Earle Mack School of Law at Drexel University
Legal Studies Research Paper Series
2010-A-09

“The Failure of Analogy in Conceptualizing Private Entity
Liability Under Section 1983”
Richard Frankel

This article will be published in:
78 UMKC L. REV. 967 (2010)

This article can be downloaded without cost from
The Drexel University Earle Mack School of Law Legal Studies Research Paper Series
at http://www.ssrn.com/link/Drexel-U-LEG.html

THE FAILURE OF ANALOGY IN CONCEPTUALIZING
PRIVATE ENTITY LIABILITY UNDER SECTION 1983
© Richard Frankel*
I. INTRODUCTION
The landmark civil rights statute, 42 U.S.C. § 1983, is a significant vehicle
for individuals who suffer violations of constitutional or federal statutory rights
at the hands of state actors to hold those government officials accountable for
their misdeeds. Over its nearly 140-year history, courts have developed a rich
body of doctrine that provides rules for determining the scope of government
liability. This body of law includes rules for determining a number of issues,
such as which government actors can be held liable under the statute, the degree
to which they can be held liable, the types of remedies that they may be required
to provide, and the scope of any immunities or other liability protections to which
they might be entitled.
Section 1983 applies primarily to state and local governments and their
employees because the statute, by its plain language, applies only to persons
acting “under color” of state law.1 Through the development of the “state action”
doctrine, however, private parties that act jointly with the government or that
perform traditional government functions are considered state actors that are
required to comply with constitutional directives and are subject to § 1983.2
Although courts have applied § 1983 to private actors for at least forty
years,3 they have struggled to develop rules for determining private-party
liability, and in particular rules for private-entities—i.e. private businesses as
opposed to the individuals working for those businesses.4 Yet, questions
concerning the nature and scope of private-entity liability are becoming
increasingly important as state and local governments privatize more and more
government services. By privatization, I generally mean the “government use of
private entities to implement government programs or to provide services to
others on the government‟s behalf.”5 The last twenty years have witnessed an

*

Associate Professor of Law, Earle Mack School of Law at Drexel University. B.A. Yale
University, 1997. J.D., Yale Law School, 2001. I wish to thank the participants at the Edward A.
Smith/Bryan Cave Symposium at the UMKC School of Law for their helpful suggestions. I also
am grateful for the able research assistance of Joseph Pecora.
1
See 42 U.S.C. § 1983 (2006).
2
A private party qualifies as a state actor when, among other things, it performs a function that was
traditionally and exclusively governmental. See Lee v. Katz, 276 F.3d 550, 554-55 (9th Cir. 2002).
State action also occurs when a private party is a “willful participant in joint activity with the State
or its agents.” Brentwood Acad. v. Tenn. Secondary Sch. Athletic Assoc., 531 U.S. 288, 296
(2001) (quotation marks and citation omitted).
3
In Adickes v. S.H. Kress & Co., 398 U.S. 144 (1970), the Supreme Court found that a plaintiff
could sue a department store under § 1983 if the plaintiff could show that the store‟s employees
engaged in a joint conspiracy with the State. Id. at 152.
4
See, e.g., Barbara Kritchevsky, Civil Rights Liabilities of Private Entities, 26 CARDOZO L. REV.
35, 36, 38 (2004) (noting that the Supreme Court has “engaged in little discussion of the rules that
govern liability” of private entities and that lower courts have not been consistent in their
application of § 1983 rules to private parties).
5
Gillian E. Metzger, Privatization as Delegation, 103 COLUM. L. REV. 1367, 1370 (2003).

968

UMKC LAW REVIEW

[Vol. 78:4

expansive growth of privatization at all levels of government, including the state
and local levels.6 State and local governments have contracted with private
parties to perform a wide array of core government services,—often very
sensitive services—including operating prisons,7 providing medical care to
prisoners,8 administering welfare and public benefit programs,9 processing
parking tickets,10 providing private security services,11 collecting government
debts,12 fighting fires,13 and overseeing foster care and child placement

6

See Jody Freeman, Extending Public Law Norms through Privatization, 116 HARV. L. REV. 1285,
1289 (2003) (observing “a discernable trend toward „privatization,‟ which in the American context
consists largely of contracting out”); Metzger, supra note 5, at 1379 (“Privatization seems likely
only to expand in the near future, fueled by increasing belief in market-based solutions to public
problems.”); Paul Howard Morris, Note, The Impact of Constitutional Liability on the Privatization
Movement after Richardson v. McKnight, 52 VAND. L. REV. 489, 493 (1999).
7
As of June 30, 2007, 7.4 % of all U.S. prisoners were housed in private facilities. 30,379 federal
prisoners and 87,860 state prisoners were in private facilities. William J. Sabol & Heather Couture,
Bureau of Justice Statistics Bulletin, Prison Inmates at Midyear 2007 5 (Dept. of Just. June 2008),
available at http://www.ojp.usdoj.gov/bjs/pub/pdf/pim07.pdf. Additionally, private prisons hold
alien detainees facing deportation or other immigration-related charges. See Margaret Talbot, The
Lost Children, NEW YORKER, Mar. 3, 2008, at 58. By the mid-1980s, private facilities held nearly
thirty-percent of detained aliens. See James Austin & Garry Coventry, Bureau of Justice
Assistance, Emerging Issues on Privatized Prisons 12 (Dept. of Just. 2001), available at
http://www.ncjrs.gov/pdffiles1/bja/181249.pdf.
8
Correctional Medical Services (“CMS”), one of the larger prison healthcare companies, alone
provides health services to over 260,000 inmates. See Correctional Medical Services, About Us,
http://www.cmsstl.com/about_us.aspx (last visited Mar. 29, 2010). Prison Health Services
(“PHS”), provides health care for an additional 270,000 prisoners. See Alfred C. Aman, Jr., An
Administrative Law Perspective on Government Social Service Contracts: Outsourcing Prison
Health Care in New York City, 14 IND. J. GLOBAL LEGAL STUD. 301, 302 (2007). CMS estimates
that forty-two percent of all prison health services are provided by private companies. See
Correctional Medical Services, About Us, supra.
9
As many as forty states have privatized aspects of their welfare and public benefits administration
and delivery programs and have spent billions on contracts with private welfare providers. See
Matthew Diller, The Revolution in Welfare Administration: Rules, Discretion and Entrepreneurial
Government, 75 N.Y.U. L. REV. 1121, 1180 (2000); Metzger, supra note 5, at 1383-88; Dru
Stevenson, Privatization of Welfare Services: Delegation by Commercial Contract, 45 ARIZ. L.
REV. 83, 87-88 (2003).
10
See, e.g., Ace Beverage Co. v. Lockheed Information Mgmt. Servs., 144 F.3d 1218, 1219 (9th
Cir. 1997) (finding that a private company that processed parking tickets for the city of Los
Angeles acted under color of state law for purposes of § 1983).
11
See, e.g., Powell v. Shopco, 678 F.2d 504, 505 (4th Cir. 1982) (finding that private security guard
could be sued under §1983); Groom v. Safeway, 973 F. Supp 987, 991-92 (W.D. Wash. 1997)
(holding that a grocery store acted under color of state law when it hired an off-duty police officer
to work as a private security guard); see also Flagg Bros. v. Brooks, 436 U.S. 149, 163-64 (1978)
(identifying police protection as a traditional public function).
12
See, e.g., Del Campo v. Kennedy, 517 F.3d 1070, 1072-73 (9th Cir. 2008) (describing how a
private company sought to collect debts owed to the state of California).
13
See Jesse McKinley, On the Fire Lines, a Shift to Private Contractors, N.Y. TIMES, Aug. 18,
2008, at A11; see also Flagg Bros., 436 U.S. at 163-64 (identifying fire protection as a traditional
public function).

2010]

THE FAILURE OF ANALOGY

969

programs.14 Moreover, many of these services, such as operating prisons or
providing security—as opposed to building roads, say—are the types that create
conditions in which constitutional violations by the private entities providing
these services.15
As privatization increases, it is important for courts to develop a clear, wellreasoned framework for addressing questions of private-entity liability in order to
ensure that citizens who receive public services from private providers do not
face undue risks to their constitutional rights.16 Currently, one of the approaches
that courts commonly take in determining private-entity liability is that of
analogy. Courts often have determined the scope of private-entity liability by
analogizing private entities to government actors. When courts determine that a
private-entity resembles a government actor, the private-entity tends to be subject
to the same liability rules, and to receive the same protections from liability, as
the relevant government actor.17 When courts determine that the character or

14

See, e.g., Donlan v. Ridge, 58 F. Supp. 2d 604, 610-11 (E.D. Pa. 1999) (holding that a private
foster care provider acted under color of state law); Bartell v. Lohiser, 12 F. Supp. 2d 640, 649-50
(E.D. Mich. 1998) (finding that a private contractor who provided foster care acted under color of
state law and could receive qualified immunity).
15
Privatization also has been popular at the federal level. At the federal level, government
contracting to private parties increased by fifty-percent from 2002 to 2005. See Kerry Korpi, Panel
Discussion, Outsourcing Government? The Privatization of Public Responsibilities (American
Constitution
Society
Annual
Convention
June
13,
2008),
available
at
http://www.acslaw.org/node/6787 (Kerry Korpi‟s comments appear in the audio recording
available at this url). Some of this involves particularly sensitive functions as well, including
intelligence gathering and national security. The State Department contracted out to private
security firms like Blackwater USA to provide security in Iraq, which included the ability to use
deadly force when necessary. See, e.g., John M. Broder & David Rohde, State Dept. Use of
Contractors Leaps in 4 Years, N.Y. TIMES, Oct 24, 2007, at A1; see also TIM SHORROCK, SPIES FOR
HIRE: THE SECRET WORLD OF INTELLIGENCE OUTSOURCING (Simon & Schuster 2008) (describing
privatization of intelligence services). However, while private parties performing state action can
be sued under § 1983, there is no corresponding cause of action for constitutional violations against
private parties who perform government functions for the federal government. See Correctional
Servs. Corp. v. Malesko, 534 U.S. 61 (2001).
16
Section 1983 provides a cause of action for violations of both constitutional rights and federal
statutory rights. For ease of reference, because many § 1983 claims arise under the Constitution,
this paper refers to § 1983 claims as involving constitutional rights.
17
For example, in Monell v. N.Y. City Dep’t of Social Servs., 436 U.S. 658, 690-94 (1978), the
Supreme Court held that municipal § 1983 defendants were exempt from the traditional tort rule of
respondeat superior liability. Lower courts have extended that same rule to private-entity
defendants, on the ground that Monell‟s reasoning also applies to private entities. See, e.g., Powell
v. Shopco, 678 F.2d 504, 506 (4th Cir. 1982) (concluding that “[n]o element of the [Monell]
Court‟s ratio decidendi lends support for distinguishing the case of a private corporation [from a
municipality].”). Most lower courts have reached the same result. See, e.g., Kritchevsky, supra
note 4, at 55 n.149 (listing cases). For critiques of exempting private § 1983 defendants from
respondeat superior liability, see Hutchison v. Brookshire Brothers, Ltd., 284 F. Supp. 2d 459, 473
(E.D. Tex. 2003); Segler v. Clark County, 142 F. Supp. 2d 1264, 1268-69 (D. Nev. 2001); Taylor v.
Plousis, 101 F. Supp. 2d 255, 263 n.4 (D.N.J. 2000); Richard H. Frankel, Regulating Privatized

970

UMKC LAW REVIEW

[Vol. 78:4

behavior of the private entity is distinguishable from that of the governmental
comparator, then courts tend not to give the private entity the same protections
from liability that the government entity receives.18
While analogy is an essential part of legal reasoning and often serves as a
useful interpretive tool, this paper suggests that in the context of private-entity
liability in § 1983 actions, judicial reliance on analogy leads to poor results that
impair victims of constitutional violations from vindicating their rights. Instead
of using analogy as a tool for understanding or applying the purposes and
principles of § 1983, the analogy itself takes primacy and becomes the basis for
the court‟s decision. The courts‟ reasoning centers on whether or not an analogy
can be drawn, often at the expense of the underlying purpose that the analogy is
supposed to further. This focus on analogy has two drawbacks. First, it can lead
to misguided and counter-productive results that subvert § 1983‟s principles and
that insufficiently protect constitutional rights. Second, where the analogies that
courts draw are themselves flawed, they have created an unprincipled doctrine
regarding private-entity § 1983 defendants.
Instead of attempting to determine whether private entities fit or do not fit
into the different boxes of various governmental defendants, courts should treat
private-entities as their own category. Private entities have certain characteristics
that make them resemble individual government entities and certain
characteristics that make them different from government entities.19 The same is
true when it comes to comparing private entities with individual government
employees.20 Courts should look at private entities as exactly that, private
entities, and should focus on the specific characteristics of private entities in
deciding which liability rules to apply to them in order to best serve § 1983‟s
purposes.
Removing analogy as the framework for determining private-entity liability,
however, does not help determine what framework should take its place. This
paper proposes that in addressing private-entity liability under § 1983, courts can
utilize the huge body of law that already exists for determining the liability of
private parties that commit injurious acts—private tort law. Although tort law
principles often have been used to restrict the scope of § 1983,21 in the privateentity context, the application of constitutional principles specific to § 1983, as
opposed to tort principles, have actually made it more difficult for individuals to

Government Through Section 1983, 76 U. CHI. L. R EV. 1449 (2009); Kritchevsky, supra note 4, at
73-76.
18
See, e.g., Del Campo v. Kennedy, 517 F.3d 1070, 1072-73 (9th Cir. 2007) (concluding that
private party sued under § 1983 was not entitled to Eleventh Amendment state sovereign
immunity).
19
See infra notes 52-57 and accompanying text.
20
Private corporations often receive certain benefits from the state that individuals do not get, in
terms of limited liability and particular tax breaks. But corporations are also distinctly private
entities with an identity separate from the state. Private entities also are unlike individuals in other
ways. The main purpose of taking on corporate status is to separate the identities of the
corporation‟s officers from the identity of the corporation itself.
21
See infra notes 144-46 and accompanying text.

2010]

THE FAILURE OF ANALOGY

971

vindicate constitutional rights and resorting to ordinary tort principles may more
effectively further § 1983‟s purposes.
This paper proceeds in four parts. Part II describes the current approach to
determining private-entity liability and explains how courts often rely on analogy
as their interpretive method. Part III offers some hypotheses about why courts
may tend to use analogy in the private-entity context. Part IV examines three
recent Supreme Court decisions in § 1983 or Bivens cases22—two involving
private defendants and one involving governmental defendants—to show how
the Court‟s reliance on analogy in each case led to an unsound or misguided
result. Part V suggests that courts should stop making analogy the primary focus
with respect to private-entity liability and proposes that courts instead look to
private tort-law principles as a guide for assessing private-entity § 1983 liability.
II. THE CURRENT APPROACH TO PRIVATE-ENTITY LIABILITY
UNDER § 1983
Section 1983 provides a cause of action to any individual whose
constitutional or federal statutory rights are violated by persons acting “under
color” of state law.23 The Supreme Court has described the main purposes of the
statute as deterring future constitutional violations and ensuring compensation for
those who suffer constitutional injury.24 Because of the “under color” of state
law requirement, § 1983 actions typically involve government defendants—state
governments, municipalities, or state or municipal employees in either their
official25 or individual capacities. Over the course of the statute‟s life, the courts
have developed various principles and doctrines governing liability for different
categories of government defendants. For example, different immunities and
protections from liability apply depending on whether the defendant is a state
government, a municipality, a government employee, or a supervisory employee;
and then even within those categories, different employees have different
protections depending on the particular government functions they perform.26
While courts have developed a robust body of law for the liability of
different government actors, they have had less opportunity to develop doctrines
regarding the liability of private entities and individuals that act “under color” of

22

A Bivens action is the federal equivalent to § 1983. In Bivens v. Six Unknown Named Agents of
Federal Bureau of Narcotics, 403 U.S. 388 (1971), the Supreme Court established an implied right
of action against federal officers for the violation of certain constitutional rights. Unlike § 1983,
only certain constitutional violations can be remedied through Bivens actions, and unlike § 1983,
Bivens does not apply to private companies acting under color of federal law. See Corr. Servs.
Corp. v. Malesko, 534 U.S. 61 (2001).
23
42 U.S.C. § 1983 (2006).
24
See Owen v. City of Independence, 445 U.S. 622, 651 (1980) (“Moreover, § 1983 was intended
not only to provide compensation to the victims of past abuses, but to serve as a deterrent against
future constitutional deprivations as well.”).
25
Suits against state officials in their official capacities are equivalent to suing the state itself. Will
v. Mich. Dep‟t of State Police, 491 U.S. 58 (1989).
26
See infra notes 37-43 and accompanying text.

972

UMKC LAW REVIEW

[Vol. 78:4

state law and thus are subject to § 1983. In part this may be because courts have
had fewer opportunities to deal with private § 1983 defendants. Although there
is a long history of private actors performing governmental functions, only in the
last couple decades has the privatization of state and local government functions
really taken off. Perhaps for this reason, when courts have faced questions
regarding private-entity liability, they have tended to take the more-developed
law that exists for government defendants and to apply it by analogy to private
defendants. In other words, where a liability issue arises with respect to a
private-entity defendant, courts appear to respond by asking whether the privateentity is akin to or distinct from the government actor. If the former then the
liability rule applying to the government actor will also apply to the private
actor,27 and if not, then the liability rule will not apply to the private-entity.28
For example, in Monell v. Department of Social Services, the Supreme
Court held that municipalities were “persons” that could be sued for damages
under § 1983, but that they were exempt from the traditional agency rule of
respondeat superior liability.29 The Court held that a municipality could be liable
for damages only if the plaintiff could show that the municipality directly caused
the constitutional violation through the application of some municipal policy or
custom.30 When private companies have faced suit under § 1983, courts have
almost universally concluded that private-entities are like municipalities and
therefore also exempt from respondeat superior liability.31 Analogy appears to be
the dominant framework by which courts assess the liability rules applicable to
private § 1983 defendants.
The decision to use analogy as an interpretive tool, however, does not
indicate the type of analogy that should be drawn. In the respondeat superior
example, courts appear to have simply assumed that a municipality is the most
comparable defendant to a private entity.32 But it is not at all clear that is the
case. In any particular privatization situation, there may be any number of
different government actors to which a private entity could be analogized. Take,
for example, a private company that contracts with a municipality to perform
certain core government functions, such as operating a municipal jail, or

27

See, e.g., Malesko, 534 U.S. at 70-71 (concluding that no Bivens cause of action exists against a
private entity because no cause of action exists against federal agencies and private corporations are
like federal agencies). Malesko is discussed in greater detail in Part IV.B.
28
See, e.g., Richardson v. McKnight, 521 U.S. 399 (1997) (concluding that private prison guards
are not entitled to qualified immunity even though state prison guards are entitled to qualified
immunity). Richardson is discussed in greater detail in Part IV.A.
29
436 U.S. 658, 691 (1978).
30
See Monell, 436 U.S. at 694.
31
See supra note 17.
32
See, e.g., Powell v. Shopco, 678 F.2d 504, 506 (4th Cir. 1982) (deciding to extend Monell to
private corporations because “[n]o element of the [Monell] Court‟s ratio decidendi lends support
for distinguishing the case of a private corporation [from a municipality].”).

2010]

THE FAILURE OF ANALOGY

973

administering welfare benefits or other social services.33 One could analogize the
private contractor to a municipality on the theory that it is standing in the shoes
of or acting as the equivalent of the municipality. But one also could analogize
the private contractor to a government employee on the theory that the private
contractor has been hired by the municipality, just as it hires employees, to
perform particular services. Alternatively, one could analogize the private
contractor to a supervisory government employee as opposed to the employee
committing the unconstitutional act, on the reasoning that while the private
company is hired by the municipality to perform a function, that function
ultimately is carried out by the company‟s employees, who are then supervised
by company officials. While each of these analogies may have their own
strength and weaknesses, it is not as self-evident as courts make it appear which
analogy is the proper one, or if any analogy should be used at all.34
Indeed, Professor Barbara Kritchevsky has shown that courts tend to
analogize to different government actors in different situations. After examining
a series of lower court decisions regarding private entities, she concludes that
when addressing the scope of entity liability, courts generally “analogize private
and government entities,” but when it comes to questions of immunity, they tend
to analogize private entities to individual government employees.35 Professor
Kritchevsky then goes on to demonstrate how the analogies that courts are
drawing lead to poor decisions in particular contexts, specifically with respect to
qualified immunity, punitive damages, sovereign immunity, and vicarious
liability.36
Although the question of which analogy to draw may seem like an
academic one, it can have significant implications for a plaintiff‟s ability to
obtain relief for the constitutional violations of private actors operating under
color of state law because of the different liability rules that apply to different
defendants. For example, while municipalities may be sued for damages under §
1983,37 state entities are protected from damages by Eleventh Amendment
immunity38 (though state officials in their official capacity can still be sued for

33

Such actions typically would qualify as state action. See, e.g., West v. Atkins, 487 U.S. 42
(1988) (holding that a private medical personnel were state actors when under contract to provide
medical services to prison inmates).
34
In Correctional Services Corp. v. Malesko, 534 U.S. 61 (2001), discussed infra in Part IV.B., the
debate between the majority and the dissent was about what type of analogy should be drawn. The
majority drew an analogy between private corporations and federal agencies to conclude that there
was no Bivens cause of action against a private company. Id. at 70-71. The dissent, by contrast,
characterized the private company as an agent of the federal government and concluded that
corporate agents, just like human agents, could be sued under Bivens. Id. at 78-83 (Stevens, J.,
dissenting).
35
Kritchevsky, supra note 4, at 38-39.
36
Id. at 50-70.
37
Monell v. Dep‟t of Soc. Servs., 436 U.S. 658, 663 (1978).
38
Will v. Mich. Dep‟t of State Police, 491 U.S. 58, 66-67 (1989) (concluding that states have
Eleventh Amendment immunity from damages under §1983).

974

UMKC LAW REVIEW

[Vol. 78:4

injunctive relief39). Individual defendants can be subject to punitive damages 40
while municipal defendants are immune from punitive damages.41 On the other
hand, individual government defendants are entitled to qualified immunity42
while municipalities are not.43
Like Professor Kritchevsky, this paper also questions the use of analogy for
determining § 1983 liability. While Professor Kritchevsky focuses on specific
liability issues facing private entities—such as qualified immunity, liability for
punitive damages, and the exemption from respondeat superior liability44—this
paper looks at analogy as an interpretive framework and asks whether it is an
effective one for addressing the liability of private § 1983 defendants.
Kritchevsky suggests that the use of analogy by lower courts is flawed because it
deviates from Supreme Court precedent indicating that courts should determine
liability based on history and policy.45 This paper, in the Sections that follow,
suggests that the Supreme Court‟s own decisions, and its own use of analogy,
highlight the failure of analogy in the § 1983 context and demonstrates why the
framework of analogy as a whole, rather than its application in specific § 1983
contexts, should be reconsidered. This paper examines three specific Supreme
Court decisions in § 1983 or Bivens cases in which analogy has played a role in
the Court‟s reasoning. Based on those cases, this paper suggests that in the §
1983 context, the analogy framework itself is flawed, not necessarily because it
leads to bad results, but because it causes the court to focus on the wrong
questions. The use of analogy is problematic because the analogy takes center
stage at the expense of the statute‟s underlying purposes. The Court tends to
focus more on whether a private-entity is like or unlike a particular government
actor than on whether applying a particular liability rule to a private entity would
further the goals of § 1983. In so doing, the Court ends up making decisions that
unduly narrow the scope of § 1983 and limit the ability of injured plaintiffs to
vindicate rights protected by the Constitution.
III. REASONS THAT COURTS TEND TO USE ANALOGY
Analogy is just one of many interpretive frameworks that courts can use to
resolve legal questions. Yet comparing private entities to government actors
seems to be a common way of addressing private-entity liability in the § 1983
context. This Part offers some thoughts on why courts might tend to use analogy
and why they have struggled with how to use analogies.

39

See Kentucky v. Graham, 473 U.S. 159, 165-67, 167 n.14 (1985) (citing Ex parte Young, 209
U.S. 123 (1908)).
40
See Smith v. Wade, 461 U.S. 30 (1983).
41
City of Newport v. Fact Concerts, Inc., 453 U.S. 247 (1981).
42
See, e.g., Saucier v. Katz, 533 U.S. 194 (2001) (describing the contours of qualified immunity).
43
Owen v. City of Independence, 445 U.S. 622 (1980).
44
Kritchevsky, supra note 4, at 70-78 (critiquing the application of particular liability rules to
private parties).
45
Id. at 38-39.

2010]

THE FAILURE OF ANALOGY

975

One reason that courts may feel tempted to use liability rules relating to
government actors as a benchmark for determining private entity liability is
because of the state action doctrine. Private entities are subject to § 1983 only if
they act “under color of state law.”46 A private party acts under color of state law
where the party‟s conduct is “fairly attributable to the State.”47 The Supreme
Court has articulated various tests for defining state action, including whether a
private party performs functions “traditionally exclusively reserved to the
State,”48 whether the private party is a “a willful participant in joint activity with
the State or its agents,”49 or whether “there is a sufficiently close nexus between
the State” and the private party‟s action such that the private party‟s action “may
be fairly treated as that of the State itself.”50
In other words, a private party is subject to § 1983 only when it is behaving
like a government actor or is standing in the shoes of the government in some
way. Given the requirement that a private party act like a state, it is only natural
to look to the liability rules regarding government actors in determining the
liability of private entities acting under color of state law. Put another way, if
private parties are only subject to § 1983 in the first place by virtue of engaging
in state action, then it seems reasonable that they should be subject to the same
liability rules as government actors, or at the very least that the liability rules for
governments should provide a starting point and a framework for addressing
private party liability.51
There is no reason, however, why the state action doctrine should determine
the framework for determining private liability rules. Whether an entity is
considered to be sufficiently similar to the State for determining whether it is
subject to § 1983 at all does not automatically mean that the entity needs to be
compared to a State in addressing the scope of liability, and in particular, the
protections from liability that it may or may not be entitled to receive.
A second reason why courts might resort to analogy in determining private
party liability is that it provides a cover to allow the court to impose its own
policy preferences on the shape of § 1983 doctrine. Because private entities are
not exactly like government entities and not exactly like government employees,
there is a lot of wiggle room to allow judges to draw almost any type of analogy
that they wish. Many scholars have criticized the Court‟s § 1983 as
indeterminate and as reflecting the Justices own individual policy preferences

46

West v. Atkins, 487 U.S. 42, 48 (1988).
Lugar v. Edmondson Oil Co., 457 U.S. 922, 937 (1982).
48
Jackson v. Metro. Edison Co., 419 U.S. 345, 352 (1974).
49
Adickes v. S.H. Kress & Co., 398 U.S. 144, 152 (1970).
50
Jackson, 419 U.S. at 351.
51
The state action doctrine has been heavily criticized on various grounds, see, e.g., Daphne BarakErez, A State Action Doctrine for an Age of Privatization, 45 SYRACUSE. L. REV. 1169, 1172 n.10
(listing articles critiquing the state action doctrine), and in particular on the ground that it is poorly
suited to deal with privatization in its modern form, see generally id. The fact that the state action
doctrine might not effectively account for current forms of privatization also suggests that perhaps
the state action requirement also should not be a reason to rely so heavily on analogy to determine
private party liability under § 1983.
47

976

UMKC LAW REVIEW

[Vol. 78:4

more than any coherent doctrinal viewpoint.52 Given the various types of
government actors to which a private entity could be analogized—municipalities,
states, supervisory employees, ordinary employees-- and given that each type of
actor is subject to different liability rules, it may be that courts use analogy so
that they can decide which government actor to analogize to based on the
outcome that it wants to reach.
A third, related, reason may be based on litigation tactics and the way that
lawyers frame the issue of private-entity liability and present it to the Court. Just
as the range of possible analogies gives courts freedom to choose a comparator
based on the result that it wants, litigants may frame private party liability
questions in terms of analogy in order to try and achieve the results that their
clients want. To the extent that private-entity liability issues remain unresolved,
litigants have an interest in analogizing to, or distinguishing from, whichever
government actor will either maximize liability (from the plaintiff‟s perspective)
or minimize liability (from the defendant‟s perspective).
Finally, the tendency to analogize and the difficulty the courts have in
settling on particular analogies may come from the fact that it is not easy to
pinpoint whether and how corporations resemble persons, states, or other entities.
Corporations have characteristics of state entities, of private entities, and of
natural persons. Thus, it is easy for one person to think that corporations are like
the government and another to say that it is unlike the government, which opens
up opportunities for courts to draw different analogies in different circumstances.
In one sense, corporations may seem more like government entities than
individuals do. Unlike natural persons, the private company‟s existence would
not exist but for the sanction of the State, and in that sense, private companies are
creatures of the state in ways that individuals are not.53 Additionally, while
corporations—at least publicly-traded corporations—exist to serve the private
ends of their shareholders, historically, particularly in the 17th and 18th
centuries, corporations were created by state charter so that they could trade and
operate in pursuit of the State‟s mercantilist goals.54 Now, by contrast, the level
of state involvement in and public purpose of corporations is much more limited.
Just as corporations have characteristics that make them seem like
government entities, they also have some characteristics that resemble individual
persons. Corporations are typically treated as persons under state and federal

52

See, e.g., Jack M. Beermann, A Critical Approach to Section 1983 with Special Attention to
Sources of Law, 42 S TAN . L. REV. 51, 53 (1988) (describing the “indeterminacy” of the Court‟s §
1983 doctrine); Richard A. Matasar, Personal Immunities Under Section 1983: The Limits of the
Court’s Historical Analysis, 40 ARK. L. R EV. 741, 774, 794 (1987); Michael Wells, The Past and
the Future of Constitutional Torts: From Statutory Interpretation to Common Law Rules, 19 CONN.
L. REV. 53, 54, 87-88 (1986).
53
See, e.g., Hale v. Henkel, 201 U.S. 43, 74-75 (1906) (distinguishing between corporate identity
and individual identity in determining that corporations do not have Fifth Amendment protections
against self-incrimination).
54
See ALAN R. PALMITER, CORPORATIONS, 7-8 (6th ed. 2009); see also JAMES D. COX & THOMAS
LEE HAZEN, CORPORATIONS 32 (2d ed. 2003) (“The dominant feature of businesses incorporated in
the eighteenth century was their public character.”).

2010]

THE FAILURE OF ANALOGY

977

law, and state and federal statutes “typically give a corporation „the same powers
as an individual to do all things necessary or convenient to carry out its business
and affairs.‟”55 On the other hand, one of the main purposes of corporate
status—the protection of limited liability—is predicated on the notion that
corporate identity is distinct from the identities of the people running it.56
Perhaps because of these dueling properties of the corporate form, courts have
struggled in areas of constitutional law in particular to decide whether
corporations should be considered to be alike or different from individuals.
Thus, corporations are persons for purposes of the Fourth Amendment‟s
protections against searches and seizures and the Fifth Amendment‟s prohibition
on double jeopardy, but not a person for purposes of the Fifth Amendment‟s
protection against self-incrimination.57 Similarly, corporations are protected by
the Due Process and Equal Protection Clauses of the Fourteenth Amendment but
not by the Fourteenth Amendment‟s Privileges and Immunities Clause.58
While these explanations may offer some insight into why analogy is often
used in the private-entity context, none of them necessarily suggests that analogy
should be used. The next Part explores whether the use of analogy may actually
be counter-productive in determining private-entity liability, by examining three
Supreme Court opinions in § 1983 and Bivens cases in which analogy played a
significant role in the Court‟s decision.
IV. THE FAILURE OF ANALOGY—THREE EXAMPLES
Three examples from recent Supreme Court decisions help show some of
the problems with the use of analogy that arise in the § 1983 and Bivens contexts.
Not all of the examples are § 1983 cases. In fact, two examples arise from
Bivens actions,59 but I think they are nonetheless appropriate because of the
similarities between Bivens and § 1983 actions and because I think the decisions
are revealing about the Court‟s decision-making process in constitutional civil
rights actions. Additionally, although one example involves a government
defendant,60 I think the example is instructive because it reveals the danger of
analogizing a case dealing with one class of defendants to another class of
defendants.

55

STEPHEN M. BAINBRIDGE, CORPORATE LAW 2 (2d ed. 2009) (quoting the M ODEL BUS. CORP.
ACT § 3.02 (2006)).
56
Id. (“In the eyes of the law, the corporation is an entity wholly separate from the people who own
it and work for it.”).
57
See, e.g., G.M. Leasing Corp. v. United States, 429 U.S. 338 (1977) (search and seizure); United
States v. Martin Linen Supply Co., 430 U.S. 564 (1977) (double jeopardy); Hale v. Henkel, 201
U.S. 43 (1906) (self-incrimination).
58
See, e.g., First Nat‟l Bank v. Bellotti, 435 U.S. 765 (1978) (due process); Ashbury Hosp. v. Cass
County, 326 U.S. 207 (1945) (Privileges and Immunities Clause); Santa Clara County v. S. Pac.
R.R., 118 U.S. 394 (1886) (equal protection).
59
See Ashcroft v. Iqbal, 129 S. Ct. 1937 (2009); Corr. Servs. Corp. v. Malesko, 534 U.S. 61 (2001).
60
See Ashcroft, 129 S. Ct. at 1937.

978

UMKC LAW REVIEW

[Vol. 78:4

A. Richardson v. McKnight, 521 U.S. 399 (1997)—How a Flawed Analogy
May Cause a Good Result to Be Limited or Opened Up to Attack
In Richardson v. McKnight,61 the Supreme Court addressed the applicability
of qualified immunity to employees of private companies performing state
functions. The Court ultimately concluded that the defendants were not entitled
to qualified immunity,62 a result with which I agree, but it reached that result in
part on reliance on what I believe to be a flawed analogy that could limit the
reach of the decision and leave it vulnerable to attack.
The defendants in Richardson were correctional officers working at a
private prison that had contracted with the State of Tennessee to incarcerate some
of the State‟s inmates.63 The plaintiff, Ronnie Lee Richardson, sued the officers
under § 1983, alleging that the officers placed him in unreasonably tight
restraints in violation of his constitutional rights.64 In defending the action, the
officers raised the defense of qualified immunity.65 Under the defense of
qualified immunity, individual government defendants are immune from
damages under § 1983—even if they violate constitutional rights—unless their
unconstitutional conduct violates “clearly established law” of which “a
reasonable person would have known.”66
The Supreme Court held that although government employees are entitled
to qualified immunity under § 1983, the private prison guards in Richardson
were not.67 The Court‟s decision rested in significant part on analogy. The Court
concluded that employees of a private company are different from government
employees in “critical” ways and that as a result they are not entitled to
immunity.68
In drawing this distinction, the Court first identified the main purpose of
qualified immunity, which in its view, is “to ensure that talented candidates were
not deterred by the threat of damages suits from entering public service.”69
Without the protection of qualified immunity, the fear of being sued would cause
employees to be overly timid in their behavior and insufficiently vigorous in the
performance of important public duties.70

61

521 U.S. 399 (1997).
Id. at 402.
63
Id. at 401-02.
64
Id. at 402.
65
Id.
66
See, e.g., Harlow v. Fitzgerald, 457 U.S. 800, 818 (1982) (“We therefore hold that government
officials performing discretionary functions, generally are shielded from liability for civil damages
insofar as their conduct does not violate clearly established statutory or constitutional rights of
which a reasonable person would have known.”).
67
Richardson v. McKnight, 521 U.S. 399, 401 (1997).
68
Id. at 409 (noting the existence of “certain important differences” between government and
private employees “that, from an immunity perspective, are critical”).
69
Id. at 408 (quoting Wyatt v. Cole, 504 U.S. 158, 167 (1992)).
70
Id. (citing Butz v. Economou, 438 U.S. 478, 506 (1978)).
62

2010]

THE FAILURE OF ANALOGY

979

The Court then concluded that these purposes did not apply to private
employees in the same way as they did to government employees because, unlike
government employees and agencies, private employees and firms are subject to
“competitive market pressures” that mitigate the risk of “unwarranted timidity.”71
Specifically, the Court noted that private companies can buy liability insurance
and can indemnify their employees,72 both of which substantially reduce an
individual‟s liability exposure for a particular act. The Court also noted that
because private firms are subject to marketplace pressures and government
monitoring, firms that employ either overly aggressive employees who violate
constitutional rights or overly timid employees who fail to carry out their duties
because of the fear of being sued can be easily replaced with another competitor
that employs better personnel.73 Thus, the Court concluded that because private
employees operate under different conditions than public employees, they should
not receive the protection of qualified immunity that public employees receive.
So far, this reasoning does not seem problematic. The availability of
insurance and employee indemnification probably does reduce the risk of
employee timidity and therefore may well justify exposing private employees to
the full measure of damages liability without the protection of qualified
immunity. But by resting its decision within the framework of analogy—
specifically that these factors make private companies different from
governments, and therefore, because private companies are different, they are
not entitled to qualified immunity—the Court‟s decision becomes vulnerable to
attack and limitation.
A main problem of the Court‟s decision, as the Richardson dissenters point
out, is that the distinction that the Court draws is flawed. While the Court
focuses on the availability of insurance and employee indemnification, it
overlooks the fact that many government institutions purchase liability insurance
and provide for employee indemnification. Research shows that plenty of
government entities, particularly municipalities and local government bodies,
purchase liability insurance.74 Conversely, not all private companies buy liability
insurance. Even some companies that face many § 1983 actions, such as private

71

Id. at 409 (“[T]he most important special government immunity-producing concern—
unwarranted timidity—is less likely present, or at least is not special, when a private company
subject to competitive market pressures operates a prison.”).
72
Id. at 411 (noting that insurance coverage and employee indemnification “reduce[] the
employment-discouraging fear of unwarranted liability potential applicants face”).
73
Id. at 409 (“Competitive pressures mean not only that a firm whose guards are too aggressive
will face damages that raise costs, thereby threatening its replacement, but also that a firm whose
guards are too timid will face threats of replacement by other firms with records that demonstrate
their ability to do both a safer and a more effective job.”).
74
See, e.g., Susan A. McManus & Patricia A. Turner, Litigation as a Budgetary Constraint:
Problem Areas and Costs, 53 PUB. ADMIN. REV. 462, 468 (1993) (noting that 73.8% of respondents
in a survey of municipalities indicated that they possessed litigation insurance, or that they were not
self-insured, and that a majority of those surveyed believed that their litigation costs affected or
would affect their insurance premiums).

980

UMKC LAW REVIEW

[Vol. 78:4

prison companies, choose to self-insure rather than to purchase insurance.75
Similarly, while private companies may often choose to indemnify their
employees when those employees are sued, it appears that employee
indemnification is also widespread among public bodies, both at the state and
local level as well as at the federal level.76 Moreover, as the Richardson
dissenters argue, what is important from an immunity perspective is not whether
governments actually insure and indemnify, but whether they can insure and

75

For example, Corrections Corporation of America (“CCA”), the nation‟s largest private prison
operator, is “subject to substantial self-insurance risk” for ongoing litigation. See Corrections
Corporation of America, Corrections Corporation of America 2008 Annual Report 25, available at
http://www.correctionscorp.com (follow “Investors” link; then follow “2008 Annual Report”
hyperlink); Gary T. Schwartz, The Ethics and Economics of Tort Liability Insurance, 75 CORNELL
L. REV. 313, 315 (1990) (noting that many defendants are self-insured). Similarly, some
government bodies may choose self-insurance over traditional insurance. See McManus & Turner,
supra note 74, at 468 (indicating that a number of municipalities have chosen self-insurance,
particularly as their liability costs have risen).
76
See, e.g., Douglas L. Colbert, Bifurcation of Civil Rights Defendants: Undermining Monell in
Police Brutality Cases, 44 HASTINGS L.J. 499, 547 n.258 (1993) (noting that many state and local
laws require indemnification of police officers); Lant B. Davis, et al., Project, Suing the Police in
Federal Court, 88 YALE L.J. 781, 810-11 (1979) (concluding from a sample of cases from
Connecticut that police departments routinely footed the cost of lawsuits against individual
officers); Richard Emery & Ilann Margalit Maazel, Why Civil Rights Lawsuits Do Not Deter Police
Misconduct: The Conundrum of Indemnification and a Proposed Solution, 28 FORDHAM URB. L.J.
587 (2000). It also appears that indemnification is common at the federal level. See Cornelia T.L.
Pillard, Taking Fiction Seriously: The Strange Results of Public Officials’ Individual Liability
under Bivens, 88 GEO . L.J. 65, 76-78 & n.51 (1999) (noting that the federal government
represented ninety-eight percent of Bivens defendants who requested counsel).
The Richardson Court‟s conclusions about indemnification may not have been entirely
their fault. The federal government suggested in its brief in Richardson that indemnification at the
federal level is hardly guaranteed. Brief for the United States et al. as Amici Curiae Supporting
Respondents, Richardson v. McKnight, No. 96-318, 1997 WL 63323, at *19 (U.S. Feb. 7, 1997)
(noting that while private employers have “broad latitude” to indemnify employees, “[t]he
government has only a limited ability to indemnify its employees for constitutional tort liability and
is unable to promise in advance to indemnify employees for judgments of unknown magnitude”);
see also Oral Argument of Jeffrey A. Lamken on Behalf of the United States, Corr. Servs. Corp. v.
Malesko, No 00-860, 2001 WL 1182728, at *24 (U.S. Oct. 1, 2001) (“The Government, for
example, does not routinely idemnify its employees before a judgment or even necessarily after
judgment. On occasion we both decline to indemnify them.”).
Additionally, it is conceivable that private employers may indemnify their employees less
frequently, or to a lesser degree, than the government. See Oral Argument of Jeffrey A. Lamken on
Behalf of the United States, Corr. Servs. Corp. v. Malesko, 2001 WL 1182728, at *24-25 (stating
that a “corporation will not necessarily pick up the tab” for lawsuits against their employees and
arguing that corporations should not indemnify automatically, but only where indemnification is
“in the corporate interest”). To the extent that private employees are less likely to be unionized
than government employees that may make it less likely that they have indemnification as one of
their employee benefits.

2010]

THE FAILURE OF ANALOGY

981

indemnify.77 Finally, the notion that private companies face replacement if their
employees are either overly aggressive or overly timid assumes the existence of
an efficiently functioning marketplace with many competitors. With respect to
the market for many government functions, however, there appear to be few
competitors and substantial barriers to changing providers in mid-stream.78 Thus,
to the extent that the Court based its decision on the differences between
government and private entities, that distinction is a flawed one.
Still, an appropriate response might be to ask “so what?” If there is little
justification for qualified immunity for private employees and if that is the result
that the Court reached, then perhaps it does not matter whether the analogy the
Court drew is weak. However, even if the Court reached the right result, by
using the framework of analogy to focus primarily on whether private entities are
alike or different from government entities instead of just focusing on whether
the policies supporting qualified immunity apply to private entities, the Court‟s
decision rests on unsound footing that makes it vulnerable to attack in at least
two ways.
First, the Court‟s reasoning leaves open the opportunity for a private
defendant to flip the decision on its head and argue that the more it behaves like a
government entity, the greater entitlement it has to qualified immunity. Because
of the Court‟s focus on analogy, a private defendant in a § 1983 action could
argue that it could show that government entities do purchase liability insurance
and indemnify their employees and that as a result, the private defendants should
receive qualified immunity because they are no different than the government
employees who receive immunity. This reasoning, however, is backwards. In
essence, a private defendant would be taking reasons not to provide qualified
immunity—the fact that employers purchase insurance and provide for
indemnification—and using them as a justification to grant qualified immunity,
on the ground that those facts make private entities analogous to government
entities. In fact, this is the approach taken by the Richardson dissent, which uses
the availability of insurance and indemnification for government entities as a

77

Richardson v. McKnight, 521 U.S. 399, 420 (1997) (“surely it is the availability of [insurance],
rather than its actual presence in the case at hand, which decreases (if it does decrease, which I
doubt) the need for immunity protection.”).
78
In the welfare context, for example, “when the Arizona state legislature mandated the
privatization of its state welfare system, only one company offered a bid; the state had no selection
of alternatives. In Connecticut, Colonial Cooperative Care, Inc., was the only bidder for its
contract to determine eligibility for disability-based cash assistance.” Dru Stevenson, Privatization
of Welfare Services: Delegation by Commercial Contract, 45 ARIZ. L. REV. 87, 92 & nn.37-38
(1992). Similarly, by the late 1990s two companies accounted for more than 75 percent of the
global private-prison market. Austin, supra note 7, at 3-4. Even if the government were inclined to
switch competitors, it is not clear that doing so would be feasible. If one company is housing
hundreds or thousands of a State‟s prisoners, the costs of moving them to different facilities or
bringing in different managers could be enormous.

982

UMKC LAW REVIEW

[Vol. 78:4

reason to extend qualified immunity to private parties.79 While this risk may
seem more theoretical than real, the fact that Richardson did not hold
categorically that private employees cannot receive qualified immunity opens the
door for private defendants to make this argument.80 Indeed, some lower courts
have granted qualified immunity to some private defendants even after
Richardson.81 In a framework where analogy is the decision-making fulcrum,
this is a natural result, even if the application of the analogy operates at crosspurposes with the underlying goal that the analogy is supposed to serve.82
A second, admittedly more speculative, possibility is that the Court
recognized that its analogy was weak and as a result may have written “caveats”
into its decision that threaten to mute the force of its rule.83 At the end of its
decision, the Court explicitly left open the possibility that private defendants may
still be able to avail themselves of “good-faith” immunity in the absence of
qualified immunity, but did not resolve the question.84 Although the Supreme
Court expressed no opinion on good faith immunity, lower courts both before
and after Richardson have permitted private defendants to claim good-faith
immunity, which protects an employee from damages unless the employee acted
out of actual malice or ill-will.85
Although tacked on at the end of the Court‟s decision, this exception
threatens to substantially undercut the force of the Court‟s decision. Because
actual malice is a relatively high standard, good faith immunity likely would
apply in many of the same cases in which qualified immunity would apply.
Moreover, given that it is a subjective standard and qualified immunity is an
objective standard, it could apply to cases where qualified immunity would not

79

See Richardson, 521 U.S. at 420 (Scalia, J., dissenting) (noting that governments can obtain
insurance just as easily as private parties in arguing that private employees should receive qualified
immunity under § 1983).
80
Id. at 413 (stating that it was “answer[ing] the immunity question narrowly, in the context in
which it arose”).
81
See, e.g., Bartell v. Lohiser, 12 F. Supp. 2d 640, 645-46 (E.D. Mich. 1998) (finding that
employees of a private contractor providing foster care services could receive qualified immunity
under § 1983).
82
Of course, one could also try and make the analogy work in the opposite direction and argue that
because government entities are actually more like private entities than the Richardson Court
acknowledged, government employees, like private employees should not receive qualified
immunity. Even if that result is a good one, that analytical approach runs into the same problems.
Determining whether government employees receive qualified immunity should turn on the nature
of public employment and its application to the purposes underlying qualified immunity. It should
not depend on whether government entities resemble private entities.
83
See Richardson, 521 U.S. at 413 (closing its decision with “three caveats”).
84
Id. at 413-14 (raising, but declining to express a view on, whether private defendants might be
entitled to good-faith immunity).
85
See, e.g., Clement v. City of Glendale, 518 F.3d 1090, 1096-97 (9th Cir. 2008) (allowing a
private towing company to assert a good faith immunity defense in a § 1983 action); Pinsky v.
Duncan, 79 F.3d 306, 311-13 (2d Cir. 1996); Vector Research, Inc. v. Howard & Howard
Attorneys, P.C., 76 F.3d 692, 699 (6th Cir. 1996); Jordan v. Fox, Rothschild, O‟Brien & Frankel,
20 F.3d 1250, 1276-77 (3d Cir. 1994); Wyatt v. Cole, 994 F.2d 1113, 1118 (5th Cir. 1993).

2010]

THE FAILURE OF ANALOGY

983

apply. In some cases, an employee could violate clearly established law for
which the employee would not be entitled to qualified immunity but without the
subjective intent required to overcome the hurdle of good faith immunity. Thus,
the possibility of the Court‟s own uncertainty about the strength of its decision in
general and its analogy between private and government entities in particular
may have caused the Court to water down its opinion in a way that may
substantially blunt its force.
In short, Richardson identifies some good reasons why private entities
should not get qualified immunity, or possibly even good faith immunity. But its
misguided focus on analogizing or distinguishing government entities from
private entities ultimately may hinder efforts to limit the reach of immunity
doctrines under § 1983 and therefore may undermine the statute‟s ability to
effectively protect and vindicate constitutional rights.86
B. Correctional Services Corporation v. Malesko, 534 U.S. 61 (2001)—
Making Analogy Primary and the Underlying Policy Secondary
A second example where the Supreme Court‟s focus on analogizing private
defendants to government parties led to a flawed result is in Correctional
Services Corp. v. Malesko.87 Although Malesko was a Bivens action involving a
private party contracting with the federal government rather than a § 1983 action
involving a party contracting with a state or local government, the case may still
be instructive because in both areas, a private party must perform state action to
become subject to a claim and because the Court has applied § 1983 law to
Bivens actions and vice versa.88 In Malesko, the Court reached a result that it
candidly acknowledged would hinder the protection of constitutional rights but
held that it was compelled to do so on the basis of an analogy it drew between the
private-entity defendant in Malesko and a federal agency.89 In other words, the
proper analogy became more important than the proper outcome. By using
analogy as the framework for reaching with the result that the Court undermined
the purpose of Bivens by using analogy as the framework for reaching its
decision.
A Bivens action—named for the Supreme Court‟s decision in Bivens v. Six
Unknown Named Agents of Federal Bureau of Narcotics90—allows a plaintiff to

86

The Richardson court hinted at the idea that private entities are simply different than government
entities, and therefore should be evaluated on their own terms, acknowledging that “government
employees typically act within a different system,” than private employees and perhaps should
simply be evaluated separately from each other. 521 U.S. at 410 (emphasis in original). But the
Court does not develop the point in great detail, and it appears to be subsumed by the Court‟s more
detailed analysis of why private entities are different from governmental entities.
87
534 U.S. 61 (2001).
88
See, e.g., Hartman v. Moore, 547 U.S. 250, 254 n.2 (2006) (noting that “a Bivens action is the
federal analog to suits brought against state officials under Rev. Stat. § 1979, 42 U.S.C. § 1983.”).
89
Malesko, 534 U.S. at 71.
90
403 U.S. 388 (1971).

984

UMKC LAW REVIEW

[Vol. 78:4

sue a federal official for certain constitutional violations.91 The Bivens cause of
action, unlike a § 1983 action, does not derive from a statute but instead is an
implied right of action.92 Consequently, not every constitutional violation is
actionable under Bivens, and courts apply several factors in deciding whether to
imply a Bivens cause of action for a particular category of constitutional
violations or for a particular category of defendants.93 Moreover, although
Bivens provides a cause of action against individual federal officers, a plaintiff
cannot bring a claim against the federal government itself.94 In FDIC v. Meyer,
the Supreme Court held that the prohibition against suing the federal government
also extended to suing specific federal agencies.95
In Malesko, John Malesko, an inmate in a private halfway house run by the
Correctional Services Corporation (“CSC”), sued CSC, which had contracted
with the federal government to house certain federal prisoners, alleging that CSC
violated his Eighth Amendment rights.96 At issue in the case was whether a
plaintiff can bring a Bivens action against a private company performing a public
function as opposed to an action against the company‟s employees.97
Given the issue in the case, as well as the Bivens doctrine that provides for
causes of action against individual officers but not against the federal
government or its agencies, it is easy to see the critical role that analogy could
play in determining the outcome of the case. Were the Court decided to
analogize CSC to an individual officer, then the Court could conclude that a

91

See, e.g., Bivens, 403 U.S. at 389 (implying a cause of action against federal officials for
violations of Fourth Amendment rights).
92
See Malesko, 534 U.S. at 66 (“In [Bivens], we recognized for the first time an implied private
action for damages against federal officers alleged to have violated a citizen‟s constitutional
rights.” (citation omitted)).
93
Compare, e.g., Carlson v. Green 446 U.S. 14, 32 (1980) (allowing a Bivens action for Eighth
Amendment Violations), and Davis v. Passman, 442 U.S. 228, 248-49 (1979) (allowing a Bivens
action for Equal Protection violations under the Fifth Amendment), with Bush v. Lucas, 462 U.S.
367, 390 (1983) (refusing to find a Bivens action for a First Amendment claim).
94
See, e.g., Randall v. United States, 95 F.3d 339, 345 (4th Cir. 1996) (“Bivens did not abolish the
doctrine of sovereign immunity of the United States. Any remedy under Bivens is against federal
officials individually, not the federal government.”).
95
FDIC v. Meyer, 510 U.S. 471, 486 (1994).
96
Malesko, 534 U.S. at 64.
97
Id. at 63 (“We decide here whether the implied damages action first recognized in [Bivens],
should be extended to allow recovery against a private corporation operating a halfway house under
contract with the Bureau of Prisons.” (citation omitted)). In Malesko, Mr. Malesko filed his
original complaint pro se, naming CSC and several unnamed CSC employees as defendants. Id. at
64. After Mr. Malesko obtained counsel, counsel amended the complaint to add the names of the
individual defendants. Id. However, because the amendment adding the individual defendants‟
names did not relate back to the original complaint within the meaning of Fed. R. Civ. P. 15, the
district court dismissed the claims against the individual defendants on statute of limitations
grounds, and that portion of the district court‟s decision was affirmed by the U.S. Court of Appeals
for the Second Circuit. Id. at 65. Thus, the only question before the Supreme Court was the
availability of an action against CSC.

2010]

THE FAILURE OF ANALOGY

985

cause of action was available. Conversely, were the Court chose to analogize
CSC to a federal agency, then no cause of action would lie.98
The Court held that no cause of action was available against a private
corporation under Bivens, concluding essentially that private corporations were
akin to federal agencies and therefore because no cause of action was available
against a federal agency, no cause of action was available against CSC either.99
The Court started out by describing the primary purpose of Bivens actions to be
to deter constitutional violations by individual federal officers.100 The Court then
cited its decision in Meyer that suits against a federal agency would purportedly
undermine deterrence of individual officers because plaintiffs would choose to
sue federal agencies rather than individual officers.101 The Court then held that a
claim against a private corporation “is, in every meaningful sense, the same” as a
claim against a federal agency.102 It concluded by saying that as with a claim
against a federal agency, suing corporate defendants would somehow undermine
deterrence of constitutional violations by individual employees because the
employees themselves would never face suit.103 The Court made it sound as if its
conclusion were compelled by Meyer, stating that “[o]n the logic of Meyer,
inferring a constitutional tort remedy against a private entity . . . is therefore
foreclosed.”104
Significantly, the Court acknowledged that allowing private corporations to
be sued under Bivens would likely deter constitutional misconduct but
nonetheless refused to find a cause of action on the ground that it would
undermine the analogy to Meyer. Mr. Malesko argued to the Court that because
CSC was a private company, making CSC responsible for its constitutional
violations would be “the best way to discourage future harms.”105 The Court
candidly conceded “[t]hat may be so,” but quickly dismissed that argument on
the ground that the reasoning was inconsistent with Meyer‟s conclusion that
federal agencies were not amenable to suit.106

98

Alternatively, the Court could have decided not to draw an analogy at all, and simply evaluate
private corporations in light of the polices and purposes of Bivens. In other words, the Court could
have decided that private corporations are different from both individual officers and from agencies
such that neither category should provide the dispositive rule for the case. That approach may have
been the most sensible approach, but it was not one followed by either the majority or the dissent.
99
See Malesko, 534 U.S. at 70-71.
100
See id. at 70 (“The purpose of Bivens is to deter individual federal officers from committing
constitutional violations.”).
101
Id. at 70-71 (“If we were to imply a damages action directly against federal agencies . . . there
would be no reason for aggrieved parties to bring damages actions against individual officers. The
deterrent effects of the Bivens remedy would be lost.” (quoting FDIC v. Meyer, 510 U.S. 471, 485
(1994))).
102
Id. at 71.
103
Id.
104
Id.
105
Id.
106
Id.

986

UMKC LAW REVIEW

[Vol. 78:4

The Court‟s decision is rather remarkable and demonstrates the primacy of
analogy over purpose. The Court refused to provide a cause of action against
private entities, even while acknowledging that doing so would lead to more
constitutional violations, because the Court felt that result was necessary in order
to maintain its analogy between private entities and federal agencies. In other
words, having decided that a private entity was analogous to the federal agency
in Meyer, the analogy dictated the outcome of the case regardless of how whether
its decision furthered the purposes of Bivens. Instead of focusing on the policies
underlying Bivens, applying those policies to private entities, and then using the
result of that analysis to determine whether private entities should be considered
similar to or different from federal agencies, the Court did the reverse, deciding
which analogy to draw and then used the analogy to control its decision.
To be fair, the Malesko Court did attempt to tie its analogy to the policies of
Bivens by asserting that the purpose of Bivens is to deter individual actors rather
than institutional actors.107 But, particularly in the private context, the notion that
suing a private entity would deter violations by the entity but would fail to deter
violations—or possibly lead to increased violations—by the entity‟s employees
makes little sense. A private entity can act only through its employees or
agents.108 Therefore, it would be seemingly impossible to reduce violations by a
private-entity without at the same time reducing the violations of its employees
and agents. In the private context, to reduce the violations of the entity is to
reduce the violation of its employees.109
The Court‟s attempt to create a line of demarcation between individual
government defendants and government entity defendants and then to label
private entities as either one or the other reveals one danger of analogizing
private entities to government actors. While such a separation may be necessary
in the government context because of sovereign immunity,110 it may be much
more difficult—and much less advisable—to try and disentangle a private entity
from its employees. On the one hand, it can act only through its employees, and

107

Id. Even that part of the analogy is subject to criticism. In Meyer, the Court concluded that the
reason why allowing a cause of action against a federal agency would cause plaintiffs to sue the
agency only and not the officers is that the officers could claim qualified immunity. See Meyer,
510 U.S. at 485 (“If we were to imply a damages action directly against federal agencies, thereby
permitting claimants to bypass qualified immunity, there would be no reason for aggrieved parties
to bring damages actions against individual officers.”). That rationale does not apply as strongly to
a cause of action against private entities, because under Richardson, private employees generally
do not receive qualified immunity, though, as explained above, they may be able to claim good
faith immunity. Richardson v. McKnight, 521 U.S. 399, 413-14 (1997).
108
For example, the Malesko complaint alleged that several individual correctional officers violated
Mr. Malesko‟s constitutional rights and that CSC, as their employer, was liable for their
misconduct. Corr. Servs. Corp. v. Malesko, 534 U.S. 61, 64-65 (2001).
109
See also id. at 80-81 (Stevens, J., dissenting) (“It cannot be seriously maintained, however, that
tort remedies against corporate employers have less deterrent value than actions against their
employees.”).
110
While sovereign immunity may be a reason to have separate rules for entities as opposed to
individuals, the Court in Meyer refused to permit a cause of action against the Federal Deposit
Insurance Corporation even though it had waived its sovereign immunity. Meyer, 510 U.S. at 483.

2010]

THE FAILURE OF ANALOGY

987

on the other hand, the corporate form exists in order to create some corporate
identity that is distinct from its employees and officers. Thus, private entities
cannot be easily categorized as being like employees or as being like government
entities, and the Court‟s attempt to try and analogize private actors to government
actors therefore can lead to misguided results.111
Moreover, the basic tort and agency principle of respondeat superior—that
an employer is liable for the torts of its employees committed within the scope of
employment112—is based in part on the idea that placing liability on an employer
will help reduce the misconduct committed by its employees, both because the
employees may be judgment-proof and therefore insufficiently deterred by the
threat of damages liability,113 and because placing liability on the employer can
induce the employer to engage in better employee training and hiring practices,
as well as engaging in other measures to encourage employees to safeguard
constitutional rights.114
Further, the Court‟s view that placing liability on the entity will undermine
deterrence of individual misconduct appears inconsistent with its prior decisions.
The Richardson Court recognized that companies have incentives to reduce
employee misconduct in order to minimize the risk that a government contractor
will replace that company with a competitor whose employees better safeguard
constitutional rights.115 Similarly, in Owen v. City of Independence,116 in which

111

The Malesko dissenters appeared to recognize this difficulty of saying that entities are somehow
categorically different from the employees who act on their behalf. The dissent focused on the law
of agency and distinguished the federal government from federal agents. The dissent argued that
while a plaintiff cannot bring a Bivens claim against the federal government a plaintiff should be
able to bring a Bivens action against an agent of the federal government, whether that agent was in
human form or corporate form. See Malesko, 534 U.S. at 75-83 (Stevens, J., dissenting) (arguing
that “corporate agents performing federal functions, like human agents doing so,” should be suable
under Bivens). The risk of such an approach, however, is that it lends itself to equating private
entities with individual government employees. While such a result may work well in the context
of who may be sued under Bivens, it may be problematic in other contexts by suggesting that both
private entities and private employees should be treated just like government employees. In
particular, such an argument could be used to assert, as the dissenters did in Richardson, that
private actors sued under § 1983 should receive qualified immunity. Richardson, 521 U.S. at 41423.
112
See DAN B. DOBBS, THE LAW OF TORTS 905 (2000).
113
See, e.g., Alan O. Sykes, The Economics of Vicarious Liability, 93 YALE L.J. 1231, 1241-52
(1984) (describing how respondeat superior might help prevent employers from making side
payments to judgment proof employees rather than take measures to reduce liability risk).
114
See, e.g., DOBBS, supra note 112, § 334 at 908 (explaining that respondeat superior liability may
encourage employers to take measures to reduce the risk that their employees will violate the law);
STEVEN SHAVELL, ECONOMIC ANALYSIS OF ACCIDENT LAW 173 (1987); Sykes, supra note 113, at
1251 (“Alternatively, vicarious liability may lead to the employment of more financially
responsible agents with an attendant increase in loss-avoidance effort.”).
115
See Richardson, 521 U.S. at 409 (noting that “[c]ompetitive pressures mean not only that a firm
whose guards are too aggressive will face damages that raise costs, thereby threatening its
replacement”).
116
445 U.S. 622 (1980).

988

UMKC LAW REVIEW

[Vol. 78:4

the Court held that municipalities (as opposed to municipal employees) are not
entitled to qualified immunity for good faith conduct, the Court explicitly held
that placing liability on the municipality would deter violations by employees. 117
The Court stated that “[t]he knowledge that a municipality will be liable for all of
its injurious conduct, whether committed in good faith or not, should create an
incentive for officials who may harbor doubts about the lawfulness of their
intended actions to err on the side of protecting citizens‟ constitutional rights.”118
Thus, by elevating the framework of analogy above the purposes of Bivens,
the Court reached a result that will undermine deterrence of constitutional
violations and that is arguably inconsistent with its own prior precedent.
C. Ashcroft v. Iqbal, 129 S. Ct. 1937 (2009)—Extending Analogy Beyond Its
Moorings
A final case that highlights the dangers of using analogy as an interpretive
tool in the § 1983 and Bivens contexts is the Supreme Court‟s decision last term
in Ashcroft v. Iqbal.119 Iqbal is also a Bivens case, and unlike the first two
examples, it involves government defendants rather than private defendants.
Nonetheless, Iqbal may be a useful example because it uses analogy to apply a
liability rule involving entity defendants—in this case municipalities—to
individual defendants, and in doing so reaches a result that extends far beyond
what the analogy itself would support.
Although Iqbal has received significant attention for its potential effect on
notice pleading, it also may have significant implications for the doctrine of
supervisory employee liability under § 1983 and Bivens. The plaintiff, Javaid
Iqbal, alleged that he was arrested by the federal government after the September
11th attacks and labeled a “high interest” detainee.120 He alleged that federal
officers exposed him to unconstitutionally severe conditions of confinement on
account of his race, religion or national origin in violation of the First
Amendment and the Fifth Amendment.121 Mr. Iqbal sued not only the officers
who were directly responsible for his mistreatment, but he also sued thenAttorney General John Ashcroft and then-Director of the Federal Bureau of
Investigation (“FBI”) Robert Mueller under a theory of supervisory liability.

117

Id. at 651-52; see also id. at 652 (“Furthermore, the threat that damages might be levied against
the city may encourage those in a policymaking position to institute internal rules and programs
designed to minimize the likelihood of unintentional infringements on constitutional rights.”
(footnote omitted)). The Court even suggested that entity liability could deter more effectively than
individual liability by reaching instances of “„systemic‟ injuries that result not so much from the
conduct of any single individual but form the interactive behavior of several government
officials.”).
118
Id. at 652.
119
129 S. Ct. 1937 (2009).
120
Id. at 1943.
121
Id. at 1943-44. Iqbal alleged, among other things, that the defendants beat him, subjected him to
serial strip searches and body cavity searches without proper justification, refused to let him pray,
and kept him and other detainees in lockdown for twenty-three hours a day. Id.

2010]

THE FAILURE OF ANALOGY

989

Specifically, Mr. Iqbal alleged that Ashcroft and Mueller had constructive
knowledge of the officers‟ behavior and failed to stop it.122 Prior to Iqbal, lower
courts had recognized that supervisory employees could be liable for damages
under § 1983 and Bivens where the supervisor was grossly negligent or
deliberately indifferent to the risk of constitutional violations.123
The Court rejected Mr. Iqbal‟s supervisory liability claim and cast doubt on
whether supervisory liability is a valid doctrine at all under either § 1983 or
Bivens.124 The Court went a step further than the parties, holding that not only
was a high-level official‟s constructive knowledge insufficient to support
supervisory liability, but also that the official‟s actual knowledge would not
support supervisory liability.125 The Court once again used the framework of
analogy as a central ground for its decision. The Court began by noting that
Bivens is the federal analog to § 1983 and that under § 1983, under the Court‟s
decision in Monell v. Department of Social Services of the City of New York,126
municipal defendants under § 1983 could not be held liable under a respondeat
superior theory.127 Although the Court also cited some very old pre-Bivens
decisions regarding the liability of federal officers, those decisions did not arise
in a Bivens context and were ambiguous with respect to supervisory liability, and
therefore did not appear to provide the primary support for the Court‟s
conclusion.128
The Court then took Monell‟s rule of no respondeat superior liability for
municipalities and extended it to limit the supervisory liability of individual
officers, essentially equating the concept of vicarious liability with supervisory
liability. The Court stated that in the § 1983 and Bivens contexts, “the term
„supervisory liability‟ is a misnomer and that supervisory liability “is

122

Id. at 1944.
See, e.g., Poe v. Leonard, 282 F.3d 123, 140 (2d Cir. 2002) (describing supervisory liability
standards); see also Iqbal, 129 S. Ct. at 1958 (Souter, J., dissenting) (describing various lower court
tests for supervisory liability and distinguishing them from vicarious liability).
124
Iqbal, 129 S. Ct. at 1948-49.
125
Id. at 1949 (rejecting the argument that a supervisor‟s “mere knowledge of his subordinate‟s
discriminatory purpose amounts to the supervisor‟s violating the Constitution”).
126
436 U.S. 658 (1978).
127
Iqbal, 129 S. Ct. at 1948 (citing Monell, 436 U.S. at 691, for the proposition that there is “no
vicarious liability for a municipal „person‟ under 42 U.S.C. § 1983”).
128
Specifically, the Court cited Dunlop v. Munroe, 11 U.S. 242, 269 (1812), a decision that is
nearly 200 years old, for the proposition that a federal officer‟s liability “will only result from his
own neglect” in failing to supervise the officer‟s subordinates. But if an officer‟s negligence in
supervising employees is sufficient to support liability, then surely an officer‟s gross negligence or
deliberate indifference—the standard typically required for supervisory liability—would also
support liability. Similarly, the Court cited Robertson v. Sichel, 127 U.S. 507, 515-16 (1888) for
the proposition that an officer generally is not liable for the acts of his or her subordinates. This
also is not inconsistent with supervisory liability, which establishes that while a supervisory
ordinarily is not automatically liable for the actions of his or her subordinates, the supervisor may
be liable when possessed with actual or constructive knowledge of the subordinates‟
unconstitutional conduct.
123

990

UMKC LAW REVIEW

[Vol. 78:4

inconsistent” with the principle that high-level officials “may not be held
accountable for the misdeeds of their agents.”129
The result of Iqbal is that the scope of supervisory liability is unclear. One
reading is that supervisory liability still exists, but that the supervisory official
must have the same state of mind as the offending subordinate. In Iqbal, because
the specific violations alleged, discrimination in violation of the First and Fifth
Amendments, requires discriminatory intent, the mere allegation of a
supervisor‟s knowledge was insufficient to support supervisory liability in the
absence of allegations that the supervisors also acted with discriminatory
purpose.130 Another reading is that the Court rejected the theory of supervisory
liability altogether. In fact, that is the reading taken by the Iqbal dissent.131
Under either reading, however, the Court‟s decision to analogize the
municipal defendant in Monell to the individual high-level officials in Iqbal and
to use Monell‟s holding about vicarious liability as a basis for significantly
curtailing—or perhaps eliminating—supervisory liability extends Monell‟s
holding far beyond its moorings. As the dissent pointed out, there is a significant
difference between the traditional form of respondeat superior liability rejected in
Monell and the supervisory liability rejected in Iqbal.132 Respondeat superior
liability is a form of strict liability in which an employer is held liable for its
employee‟s torts regardless of fault, and even if the employer took preventive
measures to reduce the risk of misconduct.133 Supervisory liability applies only
when the supervisory employee is culpable in some way, through either gross
negligence, deliberate indifference, or where the supervisor ratified the unlawful
behavior.134
Moreover, Monell never held that municipalities could never be held liable
for its employees‟ unconstitutional conduct. Rather, the Court held that a
municipality could be held liable where a municipal policy or custom caused the
violation.135 The standard for whether municipal officials established an
unconstitutional policy or custom is an objective one, and unlike in Iqbal, it does
not depend on the state of mind required for the underlying constitutional
violation.136

129

Iqbal, 129 S. Ct. at 1949.
See id. (“[P]urpose rather than knowledge is required to impose Bivens liability on the
subordinate for unconstitutional discrimination; the same holds true for an official charged with
violations arising from his or her superintendent responsibilities.”).
131
Id. at 1957 (Souter, J., dissenting) (“Lest there be any mistake, in these words the majority is not
narrowing the scope of supervisory liability; it is eliminating Bivens supervisory liability
entirely.”).
132
Id. at 1958.
133
See, e.g., Konradi v. United States, 919 F.2d 1207, 1210 (7th Cir. 1990) (identifying respondeat
superior liability as a form of strict liability that applies regardless of the employer‟s fault or lack of
fault).
134
See supra note 123
135
See Monell v. Dep‟t of Soc. Servs., 436 U.S. 658, 694 (1978).
136
In City of Canton v. Harris, 489 U.S. 378, 388-89 & n.8 (1989), the Supreme Court held that a
municipality could be liable under § 1983 for failing to adequately train its employees if the
130

2010]

THE FAILURE OF ANALOGY

991

Additionally, the Court‟s reliance on Monell underscores the dangers of
analogizing entities with individuals. Monell‟s reasoning rested on concerns
relating specifically to municipalities and its reasoning has little applicability to
individual defendants. The Monell Court relied heavily on the legislative history
of § 1983 (which itself may be reason to question Monell‟s applicability to
Bivens actions), and specifically on both constitutional concerns relating to
imposing vicarious damages liability on municipalities as well as on the rejection
of an Amendment that dealt specifically with expanding municipal liability and
which did not address individual liability.137 Indeed, the Court acknowledged
strong policy reasons for imposing respondeat superior liability, but found that
those policy rationales could not overcome the legislative history relating
specifically to municipal liability.138 The Court‟s decision therefore related only
to the scope of damages liability for municipal actors and did not address the
contours of liability for individual defendants, supervisory or otherwise.
Notwithstanding Monell‟s limitations on municipal liability, lower courts
developed a separate and distinct doctrine of supervisory liability under § 1983
that permitted liability where a supervisor knew of the risk of unconstitutional
conduct and failed to try to prevent it.139 Thus, while the Iqbal Court used
Monell to suggest that the decision was inconsistent with theories of supervisory
liability, in fact Monell‟s rules on municipal liability co-existed peacefully with
theories of supervisory liability. Because the rules regarding liability of
municipal entities differ from those for supervisory employees, the Court‟s
analogy to Monell provides a weak basis for evaluating supervisory liability.
Again, the Court‟s reasoning elevates analogy over substance. The Court
took Monell as a starting point and concluded that based on Monell, the scope of
supervisory liability necessarily had to be limited, regardless of the underlying
purposes of § 1983 and Bivens and regardless of the wisdom of the analogy to
Monell in the first place. The Court‟s reliance on a flawed analogy limited the
scope of § 1983 and Bivens, and reduced the ability of victims of constitutional
misconduct to vindicate their rights.

municipality‟s failure to train evidences deliberate indifference to the risk of a constitutional
violation. The Court specifically stated that “[t]he deliberate indifference standard we adopt for §
1983 „failure to train‟ claims does not turn upon the degree of fault (if any) that a plaintiff must
show to make out an underlying claim of a constitutional violation.” Id. at 388 n.8. Similarly, in
Baker v. District of Columbia, 326 F.3d 1302 (D.C. Cir. 2003), which involved an Eighth
Amendment claim by a prisoner for a denial of medical care, the Court held that while the prisoner
had to show subjective deliberate indifference to make out the underlying constitutional violation,
the standard for a municipal custom or policy was an objective one that could be satisfied by
showing that municipal officers knew or should have known about the risk of unconstitutional
behavior and failed to take action to prevent it. Id. at 1306-07.
137
Monell, 436 U.S. at 691-94.
138
Id. at 693-94 (noting that respondeat superior may help spread the costs of accidents across the
community and may reduce total accident costs but holding that these justifications were
“insufficient” to overcome Congress‟s constitutional objections to imposing vicarious liability on
municipalities).
139
See Ashcroft v. Iqbal, 129 S. Ct. 1937, 1958 (2009); Poe v. Leonard, 282 F.3d 123, 140 (2d Cir.
2002).

992

UMKC LAW REVIEW

[Vol. 78:4

***
All three decisions suggest that the Court‟s use of analogy to determine the
liability of private entities in the § 1983 and Bivens contexts can backfire and that
the form of analogy can take center stage at the expense of the underlying
purposes of § 1983 and Bivens. This is not to say that analogy can never be
useful. Analogy often is a valuable analytical tool, and there may be cases,
including cases arising in the Bivens and § 1983 contexts in which using the
framework of analogy can lead to sound and defensible outcomes.140 Rather, my
point is that in the context of private-entity liability, analogy is often misused in
ways that serve to overly constrain the reach of § 1983 and Bivens and that exert
substantial costs on individuals who are at risk of suffering constitutional
deprivations. For that reason, analogy may not be an effective analytical
framework with respect to questions relating to private entity liability for
constitutional violations.
V. AN ALTERNATIVE TO ANALOGY
Asserting that courts should shy away from applying an analogy framework
tells them how not to determine private-entity liability, but it does not give much
guidance about what should replace analogy as a framework for assessing
private-entity liability. The question of what is the right framework is harder to
answer than asserting what is the wrong framework, and this article offers one
possible alternative. As an initial matter, in considering how to go about
addressing private-entity liability questions, courts should start by recognizing
that private-entities are a distinct category that is different from the category of
government defendants. They may have some similarities to government
employees, but also some differences, and the same is true with respect to
supervisory employees or government entities. The first step in assessing
private-entity liability is acknowledging that private entities should be evaluated
on their own terms and not on the basis of how alike or different they are from
various government actors.
Viewing private entities as their own category suggests that one possible
way to address private-entity liability is to utilize and apply the already welldeveloped and fulsome body of law that exists to determine the liability of
private parties for their injurious acts—private tort law—to the extent that it is
consistent with § 1983‟s broad remedial purposes.141 Private tort law has already

140

For example, the Court has recognized that grand jurors, notwithstanding their status as private
individuals, are entitled the same level of immunity as other individuals—such as judges and
prosecutors—that exercise discretionary functions in carrying out the judicial function. See, e.g.,
Imbler v. Pachtman, 424 U.S. 409, 423 n.20 (1976).
141
The major purposes of § 1983 are to compensate victims of constitutional injury as well as to
deter future violations. See Owen v. City of Independence, 445 U.S. 622, 651 (1980). Many of the
members of Congress debating § 1983 explicitly stated that the statute was intended to be read
broadly in light of its remedial purpose. See, e.g., CONG. GLOBE, 42d Cong., 1st Sess. App. 68
(1871) (statement of Rep. Shellabarger) (stating § 1983 should be “liberally and beneficently

2010]

THE FAILURE OF ANALOGY

993

established rules for when private parties can claim immunity from damages,
when private-entities are vicariously liable for the torts of their employees, when
an employee‟s negligent supervision of subordinates gives rise to liability, and
when private-entities can be liable for punitive damages among other things.
Applying these rules to private-entities would avoid the difficulties and
inconsistencies created by the various special protections that Courts have
erected for § 1983 and Bivens defendants such as qualified immunity, protection
from respondeat superior liability, a particularly onerous standard of supervisory
liability, and limitations on implied rights of action in the Bivens context. Given
both that this body of private tort law already is well-developed and that it is
routinely applied to private parties in tort contexts outside of § 1983 and Bivens,
there is no reason not to apply it to private § 1983 defendants as well. Moreover,
doing so would be consistent with the Court‟s view that the framers of § 1983
were familiar with the common law of torts and that “they likely intended these
common law principles to obtain, absent specific provisions to the contrary.”142
Because tort law provides the governing common-law rules for private-party
liability, there may be good reason to apply tort law principles to private § 1983
defendants as well.
Additionally, that private entities or employees should receive special
protections from liability—such as qualified immunity or an exemption from
respondeat superior liability or supervisory liability—simply because a plaintiff
sues the private party under § 1983 rather than under state tort law would appear
to make little sense. It is not at all clear that a private party‟s incentives or
behavior is any different when performing state action than when engaging in
private behavior. For example, it is unlikely that a private debt collector would
be more timid or fearful of lawsuits so as to justify the need for qualified
immunity because it happens to be collecting a government debt rather than a
private debt, or that imposing respondeat superior liability on the debt collector
would lead to less deterrence when collecting government debts than when
collecting private debts.143
Similarly, some misconduct by private parties acting under color of state
law will give rise to both § 1983 claims and state tort claims.144 For example, a §
construed” because it is “remedial and in aid of the preservation of human liberty and human
rights”); id. at 217 (statement of Sen. Thurman) (stating that § 1983‟s language is without limit and
“as broad as can be used”).
142
City of Newport v. Fact Concerts, Inc., 453 U.S. 247, 258 (1981).
143
Debt collectors may be considered state actors for purposes of § 1983 when operating on the
government‟s behalf. In Del Campo v. Kennedy, 517 F.3d 1070 (9th Cir. 2008), a plaintiff sued a
private company running a “bad check diversion program” for the State of California under § 1983
as well as under other federal and state laws. The debt collector claimed that it was entitled to state
sovereign immunity because it was working on behalf of the State. Id. at 1072. The court rejected
the sovereign immunity argument, but it did indicate that the defendant could be considered a state
actor for purposes of § 1983. See id. at 1081 n.16.
144
While many § 1983 claims may have a state court analog, not all do. Certain constitutional
rights, such as free speech and free exercise rights, privacy and reproductive rights, procedural due
process rights including the notice and opportunity to be heard, substantive due process rights,
voting rights protections, Fourteenth Amendment equal protection guarantees, and Fourth
Amendment protections against warrantless searches may not receive equivalent protections under

994

UMKC LAW REVIEW

[Vol. 78:4

1983 claim under the Eighth Amendment against a prison health care provider
for failing to provide constitutionally adequate medical treatment also would
support a state medical malpractice claim.145 Given that the same conduct is at
issue in both the state and federal claim, there is no reason that the private
defendant would need particular protections from liability that do not exist under
traditional tort law. If imposing respondeat superior liability would promote
compensation objectives and help induce the private defendant to take actions to
reduce the risk of future torts, then the same would likely be true in the § 1983
context. Merely changing the legal label of the claim does not change the
incentives facing the private actor. Yet currently, courts give private actors extra
liability protections under § 1983 that they do not receive under ordinary tort law
by focusing on comparing private entities to government actors rather than
looking at whether liability would further the goals that the tort system is
supposed to serve.
The idea of using tort law and tort law rhetoric to determine the scope of §
1983 in general is not a new one, and it is one that has received scholarly
criticism. Professor Sheldon Nahmod has argued that the Supreme Court has
shifted away from using constitutional rhetoric in § 1983 actions in favor of
using tort rhetoric and that the effect of using tort rhetoric has been “to
marginalize § 1983 and to make it less protective of fourteenth amendment
rights.”146 Professor Nahmod identifies several examples of how tort rhetoric has
restricted the reach of § 1983. Specifically, he notes that tort rhetoric has been
used to hold that § 1983 plaintiffs are not entitled to presumed damages for
violations of constitutional rights but can only receive actual compensatory
damages.147 He also demonstrates how tort rhetoric have been used to limit
procedural due process protections by noting that the Court created an elevated
standard for showing a violation of a protected liberty interest in order to prevent
ordinary torts from becoming constitutional violations.148
I agree with many of Professor Nahmod‟s warnings about how tort rhetoric
can undermine rather than protect § 1983. In the private-entity context, however,
it appears that it is the constitutional rhetoric, or state-action rhetoric that limits
the reach of § 1983 and Bivens. Private tort defendants generally cannot claim

state tort law. For a more detailed discussion on why state tort remedies are an inadequate substitute
for § 1983 remedies, see Frankel, supra note 17, at 1510-13.
145
In Estelle v. Gamble, the Supreme Court held that prisoners have a constitutional right to a
certain level of medical care and that a state actor violates a prisoner‟s Eighth Amendment right
against cruel and unusual punishment when the state actor behaves with “deliberate indifference to
[the] serious medical needs” of the prisoner. 429 U.S. 97, 104 (1976).
146
Sheldon Nahmod, Section 1983 Discourse: The Move from Constitution to Tort, 77 GEO L.J.
1719, 1720 (1989).
147
Accord Memphis Comm. Sch. v. Stachura, 477 U.S. 299, 303 (1986) (holding that a plaintiff
cannot recover under § 1983 for the abstract violation of a constitutional right but only to
compensate for actual injuries suffered); see Nahmond, supra note 145, at 1727 (citing Carey v.
Piphus, 435 U.S. 247 (1978)); see also Beermann, supra note 52, at 70-71 (describing how the
Court applied common-law tort principles to limit the ability of plaintiffs to recover noncompensatory damages).
148
See Nahmod, supra note 146, at 1728-31.

2010]

THE FAILURE OF ANALOGY

995

qualified immunity, or allege that they cannot be held liable under respondeat
superior principles, or that there is no implied right of action against a privateentity at all. Rather, those protections have arisen specifically in the context of
constitutional torts and private entities have attempted to avail themselves of
those protections on the ground that they are defenses specific to constitutional
tort claims. Using this constitutional rhetoric also leads to the incongruous result
that private-entities have less damages exposure for constitutional torts—which
arguably are among some of the most egregious forms of torts—than they do for
ordinary torts.149 Consequently, in the context of private-entity § 1983 liability,
applying tort principles may actually further, rather than retrench, § 1983‟s
purposes.
Using tort principles as a guide certainly is no panacea. Just as courts can
use analogies as a veneer for naked policy decisions, they can do the same with
tort principles.150 Additionally, many states may adopt tort rules that do not
advance § 1983‟s purposes of deterrence and compensation. For instance, many
states have notice provisions that bar plaintiffs from bringing lawsuits against
government actors unless they first give notice within a specified time of the
injury.151 Or, state tort reform efforts may lead to damages caps or other
limitations on liability or recovery.152 Perhaps tort principles could be applied to
private entities to the extent consistent with § 1983‟s purposes of deterrence and
compensation. Just as § 1983 preempts state notice provisions,153 special state
tort rules that deviate from general common law principles and that are seen as
inconsistent with § 1983 should not necessarily govern private-entity § 1983
liability. Alternatively, it may be that applying tort law will lead to some
limitations on private-entity liability as well as some expansions. On the whole,

149

See Taylor v. Plousis, 101 F. Supp. 2d 255, 263-64 n.4 (D.N.J. 2000) (“It seems odd that the
more serious conduct necessary to prove a constitutional violation would not impose corporate
liability when a lesser misconduct under state law would impose corporate liability.”); see also
Smith v. Wade, 461 U.S. 30, 48-49 (1983) (“As a general matter, we discern no reason why a
person whose federally guaranteed rights have been violated should be granted a more restrictive
remedy than a person asserting an ordinary tort cause of action.”).
150
On the other hand, it is possible that given how well-developed private tort law is relative to the
more-undeveloped area of private-entity § 1983 liability, it may be that there will be less wiggle
room under private tort law to impose its policy preferences than there is under the framework of
analogy.
151
See, e.g., D.C. CODE § 12-309 (2009) (requiring plaintiffs suing the District of Columbia to give
notice to the Mayor within six months of the injury). Most of these statutes apply specifically to
suits against the government and therefore likely would not apply to private-entities, even privateentities performing state action.
152
See, e.g., Michael P. Allen, A Survey and Some Commentary on Federal “Tort Reform”, 39
AKRON L. REV. 909, 913 n.11 (2006) (indentifying various state tort reform statutes). For a sample
of the large number of enacted and proposed state-level tort reform statutes, see National
Association of Mutual Insurance Companies, Tort Reform: An Overview of State Legislative Efforts
to Improve the Civil Justice System, available at http://www.namic.org/reports/tortreform.
153
See Felder v. Casey, 487 U.S. 131, 134 (1988) (holding that § 1983 preempted a Wisconsin
statute requiring a plaintiff to give notice to the government within 120 days of injury as a
precondition of suit).

996

UMKC LAW REVIEW

[Vol. 78:4

however, I believe that while applying tort principles may not lead to perfectly
optimal results, it will lead to better results than the analogy framework that
many courts currently use.
Nor is tort law the only alternative to utilizing an analogy framework.
Another option is to use a historical model. Professor Barbara Kritchevsky, for
example, suggests that private-entity liability should focus on “private
corporations‟ historical liability” in conjunction with “the policies underlying §
1983.”154 Although that option is a viable one, there are two risks with taking a
historical approach. One is that determining the liability rules applicable to
private parties in 1871 when § 1983 was enacted often can be difficult or
impossible to determine. Majorities and dissents often have disagreed about
what history shows,155 and there may not always be a clear answer.156 Moreover,
in looking to history, the Court has been inconsistent about whether it wants to
apply legal rules as they existed in 1871 or as they exist today.157 Consequently,
a historical approach may be open to the same opportunities for manipulation as
are available when courts use analogies. On the whole, at least with respect to
private-entity liability, tort law principles may provide a framework that both
reduces the risk of manipulation and that can generally apply principles that
appear consistent with the deterrence and compensation objectives of § 1983.

154

Kritchevsky, supra note 4, at 39.
In Smith v. Wade, 461 U.S. 30 (1983), for example, there was significant disagreement about
whether historical evidence for requiring intentional misconduct supported an award of punitive
damages under § 1983. Compare 461 U.S. at 39-45, with id. at 78 n.12 (Rehnquist, J., dissenting).
Similarly, in Richardson v. McKnight, the majority asserted that historically, private prison
operators did not receive special immunities under tort law while the dissent cited opposing
historical evidence. Compare 521 U.S. at 404-07 (arguing that historically there was no immunity),
with id. at 417 & n.2 (suggesting that some historical evidence existed supporting a grant of
immunity). See also Oklahoma City v. Tuttle, 471 U.S. 808, 818 n.5 (1985) (identifying the
disagreement between the majority and the dissent over whether municipalities historically were
subjected to respondeat superior liability). The evolution in the Court‟s thinking about § 1983
municipal liability also reveals the malleability of history-based arguments. In its 1961 decision in
Monroe v. Pape, the Court interpreted the legislative history of § 1983 to conclude that
municipalities were not “persons” within the meaning of § 1983. 365 U.S. 167, 187-92 (1961).
Just seventeen years later, the Court reversed course, and on the basis of that same legislative
history, overruled its decision in Monroe and concluded that municipalities are “persons” under §
1983. See Monell v. Dep‟t of Soc. Servs., 436 U.S. 658, 664-89 (1978).
156
See, e.g., Beermann, supra note 52, at 75-76 (“The consensus in the commentary has been that
the history of § 1983 is too indefinite to give real guidance on most questions, and that the Court
should admit that it is making policy and not engaged in statutory construction.”). Professor
Beermann cites Justice O‟Connor‟s dissenting opinion in Smith v. Wade, which concerned rules for
punitive damages in § 1983 actions, where Justice O‟Connor noted that both the majority and
Justice Rehnquist in dissent had marshaled much historical evidence for each of their respective
sides which caused her to conclude that that the evidence as a whole was unilluminating and that
“[t]he battle of the string citations can have no winner.” 461 U.S. 30, 93 (O‟Connor, J., dissenting).
157
See Beermann, supra note 52, at 65 & nn.89-90.
155

2010]

THE FAILURE OF ANALOGY

997

VI. CONCLUSION
Because § 1983 most often is used to sue government defendants, courts are
still grappling with how to develop liability rules and doctrines applicable to
private parties that act under color of state law and therefore fall within the ambit
of § 1983. One of the more common approaches that courts use to determine
private-entity liability is that of analogy: if the court thinks that the private-entity
sufficiently resembles a particular government actor, then the court tends to give
the private-entity the same liability protections that the government actor
receives. Conversely, if the entity is perceived as different from the government
actor, then it does not receive the same liability protections. While analogy often
can be a useful tool for extending legal rules to new situations, in the context of
private-entity § 1983 liability, the use of analogy may backfire with the result
that it limits the scope of § 1983 and undermines its goal of safeguarding
constitutional rights. In relying on analogy, the courts often have made the
analogy itself the focal point of its decision-making, often at the expense of the
policies that the analogy is supposed to serve. Whether by using an analogy that
is based on an incorrect factual comparison that leaves a decision open to attack,
to letting the analogy dictate the outcome of the decision, to extending analogies
beyond what their reasoning supports, the use or misuse of analogy supports the
scholarly criticism that § 1983 doctrine is incoherent and simply a reflection of
the Court‟s own policy goals. Instead of relying on analogy, courts should
consider applying the body of law that already exists to govern injurious acts by
private parties, which is private tort law. While tort law may not provide a
perfect solution, it will avoid many of the specific liability barriers that have been
established under § 1983 and Bivens and may better advance the goal of
protecting constitutional rights.