Principal-Agent Theory of Organizations
Theory of interaction between an agent and the principal for whom they act, the point being to structure incentives so that the agent will act to benefit the principal.
The principal-agent theory of organizations (“agency theory” from here on) encapsulates the idea that public sector performance can be improved if incentive-based contracts between different actors are implemented. Principals will be more likely to achieve their desired outcomes, while agents will have clarity around work programs and goals. Agency theory has had considerable influence on the theory and practice of public administration and policy since its emergence in the 1970s. It was particularly instrumental in many high-income developed countries through the 1980s and 1990s, with often radical public sector reforms resulting. Its legacy has endured, with many public sector organizational and policy designs continuing to be underpinned by concepts derived from the theory. Based on institutional economics, agency theory has, therefore, provided a powerful and all-encompassing framework for public sector organization. As such, there has been much written about agency theory itself, and about public sector contracting, which is a central tenet (see for instance Ashton et al. 2004; Klingner et al. 2002; Lane 2001; Pallesen 2004; Pinch and Patterson 2000). There is less literature that discusses the longer-term policy outcomes when agency theory has been an overarching influence on public sector organization (Gauld 2007).
This chapter provides an overview of agency theory, including the key ideas behind it and the organizational and policy arrangements that are derived from it. The chapter outlines the benefits that might be expected to result when the theory is applied to public sector organization. It also highlights agency theory’s shortcomings. Finally, it notes areas where theoretical extension is demanded.
What Is Agency Theory?
Agency theory has its foundations in two ideas which were developed through the study of the economics of organizational and institutional behavior.
First was the notion proposed by public choice theorists that self-interest is the primary motivation behind the activities and behavior of individuals and the organizations that they work for (Mitchell and Simmons 1994; Mueller 1989). Through this lens, people in public employment and public organizations are viewed as “rational utility maximizers,” meaning that each seeks to advance their own interests as would a private business or private sector employee in pursuit of profit or an increased salary. In this way, government officials and organizations are presumed to be in pursuit of only budgetary expansion; politicians, for their part, are motivated by the prospect of an expanded share of votes; interest groups, who represent specific sector groups or services users, are only concerned with furthering their own ends and those of their members. Public choice theory suggests that the result of such behaviors is a state which is larger than it should be, along with policies which are designed primarily to serve voter preferences and boost voter support for politicians and political parties, and an economy which is skewed through meeting the demands of selected interests over and above those of the broader public interest. To counter this, public choice theorists advocate limits on the power of politicians, interest groups, and public officials, as well as the implementation of financial incentives and sanctions to ensure appropriate performances.
The second idea, again based on a presumption that self-interest and rational utility maximization drive behavior, is the view that all of life, including public work, private lives, and organizational activities, can be viewed as a set of relationships between different parties (Moe 1984; Perrow 1986). The details of these relationships, once defined, and the requirements of the various parties in any particular venture or activity can be itemized and then written into a formal contract. The contract can, in turn, be deployed for purposes such as setting expectations and objectives of contract partners and for establishing performance assessment and accountability expectations.
Advocates of agency theory and related organizational arrangements presume that contracting will align the interests of principals (those wanting something done, such as politicians, funding agencies, or chief executives) and their agents (e.g., government officials and organizations or non-government and private service providers of public services). The result, in theory, is that the achievement of principals’ objectives will be maximized, resulting in a more efficient and effective policy and service delivery outcomes. Alongside of this, the self-interested behavior of agents will be stemmed and focused on principals’ goals via various incentives and sanctions. These might include anything from withholding of service delivery payments through to organizational or individual employee performance bonuses (detailed discussion of agency and related theory can be found in Dixit 2002; Laffont and Mortimort 2002; Mueller 1989; Self 1993; Stretton and Orchard 1994; Wallis and Dollery 1999; Walsh 1995).
Factors that Complicate Application of Agency Theory
Complicating agency theory is a series of behavioral factors encapsulated by the terms “adverse selection” and “moral hazard,” as well as the very nature of the public sector and government. Adverse selection results from the existence in any relationship of what are called “information asymmetries.” This refers to the simple fact that one party (for instance, an experienced and skilled public servant) may be likely to have more knowledge, and therefore be at an advantage, than another party (such as a politician). Adverse selection can occur when, for example, a principal (a politician) is not able to gain sufficient knowledge about an agent’s (a public servant’s) background, motivations, or capabilities prior to entering into a contractual relationship (Perrow 1986). It can pose particular problems in any contracting situation and require considerable investigation to ensure that the potential for adverse selection related-difficulties is reduced.
The risk of “moral hazard” arises once a contract has been agreed to. It stems from the fact that, on a day-to-day basis, principals are not able to observe most agent activity (Moe 1984). Principals are, therefore, reliant on agents carrying out tasks and performing at a level as specified in a contract. The ever-present prospect of moral hazard means an ongoing requirement for monitoring. This can result in considerable costs to both contracting parties, as well as goal-displacement behavior on behalf of agents where they place a disproportionate emphasis on work that is specifically subject to monitoring. By goal displacement, this means that agents focus on monitored goals, to the detriment of other organizational and individual goals that may not be directly monitored.
A further factor complicating agency theory is a range of circumstances particular to government and public sector work. These include that policy refinement is frequently left to the implementation process and is routinely the responsibility of agents, being public officials and not principals (Hill and Hupe 2002); most government agencies have several and often conflicting tasks and objectives which can be difficult to define and itemize (Wilson 1989); multiple principals and agents characterize the public sector as do situations in which principals often double as agents; the public sector tends to lack competition, at least in terms of core government non-trading functions (Allison 1979); and the public workforce and agencies are motivated by a complex array of factors, only one of which might be financial incentives (Dixit 2002; Le Grand 2003; Thaler 2015; Wilson 1989).
Improving Agency Theory
There is now considerable international experience with application of agency theory to public sector organization. It has provided the foundations for public sector reforms in a range of high-income countries. While there is an absence of research into whether long-term public sector performance has improved as a result, and it could be difficult to determine direct causality, it may be fair to suggest that performances have continued to be questionable. Indeed, politicians in most high-income countries, along with international agencies, continue to demand improvement in a context where policy challenges are increasingly complex (e.g., issues ranging from how to deal with population aging and chronic disease through to inequality and climate change which require considerable cross-government coordination and long-range planning). This is despite agency theory having an ongoing and influential role as a framework for establishing the roles of principals and agents and, in turn, for managing policy development and implementation.
Principals (in this case, political leaders and public officials at different levels of government and the health system) often may not have sufficiently detailed knowledge of what they want when setting parameters and building incentives and goals for agents. If they do, and decide upon a certain policy path, then this has implications if political and policy preferences are subsequently altered. To counter such problems, principals may need a longer-term view, as discussed below.
Principals may not recognize in advance the ramifications of the directions they set. In response, it might be suggested that principals clearly need to model in detail the possible outcomes of various policy options. Such an approach, however, would be subject to the widely noted limits of “rational” policymaking (Lindblom 1959).
Agency theory fails to adequately account for situations when principals regularly change. Such changes, for instance in political leadership, can be a harbinger for change in policy directions and the organization of the public sector. Agency theory may simply need to be amended to include reference to the fact that arrangements inspired by it cannot be relied upon to produce continuity in transitions between principals.
Following the previous point, when the administrative systems within which agents work are regularly restructured by principals, with the potential to induce confusion and chaos, this undermines the assumption inherent to agency theory that principals are in control and capable of providing consistent direction over time; it also disrupts administrative processes and continuity, creating obstacles to effective individual and aggregate agent activity. A consequence is that agents may develop their own methods and systems for working and fail to coordinate with one another. If incentives are to be relied upon for complex organization and in scenarios of transition, then their design needs to be sophisticated. A simple vertical contract between principals and agents may not be enough; horizontal contracts between agents may also be required, and these may need to be coordinated by principals. By implication, this means that principals need the detailed knowledge and foresight discussed in points 1 and 2 above, along with commitment to stability implied in point 3.
Thus, for agency theory to be an effective organizing principle now and into the future, more intricacy in its development may be required (Deacon 2004). Yet even if an approach that accounted for the four points above were applied, this may be too constrained for the exigencies of politics, the multitudinous motivations of individuals and public organizations, the ever-changing and increasingly complex nature of administration and society, and the challenges for the foreseeable future. It may also prove to be administratively demanding, raising questions over transaction costs.
If so, then alternatives may need to be considered. One possibility is for increasing the level of centralized policy control and sector oversight, in short, the growth of a more comprehensive iteration of agency theory that incorporates the coordination of principals and agents. If contracting is to remain a fundamental principle of public sector organization, then longer-term contracts, combined with standardized objectives and funding levels (inherent to comprehensive policy), may promote more commitment among agents to centrally driven directions. This points to the need for development of a consensus-based variant, bringing together principals and agents, with an aim of stewardship over a subset of contractual agreements between interested parties. In practice, this may mean lengthy consultation processes that involve policymakers, key interests, industry, and provider groups with a primary aim of forging long-term goals. Such an approach might be underpinned by a focus on outcome-oriented policies being increasingly pursued by various national governments (Baehler 2003; Christensen and Laegreid 2013; Hoque 2008). Forging a long-term, cross-government view may carry with it a risk of committing to particular directions that, following any future political change, could be rejected. Consensus may, also, over time lead to policy embeddedness and establishment of a set of institutions that would be resistant to future reform (see Blank and Burau 2004; Putnam 1993: 179; Wilsford 1994). This may be appropriate when effective responses are demanded to complex policy problems such as those listed earlier in this chapter.
Short of the developments described in the previous section, which might be seen by agency theory proponents as undermining competitive incentive systems, the rudimentary method of simple contracts between principals and agents may fail to provide an effective developmental foundation for multifarious administrative issues.
- Allison G (1979) Public and private management: are they fundamentally alike in all unimportant respects? In: Shafritz J, Hyde A (eds) Classics of public administration. Brooks/Cole Publishing Company, Pacific GroveGoogle Scholar
- Blank RH, Burau V (2004) Comparative health policy. Palgrave Macmillan, HoundmillsGoogle Scholar
- Christensen T, Laegreid P (eds) (2013) Transcending new public management: the transformation of public sector reforms. Ashgate, SurreyGoogle Scholar
- Hill M, Hupe P (2002) Implementing public policy. Sage, LondonGoogle Scholar
- Laffont J-J, Mortimort D (2002) The theory of incentives: the principal-agent model. Princeton University Press, PrincetonGoogle Scholar
- Mitchell W, Simmons R (1994) Beyond politics: markets, welfare, and the failure of bureaucracy. Westview Press, BoulderGoogle Scholar
- Mueller D (1989) Public choice II. Cambridge University Press, CambridgeGoogle Scholar
- Perrow C (1986) Complex organizations: a critical essay. Random House, New YorkGoogle Scholar
- Putnam R (1993) Making democracy work: civic traditions in modern Italy. Princeton University Press, PrincetonGoogle Scholar
- Thaler RH (2015) Misbehaving: the making of behavioural economics. Allen Lane, LondonGoogle Scholar
- Wilson JQ (1989) Bureaucracy: what government agencies do and why they do it. Basic Books, New YorkGoogle Scholar