After NFIB v. Sebelius, When Does the Cost of Voting Become an Illegal Poll Tax?
TOURO LAW JOURNAL OF RACE
AFTER NFIB V. SEBELIUS, WHEN DOES THE COST OF VOTING BECOME AN ILLEGAL POLL TAX?
ANDRE L. SMITH
State, county and local legislatures in the South are enacting laws designed to reduce the participation of Blacks, Hispanics and the poor of all races in the voting franchise. Devices being used to prevent or discourage wide voter participation include photo identification laws requiring would be voters to obtain a particular type of photo identification and prohibiting forms of identification, which are more cheaply and easily obtained. They also include reducing the number of polling places and changing their location, both which impose significant direct and indirect costs on voters relating to time and travel, as well as opportunity costs with respect to missed work. The Attorney General of the United States recognizes over 15 types of election changes that could disparately impact minority voters, from voter registration procedures and language requirements to redistricting and annexation of voter precincts. The Voting Rights Act authorized the Department of Justice to review all election changes from 'covered' States.
Southern legislatures are emboldened not only by the judicial repeal of
the Voting Rights Act, but also by the Court’s decision in Crawford v. Marion
County6, upholding voter identification laws. The cost of obtaining photo
identification may be negligible and insignificant for those with a modicum of
wealth. But for those without resources, the direct, indirect and opportunity
costs can be significant, even substantial. Photo ID laws by themselves may
reduce the number of minority voters by over 700,000.7 If poor white voters
are included, this number should easily exceed one, perhaps two million
disenfranchised American citizens. Proponents of voter identification laws
cite the government’s legitimate interest in preventing voter fraud. But
while the interest is legitimate, the actual threat is minimal and references
to it are clearly pretextual. Recently, former U.S. Secretary of State, Colin
Powell asked, “how can voter fraud be widespread and undetected?”8
With the Voting Rights Act repealed, are there any other federal laws
that might serve to prevent the disenfranchisement of minorities and the
poor? Is there any other legal basis for challenging voter ID laws and other
laws that disenfranchise minority voters by increasing the indirect and direct
costs of participation? In The Cost of the Vote,9 Atiba Ellis flirts with the idea
that undue, legislatively induced costs associated with voting could fall
within the definition of poll taxes and are thus illegal pursuant to the 24th
Amendment to the U.S. Constitution10 and Harper v. Virginia (1966).11
Neither Congress nor any state legislature can require otherwise
eligible voters to pay a fee for the privilege of exercising the franchise.
Historically, poll taxes were designed to disenfranchise those without
property and marginalize their civic participation and their material
interests. The Harper Court rejected the traditional rationale that those with
property could effectively represent those who are without property and who
are supposedly uninterested in politics in the first place. The right to vote is
6 Crawford v. Marion County Election Board, 553 U.S. 181 (2008).
7 Broussard Brief supra note 2, at 17, citing Jon C. Rogowski & Cathy Cohen, Turning Back the
Clock on Voting
Rights, The Impact of New Photo Identification Requirement on Young People of Color, The
African American Youth Project, available at
8 Remarks of Colin Powell, August 22, 2013, North Carolina CEO Forum, Raleigh, North Carolina.
9 Atiba Ellis, The Cost of the Vote: Voter Identification Laws, and the Price of Democracy, 86
DENV. U. L. REV. 1023 (2009).
10 XXIV. Amend outlawed the “poll tax”; Harper v. Virginia Board of Elections 383 U.S. 663
11 Harper v. Virginia Board of Elections, 383 U.S. 663 (1966).
fundamental and one’s wealth has no rational relationship with voting
Perhaps, Professor Ellis is on to something. Can the Poll Tax
Amendment serve the same purposes as the Voting Rights Act, “to correct the
actions of States and jurisdictions which intentionally sought to exclude
citizens from exercising their right to vote and to deter States from either
deliberate attempts at future exclusion or deter legislative actions that,
veiled as ‘fraud prevention,’ deny minorities their right to vote?” 13 The
prohibition against poll taxes could not possibly reach nearly all things
covered by the Voting Rights Act, but it could apply to some.
So when do costs associated with voting mandated by a legislature
amount to an unconstitutional poll tax? Chief Justice Roberts’ functional
analysis of the ‘Affordable Care Act’ and determination that the penalty
provision, or “shared responsibility payment,” constitutes a tax for purposes
of the Constitution gives cause to consider whether voting-related laws that
impose undue burdens on the poor can be characterized as illegal poll taxes.14
The Court in NFIB v. Sebelius held that the labels a legislature affixes
to enactments do not control the determination of whether something is
properly characterized a “tax.”’15 The Court in NFIB found that “the shared
responsibility payment” of the Affordable Care Act was a tax.16 Conversely,
they have previously found that other enactments labeled as taxes, like the
“child labor tax” in Bailey v. Drexel Furniture were actually “penalties” and
not taxes at all. Just as it is hard to distinguish penalties from taxes, the
Court also has difficulty distinguishing taxes from fees.
While NFIB v. Sebelius gives cause for reconsideration, it does not
necessarily give cause for much hope. The Court did not provide a precise
definition; instead it identified factors relevant to determining when
something may reasonably be called a tax. Some of these factors support
characterizing the costs associated with voting as an illegal poll tax. Some of
the factors do not. While Justice Roberts opinion in Sebelius provides some
ammunition for voting rights advocates, it also provides ample means for a
state or local legislature to avoid the characterization.
13 Broussard Brief, supra note 2, at 19.
14 National Federation of Independent Business (NFIB) v. Sebelius, 132 S.Ct. 2566, 2582 (2013)
(opinion of Roberts, C.J.).
15 Id. at 2582.
When it comes to designating something a poll tax, the Court
ultimately requires its lower divisions to “give practical effect to the
legislature’s enactment.” And they have, to some extent. In Common
Cause/Georgia v. Billups I, the U.S. District Court for the Northern District
of Georgia struck down that state’s Photo ID Act of 2005 as an
unconstitutional poll tax because otherwise eligible voters without a driver’s
license (somewhere between 300,000 and 1,000,000 people) were required to
purchase a photo voter identification card from the state for $20.17
To avoid the poll tax characterization, the Georgia legislature amended
the law (Photo ID Act of 2006) to remove the fee for a voter identification card
and also to expand the privilege of absentee voting, which does not require
identification. It worked. The district court in Common Cause/Georgia v.
Billups II, citing an Indiana federal district court dealing with a substantially
similar photo ID law, held that indirect and opportunity costs relating to
legitimate voting regulations never constitute a poll tax.18 The court did not
question the legitimacy of the state’s interest in preventing voter fraud, or
whether the law was even rationally related to furthering that interest,
despite the fact the State Secretary testified that voter fraud was much more
likely to occur with respect to absentee voting, which the Photo ID Act of
2006 specifically expanded.19
The district court paid lip service to Harman v. Forssenius,20 which
held that the 24th Amendment prevented the state of Virginia from requiring
voters to choose between paying a fee or filing a certificate of residence.
According to the Court, “Constitutional rights would be of little value if they
could be . . . indirectly denied, or manipulated out of existence. . . . Thus, like
the Fifteenth, the Twenty-fourth [Amendment] nullified sophisticated as well
as simple-minded modes’ of impairing the right guaranteed. It hits onerous
procedural requirements which effectively handicap exercise of the franchise.”
However, without any further reference to Harman, the district court
turned around cited Burdick v. Takushi for the proposition that, “election
laws will invariably impose some burden upon individual voters. . . . Thus,
the imposition of tangential burdens does not transform a regulation into a
poll tax.”21 Then, the court overstated what was an otherwise reasonable
point: “the cost of time and transportation cannot plausibly qualify as a
17 Common Cause/Georgia v. Billups, 406 F. Supp. 2d 1326 (N.D. Ga. 2005).
18 Common Cause/Georgia v. Billups, 439 F. Supp. 2d 1294, 1354 (N.D. Ga. 2006).
19 Id. at 1356-57.
20 Id. at 1352 citing Harman v. Forssenius, 380 U.S. 528 (1965).
21 Id. at 1354 citing Burdick v. Takushi, 504 U.S. 428, 433 (1992).
prohibited poll tax because those same ‘costs’ also result from voter
registration and in person voting requirements.” 22 Harman v. Forssenius
clearly stands for the proposition that the 24th amendment prohibits the
legislature from mandating any direct cost relating to voting along with
undue indirect costs that produce the same effect, disenfranchisement of poor
voters. Thus, at some point the cost of time and transportation imposed by a
legislature must qualify as a poll tax, even if the Georgia Photo ID Act of
2006 at issue was reasonable.
That indirect and opportunity costs can never be a poll tax is a dubious
proposition. Neutrally worded voting regulations that further a legitimate
government interest still amount to an impermissible burden under the
Burdick standard if the burdens imposed far exceed marginal furtherance of
the government’s interest. These burdens are not just taxing, they are a tax.
Thus, despite the overstatements of the Georgia and Indiana district courts,
at some point, a combination of requiring specific forms of identification,
scheduling elections for days when there is little public transportation,
offering only one or a few polling locations located in obscure places with
defective machines, along with other costly regulatory devices, constitutes a
In a sense, the standard in Burdick v. Takushi for Equal Protection
Clause challenges to voter regulations is entirely consistent with the Harman
Court standard for determining whether regulations create an impermissible
poll tax. The Burdick test requires courts to weigh the interest of the
government against the burdens placed on discrete groups of voters.
Synthesizing Harman v. Forssenius and Harper v. Virginia with Burdick v.
Takushi, a direct cost to other eligible voters always outweighs any
government interest, no matter how compelling. Thus, a poll tax or any
direct cost of obtaining the voting franchise is a per se violation of both the
Equal Protection Clause and the 24th Amendment abolishing poll taxes in
Still, with respect to either Equal Protection or the 24th Amendment,
the Burdick standard is somewhat wanting. Government interest in
preventing voter fraud is legitimate, even if not compelling. But to the extent
a cost-inducing government regulation related to voting does not
demonstrably further this or some other legitimate legislative purpose, it
should constitute a violation of both the Equal Protection Clause and the 24th
Amendment. Of course, such a rule involves a bit of judicial second-guessing
of the legislature. On one hand, Shelby County, Ala. v. Holder sets a
precedent for this trend, since the Court required Congress to engage in a
copious statistical analysis of race relations before they are allowed to
legislate in favor of minorities. On the other, the Court in Crawford v.
Indiana allowed the state of Indiana to determine the prevalence of voter
fraud on pure belief and faith.