The Anderson-Burdick doctrine: Balancing the benefits and burdens of voting restrictions


This election explainer was written by experts from Election Law at Ohio State, a program of the Moritz College of Law. It is part of SCOTUSblog’s 2020 Election Litigation Tracker, a joint project with Election Law at Ohio State.

Almost any voting rule or procedure can be subjected to judicial scrutiny under doctrinal analysis known as Anderson-Burdick balancing. This balancing test requires courts to weigh burdens that a state imposes on electoral participation against the state’s asserted benefits, although the way in which courts are supposed to conduct this balancing is the subject of much dispute and uncertainty.

Named for two Supreme Court precedents involving candidates’ access to the ballot — Anderson v. Celebrezze and Burdick v. Takushi — this form of judicial inquiry has since been applied by the Supreme Court to voter identification rules, in Crawford v. Marion County Election Board, and has also been applied by lower courts to a wide variety of election administration issues. The number of days and hours of early voting, the requirements for casting and counting provisional ballots and the availability and regulation of vote-by-mail are among the myriad of issues subjected to Anderson-Burdick review in recent years. A difficulty with Anderson-Burdick balancing, however, is its largely indeterminate nature, which — in addition to causing deep divisions among the justices in the Crawford voter ID case — has led to considerable divergence among lower court judges, as well as confusion among election administrators and other participants in the electoral process. Clarification by the Supreme Court will be at some point necessary.

Anderson (1983) involved Rep. John Anderson’s effort to get on the ballot in the 1980 presidential election. Ohio had a rule that required independent candidates, like Anderson, to file in March for the November general election. The court held that this filing deadline was excessively early, violating the First and 14th Amendments.

Burdick (1992) involved Hawaii’s prohibition against write-in voting. The court upheld the prohibition as a reasonable regulation in the state’s effort to winnow the field of candidates down to a single winner. The court observed that any candidate had the opportunity to participate in Hawaii’s “open primary” process, and being in the primary gave a candidate a fair opportunity to get onto the general election ballot. Relying on a previous case, Storer v. California, the court reiterated that a state does not need to make it possible in the general election to vote for a candidate who failed to earn enough votes in the primary.

Crawford (2008) is the most significant case to apply the Anderson-Burdick analysis, but its significance derives in part from the court’s failure to produce a majority opinion for either a jurisprudential understanding of how the Anderson-Burdick inquiry is supposed to operate or the outcome it was supposed to achieve in the particular case. Crawford concerned Indiana’s voter identification law, which required in-person (but not absentee) voters to provide a government-issued form of photo identification unless the voter made a subsequent trip to a local office to attest to the voter’s indigency. The plaintiffs brought a facial challenge to the statute’s constitutionality, seeking an injunction to bar its enforcement to any Indiana voter. In what was essentially a 3-3-3 split among the justices, the court rejected the claim of facial invalidity and thus declined to order the requested across-the-board injunction against the statute’s implementation.

Justice John Paul Stevens announced the court’s judgment and wrote an opinion for himself, Chief Justice John Roberts and Justice Anthony Kennedy. This opinion represented the middle of the court, but it was an unstable middle insofar as it spoke for so few of the justices. Stevens described Anderson-Burdick as a “flexible” balancing test, to be employed for regulations of the voting process that do not categorically deny the franchise to a class of citizens based on “invidious” discrimination (the latter being governed by a stricter standard under the 14th Amendment’s equal protection clause, as reflected in such other precedents as Harper v. Virginia Board of Elections, which invalidated poll taxes). With respect to the burden imposed by Indiana’s photo identification requirement, Stevens observed that it would not be a barrier to voting for most citizens, but at most an inconvenience that could not be considered unconstitutional. As to the unknown number of citizens for whom obtaining the documentation necessary for a state-issued photo ID (like a birth certificate) would be prohibitive — Indiana provided the photo ID at no charge, but not the underlying documentation — Stevens was not prepared to invalidate the entire statute on its face based on the evidentiary record that the plaintiffs had presented. Stevens, however, indicated that the court would be open to as-applied challenges to protect specific individual voters from being unconstitutionally disenfranchised by an excessively burdensome and insufficiently justified obligation to either pay for a birth certificate they could not afford or make an unnecessary separate trip to a government office to attest to their indigency.

Justice Antonin Scalia wrote a separate concurrence for himself, Justice Clarence Thomas and Justice Samuel Alito. While agreeing in the rejection of the facial challenge to the statute, this concurrence made clear that it also would foreclose any as-applied challenges, and it emphatically rejected Stevens’ flexible conception of Anderson-Burdick balancing. Instead, Scalia viewed Anderson-Burdick as establishing a two-tiered test. If a burden on voting was not severe, Scalia argued, then the case should not be analyzed under the more stringent tier and only minimal judicial scrutiny should be necessary. Evaluating the burden of the photo ID requirement to the entire class of voters subject to it, Scalia considered it slight, hardly more than the burden of having to cast a ballot. As far as he was concerned, that was the “end of the matter.”

Justices David Souter, Ruth Bader Ginsburg and Stephen Breyer dissented, in two opinions. The three dissenters essentially agreed with Stevens that Anderson-Burdick balancing should be flexible, but they viewed that flexibility as leading to the conclusion that the Indiana statute should be facially invalidated, not sustained. Souter, unlike Stevens, was willing to look beyond the evidentiary record in the case (using what Stevens disparaged as “extensive Internet research”) to conclude that Indiana’s law threatened “nontrivial burdens” on “tens of thousands” of citizens. Breyer thought it essential that a state pay for the documents necessary to obtain a required ID in order for the requirement not to function as the equivalent of an unconstitutional poll tax.

In the dozen years since Crawford, lower courts have been as divided as the justices on the application of Anderson-Burdick to the burdens that various voting regulations and practices impose on voters. The Supreme Court has managed to avoid the need to clarify the jurisprudence on this point, although the court has fielded multiple emergency motions that have touched on Anderson-Burdick issues arising in the midst of pending elections, including several emergency motions that the court has decided already this year. But the disposition of these emergency motions has not required the court to issue a pronouncement on how Anderson-Burdick operates or applies in any particular context. Thus, the world of election law waits to see whether the court will continue to dodge this topic this year, or whether instead the unprecedented volume of litigation over the voting process that is currently percolating in the lower courts will require the court to provide some clarification of Anderson-Burdick before this November’s election is over.

Term Snapshot
At a Glance
Awards