groups—the Democratic Governors Association (DGA) and the Republican Party of Virginia (RPV)—also released ads following the nomination contests. The DGA associated Youngkin with Trump and said his main priority was, “telling the same Big Lie Trump tells about the last election and trying to restrict your right to vote in the next.” The RPV’s ad used a speech given by McAuliffe interspersed with a series of news clips from the previous eight years on topics including rates of violent crime, changes to public school curricula, and a transportation company’s decision to move out of Virginia.

A recent poll of 550 likely voters shows a close race. The poll, commissioned by CNanalytics and conducted by JMC Analytics, showed McAuliffe receiving support from 46% of respondents to Youngkin’s 42%, a difference within the poll’s ± 4.2 margin of error. The remaining 12% of the respondents were either undecided or did not respond. No third party candidates were included in the poll, though at least one candidate, Princess Blanding (I), will also appear on the general election ballot.

Every four years, the Virginia gubernatorial election is one of the first major statewide elections following the presidential election. Since 1977, the state has elected a governor from the opposite party of the president in every election except for 2013 when McAuliffe was elected governor following Barack Obama’s (D) re-election. In more recent years, Democrats have won four of the five most recent gubernatorial elections and all thirteen statewide elections in Virginia since 2012.

The gubernatorial election will determine Virginia’s trifecta status. Virginia became a Democratic trifecta in 2019 for the first time since 1994 after Democrats won majorities in the House of Delegates and the state Senate. A McAuliffe victory could continue the Democratic trifecta if Democrats also retain a majority in the House of Delegates. A Youngkin victory would make Virginia a divided government since the Democrat-controlled Senate is not holding elections this year.

Learn more about the Virginia gubernatorial election here.



Two state legislative incumbents defeated in New Jersey primary elections, a decade-high

Two members of New Jersey’s General Assembly lost to primary challengers on June 8, 2021, a decade-high number for the legislature.

Serena DiMaso (R) in the multi-member District 13 lost to Gerard Scharfenberger (R) and Victoria Flynn (R). Nicholas Chiaravalloti (D) from Assembly District 31 unofficially withdrew from the race before the primary, but his name remained on the ballot.

The Democratic primary in Assembly District 18 and the Republican primary in District 26, both featuring two incumbents each, remain too close to call as of June 11.

All four state Senate incumbents facing primary challenges won.

Primary defeats for incumbents in the New Jersey State Legislature are uncommon. Before 2021, only one state legislative incumbent had lost in a primary election: Assm. Joe Howarth (R) in 2019. No incumbent state Senator has lost in a primary since 2003.

In addition to the two primary defeats, five Democrats and three Republicans chose not to seek re-election in the General Assembly. In the state Senate, one Democrat and three Republicans opted against re-election.

Use the following links to learn more about New Jersey’s 2021 state legislative elections:



Exploring the history of Section 230 of the Communications Decency Act of 1996

You may have heard about Section 230, but what exactly is it? Section 230 is a federal law passed as part of the Communications Decency Act of 1996 (CDA) which contains two provisions that prevent internet service providers and website companies from being held liable for most content created by users of their services, including content that could be considered false or defamatory.

At the time of the bill’s passage, household internet usage was limited but rapidly growing. Societal concerns over access to pornography and sexually explicit materials by minors accompanied the rapid growth of internet usage and became notable political issues in the 1980s and 1990s. Congress sought to address concerns over pornography through legislation, including the passage of the Communications Decency Act, which became Title V of the Telecommunications Act of 1996. Shortly after passage, the American Civil Liberties Union challenged the constitutionality of the CDA on the grounds that it violated the First Amendment and Fifth Amendment. In a unanimous decision in 1997, the U.S. Supreme Court ruled in Reno v. ACLU that the Act violated the First Amendment. The decision invalidated much of the CDA with the exception of the language of Section 230, which was not the subject of the ACLU’s legal challenge, thus Section 230 was left intact and remained federal law.

The legal and social relevance of Section 230 has grown in the 21st century as internet use has expanded, causing the provision to attract both critics and defenders. Critics have argued that the law has exempted online companies from a legal responsibility to moderate what they view as harmful content and that it has protected companies that discriminate or censor content. Supporters of Section 230 have argued that it serves to promote and protect free speech and user choice on the internet.

On May 28, 2020, President Donald Trump (R) signed Executive Order 13925: Preventing Online Censorship which included a provision asking the Secretary of Commerce and Attorney General to, acting through the National Telecommunications and Information Administration (NTIA), file a petition for rulemaking with the Federal Communications Commission (FCC) requesting that the FCC clarify certain rules for the application of Section 230. The Department of Justice, on September 23, 2020, sent a letter to Congress proposing four areas for amendments to Section 230. One month later, on October 15, the FCC Chairman announced the Commission would work on clarifying the meaning of Section 230. On May 19, 2021, President Biden (D) signed Executive Order 14029: Revocation of Certain Presidential Actions and Technical Amendment, which included a provision to revoke Executive Order 13925. In addition to these actions, multiple bills have been proposed since the beginning of 2020 in Congress to amend Section 230.

Additional reading:



Seventy-nine percent of state legislative incumbents in New Jersey will not face a contested primary this year

Image of several stickers with the words "I voted"

Ninety percent of state legislative incumbents in New Jersey are seeking re-election in 2021. Of these 108 incumbents, 79%—85 legislators—will advance to the general election without a primary challenge, according to Ballotpedia’s primary election competitiveness analysis.

The remaining 23 incumbents—nine Democrats and 14 Republicans—will face contested primaries on June 8, 2021.

When an incumbent faces a contested primary, there is the chance he or she might be defeated before the general election, typically guaranteeing the seat to a newcomer. These defeats—along with retirements and general election losses—contribute to the overall incumbent turnover during each election cycle.

The most common cause of incumbent turnover is retirement, which, over the past decade, accounted for 70 percent of all state legislative turnover. Primary election defeats—at 10 percent—were the most uncommon cause of turnover.

Primary election defeats in New Jersey are especially rare. Since 2011, only one state legislative incumbent has been defeated in one: Assm. Joe Howarth, who lost in a 2019 Republican primary.

Contested incumbent primaries became more common in New Jersey during the 2017 state legislative elections. In 2019, the state saw a decade-high rate with around one-third of all incumbents facing primary challenges. The rate decreased to around one-fifth of all incumbents facing contested primaries in 2021 but remains higher than rates from the first half of the past decade.

Virginia is also holding state legislative elections in its House of Delegates in 2021. Based on preliminary data, the state is slated to see a decade-high rate of incumbents facing contested primaries at 18.1%. Out of the 100 incumbents, 94 are seeking re-election, 17 of whom will face contested primaries. Parties in Virginia use a mixture of primaries and conventions to select nominees. All primaries will take place on June 8 whereas convention dates are selected by district parties. Ballotpedia will update its primary competitiveness data for Virginia as it becomes available.

Ballotpedia is collecting primary election competitiveness statistics for all regularly-scheduled state legislative and state executive elections ongoing in 2021. Learn more here. Use the links below to view coverage of the New Jersey and Virginia state legislative elections:



Oklahoma adjourns 2021 legislative session without passing any constitutional amendments

The Oklahoma State Legislature adjourned its 2021 state legislative session on May 27, 2021. The legislature did not refer any constitutional amendments to the ballot during the 2021 legislative session. One constitutional amendment, House Joint Resolution 1001, was passed unanimously in the House but did not receive a vote in the Senate before the session adjourned. The legislature may consider the amendment again for referral to the 2022 ballot during its 2022 legislative session.

HJR 1001 would change how the Rainy Day Fund’s maximum balance is calculated. Currently, the state’s Constitutional Reserve Fund (Rainy Day Fund) has a maximum allowed balance of 15% of the General Revenue Fund certification for the preceding year. The amendment would change how the maximum balance is calculated to 15% of total expenditures for the previous fiscal year. The total expenditures would exclude money expended from revolving funds, funds with money derived from fees or non-tax revenues, and proceeds from a bond or debt obligation. In the amendment’s fiscal impact statement, legislative analysts said that “the provisions will increase the basis on which the calculation of the fund balance maximum is made, perhaps as much as a one hundred percent (100%) increase. This will result in the ability for the fund to receive more surplus funds.”

Oklahoma also allows for citizen-initiated constitutional amendments, citizen-initiated state statutes, and veto referendums, making it one of 26 states in which statewide ballot measures can qualify for the ballot through a signature petition. The number of signatures required to qualify initiatives and veto referendums for the ballot is tied to the total votes cast for governor in the last gubernatorial election. For a signature petition to initiate a constitutional amendment, proponents must collect valid signatures equal to 15% of votes cast for governor—177,958 for the 2022 ballot. For initiated state statutes, the requirement is 8%—94,911 for the 2022 ballot—and for veto referendums, the requirement is 5%—59,320 for the 2022 ballot.

One initiative petition has been filed targeting the 2022 ballot. The measure is a veto referendum that was designed to repeal House Bill 1674. HB 1674 would (1) make unlawfully obstructing use of public roads a misdemeanor and prescribe penalties for those found guilty of participation in a riot and (2) grant criminal and civil immunity in certain circumstances to vehicle operators who flee from a riot if they cause injury or death.

Once an initiative petition is filed, protests may be filed challenging the sufficiency of the petition within 10 days after the notice of filing is published. The Secretary of State sets the date for signature collection to begin at least 15 days (but no more than 30 days) after any/all protests challenging the petition have been resolved. Initiatives can be circulated for a maximum of 90 days. Signatures are due by 5:00 p.m. local time on the 90th day. State election officials in Oklahoma recommend filing a proposed initiative during the first months of the year preceding the targeted election year due to the two periods during which challenges may be filed against an initiative. For example, to qualify an initiative for the November 2022 ballot, Oklahoma election officials recommended filing the initiative proposal in the early months of 2021. For a measure to appear on the ballot, signatures must be submitted long enough in advance for the governor to issue an election proclamation, which must be issued and certified to the State Election Board at least 70 days prior to an election (August 30, 2022, to qualify for the November 8, 2022, ballot).

A total of 71 measures appeared on statewide ballots in Oklahoma from 2000 to 2020, including five measures that appeared on the ballot during even-numbered years. Of the measures, 74.65% (53 of 71) were approved, and 25.35% (18 of 71) were defeated.



Comparing 2020 presidential and senatorial vote share by party

Image of donkey and elephant to symbolize the Democratic and Republican parties.

Ballotpedia compared the performance of Joe Biden (D) and Donald Trump (R) in the 2020 presidential election to Democratic and Republican Senate candidates in each state.

Thirty-five U.S. Senate elections were held in the general election. Biden outperformed Chris Janicek (D) in Nebraska, Sara Gideon (D) in Maine, and the cumulative vote total for Democratic Senate candidates in Louisiana by the largest margins with a percentage difference of 27.0%, 22.5%, and 15.8%, respectively.

Biden underperformed Steve Bullock (D) in Montana, Doug Jones (D) in Alabama, and Mike Espy (D) in Mississippi by the largest margins with a percentage difference of 10.7%, 8.0%, and 7.0%, respectively.

The following map shows the percentage difference between Biden and Democratic Senate candidates in all states that held Senate elections. Positive numbers indicate Biden overperformed. Negative numbers indicate Biden underperformed. 

Trump outperformed Allen Water (R) in Rhode Island, Bryant Messer (R) in New Hampshire, and Lauren Witzke (R) in Delaware, by the largest margins with a percentage difference of 19.2%, 11.4%, and 7.5%, respectively.

Trump underperformed Susan Collins (R) in Maine, Mike Rounds (R) in South Dakota, and Ben Sasse (R) in Nebraska, by the largest margins with a percentage difference of 14.6%, 5.7%, and 4.7%, respectively.

The following map shows the percentage difference between Trump and Republican Senate candidates in all states that held Senate elections. Positive numbers indicate Trump overperformed. Negative numbers indicate Trump underperformed.

Additional reading: