Deanna Malatesta is associate
professor of public affairs in the School of
Public and Environmental Affairs at Indiana
University–Purdue University Indianapolis.
Her research addresses contracting, public
management and governance, and public
Julia L. Carboni is assistant professor
of public affairs in the School of Public and
Environmental Affairs at Indiana University–
Purdue University Indianapolis. Her research
focuses on governance, organizational
networks, and cross-sector collaboration.
The Public–Private Distinction: Insights for Public Administration from the State Action Doctrine 63
Public Administration Review,
Vol. 75, Iss. 1, pp. 63–74. © 2014 by
The American Society for Public Administration.
Deanna Malatesta Julia L. Carboni
Indiana University–Purdue University Indianapolis
The public–private distinction is an enduring theme in public administration (PA). While scholars debate the concept for their purposes, so do the courts, which must distinguish private actors from state actors to rule on constitutional claims. Specifi cally, the Fourteenth Amendment to the U.S. Constitution prohibits government from depriving persons of life, liberty, or property without due process. Because due process protections only apply to the coercive force of the state, courts must decide when conduct is attributable to government and not to a private entity. Th is is the essence of the legal principle known as the state action doctrine (42 U.S.C. § 1983, 2000).
We compare public–private distinctions discussed by PA scholars to the dichotomy applied by the judiciary. In revealing similarities and diff erences in the two domains, we add to the body of research attempting to capture the complexities of the public–private dis- tinction. Th ere are several steps to our application. We survey the public administration literature, law review articles, and case law to summarize and compare the most frequently noted characteristics that diff erenti- ate “public” from “private” as used in the diff erent domains. We code the text of seminal court cases and
use qualitative comparative analysis (QCA) to uncover necessary and/or suffi cient factors for fi nding state action. We derive a minimum and maximum defi ni- tion (ideal type) of “public” as used in the social sci- ences and the judiciary. We conclude with suggestions for applying our fi ndings to the outsourcing of public services with the aim of preserving public norms and constitutional protections when designing contracts.
The Literature Do public and private organizations diff er in any meaningful way? Social scientists, including scholars of public administration and management, have grap- pled with the question for decades (see, e.g., Bozeman 1987; Moulton 2009; Perry and Rainey 1988; Rainey, Backoff , and Levine 1976). Th e inquiry has important implications for those who debate the limits of privatiza- tion and outsourcing as well as for understanding what makes some organizations more eff ective than others. Scholars take varied positions on this issue: some down- play that there is any distinction at all, while others emphasize the attributes that diff erentiate public organi- zations from private organizations. A third group, which we call the “reformers,” have taken both positions. Reformers are on record claiming the virtues of the market, yet they have also contributed to the perception
Th e Public–Private Distinction: Insights for Public Administration from the State Action Doctrine
Abstract: Public administration scholars continue to grapple with how and why public organizations diff er from private organizations. Th e judiciary deals with similar questions in ruling on constitutional claims that apply exclusively to state actors. Th e authors consider similarities and diff erences between scholarly and judicial approaches, adding to the body of research attempting to capture the complexities of the public–private distinction. Th e application in this article includes the coding of seminal court decisions and qualitative comparative analysis (QCA) to fi nd combinations of causal conditions that lead to state action rulings. Th e specifi cs revealed through QCA provide valuable lessons for extending public norms and preserving constitutional protections when outsourcing public services.
Practitioner Points • Public managers continue to grapple with distinguishing “public” from “private.” Meanwhile the courts
apply the State Action Doctrine to distinguish state actors from private actors in order to determine the boundaries for constitutional protections.
• Public administrators should be knowledgable about the State Action Doctrine; its application has critical implications for contracting and the extension of constitutional protections.
• When public managers outsource services to private actors they can craft contract language to guarantee consti- tutional protections, thereby fi lling any constitutional voids that may result from defi ning an actor as “private.”
64 Public Administration Review • January | February 2015
are testaments to these objectives. Th e texts used in MPA programs underscore a similar point—that understanding management, law, and politics is somehow diff erent in the public sector than in the private sector (see, e.g., Rainey’s Understanding and Managing Public Organizations, a frequently used textbook in top MPA programs, now in its fi fth edition; see also Rosenbloom, Kravchuk, and Clerkin 2009). Notable public fi gures share similar goals. For example, Paul Volcker, chair of the Federal Reserve from 1979 to 1987, leads the Volcker Alliance, a nonprofi t alliance dedicated to “improving government performance at all levels” (Volcker 2014).
Th e reformers, most notably those associated with New Public Management (NPM) and the National Performance Review (NPR) and initiatives of the 1980s and 1990s, can be characterized as strad- dling both positions (see, e.g., Osborne and Gaebler 1992). On the one hand, reformers contribute to the blurred line between “public” and “private,” as market solutions are often equated with public interest (Verkuil 2005). However, the idea that public and private organizations are characteristically distinct is also embodied in the reforms. After all, why copy business if it is not somehow better than government? In general, both NPM and NPR portray markets as the utopian solution for exchanges. NPM advocates propose more businesslike practices and budget controls, as well as formal constraints to temper the eff ects of political infl uences (Boston et al. 1996). Many of their systems-improvement principles are along the lines of scientifi c management (Box et al. 2001; Pollitt 1993). NPR advocates also cast public agencies in an even less favorable light, for example, by questioning the fundamental role of government (e.g., Goodsell 1994; Rosenbloom 1993). Likewise, champions of privatization contend that markets have superior incentive proper- ties that cannot be matched in the public sphere; market competi- tion controls self-interest and induces sellers to provide quality services and products at optimum prices (e.g., Savas 1987, 2000).
As with all reforms, there are rebuttals. Th e New Public Administration (NPA) paradigm holds that businesslike practices fall short of addressing social and political crises. Other scholarship resists the technical and instrumental leanings of NPM and NPR (e.g., Box and King 2000). NPA proponents also see an active role for bureaucrats as one of serving rather than just steering (Denhardt and Denhardt 2000). According to this perspective, public servants are obligated to protect democratic values. Consistent with these ideals, Frederickson (1980) calls on public administration scholars and practitioners to renew the emphasis on social equity values. Likewise, public value theorists argue that reformers have sidelined the citizenry and the relationship between individuals and society; they equate managerial and organizational success with advanc- ing the rights and obligations of citizens (e.g., Benington 2011; Bozeman 2007; Meynhardt 2009; Moore 1995, 2014). Public agencies are portrayed as complex adaptive systems, better situ-
ated to work as part of a network and with citizens to advance public values. Th us, in the public value literature public agencies are not only qualitatively diff erent from private organizations, they are in some ways superior.
Of course, it is easier to claim that public organizations are distinctly diff erent from private organizations than it is to precisely
that government and business are fundamentally alike. Th is section summarizes the varied positions and approaches to the question.
Literature that downplays the distinction includes some of the classics as well as recent research in organizational economics. For example, Weber uses the term “bureaucracy” broadly to include government agencies and businesses (Shafritz, Hyde, and Parks 2004, 50–55). Taylor’s scientifi c management principles are widely applied in both sectors (Rainey 2009, 60). Simon discusses the political context of organizations in some of his research but does not reserve his recommendations for dealing with politics in govern- ment agencies (Rainey 2009, 60, referring to Simon, Th ompson, and Smithburg 1950). Similarly, some contemporary scholars claim that distinctions are diffi cult to draw because of the overlapping nature of the sectors (Cooper 2003; Kettl 2002; Moe 1994), while others contend that any diff erences that do exist do not dictate performance outcomes (Haas, Hall, and Johnson 1966).
A few economists can be grouped with those that underempha- size the distinctions, although they frame the debate diff erently. Williamson (1975, 1985, 1999) identifi es markets, hybrids, and public bureaucracies as alternative governance forms and deline- ates the main attributes on which they diff er: incentive intensity, administrative control, adaptation, and the relevance of contract law (Williamson 1999, 314–36). Williamson emphasizes that all exchanges have associated transaction costs. According to the transaction cost approach, determining the optimum organizational form involves comparing transaction costs associated with each governance mode, which is accomplished by aligning organizational attributes with the related costs of the exchange. Th e literature refers to the approach as “discriminating alignment” (Williamson 1996, 46–47). Williamson acknowledges the distinct nature of some public goods and argues that the attributes of the good or service and the extent of competition are crucial to the assessment (Williamson 1996, 84–85). Williamson also acknowledges that public bureaucra- cies are prone to political costs, which “more easily allow for conver- gent expectations,” another consideration, yet one that is not itself determinative (Williamson 1999, 319–20). Th us, in a transaction cost world, one type of organization is not inherently more capable than another type; it all depends on the particulars of the exchange.
Jensen and Meckling (1976) off er a complementary argument, emphasizing the importance of agency costs, one specifi c type of transaction cost, as well as the distribution of decision rights in determining outcomes. Th ese scholars view organizational bounda- ries as a mere legal fi ction, serving as a sort of invisible “nexus for contracting relationships” (Jensen and Meckling 1976, 8). From this perspective, it makes no sense to distinguish public organizations from private organizations; the relationship specifi cs determine the incentive structure, which in turn accounts for any diff erences in organizational outcomes.
Another group of scholars acknowledge organizational distinctions mainly to advance the theory and practice of public management (e.g., Allison 1980; Wilson 1989). Th e titles of several peer-reviewed journals, including Public Administration Review and the Journal of Public Administration Research and Th eory,
It is easier to claim that pub- lic organizations are distinctly diff erent from private organi- zations than it is to precisely
identify the attributes that make them distinct.
The Public–Private Distinction: Insights for Public Administration from the State Action Doctrine 65
public value model include those of Cole and Parston (2006) and Moulton (2009).
In sum, views on the commonalities and/or distinctions associated with public and private organizations have a long history. As with all normative research, diff erences in opinion can often be traced to values and the implications of siding one way or another. As the saying goes, “where you stand depends on where you sit.”1 Empirical research on the topic remains today as it was described by Perry and Rainey in 1988: “Th e many taxonomies, survey comparisons, and other approaches have not yet accumulated into an acceptable explanation of how or why public organizations diff er from private ones” (Rainey 2009, 192). However, these eff orts have produced a list of frequently noted distinguishing attributes of public organiza- tions that provide a foundation for future research. Th ese include coercion authority, the absence of economic markets, the presence of constraints (authoritative, political, decision-making procedures, civil service rules, goal complexity, goal ambiguity, low-powered incentives), broad impacts for activities, the unique nature of public goods, higher expectations for fairness, and the unique motivations of employees. In the next section, we summarize the progress made by the judiciary in distinguishing between public and private.
The Legal Dichotomy Adhering to the legal principle known as the state action doctrine, courts distinguish between public and private actors. Simply, distinguishing public (state) actors from private actors is an una- voidable threshold inquiry when courts review claims of due process
violations. Courts take a two-step approach in these cases. First, they answer the state action question: is the conduct attributable to the state? If the requirement is met, the court addresses the merits of the claim. If it is not met, the case is dismissed. Th us, in all state action cases, the court derives a workable defi nition of “public,” if only for the par- ticular facts of the case before it. State action
cases have relevance for public administration scholars who grapple with the public–private distinction for their own purposes. Here, we provide an overview of the most important U.S. Supreme Court state action cases to elucidate the range of conditions and the basis for the rulings.
State action cases that come before the U.S. Supreme Court vary with respect to the parties involved in the underlying action. All cases involve a formal government body, for example, a municipality or a legislature, and most cases also involve another party actor, such as a business or a nonprofi t (e.g., Brentwood Academy v. Tennessee Secondary School Athletic Assn. 2001). In addition, many cases before the Court implicate traditional contract forms of government pro- curement (e.g., Blum v. Yaretsky 1982), suggesting that the question of distinction arises in contexts similar to what we fi nd in the public management literature—situations involving agency decisions to pursue less direct modes of public service delivery. Interestingly, the Court does not always fi nd state action in cases involving only a formal government body or government employee. For example, the Court found “no state action” in a case focusing on the conduct of a public defender (Polk County v. Dodson 1981) in which there was no involvement of a contractor. In at least one case, the Court explicitly
identify the attributes that make them distinct. Nevertheless, numerous scholars have attempted to classify organizations by their attributes. We describe the most frequently cited attempts in the next section.
Attempts to Classify Organizations by Their Attributes Dahl and Lindblom (1953), Wamsley and Zald (1973), and Blau and Scott (1962) are among the widely cited early attempts to diff erentiate public and private organizations. Dahl and Lindblom propose a continuum with public agencies (government control- led) on one pole and “enterprises” (market controlled) on the other. Dahl and Lindblom reason that organizations have varied decision systems that direct economic activity, which in turn infl uences their behavior. In their schema, public organizations are subject to high levels of political authority, typically in the form of elaborate rules and procedures. Dahl and Lindblom refer to these evolving enti- ties as political hierarchies or “polyarchies.” As a consequence of the multilevel political constraints imposed on them, polyarchies contend with diverse, multiple, and ambiguous goals that compli- cate eff ectiveness and performance measurement. In contrast, the organizations that they refer to as “enterprises” are more fl exible and effi cient because they rely on market incentives and the pricing sys- tem to induce exchanges and allocate resources. Th us, for Dahl and Lindblom, the main diff erentiating attributes of public and private organizations are political and decision constraints.
Building on this idea, Wamsley and Zald (1973) suggest that an organization’s place on the continuum is attributed mainly to either ownership and/or funding sources (see Rainey 2009, chap. 3). Organizations that are pri- vately owned and funded, primarily through sales and private donations, resemble the pure type of private fi rm, while those that are pub- licly owned and receive most of their fund- ing from tax revenues refl ect the traditional public agency. Hybrids have more mixed characteristics, for example, privately owned but publicly funded organizations (e.g., some defense contractors). In their schema, Blau and Scott (1962) focus on the impact of an organization’s activities, with a specifi c focus on public interest. Public organizations—in their lexicon, “commonweal organiza- tions”—serve the interests of the general public, while private fi rms serve their owners.
More recent scholarship in the tradition of public administration builds on the earlier taxonomies described here. Bozeman (1984, 1987, 2007) focuses on public interest as a diff erentiator but also proposes that organizations are constrained or enabled by politi- cal and economic authority. Bozeman conceptualizes political and economic authority as independent of any legal status. One of his main claims is that organizations are more or less public or private, depending on the extent to which they are subject to political or economic authority. Th at is, nearly all organizations have some degree of “publicness,” which he conceptualizes as a cumulative rather than a categorical attribute. Subsequent research by Bozeman and his colleagues aims to identify the distinctive features of organizations perceived as being more public or more private (e.g., Bozeman and Bretschneider 1994; Coursey and Bozeman 1990; Rainey, Pandey, and Bozeman 1995). Other variations on the
In all state action cases, the court derives a workable defi – nition of “public,” if only for the particular facts of the case
66 Public Administration Review • January | February 2015
with public funds. However, the Court was not persuaded by this fact alone. In a separate decision (Blum v. Yaretsky 1982), the Court noted that complete reliance on government funding does not make an actor public. Th us, it is not clear from cases that focus on the extent of state involvement with the actor committing the alleged violation (government–actor nexus) that any specifi c circumstances will always leads to a fi nding of “symbiosis” or “entwinement.” Moreover, in cases that focus on the government–actor nexus, we cannot conclude that the attribute is cumulative or that a level of the connection triggers a court fi nding of state action.
At this point, we can identify the range of conditions considered relevant when courts take on state action cases, whether or not the court fi nds state action present. Th e conditions include (1) the nature of the party at the center of the action (formal government agency, legislature, government employee and/or other party- contractor, nonprofi t, etc.); (2) whether the case involved a “public function” (3) whether the state compelled or encouraged the alleged constitutional violation; (4) whether there was a suffi ciently close nexus between the state and the alleged action; (5) whether there was joint participation between the state and the actor; (6) whether the party was controlled by the state; (7) whether the actor and the state relationship was symbiotic; and (8) whether the state and actor were intertwined. We add a fi nal condition to capture other pos- sibilities: (9) any countervailing circumstances relevant to the ruling.
Having reviewed the social science literature2 and seminal court cases, some common ground begins to appear. Before discussing the comparisons, we note the importance of context: the social science
discussions of a public–private dichotomy are grounded in concerns about privatiza- tion and management, whereas the legal dichotomy is grounded in constitutional concerns. “Defi nitions can only be elucidated by considering typical contexts where these words are at work” (Hart 1954). Keeping the diff erent objectives and approaches in mind, table 1 compares the attributes that frequently surface in public administration to the range of considerations applied by the courts. Th e
attributes appear side by side with the original public administration literature and the case law to provide an idea of the relative impor- tance of the attributes and factors in each domain.
We highlight a few notable points in table 1. Like public adminis- tration scholars, the judiciary has acknowledged that private corpo- rations and nonprofi ts can eff ectively become “public.” In addition, both public administration and the judiciary include government functions as an attribute of “public.” Th e attributes of political control and coercion that often surface in public administration as distinguishing public organizations also appear to be important to courts when the state action decision focuses on the government’s involvement with the conduct alleged to be unconstitutional.
We can also draw a few tentative conclusions about the extent to which some considerations matter in the court system. For exam- ple, the legal status of the actor (government employee, nonprofi t, corporation) does not determine state action by itself. Likewise, complete reliance on government funding does not determine state
noted that the actor’s legal designation is one factor to be considered but does not itself determine the ruling (Marsh v. Alabama 1946). Ownership is not a decisive factor.
A subset of cases focus on public functions. Th e judiciary has set a high bar in determining which functions fall into this category. For example, in Jackson v. Metropolitan Edison Co. (1974 at 352–53), the Court did not view electricity service as a public function because the service was not “traditionally the exclusive prerogative of the state” or “traditionally associated with sovereignty to be deemed a public function.” Similarly, in Flagg Bros., Inc. v. Brooks (1978 at 163), the criteria for public function were not met; the Court rea- soned that education, fi re, police protection, and even tax collection could not meet the high threshold established. However, in Marsh v. Alabama (1946), the Court found state action when a company- owned town denied the right of Jehovah’s Witnesses to distribute religious materials, noting the government’s traditional responsibil- ity in town governance.
In cases involving both a formal government body as well as an additional private party who allegedly acted unconstitutionally, the Court probes the extent of state involvement in the action and/or the extent of state involvement with the other party. In other words, the state involvement question can take two diff erent forms: to what extent is a formal government body involved in the conduct (gov- ernment–action nexus) alleged to be unconstitutional? Or, to what extent is the formal government body involved with the actor (gov- ernment–actor nexus) alleged to have committed the constitutional violation? Cases addressing the government–action nexus range from instances in which the conduct was compelled by the state to instances in which the state was simply passive in allowing another party to act. For example, in Burton v. Wilmington Parking Authority (1961 at 725), the Court said that the “state so far insinuated itself . . . that it must be recognized as a joint participant” in the conduct. In Public Utilities Commission (PUC) of the District of Columbia v. Pollak (1952), government inspection and approval of an activity led to a ruling of state action. However, in Polk County v. Dodson (1981), the Court found no state action where the prosecutor had the ability to act independently. On the surface, the facts and the decisions in Burton and in PUC suggest the government–action nexus is a cumulative attribute, but the out- come in Polk can be interpreted as contradicting this conclusion.
In considering the extent of state involvement with the actor committing the alleged violation (government–actor nexus), the Court has described the relationship using terms such as “symbi- otic” or “entwinement.” In Burton v. Wilmington Parking Authority (1961), the government and the actor had a lessor–lessee relation- ship, which, according to the Court, conferred mutual benefi ts. In Brentwood (2001) the Court determined the actors to be eff ectively “entwined” because government and leadership of the private school were interdependent. Reliance on government funding is another factor the Court has considered in ruling on the nature of the relationship. In Rendell-Baker v. Kohn (1982), a case involving a privately operated high school where counselors alleged a wrongful dismissal, the Court focused on the fact that the school operated
Th e social science discussions of a public–private dichotomy
are grounded in concerns about privatization and management,
whereas the legal dichotomy is grounded in constitutional
The Public–Private Distinction: Insights for Public Administration from the State Action Doctrine 67
pathways (equifi nality). QCA is a case-based analytical tool that relies on set relations and Boolean algebra to identify diff erent causal pathways that lead to a specifi ed outcome. Applying QCA to the court decisions, we aim to answer these questions:
• Can we go beyond establishing general criteria for diff erentiat- ing public from private? Th at is, are there specifi c “tests” for fi nding state action?
• What is the relative infl uence of diff erent conditions consid- ered by the courts in state action cases? Are some factors or combinations of factors more important than others?
• Has the defi nition of the state (public) changed over time, or is it static?
• Does “public” have a cumulative quality? Th at is, do factors such as economic and political authority combine in degree such that at a point an entity becomes “public”?
action, but mutually conferred fi nancial benefi ts are a consideration in determining state action. Extensive regulation is also insuffi cient for fi nding state action, but government inspection and approval of conduct has led to state action, and the ability to act independ- ently has led to fi nding no state action. Rather than state action being the result of a single factor, it appears to result from mul- tiple factors present in a single case. In other words, it is not the ingredients but the recipe that matters. Finally, although few of our court decisions imply a possible cumulative eff ect of attributes for fi nding state action, some of our court decisions contradict such a conclusion.
Qualitative Comparative Analysis (QCA) To better understand how causal factors combine to produce state action, we use qualitative comparative analysis (Ragin 1987, 2008) to analyze cases. State action can be the result of multiple causal
Table 1 The Public–Private Distinction: Comparing Public Administration and the Judiciary
Basis for Distinction in Public Administration Relevance in Case Law
1 Absence of economic markets for outputs; reliance on governmen- tal appropriations (Alchian 1967; Dahl and Lindblom 1953; Niskanen 1971; Wamsley and Zald 1973; Weidenbaum 1969)
Government funding is considered in determining whether there is a suffi cient nexus between government and the other actor, but the Court has clearly stated that complete reliance on government funding does not determine the presence of state action (Blum v. Yaretsky; Rendell-Baker v. Kohn). Mutually conferred benefi ts; symbiotic relationship (lease agreement) has led to deci- sion of entwinement and, in turn, state action (Burton v. Wilmington Parking Authority).
2 Presence of elaborate constraints, government oversight (Antonsen and Jørgensen 1997; Bozeman 1987; Dahl and Lindblom 1953; Thomp- son 1967; Wamsley and Zald 1973; Woll 1963)
Regulation is another factor considered in determining whether there is a suf- fi cient nexus between government and the actor, but the Court has clearly stated that even extensive regulation is insuffi cient for determining state action (American Manufacturers Mutual Insurance v. Sullivan; Jackson v. Metropolitan Edison Co.; Moose Lodge No. 107 v. Irvis; Smith v. Allwright). In determining state involvement with the action, government inspection and approval of an activity have led to ruling of state action (PUC v. Pollak).
3 Ownership (Dahl and Lindblom 1953; Wamsley and Zald 1973) Courts have ruled that ownership is not a decisive factor (Marsh v. Alabama). Likewise, it is not the label of the party (public/private/nonprofi t) but rather whether the actor carries a badge of state authority.
4 Presence of political infl uence (Bozeman 1987; Gawthorp 1971; Golem- biewski 1969; Woll 1963)
No state action cases found that expressly consider the presence of political infl uence as a factor distinguishing private actors from public actors, although coercion has been recognized as unique to the state and has been a determin- ing factor in fi nding a suffi cient nexus between government and alleged un- constitutional conduct. State authorization of the activity is not enough (Flagg Bros., Inc. v. Brooks). Conversely, the ability to act independently warrants a fi nding of no state action (Columbia Broadcasting System, Inc. v. Democratic National Committee; Polk County v. Dodson).
5 Acts can be coerced, monopolistic, or unavoidable (Banfi eld 1976; Caiden 1971; Lowi 1969; Meyer 1972; Stahl 1971; Weidenbaum 1969)
6 Strategic decision-making processes subject to stimuli, interruptions, constrained (Appleby 1945; Charlesworth 1968; Dahl and Lindblom 1953; Mainzer 1973; Mintzberg 1973)
7 Administrative authority constrained by law (Buchanan 1975; Gaw- throp 1971; Golembiewski 1969; Siffi n 1963)
8 Involved in production of public goods (Burkhead and Miner 1971; Hinrichs and Taylor 1972)
Courts use the term “government function” to distinguish relevant cases but have considered very few functions to be “traditionally the exclusive preroga- tive of the state” (Jackson v. Metropolitan Edison Co.).
9 Broader impact for activities/unique expectations of fairness, and other public values (public interest) (Blau and Scott 1962; Caiden 1971; Dewey 1927; Mainzer 1973; Wamsley and Zald 1973).
Some race-based cases appear to consider citizens’ expectations and public values.
State’s abdication of responsibility has been a factor leading to ruling of state action (Terry v. Adams); state cannot evade constitutional duties by contract- ing (Evans v. Newton).
10 Goal ambiguity, multiplicity, confl ict (Banfi eld 1976; Gawthrop 1971; Rainey 1983, 1993; Seidman 1970; Weiss 1974)
No direct corollary for this group in our court decisions. No state action cases expressly address these factors in attributing action to the state.
11 Organizational structure more elaborate (Banfi eld 1976; Woll 1963) 12 Incentives and incentives structures more constrained in public (Ban-
fi eld 1976; Lawler 1971; Rawls, Ullrich, and Nelson 1975; Schultze 1970) 13 Work-related attitudes and behaviors vary; higher levels of public
service motivation and lower valuation of monetary incentives in public (Banfi eld 1976; Lawler 1971; Perry 1986, 1996, 2000; Rainey 1979, 1982, 1983)
14 Organizational and individual performance; public less effi cient (Alchian 1967; Downs 1967; Gawthrop 1971)
Note: The list in the left column is based on Rainey (2009, 83, exhibit 3.1, Distinctive Characteristics of Public Management and Public Organizations: A Summary of Common Assertions and Research Findings). The right column is based on a selection of U.S. Supreme Court cases in our sample. The Cases Cited section in the article includes complete text citations.
68 Public Administration Review • January | February 2015
QCA Analysis and Results We examined our data using fsQCA software (Ragin and Davey 2012). FsQCA identifi es necessary and suffi cient conditions as well as relevant combinations of causal conditions “that are insuffi cient but necessary parts of causal recipes which are themselves unneces- sary but suffi cient” (Ragin 2000). Th e software also subjects the data to a minimization procedure that identifi es the simplest set of conditions that can account for the outcome (in this case, a fi nding of state action). We also used Stata code developed by Longest and Vaisey (2008) to confi rm our results. In table 3, we show a truth table, a matrix of conditions and their occurrences produced by the software. Following QCA guidelines, the truth table shows combi- nations with at least two cases (Ragin 2008).
Before discussing the specifi cs in the truth table, we note that there is no single necessary or suffi cient condition for a fi nding of state action (software output not shown). Th is QCA fi nding comports with what the U.S. Supreme Court has said itself (see Gilmore v. City of Montgomery 1974; Brentwood Academy v. Tennessee Secondary School Assn. 2001). Although no condition by itself can produce state action, some individual conditions are more relevant than oth- ers. Specifi cally, condition A (government–action nexus) is present in 74 percent of cases leading to state action, coverage indicating more relevance than any other condition. Sixty-eight percent of cases leading to state action involve condition F (public function), indicating this condition is also highly relevant to state action decisions.4
Turning to the truth table, each combination of conditions leading to state action is represented as a row. From the truth table, consist- ency and coverage scores are calculated. Consistency measures how often a combination is the subset of the outcome. For example, a combination with a consistency of 0.8 indicates that 80 percent of cases with the solution result in the outcome, which in this research is a fi nding of state action. Coverage measures how much of the outcome is explained by the combination of conditions. For example, a combination with coverage of 0.5 means that the solution explains 50 percent of the cases in which the outcome is present. Consistency and coverage may not always agree with each other. For example, a combination may have high consistency and low coverage, indicating that cases with that combination of causal conditions are likely to have a specifi c outcome, but the same cases may only explain a small portion of the outcome, resulting in low coverage.
Th e combination PFAB, appearing in the fi rst row of the table, is the most common confi guration, with 23.08 percent of the sample best fi tting it. Th e notation of all capital letters indicates the presence of all four conditions in our coding scheme. Th at is, the Court consid- ered the entire range of possible conditions that we coded in arriving at its decision. In QCA terms, these cases implicate the presence of P (case includes a private actor, not just a formal government body) and the presence of F (public function discussed) and the presence of A (government’s involvement with the alleged conduct consid- ered) and the presence of B (the Court considered the relationship between a formal government body or government employee and a private party). Th e last row includes fi ve cases (19.23 percent of the sample) with the specifi c combination pfab; the notation of all low- ercase letters indicates the absence of all conditions. Th e cases listed
Case Selection and Coding Scheme To select cases consistent with QCA recommendations, we reviewed law journals to fi nd scholarly discussions about the state action doctrine. Next, we gained an in-depth familiarity with the text of each case. We added cases to our sample that were noted as precedents within the text of the cases we initially examined. Our fi nal sample includes 27 cases dating from 1896 to 2001 and including both positive and negative cases. Positive cases are those that exemplify the outcome, that is, cases in which the Supreme Court found state action (outcome = 1). Negative cases are can- didates for the outcome but failed to display it; these are cases in which the Court considered the familiar conditions but did not fi nd state action (outcome = 0). To develop our coding scheme, we took the initial set of nine conditions (see previous section) and collapsed them into four conditions. QCA methodologists recom- mend combining substitutable conditions because too many vari- ables can pose a problem of uniqueness (Varone, Rihoux, and Marx 2006). Th e four resulting conditions are (1) the nature of the party, (2) whether the case involved a public function, (3) whether the decision centered on the actions of the state and the party, and (4) whether the decision focused on the actors’ (government and other party) relationship.3 Th ese four conditions refl ect our main variables of interest for analysis. Our coding values are conventional binary measures (0/1). In the language of QCA, our analysis is based on crisp sets, and each case is either a member of the set or not a member of the set. To be clear, we can only consider whether a condition was or was not considered by the Court, not how much it mattered, as the weight given to each condition is not always evi- denced in the Court’s decisions. Absence in the set is coded as zero (0). Presence in the set is coded as one (1). Importantly, our cases constitute diversity with regard to our conditions of interest, which is evidenced in our summary statistics. Table 2 summarizes codes, condition descriptions, and key statistics.
Table 2 Codes, Condition Descriptions, and Summary Statistics
Code Condition Description Obs. Mean SD Min. Max.
O Outcome: State action found 27 0.70 0.47 0 1 P Actor/defendant is private/not
formal government agency+ 26 0.73 0.45 0 1
F Public function 27 0.33 0.48 0 1 C* State compelled or encouraged
act complained of 27 0.37 0.49 0 1
N* Suffi ciently close nexus between state and challenged action
27 0.44 0.51 0 1
J* Joint participation between state and other actor
27 0.30 0.47 0 1
R* Entity controlled by state 27 0.15 0.36 0 1 S* Symbiosis 27 0.15 0.36 0 1 E* Entwinement 27 0.11 0.32 0 1 V* Countervailing circumstances 27 0.04 0.19 0 1 A* Court decision centered on action
(C, N, or J) 27 0.81 0.40 0 1
B* Court decision centered on actor (R, S, or E)
27 0.41 0.50 0 1
Y Case decided before 1946 27 0.26 0.45 0 1 Z Case involves race issues 27 0.48 0.51 0 1 H Case involves First Amendment
issues 27 0.19 0.40 0 1
+n = 26 because the Court did not discuss party status in reaching a decision in Nixon v. Condon. *The Court considered this condition, but parties may or may not have satisfi ed this condition.
The Public–Private Distinction: Insights for Public Administration from the State Action Doctrine 69
use QCA to check claims about subset relations (Schneider and Wagemann 2010; see also Kvist 2006, applying QCA as a tool to develop for operationalizing theoretical concepts and constructing ideal types). Accordingly, we draw conclusions by going back to the text of these cases. Referring to the combination pfab (last row of the table), which has perfect consistency (state action always = 1), we fi nd these cases range from years 1897 to 1940. Th us, the earliest cases in our sample also constitute the purest form of state action. We see this as evidence that the determination of what constitutes state action became more diffi cult in later years. Applying the more familiar language of public administration, it appears that the defi – nition of “public” has evolved over time.
Th e combination PfAb (row 4) also has perfect consistency; in all cases, the outcome equals 1. Th ese cases involve a private party in addition to a formal government body (denoted by the capital let- ter P ) and the Court did not discuss or consider a public function (denoted by the lowercase f ). Th e Court’s focus was on govern- ment’s involvement in the conduct complained of (denoted by the capital letter A) and not on other ways in which the two parties might be intertwined (denoted by the lowercase b). Considering the combination of conditions along with the case text, we see that cases range from 1948 to 1988, so they all appear after the earlier “easy cases.” Th e government involvement is not as direct in these cases as they all involve a private actor in addition to the formal govern- ment body. Th e courts’ main focus in arriving at a decision appears to be on government involvement in the alleged constitutional violation, although the involvement takes many diff erent forms (see discussion in previous section). However, we cannot conclude from the text of the cases that involvement can be measured in degree or that a threshold level of government involvement with the conduct gives rise to a fi nding of state action.6 For example, in PUC v. Pollak (1952), the state authorized the transit company’s broadcasting policy, which was the subject of the underlying complaint, and in Reitman v. Mulkey (1967), the state involved itself in discrimination by passing a statute permitting property owners to sell, lease, or rent to anyone for any reason. Th e challenge of determining a degree of involvement or a threshold is exemplifi ed even more in Adickes v. S. H. Kress & Co. (1970). Adickes centers on a private restaurant owner’s discrimination of a teacher and her black students at a lunch counter. By most accounts, the Court reached to fi nd state involve- ment in the conduct when it found state action based on govern- ment passivity, which, according to the Court, established “a custom of segregation” (see Madry 2001, 381).
Th e truth table also shows combinations that produce diff erent out- comes (contradictions). Too few variables can increase the number
in row 6 (the last row) are the easiest cases for the Court, as fewer considerations suggest limited inquiry to arrive at a decision. In these cases, the Court’s ruling revolves around one main actor, which in each case is a formal government body or no additional private actor (denoted by a lowercase p in the confi guration). Th e easy cases also do not involve public functions (denoted by a lowercase f ). Moreover, because the main actor is a formal government body or state offi cial carrying out state policy, the Court had no reason to consider factors such as whether the government compelled the act of another party (hence the lowercase a) or the nature of the interde- pendence between actors (hence the lowercase b).
Th ere are two sets of conditions in the truth table (pfab in row 6 and PfAb in row 4) with perfect consistency, that is, the same outcome is always observed. When these conditions appear as a set, the Court consistently attributes the action to the state (outcome =1). Perfect consistency is evidence of an integral connection among cases dis- playing the associated set of conditions, but it does not guarantee a meaningful set-theoretic relationship exists (Ragin 2008). Our two condition sets with perfect consistency result in the following logical notation:
pfab + PfAb 5,
where lowercase letters represent negation or the absence of a condition; capital letters represent the presence of a condition; and + represents the logical or. In plain language, the equation simply states that courts consistently fi nd state action in two kinds of cases:
1. Cases solely that involve a formal government agency or government employee as the main actor and that do not require the court to consider the government’s involvement in the conduct complained of or the government’s relation- ship with another actor and where public function is also not an issue.
2. Cases that involve both a formal government body or gov- ernment employee as well as another “private” actor such as a nonprofi t or business entity, and where facts before the court center on the government’s involvement in the conduct complained of, and where public function is also not an issue.
Th e QCA approach is intended to be iterative. Table 3 illuminates cases that are connected in some way, thereby prompting the researcher to go back to each case to understand the idiosyncrasies that may lie within. Consistent with other QCA applications, we
Table 3 Truth Table
Party Function Nexus between State and Action
Nexus between State and Actor
Combination of Con- ditions Considered Freq. Cases+
1 1 1 1 PFAB 6 Jackson, San Francisco Arts & Athletics, Evans v. Newton, Terry, Blum, Rendell 1 1 1 PFAb 3 Marsh, Flagg, Edmonson 1 0 1 1 PfAB 5 Moose, Columbia, Smith, Burton, Brentwood 1 0 1 PfAb 5 Reitman, Adickes, Shelley, PUC, West 1 0 1 pfAb 2 Polk, Gilmore 0 0 pfab 5 Chicago, Lovell, Carter, AFL, Rogers Total 26
+ Full case names are given in the Cases Cited section of the article.
70 Public Administration Review • January | February 2015
Minimum and Maximum Defi nitions A minimum defi nition identifi es the “bare essentials of the con- cept” (Gerring and Barresi 2003, 207). In set theory, an organiza- tion must possess these attributes to be considered a member of any set labeled “public.” Our QCA results also suggest a minimum and maximum defi nition of public that can be employed to further compare and contrast the conceptualization of “public” in public administration with the judiciary’s “state actor.” For example, QCA results signal the essential nature of state involvement when the conduct or alleged constitutional violation involves a party other than a formal government body, such as a contractor or nonprofi t. Th e language of the Supreme Court in Adickes (1970 at 152) supports this conclusion. Th e Adickes Court stated that “with the possible exception of an exceedingly opaque district court opin- ion, every lower court opinion of which we are aware . . . has con- cluded that [state action] . . . requires state involvement” (emphasis added). By contrast, no consensus easily emerges from the public administration literature consistent with minimum defi nition of “public.” However, as an alternative to the conventional defi nition of “minimal.” Collier and Levitsky (1997) propose that a mini- mal defi nition can also identify elements that most would agree, similar to a “procedural minimum.” Recalling our interpretation of QCA results, there appears to be a higher bar for meeting the defi nition “public” in judicial applications because a government employee is not automatically public (see Polk County v. Dodson 1981). Th e Court always considers the extent of state involvement in the action in determining whether an action is fairly attributable to the state. (Note: In the truth table, the presence of condition A always appears with the presence of condition P). On the other hand, there is a presumption in the public administration literature that a government employee is public. Accordingly, we propose the following minimum defi nitions:
JUDICIARY MINIMUM: An actor/organization that is not a formal government body is public if the state is connected with or involved in the alleged violation of due process.
PA MINIMUM 1: An actor is public if it is a formal govern- mental body or government employee.
PA MINIMUM 2 (ALTERNATIVE MINIMUM): An actor is public if it is somehow connected to government either through ownership, funding, or authority.
QCA results also aid in deriving a maximum defi nition, also referred to as a Weberian or ideal type. An ideal type is achieved when one takes traits meaningful in their essential features and distills them into a unifi ed ideal construct (Weber 1949, 76). An ideal type is “formed . . . by the synthesis of a great number of diff use, discrete, more or less present and occasionally absent individual phenomena, which are arranged according to those one-sidedly emphasized view- points into a unifi ed analytical concept” (90). “In constructing an ideal type, the objective is to reveal the qualities of interdependence, causality, and signifi cance” but not to judge as right or wrong (12). Ideal types can also help clarify what is excluded from the defi nition. In applications of the state action doctrine, the ideal/maximum legal defi nition goes beyond the minimum defi nition, specifying not only the existence of a relationship between the actors but also an inti- mate relationship. In contrast, the public administration literature
of contradictions (Varone, Rihoux, and Marx 2006). Original case text is also useful for identifying possible conditions not identi- fi ed earlier to solve contradictions. According to Schneider and Wagemann (2010, 405), researchers using the QCA method should rethink the case selection or the conditions if the outcome cannot be explained by the conditions used. However, in the present research, we recognize that courts apply their criteria in diff erent contexts. In other words, the fact-bound nature of court decisions means that diff erent facts may give rise to diff erent outcomes.7 Th us, given the exploratory nature of this research, contradictions are more valu- able than problematic as they help paint a more complete picture of casually relevant conditions underlying state action decisions. Accordingly, for all combinations that produced contradictions, we went back to the case text to fi nd additional factors that we may have overlooked that connect these cases. We fi nd that some cases involve issues of racial discrimination but others do not. Likewise, some contradiction cases involve First Amendment considerations and others do not. Yet it is also clear that some of the decision anomalies fi rst appear in cases involving civil rights (Van Alstyne and Karst 1961), another indication of the importance of context in explaining diff erent outcomes.
We also looked for commonalities by U.S. Supreme Court period to see whether we could link case outcomes to eras of conservative courts and/or liberal courts. Although there are some recognizable changes in standards by period, for example, during the tenures of the chief justices, these standards vary even more by individual justices on the Court than by a majority or minority ideology (see also Strickland 1991). Because inclusion of these factors in our analysis does not resolve contradictions, we conclude that there is no clear standard for fi nding state action. Th is is more evidence that state action cases can be reduced to a general range of criteria but not a set of standards, a fi nding somewhat inconsistent with that of legal scholars who refer to “tests” (e.g., Brown 2008; Elkind 1974) or theories (Choper 1979; Strickland 1991) for determining state action.
In the present research, contradictions also reveal information relevant to comparing social science approaches to the dichotomy to applications by the judiciary. For example, pfAb (table 3, row 5), we see two cases; the Gilmore (1974) case has outcome 1 (fi nding of state action), and the Polk case (1981) has outcome 0 (fi nding of no state action). Polk County v. Dodson appears to be an anomaly because it is the only case involving the direct act of a government employee that does not constitutive state action (note all other cases in the truth table with a lowercase p also have outcome = 1). In Polk, the Court determined that a public prosecutor’s actions were not attributable to the state even though the prosecutor was a government employee. Th e decision turned on consideration of the nature of the governmental connection to the underlying conduct (denoted by the capital A in the combination pfAb). Th e Court ulti- mately ruled that the prosecutor had discretion in carrying out his duties, and therefore the government was not intimately connected to the conduct. Our preliminary conclusion is supported by QCA: cases involving the direct act of a government offi cial will not always constitute state action. Inasmuch as the public administration litera- ture assumes that government employees are “public,” the threshold for defi ning an actor as public appears higher in the judiciary than it is in public administration.
The Public–Private Distinction: Insights for Public Administration from the State Action Doctrine 71
potential suppliers to determine ex ante whether potential suppliers’ policies and/or missions complement or run counter to democratic values. Managers can then specify contracts to include fair proce- dures to fi ll voids.
Diligently applied, set-theoretic methods can help researchers avoid some of the methodological pitfalls of small-N designs so they can tap into the rich information embedded in cases (Peters 1998, 44). Researchers from a wide range of disciplines appear to be catching on to the methods’ benefi ts (see Th eim and Dusa 2013 for a trend analysis of QCA applications and an overview of disciplines now benefi ting from the related software). However, as far as we can tell, this research represents the fi rst attempt in public administration to apply set-theoretic methods and confi guration logic to the question of what constitutes “public.” To date, research on the topic in public administration has been guided by simple frameworks that identify attributes of public or more recently, the quality of “publicness.” Using QCA, this research off ers systematic evidence that there is no single attribute that is necessary or suffi cient for diff erentiating public from private. Instead, the dichotomy rests on confi gurations of attributes, at least in the judicial context.
Th e fact that the U.S. Supreme Court has not applied a clear set of standards for fi nding state action may or may not be a concern for maintaining democratic norms. On the one hand, the absence of a standard may mean that decisions by the Court change to conform
to evolving societal values. Outcomes in race-based cases suggest this possibility. On the other hand, state action decisions may be purely ideological. Th at is, decisions may not actually refl ect the application of objec- tive criteria but rather the normative values of the judges. Viewed this way, the justices’ ideological positions are either embedded in the condition sets or the outcome. We have only scratched the surface of this question, but we believe more research can uncover an actual theory of “public,” even if that theory is in essence the uncovering of a value basis
for court decisions. Th is would require more extensive coding of the other case features, including the implications of each decision. For example, researchers could code decisions to identify the various competing interests associated with the case and to identify more specifi cally who benefi ts and who loses in the decision and whether specifi c governmental authority or individual rights are abridged or expanded by the decision.
Th ere are some limitations to this research. With respect to the comparison of the public administration literature with judicial applications, we acknowledge that we are comparing two diff erent contexts: public administration scholars are concerned with the characteristics of organizations that make them more eff ective and/ or more appropriate for delivering public services, whereas courts are mainly concerned with the boundaries for establishing constitu- tional due process. Th e two approaches also diff er: social scientists focus on what attributes make an actor/organization public, while the courts focus on whether an action can be fairly “attributable to the state” or whether one is acting “under color of the law.” However, we believe the court decisions do have relevance for public
suggests an ideal type that includes a governmental connection yet also acknowledges the impact of the actors’ activities. Accordingly, we propose the following maximum/ideal type consonant with the judicial applications and public administration literature, respectively:
JUDICIARY MAXIMUM: Public actors include those who “under the color of state law” cause the deprivation of a feder- ally protected right, typically but not always including those who have a close relationship with government and/or those who are involved in public functions.
PUBLIC ADMINISTRATION MAXIMUM: Public actors include employees of government as well as agents of govern- ment and others that are involved in the work of the citizenry.
Conclusion We believe our review of judicial decisions on the state action doc- trine is valuable in itself. To date, only a few scholars (Gilmour and Jensen 1998; Moe and Gilmour 1995; Rosenbloom and Piotrowski 2003; Verkuil 2007) have given attention to the constitutional and democratic norms associated with third-party governance decisions. At a minimum, public administration scholars and public managers should understand the reach of federal and state law in guarantee- ing fair treatment, and they should be aware that constitutional protections rarely apply in the privatization and contracting setting. Our QCA fi ndings support this conclusion: state action cases suggest a high threshold for determining state action. Th e courts have also established a high bar for determining what functions are “public functions.” Th is is par- ticularly salient given governments’ increasing reliance on contractors to operate bridges, municipal parking garages, state parks, and toll roads—functions that many public administration scholars assume are “public functions.” Likewise, constitutional due proc- ess is not guaranteed when governments con- tract with a business to provide social services or with a nonprofi t to serve at-risk youth. Th us, public managers should assume that the state action doctrine insulates these private actors from constitutional obligations that would apply to govern- ment if it performed the same functions.
However, contracting out traditional public functions and services does not mean the end of fairness. When the state action doctrine insulates private actors from constitutional obligations, there may be other avenues for extending constitutional protections that can be applied in designing and managing the contracting proc- ess. As Freeman suggests, instead of constraining the private role in governance, agencies can aim to “facilitate and direct it” (2003, 1289). Agencies can “privatize due process itself by eff ectively “enlist[ing] private actors in the project of protecting democratic norms” (1327). Th e contract negotiation process and the selection process provide opportunities for this. For example, nonprofi ts and charities that receive federal grants are required to adhere to some public norms as a condition of receiving funding or because of their tax status. Similarly, nonprofi t hospitals are often required to pursue public goals (Horwitz 2004). Knowing this, managers can scrutinize
Public administration scholars and public managers should
understand the reach of federal and state law in guaranteeing
fair treatment, and they should be aware that constitutional
protections rarely apply in the privatization and contracting
72 Public Administration Review • January | February 2015
Edmonson v. Leesville Concrete Co., 500 U.S. 614 (1991). Evans v. Abney, 396 U.S. 435 (1970). Evans v. Newton, 382 U.S. 296 (1966). Flagg Bros., Inc. v. Brooks, 436 U.S. 149 (1978). Gilmore v. City of Montgomery, 417 U.S. 556 (1974). Goldberg v. Kelly, 397 U.S. 254 (1970). Jackson v. Metropolitan Edison Co., 419 U.S. 345 (1974). Lombard v. Louisiana, 373 U.S. 267 (1963). Lovell v. City of Griffi n, 303 U.S. 444 (1938). Lugar v. Edmondson Oil Co., 457 U.S. 922 (1982). Marsh v. State of Alabama, 326 U.S. 501 (1946). Moose Lodge No. 107 v. Irvis, 407 U.S. 163 (1972). National Collegiate Athletic Association v. Tarkanian, 488 U.S. 179 (1988). Nixon v. Condon, 286 U.S. 73 (1932). Pennsylvania v. Board of Directors of City Trusts, 353 U.S. 230 (1957). Polk County v. Dodson, 454 U.S. 312 (1981). Public Utilities Commission of District of Columbia v. Pollak, 343 U.S. 451 (1952). Reitman v. Mulkey, 387 U.S. 369 (1967). Rendell-Baker v. Kohn, 457 U.S. 830 (1982). Rogers v. Alabama, 192 U.S. 226 (1904). San Francisco Arts & Athletics v. United States Olympic Committee, 483 U.S. 522 (1987). Shelley v. Kraemer, 334 U.S. 1 (1948). Smith v. Allwright, 321 U.S. 649 (1944). Terry v. Adams, 345 U.S. 461 (1953). West v. Atkins, 487 U.S. 42 (1988).
References Alchian, Armen. 1967. Cost Eff ectiveness of Cost Eff ectiveness. In Defense Management,
edited by Stephen Enke, 74–86. Englewood Cliff s, NJ: Prentice Hall. Allison, Graham T. 1980. Public and Private Management: Are Th ey Fundamentally
Alike in All Unimportant Respects? In Classics of Public Administration, 3rd ed., edited by Jay M. Shafritz and Albert C. Hyde, 457–75. Pacifi c Grove, CA: Brooks/Cole, 1992.
Antonsen, Marianne, and Torben Beck Jørgensen. 1997. Th e “Publicness” of Public Organizations. Public Administration 75(2): 337–57.
Appleby, Paul H. 1945. Big Democracy. New York: Alfred A. Knopf. Banfi eld, Edward C. 1976. Corruption as a Feature of Government Organization.
Journal of Law and Economics 18(3): 587–606. Benington, John. 2011. From Private Choice to Public Value? In Public Value: Th eory
and Practice, edited by John Benington and Mark H. Moore, 31–51. New York: Palgrave Macmillan.
Blau, Peter M., and W. Richard Scott. 1963. Formal Organizations: A Comparative Approach. San Francisco: Chandler.
Boston, Jonathan, John Martin, June Pallot, and Pat Walsh. 1996. Public Management: Th e New Zealand Model. Auckland: Oxford University Press.
Box, Richard C., and Cheryl Simrell King. 2000. Th e “T”ruth Is Elsewhere: Critical History. Administrative Th eory and Praxis 22(4): 751–71.
Box, Richard C., Gary S. Marshall, B. J. Reed, and Christine M. Reed. 2001. New Public Management and Substantive Democracy. Public Administration Review 61(5): 608–19.
Bozeman, Barry. 1984. Dimensions of Publicness: An Approach to Public Organization Th eory. In New Directions in Public Administration, edited by Barry Bozeman and Jeff rey Straussman, 46–62. Monterey, CA: Brooks/Cole.
———. 1987. All Organizations Are Public: Bridging Public and Private Organizational Th eories. San Francisco: Jossey-Bass.
administration scholars who grapple with the public–private distinc- tion: when courts rule an action is fairly attributable to the state, the actor is considered a mere extension of the state for constitutional purposes—eff ectively a public, not a private, entity.
As for the QCA application and fi ndings, we could not determine the weight of considerations from the court decisions. However, the language of the Supreme Court suggests that some factors were more important in some (but not all) cases. QCA is also a tool for fuzzy set analysis (Ragin 2000) and there may be opportunities for fi ne-tuning the analysis with fuzzy set coding if suitable ordinal scales could be developed for the diff erent considerations. As another extension of this research, United States Circuit Court cases could be added to the sample to determine whether the vari- ous circuit courts have been consistent in their fi ndings or whether they have established threshold requirements for the diff erent categories.
Notes 1. Th e authors thank one of the anonymous reviewers for suggesting the use of
problematization for distilling the literature in this section. 2. Rainey (2009, chap. 3) provides a thorough review of the distinction literature
and related research. 3. Government compulsion, government’s encouragement of the conduct, and
questions of joint participation all involve questions of government’s complicity in the action or conduct and thus are merged into one condition. Court decisions that focus on government control of another actor, intertwinement, and symbi- otic relationships all involve government’s relationship with the actor (as opposed to the conduct) and thus are merged into one condition. Finally, we dropped the condition “countervailing circumstances” discussed in law journals, as it was not clearly linked to a court case.
4. To conserve space, we omit the coincidence matrix with coverage statistics pro- duced with Stata code.
5. Th e alternative notation is ~p *~f *~a *~b + p * ~f * a * ~b, where negation represents the absence of the condition; * is the logical and; and + is logical the or. Th e reduction equation is reached using the Quine-McCluskey algorithm, a procedure described by Ragin (1987, 85–124).
6. Although our analysis is based on crisp set dichotomous coding, QCA also includes tools for fuzzy set analysis, which we cannot use because our conditions do not have a clear ordinal quality to them.
7. On the other hand, when the range of general conditions considered changes from decision to decision, we have to ask, is there really a condition set or are decisions actually value judgments, where justices interpret the facts to fi t their ideological position?
Cases Cited Adickes v. S. H. Kress & Co., 398 U.S. 144 (1970). AFL v. Swing, 312 U.S. 321 (1940). American Manufacturers Mutual Insurance v. Sullivan, 526 U.S. 40 (1999). Blum v. Yaretsky, 457 U.S. 991 (1982). Brentwood Academy v. Tennessee Secondary School Athletic Assn., 531 U.S. 288 (2001). Burton v. Wilmington Parking Authority, 365 U.S. 715 (1961). Carter v. Texas, 177 U.S. 442 (1900). Chicago, Burlington, & Quincy Railroad Company v. City of Chicago, 166 U.S. 226 (1897). Columbia Broadcasting System, Inc. v. Democratic National Committee, 412 U.S. 94 (1973).
The Public–Private Distinction: Insights for Public Administration from the State Action Doctrine 73
Hart, H. L. A. 1954. Defi nition and Th eory of Jurisprudence. Law Quarterly Review 70: 37.
Hinrichs, Harley H., and Graeme M. Taylor. 1972. Systematic Analysis: A Primer on Benefi t-Cost Analysis and Program Evaluation. Pacifi c Palisades, CA: Goodyear.
Horwitz, Jill R. 2004. Why We Need the Independent Sector: Th e Behavior Law and Ethics of Not-for-Profi t Hospitals. Research Paper no. 3, Law and Economics Working Papers, Olin Center for Law and Economics, University of Michigan.
Jensen, Michael C., and William H. Meckling. 1976. Th eory of the Firm. Journal of Financial Economics 3(4): 305–60.
Kettl, Donald F. 2002. Th e Transformation of Governance: Public Administration for Twenty-First Century America. Baltimore: Johns Hopkins University Press.
Kvist, Jon. 2006. Measuring the Welfare State—Concepts, Ideal Types and Fuzzy Sets in Comparative Analysis. http://www.compasss.org/wpseries/Kvist2006.pdf [accessed August 4, 2014].
Lawler, Edward E. 1971. Pay and Organizational Eff ectiveness: A Psychological Review. New York: McGraw-Hill.
Longest, Kyle C., and Stephen Vaisey. 2008. Fuzzy: A Program for Performing Qualitative Comparative Analyses (QCA) in Stata. Stata Journal 8(1): 79–104.
Lowi, Th eodore. 1969. Th e End of Liberalism: Ideology, Policy, and the Crisis of Public Authority. New York: W. W. Norton.
Madry, Alan R. 2001. Statewide School Athletic Associations and Constitutional Liability: Brentwood Academy v. Tennessee Secondary School Athletic Association. Marquette Sports Law Review 12(1): 364–95.
Mainzer, Lewis C. 1973. Political Bureaucracy. Glenview, IL: Scott, Foresman. Meyer, Marshall W. 1972. Bureaucratic Structure and Authority: Coordination and
Control in 254 Government Agencies. New York: Harper & Row. Meynhardt, Timo. 2009. Public Value Inside: What Is Public Value Creation?
International Journal of Public Administration 32(3–4): 192–219. Mintzberg, Henry. 1973. Th e Nature of Managerial Work. New York: Harper & Row. Moe, Ronald C. 1994. Th e “Reinventing Government” Exercise: Misinterpreting the
Problem, Misjudging the Consequences. Public Administration Review 54(2): 111–22.
Moe, Ronald C., and Robert S. Gilmour. 1995. Rediscovering Principles of Public Administration: Th e Neglected Foundation of Public Law. Public Administration Review 55(2): 135–46.
Moore, Mark H. 1995. Creating Public Value: Strategic Management in Government. Cambridge, MA: Harvard University Press.
———. 2014. Public Value Accounting: Establishing the Philosophical Basis. Public Administration Review 74(4): 465–77.
Moulton, Stephanie. 2009. Putting Together the Publicness Puzzle: A Framework for Realized Publicness. Public Administration Review 69(5): 889–900.
Niskanen, William A., Jr. 1971. Bureaucracy and Representative Government. Chicago: Aldine, Atherton.
Osborne, David, and Ted Gaebler. 1992. Reinventing Government: How the Entrepreneurial Spirit Is Transforming the Public Sector. Reading, MA: Addison-Wesley.
Perry, James L. 1986. Merit Pay in the Public Sector: Th e Case of a Failure of Th eory. Review of Public Personnel Administration 7(1): 57–69.
———. 1996. Measuring Public Service Motivation: An Assessment of Construct Reliability and Validity. Journal of Public Administration Research and Th eory 6(1): 5–24.
———. 2000. Bringing Society In: Toward a Th eory of Public-Service Motivation. Journal of Public Administration Research and Th eory 10(2): 471–88.
Perry, James L., and Hal G. Rainey. 1988. Th e Public–Private Distinction in Organization Th eory: A Critique and Research Strategy. Academy of Management Review 13(2): 182–201.
Peters, B. Guy. 1998. Comparative Politics: Th eory and Methods. New York: New York University Press.
———. 2007. Public Values and Public Interest: Counterbalancing Economic Individualism. Washington, DC: Georgetown University Press.
Bozeman, Barry, and Stuart Bretschneider. 1994. Th e “Publicness Puzzle” in Organization Th eory: A Test of Alternative Explanations of Diff erences between Public and Private Organizations. Journal of Public Administration Research and Th eory 4(2): 197–223.
Brown, Julie K. 2008. Less Is More: Decluttering the State Action Doctrine. Missouri Law Review 73(2): 560–81.
Buchanan, Bruce. 1975. Government Managers, Business Executives, and Organizational Commitment. Public Administration Review 35(4): 339–47.
Burkhead, Jesse, and Jerry Miner. 1971. Public Expenditure. Chicago: Aldine, Atherton.
Caiden, Gerald E. 1971. Th e Dynamics of Public Administration: Guidelines to Current Transformation in Th eory and Practice. New York: Holt, Rinehart and Winston.
Charlesworth, James C., ed. 1968. Th eory and Practice of Public Administration: Scope, Objectives, and Methods. Philadelphia: American Academy of Political and Social Science.
Choper, Jesse H. 1979. Th oughts on State Action: Th e “Government Function” and “Power Th eory” Approaches. Washington University Law Review 3(8): 757–82.
Cole, Martin, and Greg Parston. 2006. Unlocking Public Value: A New Model for Achieving High Performance in Public Service Organizations. Hoboken, NJ: Wiley.
Collier, David, and Steven Levitsky. 1997. Democracy with Adjectives. World Politics 49(3): 430–51.
Cooper, Phillip J. 2003. Governing by Contract: Challenges and Opportunities for Public Managers. Washington, DC: CQ Press.
Coursey, David, and Barry Bozeman. 1990. Decision Making in Public and Private Organizations: A Test of Alternative Concepts of “Publicness.” Public Administration Review 50(5): 525–35.
Dahl, Robert A., and Charles E. Lindblom.1953. Politics, Economics, and Welfare: Planning and Politico-Economic Systems Resolved into Basic Social Processes. New York: Harper.
Denhardt, Robert B., and Janet V. Denhardt. 2000. Th e New Public Service: Serving Rather than Steering. Public Administration Review 60(6): 549–59.
Dewey, John. 1927. Th e Public and Its Problems. New York: Henry Holt. Downs, Anthony. 1967. Inside Bureaucracy. Boston: Little, Brown. Elkind, David S. 1974. State Action: Th eories for Applying Constitutional
Restrictions to Private Activity. Columbia Law Review 74(4): 656–705. Frederickson, H. George. 1980. New Public Administration. University: University of
Alabama Press. Freeman, Jody. 2003. Extending Public Law Norms through Privatization. Harvard
Law Review 116(5): 1285–1352. Gawthrop, Louis C. 1971. Administrative Politics and Social Change. New York: St.
Martin’s Press. Gerring, John, and Paul A. Barresi. 2003. Putting Ordinary Language to Work:
A Min-Max Strategy of Concept Formation in the Social Sciences. Journal of Th eoretical Politics 15(2): 201–32.
Gilmour, Robert S., and Laura S. Jensen. 1998. Reinventing Government Accountability: Public Functions, Privatization and the Meaning of “State Action.” Public Administration Review 58(3): 247–58.
Golembiewski, Robert T. 1969. Organization Development in Public Agencies: Perspectives on Th eory and Practice. Public Administration Review 29(4): 367–68.
Goodsell, Charles T. 1994. Th e Case for Bureaucracy: A Public Administration Polemic. 3rd ed. Chatham, NJ: Chatham House.
Haas, J. E., R. H. Hall, and N. J. Johnson. 1966. Toward an Empirically Derived Taxonomy of Organizations. In Studies on Behavior in Organizations, edited by Raymond V. Bowers, 157–80. Athens: University of Georgia Press.
74 Public Administration Review • January | February 2015
Pollitt, Christopher. 1993. Managerialism and the Public Services: Cuts or Cultural Change in the 1990s? 2nd ed. Cambridge, MA: Blackwell.
Ragin, Charles C. 1987. Th e Comparative Method: Moving beyond Qualitative and Quantitative Strategies. Berkeley: University of California Press.
———. 2000. Fuzzy-Set Social Science. Chicago: University of Chicago Press. ———. 2008. Redesigning Social Inquiry: Fuzzy Sets and Beyond. Chicago: University
of Chicago Press. Ragin, Charles C., and Sean Davey. 2012. fs/QCA [computer program], v. 2.5. Irvine:
University of California. Rainey, Hal G. 1979. Perceptions of Incentives in Business and Government:
Implications for Civil Service Reform. Public Administration Review 39(5): 440–48.
———. 1982. Reward Preferences of Public and Private Managers: In Search of the Service Ethic. American Review of Public Administration 16(4): 288–302.
———. 1983. Public Agencies and Private Firms: Incentive Structures, Goals, and Individual Roles. Administration & Society 15(2): 207–42.
———. 1993. Toward a Th eory of Goal Ambiguity in Public Organizations. In Research in Public Administration, vol. 2, edited by James L. Perry, 121–66. Greenwich, CT: JAI Press.
———. 2009. Understanding and Managing Public Organizations. 4th ed. San Francisco: Jossey-Bass.
Rainey, Hal G., Robert W. Backoff , and Charles H. Levine. 1976. Comparing Public and Private Organizations. Public Administration Review 36(2): 233–44.
Rainey, Hal G., Sanjay Pandey, and Barry Bozeman. 1995. Public and Private Manager’s Perceptions of Red Tape. Public Administration Review 55(6): 567–74.
Rawls, James R., Robert A. Ullrich, and Oscar T. Nelson. 1975. A Comparison of Managers Entering or Reentering the Profi t and Nonprofi t Sectors. Academy of Management Journal 18(3): 616–22.
Rosenbloom, David H. 1993. Editorial: Have an Administrative Rx? Don’t Forget the Politics! Public Administration Review 53(6): 503–7.
Rosenbloom, David H., and Suzanne J. Piotrowski. 2003. Outsourcing Constitutional and Administrative Law Norms. American Review of Public Administration 33(1): 1–19.
Rosenbloom, David H., Robert S. Kravchuk, and Richard M. Clerkin. 2009. Public Administration:. Understanding Management, Politics, and the Law in the Public Sector. 7th ed. Boston: McGraw-Hill.
Savas, E. S. 1987. Privatization: Th e Key to Better Government. Chatham, NJ: Chatham House.
———. 2000. Privatization and Public–Private Partnerships. New York: Chatham House.
Schneider, Carsten Q., and Claudius Wagemann. 2010. Standards of Good Practice in Qualitative Comparative Analysis (QCA) and Fuzzy Sets. Comparative Sociology 9(3): 397–418.
Schultze, C. L. 1970. Th e Role of Incentives, Penalties, and Rewards in Attaining Eff ective Policy. In Public Expenditures and Policy Analysis, edited by Robert H. Haveman and Julius Margolis, 145–72. Chicago: Markham.
Seidman, Harold. 1970. Politics, Position, and Power: Th e Dynamics of Federal Organization. New York: Oxford University Press.
Shafritz, Jay M., Albert C. Hyde, and Sandra J. Parkes, eds. 2004. Classics of Public Administration. Belmont, CA: Wadsworth/Th omson Learning.
Siffi n, William J. 1963. Business Administration ≠ Public Administration. Business Horizons 5(4):69–76.
Simon, Herbert A., Victor A Th ompson, and Donald W. Smithburg. 1950. Public Administration. New York: Alfred A. Knopf.
Stahl, O. Glenn. 1971. Public Personnel Administration. New York: Harper & Row. Strickland, Henry C. 1991. Th e State Action Doctrine and the Rehnquist Court.
Hastings Constitutional Law Quarterly 18: 587–666. Th eim, Alrik, and Adrian Dusa. 2013. QCA: A Package for Qualitative Comparative
Analysis. Th e R Journal 5(1): 87–97. Th ompson, James D. 1967. Organizations in Action: Social Science Bases of
Administrative Th eory. New York: McGraw-Hill. Van Alstyne, William W., and Kenneth L. Karst. 1961. State Action. Stanford Law
Review 14(1): 3–58. Varone, Frederic, Benoit Rihoux, and Axel Marx, A. 2006. A New Method
for Policy Evaluation? Longstanding Challenges and the Possibilities of Qualitative Comparative Analysis (QCA). In Innovative Comparative Methods for Policy Analysis, edited by Benoit Rihoux and H. Grimm, 213–36. New York: Springer.
Verkuil, Paul R. 2005. Privatizing Due Process. Administrative Law Review 57(4): 963–94.
———. 2007. Outsourcing Sovereignty: Why Privatization of Government Functions Th reatens Democracy and What We Can Do about It. New York: Cambridge University Press.
Volcker, Paul A. 2014. Vision without Execution Is Hallucination. Public Administration Review 74(4): 439–41.
Wamsley, Gary L., and Mayer N. Zald. 1973. Th e Political Economy of Public Organizations: A Critique and Approach to the Study of Public Administration. Lexington, MA: Lexington Books.
Weber, Max. 1949. Th e Methodology of the Social Sciences. Translated and edited by Edward A. Shils and Henry A. Finch. New York: Free Press.
Weidenbaum, Murray L. 1969. Th e Modern Public Sector: New Ways of Doing Government’s Business. New York: Basic Books.
Weiss, H. L. 1983. Why Business and Government Exchange Executives. In Public Management, edited by James L. Perry and Kenneth L. Kraemer, 41–54. Palo Alto, CA: Mayfi eld.
Williamson, Oliver E. 1975. Markets and Hierarchies, Analysis and Antitrust Implications: A Study in the Economics of Internal Organization. New York: Free Press.
———. 1985. Th e Economic Institutions of Capitalism: Firms, Markets, Relational Contracting. New York: Simon & Schuster.
———. 1996. Th e Mechanisms of Governance. New York: Oxford University Press.
———. 1999. Public and Private Bureaucracies: A Transaction Cost Economics Perspective. Journal of Law, Economics, and Organization 15(1): 306–42.
Wilson, James Q. 1989. Bureaucracy: What Government Agencies Do and Why Th ey Do It. New York: Basic Books.
Woll, Peter. 1963. American Bureaucracy. New York: W. W. Norton.
Copyright of Public Administration Review is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder’s express written permission. However, users may print, download, or email articles for individual use.