“State action” is a general term used to describe the kinds of actions specifically prohibited by Section 1 of the Fourteenth Amendment, and by extension, the kinds of state or federal actions prohibited by other provisions of the Constitution. Since each prohibition in Section 1 is addressed directly to states, the generalization of “state action” makes logical sense, but it leaves unanswered two questions of interpretation: First, are the directives in Section 1 aimed only at states and those acting under state authority? Second, is Congress’s power under Section 5 to enforce Section 1 limited to enacting laws aimed only at states and those acting under state authority?

The words of the provision identify three kinds of prohibited conduct: (1) a state’s making or enforcing any law “which shall abridge the privileges or immunities of citizens of the United States”, (2) a state’s depriving “any person of life, liberty, or property, without due process of law”, and (3) a state’s denying “to any person within its jurisdiction the equal protection of the laws.” Although each prohibition seems to forbid a different kind of conduct, all three seem aimed at state action or conduct attributable to a state.

Soon after Congress had overturned President Andrew Johnson’s veto of the Civil Rights Act of 1866, it passed the Fourteenth Amendment and sent it to the states for approval. This was to counter President Andrew Johnson’s claim when he vetoed the Civil Rights Act that it was beyond Congress’s constitutional powers. The drafters of Sections 1 and 5 of the Fourteenth Amendment understood Section 5 as providing an unquestionable constitutional base for the 1866 Civil Rights Act, and Section 1 as embedding the substantive proscriptions of that act in the Constitution itself, safe from subsequent repeal by mere legislative action.

In fact, Congress had enacted the 1866 Civil Rights Act to overturn the effects of the “Black Codes” enacted by the reconstituted Southern state governments in 1865 and 1866 under President Johnson’s Reconstruction policies. Those codes limited the basic civil rights of freed slaves to contract, to own property, and to sue. Republicans in Congress viewed the Black Codes as reducing freed slaves to an inferior legal status, close to slavery, so they believed the Civil Rights Act was constitutionally authorized by the Thirteenth Amendment, which abolished slavery throughout the United States and gave Congress power to enforce that prohibition.

Reading the words of Section 1 against this history, the contemporary reader would likely have concluded that they were aimed at the state actions that the Civil Rights Act of 1866 intended to prohibit: state laws or actions by state officials that treated freed slaves differently from whites with respect to basic litigation, contract, and property rights, or that imposed on freed slaves different punishments, pains, or penalties than would be imposed on whites for the same conduct. Section 5 gave Congress the important but limited power to pass legislation enforcing the protections of Section 1.

The Fourteenth Amendment was integral to the Reconstruction policies of Republicans in the election of 1866: constitutional limitations on state power to withhold basic civil rights from the newly freed slaves; recognition and restoration of the fundamental features of the federal union, in which states hold primary lawmaking authority and Congress is given defined, limited powers; and restoration of the secessionist states to their full powers in a federal union only after they ratified the Fourteenth Amendment. Limiting the prohibitions in Section 1 to state actions was thus not mere happenstance. The limitation had an overriding constitutional aim: to preserve the federal system in which states had broad authority to legislate and act for the common good and the federal government had limited, defined powers.

However, a broad interpretation of Congress’s power under Section 5 was proposed not long after its ratification. Section 1 forbids a state to deny a person the “equal protection of the laws.” The “laws” in that directive should include the long-established common law that protected the right of any member of the public to be served by those private entities that hold themselves out to provide a service to the public: public accommodations, public transportation, and public places of amusement. For the courts of a state to refuse to enforce this common law right when asserted by a newly freed slave (while enforcing the law for a white customer) would be to deny the former slave the equal protection of the laws. One way for Congress to prevent this kind of violation of Section 1 would be to enact a general law under Section 5 protecting everyone’s common law right of access to public accommodations, conveyances, and places of amusement. This could reasonably be understood as Congress’s “enforcing” the right to equal protection of the laws recognized in Section 1. Congress did exactly that in the Civil Rights Act of 1875, which mandated equal access to public accommodations, common carriers, and places of amusement.

Nonetheless, the Supreme Court in The Civil Rights Cases (1883) held the Civil Rights Act of 1875 to be unconstitutional. In holding that Congress had no power under Section 1 and Section 5 of the Fourteenth Amendment to enact that legislation, the Court declared,

The prohibitions of the amendment are against State laws and acts done under State authority. [But the Civil Rights Act of 1875] makes no reference to any supposed or apprehended violation of the Fourteenth Amendment on the part of the State. . . . [It] lays down rules for the conduct of individuals in society towards each other, and imposes sanctions for the enforcement of those rules, without referring in any manner to any supposed action of the State or its authorities.

The Court turned the “state common law” argument against its proponents. If private actors deprive anyone of their traditional common law rights of access to public accommodations, transportation, or places of amusement, those deprived should enforce their rights in the state courts.

Dicta in later cases have questioned the central holding of The Civil Rights Cases but never overruled it. In United States v. Guest (1966), the Supreme Court held that Congress under Section 5 could regulate the conduct of private individuals who conspired with state officials to deprive persons of their rights under Section 1, because a private individual’s conspiracy with state officials brought his conduct within the state-action prohibition. Justices in two concurring opinions suggested that Congress could use its Section 5 powers to reach purely private conduct. But in no case since then has the Court embraced the Guest concurrences. In United States v. Morrison (2000), the Supreme Court reaffirmed the holding of The Civil Rights Cases, explicitly rejected the Guest dicta, and struck down the part of the federal Violence Against Women Act of 1994 that had provided a federal civil remedy for victims of “gender-motivated” violence. The Court reasoned that the act exceeded Congress’s power under Section 5 of the Fourteenth Amendment because it was “directed not at any state or state actor, but at individuals who have committed criminal acts motivated by gender bias.”

The Supreme Court has thus consistently held that some sort of state action is a prerequisite to judicial enforcement of the prohibitions in Section 1 of the Fourteenth Amendment. The remaining question, then, is what counts as state action. Of course, there are easy cases: “state action” clearly includes state legislative, judicial, and executive actions, as well as actions by entities expressly vested with governmental powers, such as municipalities and administrative agencies. The hard cases arise because it is difficult to discover clear standards for determining when conduct of a private entity may constitute “state action” under Section 1 of the Fourteenth Amendment.

The courts have determined that the question is whether, on all the facts and circumstances, including the private entity’s relationship to the state and any of the state’s acts, it is fair to say that the private entity’s conduct is “state action” and thus attributable to the state. Blum v. Yaretsky (1982). Over the years, the Supreme Court has found that a private entity can be a “state actor” if it is acting for the state, acting like a state, or becoming too involved with the state.

Early on, the Court found that whether a person acts for the state is determined by the common law principles of vicarious responsibility for the conduct of others, embedded in the common law of agency, apparent agency, conspiracy, and joint venture.

At common law, for example, a principal is legally responsible for the actions of an agent even outside the scope of his express authority if the agent was clothed by the principal with “apparent” authority. After one false start, the Court concluded in 1912 that a state official who violated a state law or the Constitution and hence exceeded his actual authority nevertheless engaged in state action within the Fourteenth Amendment when the exercise of his apparent authority caused the harm that the Fourteenth Amendment intended to prevent. Home Telephone & Telegraph Co. v. City of Los Angeles (1913). See also Monroe v. Pape (1961). Also, at common law, each member of a conspiracy is responsible for the actions of his co-conspirators. So if a state official conspires with a private person to deprive another of a constitutional right, the action of the private person is “state action” under the Fourteenth Amendment. United States v. Guest. See also Dennis v. Sparks (1980).

The Court after Home Telephone relied on explicit or implied delegation to act for the state in finding state action without requiring that all the technical requirements of agency be met. In the first of two “white primary” cases, the Court held that the allegedly private Democratic Party’s primary elections were held using authority delegated by the state, so the party’s exclusion of black voters was state action in violation of the Fifteenth Amendment, which forbids a state to deny or abridge the rights of citizens of the United States to vote “on account of race, color, or previous condition of servitude.” Nixon v. Condon (1932); Smith v. Allwright (1944).

In the other white primary case, an all-white club held a pre-primary vote of its members, whose approved candidates usually won the subsequent Democratic Party primary. The Court held that this private, segregated pre-primary was state action because it effectively chose the Democratic Party’s candidate for the general election. In conducting that primary, therefore, the club exercised a public function. Terry v. Adams (1953).

When a private party acts like a state, the Court determines that it is performing a “public function.” In Marsh v. Alabama (1946), the Court held that a company town that prohibited the distribution of religious flyers in public violated the First Amendment of the Constitution, (applicable to the states through the Fourteenth Amendment’s due process clause). Marsh held that the company’s exercise of traditional public functions on property that it owned and developed, which appeared to be a town, was subject to the Fourteenth Amendment’s restrictions on state action.

The public function test reached its high-water mark in Evans v. Newton (1966). The Court held that a private entity (a testamentary trust with private trustees) operating a park open to white members of the public was engaged in a public function—it looked like a public park except that it excluded all blacks. This holding was narrowed by a subsequent case, Evans v. Abney (1970), which held that once the park had reverted to private heirs, they could thereafter exclude blacks from park.

The public function test was further restricted by subsequent cases distinguishing Marsh v. Alabama. The Court held that an owner of a private shopping mall was not exercising a public function. A mall is not an entire city, said Hudgens v. National Labor Relations Board (1976). The Court also held that a privately owned utility was not exercising a public function in providing electric power to the public, even though the utility was publicly licensed and regulated, and even though some local governments also provided electricity to the public. Jackson v. Metropolitan Edison Co. (1974). In the course of whittling away at the public function test, the Jackson Court explicitly limited its application to cases in which a private entity performs a necessary public function or exercises powers traditionally exercised exclusively by the state.

The key case for determining when a private entity has become too involved with the state is Burton v. Wilmington Parking Authority (1961). There, a privately owned restaurant and bar, the Eagle Coffee Shoppe, refused to serve a black prospective customer because of his race. The Eagle leased its space from the public parking garage authority and located its restaurant within the parking garage owned and operated by the authority. The Court, in reaching its conclusion, listed many factors, even that a flag was flying from the building. It found that the Eagle’s yearly lease payments and that of other tenants were an indispensable part of the state’s plan to operate the parking garage as a self-sustaining unit, that the parking garage benefitted from patronage by the Eagle’s customers, and that the Eagle’s profits earned through its discriminatory practice were important for its financial success, so that profits “earned by discrimination [were] indispensable elements in the financial success of the governmental agency.” The Court concluded that the state through its public parking authority had become a “joint participant” in the discriminatory practice. The Court explicitly denied, however, that government leasing of property to a private entity always made the government lessor a joint participant in the lessee’s conduct.

Often, private entities act pursuant to a license issued by the state or one of its regulatory agencies. The general rule seems relatively clear: a private entity acting pursuant to a state license is not engaged in state action unless the licensing authority specifically requires, approves, or encourages the licensee to infringe the constitutional rights of others. In Moose Lodge No. 107 v. Irvis (1972), the Court found that a state liquor license granted to a private club that discriminated against blacks did not make that discrimination state action.

The Court has yet to fashion a clear line of analysis in cases in which close involvement by the state makes private action into state action.

For example, the Burton case was thought to stand either for a “reciprocal benefit” test (symbiotic between the state and the private entity) or a test of whether the state was “inextricably entwined” with the private conduct. In Brentwood Academy v. Tennessee Secondary School Athletic Association (2001), the Court found state action by focusing on the inextricable connections between state secondary schools and a private association voluntarily formed by agreement among public and private secondary schools to regulate interscholastic athletic contests. At the same time, the Court also applied the reciprocal benefit test, relying on the nominal dues paid by public high schools to the association and the portion of the ticket sales the schools received attributable to membership in the association, all of which seemed to be de minimis.

Earlier, in Shelley v. Kraemer (1948), the Supreme Court held that judicial enforcement of a private restrictive covenant barring occupancy by “any person not of the Caucasian race” was state action denying equal protection of the laws to the black buyer of the property. The Court held that the lower court’s application of the law of restrictive covenants was state action. Commentators quickly pointed out that the implications of Shelley were too broad: any private contract would become state action, for instance, when enforced by a court. To avoid this overbreadth, later commentators suggested that Shelley should be limited to its facts: a court’s action in enforcing a racially discriminatory agreement would be in violation of the Equal Protection Clause only if it coerced someone to discriminate when he did not choose to do so. In fact, that is what the case said. It referred to a willing buyer (the black purchaser) and the willing seller (the while property owner). The state court sought to prevent that sale, and it was that judicial action to prevent a sale that constituted the state action.

In Reitman v. Mulkey (1967), California had repealed a fair housing act by adopting a constitutional amendment that barred the enactment of any law limiting the right of any property owner to refuse to sell his or her property to any buyer for any reason. The Court held the amendment unconstitutional even though the Fourteenth Amendment did not require the state to enact fair housing legislation. A simple legislative repeal of the California fair housing act, therefore, probably would have been constitutional, and the California amendment was on its face racially neutral. The Court justified its holding by arguing that the amendment would forbid a city or the legislature from enacting laws allowing open housing or forbidding private racial discrimination in the sale or rental of property. The California amendment did more than simply repeal an open housing act; it prohibited passage of another such act in the absence of another amendment to the state constitution. Emphasizing this difference, the constitutional scholar Charles Black subsequently explained the Reitman case as follows: the amendment was unconstitutional because it placed a larger, discriminatory burden on those seeking to obtain legislation that would remove racial discrimination from the housing market. Before obtaining a fair housing act they first had to obtain a constitutional amendment repealing this constitutional amendment. The Court adopted this argument in the subsequent case of Romer v. Evans (1996).

The Court emphasized this point in James v. Valtierra (1971). Justice Hugo Black, for the Court, upheld a California constitutional amendment that forbade the construction of any low-rent housing project unless a majority of those voting in the relevant area approved the low-rent housing. This amendment, unlike the one at issue in Reitman, did not rest “on distinctions based on race.” Reitman imposed a special hurdle on black citizens who sought open housing, and the Court reviews such laws applying strict scrutiny. In contrast, Valtierra imposed a special hurdle on poorer people. The Court applies strict scrutiny to racial distinctions but evaluates wealth distinctions using a rational basis test.

Two cases decided in the last thirty years present puzzling questions of consistency with seemingly settled state action doctrine. In Lugar v. Edmondson Oil Co. (1982), a creditor sought to attach a debtor’s property. Pursuant to state law, the county clerk of the court issued a writ of attachment, which the sheriff served on the debtor. The Court held that the creditor’s petition for a writ of attachment was “state action” because the creditor “jointly participated” with the sheriff in seizing the debtor’s assets without due process of law. Some argue that the Lugar holding seems inconsistent with prior state action cases, which had held a private litigant’s actions to be state action only when the litigant corruptly conspired with a state official in a jointly beneficial scheme to deprive another litigant of his right to a fair trial. Dennis v. Sparks. Without the elements of corruption and conspiracy, the Lugar holding seems to subject any litigant, acting in good faith under presumptively valid laws, to a successful federal civil rights action for damages if a court subsequently determines that the requested judicial action violated the Constitution. One way of looking at this case is that, in general, only the state can use violence to seize property. If the creditor uses the state powers to seize the debtor’s property, that is state action. In some cases, the creditor can use self-help that does not involve violence, such as seizing the car without the help of a sheriff when the debtor is in default of a loan secured by the car. In such a case, there is no state action.

The Court followed Lugar nine years later in Edmonson v. Leesville Concrete Co., Inc. (1991), holding that a private litigant’s racially discriminatory use of peremptory challenges to remove potential jurors in a civil trial constituted “state action.” Reading Lugar as holding that a private litigant’s extensive use of state procedures with the overt and significant assistance of state officials constitutes state action, the Court found significant assistance here by the judge in implementing the litigant’s racially discriminatory peremptory challenges. The Court went on to conclude that the private litigant’s exercise of peremptory challenges was the exercise of a public function. This seems to contradict the Jackson Court’s limitation of the public function test to “activities that have traditionally been undertaken exclusively by the government,” because unhindered choice by private litigants had always been the hallmark of the peremptory challenge procedure in civil cases, justified on grounds of fairness to both litigants. On the other hand, the Court majority said that it is the trial judge who removes the juror, and what a judge does is state action.