Concept
Measuring the Burden
Right to Vote
Measuring the Burden
The Crawford opinions disagree about the burden before they disagree about the state's interests, and the first disagreement shapes the second. Under the balancing approach, a court weighs the asserted injury to the right to vote against the precise interests the state offers. The scrutiny varies with the effect of the regulation, no litmus test measures severity, and even a slight burden must be justified by interests sufficiently weighty.1 A severe burden brings strict scrutiny,2 a demand like the exacting scrutiny the Court applies to compelled disclosure, where a mere legitimate interest cannot justify a significant encroachment and the state's interest must bear a substantial relation to what the rule demands.3 A limited burden calls instead for a deferential look at whether the state's regulatory interests are important.4 Older eligibility cases show how much the choice matters: one Justice would have upheld a classification so long as it was rationally related to a permissible legislative end, while the Court reversed because the classification included many persons with a remote and indirect interest and excluded others with a direct one.5
The Crawford Justices disagreed about the burden on two levels. The first is how heavy it is. The lead opinion found it limited for most voters and unproven for a narrow group who may bear more, while the dissenting Justices weighed the same law and found its burden on poor, elderly, and disabled voters too great for Indiana's interests to justify.6 The second is whose burden counts. Justice Scalia, concurring in the judgment, would measure only the burden the law places on everyone, and he treated differences among voters as variations of that single burden.7 The three subsections below follow those questions from the typical voter to the vulnerable voter to the choice of yardstick.
Burden on Most Voters
The lead opinion sets the voter's side of the scale by asking what the identification requirement adds to what voting already asks of most people. For most voters who need identification, it concluded, the trip to the Bureau of Motor Vehicles (BMV), the gathering of documents, and the photograph “surely does not qualify as a substantial burden on the right to vote, or even represent a significant increase over the usual burdens of voting.”8
The opinion first fixes which burdens count: those imposed on persons who are eligible to vote but lack a current photo identification that satisfies the statute.9 A voter who already holds a valid driver's license is unaffected, so the question is what it costs to obtain the missing document. Harper sets the limit. The fact that most voters already hold acceptable identification would not save the statute if the state required voters to pay a tax or a fee for a new photo identification, because the convenience of the majority cannot excuse a price charged to a minority.10 Indiana's cards are free, as other states' voter registration cards are.11 That comparison places the card among the ordinary administrative steps of voting and away from the wealth conditions Harper forbids.
The phrase “usual burdens of voting” supplies the yardstick. Voting has never been effortless: a voter must register, find the polling place, and travel there. The question is therefore what the new rule adds, and a modest errand that resembles those existing steps is no “significant increase.”12 Measuring the increment lets a court separate rules that change what it takes to vote from rules that ask for the kind of effort voting already involves.
The missing fee carries the analysis. The same errand to an office that charged for the card would run into Harper, and the free price is what lets the lead opinion treat the remaining steps as inconvenience.13
A different burden falls on the voter who holds the right identification and meets trouble on election day, such as a lost card or a stolen wallet. The lead opinion held that burdens of that sort, arising from life's vagaries, are neither so serious nor so frequent as to raise any question about the statute's constitutionality, because the right to cast a provisional ballot is an adequate remedy.14 A provisional ballot is cast on election day and counted once the voter completes a later step. In Indiana that step is a trip to the circuit court clerk's office within ten days to execute an affidavit.15 The ballot survives, and the mishap costs an errand and leaves the vote intact. Because such mishaps scatter across the electorate and the provisional ballot cures them, they do not describe the law's effect on any identifiable group.
The elderly voter born out of state sits on the other side of the line the lead opinion drew. That voter may have difficulty obtaining the birth certificate needed to get the free card, and the opinion recognized that the requirement may have imposed a special burden on such a person's right to vote.16 The stolen wallet is a one-day accident that a provisional ballot absorbs, while this obstacle stands in front of the card itself, so the free price does not solve it.
A finding that the burden is insubstantial for most voters settles half of the question. The other half is how heavily the law falls on the voters for whom compliance is hardest, and there the lead opinion and the dissenting Justices part ways.
Burden on Vulnerable Voters
Having found the burden insubstantial for most voters, the lead opinion turned to the voters for whom it is heavier. Record evidence and facts subject to judicial notice, it said, show that “a somewhat heavier burden may be placed on a limited number of persons.”18 It named four groups: elderly persons born out of state, who may have difficulty obtaining a birth certificate; persons who, because of economic or other personal limitations, find it hard to secure a birth certificate or assemble the other required documents; homeless persons; and persons with a religious objection to being photographed.19 If some members of those classes were registered voters when the statute was enacted, the requirement may have imposed a special burden on their right to vote.20
The free card explains why these groups stand apart. A free card ends the Harper fee problem, but it still depends on underlying documents, a trip to the BMV, and a photograph, and each group meets a different one of those steps. A person born elsewhere must obtain a certificate from another state's records. A person with few resources faces a real cost, since the record acknowledges that obtaining a birth certificate carries a financial cost.21 The one homeless woman in the record had a copy of her birth certificate and was still denied a card because she had no address.22 For a person whose faith forbids being photographed, the photograph itself is the obstacle. The phrase “limited number” says how many people are affected and leaves open how heavy the burden is for each of them.
The lead opinion treated the provisional ballot as the main mitigation. Eligible voters without photo identification may cast provisional ballots that will ultimately be counted, but they must travel to the circuit court clerk's office within ten days to execute an affidavit, and the opinion thought it unlikely that such a requirement would pose a constitutional problem unless it is wholly unjustified.23 The reasoning is that a post-election errand preserves the vote, so the cost is time and travel and the ballot is not lost. A footnote adds that most provisional voters will presumably be able to obtain photo identification before the next election. It also finds it hard to understand why voters with a faith-based objection to being photographed must vote provisionally in every election when the BMV already issues such citizens special licenses that let them drive without any photo identification.24
The challengers asked the Court to weigh the burden on voters who cannot afford or obtain a birth certificate, and who must make a second trip to the clerk, against the state's interests. The lead opinion answered that the record made it impossible to quantify either the magnitude of the burden on that narrow class or the portion of it that is fully justified.25 That answer concerns proof. A balancing test weighs the injury, and a court that cannot size the injury has nothing to place on the voters' side of the scale. The challengers also attacked the statute in all its applications, which carries a heavy burden of persuasion,26 and the lead opinion added that a burden unjustified as to a few voters would not by itself establish the relief they sought.27
The gaps in the record were specific. It gave no count of registered voters without photo identification, because the district judge found the challengers' expert report “utterly incredible and unreliable.” It held one affidavit from a homeless woman with a birth certificate and no address, and that single affidavit says nothing about how common her problem is.28 It offered no rough estimate of how many indigent voters lack copies of their birth certificates, and supposition based on extensive Internet research cannot substitute for admissible evidence subject to cross-examination.29 Elderly voters in Indiana, finally, may have an easier time obtaining a card than the nonelderly and may vote absentee without presenting photo identification, though the opinion conceded that this may not be a completely acceptable alternative. On that record the opinion could not conclude that the statute imposes “excessively burdensome requirements” on any class of voters.30
The lead opinion therefore decides on proof and leaves the merits of the vulnerable voters' claims open. It accepts that the groups exist and that the requirement may have burdened them specially, and it holds that this record shows neither how heavy nor how widespread that burden is. A common mistake is to report that the lead opinion rejected those claims on the merits.
The dissenting Justices weighed the same law differently. Justice Souter, joined by Justice Ginsburg, agreed with the plurality that a balancing analysis governs and described it as a sliding scale in which the scrutiny varies with the effect of the regulation.31 Justice Breyer's separate dissent applied that balancing and found that the burden of Indiana's unusually restrictive law, particularly on poor, elderly, and disabled voters, was not adequately justified by the state's asserted interests.32 Both the lead opinion and the dissents balanced, then. They differed on what the record showed about the burden and on how heavily it weighed against Indiana's reasons.
Whose Burden Counts
Before a court can weigh a burden, it has to identify it. Justice Scalia, joined by Justices Thomas and Alito and concurring in the judgment, identified the burden categorically: everyone who votes in person in Indiana must have and present a photo identification that can be obtained for free.36 What the challengers described as the law's several light and heavy burdens were, in his account, only the different impacts of that single burden on different voters.37 Voters who already hold identification do not escape the burden either, since they must keep the information on the card accurate, renew it before it expires, and replace it if it is lost.38
The lead opinion took a different route. It assumed the challengers' premise that the law may have imposed a special burden on some voters, but held that they had not assembled evidence showing that burden severe enough to warrant strict scrutiny. Justice Scalia preferred to hold the premise irrelevant and the burden minimal and justified.39 The choice of method decides how much the evidence about vulnerable voters in the last subsection can matter.
The categorical method begins with the question that sorts the two tracks. Strict scrutiny is appropriate only if the burden is severe, so a court first decides whether the challenged law severely burdens the right to vote. Ordinary and widespread burdens that require “nominal effort” of everyone are not severe, and burdens are severe only if they go beyond the merely inconvenient.40 A law that is “virtually impossible” to satisfy crosses that line, as Storer v. Brown described the law in Williams v. Rhodes.41 Nonsevere, nondiscriminatory restrictions then receive a deferential “important regulatory interests” standard.42
Justice Scalia gave the method three supports. The first is that the precedents measured burdens generally. In Burdick v. Takushi, the Court found that Hawaii's write-in ban imposed only a limited burden by considering the laws and their reasonably foreseeable effect on voters generally, and it never asked how the ban affected Mr. Burdick given his circumstances. Looking at the effect on “some voters” was the approach of the Burdick dissenters.43 Storer measured severity by a regulation's “nature, extent, and likely impact” and told the lower court to ask whether a reasonably diligent independent candidate could be expected to satisfy the signature requirements, without asking whether one of the petitioners would find them harder. What mattered was the general assessment of the burden.44
The second support is that the rule for ordinary burdens would mean nothing otherwise. Clingman's holding that burdens are not severe if they are ordinary and widespread would be rendered meaningless if a single plaintiff could claim a severe burden.45 Nearly any election rule falls harder on someone, through distance, disability, or poverty. If one voter's hardship could relabel the rule severe, every ordinary requirement would face strict scrutiny on the strength of a single sympathetic plaintiff.
The third support is structural. The Constitution commits the task of election regulation to the States in Article I, § 4, and detailed judicial supervision of the election process would flout that commitment.46 State legislatures weigh the costs and benefits of changes to their election codes, and their judgment prevails unless it imposes a severe and unjustified overall burden or is intended to disadvantage a particular class. Judicial review must therefore apply an objective, uniform standard that lets them determine, ex ante, whether a burden is too severe.47 On this account Indiana's law draws no classifications, let alone discriminatory ones, apart from the optional absentee and provisional ballots it offers certain voters.48 Those accommodations are an indulgence and carry no constitutional mandate.49
The categorical method does not immunize every law. A law that is virtually impossible for voters generally to satisfy is severe on this method, and even a nonsevere burden must be justified by important regulatory interests. What the method removes is the adjustment of the severity finding for one voter's circumstances.50
Justice Scalia attacked the lead opinion's record-based resolution on the ground that it neither rejects nor embraces the rule of the precedents, provides no certainty, and will embolden litigants who suspect the precedents have been abandoned.51 The lead opinion found the burden “neither so serious nor so frequent as to raise any question about the [law's] constitutionality,”52 and it added that a burden unjustified as to a few voters would not establish the relief the challengers sought.53 The dissenting Justices, who balanced as the previous subsection describes, weighed how heavily the law fell on the poor, elderly, and disabled voters it affects most, and Justice Breyer found that burden inadequately justified.54
The three methods place the cost of uncertainty differently. Under the categorical rule, a legislature can know before an election whether a rule's general effect is acceptable, and the state does not have to defend each rule against the circumstances of every voter. Under the record-based and ground-level methods, a neutral rule can still be tested against the cost it imposes on an identifiable group, so the voters with the hardest path to the ballot are not shut out of the analysis, but the state must expect litigation over each rule's real-world effect. The lead opinion and the concurrence reached the same judgment by different routes, so the choice of method did not change the outcome in Crawford. It matters for how later courts choose among the approaches, a point taken up in a later section.
Two mistakes recur on exams. The first describes the categorical view as ignoring voters' hardship altogether, when it still measures the law's general weight and still requires justification for a nonsevere burden. The second attributes that view to the lead opinion, which treated the burden on a few voters as possibly special and rejected invalidation for want of proof.