Concept
The Anderson-Burdick Framework
Right to Vote
The Anderson-Burdick Framework
The Anderson-Burdick framework supplies the flexible review that the preceding section promised for election laws that regulate how eligible people vote rather than exclude them. It comes from two decisions, Anderson v. Celebrezze and Burdick v. Takushi. Its most prominent recent use came in Crawford v. Marion County Election Board, where six Justices voted to uphold Indiana’s photo identification law against a facial challenge although no opinion commanded a majority.1 This section teaches the method behind that result. Exclusions such as the poll tax in Harper draw heightened scrutiny because they are invidious, while other voting rules receive a more deferential standard;2 the question here is how closely a court must look once it has put a law in the second group.
The framework replaces a fixed tier of scrutiny with a sliding scale. A court first measures how much the challenged rule injures voting and associational rights under the First and Fourteenth Amendments, and it then asks how well the state’s precise interests justify that injury. The heavier the injury, the more the state must show. Because legitimate interests lie on both sides, the Court has avoided pre-set levels of scrutiny in favor of a balancing analysis in which the scrutiny varies with the effect of the regulation.3 The design answers a practical problem. Every registration deadline, ballot-access rule, and polling-place procedure burdens someone, so strict scrutiny for all of them would leave states unable to run orderly elections, while deference for all of them would let a state impose heavy costs on a fundamental right by pointing to any plausible interest.
The scale has two familiar ends. A severe restriction must satisfy strict scrutiny, but when a state election law imposes only “reasonable, non-discriminatory restrictions” on the rights of voters, “the State’s important regulatory interests are generally sufficient to justify” the restrictions.4 Burdens between the extremes call for a court to weigh the particulars of the rule and the record. The next subsection teaches that weighing, and the two after it take up the strict and deferential ends. The later sections apply the framework to Indiana’s identification law, where Justices who agreed on the method divided over how heavy the burden was and how much weight the state’s interests could carry against it.
The Balancing Inquiry
The balancing inquiry is the core of the framework, and the lead opinion in Crawford presents it as a method rather than a formula. It explains that Anderson declined to apply any “litmus test” that would neatly separate valid from invalid election restrictions; a court must instead identify and evaluate the interests the State offers as justification for the burden its rule imposes, and then make the “hard judgment” that the adversary system demands.5 Burdick supplies the fullest statement of that judgment, and Justice Souter’s dissent quotes it: a court “must weigh ‘the character and magnitude of the asserted injury to the rights protected by the First and Fourteenth Amendments that the plaintiff seeks to vindicate’ against ‘the precise interests put forward by the State as justifications for the burden imposed by its rule,’ taking into consideration ‘the extent to which those interests make it necessary to burden the plaintiff’s rights.’”6 The lead opinion and the dissent reached opposite results in Crawford, yet both used this weighing as the governing standard, which is why it works as the framework’s shared core.7
A sequence of questions replaces a fixed tier because both sides of the scale carry constitutional weight. The Court treats suffrage as fundamental and insists that any alleged infringement be carefully and meticulously scrutinized,10 yet government must also regulate elections substantially if they are to be fair and orderly.11 A formula that fixed the level of review in advance would ignore one side or the other. Strict review for every rule would put routine administration under the heaviest test, and deference for every rule would let a state justify a heavy burden by naming any plausible goal. The Court therefore lets the scrutiny vary with the effect of the regulation at issue.12
The first step asks how much the rule hurts, and the answer sets how hard the court looks at everything after it. The rigor of the inquiry depends on the extent to which the regulation burdens First and Fourteenth Amendment rights, which calls for assessing the character and magnitude of the threatened injury and estimating how many voters it will likely affect.13 The Court has identified no litmus test for measuring severity, and even a burden that looks slight must be justified by relevant and legitimate state interests “sufficiently weighty to justify the limitation.”14 No election rule therefore escapes review, and none draws strict review automatically. Justice Souter added that a plaintiff need not establish his claims beyond mathematical doubt; showing that serious burdens are likely is enough.15
Anderson shows the weighing with the state on the losing side. Ohio required an independent candidate for President to file a nominating petition in March, months before the general election and before the major parties had to name their candidates.16 A five-Justice majority weighed the impairment of voters’ First and Fourteenth Amendment rights against the degree to which the deadline advanced important state interests. It concluded that a state could justify some cutoff but that its interests in political stability, voter awareness, and equal treatment of candidates did not justify so significant a restriction at that early date.17 The same inquiry can yield the opposite result when the findings differ. The lead opinion in Crawford sustained Indiana’s photo identification law against a facial challenge, finding that it “imposes only a limited burden on voters’ rights.”18 Justice Souter, applying the same Burdick formulation to the same statute, found nontrivial burdens on tens of thousands of citizens, a significant percentage of whom were likely to be deterred from voting.19 The two opinions share one test and differ in the first-step finding, and that finding shaped how closely each examined the state’s interests.
The second step turns on one word in the formulation: precise. Naming a value that elections serve does not fill the state’s side of the scale. Justice Souter stated the consequence directly: a State may not burden the right to vote merely by invoking abstract interests, whether legitimate or even compelling, but must make a particular, factual showing that threats to its interests outweigh the particular impediments it has imposed.20 The reason is fit. Preventing fraud describes why states have election laws at all, so the same goal could defend a rule that burdens ten voters and a rule that burdens ten thousand. The state’s generalities must therefore be narrowed to the aspect of each interest that the challenged law actually addresses, and only the weight of that aspect goes on the scale. A common error is to stop at the observation that preventing fraud is a legitimate interest.
How hard a court presses on precision depends on the burden found in step one. Justice Souter observed that the lead opinion, having found only limited burdens, avoided a hard look at the State’s claimed interests, whereas his own finding of burdens far from trivial obliged him to assess the precise interests rigorously. Step three discounts those interests a second time. Even when a state ties its interest to the aspect its rule addresses, the interest counts only to the extent that it makes the burden necessary, because a goal that a gentler rule could serve equally well explains little of the burden the state chose.21
The weight that necessity carries depends on where the burden sits on the scale. A regulation that severely restricts voting rights must be narrowly drawn to advance an interest of compelling importance, whereas one imposing only reasonable, nondiscriminatory restrictions is generally justified by the state’s important regulatory interests.22 A gentler alternative therefore counts far more against the first kind of rule than against the second. The next two subsections take up those two ends of the scale in turn.
The inquiry reviews regulations of how eligible people vote, so a court decides first whether a law regulates or excludes. A rule that bars a class of people from the ballot, such as the poll tax in Harper, is treated as invidious and draws heightened scrutiny.23 Only for a regulation does the court run the four questions above.
Severe Burdens
At the demanding end of the scale, a regulation that severely burdens the right to vote or to associate must be narrowly drawn to advance a state interest of compelling importance. The lead opinion in Crawford illustrates the track through Norman v. Reed: after identifying the burden Illinois placed on a political party’s access to the ballot, the Court demanded an interest “sufficiently weighty to justify the limitation” and concluded that the “severe restriction” was not justified by a narrowly drawn state interest of compelling importance.25 Severity is therefore the threshold fact. It decides which track a court takes before the court weighs any interest the state offers.26
Strict scrutiny is the Court’s answer when a restriction falls hard on a right it treats as fundamental. Suffrage preserves other basic civil and political rights, and any alleged infringement must be carefully and meticulously scrutinized.27 A state that severely burdens voting or association must therefore show a compelling interest and a rule drawn narrowly to serve it, so a restriction that a less burdensome alternative could replace fails however sensible its stated purpose sounds.
Ballot-access cases reach this track through the voters. The Constitution contains no express guarantee of a right to become a candidate, so states may restrict candidacy so long as the restrictions do not violate provisions of general application.28 A candidate likewise has no fundamental right of access to the ballot, and restrictions on access draw heightened scrutiny only when they substantially interfere with the fundamental right to vote for a candidate who reflects one’s views or with the First Amendment right to associate for political purposes and advance a candidate of one’s choosing.29 A rule aimed at new parties can therefore burden voting severely, since the people it shuts out are the supporters who would form the party and choose its candidates.
Williams v. Rhodes is the leading illustration. Ohio required a new party to submit petitions signed by voters equal to fifteen percent of the last gubernatorial vote and to file them by an early deadline, while a party that had won ten percent of the previous gubernatorial vote needed no petition at all.30 The Court applied strict scrutiny because the scheme burdened both the right to vote and the right to associate, so Ohio had to justify it with a compelling interest.31
Ohio offered three interests, and the Court’s answer to each shows what “narrowly drawn” demands. The interest in promoting a two-party system was served only by favoring two particular parties, the Democrats and Republicans. The interest in electing majority candidates was served only by suppressing the growth of new parties. The interest in avoiding voter confusion was legitimate, but the means Ohio chose were not necessary to achieve it.32 A real and even sensible goal therefore does not carry the burden alone, because the state must also show that this particular restriction was needed to reach it.
Norman reached the same result by a different route. Illinois conditioned ballot access for a new party on securing 25,000 signatures in each political subdivision, and it forbade a party established in one subdivision to use its name in another, even with the party’s permission.33 Williams turned on a high, early entry price that favored established parties; Norman turned on a requirement that a party prove its support again in every subdivision and could not carry its identity from one to the next. Because regulation restricting new parties’ ballot access must be narrowly drawn to advance interests of compelling importance, the Illinois scheme failed.34 A burden can become severe through the numbers demanded, the timing, the geography, or the control of the party’s name. What matters is the work the rule does to keep a group of voters from its preferred choice.
A condition tied to wealth reaches the same track. In Bullock v. Carter the Court struck down candidate filing fees as high as $8,900, and in Lubin v. Panish it struck down fees set at two percent of the office’s annual salary, because both laws prevented less affluent groups from voting for a candidate who shared their views.35 Applying strict scrutiny, the Court noted that a signature requirement would control the length of the ballot and discourage frivolous candidacies with less discrimination, and that a fee scheme must give indigent candidates an alternative way to qualify.36 Signature requirements themselves usually survive, since a state may demand a significant modicum of support before it lists a candidate, and the Court has sustained a requirement of signatures from five percent of eligible voters within 180 days. Strict scrutiny enters only when the number is set so high that it keeps independent candidates and minor parties off the ballot.37 The dividing line is whether the ballot remains reachable, not whether a petition or deadline exists.
Justice Scalia’s concurrence in Crawford gives the clearest account of what makes a burden severe. The first step is to decide whether the law severely burdens the right to vote, and ordinary, widespread burdens that require only “nominal effort” of everyone are not severe.38 A burden is severe when it goes beyond the merely inconvenient; he read Storer v. Brown as treating the law in Williams as severe because it was so burdensome as to be “virtually impossible” to satisfy.39 The lead opinion, by contrast, said that the Court has identified no litmus test for measuring severity and that even a slight burden must be justified by relevant and legitimate interests.40 The two approaches differ on how much the severity finding controls, but Williams and Norman fall on the severe side under both.
Crawford itself shows the filter at work. The lead opinion assumed that the law may burden some voters specially but found that the challengers had assembled no evidence showing the burden severe enough to warrant strict scrutiny,41 and the record said virtually nothing about the difficulties of indigent voters or voters with religious objections to being photographed.42 Justice Scalia went further and called the law a single uniform burden: everyone voting in person must have and present a photo identification that can be obtained for free.43 Voting’s importance explains why a court weighs the burden carefully; it does not by itself set the standard at strict scrutiny, which is reserved for burdens beyond the inconvenient.
The hypothetical lands on the strict side for the same reasons Williams does. The free photo identification in Crawford lands on the other side for the reasons Justice Scalia gave: a document available at no cost, required of everyone, asks for ordinary effort and leaves every eligible voter able to reach the ballot.
Nonsevere Burdens
The deferential end of the scale governs reasonable, nondiscriminatory rules, and Burdick supplies the standard. Justice Scalia’s concurrence in Crawford describes it as a deferential “important regulatory interests” standard for nonsevere, nondiscriminatory restrictions, with strict scrutiny reserved for laws that severely restrict the right to vote.49 Burdick’s own formulation is that when a state election law imposes only “reasonable, nondiscriminatory restrictions” on the rights of voters, “the State’s important regulatory interests are generally sufficient to justify” them.50 The lead opinion in Crawford landed on this side of the scale when it concluded that, viewed in its broad application to all Indiana voters, the statute “imposes only a limited burden on voters’ rights.”51 Before asking whether an identification law belongs there, watch the track operate in the cases that built it: Burdick, Timmons, and Clingman.
Burdick’s formulation names two conditions, and each does separate work. The restriction must be nondiscriminatory, meaning that it applies evenhandedly. Indiana’s identification rule drew no classifications beyond optional absentee and provisional balloting for certain voters,52 and Minnesota’s fusion ban applied to major and minor parties alike.53 The restriction must also be reasonable, and in the cases below reasonableness turned largely on whether voters kept other adequate ways to reach the ballot and express a choice.
Burdick shows the second condition at work. Hawaii prohibited write-in votes in general elections, and the motive for challenging the ban was plain: a third of the elections for the state legislature were unopposed, and many voters preferred to cast blank ballots rather than endorse the only candidate listed.54 The Court nonetheless treated the ban as a mild burden, because Hawaii offered several easy paths onto the ballot. The state ran open primaries, new parties could qualify through primary petitions carrying the signatures of one percent of voters, and independent candidates could file on the nonpartisan primary ballot with nominating papers bearing fifteen to twenty-five signatures.55 In light of that adequate ballot access, the ban imposed only a limited burden, so Hawaii did not need a compelling interest. Its interest in avoiding “unrestrained factionalism” at the general election outweighed the petitioner’s “limited interest in waiting until the eleventh hour to choose his preferred candidate.”56
The Court said that when a state’s ballot access laws pass constitutional muster, a prohibition on write-in voting is presumptively valid, because the objection then amounts to an insistence that the state record, count, and publish individual protests against the election system or the choices on the ballot.57 Adequate access earlier in the process is what made the late closing tolerable. Justice Kennedy, joined by Justices Blackmun and Stevens, dissented on the ground that, with so many candidates running unopposed, the ban left dissidents “no way to cast a meaningful vote.”58 The split shows that choosing the deferential track does not end the argument; the dispute moves to how heavy the burden really is.
Timmons v. Twin Cities Area New Party applies the same track to a different rule and brings the evenhandedness condition forward. Fusion is a practice, dating to the nineteenth century, in which a minor party lists another party’s nominee as its own, and it has historically helped third parties build coalitions with larger ones.59 Minnesota banned it, and the Court held that the ban was not a severe burden. Chief Justice Rehnquist reasoned that the ban applies to major and minor parties alike and simply keeps a candidate already nominated by one party from appearing on the ballot as another party’s candidate, leaving the minor party free to persuade someone else to run as its candidate.60 The burdens, though “not trivial,” were not severe, so Minnesota needed only an important interest, and the Court found one in the integrity, fairness, and efficiency of the election process and in the stability of the political system.61
The party argued that fusion is necessary for minor parties to thrive, and the Court was unmoved. Many features of the American system, such as single-member districts, first-past-the-post elections, and the high cost of campaigning, already make life hard for third parties, and the Constitution no more requires a state to permit fusion than it requires proportional representation or public financing of campaigns.62 The party also claimed a right to use the ballot to send a message about the nature of its support for a candidate, but ballots serve primarily to elect candidates and are not forums for political expression.63 The stability interest has a ceiling. It does not permit a state to completely insulate the two-party system from the competition of minor parties and independents, yet it does permit reasonable regulations that may in practice favor the traditional two-party system and temper the destabilizing effects of party-splintering and excessive factionalism.64 The reasoning reflects a picture of democracy in which two major parties dominate while the threat of outside competition keeps them responsive, so a state may protect the system’s stability but may not eliminate the pressure that keeps it responsive.65
Clingman v. Beaver shows how far the deferential track reaches. Oklahoma’s semiclosed primary let a party invite only its own members and registered Independents, so the Libertarian Party could not invite registered Republicans and Democrats to vote in its primary. The Court called these “minor barriers between voter and party” that do not compel strict scrutiny, and it found Oklahoma’s interests in preserving party identity and aiding party-building reasonable and politically neutral.66 Justice Scalia reads Clingman as reaffirming that strict scrutiny is appropriate only if the burden is severe, and he treats ordinary, widespread burdens that require only “nominal effort” of everyone as nonsevere.67
Williams v. Rhodes marks the point where the deferential track ends. Justice Scalia reads that case as severe because the Ohio law was so burdensome as to be “virtually impossible” to satisfy, the opposite of the several easy paths that kept Hawaii’s write-in ban mild.68 Which track a rule lands on therefore depends on available alternatives as well as on evenhandedness.
The same sorting decides where voter identification belongs. Justice Scalia would treat Indiana’s law as a single burden that falls uniformly on everyone: to vote in person, every voter must have and present a photo identification that can be obtained for free, and the state draws no classifications except to create optional absentee and provisional balloting for certain poor, elderly, and institutionalized voters and for religious objectors.69 A uniform condition that demands ordinary effort meets both conditions of the deferential track, so on his account the state’s important regulatory interests suffice. The lead opinion reached the same result through the weighing inquiry. It assessed the statute’s broad application to all voters, found a limited burden, and held that the state’s precise interests defeated the facial challenge.70 Whether that characterization survives when attention turns to the voters for whom the law is heaviest, among them elderly people born out of state, indigent voters, and voters with religious objections to being photographed, is a question the later sections take up. The record the lead opinion reviewed said virtually nothing about the last two groups, which is part of why the facial challenge failed.71
The usual error on this track is to hear “nonsevere” as “unreviewable.” Burdick still weighed the petitioner’s interest in a late write-in against the state’s interest in avoiding factionalism, and Timmons still required an important interest and set a ceiling on the stability rationale. The word “generally” in Burdick’s formulation leaves room for a reasonable-looking rule to fail when the state’s interest is thin or the rule targets a particular group. The deferential track gives the state a presumption in its favor, but the state must still point to an interest the rule serves.