STUART MINOR BENJAMIN 0 1
0 Copyright © 2012 by Stuart Minor Benjamin. This article is also available at
1 Douglas B. Maggs Professor of Law, Duke Law School. I would like to thank Matt Adler , Randy Barnett, Mitch Berman, Joseph Blocher, Jamie Boyle, Curt Bradley, Josh Chafetz, Guy Charles, Bob Cooter, Fred Kameny, Arti Rai, Stephen Sachs, Fred Schauer, Neil Siegel, Ilya Somin , and Ernie Young for useful comments and conversations, and Abby Liebeskind for research assistance. All errors are mine, of course
Law often develops through accretion. The common law and precedent involve courts building on what came before as they reach new decisions. The earlier decisions lay the groundwork for later ones. Slippery slopes are a particular kind of accretion: a conclusion that might seem unreasonable if taken in one fell swoop becomes reasonable because of the many small steps that precede it. In both situations, a decision depends on earlier ones-or, to put the point differently, earlier decisions enable the later one. This is of course a familiar story, and the accretive process of precedent is largely taken for granted in the United States. Accretion often involves many different actors and many different steps. Consider a more directed, and direct, process: an actor undertakes action Y that enables that actor's action Z. As with the development of precedent, the later action depends on the earlier one, but in this case a given actor is creating the conditions for its later actions. The actor is engaged in bootstrapping-using its own volitional actions to empower it to take other actions. Every action a person can take depends on the circumstances created by prior actions. One can walk into a building because someone built that building, one can own money because a monetary system was created, etc. Some of those conditions precedent are created by the actor herself. For example, walking into a given building depends on earlier decisions to be in that city and to go to that building. A person routinely enables her own action Z by first undertaking action Y without which Z would be impossible. Government institutions operate similarly: they rely on the earlier acts of others, for instance those who founded the nation or those who created the businesses they regulate. And government institutions rely in part on the acts of their predecessors. A Supreme Court holding on any issue depends on a series of decisions that its predecessors made. This article focuses on bootstrapping based not on the physical enablement of Z but instead on its legal enablement-situations in which action Y transforms Z from legally problematic, if not clearly impermissible, to legally
permissible. That is, an actor undertakes permissible action Y and thereby
renders its action Z legally permissible, as the actor’s undertaking of Z absent Y
would raise serious legal problems. For example, the Supreme Court’s ruling in
Bolling v. Sharpe1 would have been difficult to defend absent Brown v. Board of
Education.2 Brown in turn relied on Sweatt v. Painter3 and other earlier cases
that had chipped away at the “separate but equal” doctrine, and later cases that
would have seemed barely conceivable absent Brown relied on Brown for
The idea behind bootstrapping is that by undertaking Y, an actor creates the
conditions that enable that actor to undertake some further action Z. This is
different from estoppel, in which others’ failure to object to earlier Ys is treated
as preventing them from objecting to a new iteration of Y. In that situation, the
earlier Ys have played an important role for the actor, but that role is to
consolidate its ability to undertake similar actions.4 Bootstrapping is different in
that the actor is undertaking a further action that is beyond its precedents, and
it does not rely on others’ failure to object. The point of estoppel is that the
actor says, in effect, “I am doing the exact same thing that I did in the past, and
your lack of concern about those earlier actions prevents you from objecting
now.” Bootstrapping, on the other hand, involves a new action that goes beyond
what was done in the past.
Bootstrapping is also different from estoppel in that estoppel may rely on Ys
undertaken by others, but in bootstrapping the same actor undertakes Y and Z.
The same point applies in distinguishing bootstrapping from path dependence
or a slippery slope.5 Neither path dependence nor a slippery slope necessarily
involves a given actor building on its own prior actions. A key element of
bootstrapping is an actor using two or more steps to achieve an outcome it
could not achieve with a single step. The actor does not merely build on
conditions precedent that have arisen, but instead creates those conditions
precedent. A less important difference for my purposes between bootstrapping
and path dependence or a slippery slope is that path dependence often, and a
slippery slope by definition, entails many small steps that over time produce a
given result. Slippery slopes in particular involve an accretion of small measures
that results in a significant change. Bootstrapping, by contrast, often entails a
simple step function rather than a slope. Bootstrapping can involve a series of
No. 3 2012]
small changes that build up over time—and insofar as it does so, it overlaps with
a slippery slope created by a single actor—but it need not do so.
Bootstrapping has particular current salience in the context of the Patient
Protection and Affordable Care Act (ACA) of 2010.6 The main response to the
argument that the ACA exceeds Congress’s commerce power relies on two
apparent bootstraps. First, in the Emergency Medical Treatment and Active
Labor Act (EMTALA), Congress required emergency rooms to treat patients
whether they could pay or not, which means that individuals can receive health
care paid by others.7 Congress could then say that forgoing insurance affected
interstate commerce, because the guarantee of emergency room service meant
that the non-purchase of insurance shifted costs from some individuals onto
others. Second, in the ACA, Congress prohibited insurance companies from
disallowing coverage based on preexisting conditions and from charging higher
premiums because of a person’s medical condition or history.8 As many
commentators have noted, these prohibitions effectively necessitate an
individual mandate or something like it (or else many people would forgo
insurance in good health and sign insurance forms on their way to the hospital).9
What is wrong with bootstrapping? It seems to hit a constitutional nerve, as
it were, when government institutions engage in it. An entity takes action Y that
empowers it to take action Z that might further empower it. The institution is
able to ratchet up its power by itself, creating an upward spiral of its power
based on its own actions. The obvious concern is that an institution is able to
create and control the conditions that empower it. Giving government actors
the ability to determine their own limits—or lack thereof—allows for
aggrandizement. The key is that an entity is empowering itself, rather than
being part of a process involving many different entities. This raises fears about
an actor concentrating its own power, contrary to foundational principles
articulated in The Federalist10 and embodied in constitutional provisions
creating a government of limited, separated powers. Some bootstraps by
nongovernment actors may trouble us (to tweak the definition of chutzpah,
imagine that a person who kills her parents thereby becomes automatically
eligible for benefits as an orphan). But this article’s focus is on governmental
6. Pub. L. No. 111-148, 124 Stat. 119 (2010) (codified as amended in scattered sections of 42
7. 42 U.S.C.A. § 1395dd (West 2010).
8. Id. § 300gg-3a (“A group health plan and a health insurance issuer offering group or individual
health insurance coverage may not impose any preexisting condition exclusion with respect to such plan
or coverage.”); id. § 300gg(a)(
) (prohibiting insurers from charging higher premiums except on the
basis of age, rating area, tobacco use, and whether the policy covers individuals or families). The ban on
exclusions based on preexisting conditions is sometimes called “guaranteed issue” (because it
guarantees the issuance of insurance to those with preexisting conditions). The limitation on the bases
for charging higher premiums is usually called “community rating” (because everyone in a given
community is treated similarly, even though some people would be expected to have higher health care
9. See infra notes 17–18 and accompanying text.
10. See THE FEDERALIST NO. 51 (James Madison).
institutions, because a key element of bootstrapping is self-empowerment, and
that practice raises fears of government aggrandizement that have a particular
historical and constitutional pedigree.
One can and probably should distinguish among aggrandizing outcomes.
The Supreme Court, for example, has treated aggrandizing actions that invade
another branch’s authority (as opposed to aggrandizement that does not so
invade) as particularly troubling.11 This is a relevant axis, but it is orthogonal to
the issues this article addresses. The possible illegitimacy of an outcome does
not automatically condemn the process that produced it. Any process can
produce a bad outcome. The question on which this article focuses is the
legitimacy of the bootstrapping process. The unconstitutionality or illegitimacy
of action Z is an independent reason to reject Z, but not a reason to reject the
process by which all Zs are created.
Many courts and commentators have addressed the sufficiency of a
connection between conditions precedent and what they allegedly enable,
particularly in the Commerce Clause context. For example, should the Supreme
Court find a sufficient connection to interstate commerce if Congress had a
rational basis for finding a substantial connection between gun-related school
violence and interstate commerce?12 And, once a test is chosen, is it satisfied—is
there a sufficient connection between the allegedly enabling condition and what
it enables? Even a rational basis test likely would not be satisfied if Congress
tried to regulate backyard cookouts based on its past declarations of a
commemorative National Ice Cream Day. The connection between the two is
just too tenuous.13
The issue addressed in this article is different: Are there categories of
situations in which there is a sufficient empirical connection between the
enabling condition and what it enables, but the bootstrap itself is illegitimate?
So my focus is not on the test for sufficiency or its application, but instead on
whether there are forms of bootstrapping that should be rejected even though Y
in fact enables Z (and any reasonable decisionmaker would so conclude). More
specifically, my question is whether there are situations in which a particular
category of bootstrap is particularly troubling even though Y is permissible and
justifies Z, and Z does not violate an independent constitutional right. If Z is
unconstitutional or otherwise illegitimate whether or not Y exists, then Y is not
11. See Buckley v. Valeo, 424 U.S. 1, 122 (1976) (emphasizing the importance of “encroachment or
aggrandizement of one branch at the expense of the other”); N. Pipeline Const. Co. v. Marathon Pipe
Line Co., 458 U.S. 50, 83 (1982) (“The constitutional system of checks and balances is designed to
guard against ‘encroachment or aggrandizement’ by Congress at the expense of the other branches of
government.”) (quoting Buckley, 424 U.S. at 122); Bowsher v. Synar, 478 U.S. 714, 727 (1986)
(emphasizing “[t]he dangers of congressional usurpation of Executive Branch functions”).
12. See United States v. Lopez, 514 U.S. 549, 618 (1995) (Breyer, J., dissenting) (“[W]e must ask
whether Congress could have had a rational basis for finding a significant (or substantial) connection
between gun-related school violence and interstate commerce.”).
13. Cf. id. at 567 (“The possession of a gun in a local school zone is in no sense an economic
activity that might, through repetition elsewhere, substantially affect any sort of interstate
No. 3 2012]
legally enabling Z and there is no bootstrap, because Z is illegitimate anyway.
Put differently, this article addresses the circumstances under which the act of
bootstrapping is a problem and thus delegitimizes the enabled decision.
Governmental bootstrapping is not limited to the Commerce Clause
context, or even to congressional enactments. Examples abound of government
institutions relying on their decisions to legitimate other decisions they
undertake. Precedent can serve this function—allowing a court (or the
President) to do in a few steps what it could not do in one.
My inquiry is not limited to whether particular categories of bootstrapping
violate the Constitution, much less whether the Supreme Court would find
those categories unconstitutional. I want to ask a broader question: Are there
categories of bootstrapping that we as lawyers and citizens should treat as
particularly troubling forms of aggrandizement, even if they would not run
afoul of a specific constitutional prohibition?14 What is troubling will differ from
person to person, of course, but what I mean to ask is whether the
aggrandizement concerns that motivate fears about governmental
bootstrapping are sufficiently greater in some contexts that we can distinguish
them from other forms of bootstrapping. As I will discuss below, bootstrapping
is pervasive. My question is whether we can identify differences among
bootstraps that should lead lawyers and citizens to reject bootstraps of
Part II sets up the analysis by focusing on the most prominent recent
examples of bootstrapping, arising out of the ACA’s mandate that individuals
purchase health insurance. The arguments against bootstrapping are
consequentialist ones, grounded in fears about the results of bootstrapping. Part
III discusses some key features of bootstraps, including their ubiquity, and thus
the undesirability of disallowing all of them. Part IV turns to the question of
distinguishing problematic bootstraps from other bootstraps, based either on
substantive distinctions or on the purposes of the bootstrapper. I find that there
are no useful substantive distinctions, and that purpose also does not give rise to
useful distinctions, with one possible exception. That possible exception is a
distinction based on the certainty of the bootstrap. A focus on certainty gives
rise to a possible distinction between simultaneous and nonsimultaneous
bootstraps, as Y’s enablement of Z will be clear when they are simultaneous,
but the promulgation of Z will not be certain when they are separated in time. I
find that this distinction would impose no meaningful hurdles to the President’s
ability to bootstrap but would impose very significant costs on Congress, and
that those costs outweigh its benefits. The cure is worse than the disease.
14. So violating a constitutional norm (like the limits imposed by the Interstate Commerce and
Necessary and Proper Clauses) is sufficient but not necessary.
BOOTSTRAPPING AND THE INDIVIDUAL MANDATE
My launching point is the argument that the individual mandate15 in the
ACA exceeds Congress’s power under the Commerce Clause. Many arguments
have been put forward against the individual mandate, some focusing on
particular cases, some focusing on the Commerce Clause versus the Necessary
and Proper Clause, etc. I am interested in one particular species of argument
against the individual mandate—that focused on bootstrapping.
One justification for the individual mandate that operates as a bootstrap is
the set of ACA provisions that forbid exclusions based on preexisting
conditions and that sharply limit the bases upon which insurers can charge
higher premiums.16 The individual mandate is designed to counteract the
incentives created by these provisions.17 This is particularly clear with respect to
preexisting conditions: Without a coverage mandate to accompany the ban on
exclusions for preexisting conditions, “many people would simply wait to
purchase insurance until they needed care. This ‘adverse selection’ would force
the price of insurance higher for sick people who want to maintain continuous
coverage.”18 As Randy Barnett has stated, “Congress exercises its commerce
power to impose mandates on insurance companies and then claims these
15. 26 U.S.C.A. § 5000A(a) (West 2010) (“An applicable individual shall for each month
beginning after 2013 ensure that the individual, and any dependent of the individual who is an
applicable individual, is covered under minimum essential coverage for such month.”).
16. See supra note 8 and accompanying text.
17. See 42 U.S.C.A. § 18091(a)(
)(I) (West 2010) (“[I]f there were no [individual mandate], many
individuals would wait to purchase health insurance until they needed care. By significantly increasing
health insurance coverage, the requirement, together with the other provisions of this Act, will
minimize this adverse selection . . . . The [individual mandate] is essential to creating effective health
insurance markets in which improved health insurance products that are guaranteed issue and do not
exclude coverage of pre-existing conditions can be sold.”).
18. Mark A. Hall, Commerce Clause Challenges to Health Care Reform, 159 U. PA. L. REV. 1825,
1841 (2011); see also Anthony T. Lo Sasso, Community Rating and Guaranteed Issue in the Individual
Health Insurance Market, NIHCM Foundation, Jan. 2011, available at
http://nihcm.org/pdf/EVLoSassoFINAL.pdf (“The primary concern with community rating and guaranteed issue is that these
regulations can lead to adverse selection, destabilizing the insurance market and potentially causing
total market collapse. This scenario is commonly termed an adverse selection death spiral. Aware of
these concerns, policymakers included a provision in the ACA mandating all individuals to have health
insurance.”); Making Health Care Work for American Families: Hearing Before the Subcomm. on
Health of the H. Comm. on Energy & Commerce, 111th Cong. 11 (2009) (statement of Uwe Reinhardt,
Professor of Economics and Pub. Affairs, Princeton Univ.) (“[C]ommunity-rating and guaranteed issue,
coupled with voluntary insurance, tends to lead to a death spiral of individual insurance.”); Brief for
Petitioners, Dep’t of Health & Human Servs., No. 11-398 (U.S. Jan. 6, 2012), 2012 WL 37168, at *18
(“[G]uaranteed-issue and community-rating enacted in isolation create a spiral of higher costs and
reduced coverage because individuals can wait to enroll until they are sick.”); Adele M. Kirk, Riding
the Bull: Experience with Individual Market Reform in Washington, Kentucky, and Massachusetts, 25 J.
HEALTH POL. POL’Y & L. 133, 152 (2000) (stating that all but one private insurer left Kentucky after
the state limited exclusions based on preexisting conditions); Mark A. Hall, An Evaluation of New
York’s Reform Law, 25 J. HEALTH POL. POL’Y & L. 71, 91–92 (2000) (after New York enacted
legislation containing guaranteed issue and community rating for health insurance without an
individual mandate, “[t]here was a dramatic exodus of indemnity insurers from New York’s individual
No. 3 2012]
insurance mandates will not have their desired effects unless it can impose
mandates on the people, which would be unconstitutional if imposed on their
A second bootstrap involves EMTALA.20 That statute prohibits hospitals
from refusing to stabilize emergency patients on the basis of their ability to pay
or insurance status, thereby effectively turning every person into a potential
cost that is shifted from the individual to hospitals.21 The individual mandate
forces individuals to internalize that cost. As Randy Barnett further explains,
it is claimed that because everyone will one day need health care and may not be able
to afford it when that day arrives, and because emergency rooms are obligated by law
to provide care regardless of ability to pay, it is necessary to require that all persons
purchase health insurance today to avoid shifting costs to others.22
So, once EMTALA was passed, an individual mandate could fairly be
characterized as simply shifting economic impacts from providers to consumers.
Without EMTALA, the ACA’s individual mandate would be harder to justify
as a cost-shifting mechanism because hospitals would not be legally required to
provide uncompensated emergency care, and thus there would be no legal
burden of costs to shift. Again, opponents have decried this bootstrap. As the
Revere America Foundation wrote in an amicus brief about EMTALA and the
resulting free riders as a justification for the individual mandate, “that problem
is of Congress’ own creation, and Congress cannot bootstrap itself into powers
not enumerated by the Constitution simply because it deems the exercise of
those powers expedient in light of other regulations that it has previously
19. Randy E. Barnett, Turning Citizens into Subjects: Why the Health Insurance Mandate Is
Unconstitutional, 62 MERCER L. REV. 608, 614 (2011); see also Opening/Response Brief of
Appellee/Cross-Appellant States, Florida ex rel. Att’y Gen. v. U.S. Dep’t of Health & Human Servs.,
648 F.3d 1235 (11th Cir. 2011) (Nos. 11-11021, 11-11067), 2011 WL 1944107, at *39–40 (“The
government also contends that the individual mandate is incidental to ‘the requirement that insurers
extend coverage and set premiums without regard to pre-existing medical conditions.’ The government
insists that this requirement ‘would not work without’ the individual mandate because the requirement
will encourage consumers to refrain from buying insurance until they are injured or sick. . . . The
Constitution does not permit this type of blatant bootstrapping—create a problem and then assert that
it is necessary and proper to fix the problem by asserting an authority the Constitution otherwise denies
the federal government.”) (internal citation omitted).
20. 42 U.S.C. § 1395dd (2006).
21. 42 U.S.C.A. § 1395dd(h) (West 2010) (“A participating hospital may not delay provision of an
appropriate medical screening examination . . . or further medical examination and treatment . . . in
order to inquire about the individual’s method of payment or insurance status.”). Before the passage of
EMTALA, some hospitals provided emergency care regardless of ability to pay and others did not. See
Tiana Mayere Lee, An EMTALA Primer: The Impact of Changes in the Emergency Medicine
Landscape on EMTALA Compliance and Enforcement, 13 ANNALS HEALTH L. 145, 146–47 (2004).
EMTALA created a uniform requirement of basic coverage.
22. Barnett, supra note 19, at 614; see also Florida ex rel. Att’y Gen. v. U.S. Dep’t of Health &
Human Servs., 648 F.3d 1235, 1295 n.101 (11th Cir. 2011), cert. granted sub nom. Nat’l Fed’n of Indep.
Bus. v. Sebelius, 132 U.S. 603 (Nov. 14, 2011) (No. 11-393); Dep’t of Health & Human Servs. v. Florida,
132 U.S. 604 (Nov. 14, 2011) (No. 11-398); Florida v. Dep’t of Health & Human Servs., 132 U.S. 604
(Nov. 14, 2011) (No. 11-400) (noting the government’s reliance on EMTALA and stating “the plaintiffs
point out that the government’s contention amounts to a bootstrapping argument. Under the
government’s theory, Congress can enlarge its own powers under the Commerce Clause by legislating a
market externality into existence, and then claiming an extra-constitutional fix is required”).
Those who attack this form of bootstrapping contend that the relevant
bootstrap is illegitimate because it goes too far. Critics fear that allowing this
bootstrapping would give Congress the power to compel economic activity,
which it can then regulate under the Commerce Clause.24
It is worth unpacking these bootstrapping arguments a bit. The argument
against bootstrapping is not that the individual mandate violates some
independent constitutional right (e.g., the substantive component of the Due
Process Clause).25 The opponents of bootstrapping are not arguing that the
enabling provision (e.g., EMTALA) is illegitimate. They are not denying that
there is a sufficient empirical connection between the enabler (e.g., EMTALA)
and what it enables (a requirement that people who can impose costs on
emergency rooms bear those costs themselves through insurance).26 Instead,
they are focusing on the illegitimacy of relying on Y to justify Z because of the
implications of accepting such reasoning. The problem seems to be one of
consequences: Allowing Y to enable Z allows for aggrandizement, possibly
unending aggrandizement.27 If we allow this bootstrapping, what is the limit?
23. Brief of Revere America Foundation as Amicus Curiae in Support of
PlaintiffsAppellees/Cross Appellants, Florida ex rel. Att’y Gen., 648 F.3d 1235 (Nos. 11-11021, 11-11067), 2011
WL 2530512 at *10; see also Brief for the Am. Legislative Exch. Council as Amicus Curiae in Support
of Plaintiffs-Appellees, Dep’t of Health & Human Servs. v. Florida, No. 11-398 (U.S. filed Feb. 13,
2012), 2012 WL 504620 at *14–15 (“[T]he cost-shifting in this case is largely of Congress's own creation,
stemming from the requirement in the Emergency Medical Treatment and Labor Act, 42 U.S.C. §
1395dd, that hospitals provide emergency medical care regardless of ability to pay. Congress cannot
bootstrap a radical expansion of its Commerce power simply by legislating cost-shifting measures.”);
Brief of Amici Curiae Docs4patientcare, the Benjamin Rush Society, and the Pacific Research
Institute, Dep’t of Health & Human Servs. v. Florida, 648 F.3d 1285 (11th Cir. 2011) (No. 11-11021),
2011 WL 2530507, at *22 (characterizing the argument that EMTALA justifies the individual mandate
as “disingenuous bootstrapping” and adding: “The federal government cannot expand its own authority
by manufacturing the circumstances of uncompensated care and claiming that such circumstances now
affect interstate commerce.”); Opening/Response Brief for Appellee, Virginia ex rel. Cuccinelli v.
Sebelius, 656 F.3d 253 (4th Cir. 2011) (Nos. 11-1057, 11-1058), 2011 WL 1115016 at *43–44 (“While it is
true that Congress has directly regulated aspects of the health care system, principally by mandating
emergency room treatment by hospitals receiving federal funds . . . the question in this case is whether
Congress can command a citizen to purchase insurance solely for the convenience of the government in
regulating market distortions caused, at least in part, by previous congressional regulation.”).
24. See Brief of Revere America Foundation, supra note 23, at *24 (“Under the Government’s
theory, Congress can impress unwilling individuals into commerce and compel them to buy unwanted
products whenever doing so is deemed by Congress to be essential to some larger regulatory plan.”);
see also Florida ex rel. Att’y Gen., 648 F.3d at 1351 n.14 (Marcus, J., concurring in part and dissenting in
part) (“For reasons that remain inexplicable to me, the majority opinion seems to suggest that the
individual mandate is a ‘bridge too far’—in the words of the district court—not because it conscripts the
inactive, but rather for some inchoate reason stated at the highest order of abstraction.”).
U.S. CONST. amend. V; id. amend. XIV, § 1.
26. Some bootstrapping opponents make some or all of these arguments in addition to their
complaints about bootstrapping, but I am focusing on the bootstrapping arguments specifically.
27. See Reply Brief for Appellants, Thomas More Law Ctr. v. Obama, 651 F.3d 529 (6th Cir. 2011)
(No. 10-2388), 2011 WL 1653756 at *16 (“[I]f Congress can force a private citizen to engage in
commercial or economic activity under penalty of federal law—whether exercising its authority under
the Commerce Clause or the Necessary and Proper Clause—then it can regulate virtually anything, and
the federal government is no longer one of limited and enumerated powers.”); Robert A. Levy, The
This is a consequentialist argument, one that might resonate for many
judges. The argument here is not that bootstrapping is illogical or immoral or
unethical, but rather that it produces, or can produce, undesirable outcomes.
The particular undesirable outcome, in the context of the individual mandate, is
limitless congressional power: If we allow Congress to use EMTALA and the
provisions on preexisting conditions and community rating to enable it to
require an individual mandate, Congress can mandate anything the
Constitution does not specifically prohibit.28 More generally, a governmental
Case Against President Obama’s Health Care Reform: A Primer for Nonlawyers, CATO INST., 6–7 (Apr.
25, 2011), http://www.cato.org/pubs/wtpapers/ObamaHealthCareReform-Levy.pdf (“Essentially, the
insurance mandate is regulatory bootstrapping of the worst sort. Congress forces someone to engage in
commerce, then proclaims that the activity may be regulated under the Commerce Clause. If Congress
can do that, it can prescribe all manner of human conduct.”); The Constitutionality of the Affordable
Care Act: Hearing Before the S. Comm. on the Judiciary, 112th Cong. 7 (2011) (statement of Michael
Carvin, Partner, Jones Day) (“[A]llowing Congress to impose an ‘individual mandate’ on Americans to
offset the costs created by Congressional regulation would mean that the Necessary and Proper Clause
eviscerates all limits on Congress’s enumerated powers.”); Memorandum of Amici Curiae, Former
United States Attorneys General William Barr, Edwin Meese, and Dick Thornburgh, in Support of
Plaintiff’s Motion for Summary Judgment at 18, Virginia ex rel. Cuccinelli v. Sebelius, 728 F. Supp. 2d
768 (E.D. Va. 2010) (No. 3:10cv001880), 2010 WL 4168827 (“If one does not apply the Necessary and
Proper Clause according to its meaning, the government could claim virtually any power simply by
citing one of the myriad federal laws that—by being carried out—affects the economy, and then
claiming that it wishes to force individuals to participate in a certain fashion to adjust for the
consequences of those laws.”); Opening/Response Brief for Appellee, Virginia ex rel. Cuccinelli, 656
F.3d 253 (4th Cir. 2011) (Nos. 11-1057, 11-1058), 2011 WL 1115016 at *44–45 (“[T]he claim that citizens
can be commanded to purchase goods or services from another citizen in order to increase the
efficiency of the federal government’s regulation of commercial actors goes beyond the negative outer
limits of the Commerce Clause, even as aided by the Necessary and Proper Clause, because the claimed
power would be unlimited and indistinguishable from a national police power.”); Brief for the Am.
Legislative Exch. Council, supra note 23, at *14 (“The Government’s theory in this case reduces to the
idea that an individual’s failure to purchase health insurance has a negative effect on the interstate
market for health insurance, as compared to when individuals uniformly purchase such insurance under
Government compulsion. But this bootstrapping rationale is equally true—that is, trivially so—with
respect to nearly every individual decision not to participate in commerce: Congress can always
conceive of some activity which, if mandated, would substantially affect interstate commerce. . . . Such a
conception of the Commerce Clause is indistinguishable from a plenary federal police power.”); Reply
Brief for Plaintiffs in Support of Motion for a Preliminary Injunction, Thomas More Law Ctr. v.
Obama, 720 F. Supp. 2d 882 (E.D. Mich. 2010) (No. 2:10-cv-11156), 2010 WL 4784262 (arguing that the
doctrine of enumerated powers would be eviscerated if Congress could expand its powers by issuing a
finding of “essentiality” in a regulatory scheme); Brief of Members of the U.S. Senate and Speaker of
the House of Representatives John Boehner as Amici Curiae in Support of Plaintiffs-Appellees,
Florida ex rel. Att’y Gen., 648 F.3d 1235 (No. 11-11021), 2011 WL 2530520 at *31 (“The [Necessary and
Proper] Clause is not a blanket grant of Congressional power to be invoked by Congress whenever its
constitutionally-permissible provisions have bad real-world results that can only be mitigated by
otherwise unconstitutional provisions.”); see also Florida ex rel. Att’y Gen., 648 F.3d at 1351 (Marcus,
J., concurring in part and dissenting in part) (“Perhaps at the heart of the plaintiffs’ objection to the
mandate—adopted by the majority opinion in conclusion, if not in reasoning—is the notion that
allowing the individual mandate to stand will convert Congress’ commerce power into a plenary police
power, admitting of no limits and knowing of no bounds. The parade of horribles said to follow
ineluctably from upholding the individual mandate includes the federal government’s ability to compel
us to purchase and consume broccoli, buy General Motors vehicles, and exercise three times a week.”).
28. See supra note 27; Florida ex rel. Bondi v. U.S. Dep’t of Health & Human Servs., 780 F. Supp.
2d 1256, 1298 (N.D. Fla. 2011), aff’d in part, rev’d in part sub nom. Florida ex rel. Att’y Gen. v. U.S.
Dep’t of Health & Human Servs., 648 F.3d 1235 (11th Cir. 2011), cert. granted sub nom. Nat’l Fed’n of
institution, finding itself with less power than it would like, and in particular the
inability to implement Z, could make decision X and thereby empower itself to
make decision Y, which would further empower it to undertake Z. The
governmental institution could thus engage in repeated stepwise increases in
power. Randy Barnett succinctly stated the possible result of allowing the
bootstraps in the ACA: “By this reasoning, Congress would now have the
general police power the Supreme Court has always denied it possessed.”29
BOOTSTRAPS AS A CATEGORY
Given the fears of aggrandizement, one obvious response would be to
disallow any actions that have this aggrandizing form. If we are worried about
increases in power, why not prohibit all of them?
At the outset, note a basic textual problem for a prohibition on all
congressional bootstrapping. The language of the Necessary and Proper Clause
necessarily means that at least some bootstrapping is constitutionally
authorized. The text of the clause gives Congress the authority to “make all
Laws which shall be necessary and proper for carrying into Execution” the
specifically enumerated powers.30 To have meaning, this language must, like any
Indep. Bus. v. Sebelius, 132 S. Ct. 603 (Nov. 14, 2011) (No. 11-393); Dep’t of Health & Human Servs. v.
Florida, 132 S. C.t 604 (Nov. 14, 2011) (No. 11-398); Florida v. Dep’t of Health & Human Servs., 132 S.
Ct. 604 (Nov. 14, 2011) (No. 11-400) (“If Congress is allowed to define the scope of its power merely by
arguing that a provision is ‘necessary’ to avoid the negative consequences that will potentially flow from
its own statutory enactments, the Necessary and Proper Clause runs the risk of ceasing to be the
‘perfectly harmless’ part of the Constitution that Hamilton assured us it was, and moves that much
closer to becoming the ‘hideous monster [with] devouring jaws’ that he assured us it was not.”).
Indeed, some opponents have suggested that allowing the bootstrap will produce particularly
pernicious forms of aggrandizement. House Republican Leader John Boehner argued in an amicus
brief that this bootstrapping could create the wrong incentives for Congress. Brief of House Republican
Leader John Boehner as Amicus Curiae in Support of Plaintiffs’ Motion for Summary Judgment at 6,
Florida ex rel. Bondi, 780 F. Supp. 2d 1256 (No. 3:10-cv-91) (“If adopted by the court, this
interpretation of the Necessary and Proper Clause would create incentives for Congress to pass
illconceived or unrealistic statutes.”).
In his opinion, Judge Vinson emphasized this possibility, writing,
Such an application of the Necessary and Proper Clause would have the perverse effect of
enabling Congress to pass ill-conceived, or economically disruptive statutes, secure in the
knowledge that the more dysfunctional the results of the statute are, the more essential or
“necessary” the statutory fix would be. Under such a rationale, the more harm the statute
does, the more power Congress could assume for itself under the Necessary and Proper
Florida ex rel. Bondi, 780 F. Supp. 2d at 1297; see also Levy, supra note 27, at 10 (“Once again, Judge
Vinson saw through the sophistry: The mandate is artificially necessary—required only because
Congress went down a particular path that left few if any alternatives.”).
29. Barnett, supra note 19, at 614. Perhaps some might suggest that the claimed bootstraps are
overstated, reasoning that EMTALA and the ban on preexisting condition exclusions do not change
the legal landscape very much, so the individual mandate does not depend on them. If that is true, then
the concerns created by bootstrapping are overstated as well. That is, the less work that is done by
bootstrapping, and the less common they are, the less we need to be worried about their aggrandizing
30. U.S. CONST. art. I, § 8, cl. 18.
grant of power, authorize Congress to enact legislation that Congress would not
have the authority to enact in the absence of the language. Whatever the limits
of its meaning, the Necessary and Proper Clause at its textual core means that
laws passed within Congress’s enumerated authority can authorize other laws
(those “necessary and proper for carrying into execution” those enumerated
powers) that would otherwise be outside of Congress’s authority. In other
words, legislation enacted pursuant to enumerated powers is the Y that enables
the Z that the Necessary and Proper Clause covers at a minimum.31
The Necessary and Proper Clause empowers only Congress, and there is no
similar empowerment of the President or the Supreme Court. There are broad
grants of power.32 But there is no provision with respect to the President or the
Supreme Court that clearly authorizes bootstrapping. So the only entity that is
textually guaranteed some ability to bootstrap is Congress.
The textual argument, however, is sufficient but not necessary. Regardless
of the Necessary and Proper Clause, any suggestion of the illegitimacy of
bootstrapping cuts much too broadly. No government entity could meaningfully
function without bootstrapping. Congress provides the most obvious
illustration. Every enactment depends in part on conditions precedent that
previous Congresses helped to create. The multitude of federal regulations of
interstate banks relies on the impact of earlier federal actions providing for
interstate banks.33 Regulation of activities on federal properties can exist
because the federal government chose to have those properties in the first
place. For that matter, Congress’s ability to regulate any company depends on a
series of earlier legislative decisions allowing the company to enter its line of
business, not to mention earlier enactments that protected the United States
from its enemies and allowed companies to exist. Simply stated, every piece of
legislation is predicated in part on a state of affairs that previous Congresses
helped to create.
Another point bears emphasizing. Bootstrapping involves the use of Y to
legitimate Z—doing Z directly would be problematic because of some hurdle,
and Y enables Z by overcoming that hurdle. This highlights the fact that
bootstrapping depends on the existence of a hurdle in the first place. An actor
with plenary power (say, an absolute monarch) has no need to bootstrap,
because it can always undertake Z, whether or not it has undertaken Y first. Or,
to use an example closer to home, in a world in which everyone agreed that the
interstate commerce power authorized any regulation Congress deemed
appropriate, there would be no role for bootstrapping in the Commerce Clause
context. Congress could simply enact any Z that it wanted, regardless of the
presence or absence of any conditions precedent. By the same token, the hurdle
must be superable. If there is no Y that could enable a particular Z—say, the
goal of the President attaining absolute power—then bootstrapping is not an
issue. An institution’s ability to use Y to legitimate Z is a function of the
relevant legal test.
The centrality of the relevant test relates to another point: For many tests,
inaction is as important as action. This puts enormous weight on the decisions
of an actor’s predecessors—either to act or not to act. For instance, in Reno v.
ACLU,34 the Supreme Court ruled that the modest First Amendment
protections for broadcast television were inapplicable to the Internet.35 A
central distinction for the Court was the long history of broadcast regulation, as
contrasted with the absence of such a history with respect to the Internet.36
Regulation of the broadcast spectrum helped to enable later broadcast
regulation, but the failure to regulate the Internet helped to doom later Internet
regulation. Public forum doctrine is another example. If the government places
limits on the use of a forum when it creates the forum, courts will treat those
limits as demarcating its openness.37 “Once it has opened a limited forum,
however, the State must respect the lawful boundaries it has itself set.”38
Deciding exactly what if any limits to impose at the outset thus has major
implications for later regulation. Beyond that, the Supreme Court sometimes
relies on the existence, or nonexistence, of early congressional enactments in
determining the constitutionality of a recent enactment. Printz v. United States,39
for example, in invalidating a 1993 statute’s imposition of responsibilities on
states without their consent, placed great emphasis on the absence of early
congressional enactments imposing similar responsibilities.40 The failure of
earlier Congresses to enact such statutes undermined its successors’ ability to do
Bootstrapping’s emphasis on conditions precedent—the existence
34. 521 U.S. 844 (1997).
35. Id. at 867.
36. Id. (distinguishing Fed. Commc’n Comm’n v. Pacifica Found., 438 U.S. 726 (1978), because
“the order in Pacifica [was] issued by an agency that had been regulating radio stations for decades”);
id. at 868 (noting that the Court “relied on the history of extensive Government regulation of the
broadcast medium” in its broadcasting jurisprudence); id. at 868–69 (“Neither before nor after the
enactment of the CDA have the vast democratic forums of the Internet been subject to the type of
government supervision and regulation that has attended the broadcast industry.”).
37. See Rosenberger v. Rector & Visitors of Univ. of Va., 515 U.S. 819, 829 (1995) (“The
necessities of confining a forum to the limited and legitimate purposes for which it was created may
justify the State in reserving it for certain groups or for the discussion of certain topics.”).
39. 521 U.S. 898 (1997).
40. Id. at 907–08.
41. The emphasis on history is particularly notable in the jurisprudence of Justice Scalia, who has
emphasized the importance of “universal and long-established American legislative practice” as a
relevant consideration in interpreting the First Amendment. McIntyre v. Ohio Elections Comm’n, 514
U.S. 334, 375–77 (Scalia, J., dissenting).
nonexistence of a given set of prior actions—means not only that a previous
undertaking of those actions will enable further action, but also that the failure
to undertake those earlier actions will make it harder to take further action. In
this way, and importantly, allowing (or disallowing) bootstrapping will often
have no clear valence. And bootstrapping can occur with individual rights as
well as congressional power, as the path to Brown v. Board of Education
reveals.42 Sometimes bootstrapping aggrandizes congressional power and
sometimes it doesn’t.
Bootstrapping involving Congress’s commerce power is about congressional
power and does have a clear valence, given the tests that the Supreme Court
uses to determine the limits of that power. Congress’s failure to enact
EMTALA in year A in no way made it harder for Congress to pass EMTALA
in year A+B, and therefore had no impact on the ability of Congress to enact an
individual mandate in year A+B+C. But many tests will place weight not only
on action but also on inaction, as the example of broadcasting and the Internet
DISTINGUISHING AMONG BOOTSTRAPS
Can’t one separate problematic from unproblematic bootstrapping? The
fact that bootstrapping is pervasive doesn’t mean there are no clear distinctions
among forms of bootstrapping that allow us to identify the troubling versions.
A. Substantive Distinctions
Perhaps some forms of Y are substantively different from others in some
dispositive way. But if Y is illegitimate, it should be rejected: It cannot
legitimately enable anything. If, say, the Supreme Court ruled first that it had
plenary jurisdiction to decide the outcome of any election and then chose the
winner of a bitterly contested election, the allegedly enabling decision would be
illegitimate and should be rejected in its own right. There would be no need to
consider the legitimacy or implications of the choice of a winner, because the
entire foundation was infirm.
And if Y is permissible, distinguishing among more and less appropriate Ys
is a fool’s errand. On what basis would we place congressional enactments or
Supreme Court holdings on this continuum? By how many times they have
been favorably cited? By how central they seem to have become in our lives?
By how much controversy they have created? There is no answer to these
questions.44 Wisely, those concerned about bootstraps do not ask us to consider
them. The stated concern is not with the legitimacy of Y. Concerns about
42. 347 U.S. 483 (1954); see supra notes 1–3.
43. See supra notes 34–36 and accompanying text.
44. We could try to create different categories of Ys based on how generative they have been of
Zs, but that assumes the conclusion that the Zs are permissible and are a useful measuring stick.
Congress bootstrapping EMTALA into an individual mandate do not depend
on the propriety of EMTALA. The point is that EMTALA may be totally
appropriate and legitimate, but using EMTALA to justify an individual
mandate is illegitimate. Why? Apparently, because there is no stopping point:
once one allows EMTALA to be used in this way, Congress can claim whatever
powers it wants.45
Some might object that this rejection of distinctions among categories of Ys
is too hasty. Aren’t there meaningful distinctions we can draw among them?
Each of us can of course identify enactments and opinions we prefer (e.g., ones
that we think are more public-regarding, or ones with better provenance),46 but
that is possible for all legal decisions. If we accept such grounds, then we can
apply our criteria directly to Zs and dispense with any focus on bootstraps. One
possible distinction is between bootstraps in which the entity relies on the
impact of its own action and those in which it does not.47 But, as the discussion
so far indicates, in every bootstrap an actor is relying on the impact of its first
action to enable a later action, usually directly.
A more nuanced variation on this argument is that we might try to
distinguish Ys arising on their own from those that Congress creates. The idea is
that with Ys that have arisen on their own via private conduct, there is only one
jump (between Y and Z), but with congressionally created Ys, there are two
jumps (to Y, and then between Y and Z). But this distinction depends on a false
dichotomy between Ys that arise on their own and those mandated by the
government. No Y arises entirely on its own. Every private activity relies on a
backdrop created in part by prior government actions. Some Ys might seem to
have arisen more naturally than others, but the Ys will be on a continuum in
this regard. And even statutory Ys designed to codify an existing practice police
outliers and circumscribe future conduct, and thus change the playing field.48
45. See supra notes 25–29.
46. See, e.g., McCulloch v. Maryland, 17 U.S. (4 Wheat) 316, 402 (1819) (emphasizing that “[t]he
bill for incorporating the Bank of the United States . . . [was] supported by arguments which convinced
minds as pure and as intelligent as this country can boast”); Akhil Reed Amar, Intratextualism, 112
HARV. L. REV. 747, 755 (1999) (“[Marshall] opens his opinion by reminding us that the 1791 bill
creating the first bank was supported by ‘minds as pure and as intelligent as this country can boast.’ The
reference here is to the sainted Washington, who added his name to the bill and thus made it law only
after satisfying himself of its constitutionality.”) (internal citation omitted).
47. John Eastman made a more specific argument analogous to this, focusing on the Supreme
Court and the Commerce Clause. John Eastman in the Sacramento Bee: Health Care Reform
Challenge? It’s the Law, JOHN EASTMAN FOR ATTORNEY GENERAL (Apr. 6, 2010),
http://www.eastmanforag.com/news/john-eastman-in-the-sacramento-bee-health-care-reformchallenge-its-the-law (“[T]he Supreme Court has never before allowed Congress to bootstrap its way
into Commerce Clause authority by relying on the impact of its own regulation.”).
48. Consider EMTALA in this regard. Before EMTALA, some emergency rooms voluntarily
treated patients who could not pay, and some federally funded emergency rooms had some treatment
obligations. See Maria O’Brien Hylton, The Economics and Politics of Emergency Health Care for the
Poor: The Patient Dumping Dilemma, 1992 BYU L. REV. 971, 980 (1992) (discussing 1946 legislation
known as the Hill-Burton Act that required hospitals “receiv[ing] federal funds for construction and
capital improvements to furnish a ‘reasonable’ amount of free or reduced-cost care to indigent
patients,” and the “widespread agreement that this program has been a complete failure with respect to
The degree of naturally occurring activities underpinning Y may lead us to
conclude that a given Y is undesirable or unwise. But that is not an argument
about bootstrapping; it is an argument about the underlying attractiveness of Y.
This brings us to the most prominent of the arguments against the individual
mandate—that it is regulating inaction instead of action. I will assume, for
purposes of this discussion, that the opponents are correct in asserting that the
individual mandate involves a different form of regulation than Congress has
attempted before. That may affect some aspects of the argument about the
constitutionality of the individual mandate, but it has no logical relationship to
arguments about the legitimacy of this bootstrap. Bootstraps do not work
differently when they enable regulation of action rather than inaction, any more
than they do when they enable government actions that enhance rather than
constrict civil liberties. This particular bootstrap may produce consequences we
do not like (here, by hypothesis, allowing regulation of inaction), but the
mechanism of the bootstrap is the same—one regulation has enabled another.49
Suppose the Supreme Court relied on its statement that the First
Amendment “rests on the assumption that the widest possible dissemination of
information from diverse and antagonistic sources is essential to the welfare of
the public”50 to hold that the First Amendment has a positive component that
empowers Congress to regulate speech (rather than merely prohibiting
government abridgements of speech). Some might consider this conclusion to
be wrongheaded and a misreading of the First Amendment, but the mechanism
of building upon previous opinions would be the same as it has been in
hundreds of other cases. The earlier cases would enable the “wrong”
conclusion, but the mechanism of that enablement would be the same one that
courts routinely use.
One last substantive distinction bears discussion: One might want to
distinguish the Supreme Court from other actors, on the grounds that the Court
holds Y and then Z because the issue was presented to it and it felt compelled
to reach that result. Congress and the President, on the other hand, freely
increasing the supply of indigent care”); W. Adam Malizio, Note, Moses v. Providence Hospital: The
Sixth Circuit Dumps the Federal Regulations of the Patient Anti-Dumping Statute, 27 J. CONTEMP.
HEALTH L. & POL’Y 213, 215 (2010) (recounting the failure of the Hill-Burton Act to ensure care and
adding that “[i]n an effort to address the problem of patient dumping on a broad scale, Congress passed
[EMTALA]”); Jennifer M. Smith, Screen, Stabilize, and Ship: EMTALA, U.S. Hospitals, and
Undocumented Immigrants (International Patient Dumping), 10 HOUS. J. HEALTH L. & POL’Y 309, 321
(2010) (recounting the “ineffective[ness]” of the Hill-Burton Act and stating that “Congress’ enactment
of EMTALA was the direct response to the American public’s outcry over the inhumane treatment of
the poor”). But whatever percentage of emergency rooms already provided uncompensated care,
EMTALA changed the legal landscape by compelling free care.
49. One can draw a distinction between bootstraps enabling regulation of action and bootstraps
enabling regulation of inaction, but that would be an ad hoc judgment reflecting a desire not to allow
this bootstrap (or maybe to stop the continued growth of bootstrapping). That is, the distinction would
reflect a distaste for Z, but it would not reflect anything different about the bootstrapping process.
50. Associated Press v. United States, 326 U.S. 1, 20 (1945); see also Turner Broad. Sys., Inc. v.
FCC, 512 U.S. 622, 663–64 (1994) (quoting some of this language from Associated Press); Leathers v.
Medlock, 499 U.S. 439, 459 (1991) (same).
LAW AND CONTEMPORARY PROBLEMS [Vol. 75:115
choose what topics to act on and how to act on them. In this way, the Court
arguably is not bootstrapping in the same way as Congress and the President,
because its holdings do not represent the same sort of volition.
This of course implicates one of the fundamental debates in law about the
nature of judging, one that I will not attempt to resolve here. A few points bear
noting, however. First, as to what matters are addressed, the Supreme Court has
a great deal of control over which cases it hears. The Court cannot choose to
decide an issue if no petition presents it, but it can choose among the issues
presented to it. Second, as to the decisions it reaches, few if any people believe
that Justices are completely without volition. The only question is how much
volition Justices exercise in reaching their conclusions. And the relevant
question for my purposes is whether the divergence in volition between Justices
on the one hand and members of Congress and the President on the other
distinguishes the bootstrapping of the former from that of the latter in some
relevant way. After all, once they have some ability to shape their agenda and
some ability to exercise volition in reaching their conclusions, actors have the
ability to aggrandize their power through bootstrapping. The degree of that
ability differs for Justices versus members of Congress and the President, but it
is far from clear that this difference in degree translates into a dispositive
difference in the approach one should take to their bootstrapping. For my
purposes, relatively little turns on the answer to that question. In the remainder
of this article, I consider the Supreme Court’s bootstrapping alongside that of
Congress and the President. But the key contentions in this article stand even if
one rejects the applicability of bootstrapping to the Supreme Court.
I can imagine forms of bootstrapping that virtually everyone would
condemn—for example, a President who, after a terrorist attack, issues an
executive order that declares, “In this order I make a finding of a national
emergency so that I have a justification for arrogating to myself near-dictatorial
authority to respond to terrorism.” And we might extend the category beyond
such brazen situations to any in which a political actor makes decision Y solely,
or at least centrally, to allow it to make a decision Z that would be legally
problematic absent Y. If, for instance, Congress’s sole purpose in enacting
legislative provision Y is to pave the way for a legislative provision Z that would
be impermissible but for Y, then let us assume that Congress has acted
This implicates longstanding debates about the circumstances under which it
is useful to attempt to determine a lawmaker’s subjective motivation or
apparent purpose in making a particular decision.51 One element of these
51. I am referring here to a distinction made by many commentators between the decisionmaker’s
subjective motivation and its objective purposes. See, e.g., Jed Rubenfeld, The First Amendment’s
Purpose, 53 STAN. L. REV. 767, 793–94 (2001) (distinguishing legislators’ subjective motives from
objective legislative purpose); Ira M. Heyman, The Chief Justice, Racial Segregation and the Friendly
debates involves the difficulty of an observer’s determining the lawmaker’s
actual purpose. Assuming that a lawmaker has a purpose, an observer (notably,
a court) may not be able to reliably determine that purpose. A second,
antecedent element is the usefulness of the notion of a purpose, particularly in a
multimember context. Kenneth Shepsle noted twenty years ago that Congress is
a “they,” not an “it,” and contended that “[i]ndividuals have intentions and
purpose and motives; collections of individuals do not. To pretend otherwise is
I do not take a position on these broad questions about the ontological
significance and epistemic value of lawmakers’ purpose, because it is not
necessary for my purposes.53 Whatever one’s answers to these questions,
aggrandizement is a bad candidate for a focus on purpose. There may be some
goals of such importance to both constituents and representatives that they are
central to the legislators who vote for legislation that achieves those goals. If,
for example, a majority of a given set of constituents and legislators desired the
subordination of one or more racial groups, we might feel confident in
identifying racism as the purpose of legislation that clearly harmed the
disfavored race(s). Aggrandizement is different in two ways. First, there is little
basis for concluding that aggrandizement is of central importance to
legislators.54 Congress has had ample opportunities to aggrandize its power in
Critics, 49 CAL L. REV. 104, 115–16 (1961) (same). Others reject the distinction. See, e.g., John Hart
Ely, Legislative and Administrative Motivation in Constitutional Law, 79 YALE L.J. 1205, 1217–21
(1970); Elena Kagan, Private Speech, Public Purpose: The Role of Governmental Motive in First
Amendment Doctrine, 63 U. CHI. L. REV. 413, 426 n.40 (1996); Donald H. Regan, The Supreme Court
and State Protectionism: Making Sense of the Dormant Commerce Clause, 84 MICH. L. REV. 1091, 1143
(1986). As my arguments make clear, I do not think anything turns on a distinction between subjective
motivation and objective purpose in the bootstrapping context, so those who reject this distinction can
ignore this footnote and erase it from their memories (if it is possible to do so, see The Eternal
Sunshine of the Spotless Mind (2004)).
52. Kenneth A. Shepsle, Congress Is a “They,” Not an “It”: Legislative Intent as Oxymoron, 12
INT’L REV. L. & ECON. 239, 254 (1992); see also Max Radin, Statutory Interpretation, 43 HARV. L. REV.
863, 870 (1930) (“The least reflection makes clear that the law maker . . . does not exist, and only worse
confusion follows when in his place there are substituted the members of the legislature as a body. A
legislature certainly has no intention whatever in connection with words which some two or three men
drafted, which a considerable number rejected, and in regard to which many of the approving majority
might have had, and often demonstrably did have, different ideas and beliefs.”); Frank H. Easterbrook,
Statutes’ Domains, 50 U. CHI. L. REV. 533, 547 (1983) (“Because legislatures comprise many members,
they do not have ‘intents’ or ‘designs,’ hidden yet discoverable. Each member may or may not have a
design. The body as a whole, however, has only outcomes.”).
The Supreme Court is smaller, but the same concern arises about divining its purposes. See
Ernest A. Young, The Rehnquist Court’s Two Federalisms, 83 TEX. L. REV. 1, 94 (2004). The President,
as a single person, avoids the multiperson difficulty. The problem of identifying a sole, or even central,
purpose remains for presidential decisions, but at least the President is a single decisionmaker. See infra
notes 75–79 and accompanying text.
Others disagree, of course, and contend that inquiries into purpose, including the purpose of
multimember bodies, is useful and appropriate. E.g., Regan, supra note 51, at 1143 (“[T]here is no good
general objection to motive review.”); Ely, supra note 51, at 1220–21; Kagan, supra note 51, at 439–41.
53. Pardon the pun.
54. As Joseph Blocher notes in his response to this article, “Congress A might not have sufficient
incentive to aggrandize the power of Congress B; the personal and party interests of individual
LAW AND CONTEMPORARY PROBLEMS
exactly the ways that opponents of the individual mandate fear, but its actions
have been fairly modest.55 Aggrandizement simply has not manifested itself as a
clear and central goal of members of Congress. The second and more important
difference flows from the fact that constituents will not deeply desire
aggrandizement. The fear of aggrandizement is grounded in agency costs:
Legislators will have an interest in aggrandizing their power despite their
constituents’ lack of interest in it, and will achieve aggrandizement by
smuggling aggrandizement into legislation that constituents find acceptable for
other reasons.56 The key point is that there will be other reasons for constituents
to support the legislation. Aggrandizement comes packaged in a substantive
proposal, and some aspect of that substantive proposal is what makes it
acceptable, if not desirable, for constituents. And many members of Congress
may support the legislation for the same reason their constituents do (or in
response to their constituents’ support) and have no particular interest in
aggrandizement. The aggrandizement will be part of an initiative with aspects
attractive to those uninterested in aggrandizement. And the particular balance
for the decisionmaking body, or for any member of it, may not be knowable or
Consider the following hypothetical: A powerful organization in a given
congressional district visits the district’s member of Congress and persuades her
that change Z in the law will allow it to expand its beneficial activities in her
district while doing no harm to any constituency the member of Congress cares
about. Z has no obvious connection to commerce, but it can be packaged with
another proposal Y that clearly affects commerce and might have modest
benefits, and that neither the organization nor any significant constituency
opposes. The member of Congress decides to introduce legislation that would
achieve the organization’s aims. The resulting legislation builds on the existing
state of the world; adds to an existing regulatory regime; packages Y and Z
together into a broad initiative; and has findings containing assertions that
support treating the legislation as a permissible exercise of Congress’s
Commerce Clause authority. In other words, the legislation is filled with forms
Let us imagine that the member of Congress drafted the legislation herself
and has a sufficiently deep background in constitutional law to understand the
need for the bootstraps. By hypothesis, she understands that Z would be legally
problematic absent Y, and she is aware that her inclusion of Y enables her
inclusion of Z. So we might reasonably infer that she has an apparent purpose
representatives probably outweigh their interests in expanding the power of the institution itself.”
Joseph Blocher, What We Fret About When We Fret About Bootstrapping, 75 LAW & CONTEMP.
PROBS., no. 3, 2012 at 146, 151; see generally Daryl J. Levinson & Richard H. Pildes, Separation of
Parties, Not Powers, 119 HARV. L. REV. 2311 (2006) (arguing that members of Congress are primarily
loyal to their party, not to Congress).
55. See infra notes 85–88 and accompanying text.
56. Unless we believe that constituents want aggrandizement, in which case the push against
aggrandizement is both antidemocratic and doomed to failure.
of bootstrapping by including Y in the legislation. But an apparent purpose to
bootstrap does not necessarily entail a purpose to aggrandize. Here, the
aggrandizement seems to be a byproduct of the legislator’s desire to help the
organization. And, of course, even if there were a purpose to aggrandize, that
does not mean that this was the sole or central purpose for the drafter (much
less for her colleagues who voted for it).57
One could respond that this is setting the bar too high: Even if
aggrandizement is a byproduct, the member was aware of the aggrandizement
and it is still fairly treated as an apparent purpose—and that is sufficient to
render the bootstrapping problematic. In other words, instead of focusing on a
sole or central purpose, one could say bootstrapping is illegitimate if the actor
had an awareness of aggrandizement or an apparent purpose to aggrandize. But
that cuts much too broadly. As I have noted, every piece of legislation relies on
various forms of bootstrapping, and drafters are aware of the ways in which
they rely on existing regulatory schemes and provisions in the legislation at
issue—that is part of the drafters’ job (and a reason why members of Congress
do not draft on their own, but instead rely on their staff and the Office of the
The paragraph above may seem to suggest that illegitimate bootstrapping
can be distinguished by creating a threshold that is less than the predominant
purpose but more than a purpose.58 But in fact this discussion highlights an
additional problem with attempting to consider any chosen threshold for
57. See, e.g., Edwards v. Aguillard, 482 U.S. 578, 636–40 (1987) (Scalia, J., dissenting). Justice
[A] particular legislator need not have voted for the Act either because he wanted to foster
religion or because he wanted to improve education. He may have thought the bill would
provide jobs for his district, or may have wanted to make amends with a faction of his party he
had alienated on another vote, or he may have been a close friend of the bill’s sponsor, or he
may have been repaying a favor he owed the majority leader, or he may have hoped the
Governor would appreciate his vote and make a fundraising appearance for him, or he may
have been pressured to vote for a bill he disliked by a wealthy contributor or by a flood of
constituent mail, or he may have been seeking favorable publicity, or he may have been
reluctant to hurt the feelings of a loyal staff member who worked on the bill, or he may have
been settling an old score with a legislator who opposed the bill, or he may have been mad at
his wife who opposed the bill, or he may have been intoxicated and utterly unmotivated when
the vote was called, or he may have accidentally voted “yes” instead of “no,” or, of course, he
may have had (and very likely did have) a combination of some of the above and many other
motivations. To look for the sole purpose of even a single legislator is probably to look for
something that does not exist.
Id. at 637 (Scalia, J., dissenting) (emphasis in original).
58. This is akin to a distinction drawn in some Establishment Clause cases. See, e.g., McCreary
Cnty. v. ACLU of Ky., 545 U.S. 844, 860 (2005) (“When the government acts with the ostensible and
predominant purpose of advancing religion, it violates th[e] central Establishment Clause value of
official religious neutrality.”); id. at 901 (Scalia, J., dissenting) (“[T]he Court replaces Lemon’s
requirement that the government have ‘a secular . . . purpose,’ with the heightened requirement that
the secular purpose ‘predominate’ over any purpose to advance religion.”) (quoting Lemon v.
Kurtzman, 403 U.S. 602, 612 (1971)); Edwards, 482 U.S. at 599 (Powell, J., concurring) (“A religious
purpose alone is not enough to invalidate an act of a state legislature. The religious purpose must
LAW AND CONTEMPORARY PROBLEMS
purpose in the bootstrapping context: With one possible exception discussed
below, this approach will not usefully distinguish among bootstraps. There are
myriad elements in any given piece of new legislation and in prior legislation
that enable all the elements in that new legislation. No one believes that all such
bootstraps are illegitimate (if so, every piece of legislation would be
illegitimate). The question is whether one can distinguish some bootstraps from
others, and the problem is that there is no reason to believe that the drafter(s),
sponsor(s), or supporter(s) will have different purposes for different bootstraps
that achieve the same goal. To return to the example above, insofar as one or
more members of Congress have some set of purposes in pushing to enact Z,
they will have the same purposes with respect to all the bootstraps necessary,
because the bootstraps would serve the same instrumental goal. All the
conditions precedent enable Z, and thus all have instrumental value to the
proponents of a given initiative. Some of the conditions precedent might be
more important to the drafter because they seem truly necessary for the
initiative to pass muster, as opposed to other conditions precedent that might
seem helpful but not absolutely necessary. But that is just another way of saying
that to achieve one’s purposes (whatever they may be), some building blocks
are more essential than others. And this tells us nothing about those purposes.
If Y is useful but not absolutely necessary to justify Z and some other aspect of
the legislation is absolutely necessary, then the latter may be more important to
the drafter and to the sponsors. This gives us no information about the degree
to which their purpose was aggrandizement.
C. The Significance of Simultaneity
The one possible exception to the inability to distinguish among bootstraps
based on purpose (or anything else) involves uncertainty. Some acts contain
both Y and Z, meaning that Y’s enablement of Z is certain. At the other
extreme, some acts contain what turns out to be a Y (that is, an enablement of
some Z), but not only is the promulgation of Z unclear—even the status of Y as
enabling that possible Z is unclear. In this second situation, the bootstrap would
not merely be uncertain but in fact hard to foresee. Fears about a purpose to
aggrandize would not be founded in this second situation, whereas they are at
least plausible in cases of certain bootstraps.
Uncertain bootstraps arise most notably in situations in which Y and Z are
not simultaneous. There is often a significant time lag between Y and Z, and
often the significance of Y as enabling Z will not have been apparent at the time
Y occurred. In effect, the bootstrapping is not only difficult to foresee but also
retroactive. Indeed, many of the most prominent examples of congressional
bootstrapping involve situations in which there was little incentive to bootstrap
because at the time of action Y people did not believe that it was in fact
necessary for action Z.
Reno v. ACLU provides one such example.59 When Congress enacted the
Radio Act of 192760 and then built on that legislation in the Communications
Act of 1934,61 there was no hint in the discussion in Congress or among
commentators about the possible relevance of the regulations’ breadth to any
First Amendment challenges. The most widely held assumption was that the
First Amendment had little, if any, impact on government regulation of
broadcasters’ speech. And early court cases embodied this view.62 It is unlikely
that members of Congress imagined that the breadth of the regulations they
imposed on broadcasters would have any impact on the likelihood of courts
invalidating their regulations on First Amendment grounds. The danger of
invalidation would have seemed remote, and there was no particular reason to
assume that the breadth of the other regulations would be relevant. Decades
later, however, the Supreme Court made this distinction dispositive when it
invalidated the Communications Decency Act (applicable to the Internet) in
Reno v. ACLU, despite Federal Communications Commission v. Pacifica
Foundation63 (which had permitted indecency regulation of broadcasts).
Another instance of retroactive bootstrapping arose from a different line of
Supreme Court cases. In Glickman v. Wileman Brothers & Elliott, Inc.,64 the
Supreme Court upheld a program that compelled California peach growers to
contribute to a fund used to advertise California peaches. Four years later, in
United Foods v. United States,65 the Supreme Court invalidated a very similar
program compelling mushroom growers to contribute to a fund to advertise
mushrooms. The difference, according to the Court, lay in the fact that
Glickman involved a wide-ranging regulatory program and United Foods did
not.66 The Court thus found the presence of comprehensive regulation
dispositive. Indeed, here the significance of such regulation was quite great: The
LAW AND CONTEMPORARY PROBLEMS [Vol. 75:115
Court in Glickman concluded that the challenged regulation did not have to
survive any level of First Amendment scrutiny, because the advertising program
was part of an economic regulation that was not subject to First Amendment
scrutiny in the first place;67 by contrast, the First Amendment did apply to (and
invalidate) the program challenged in United Foods.68
At the time Congress created these programs, it was not obvious (to put it
mildly) that the applicability of the First Amendment to each fund would turn
on the comprehensiveness of the program. Indeed, it was not clear how, if at all,
the First Amendment might apply to them. There is no reason to believe that
members of Congress thought that anything, as a First Amendment matter,
turned on how pervasive their scheme was. And seven of the Justices on the
Supreme Court saw no significance to this distinction at the time these cases
were decided. But the two other Justices drew a distinction based on
comprehensiveness. As a result we now know that by the Supreme Court’s
lights, a comprehensive program will enable Congress to require food producers
to contribute to a fund to advertise their product. The status of Y (the
comprehensive program) as an enabler thus only became clear years after the
programs were created.
In other situations, decisionmakers may take action Y with no particular
intention of its being treated as an enabling precedent but with some awareness
that it could be so treated, and a later action Z in fact treats it as an enabling
precedent. Arguably, some of the decisions of the First Congress were in that
position. Most of those in the First Congress made at least some of their
decisions without intending them to bind, or even greatly influence, later
Congresses, but they were likely aware of the possibility that their decisions
could be treated as dispositive precedents (and later decisionmakers did treat
them as enabling some actions and disabling others).69
1. The Impact of a Simultaneity-Based Distinction on the Different
This discussion highlights the way that bootstraps depend on factors made
relevant by other actors, often acting much later. That is the nature of tests. But
this discussion also suggests a possible distinction among bootstraps: that
between simultaneous and nonsimultaneous bootstraps. If a government
institution takes actions Y and Z at the same time, uncertainty is eliminated.
The actor ensures that Y leads to Z. This answers any question of whether the
67. Glickman, 521 U.S. at 475.
68. United Foods, 533 U.S. at 415–16.
69. This point applies even to major constitutional debates in the first Congress, such as that on
the President’s ability to remove executive officers. In Myers v. United States, 272 U.S. 52 (1926), the
Supreme Court placed great emphasis on this debate and heralded it as “the decision of 1789.” But as
Sai Prakash has noted about that debate, “[t]he vast majority of Representatives [in the first Congress]
understood that Congress could express its views about the Constitution’s meaning and that, while the
Decision of 1789 would be deserving of some deference, it would not decide the removal question for
all time.” Saikrishna Prakash, New Light on the Decision of 1789, 91 CORNELL L. REV. 1021, 1075
No. 3 2012]
bootstrap was foreseeable. It also eliminates questions about whether Z will in
fact occur, and whether it will be Z or some variant of Z, by simply creating Z at
the time it creates Y. The governmental entity acts with a presumed awareness
that it has engaged in bootstrapping, and this arguably gives rise to an
appearance of a purpose to bootstrap. So, insofar as one thinks that purpose is
relevant, maybe we should distinguish simultaneous bootstraps from
A distinction between simultaneous and nonsimultaneous bootstraps could
lead to an absolute bar to simultaneous bootstraps, a strong presumption
against, a weak presumption against, etc. But this simply raises more questions,
about not only the basis for a particular level of presumption but also what
would overcome the presumption. And to answer that question we are back to
the problems with drawing distinctions among bootstraps. One can distinguish
simultaneous versus nonsimultaneous bootstraps, but that provides no basis for
distinguishing among simultaneous bootstraps. And the difficulties discussed
above with respect to possible distinctions among all bootstraps apply as well
with respect to possible distinctions among simultaneous bootstraps.
Whatever the form of the distinction, increasing the cost of simultaneous
bootstraps will have profound consequences. The implications for Congress
would be greatest. At the outset, note that Congress would be disabled from
responding quickly to an emergency with a single piece of legislation that
contained a bootstrap (for example, legislation after September 11 creating a
federal cadre of airport screeners and regulations governing their behavior). Do
we really want to tell Congress that even if there seems to be a crying need for a
particular Z, it must respond in two steps? We could create an exception for
particularly important or urgent bootstraps. But how would we define those
with any clarity? Would a horribly lethal pandemic qualify as such an
emergency?70 The collapse of Lehman Brothers? More importantly, once we are
distinguishing among bootstraps based on their perceived importance or
urgency, we are simply distinguishing among Zs. That is, we are back to
distinguishing among bootstraps based on their outcome, not based on any
element of the bootstrapping process. To put the point differently, there would
be no difference in the way that Y enabled Z, or in the nature of the
relationship between Y and Z; what would distinguish the permissible from the
impermissible bootstrap would be the importance or urgency of Z. That is not
an argument about bootstrapping. It is an argument about outcomes.
A further problem for Congress is less obvious, but no less real. Any
presumption or bar against simultaneous bootstraps will pose particular barriers
for Congress. This flows from the fact that Congress has less ability to engage in
a long-term process of aggrandizement than the other branches do. As many
commentators have noted, the hurdles to passage of a given piece of legislation
70. See Mark A. Hall, Commerce Clause Challenges to Health Care Reform, 159 U. PA. L. REV.
1825, 1829 (2011) (noting relatedly that “authority under the commerce power to compel purchases or
other actions could well be essential to combat a horrifically lethal pandemic”).
LAW AND CONTEMPORARY PROBLEMS
are numerous and subject to change—support by the relevant committees and
their chairs, support not only by a majority of the House but also by a majority
of the majority party in the House, in many cases support by a filibuster-proof
majority in the Senate, and support by the President.71 Legislative coalitions
change during each session of Congress, in response to the electoral calendar as
well as public opinion, and of course the makeup of the House and Senate
changes every two years. A second piece of legislation means a changed
political environment, a new legislative coalition to bring together, and a new
round of veto gates to overcome. No member of Congress can have any real
confidence that the stars will align for a given future piece of legislation.
Indeed, a ban or strong presumption against simultaneous bootstrapping
could completely disable Congress from crafting legislation on some issues,
because often the only way legislation can pass is if multiple elements are
combined into one act. Starting with constitutional constraints, the best
example arises from strict scrutiny. For any legislation that would be subject to
strict scrutiny, a key element of that scrutiny is whether Congress has regulated
comprehensively. A court will invalidate the legislation if the claimed
compelling interest implicates activities that the legislation does not cover (that
is, if Congress regulates only some of the activities that implicate that interest).72
So breaking up legislation into separate parts could prevent a legislative
response to an important issue by preventing constitutionally required
Furthermore, both theory and practice indicate that successful legislation
(especially on complex issues) often requires multiple related elements in order
to satisfy the different groups needed to create a sufficiently large legislative
coalition. Prominent among recent examples is the ACA.73 The individual
mandate relied on the ban on denying coverage to those with preexisting
conditions as an empirical matter, but the provisions on preexisting conditions
relied on the individual mandate as a political matter. Insurance companies and
other major stakeholders believed that without an individual mandate, the
provisions on preexisting conditions would lead the sickest people to pay for
insurance and many of the healthy remainder to take their chances (until they
No. 3 2012]
were on the way to the emergency room, when they would sign up).74 The ban
on denying coverage based on preexisting conditions had been popular for
many years, but Congress could never enact it because of this opposition.
Insurance companies dropped their opposition, allowing the provisions on
preexisting conditions to be enacted, only when these provisions were paired
with the individual mandate. In other words, if Congress had been required to
enact the provisions on preexisting conditions and the individual mandate in
separate pieces of legislation, then likely neither would have been enacted. And
because the provisions on preexisting conditions were key (and among the most
popular) elements of the health care reform proposals, without the linkage
between the provisions on preexisting conditions and the individual mandate it
seems unlikely that any major health care reform would have been enacted.
Congress could have overcome this obstacle by establishing an unbreakable
linkage between its (now separate) legislation on preexisting conditions and on
the individual mandate. But an unbreakable linkage raises the same problems
as are created by simultaneity in the first place—the members of Congress
would know the result of their bootstrapping. By hypothesis, bootstraps with
that level of certainty are different from other bootstraps and should be
disfavored. So the way to avoid these seemingly problematic bootstraps is to
eliminate that certainty. And once there is no certainty, complex legislation like
health care reform becomes more difficult, if not impossible.
With respect to the President, disfavoring simultaneous bootstraps will have
very little impact. The President needs no one else’s vote in order to
promulgate an executive order.75 A President with any meaningful time left in
his term knows that he can issue two or three executive orders almost as easily
as he can issue one, and more generally can make multiple decisions that build
on each other.76 Indeed, this was precisely the fear that some commentators
articulated during the George W. Bush Administration—that the Bush White
House was pursuing a strategy of aggrandizing its power by playing the long
game. Recall, for instance, the Bush Administration’s unwillingness to release
the names of people who met with Vice President Cheney about energy policy.77
As was widely noted at the time, it seems unlikely that the Administration was
worried about releasing that information per se, but rather that it was using the
matter to set a precedent that it could later expand.78 A more recent example
74. See supra notes 17–18 and accompanying text.
75. See Terry M. Moe & William G. Howell, The Presidential Power of Unilateral Action, 15 J. L.
& ECON. 132, 133 (1999) (discussing “the president’s formal capacity for taking unilateral action and
thus for making law on his own”).
76. I say “almost as easily” because Presidents have always had an internal executive branch
review process for executive orders, and so two orders entail two iterations of that process. But
Presidents choose such review and thus can dispense with it, the review is not very onerous, and the
President knows at the end of the day that the decisions on all aspects of a given executive order are his
77. See Joseph Kahn, Cheney Refuses to Release Energy Task Force Records, N.Y. TIMES, Aug. 4,
2001, at A10.
78. See Michael Abramowitz & Steven Mufson, Papers Detail Industry’s Role in Cheney’s Energy
LAW AND CONTEMPORARY PROBLEMS [Vol. 75:115
arises from the United States’ role in Libya. President Obama concluded in
June 2011 that our support for bombing in Libya did not constitute “hostilities,”
based in significant part on the absence of United States ground forces, active
exchanges of fire, or a serious threat of United States casualties.79 A President
who wanted to gut the War Powers Act could build on that in future decisions
to go much further, leading to the conclusion that he could use air power in any
way he saw fit, even to the extent of making a long-term direct commitment of
the United States Air Force, because the United States has total air superiority.
This result would have seemed absurd pre-Libya. (Could the United States
claim that a re-run of the World War II air attack on Dresden did not constitute
“hostilities”?) But a President who sought to expand his power could begin with
Libya and build on that precedent, ending up with a result that would have
seemed absurd at the outset.
For courts, in particular the Supreme Court, the costs of disfavoring
simultaneous bootstraps will be significant, though not nearly as high as for
Congress. In some circumstances the Supreme Court Justices can reasonably
assume that there will be a durable majority for a result and that there will be
ready future opportunities to reach that result.80 But that will often not be the
Report, WASH. POST, July 18, 2007, at A01.
79. See White House, United States Activities in Libya, at 25 (June 15, 2011), available at
_--_6_15_11.pdf (“U.S. operations do not involve sustained fighting or active exchanges of fire with
hostile forces, nor do they involve the presence of U.S. ground troops, U.S. casualties or a serious
threat thereof.”); Charlie Savage & Mark Landler, White House Defends Continuing U.S. Role in Libya
Operation, N.Y. TIMES, June 15, 2011, at A16 (“Jack L. Goldsmith, who led the Justice Department’s
Office of Legal Counsel during the Bush administration, said the Obama theory would set a precedent
expanding future presidents’ unauthorized war-making powers, especially given the rise of
remotecontrolled combat technology. ‘The administration’s theory implies that the president can wage war
with drones and all manner of offshore missiles without having to bother with the War Powers
Resolution’s time limits,’ Mr. Goldsmith said.”).
80. Consider Nw. Austin Mun. Util. Dist. No. One v. Holder (NAMUDNO), 557 U.S. 193 (2009). I
think it is plausible to imagine that at least one of the five Justices most likely to invalidate the Voting
Rights Act thought there would be unacceptable costs to the Court if it invalidated it in NAMUDNO,
but wanted ultimately to invalidate it. In such circumstances, what might such a Justice do? One
possible answer is to bootstrap—to write the NAMUDNO opinion in a way that enables the future
invalidation of the Voting Rights Act without the high costs to the Court. If our hypothetical Justice
believed both that the majority of which he was a part was durable and that a future Voting Rights Act
challenge would be brought, then the risks of this strategy would be fairly low and the benefits
Whether or not this was in fact any Justice’s intention, I believe that the Court’s opinion in
NAMUDNO has that effect. It sounds warnings about the Voting Rights Act and urges Congress to
revise the act to save the Court from the need to do so. No one expects Congress to revise the Voting
Rights Act—given the differing views in Congress and the many barriers to any legislation, it would be
a minor miracle if the revisions the Court seeks were enacted. But the lack of a response to the Court’s
call makes it easy for the Court in the future, when facing a constitutional challenge to the Voting
Rights Act, to say that because Congress failed to revise the Act, the Court will invalidate the Act.
What would seem illegitimate in one step will seem much less objectionable when done in two. This is a
bootstrap that relies on congressional inaction, but it is a safe bet. Meanwhile, the Justices most likely
to invalidate the Voting Rights Act knew, at the time NAMUDNO was decided, that their majority was
likely to persist for at least five to ten years, either because none of them would retire (the most likely
outcome) or because any Justice who did retire would be replaced by someone like-minded (since
case; and in any event disfavoring simultaneous bootstraps will impose
significant costs. In a single opinion, the Supreme Court often has a significant
holding and another significant holding that builds on the first. That is, the
process of bootstrapping via precedent sometimes occurs in a single opinion:
the Court thus undertakes Y and Z simultaneously. In Marbury v. Madison,81
for instance, the Court’s significant holding on the scope of Article III’s
original-jurisdiction provision enabled the Court to both consider the
constitutionality of the Judiciary Act and assert the Court’s power of judicial
review. The Court could have decided each of those issues separately, but that
would have produced piecemeal opinions and piecemeal processes.
Judges could try to minimize the inefficiencies by deciding all the issues at
the same time and then simply splitting the resolution into different opinions. In
effect, that is what happened in Brown v. Board of Education82 and Bolling v.
Sharpe,83 decided on the same day. But that raises the same problems as
simultaneity, namely the certainty of Z at the time that Y is decided. If judges
decide all the issues at the same time, they have engaged in simultaneous
bootstrapping. To avoid simultaneity, judges must decide the various steps
separately, and that imposes significant costs on judges and lawyers (not to
mention the parties involved in the cases).
2. The Benefits of Such a Distinction
Insofar as the fear of bootstrapping is a fear of aggrandizement, a focus on
simultaneous bootstraps is a poor fit. A bar to or strong presumption against
simultaneous bootstraps would impose enormous costs on Congress and
significant costs on the Supreme Court, but it would impose no meaningful limit
on the President. The President’s ability to aggrandize would be unaffected.
Insofar as the fear of bootstrapping is focused more directly on congressional
aggrandizement (which, of course, is the focus of the challenges to the
individual mandate), it is not obvious why presidential aggrandizement should
be exempted from this fear, and the costs imposed on courts are difficult to
defend. And a fear of congressional and judicial but not presidential
aggrandizement seems entirely arbitrary.84
Justices can time their retirements). It was a high-probability bet, at the time NAMUDNO was decided,
that if a majority was ready to invalidate the Voting Rights Act in 2009 but for the seeming
inappropriateness of acting precipitously, a majority would be just as ready to invalidate it in 2014 or
2019, at which point the Court could point to years of congressional inaction as a justification. And
NAMUDNO was a sufficiently clear hint to litigants that the Justices could reasonably expect future
challenges that would provide vehicles for considering the constitutionality of the Voting Rights Act.
81. 5 U.S. 137 (1 Cranch) (1803).
82. 347 U.S. 483 (1954).
83. 347 U.S. 497 (1954).
84. It may be that some critics of simultaneous bootstrapping are actually motivated by a desire to
make complex legislation more difficult to enact (and perhaps complex Supreme Court opinions more
difficult to create), perhaps on the theory that simultaneous bootstraps are likely not only to be
complex but also to contain policy elements that they generally will not favor, or perhaps because they
think that states should have more power relative to the federal government. If so, bootstrapping is a
proxy for policy results they do not like. In any event, if that is their concern then critics should so state,
LAW AND CONTEMPORARY PROBLEMS [Vol. 75:115
More fundamentally, this seems to be a situation in which the cure is worse
than the disease. The argument against bootstrapping is based on its
consequences. Here, the costs avoided (or benefits created) by disfavoring
simultaneous bootstraps are outweighed by the costs of disfavoring them.
Daryl Levinson has persuasively argued against the notion that our
governing institutions, and in particular Congress, have great incentives to
aggrandize their power.85 As Levinson notes, “neither the policy interests of
constituents nor the self-serving goals of officials are likely to correlate in any
systematic way with the power or wealth of government institutions.”86 This is
not a matter of mere theorizing. In the years between Wickard v. Filburn87 and
United States v. Lopez,88 the widespread assumption was that the Commerce
Clause posed no meaningful hurdle for congressional legislation, but Congress
left many areas subject only to state regulation. Congress did of course enact
many regulations, but I do not think anyone could fairly argue that Congress
inexorably increased its power, much less ran amok. Indeed, Congress increased
its role in some areas but reduced its role in others.
Note in this regard that bootstrapping is costly for the bootstrapper, because
undertaking Y is by definition a significant step in its own right. (If it is not a
significant step, then there is no bootstrap.) An actor that wants to achieve Z
has to incur the cost of Y to do so, and that cost will usually be substantial.
Consider pervasive regulation. It is true that if Congress’s goal is to regulate
farmers’ collective speech about their products, it now knows what to do:
Create a pervasive regulatory scheme and embed speech regulations as part of
it.89 So a Congress that wants to regulate farmers’ collective speech can do so,
but only if it is willing to incur the costs associated with creating the much larger
regime in which to embed it. To use another example, even if Congress had
designed EMTALA to pave the way for an individual mandate (quite unlikely,
but let’s indulge the assumption), we still have the fact that it did pass
EMTALA, and enacting any regulation entails nontrivial costs for Congress.
Indeed, simultaneous bootstraps have a significant advantage over
nonsimultaneous bootstraps. One fear about bootstraps is that an actor will
move in degrees (as in a slippery slope) and thus make its destination difficult
to determine. We the citizens will be boiled like a frog in slowly heating water.90
Simultaneous bootstraps avoid such an accretive process by containing Y and Z
so that we can then assess that argument on its merits.
85. See Daryl J. Levinson, Empire-Building Government in Constitutional Law, 118 HARV. L.
REV. 915 (2005).
86. Id. at 916.
87. 317 U.S. 111 (1942).
88. 514 U.S. 549 (1995).
89. See Glickman v. Wileman Bros. & Elliott, Inc., 521 U.S. 457 (1997); United Foods v. United
States, 533 U.S. 405 (2001); supra text accompanying notes 64–68.
90. Apparently, this is not actually true of frogs, but the metaphor has taken root nonetheless. See,
e.g., Next Time, What Say We Boil a Consultant, FAST COMPANY (Oct. 31, 1995),
http://www.fastcompany.com/magazine/01/frog.html (quoting numerous experts who said that a frog
would, in fact, jump out of the water).
in a single act. They thus have the significant advantage of transparency.
Whatever other complaints citizens may have about the ACA, they cannot
complain that Congress enacted a popular ban on preexisting conditions while
hiding its intention of enacting a less popular individual mandate. Congress
revealed both steps at the same, a benefit that would be lost if simultaneous
bootstraps were disallowed.
Interestingly, focusing on simultaneous bootstraps would divert attention
and energy from a greater danger of aggrandizement. As Levinson
acknowledges, the constraints on aggrandizement have less force with respect to
the President.91 And the President’s ability to engage in aggrandizement via
bootstrapping would be only marginally affected by a limit on simultaneous
bootstrapping. So not only would a limit on simultaneity have modest benefits,
but it would also ignore the potentially greater aggrandizer.
The point here is that the benefits of disfavoring simultaneous bootstraps
will be much smaller than its costs. The incentives to engage in bootstrapping
are not very great, and the costs of disfavoring bootstraps are substantial.
Concerns about bootstrapping do not arise from any provision in the
Constitution. They arise from a fear about the consequences of bootstrapping.
The problem, as this article has demonstrated, is that bootstrapping is
pervasive. The consequences of freely allowing bootstrapping are modest. It is
the world that we inhabit, and the sky has not fallen. The consequences of
disfavoring bootstrapping would be massive. The one distinction that has any
traction—between simultaneous and nonsimultaneous bootstraps—still creates
costs that greatly outweigh the benefits.
Whether or not those who have raised concerns about bootstrapping in the
ACA have put forward persuasive arguments against the legitimacy of the
individual mandate, the problem is not the legitimacy of bootstrapping. If the
final destination is impermissible, so be it. But that does not mean that the
pathway should be abandoned as well.
1. 347 U.S. 497 ( 1954 ).
2. 347 U.S. 483 ( 1954 ).
3. 339 U.S. 629 ( 1950 ).
4. See, e.g., Medellin v . Texas , 552 U.S. 491 , 531 ( 2008 ) (“[I]f pervasive enough, a history of congressional acquiescence can be treated as a 'gloss on “Executive Power” vested in the President by § 1 of Art . II.'”) (quoting Dames & Moore v . Regan , 453 U.S. 654 , 686 ( 1981 ) ) . For a discussion of the significance of acquiescence in creating a form of estoppel, see Curtis A . Bradley & Trevor W. Morrison , Historical Gloss and the Separation of Powers , 126 HARV. L. REV. (forthcoming 2012 ). 5. For a discussion on path dependence , see Mark J. Roe, Chaos and Evolution in Law and Economics , 109 HARV. L. REV. 641 , 643 - 62 ( 1996 ). For a discussion on slippery slopes, see generally Eugene Volokh, The Mechanisms of the Slippery Slope , 116 HARV. L. REV. 1026 ( 2003 ) and Frederick Schauer , Slippery Slopes, 99 HARV. L. REV. 361 ( 1985 ).
31. See McCulloch v . Maryland , 17 U.S. 316 ( 1819 ) ; United States v . Comstock , 130 S. Ct . 1949 ( 2010 ); Gonzales v . Raich , 545 U.S. 1 ( 2006 ) (Scalia , J., concurring).
32. U.S. CONST. art. II, § 1 (“The executive Power shall be vested in a President of the United States of America.”); id. art. III, § 1 (“The judicial Power of the United States, shall be vested in one supreme Court, and in such inferior Courts as the Congress may from time to time ordain and establish .”).
33. See , e.g., National Currency Act, ch. 106 , 13 Stat . 99 ( 1864 ) ; McFadden Act, ch . 191 , 44 Stat . 1244 ( 1927 ).
59. 521 U.S. 844 ( 1997 ).
60. Radio Act of 1927 , ch. 169 , 44 Stat . 1162, repealed by Communications Act of 1934 , ch. 652 , 48 Stat . 1064 (codified as amended at 47 U.S.C. §§ 151 - 620 ( 2006 )).
61. Ch . 652 , 48 Stat . 1064 (codified as amended at 47 U.S.C. §§ 151 - 620 ( 2006 )).
62. See , e.g., Trinity Methodist Church, S. v. Fed . Radio Comm'n, 62 F.2d 850 , 851 (D.C. Cir . 1932 ) (stating that the First Amendment applies to prior restraints, and that refusing to renew a broadcaster based on his earlier broadcasts does not implicate the First Amendment) .
63. 438 U.S. 726 ( 1978 ).
64. 521 U.S. 457 ( 1997 ).
65. 533 U.S. 405 ( 2001 ).
66. The programs at issue in the two cases were very similar. Both required growers to contribute to a program that would fund generic advertising of the product grown. Both programs were mandatory, once a majority of farmers voted to initiate it. But the Supreme Court found one program constitutional and the other unconstitutional. The difference, according to the Court in United Foods, was in the larger regulatory scheme surrounding the two advertising programs: The program sustained in Glickman differs from the one under review in a most fundamental respect. In Glickman the mandated assessments for speech were ancillary to a more comprehensive program restricting marketing autonomy. Here, for all practical purposes, the advertising itself, far from being ancillary, is the principal object of the regulatory scheme . United Foods , 533 U.S. at 411- 12 . The Sixth Circuit had similarly distinguished the mushroom program . See United Foods v. United States , 197 F.3d 221 , 222 ( 6th Cir . 1999 ).
71. See , e.g., William N. Eskridge , Jr., Vetogates , Chevron, and Preemption, 83 NOTRE DAME L. REV . 1441 , 1444 - 48 ( 2008 ) (listing nine different vetogates that legislation must overcome); McNollgast, Positive Canons: The Role of Legislative Bargains in Statutory Interpretation, 80 GEO . L.J. 705 , 720 ( 1992 ) (noting the many vetogates that can block legislative initiatives).
72. The Court has emphasized the importance of comprehensiveness in other contexts, notably including the Commerce Clause. See, for example, Hodel v . Indiana , 452 U.S. 314 , 329 n. 17 ( 1981 ), stating, A complex regulatory program such as established by the Act can survive a Commerce Clause challenge without a showing that every single facet of the program is independently and directly related to a valid congressional goal. It is enough that the challenged provisions are an integral part of the regulatory program and that the regulatory scheme when considered as a whole satisfies this test .
73. Patient Protection and Affordable Care Act, Pub. L. No. 111 - 148 , 124 Stat. 119 ( 2010 ) (codified as amended in scattered sections of 42 U .S.C.).