Journal of Public Administration Research and Theory

Journal of Public Administration Research and Theory Advance Access originally published online on September 14, 2006
Journal of Public Administration Research and Theory 2007 17(2):157-187; doi:10.1093/jopart/mul012

This Article Abstract FREE Full Text (PDF) All Versions of this Article: 17/2/157    most recent mul012v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (2) Request Permissions Google Scholar Articles by Van Slyke, D. M. Search for Related Content Social Bookmarking          What's this?

© The Author 2006. Published by Oxford University Press on behalf of the Journal of Public Administration Research and Theory, Inc. All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org.

Agents or Stewards: Using Theory to Understand the Government-Nonprofit Social Service Contracting Relationship

David M. Van Slyke

Department of Public Administration, The Maxwell School at Syracuse University

Address correspondence to the author at vanslyke{at}maxwell.syr.edu.

	Abstract

TOP Abstract INTRODUCTION BACKGROUND METHODS AND DATA FINDINGS: AGENTS OR STEWARDS? CONCLUSION References Notes

 
Using agency and stewardship theories, this study examines how public administrators manage contracting relationships with nonprofit organizations. Interviews were conducted with public and nonprofit managers involved in social services contract relationships at the state and county level in New York State. The use of trust, reputation, and monitoring as well as other factors influence the manner in which contract relationships are managed. The findings suggest that the manner in which nonprofits are managed evolves over time from a principal-agent to a principal-steward relationship but with less variance than the theories would suggest. This results in part from the contextual conditions that include the type of service, lack of market competitiveness, and management capacity constraints. The intergovernmental environment in which social services are implemented and delivered presents complex challenges for public managers responsible for managing contract relationships. The findings from this study document those challenges and the corresponding management practices used with nonprofit contractors.

	INTRODUCTION

TOP Abstract INTRODUCTION BACKGROUND METHODS AND DATA FINDINGS: AGENTS OR STEWARDS? CONCLUSION References Notes

 
Scholars and practitioners within the public management, public policy, and nonprofit communities have called for theories that better explain the nature of and variation in the government-nonprofit contracting relationship and the implications of contracting for public management (Cooper 2003; DeHoog 1984; Donahue 1989; Johnston and Romzek 1999; Kettl 1993; Saidel 1991; Salamon 1989; Shleifer and Vishny 1998; Smith and Lipsky 1993). This call stems largely from the effects of devolution and privatization on government service provision. Philosophically, policy makers are enacting decisions that not only restrict but in many cases remove government from providing services directly to citizens. In part this is being driven by market and political ideologies that have their roots in perceptions about greater efficiencies and innovations arising from the private sector and broader support for smaller government. Thus, contextual changes at the national and local level have led to a transformation from governance by authority to governance by contract. They have also created conditions under which it is all the more important to think about problems of public management and therefore investigate alternative models for effectively managing government's relationships with its contractors (Cooper 2003).

Privatization, as both a tool and process, involves "changing from an arrangement with high government involvement to one with less" (Savas 1987, 88). Privatization advocates argue that government will receive better services at lower costs because of the expertise and innovation of private providers. This argument rests on "introduc[ing] competition and market forces in [to] the delivery of public services" (Savas 2000, 122). The most frequently used form of privatization in the United States is contracting with a third party for the production of goods and provision of services. Some of the leading scholars involved in efforts to understand the effectiveness of market-based alternatives to government provision have focused on the extent to which markets can be developed, contracts designed, and contractors incentivized to better capitalize on the advantages associated with contracting as a result of competition (Cooper 2003; Goldsmith 1997; Heinrich 2000, 2002; Savas 2005). However, for government to realize the advantages associated with privatization, public managers need to have expertise in contract management and the ability to negotiate, monitor, and communicate expectations and technical information (Kettl 1993; Van Slyke 2003). Yet, Kettl suggests that policy makers "look at puzzles like contract management [by racing] past the details to get to what they see as the real issues ... Details about how to manage contracts are unimportant, to be left to third-level administrators to sort out" (Cooper 2003, xi). Kelman (2002a, 90) similarly notes that "the administration of contracts once they have been signed has been the neglected stepchild of these [contract management] efforts." Consequently, the quality of public services delivered by contractors depends largely on the quality of contract management provided by public managers. While this may not seem surprising, much of the research on contracting has been on issues other than how public managers manage their contractors. There has also been no shortage of media-sensationalized government-contractor corruption and abuse stories¹ and a litany of Government Accountability Office (GAO) recommendations for strengthening different aspects of government contract management.² As previous research suggests our intergovernmental system of publicly financed and privately produced and delivered services could benefit from more public management attention and scholarship on the contract environment and the manner in which the contracting relationship is managed.

In this study, I use agency and stewardship theories to examine the manner in which the government-nonprofit social services contracting relationship is managed. Each theory focuses on using tools, such as monitoring, trust, reputation, incentives, and sanctions in contract relationships in order to achieve goal alignment between the parties to the contract. Agency theory has been described as the central approach to a theory of managerial behavior (Ross 1987), whereas stewardship theory has been framed as the organizational behavior counterweight to rational action theories of management (Davis, Donaldson and Schoorman 1997a). The theories however start with different initial assumptions about managing contractual relations. Agency theory assumes goal divergence on the part of the contracted agent and stewardship theory assumes convergence because of shared collective interests with the contracted steward. Three questions are examined in this study: First, what public management contract practices are applied to nonprofits in the contracting relationship? Second, what conditions affect the public management contract practices applied to nonprofit organizations in the contracting relationship? Third, to what extent are these management practices consistent with the tenets of agency and stewardship theories? Given the lack of theory used to explain contract management practices for social services, this research is important considering government's extensive use of contracting. To begin, a substantive review of the government-nonprofit contracting relationship, agency theory, and stewardship theory is presented. The methods and data follow, the findings presented, and public management implications are discussed in the conclusion.

	BACKGROUND

TOP Abstract INTRODUCTION BACKGROUND METHODS AND DATA FINDINGS: AGENTS OR STEWARDS? CONCLUSION References Notes

 
Government-Nonprofit Social Service Contracting Relationships
Social services are often characterized along several dimensions by:

• Varying levels of tractable and intractable client problems
  
• Clients that require specialized treatment expertise, sometimes over long time periods, and who have varying levels of motivation
  
• Markets consisting of varying levels of competition
  
• Legislatively mandated services with imposed time and funding constraints for program development and implementation
  
• Programs that consist of ambiguous policy directives
  

Social services are complex services for which government may have some expertise but often requires additional expertise that it must either hire or contract for in order to treat and serve clients. Social services are prime candidates to be privatized in part because alternatives to government provision do exist, through the use of nonprofits and increasingly for-profit firms for service provision.³ There is also generally strong support among policy makers to follow through on historical precedent to remove government from providing services that nongovernmental organizations already or can potentially provide. There are a number of challenges for government in providing social services through contractual arrangements. I identify four broad categories of challenges here that have frequently been cited.

First, there is often a lack of competition by geographic market (rural, suburban, urban) and service type (refugee resettlement, substance abuse, and addiction programs) that makes it difficult for public managers to correct for supply-side imperfections (too few market providers) and thus limits their use of contract termination and rebidding as a management strategy (DeHoog 1984, 1990; Johnston and Romzek 1999).⁴ Second, previously noted ideological motives to contract can also contribute to a lack of administrative capacity in government agencies. This can have the effect of limiting the ability of public managers to develop competition, solicit bids, rebid contracts, develop performance measures, and monitor and hold contractors accountable for contract goals, service quality, and client satisfaction. An increasingly persistent problem for government agencies is having too few personnel with contract management and policy and program expertise to manage the contract process (Kelman 2002a; Meyers, Glaser, and MacDonald 1998; Milward and Provan 2000; Romzek and Johnston 2002; Smith and Smyth 1996; Van Slyke 2003).⁵ Third, policy directives, policy goals, and corresponding program requirements are often ambiguously defined and infrequently monitored, creating contract implementation situations in which public managers can find it difficult to evaluate the frequency, consistency, and quality of service delivery among their contractors.⁶ The attributes of social services thus afford or require, depending on your perspective, that public managers use discretion in the implementation of services. This can lead to goal divergence between policy directives and implementation practices, presenting genuine accountability concerns for public managers (Meyers, Riccucci, and Lurie 2001; Riccucci 2005; Sandfort 2000). And, fourth, contracting relationships between government and nonprofit organizations can have the unintended effect of altering nonprofit governance practices, causing mission drift, deprofessionalization of staff, and contributing to a position of government funding dependency (Alexander, Nank, and Stiffers 1999; Kramer 1994; Saidel 1991).

Notwithstanding these challenges, however, there have been a number of developments in government's contracting with nonprofits for social services in certain markets and for specific types of services. In large cities, such as New York City, there has been a significant effort to develop and stimulate competition through more aggressive outreach by public managers to nonprofit and for-profit providers using a variety of contract solicitation tools (Savas 2002, 2005). The use of performance-based contracts has also increased especially at the state level in an effort to incentivize contractors and ensure alignment with program and funding goals and targets (Frumkin 2001; Martin 2004).⁷ This has been the case in a number of states⁸ and in service areas such as workforce investment training, rehabilitative services, and adoption services. At the federal level, there have been a number of contract management innovations for procuring a broad range of goods and services as a result of changes to A-76, federal acquisition regulations (FARs), and competitive sourcing requirements. Other contract innovations are being used across federal agencies, but remain in their infancy in terms of widespread use, such as share in savings contracts, best value contracting, and a variety of strategic sourcing arrangements (Kelman 2002b).⁹

Although there has been considerable progress in some areas of contracting, two important factors remain in considering how contracts are to be developed and managed. They are the environment in which the contracted service exists and the specific nature of that service and whether it is deemed to be inherently governmental. There are significant differences among contracted services, such as social services and refuse collection, in terms of the market and political ideology supporting or opposing alternative service delivery arrangements, the level of market competition that exists, and the ease of defining, measuring, and observing outputs and outcomes (See Brown and Potoski 2004). In addition, the complexity of managing networked providers involved in joint production and delivery efforts for bundled services (such as drug treatment, vocational education, mental health, and housing supports), as opposed to overseeing a single contractor for a single service (refuse collection), is an important factor to consider as is the ability to monitor contractor performance and service quality (Milward and Provan 2000). In addition "buying" management activities through contracts is more accessible for single services that are more easily measured and monitored, such as refuse collection, than for more integrated and complex services, such as social and mental health services (Brown and Potoski 2004). Despite these challenges, contract relationships are developed and contractors managed, not simply because they may be economically feasible but just as often because they are politically popular (Donahue 1989; Kettl 1993; Milward and Provan 2000; Sclar 2000; Smith and Lipsky 1993). Yet, state and local governments continue to contract for services best characterized as incomplete contracts in imperfect markets.

As a result political ideology and uncertainty can envelop the make or buy decision. The question then is how do public managers manage contracts with nonprofits for programs like social services in which performance is not always easily observed and measured, where there may be fewer provider alternatives given supply-side imperfections in local markets, demand-side imperfections (in that government is the sole buyer of services) that create monopsonistic pressures in markets, and competing positions regarding whether a service is or is not inherently governmental.¹⁰ Public administration scholarship has for some time asserted that research on this relationship needs to move beyond the single organizational case study, be linked to social science theory, and consist of multimethod, multisite examinations (Kettl and Milward 1996; Lynn 1996). To advance public management research and knowledge about the manner and variation in which public managers manage their contractors, theories must be tested and empirical data collected if the government-nonprofit relationship is to be more fully understood in all its complexity, variation, and sophistication. A brief review of agency and stewardship theories is presented in order to better understand this type of contract scenario and the question of what types of conditions give rise to which contract practices in managing the government-nonprofit contract relationship.

Agency Theory
Agency theory, also frequently referred to as the principal-agent model, is used in the organizational economics and management literature as a theoretical framework for structuring and managing contract relationships and to explain the behaviors of principal and agent.¹¹ As a theory the focus is on accountability by correcting for opportunistic behavior that can result from exploiting asymmetric information (Alchian and Demsetz 1972; Eisenhardt 1989; Jensen and Meckling 1996; Sappington 1991). The principal-agent model has been applied extensively to a range of contractual relations between organizations, boards and directors, and managers and employees in for-profit, public, and nonprofit organizations (Bogart 1995; Brody 1996; Dharwadkar, George, and Brandes 2000; Lee and O'Neill 2003; McGubbins, Noll, and Weingast 1987; Wood and Waterman 1991).

In agency theory, a principal chooses to contract with an agent for reasons of cost and expertise. The principal may decide that their organization lacks the expertise or resources required to produce a good or service and that the cost of hiring or developing that expertise in-house exceeds the costs associated with contracting for the expertise. The principal and agent agree on the terms of the contract including the inputs, processes, outcomes, quality and satisfaction parameters, monitoring and performance-reporting requirements, how the agent is to be compensated for doing the work of the principal, and the sanctions that will result if the principal detects the agent pursuing his/her own goals above the principal's objectives. The main tenets of agency theory focus on information asymmetry (when one party has information the other party does not possess), adverse selection or precontractual opportunism (when one party knows more about attributes of a product or service than another and, as a result, the uninformed party runs the risk of purchasing a product or service of low quality), and moral hazard or postcontractual opportunism (when a party to the contract uses information and expertise and acts opportunistically, in its own self-interest, to the exclusion of the agreed upon contract goals).¹² Two assumptions characterize the principal-agent model: (1) there is goal conflict between the power (bureaucratic and political) and budget/wealth-maximizing behavior of the principal and the utility-maximizing behavior of the agent and (2) agents have more information than principals, which agents can exploit for self-gain rather than for the collective interests of the contracting parties leading to moral hazard problems. Important to the agency argument is the issue of uncertainty and the costs associated with measuring agent behavior and the outcomes produced.

To correct for the inefficiencies that can arise from contracts because of the aforementioned assumptions (namely the unobservable behavior of the agent due to adverse selection and moral hazard) and to align the actions of the agent with the goals of the principal, a mix of incentives, sanctions, information systems (such as reporting procedures), and monitoring mechanisms are employed.¹³ Frey (1993) suggests that the monitoring intensity—frequency, formality, and precision of the performance criteria—as well as the use of these other contractual tools can serve to ensure alignment but can also be perceived as distrust that may lead agents to reduce their work efforts.

Critics of the principal-agent model, such as Perrow (1986) and Donaldson (1990), have argued that the model is one sided because it negatively characterizes an individual agent's moral and collective behavior as self-seeking and focused on obtaining power and wealth; ignores worker loyalty, pride, and identification with the organization's mission and goals; and omits opportunistic behavior by principals. Waterman and Meier (1998) assert that a weakness of the model is that the interaction of these variables with other variables is often discounted in terms of how contract relationships may evolve from one form to another, such as hierarchical and legal to cooperative and interdependent relationships. A third criticism is that agency theory is an inappropriate framework for contractual services that are not easily measured and observed. Yet, as Eisenhardt (1989, 71) notes "agency theory is most relevant in situations in which contracting problems are difficult" in terms of goal conflict, measurability, and observability.

Despite its critics, agency theory has been usefully applied to understanding a variety of organizational phenomena. Eisenhardt (1989) suggests that the assumptions can be relaxed depending on the context of the contract, the length of time the contracting parties have been engaged, the programmability of the contracted service/product, and the level of conflict that may exist between the parties. This observation and recommendation is not divergent with the theory but rather is an overlooked flexibility of the model and one that is underinvestigated in previous research. Thus, we would expect that principals who provide clear incentives, such as contract renewal and stability, discretion and flexibility in program implementation, reputational enhancement, and vigilant monitoring through a myriad of formal and informal mechanisms, would experience less agent opportunism and contract goal divergence. By investing at the outset of the contractual relationship in the use and enforcement of agency mechanisms, the principal can increase the opportunities for achieving contractual alignment and, over time, lessen the managerial transaction costs associated with contracting out. For the agent, convergence with the principal's goals can over time lead to trust and reputational enhancement as well as less monitoring and fewer reporting requirements. An evolved principal-agent relationship based on alignment may also lead to contractual stability and therefore fewer instances of bid letting. For purposes of this research, public managers representing government social services agencies will be referred to as principals and executive directors of nonprofit organizations as agents.¹⁴

Eisenhardt (1989, 71) notes that the application of agency theory should be "expanded to [include] a richer and more complex range of contexts." The government-nonprofit social services contracting relationship certainly provides a context in which a decision to contract can be as politically motivated as market driven.¹⁵ In such a context principals may lack the ability to renegotiate or terminate contractors because of market imperfections, outcomes that can be difficult to measure and evaluate, and agents which have different goals and preferences than their principals. Such principal-agent differences may be based less on resource maximization and more on mission and programmatic philosophies and organizationally stylized modalities of treatment intervention.

One of the criticisms leveled by Eisenhardt (1989, 64 and 71) about much of the organizational research done today is that it is "topic centered" rather than "theory centered" and that organizational scholarship would be enhanced if more focus were on "theory-relevant contexts." She goes on to suggests that "it seems reasonable to urge the adoption of an agency theory perspective when investigating the many problems that have a principal-agent structure" (1989, 70) and that it is "an empirically valid perspective, particularly when coupled with complementary perspectives" (57). The government-nonprofit social services contracting relationship qualifies as an important topical area of public management that has been addressed but in which analysis of contract management practices could benefit from a more substantial theory-grounded examination. A complementary theory for examining managerial behavior and goal alignment between contracting parties is stewardship theory.

Stewardship Theory
Stewardship theory examines relationships and behaviors often discounted in organizational economic theories, emphasizing collective, pro-organizational, contractual behavior in which a higher value is placed on goal convergence than on agent self-interest. Stewardship theory "defines situations in which managers are not motivated by individual goals, but rather are stewards whose motives are aligned with the objectives of their principals" and was developed as a management alternative to agency theory (Davis, Donaldson and Schoorman 1997a, 21). The assumptions of stewardship theory are that long-term contractual relations are developed based on trust, reputation, collective goals, and involvement where alignment is an outcome that results from relational reciprocity. In the case of the government-nonprofit social services contracting relationship, stewardship theory is an appropriate model because nonprofits by virtue of their organizational form, specialized missions focused on poverty reduction and client stability, governance structures, the resource-interdependent nature of their funding relationship with government,¹⁶as well as the incomplete nature of social services contracts may well contribute to their being a closer alignment with government's goals.

This may be less true in other policy areas, for markets, and for different types of programs and services, but in the area of social services there is generally acknowledged to be a set of shared goals between the contracting parties. The differences between contracting for social services as opposed to refuse collection provide additional credence to the proposition that the managerial starting point in government-nonprofit social services relationships may in fact be different than in other contract relationships. This relationship merits examination using management theories with different assumptions about managerial behavior, agent motivation, and the variation in which management tools are used for ensuring goal alignment between the parties. Although the public administration literature has used the terms steward and stewardship to describe some forms of bureaucratic behavior (Bundt 2000; Dicke 2002), the theory remains acknowledged but largely untested. Therefore, it has not been used as a complementary theory to agency theory in public management or social services contracting research, areas important for further theory building and for new understanding into contract management.¹⁷

In contrast to the "agent" of agency theory, a steward places greater value on collective rather than individual goals, makes decisions he/she perceives to be in the best interests of his/her principals, and views the successes of the organization or contract as accomplishment and incentive for achieving goal alignment, absent any immediate financial payoff or maximizing of individual utility (Davis, Donaldson and Schoorman 1997b). Stewards are motivated by intrinsic rewards, such as trust, reputational enhancement, reciprocity, discretion and autonomy, level of responsibility, job satisfaction, stability and tenure, and mission alignment. Fundamentally, stewardship theory relies significantly on the principal's and steward's initial trust disposition. As the research of Mayer, Davis, and Schoorman (1995) identifies, trust is the willingness and risk of being vulnerable, on the part of both actors, to the possibility that one actor in the contract may pursue his/her own self-interest to the exclusion of the collectively agreed upon goals of the contract. A steward places greater value on cooperation, even when his/her goals are not perfectly aligned with the principal, over defection and other expressions of self-serving behavior. This is because of the steward's perception "that the utility gained from [contractually aligned] behavior is higher than the utility that can be gained through individualistic, self-serving behaviors" at the expense of the principal's goals (Davis, Donaldson, and Schoorman 1997a, 25). However, as noted, the initial disposition of both the principal and the contracted steward has to be toward trust and the realization of collective interests. This differs sharply with the initial disposition of the principal in agency theory of distrusting the intentions of the agent. The economic payoff for the principal in a principal-steward contracting relationship may come over time in the form of lower transaction costs associated with contracting out.

Initially, a principal-steward relationship may involve higher transaction costs than a traditional principal-agent relationship because there will be a greater investment of time on the part of the principal by involving the steward in problem formulation, joint decision making, information exchange, and generally attempting to understand the needs of the steward and therefore managing the relationship in a collectively interested manner.¹⁸ Over time, the transaction costs may well decline as each party better understands the others' motives, actions, and signals. As such, a relationship focused on collective goals and objectives should reduce the need for frequent rebidding of the contract and close monitoring of the stewards' organizational and programmatic activities. Although the principal can certainly benefit over time from this type of long-term relationship, the steward can also benefit by reducing the organizational and personnel costs associated with continuous bidding and proposal preparation and by having greater funding stability, fewer monitoring and reporting requirements, and more involvement in how the contract is defined, structured, and implemented. Trust and reputation are used as incentives for alignment and monitoring as a potential sanction but one that is less coercive and directed toward relational alignment. The costs associated with accountability should be reduced over time when there is mutual goal alignment and trust between the contracting parties. Thus, whereas the principal in a principal-agent relationship invests in coercive and compliance-based monitoring and reporting mechanisms and uses incentives and sanctions for achieving goal alignment, the principal in a principal-steward relationship invests in developing trustworthy relations with the steward through other types of contractual mechanisms that may cost more in the short run but offer long-term goal alignment. Using this theory, public managers are assumed to act as principals and nonprofit executive directors as stewards.

The main themes, tenets, and applications of agency, stewardship, and privatization theory are summarized in table 1. Agency theory leads us to postulate that given conditions of low trust, a provider would be monitored with greater intensity than a trusted provider and that sanctions would be enforced with any provider identified as engaged in opportunistic behavior. At the same time, stewardship theory would lead us to postulate that when information is not used for self-interest and goal alignment is achieved, the relationship may be characterized more as a principal-steward relationship and result in providers being monitored less and receiving rewards in the form of enhanced reputation and involvement in goal setting and program evaluation. Both theories assume that managers have sufficient capacity for monitoring and oversight, that there are alternative providers from which to choose, and that what the agent/steward is providing is in many instances measurable and observable by the principal. To a large extent, the respective theories assume rationality and therefore do not control for prevailing politics and ideologies that may be counterintuitive to the decisions enacted.

View this table: [in this window] [in a new window]   Table 1 Theoretical Tenets and Applications of Agency and Stewardship Theory

 
The theories are appropriate for studying the contract management practices in the government-nonprofit social services context precisely because prevailing political ideologies may support contracting even in cases in which the service and market characteristics are antithetical to this type of external provision. And yet, it is well documented that social services contracting takes place in municipalities around the country where the Goldsmith test of market competitiveness does not exist.¹⁹ In such cases, contract management is of significant importance because of market imperfections and reduced levels of measurability and observability in contractor-provided services. Therefore, two countervailing management theories of contractual relations are used together to investigate how public managers manage government social service contracts with nonprofits. These issues are explored empirically below.

	METHODS AND DATA

TOP Abstract INTRODUCTION BACKGROUND METHODS AND DATA FINDINGS: AGENTS OR STEWARDS? CONCLUSION References Notes

 
Social services in New York State provide an ideal research context for investigating government-nonprofit contracting relationships from the agency theory and stewardship theory perspectives. New York State consistently uses contracting as a vehicle for program implementation and is one of only 12 states in which counties have responsibility for administering federal programs (Riedinger et al. 1999). The state is frequently cited as one of the best examples of a state where counties provide general assistance as a result of a state constitutional mandate.²⁰ And, nonprofit providers are active participants and often legislatively mandated as one of the few groups eligible to participate in publicly funded service provision.²¹ In 2001, the total value of TANF contracts in New York State was estimated at $150 million, and 75% of those contract funds were with nonprofit organizations (GAO 2002, 12).

A purposive sampling strategy was employed in a cross-site analysis in New York State because it assists in identifying cases of interest from individuals who know the cases that are information rich (Denzin and Lincoln 1998; Lofland and Lofland 1995; Miles and Huberman 1994). A 15-question semistructured interview instrument was administered to a purposive sample of 12 county-level public managers, 11 state-level public managers, and 12 nonprofit executive directors.²² The interviewees were in five different counties that included one urban, three suburban, and one rural.²³ In all, 35 complete interviews were used in this analysis.²⁴ Each interview took approximately 1 h and 30 min. The interviews were tape-recorded, transcribed, and coded based on interviewee responses.²⁵ The data were analyzed using Ethnograph v. 5.0.

The counties included in the sample were purposely selected based on size, ideological preference for and opposition to contracting, history of contracting, demand for services, service supply, and county reputation for contracting innovations. This sampling strategy was used to specifically examine the extent to which these market, political, and contract conditions might affect the type of contract management practices used with government's nonprofit contractors. Additionally, the services selected were from the broad range of social service for the intent purpose of identifying variation in management practices by service type (in terms of task programmability, measurability, observability, market supply, complexity and duration of intervention, client characteristics, and political and funding support). A second reason for this strategy is that many of these service and market characteristics are in fact antithetical to the conditions often prescribed as advantageous for contracting. Such an approach therefore provides an ideal opportunity to identify conditions under which there is variation in contract management practices. The five New York State agencies included the Office of Children and Family Services (OCFS), Office of Temporary and Disability Assistance, Office of Mental Health (OMH), Office of Mental Retardation and Developmental Disabilities (OMRDD), Department of Health, and Office of Alcohol and Substance Abuse Services. Each of the individuals interviewed was involved in social service contracts in one of the five New York State counties selected for this sample.

Analysis of internal data was complemented by a content analysis of contract documents, government reports and audits, news sources, and organizational materials. These included contract data, internal program evaluation memos, budget materials, program award criteria with weights, progress reports, quarterly performance reports, committee memos, legislative amendments, and internal process and procedure documents. In all 152 contracts were administered and examined across the five counties in 24 different program areas, such as refugee assistance, home-based intensive supervision, homeless housing assistance, domestic violence, youth centers, and foster and adoptive care. Reviewing these documents provided data that are used to generalize across interviewees regarding the contract implementation environment and the corresponding management practices employed across public social service agencies in New York State.²⁶

The above discussion of principal-agent and stewardship theories suggests several propositions that can be tested against the New York case data to reveal which theory more accurately characterizes the contracting relationships. If the initial disposition of the public managers toward the nonprofit contractors is on control through hierarchical authority, using reporting and monitoring mechanisms, and initially assuming a lack of trust and goal alignment, then the relationship would be consistent with the tenets of agency theory. If the initial disposition of the public managers is on extending trust to its nonprofit contractors, involving the organization in the development of the contract goals, and enhancing the contractor's reputation, then the contract management practices would be consistent with the tenets of stewardship theory. If the relationship develops over time, trust is built and established between the contracting parties, and goal alignment ensues, then we would expect the relationship to be characterized by less monitoring and greater relational exchange (Beinecke and DeFillippi 1999). A more difficult question is whether the management practices of an evolved relationship are more representative of a principal-agent or principal-steward relationship (Albanese et al. 1997). A proposition for consideration therefore is that in evolved principal-agent relationships the contract management practices employed will be strongly correlated with those practices set forth under stewardship theory. Finally, the context is hypothesized to strongly influence the intensity and variation of the contract management practices applied by public managers with their nonprofit contractors. In these cases, context is likely to prevail over theory-practice alignment.

Therefore, trust, reputation, and monitoring are expected to be used as contract management practices and will consist of variation in their form and intensity based on the level of goal alignment. Given recent scholarship, it is expected that the contextual conditions set out in this study will strongly influence those practices. Such cases then are likely to suggest an evolution in contract relationships from a principal-agent to principal-steward framework but for contextual reasons rather than those suggested by theory.

	FINDINGS: AGENTS OR STEWARDS?

TOP Abstract INTRODUCTION BACKGROUND METHODS AND DATA FINDINGS: AGENTS OR STEWARDS? CONCLUSION References Notes

 
The focus of this research centers on three questions: First, what public management contract practices are applied to nonprofits in the contracting relationship? Second, what conditions affect the public management contract practices applied to nonprofit organizations in the contracting relationship? And, third, to what extent are these management practices consistent with the tenets of agency and stewardship theories?

Public and nonprofit managers exhibited different initial dispositions toward each party with respect to contractual compliance and described a number of practices employed to manage contractual relationships with nonprofit organizations. Variation in how public management practices were applied and the conditions under which they were applied is examined. Variation does exist in the initial trust disposition among the contracting partners with public managers viewing contracted nonprofit organizations and their leaders as agents in any new relationship for service delivery. Nonprofit leaders preferred to view their relationships with government agency managers as that of a steward involved in service provision with shared goals based on collective missions. The management practices used with nonprofit contractors are largely consistent with more relaxed assumptions of an evolved principal-agent relationship. An evolved principal-agent relationship mirrors some of the practices put forth under stewardship theory. However, further analysis reveals that the practices applied may be as much a result of the market conditions, service characteristics, and the public manager's own internal agency contract management capacity as they are with existing management theory. The management practices used to oversee contract relationships and the conditions that would suggest variation in those practices were not as expansive as the theories would seem to suggest. The public managers were well aware of the need to vary their management practices. Yet, they often cited the types of contextual constraints mentioned including the political environment that shaped the decision to contract as central to the contract governance tools they employed.

What materialized from this investigation is a complex managerial environment, an environment in which county size and market conditions gave rise to and afforded managers different tools and responses to managing contract relationships with their nonprofit contractors. In the urban county, public managers developed principal-agent relationships with new contractors and existing providers they did not yet quite trust. They also managed evolved principal-agent relationships with "preferred" providers that implemented contracts where there was some competition and service measurability and observability. In the suburban counties, competition, management capacity, and service characteristics played a very important role in how public managers governed their contract relationships. The contract relationships often began as principal-agent–type relationships and evolved into steward-type relations in which trust, information exchange, involvement, and discretion were used. In the rural county, there was no variation to speak of, other than the public management rhetoric voiced by those local contract managers who saw themselves engaged in hierarchical and authoritative relationships over their nonprofit contractors. The reality was that these public managers possessed little latitude in how they managed their contractors. If unhappy, they could not easily terminate the contract because they lacked the capacity to take the service over, there was not another in-county provider, and out-of-county providers were not immediately available or even frequently solicited because of the great maxim "the one you know is better than the one you don't." If public managers were happy with the contractor, it was difficult to reward one contractor more than another because "everyone talked and knew everyone else's business" and performance-based contracts were not the norm.

The contract management practices used by public managers as incentives, sanctions, and decision heuristics as a result of administrative capacity shortages, a dearth of providers, and measurement uncertainty are trust, reputation, and monitoring. These practices are relied upon because they reflect the limited capacity levels and markets in which public managers operate. Public managers draw confidence in the efficacy of contracting with nonprofit organizations because of their organizational form and mission. They also rely on supply-side imperfections in the market to mitigate against contractor opportunism because of the resource interdependent-needs of the parties. In this section, I will discuss my findings about trust, reputation, and monitoring from the perspectives of agency and stewardship theories.

Trust
Trust is a public management contract practice used by public managers to manage nonprofit contractors. There are different types of trust, such as general and particularized trust, though in this section the focus is on strategic trust, to be denoted as trust that is based on knowledge of and experience with the other party and a mutual expectation of reciprocity (Hardin 2002; Yamagishi and Yamagishi 1994). Trust is a major psychological and social process that underlies developing, maintaining, changing, and discontinuing contractual relations (Rousseau 1995). The issue of trust is at the center of agency and stewardship theories with agency theory assuming agent self-interest (the agent is not to be completely trusted) and stewardship theory assuming a steward's actions are aligned with the principal's interests (the steward can be trusted, especially if it is a nonprofit provider with similar goals). As Whitener et al. (1998, 513) note "trust involves some level of dependency on the other party so that the outcomes of one individual [actor] are [is] influenced by the actions of another." Trust has cost implications and is integrally linked to reputation, information exchange, the discretion and legitimacy afforded to contractors by public managers, the types of rewards and sanctions applied to providers, and how providers are monitored. It can also be argued that an incentive for the agent to be in alignment with the principal's goals is the trust that may be extended to them and the positive externalities that may result from being perceived as a trusted partner. This is congruous with an evolved principal-agent relationship. The attitudes, values, and trust dispositions of public managers about contractors, program areas, and clients affect how trust is developed, extended, used, and rewarded. Trust is, however, fundamentally an assumption about the level of risk one exposes oneself to in relationships of expected reciprocity and goal alignment. Agency theory and stewardship theory have fundamentally different starting positions in terms of the level of trust principals extend in contractual relationships.

Public managers supported the proposition that trust is at the center of contractual relationships and is the single most important criterion in how and under what conditions providers are to be managed.

Trust is outcome based, based on success. You could have a history, but that history is built on success. We're not a very trusting agency.

This statement reflected the viewpoints offered by all the public managers interviewed and is consistent with the work to date on the issue of strategic trust and with the tenets of agency theory in that trust evolves between parties, but is not simply accepted as a relational starting position, as is the case with stewardship theory. Trust is achieved based on repeated interactions and transactions over time in which goals are achieved with consistent responses given to questions and information that is made available in a timely fashion upon request. Many of the nonprofit executive directors described trust as an "attitude." An attitude of "what can I do to make this better versus a woe is me, this isn't right attitude." As an executive director stated, "they [the public manager and bureau director] trust us to be politically sophisticated and not to say or do anything that is going to embarrass the governor's administration." This viewpoint was offered by approximately 83% of the executive directors and is more reflective of a principal-steward relationship in which the steward seeks alignment with the principal's goals. These goals are less about outcomes produced and more about individual and organizational behavior that is both astute and adaptive to the political environment in which contracting takes place. Initially, therefore, public managers come to contract relationships with a trust and management disposition consistent with agency theory, whereas a majority of the nonprofit executive directors approach the relationship with a steward-like disposition.

Trust, as expressed by a commissioner, was "built on knowing each others' motivations and limitations." More than 87% of the public managers genuinely embraced this idea and reported that at the outset of new contract relationships, they worked to understand their contractors' motivations and build that information into their contract governance practices. Information exchange was frequently identified as a key building block for trusting relationships. The public managers continuously reiterated that trust is built one day at a time through communication, interaction, the repeated articulation of goals and outcomes, and by measuring client satisfaction.²⁷ Involving nonprofit providers as stakeholders in decisions, engaging them in identifying need, and showing them respect were all components identified by public managers as necessary for building trust. Public managers spoke of being able to go to an executive director and individual board members they personally trusted and getting their objectives achieved. These actions are reflective of an evolving principal-agent relationship. In managing contract relationships, public managers spoke of the rewards they use for trusted providers and the sanctions applied to those they do not trust.

How do public managers reward providers they trust and sanction those they do not? Sixty-five percent of the public managers spoke of sharing information with their nonprofit providers that could assist them in the preparation of contract proposals. Information exchange took the form of providing "inside" information on a particular clientele, new funding being proposed in the legislature, suggestions to providers on leveraging additional contract funding, and specific information a bureau director and his/her staff look for in request for proposals (RFPs). With the exception of issues about confidentiality which each contracting party adhered to, many public managers would assist executive directors whom they trusted with information about how to navigate the bureaucracy, who they should talk to, and what political strategies to enact in order to secure support from agency executives and other officials. In some cases, nonprofit organizations would share true cost data with public managers in an effort to assist them in their own strategic resource planning and budget preparations. Contract relationships that reached this stage, frequently taking many years and built on histories of success, were characterized by low levels of monitoring, enhancing a provider's reputation, and involving nonprofit providers as partners in the formation, implementation, and evaluation of policy. As approximately 74% of the public managers noted, "when you trust that they're doing what they say they're doing you don't have to watch them as closely." Over a period of time, 83% of the county-level public managers and approximately 64% of the state-level public managers described a transformation in the contract relationship. When providers achieved this level, they were referred to as "preferred" providers, those easy to work with and who were manageable, reputable, and qualified. Two-thirds of the nonprofit executive directors suggested that when they were perceived and affirmed by public managers as trusted, "preferred," they were able to be more creative and innovative and could make bolder, more assertive policy and programmatic decisions using the discretion they earned based on the trust developed. As one provider said, "we could not have taken risks without trust and experience which led to rewards in the form of enhanced reputation." Trust in the relationship was used to publicly demonstrate the effectiveness of public-private partnerships. This transformation can be viewed as an evolved principal-agent relationship in which trust is used as an incentive alignment mechanism and reward comparable to the tenets of stewardship theory.

In instances where contractor malfeasance was identified, public managers confronted the providers but did not formally sanction them. Informally, public managers spoke of no longer trusting the provider, affording them less discretion and legitimacy, allocating funding incrementally rather than all at once, and scrutinizing their performance reports. State-level public managers reported being even less trusting and utilizing more formal sanctions, such as contract termination, in part because they had the luxury of putting RFPs out for bid in competitive markets. This is in contrast to county-level managers who relied on more informal mechanisms given their own market supply-side imperfections and human capital capacity constraints. In suburban counties, such malfeasance was dealt with not by putting the bid immediately out for contract but in stripping the provider of discretion, stopping all informal advocacy on the provider's part within the public manager's network, and over time cultivating alternative providers from other counties or out of state to replace the provider that was no longer trusted. These public managers described having a "long memory" for this type of nonprofit action. In the rural county, the public managers were more boisterous and confident that such malfeasance does not occur on their watch because of the "small town environment of knowing what one another is doing." The sanctions are consistent with the types of mechanisms suggested by agency theory when trust is breached but inconsistent in the intensity of their use because of the contextual conditions of the contracting environment. Those conditions strongly influence the variation and intensity of the applied sanctions. This is especially evident with the suburban county managers. The sanctions used by the principal are divergent from the type of prorelational intervention designed to realign a steward with the principal's goals as put forth under stewardship theory.

It is clear from these findings that trust is a significant contract management practice and decision heuristic used by public managers. Trust is not the default position for public managers, consistent with agency theory, whereas it more often is for a nonprofit, which is consistent with stewardship theory. This is not meant to oversimplify the complex nature of trust as an integral construct to how contractual relations are managed but rather to illustrate where different parties begin in their initial contract relationships. There is variation in how providers are rewarded and more modest variation in whether and how they are sanctioned based on other contextual factors, such as market competition. There is strong evidence that trust evolves over time from a legal and transactional arrangement, consistent with agency theory, to an involvement-oriented steward-type relationship. Trust as a tool for goal alignment and a contractual reward is more reflective of an evolved principal-agent and principal-steward relationship in which reciprocal benefits, such as information exchange and involvement, are used.

Reputation
Reputation is an organizational, programmatic, and individual attribute used by public managers for assessing contractors. Milgrom and Roberts (1992) state, "if it is possible but costly to write detailed contracts, a good reputation can often allow the decision maker to avoid the expense as well as the use of costly and error-prone legal contract enforcement mechanisms." Reputation is frequently referred to in the economics and agency theory literature as a contract enforcer or bonding mechanism because of its historical role in promoting and ensuring goal alignment (Milgrom and Roberts 1992). They go on to suggest that, "in each transaction, the decision maker honors trust in order to encourage future trading partners to offer trust or, in other words, to maintain his or her reputation for honesty [sic]" (263). The role of reputation is accentuated in authority relations where there are repeated transactions. Reputation is a mechanism used by public managers for evaluating the past performance of nonprofit organizations, for identifying future providers, and for partially mitigating the opportunity for moral hazard that can result from incomplete contracts.

Public managers reward providers by enhancing their organization's reputation and sanction providers especially by questioning publicly their credibility and service quality or by remaining silent about the provider's reputation to others. Reputation is established in a number of ways. It is most often built through past performance, meeting goals and achieving outcomes, and attaining goal convergence. It is a factor that can, over time, transform contractual relationships from a hierarchical and control-oriented principal-agent relationship to a more involvement-oriented principal-steward arrangement based on shared goals and trust. Reputation serves as a commodity for many nonprofit organizations because of its causal relationship with trust.

In a small county your reputation is all that you have. Competition is not as important as reputation.

These insights from public managers in a rural county reflect the importance that reputation plays in the contracting process. A deputy commissioner in a suburban county expressed a perspective confirmed by 74% of the public managers in commenting "enhancing a nonprofit organization's reputation is more than anything I can give them." As public managers expressed:

Those providers, that do what they say they are going to do, deliver what they say they will deliver, and meet time lines, goals, and produce measurable outcomes are the organizations whose reputations will be enhanced.²⁸

Executive directors of reputable providers were invited to attend and participate in public events with agency executives and elected officials by serving in highly visible roles such as executive members of statewide professional associations; speaking at professional boards, committees, conference panels, and public hearings; and participating in the formulation of policy. By participating in these types of events, providers gained legitimacy, credibility, visibility, additional funding opportunities, and enhanced reputation. The visibility and legitimacy provided nonprofits with a vehicle they could use to give voice to their client's issues in a public forum where respect and credibility were accorded to the providers by public managers, elected officials, and the media. Nonprofit leaders recognized the value of these nonfinancial rewards. As an executive director noted

These opportunities give me access to the types of people I would not normally interact with. I can educate and advocate with credibility without being perceived as a threat ... this is where I can get involved in the legislative process ... and that's how we're going to benefit as an organization. That's my job as the ED, and these are the relationships I need to establish.

This executive director's comments capture the sentiments of 83% of the nonprofits leaders regarding the importance and benefits of being perceived as reputable. In these ways, public managers were managing and rewarding their contractors in a manner more consistent with stewardship theory. Agency theorists argue that reputational enhancement is an incentive alignment tool, especially where nonfinancial rewards are limited. It might therefore be concluded that reputation, as a reward strategy employed by public managers, is more reflective of an evolved principal-agent/steward relationship in which there has been a history of trust and goal alignment.

Public managers also rely on reputation and often use existing providers they are happy with to deliver services even in programmatic areas in which the provider may not have direct expertise and experience. This is a more common practice than the contracting literature has generally acknowledged. Each nonprofit executive director described reputation as, "you're buying a history and stability, reliability, community presence, and quality assurance." Public managers confirmed that in many cases their use of reputation as a decision heuristic is consistent with the characterization by executive directors. They did insist that it is not because they are indolent but rather it is a proactive strategy to minimize instances in which contract providers may act in their own interest to the exclusion of the contract goals. A commissioner captured a sentiment shared by all the public managers in stating:

the risk of rebuke [resulting from public awareness of contractor malfeasance] is too high. Where we have a known quantity that we can work with, that's who we're going to go with; it cuts risk for us.

This statement affirms public managers' concern about exposure to fraud and their strategic decision not to seek additional providers, even when they are available, because of their risk-averse disposition. This statement was overwhelmingly acknowledged by all the public managers, regardless of venue and market characteristics. Public managers cited another reason for relying on reputation; it provides the public agency with a degree of legitimacy from which to promote or blame depending on how the relationship develops. If a reputable provider is exposed as acting in an opportunistic manner, then the public agency has some leverage in being able to say publicly that this was a reputable organization and there was no way of knowing it would act this way. The risk-averse behavior of public managers reflects the management capacity constraints that can limit their ability to develop competition and exercise oversight. This use of reputation is more representative of agency theory because it is being used to reduce risk for the principal. Stewardship theory is also concerned about reducing risk but takes a prorelational and involvement-oriented approach to developing goal alignment with its contractors. Agency theory is explicit about redistributing risk through a variety of incentive and sanction mechanisms. Here, that mechanism is reputation. Public managers, therefore, rely on reputation as a proxy for service quality and expected goal alignment, consistent with agency theory.

Sanctioning of provider reputation was found not to have the same economic costs as having no reputation at all. When public managers did not comment on a contractor, such as a provider with a noted community name and reputation, this was intended to signal a sanction. Repeated incidences of not commenting on a provider were strategically intended by public managers to erode the high level of support and reduce public pressure from elected officials and agency executives to contract with specific organizations. Yet, in markets devoid of competition, a diminished reputation does not necessarily have an adverse effect on a provider's opportunity for continued contracting in the way that agency theory strongly suggests it should. Therefore, the higher bonding costs associated with a poor reputation are negated in large part, especially in small- and medium-sized counties, because of supply-side market imperfections. The management technique of signaling and reputational diminution is consistent with the tenets of agency theory, but the intensity and effect is weakened because of the contextual conditions of the contracting environment.

Public managers were quick to acknowledge that the rewards they can legitimately and publicly offer are not going to be valued and appreciated by all providers in the same way. One-third of the smaller nonprofit organizations reported that they did not feel that these "reputational rewards" were rewards at all. As one executive director commented, "profiling us hasn't brought us any additional funding and that's what this organization needs." For several of the smaller nonprofit organizations, "intangible, non-financial, psychological rewards do not pay the bills, provide for staff pay increases, or assist in hiring additional staff and offering additional services." Larger, more established, and better positioned nonprofit providers claimed they were able to use the benefits representative of a principal-steward relationship (trust, recognition, reputation) to leverage additional resources from other revenue sources. Smaller nonprofits lacked this capability and were more sanguine about the financial benefits or otherwise that could be derived from a principal-steward relationship. As one public manager observed, by at least giving them the nonfinancial and psychological rewards, they may be successful in attracting additional funding in the long term from other revenue sources. The public managers were keenly aware that their limited cache of incentives and rewards might be insufficient in curbing opportunistic behavior among some providers. Yet in the end, they used the level of market competition, their own administrative capacity constraints, and trust and reputation as tools for contract alignment. Public managers are using reputation as a management practice in creative ways but are also using the contextual conditions that so often limit their contract governance approach in a manner that may actually encourage alignment. As the managers noted, some providers will see their use of reputation as a reward and therefore stay aligned with government's goals. Other providers, especially smaller providers, may see the managerial practice of using reputation as a one-size-fits-all approach by public managers.

Reputation is used as a management practice, incentive, and sanction. However, the question public managers may wish to consider is whether there are other incentive alignment mechanisms that may be more appropriate for smaller providers compared to larger providers, in urban counties compared to rural counties, and for services that are more competitive such as welfare-to-work services compared to services for persons in need of supervision. The findings demonstrate that large providers view reputation enhancement as a reward and one that can lead to positive externalities. Smaller providers with more limited resources and opportunities perceive the use of reputation as incongruous with their organizational, financial, and capacity needs. The use of reputation by public managers as an incentive is clearly consistent with agency and stewardship theories, but the limited variation in its use as a sanction because of market conditions does not illustrate congruity with either theory.

Monitoring
Monitoring is a public management contract practice used by public managers to manage nonprofit contractors. One of the claims made by advocates of privatization is that the quality and value of a product and service is enhanced because more monitoring is present than is the case with internal production. Monitoring and oversight are important components of any management theory and integral to agency and stewardship theories. Frequently, the literature on this topic suggests formal mechanisms. Monitoring can also occur, however, through the use of informal mechanisms as has been shown in previous contracting research (Smith and Lipsky 1993). In this section, I will describe not only the formal monitoring mechanisms in place to provide oversight but perhaps more importantly the informal oversight tools enacted to increase transparency and ensure accountability. The variation and intensity of the various monitoring mechanisms will also be discussed.

A number of formal monitoring strategies were found to be available for overseeing the actions of nonprofit providers and evaluating the quality of the services they deliver. These include, but are not limited to, colocating public staff at a nonprofit organization's office, using program staff to conduct performance audits, relying on financial audits conducted by the Comptroller's office, contacting providers on a consistent basis, performing site visits on capital projects, requiring quarterly reports such as the Consolidated Fiscal Report, enforcing corrective action plans from providers placed on probation for performance or compliance issues, mandating client surveys as an evaluation tool for demonstrating contractor performance and service quality, reviewing a provider's diagnosis and recommended treatment plans for clients, and relying on feedback from caseworkers in the field and other providers. Each of these formal and direct monitoring mechanisms is clearly aligned with the mechanisms for oversight suggested by agency theory. Frequent contact, feedback, and periodic reporting would also be consistent with techniques suggested by stewardship theory for achieving goal congruence. Public managers also relied on regulatory enforcement from other state agencies and agency inspector generals to ensure provider compliance with contract requirements. In particular service areas, such as child care, public managers contracted for oversight functions with third-party accreditation organizations.²⁹ In addition, owing to the distinct organizational form of nonprofit contractors, public managers rely on the use of nonprofit boards of directors and charitable donors to ensure that contracted providers are aligned with their organization's missions, deliver quality services, and faithfully adhere to the terms of their contracts for which they received funding. This type of oversight is commonly referred to as proxy monitoring. In other cases, monitoring occurs through media and citizen feedback in identifying cases of malfeasance. This type of oversight is frequently referred to as fire alarm monitoring and is the form of oversight that is least desired by public managers because of the sanctions that will result to them and their agencies.

Low levels of monitoring are the norm in agencies without a culture of regulatory compliance.³⁰ One public manager's explanation resonates consistently with the views expressed by 91% of the public managers interviewed in commenting that

... we wouldn't like to admit it, or say that we're not monitoring. Part of it stems from staffing and capacity issues. Part of it is that we build in monitoring, but the burden is on the provider. So we're asking for reporting mechanisms. We have over 500 providers and 12 people to follow up. How much do you do that? ... Of all the functions we need to do, getting budget proposals, making contract decisions, getting this service completed, signing off, which one will people not notice if we don't do? Monitoring.... Contractors will notice if they're not getting paid, clients will notice if they have nowhere to go, few will notice an absence of monitoring. Our management certainly notices if we have the appropriation of funds and yet we don't get the RFP out or we say we have an RFP, but we're not ready to make that decision yet. Question is why haven't we spent the money yet? But the monitoring is the one that everyone agrees is critical and important and is not happy if it doesn't happen, but is also the one they're not asking about every day.

This candor reflects many of the frustrations expressed by public managers about oversight and is integral to the decisions they make about how, where, and when to allocate their scarce human and financial resources to this function. Agency theory is prescriptive in suggesting that monitoring is critical for principals to ensure the alignment of agents. In a principal-steward relationship, it would be acceptable for a principal to monitor a steward less frequently if the principal had contracted with the steward over a long period of time, the steward has a strong reputation, and a history of goal alignment with the government agency. This could also be true of a principal-agent relationship that has evolved over time and one in which the core management assumptions were relaxed. The quote above reveals that management theory aside, administrative capacity constraints contribute to a contracting environment in which public managers do not use monitoring as a contract management tool as much as they desire regardless of the evolution of the contract relationship from one of agent to steward. There is monitoring present through the use of performance reports. However, other formal monitoring mechanisms, such as site visits, are described as being limited because of human capital constraints.

Conversely, public managers interviewed at the OMH suggested that oversight is very important in their agency. This occurs in part because of their culture of regulatory compliance and that resources are allocated to monitor the contracted providers. They also suggested that there is very little variation in how providers are monitored. Long-term providers are monitored as frequently as new providers, though a bit more flexibility and discretion are afforded to reputable providers. Consistent with the regulatory culture of this agency, legislative mandates also set up and institutionalized the Commission on Quality of Care (COQC) which serves to oversee and monitor OMH's ability to oversee its providers and services.³¹ This is a case of multiple levels of oversight existing to insure that opportunism does not exist and that a standard of service quality is being produced. Yet, providers who contracted with this public agency perceived the oversight actions of COQC and OMH as duplicative, costly, and excessive. The monitoring, according to the contract parties, was motivated by low trust, fear of culpability, and the threat of exposure by the public and media to instances of malfeasance and abuse. This finding is at odds with stewardship theory but could be argued as an example of a principal-agent relationship that is fundamentally premised on mistrust and fear of moral hazard. However, the lack of variation in how and the intensity with which monitoring tools are used points more to the contextual conditions of a politicized contract environment and less to either theory. The theories are not particularly discriminating in the degree to which application is shaped by organizational, environmental, and contextual conditions. The findings suggest that public managers are aware of different monitoring tools but are constrained by their capacity and environment.

To this point only formal monitoring mechanisms have been discussed. However, the public managers did use informal monitoring mechanisms. They relied on the reputation of providers, their past experiences with providers, and the past experiences that colleagues who were part of their network had with certain providers. In each of these cases the argument could be made that some monitoring is happening, but it is more representative of the contracting environment than the prescripts of either theory. What we can conclude is that managers are making decisions in an interconnected and networked environment. Did the public agency, type of service, number of clients, and funds allocated have any bearing on the intensity of the formal and informal monitoring practices? Yes, in cases where particular client populations were receiving services for which political support had waned, there was less monitoring. In cases where capital projects were underway, the frequency of monitoring was increased because the stakes were perceived as higher. In cases in which the type of service activity lended itself to the reporting of performance measures, public managers would randomly select providers and performance reports for review. However, this selection was not scientific. If it were, that would be a useful tool. The selections, as described by the public managers, were very random, like sticking one's hand into a pile of contractor-submitted performance reports, pulling one out, and quickly assessing whether aggregate benchmarks were roughly aligned with contract expectations. A county deputy commissioner summarized the monitoring issue expressed by 70% of public managers in stating, "there is no current or consistently applied process ... government has never been taught how to do this in terms of oversight." This acknowledgment, though, was not the only manner in which monitoring was addressed. Public managers also expressed skepticism that nonprofit organizations were sophisticated enough to "pull the wool over our eyes" and engage in opportunistic behavior.³²

Internal state agency documents included a number of Comptroller audits that consistently highlighted "the absence or severe lack of monitoring." Report highlights suggested a lack of project documentation, capacity, and information systems necessary for oversight; over reliance on providers to document community need; and few, if any, systematic attempts at data collection. Comptroller audits that identified similar weaknesses in county oversight actions were equally harsh. Other types of reports were also critical of the oversight exercised by state and local agencies.³³

The important findings to emerge have less to do with the monitoring mechanisms available to oversee providers and more to do with the lack of variation and intensity of formal and informal practices. Intuitively, one proposition we would expect, given the theoretical tenets of agency theory, is that conditions of low trust would result in greater levels of monitoring and greater levels of trust would result in less intense and demanding forms of monitoring, consistent with both agency and stewardship theories. The findings from this analysis suggest there is actually much less variance in the level of monitoring used to oversee trusted and reputable versus less trusted and less reputable providers. Monitoring and sanction enforcement as management activities were used more as a measure of last resort rather than as contract governance tools for reducing fraud and abuse and achieving accountability through goal alignment. Monitoring is a critical public management responsibility, one written about in most of the contracting literature, a vital component of most management theories and styles, and yet an area of public management that is constantly highlighted for what has not been achieved, implemented, and enacted (See GAO-02-245). Agency and stewardship theories say very little about the degree to which organizational, contextual, and environmental conditions can strengthen and weaken the use and intensity of oversight practices in contract relations.

	CONCLUSION

TOP Abstract INTRODUCTION BACKGROUND METHODS AND DATA FINDINGS: AGENTS OR STEWARDS? CONCLUSION References Notes

 
This study uses social science theory and a multisite methodological approach to examine incomplete contracting in service areas in which program effectiveness is neither easily measured nor observed in largely noncompetitive markets and in public agencies where there are management capacity constraints. Agency and stewardship theories are used to examine three questions about the government-nonprofit social services contracting relationship. First, what public management contract practices are applied to nonprofits in the contracting relationship? Second, what conditions affect the public management contract practices applied to nonprofit organizations in the contracting relationship? And, third, to what extent are these management practices consistent with the tenets of agency and stewardship theories?

Agency theory has been used extensively to study a range of principal-agent contract relationships, but few empirical studies have "relaxed the agency theory assumptions" suggested by Eisenhardt (1989) to assess the extent to which principal-agent relationships evolve over time. The development of stewardship theory is more recent and serves according to Davis, Donaldson, and Schoorman (1997b) as the counterweight to agency theory. They suggest that not all contractual relationships are based on individual utility maximization at the expense of goal alignment. The findings from the research interviews suggest that trust and reputation are used with some variance and that the effects of other contextual variables, such as service characteristics, market competition (supply-side imperfections), and management capacity constraints, have a significant impact on how contract relationships are managed. The contextual conditions strongly influence the lack of variation and intensity in which monitoring is used to achieve goal alignment in contract relations.

If we examine the social services contracting relationship strictly from an agency theory perspective, we find that public managers do not initially trust their contractors and are thus more formal, control oriented, and hierarchical in their relationships with nonprofits. However, over time, the public managers do appear to develop relationships that are more relaxed versions of a classic principal-agent relationship. There is still a focus on control, compliance, and hierarchical authority but largely because the public managers see their role as funders/buyers in a monopsonistic market environment. Trust is extended and rewarded in some creative ways and used as a mechanism not only to foster goal alignment but also to informally monitor the actions of their contractors. Reputation is also used in the spirit of agency theory as a contract enforcer, but again with a more relaxed set of assumptions. It is employed as a decision heuristic, reward, and sanction. As rewards, trust and reputation are used by public managers to reward preferred providers for their goal alignment. Yet as sanctions, trust and reputation are used less frequently and without the intensity that agency theorists might prescribe. Formal and informal monitoring mechanisms are used but without significant variation and intensity. The lack of use or excessive use as in the case of mental health contracts is inconsistent with both agency and stewardship theories. In part, this underutilization must be considered within the contextual conditions of the government-nonprofit contract relationship.

If we examine the social services contracting relationship strictly from a stewardship theory perspective, we find nonprofit executive directors who sought relationships where they were trusted as partners to government based on their organizational form and mission, as well as their expertise and commitment to serving persons in need. Nonprofits that entered into new contracts with government agencies anticipated that their positive organizational reputations would create the context for a principal-steward relationship. And although the creative use of trust and reputation and fewer instances of overt and coercive monitoring practices were identified in the contracting relationships, care has to be taken not to ascribe these practices only to the dispositions of public managers and the instruments in their contract management toolbox. For underlying the contract management practices are contextual conditions that give rise to the practices employed.

The context of this study suggests certain limitations for further consideration. First are the risk-averse principals and risk-averse agents. The risk-averse disposition of the public managers varied by county, with the urban county- and state-level managers exhibiting the least aversion to risk because of more competitive markets as well as the rural county managers who believed the resource-interdependent nature of the contract relationships with local nonprofits would largely preclude goal divergence and opportunism. The greatest level of risk aversion was in the suburban counties where competition could be developed in some cases and where a legitimate risk existed for nonprofits of contract termination but also because of the administrative capacity constraints of public managers which limited the scope of management activities they could exercise, such as market development and contractor oversight. The suburban managers also believed they lacked the agency capacity to internalize services they were presently contracting for. To some extent, the managers considered the idea of joint contracting and the advantages of some level of internal production and some portion of external provision. While not articulating the issue as such, the managers clearly perceived the transaction costs associated with creating some level of internal capacity for joint contracting to outweigh the benefits. Thus, all the public managers and especially those in the suburban markets expressed concern about diversifying their contract investments even given their monopsonistic position in local social services markets.

Given the dearth of research about evolved principal-agent relationships and general lack of empirical evidence using stewardship theory in nonproprietary firm settings, the following conclusions can be drawn. First, the initial disposition of public managers in contract relationships is consistent with agency theory. The initial disposition and desire of nonprofit executive directors in contract relationships are consistent with stewardship theory. Second, the lack of financial incentives and inconsistent use of monitoring is incongruous with agency theory. Third, the use of trust and reputation is clearly consistent with the tenets of stewardship theory but may also be consistent with an evolved principal-agent relationship. Fourth, the two theories used in concert together are complementary. Stewardship theory captures the state of what evolved and aligned principal-agent contract management practices might look like between government and nonprofit organizations. Fifth, the contextual characteristics that color much of the application of public management contract practices are not well accounted for in either theory. Therefore, the theories only explain part of the government-nonprofit relationship. More work is needed on evolved principal-agent relationships and on the development of principal-steward relations. The variables that comprise the set of contextual conditions are a very important set of factors to be considered and controlled for if these theories are used to explain how government manages its social services contracting relationships with nonprofits. In this respect, the line between an evolved principal-agent and principal-steward relationship is less precise than desired. This exploratory research suggests that the contract relationships examined are more consistent with an evolved principal-agent relationship. The theories used in combination do explain part of the puzzle of how government-nonprofit social service contract relationships are managed.

The theories present limitations that need to be considered within the context of this study. Agency theory is silent on the relevance of competition, management capacity, and availability or lack thereof of strong incentives, particularly financial, for aligning an agent's actions. The theory's dominant focus on executive-board relations and financially aligning their interests also limit the theory's utility to other units of analysis where such incentive alignment devices are more constrained. Stewardship theory fails to consider the lack of trust within government agencies, the risk-averse dispositions of public managers, and the lack of incentives for extending trust to contracting parties. The deeply politicized environment and scrutiny for oversight and accountability create pressures for not developing trusting relationships because of external perceptions of corruption. Stewardship theory does not say much about the degree to which a principal extends trust to a steward in a new relationship or the degree to which that trust is further extended in an evolved relationship. Consequently, given the lack of empirical tests using stewardship theory, little is known about the extent to which trusting principal-steward relationships evolve, the prorelational attributes that are developed and used, and the quality and cost of the outcomes that result from principal-steward convergence.

The findings suggest implications for public affairs education, namely that public managers need specific training on how to manage increasingly networked third-party contracts in integrated service markets. Increasingly, there is joint production among many actors for the provision of governmental services that may be inadequately specified and for a clientele that may have varying levels of motivation and acceptance for intervention. Public managers need training on contract design, solicitation, and management in which they understand the trade-offs among a range of factors. Such factors may include service, client, and agent characteristics, as well as market competitiveness. They also need more training on when and how best to involve contractors in the joint formulation of goals, objectives, measures, quality standards, reporting mechanisms, modes of oversight, and evaluation. Too often, contract management training has been underinvested in by government agencies and not taught in public affairs programs. Skills such as bargaining and negotiation, communicating policy and program goals, aligning contractor goals with agency goals, conducting oversight, providing technical assistance, and evaluating program and client outcomes are needed for contract managers to be successful at managing relationships. These skill sets are not going to abate with time. On the contrary, demand for these skills is going to grow with continued devolution and increasing reliance on market alternatives to government service provision. Contracting is taking place at all levels of government, and skilled contract managers are more needed than ever.

The findings presented here suggest an alternative explanation. A relational contracting model combined with attributes of resource interdependence may better capture the middle ground between agency and stewardship theories. To resolve the contract management problems that can arise because of the contextual conditions discussed, government and its nonprofit social service contractors enter into long-term negotiated relationships that involve trust, discretion, joint problem solving, and information exchange. Each of these variables contributes to promoting goal alignment and reducing the measurement difficulty and problem intractability in service provision to clients. Trust is not necessarily assumed at the beginning of the contract relationship but is built through extensive interaction and involvement focused on communicating each other's goals and approaches to service intervention. Such efforts in which the transaction costs are higher at the outset of the relationship are intended to build alignment through approaches that resemble stewardship theory but without the same emphasis on traditional arms-length principal-agent prescriptions of incentives, sanctions, and monitoring. Therefore, fewer instances of opportunism and moral hazard are to be anticipated because of the interdependent needs of the parties. This interdependence gives rise to more relational forms of contracting that is built on mutual exchange and reciprocity.

Future research should examine the extent to which public managers account for the range of contextual, organizational, and contractor characteristics and develop relational contracts. Whether such a contract management approach would actually contribute to lower costs and higher quality service provision because of the coordinated approach to joint problem solving and decision making, information exchange, flexibility and discretion, and reduced monitoring is an empirical question. These are the types of contracting questions that need to be better understood, especially as contracting and managing contractual relationships become more embedded in our execution of democratic governance.

	Notes

TOP Abstract INTRODUCTION BACKGROUND METHODS AND DATA FINDINGS: AGENTS OR STEWARDS? CONCLUSION References Notes

 
For their helpful suggestions, I am grateful to Trevor Brown, Matt Potoski, Stu Bretschneider, Carolyn Heinrich, Michael Rushton, Ravi Dharwadkar, Kristin Byron, and the anonymous reviewers. I would like to thank Norma Riccucci, Judith Saidel, and Frank Thompson for their continuing support. The usual caveat applies.

¹ For a very recent case, see http://www.govexec.com and review the stories involving the Air Force, Darlene Druyun, and the Boeing Corporation. Examples include the January 12, 2005, story titled "Druyun scandal may lead to more vacancies at the Pentagon" and the February 14, 2005, story titled "More Air Force contracts under scrutiny in procurement scandal."

² See for example, GAO-02-245 "Welfare Reform: Interim Report on Potential Ways to Strengthen Federal Oversight of State and Local Contracting." April 2002.

³ The GAO (2002, 3) documents that of the 88% of total Temporary Assistance for Needy Families (TANF) funds contracted by state governments, 73% are with nonprofit providers.

⁴ Demand-side imperfections are also a problem given that some types of social services are monopsonistic where government agencies are the only buyers of services.

⁵ Milward (1994) and Milward and Provan (2000) refer to this challenge as governing the hollow state.

⁶ Shleifer and Vishny (1998) refer to contracts that lack contract specificity as incomplete contracts.

⁷ Even within the field of social services, there is considerable variation among type of service and expected outcomes. For example, workforce-training services are more easily measured, thus increasing the use of performance-based contracts, whereas other services, such as AIDS services, can be more difficult to measure.

⁸ Examples include Arizona, Florida, Illinois, Maine, Massachusetts, and North Carolina (see Martin 2003).

⁹ At the Department of Health and Human Services, competitive sourcing and other contract innovations that have taken place under A-76 and revisions to the FARs are primarily for commercial activities, such as data processing.

¹⁰ A related point on the issue of how to manage is a second question of how public managers should contract for other less tangible, uncontractible issues, such as the courtesy and respect providers are to offer to clients, the use of judgment and discretion beyond the formalized parameters of the contract in order to better serve individuals, and efforts focused on searching for innovation in program intervention and treatment.

¹¹ Eisenhardt (1989, 64) notes that "agency theory reminds us that common problem structures do exist across research domains."

¹² It should be noted that the principal-agent relationship is not unidirectional. Just as agents may act opportunistically by exploiting information for their own self-interest, principals may also use information that agents do not possess for their own gain over and above the goals stated in the contract with the agent.

¹³ Conflict is resolved through the coalignment of incentives, the price mechanism, in economic models of agency theory while political power is the mechanism for resolving conflicts in political models of agency theory. See Eisenhardt (1989, 63).

¹⁴ There exists however, the opportunity for multiple principal-agent relationships in every contracting relationship. In some cases, government may formally be the principal in the contract framework, but the nonprofit may really possess information resources that give it the advantage in the relationship. In addition, government (principal) may contract with a nonprofit (agent) for service delivery, and the nonprofit (principal) may subcontract with another nonprofit (agent) to deliver a part of the overall service for which it has specialized expertise. To some extent, modeling multiple principal-agent relationships can appear as complex as modeling a network analysis of the multiple linkages within a policy area.

¹⁵ This could certainly lead to uncertainty, especially with respect to political support and funding availability.

¹⁶ Salamon (1995) has estimated that nonprofit social service providers receive approximately a third of their revenue from government through grants and contracts.

¹⁷ The term stewardship is a popular term in the nonprofit management and philanthropy literatures and does convey the notion of a steward acting on behalf of a benefactor in terms of fulfilling mission goals with accountability (see Fogal 2005). However, moving from the term to a theory of stewardship as an organizing framework for thinking about alignment between contractual parties is a relatively recent development.

¹⁸ In some respects, these attributes mirror some of the characteristics cited in the developing literature on relational contracting. This is an important area for future research but one that is beyond the scope of the current study.

¹⁹ Former Indianapolis Mayor Goldsmith used what he described as the yellow pages test of service selection when considering contracting. If more than five agencies (private or nonprofit) could be identified as providing the service, then that service became a candidate for contracting. However, Adrian Moore of the Reason Foundation disputes the notion that a specific number of providers are needed and argues that there must be at least two competitive providers and an open solicitation and bidding process. Van Slyke (2003) finds that contracting often takes place for social services even when neither of these conditions are met. I thank the reviewers for this suggestion.

²⁰ The National Association of Counties cites California and New York as two of the best examples of counties providing general assistance as a result of a state mandate.

²¹ See the New York State Catalog of State and Federal Programs Aiding New York's Local Governments at http://assembly.state.ny.us/comm/StateLocal/20031120/.

²² Due to issues of confidentiality and requests for anonymity, all the counties, interviewee names and titles, and names of the nonprofit organizations have been omitted. In order to gain access to "elites" with specialized knowledge, the respondents were guaranteed complete anonymity given their concerns that the nature of their comments could have adverse affects on the contract relationships they are involved in and repercussions from agency superiors and elected officials. The following terms are used to describe the respondents. The term "public managers" represents all those individuals interviewed at either the state or county level with responsibility and authority for managing social service contract relationships with nonprofit organizations. The individuals were political appointees and civil servants. In four of the five counties this meant interviewing the commissioner and deputy commissioner, separately, as well as division directors. A commissioner and deputy commissioner would be roughly equivalent to the number one and two positions within a social service agency. At the state level all the managers were bureau chiefs or division directors. The respondents representing nonprofit organizations were all executive directors.

²³ The Bureau of the Census defines a small county as having a population less than 100,000 based on 1998 total resident population. A medium county ranges from 100,000 to 250,000. A large county is 250,000 and above. Based on these population ranges as a definition of county size, there are 15 large counties, 11 medium counties, and 36 small counties in New York State. (Bureau of the Census New York State population data, July, 1998). Two large, two medium, and one small county were examined in this research study.

²⁴ Other interviews were conducted; however, there is missing data for some of the questions. Therefore, the 35 interviews used here represent complete data cases. This lends itself to greater levels of generalizability. Of the 35 interviewees, 19 were male and 16 female, 5 of whom were between the ages of 30 and 39, 21 between 40 and 49, and 9 between 50 and 62. Thirty-four percent had completed graduate work and 97% had a bachelor's degree.

²⁵ The coding was guided by the theories identified and involved memoing and concept mapping, two techniques specifically designed to avoid interjecting subjective bias by the researcher. In addition, both techniques contribute to the sorting of some codes and integration of others.

²⁶ This excludes New York City and Long Island.

²⁷ Such satisfaction measures though were largely absent or unavailable.

²⁸ All the county-level public managers' remarks on this issue were in general agreement with this statement, though few gave examples of how they personally enhance their contractors' reputations.

²⁹ In this study, this type of monitoring was fairly limited. But this type of contract does raise questions about arms-length transactions. In contracts with trade associations whose missions consist of an accreditation component, local governments contract with the associations for its expertise in evaluating contractor compliance with accreditation standards. The providers being evaluated are also in many cases members of the very association conducting the evaluation and oversight. Although no dereliction of duty and divergence had been identified, this form of contract relationship points to the potential conflict of interest that accompanies contracting for oversight and compliance.

³⁰ In GAO-02-245 (28), the following statement appears: "According to New York State program officials, contracting agencies in the state continue to experience ongoing shortfalls in staff resources necessary to provide sufficient oversight of contractor performance."

³¹ The COQC is a classic principal-agent problem in that the Commission serves at the pleasure of the legislature and is charged with uncovering fraud and abuse among mental health providers. Therefore, auditors from the COQC and auditors from OMH are simultaneously monitoring the activities of OMH providers under contract. The OMH managers provide technical assistance. For the COQC auditors though, there is a disincentive to offer assistance because their mission, largely regulatory, is to identify and sanction providers that are out of compliance. The more they identify the more they fulfill their own mission and secure their continued existence. In addition, the more malfeasance the COQC identifies that has not been identified by OMH, the more they demonstrate the need and credibility of their legislatively mandated service. Their Web site http://www.cqc.state.ny.us/ provides a listing of reports in which contractor malfeasance and goal divergence have been identified. See for example, "Exploiting Not-For-Profit Care in an Adult Home: The Story Behind Ocean House Center, Inc." December 2001.

³² Certainly, there were agents/stewards who played out during the course of the interviews the degree to which they could avail themselves of opportunities to use information for their own interest with little risk of the principal learning of their actions. Public managers reluctantly agreed. Even somewhat benign examples provide an ethical paradox to be considered by public managers. Case in point was a nonprofit provider under contract to provide a particular type of service. The provider was not monitored closely and was managed in a steward-like manner with regard to information exchange, the letting of reimbursements, and the trust and reputation conferred on preferred providers. In this case, the provider was maximizing the per-client reimbursement rates from the state and using the positive differential between what it actually cost the provider to treat a client and what it received from the state in reimbursement. The provider used that differential to begin funding an innovative and experimental program, one akin to a demonstration project. The provider did not share this information with the public agency and manager funding and overseeing the contract. The new program designed by the provider turned out to be successful and quite innovative. The provider then tried to sell the program to the public manager. When asked by government officials how the provider organization was able to pay for the development and testing of this new program, the nonprofit executives essentially responded by saying they used contract monies to fund this new program. On the surface of this issue, did a moral hazard problem develop? Was there benign intent underlying the providers' actions? In the eyes of the public managers overseeing this contract, moral hazard did occur. An agent had used asymmetric information for its own gain, and opportunism did take place. The result: the provider was informally sanctioned in that reimbursement funds were released incrementally and often delayed, the provider was referred to as "scant" within the public management contract networks, its reputation and trust tarnished, its discretion and legitimacy stripped, and formally sanctioned in that the contract was rebid and they lost the contract to a preferred provider. The innovative program did not receive funding. These sanction mechanisms were the outlier and not the norm when monitoring.

³³ One example is a report published jointly by the Assembly Children and Families Committee and The Oversight, Analysis, and Investigation Committee titled "Losing Our Children." In this report, a critical finding is that "the state agency culture leaves operations to the districts, without ensuring that laws are being implemented and clients are being well served" (Green and Parment 2001, 16).

	References

TOP Abstract INTRODUCTION BACKGROUND METHODS AND DATA FINDINGS: AGENTS OR STEWARDS? CONCLUSION References Notes

 

Albanese Robert, Dacin Tina M., Harris Ira C. (1997) Agents as stewards. Academy of Management Review 22:609–11.

Alchian Armen and Demsetz Harold. (1972) Production, information costs, and economic organization. American Economic Review 62:777–95.

Alexander Jennifer, Nank Renee, Stivers Camilla. (1999) Implications of welfare reform: Do nonprofit survival strategies threaten civil society? Nonprofit and Voluntary Sector Quarterly 28:452–75.[Abstract/Free Full Text]

Beinecke Richard H. and DeFillippi Robert. (1999) The value of the relationship model of contracting in social services reprocurements and transitions: Lessons from Massachusetts. Public Productivity and Management Review 22:490–501.

Bogart William T. (1995) Accountability and nonprofit organizations: An economic perspective. Nonprofit Management and Leadership 6:157–69.

Brody Evelyn. (1996) Agents without principals: The economic convergence of the nonprofit and for-profit organizational forms. New York Law School Law Review 40:457–536.

Brown Trevor L. and Potoski Matthew. (2004) Managing the public service market. Public Administration Review 64:656–68.

Bundt Julie. (2000) Strategic stewards: Managing accountability, building trust. Journal of Public Administration Research and Theory 10:757–77.[Abstract/Free Full Text]

Cooper Phillip J. (2003) Governing by contract: Challenges and opportunities for public managers(CQ Press, Washington, DC).

Davis JH, Donaldson L, Schoorman FD. (1997a) Toward a stewardship theory of management. Academy of Management Review 22:20–47.

———. (1997b) Davis, Schoorman, and Donaldson reply: The distinctiveness of agency theory and stewardship theory. Academy of Management Review 22:611–3.

DeHoog Ruth Hoogland. (1984) Contracting out for human services: Economic, political, and organizational perspectives(State Univ. New York Press, Albany, NY).

———. (1990) Competition, negotiation, or cooperation: Three models for service contracting. Administration and Society 22:317–40.

Denzin Norman K. and Lincoln Yvonne S. (1998) Collecting and interpreting qualitative materials(SAGE Publications, Newbury Park, CA).

Dharwadkar Ravi, George Gerard, Brandes Pamela. (2000) Privatization in emerging economies: An agency theory perspective. Academy of Management Review 25:650–69.

Dicke LA. (2002) Ensuring accountability in human services contracting: Can stewardship theory fill the bill? American Review of Public Administration 32:480–95.

Donahue John D. (1989) The privatization decision: Public ends, private means(Basic Books, New York).

Donaldson L. (1990) The ethereal hand: Organizational economics and management theory. Academy of Management Review 15:369–81.

Eisenhardt Kathleen M. (1989) Agency theory: An assessment and review. Academy of Management Review 14:57–74.

Fogal Robert E. (2005) Designing and managing the fundraising program. In Herman Robert D. and Associates Robert D. (Eds.). The Jossey-Bass handbook of nonprofit leadership and management(John Wiley and Sons, San Francisco, CA) pp. 419–35.

Frey Bruno S. (1993) Does monitoring increase work effort? The rivalry with trust and loyalty. Economic Inquiry 31:663–70.

Frumkin Peter. (2001) Managing for outcomes: Milestone contracting in Oklahoma. (The Price Waterhouse Coopers Endownment for the Business of Government., Arlington, VA) http://www.businessofgovernment.org/main/publications/grant_reports/details/index.asp?GID=56 (accessed January).

General Accounting Office. (2002) Welfare reform: Interim report on potential ways to strengthen federal oversight of state and local contracting GAO–02–245.

Goldsmith Stephen. (1997) The Twenty-First Century City: Resurrecting Urban America(Regnery Publishing, Washington, DC).

Green Roger L. and Parment William L. (2001) Losing our children: An examination of New York's foster care system(The New York State Assembly Committee on Children and Families and the Committee on Oversight, Albany, NY).

Hardin R. (2002) Trust and trustworthiness(Russell Sage Foundation, New York, NY).

Heinrich Carolyn J. (2000) Organizational form and performance: An empirical investigation of nonprofit and for-profit job-training service providers. Journal of Policy Analysis and Management 19:2233–61.

———. (2002) Outcomes-based performance management in the public sector: Implication for government accountability and effectiveness. Public Administration Review 62:712–25.

Jensen Michael C. and Meckling William H. (1996) Theory of the firm: Managerial behavior, agency costs, and ownership structure. In Buckley Peter J. and Michie Jonathan (Eds.). Firms, organizations, and contracts: A reader in industrial organization(Oxford Univ. Press, Oxford, NY) pp. 103–67.

Johnston Jocelyn M. and Romzek Barbara S. (1999) Contracting and accountability in state medicaid reform—Rhetoric, theories, and reality. Public Administration Review 59:383–99.

Kelman Steven J. (2002a) Contracting. In Salamon Lester M. (Ed.). The tools of government: A guide to the new governance(Oxford Univ. Press, New York, NY).

———. (2002b) Strategic contracting management. In Donahue John D. and Nye Joseph S. Jr (Eds.). Market-based governance: Supply side, demand side, upside, and downside(Brookings Institution Press, Washington, DC).

Kettl Donald F. (1993) Sharing power: Public governance and private markets(Brookings Institution, Washington, DC).

Kettl Donald F. and Brinton Milward H. (1996) The state of public management(Johns Hopkins Univ. Press, Baltimore, MD).

Kramer Ralph M. (1994) Voluntary agencies and the contract culture: Dream or nightmare? Social Service Review 68:33–60.

Lee Peggy M. and O'Neill Hugh M. (2003) Ownership structures and R&D investments of U.S. and Japanese firms: Agency and stewardship perspectives. Academy of Management Journal 46:212–25.

Lofland John and Lofland Lyn H. (1995) Analyzing social settings: A guide to qualitative observation and analysis(Wadsworth Publishing Company, Belmont, CA).

Lynn Laurence E. (1996) Public management as art, science, and profession(Chatham House Publishers, Chatham, NJ).

Martin Lawrence L. (2003) Performance-based contracting for human services: A review of the literature. Working Paper No. 1 (December), Center for Community Partnerships, University of Central Florida.

———. (2004) Performance-based contracting for human services: Does it work? Administration in Social Work 29:163–77.

Mayer Roger C., Davis James H., David Schoorman F. (1995) An integrative model of organizational trust. Academy of Management Review 20:709–34.

McGubbins Matthew D., Noll Roger G., Weingast Barry R. (1987) Administrative procedures as instruments of political control. The Journal of Law, Economics, and Organization 3:243–86.

Meyers Marcia K., Glaser Bonnie, MacDonald Karin. (1998) On the front lines of welfare delivery: Are workers implementing policy reforms? Journal of Policy Analysis and Management 17:1–22.

Meyers Marcia K., Riccucci Norma M., Lurie Irene. (2001) Achieving goal congruence in complex environments: The case of welfare reform. Journal of Public Administration Research and Theory 11:165–201.[Abstract/Free Full Text]

Miles Matthew B. and Huberman Michael A. (1994) Qualitative data analysis: An expanded sourcebook 2nd ed (SAGE Publications, Newbury Park, CA).

Milgrom Paul and Roberts John. (1992) Economics, organizations, and management(Prentice Hall, Inc, Englewood Cliffs, NJ).

Milward H. Brinton. (1994) Implications of contracting out: New roles for the hollow state. In Ingraham Patricia and Romzek Barbara S. (Eds.). New paradigms for government: Issues for the changing public service(Jossey-Bass, San Francisco, CA) pp. 41–62.

Milward H. Brinton and Provan Keith G. (2000) Governing the hollow state. Journal of Public Administration Research and Theory 10:359–80.[Abstract/Free Full Text]

Perrow C. (1986) Complex organizations(Random House, New York).

Riccucci Norma M. (2005) Street-level bureaucrats and intrastate variation in the implementation of Temporary Assistance for Needy Families policies. Journal of Public Administration Research and Theory 15:89–111.[Abstract]

Riedinger Susan A., Aron Laudan Y., Loprest Pamela J., O'Brien Carolyn T. (1999) Income support and social services for low-income people in New York. Urban Institute http://newfederalism.urban.org/html/isss_ny.html (accessed June21, 2000).

Romzek Barbara S. and Johnston Jocelyn M. (2002) Effective contract implementation and management: A preliminary model. Journal of Public Administration Research and Theory 12:423–53.[Abstract/Free Full Text]

Ross SA. (1987) The interrelations of finance and economics: Theoretical perspectives. American Economic Review 77:29–34.

Rousseau Denise M. (1995) Psychological contracts in organization: Understanding written and unwritten agreements (SAGE Publications, Thousand Oaks, CA).

Saidel Judith R. (1991) Resource interdependence: The relationship between public agencies and nonprofit organizations. Public Administration Review 51:543–53.

Salamon Lester M. (1989) Beyond privatization: The tools of government action(The Urban Institute Press, Washington, DC).

———. (1995) Partners in public service: Government-nonprofit relations in the modern welfare state(Johns Hopkins Univ. Press, Baltimore, MD).

Sandfort Jodi R. (2000) Moving beyond discretion and outcomes: Examining public management from the front lines of the welfare system. Journal of Public Administration Research and Theory 10:729–56.[Abstract/Free Full Text]

Sappington David E. M. (1991) Incentives in principal-agent relationships. Journal of Economic Perspective 5:245–66.

Savas ES. (1987) Privatization: The Key to Better Government(Chatham House Publishers, Inc, Chatham, NJ).

———. (2000) Privatization and public-private partnerships(Chatham House Publishers, Inc, Chatham, NJ).

———. (2002) Competition and choice in New York City social services. Public Administration Review 62:82–91.

———. (2005) Privatization in the city: Successes, failures, lessons(CQ Press, Washington, DC).

Sclar Elliott D. (2000) You don't always get what you pay for: The economics of privatization(Cornell Univ. Press, Ithaca, NY).

Shleifer Andrei and Vishny Robert W. (1998) The grabbing hand: Government pathologies and their cures(Harvard Univ. Press, Cambridge, MA).

Smith Steven Rathgeb and Lipsky Michael. (1993) Non-profits for hire: The welfare state in the age of contracting(Harvard Univ. Press, Cambridge, MA).

Smith Steven Rathgeb and Smyth Judith. (1996) Contracting for services in a decentralized system. Journal of Public Administration Research and Theory 6:277–96.[Abstract/Free Full Text]

Van Slyke David M. (2003) The mythology of privatization in contracting for social services. Public Administration Review 63:277–96.

Waterman Richard W. and Meier Kenneth J. (1998) Principal-agent models: An expansion? Journal of Public Administration Research and Theory 8:173–202.[Abstract/Free Full Text]

Whitener Ellen M., Brodt Susan E., Audrey Korsgaard M, Werner Jon M. (1998) Managers as initiators of trust: An exchange relationship framework for understanding managerial trustworthy behavior. Academy of Management Review 23:513–30.

Wood B. Dan and Waterman Richard W. (1991) The dynamics of political control of the bureaucracy. American Political Science Review 85:801–28.[CrossRef][Web of Science]

Yamagishi T and Yamagishi M. (1994) Trust and commitment in the United States and Japan. Motivation and Emotion 18:129–66.

CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				J. C. Thomas, T. H. Poister, and N. Ertas Customer, Partner, Principal: Local Government Perspectives on State Agency Performance in Georgia J. Public Adm. Res. Theory., September 2, 2009; (2009) mup024v1. [Abstract] [Full Text] [PDF]

				A. A. Amirkhanyan Collaborative Performance Measurement: Examining and Explaining the Prevalence of Collaboration in State and Local Government Contracts J. Public Adm. Res. Theory., July 1, 2009; 19(3): 523 - 554. [Abstract] [Full Text] [PDF]

				J. Alexander and R. Nank Public--Nonprofit Partnership: Realizing the New Public Service Administration Society, May 1, 2009; 41(3): 364 - 386. [Abstract] [PDF]

				K. T. Lambright Agency Theory and Beyond: Contracted Providers' Motivations to Properly Use Service Monitoring Tools J. Public Adm. Res. Theory., April 1, 2009; 19(2): 207 - 227. [Abstract] [Full Text] [PDF]

				K. T. Lambright Getting What You Ask For: Barriers to Proper Use of Service Monitoring Tools The American Review of Public Administration, September 1, 2008; 38(3): 362 - 379. [Abstract] [PDF]

				J. M. Johnston and B. S. Romzek Social Welfare Contracts as Networks: The Impact of Network Stability on Management and Performance Administration Society, April 1, 2008; 40(2): 115 - 146. [Abstract] [PDF]

This Article Abstract FREE Full Text (PDF) All Versions of this Article: 17/2/157    most recent mul012v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (2) Request Permissions Google Scholar Articles by Van Slyke, D. M. Search for Related Content Social Bookmarking          What's this?

Online ISSN 1477-9803 - Print ISSN 1053-1858

Copyright © 2009 Public Management Research Association

Oxford Journals Oxford University Press

• Site Map
• Privacy Policy
• Frequently Asked Questions

Other Oxford University Press sites: Oxford University Press Oxford Journals China Oxford Journals Japan American National Biography Booksellers' Information Service Children's Fiction and Poetry Children's Reference Corporate & Special Sales Dictionaries Dictionary of National Biography Digital Reference English Language Teaching Higher Education Textbooks Humanities International Education Unit Law Medicine Music Online Products Oxford English Dictionary Reference Rights and Permissions Science School Books Social Sciences Very Short Introductions World's Classics