Keywords

1.1 An Overview of the Agency Theory

Agency theory, in its modern form, largely originates from the works of Mitnick (1973) and Ross (1973) and embraces the areas of political science and economics. Nevertheless, many previous works on economic risk analysis and individual–group goal incongruence have contributed to its evolution (e.g. Arrow 1985; Wilson 2006). In the management literature, the tenets of agency theory have been adapted and developed by Eisenhardt (1989) and Jensen and Meckling (1976).

In agency relationships, one party (the principal) delegates a task/activity to another party (the agent). Agency theory explicitly addresses under which contractual arrangements the relationship between principal and agent operates most efficiently. The notion of the contract is used as a metaphor to describe agency relationships (Jensen and Meckling 1976). The contract is designed considering the outcome or behaviour of the agent (Eisenhardt 1989). In agency relationships, typically, the principal seeks to minimise the agency costs—such as specifying, rewarding and monitoring the agent’s behaviour—while the agent works towards maximising rewards and reducing the principal’s control (Fayezi et al. 2012). The agency theory considers some basic assumptions about people (e.g. self-interest, bounded rationality, risk aversion), organisations (e.g. goal conflict among members) and information (e.g. information is a commodity which can be purchased) (Eisenhardt 1989). In each principal–agent relationship, the basic question becomes: “Is a behaviour-oriented contract (e.g., salaries, hierarchical governance) more efficient than an outcome-oriented contract (e.g., commissions, stock options, transfer of property rights, market governance)?” (Eisenhardt 1989: p. 58). Efficient management of agency problems, such as information acquisition, preference mismatch (or conflict of interest), effort (or moral hazard) and capability (or adverse selection), mainly associated with the agent (Fleisher 1991), is also imperative to any principal–agent relationship (Fayezi et al. 2012). It is important to note that the original formulation of agency theory defines efficiency from the principal’s point of view. This assumption is based on the recognition of the principal as the dominant party in the relationship. Thus, an efficient contract is “one that brings about the best possible outcome for the principal given the constraints imposed by the situation, rather than one that maximises the joint utility of both principal and agent” (Bergen et al. 1992).

The recent development of agency theory can be categorised into two streams of inquiry: the principal–agent research and the positivist agency theory. The principal–agent approach assumes that the principal and agent will attempt to maximise their positions through individual interpretation of the contract. This approach is grounded in economic research and widely adopts mathematical modelling for relationship building (Eisenhardt 1989). In contrast, the positivist agency theory adopts a descriptive logic to interpret the real-world behaviours by using the agency theory logic and assumptions. Even if both approaches have been widely discussed in academic settings, “the mathematical and non-empirical orientation of principal-agent research have been the primary causes of this stream of research stagnating academically, particularly in organisational research” (Fayezi et al. 2012: p. 557).

Positivist agency theory has been largely extended in order to synthesise different disciplines—such as political science, expert agency and sociology—into a comprehensive framework. It aims to explain the main features of business and government relationships and offer suggestions to improve their efficiency and effectiveness (Shapiro 2005). This stream of research also provides explanations for the minimisation of a number of economic and business issues, such as separation of control from ownership (Fama and Jensen 1983; Halldórsson and Skjøtt-Larsen 2006) and the non-rational behaviour of agents (Jensen and Meckling 1976).

A principal faces two kinds of (agency) problems when entering a relationship with an agent: pre-contractual and post-contractual. The first arises before the principal decides to offer an agent a contract. The major issues here are: (a) the screening of the agent’s abilities and intentions in order to find the most appropriate characteristics for the required task and (b) the definition of the strategy adopted by the principal to detect and control such abilities. The post-contractual problems emerge after the principal and agent engage in a contractual relationship. The major issues here are: (1) the definition of the incentive structure able to motivate the agent to behave in a manner consistent with the principal’s goal and (2) the definition of the principal’s information strategy essential for the evaluation and reward of the agent’s performance. Compensations or incentive programmes for employees are typical outcomes of such decisions (Bergen et al. 1992). The literature on agency theory refers to contractual issues as problems of “adverse selection or hidden information” and post-contractual issues as problems of “moral hazard or hidden action” (Arrow 1985; Hart and Holmström 1987).

1.1.1 Hidden Information and Action Models

In order to address the issues related to misrepresentation of ability (adverse selection) and lack of effort (moral hazard) of the agent, agency theory research proposes two different approaches, respectively, designed to help the principal in designing the appropriate contract: the hidden information and the hidden action models (Bergen et al. 1992; Fayezi et al. 2012). Both models work under the informative asymmetry assumption according to which the principal is assumed to know the nature and characteristics of the task as well as the abilities needed for the agent to successfully perform the task.

Hidden information models recognise the impossibility for the principal to know if the agent has the required abilities to accomplish the contractual obligations and if he/she has the intention to do so. The principal can attempt to overcome these problems in at least three ways: screening, examining signals from the potential agents or providing opportunities for self-selection (Bergen et al. 1992). The screening strategy refers to the principal’s attempt to unveil the actual abilities and attributes of the potential agent by using different means such as, for example, personal interviews and references. Alternatively, the principal might hire the agent on the basis of the imperfect information provided by the agent’s signals and then learn from subsequent experience the true abilities and intentions (Bergen et al. 1992). It follows that screening is most likely to be an efficient solution to the hidden information problem when it is relatively easy to obtain information about the agent.Footnote 1 In the signalling strategy, the agent (who knows his/her own capabilities) engages in actions with the only aim being to communicate (i.e. signal) to the principal that he/she has the abilities required to perform the principal’s task successfully. For such actions to yield a worthwhile signal from the agent’s point of view, however, “it must meet the criteria of individual rationality and incentive compatibility. The action must make the agent better off than he/she would be by doing nothing and the benefits to be derived from the action must exceed its cost” (Bergen et al. 1992: p. 6). However, the principal is not a merely passive receiver of signals from different potential agents. He/she might also proactively construct choices or provide opportunities that enable potential agents to signal their abilities and willingness to expend effort through self-selection. A training programme for new recruits is a good example of opportunities for agents’ self-selection.

Hidden action models help the principal in choosing an appropriate contract to mitigate moral hazard and motivate the agent to take the desired action given three basic assumptions: (1) principal and agent are assumed to be motivated by self-interest; (2) the principal takes decisions under conditions of incomplete information; and (3) the relationship’s outcomes are partly determined by environmental factors such as economic conditions, technological changes and environmental uncertainty in addition to the agent’s behaviour. According to Bergen et al. (1992), another assumption undermines the efficiency of principal–agent relationships: the different risk preferences of the two actors. In fact, “in most hidden action models the principal is assumed to be risk neutral—or at least more neutral than the agent—whereas agents are typically assumed to be risk averse”Footnote 2 (Bergen et al. 1992: p. 4). Different risk preferences between principal and agent lead to conflicting goals, thus orienting the two actors to different strategies and courses of action (Arrow 1985). The risk preference assumption suggests that the principal has more possibilities to diversify his/her investments than the agent. The latter is supposed to be more dependent on the principal and thus more inclined to inappropriate behaviours. However, anecdotal evidence suggests that agents are often prepared to accept unbalanced risks and have a number of different choices to reduce the dependence on—and thus the control from—the principal (see, for example, Basov and Bardsley 2005). The arguments above are particularly important in supply chain contexts, where factors relating to commitment and trust compensate for goal incongruence and different risk perceptions of the actors involved, and also outweigh in many situations contractual relationships and formal governance (Fayezi et al. 2012; O’Loughlin and Clements 2007).

1.2 Descriptive Analysis of the Literature Review on Agency Theory in Supply Chain Management Contexts

The need for an agency theory perspective in supply chain management contexts arises from the consideration of several shortcomings of Transaction Cost Economics (TCE) in explaining supply chain dynamics (see Fayezi et al. 2012 for additional details). Essentially, agency theory is supposed to provide a “mechanism that may be used to explain how players (both independently and as a collective) within the [supply chain] respond to transaction cost dilemmas where rational and non-rational behaviour occurs” (p. 558). According to Stock (1997), agency theory may provide alternative explanations for a number of issues relevant for both research and practice. Some of these are: (1) the development of inter- and intra-organisational relationships; (2) the maintenance of complex relationships between suppliers and customers; (3) the effects of supply chain decisions concerning risk sharing, conflictual relationships and power; and (4) identifying the costs and benefits of make-or-buy decisions (Logan 2000). In addition, there have been arguments in support of the combination of TCE and agency theory outside the supply chain management context (Williamson 1988) but additional research is necessary to show the real potential of such a combination.

In spite of its recognised explanatory power, in the supply chain management literature only a limited number of studies have adopted agency theory as a theoretical view.

Through a documentary research using electronic databases and a data reduction process, we have updated the literature review by Fayezi et al. (2012) on agency theory and its application in supply chain management contexts. A final list of 22 articles has been identified as relevant for the literature review’s objective, and these are analysed through a detailed content analysis. The literature survey has been undertaken using the following online databases: Emerald, Science Direct, Inderscience and ABI/Inform Global ProQuest. As suggested by Fayezi et al. (2012) and Delbufalo (2012), a three-stage process using data reduction has been adopted. The selected articles have been included/excluded in the process using keywords, title, abstract and conclusion. The inclusion criteria used in the literature screening process aim to unveil whether supply chain management (and related issues such as manufacturing, procurement and logistics) and agency theory arguments are contextually discussed in the abstract and conclusion of the selected articles. A synthesis of the content analysis is reported in Table 1.1.

Table 1.1 Summary of key SCM researches adopting agency theory
Full size table

Figure 1.1 shows the journals with agency theory applications within supply chain management contexts. Only a few outlets collected more than one study addressing the review topic. The International Journal of Logistics Management was the only journal hosting three of the selected articles. The information collected in the articles was then used to identify the main themes for agency theory and supply chain management literature shown below.

Fig. 1.1
figure 1

Journals with agency theory application within supply chain management

Full size image

1.3 Content Analysis and Stream of Research

From the analysis described above, a number of important streams of research and issues can be identified. In general, the literature review suggests two main considerations: (1) agency theory has been adopted in supply chain management research to address various areas such as procurement, manufacturing, logistics and new product development (NPD) and (2) positivist agency theory plays a dominant role in supply chain management, being adopted in the vast majority of the selected studies. The latter consideration corroborates the thesis according to which positivist agency theory can be more easily adopted by organisational researchers “mainly due to its non-mathematical, real-world oriented nature” (Fayezi et al. 2012: p. 562; Fama and Jensen 1983).

The majority of the selected studies (18 out of 22) adopted agency theory to address problems concerning buyer–supplier relationships. The remaining articles addressed inter-organisational problems (e.g. employer–employee) and other inter-organisational situations (e.g. supplier–retailer or supplier–third party). However, considering that in a supply chain multiple actors interact simultaneously, a discrete number of studies (3 out of 22) decided to address supply chain issues taking a triadic and tetradic perspective (Agrell and Norrman 2004; Agrell et al. 2004; Cheng and Kam 2008; Hornibrook 2007). The application of agency theory in networks is useful for researchers and managers aiming to understand how and under what conditions multiple principals and/or multiple agents interact with each other and impact on supply chain behaviours and intentions (Delbufalo and Bastl 2018; Surana et al. 2005).

The application of agency theory in supply chain management contexts follows the indications by Eisenhardt (1989) concerning the “theory-relevant contexts” (p. 70). Specifically, in her 1989 seminal work, Eisenhardt suggests that agency theory can be usefully adopted when:

  • principal and agent have different goals (i.e. goal incongruence or goal conflict) such as in buyer–supplier situations (Prosman et al. 2016);

  • evaluation of behaviours is difficult due to market or technological uncertainty or task complexity (Morgan et al. 2007; Simatupang and Sridharan 2002; Whipple and Roh 2010);

  • there is sufficient outcome uncertainty to increase the performance and relational risks (Fayezi et al. 2012; Melek et al. 2015).

One of the most appealing applications of agency theory in supply chain contexts is in regard to the development of risk mitigation strategies against suppliers’ threats of opportunism. Zsidisin and Ellram (2003) address this point by arguing that managing agents’ activities using buffers (such as safety stock and multiple sourcing) in conjunction with incentives can be more effective than manipulating agents’ behaviour to increase compliance by using, for example, certifications and supplier involvement in target costing development. Zsidisin et al. (2004) and Zsidisin and Smith (2005) corroborate this view by suggesting that the most common principals’ strategies to increase compliance (e.g. early supplier involvement) have been used not only to collect information on suppliers’ behaviour and mitigate supply chain risks but rather “to lock” suppliers into the supply chain and increase their dependence on the principals (Fayezi et al. 2012: p. 563).

Hornibrook (2007) suggests adopting incentives to moderate the agent’s threat of opportunism. He recommends using financial and social incentives in different ways as they have different temporal returns (short term the former and long term the latter). In a similar way, Norrman (2008) asserts that agency relationships tend to be more effective when they are governed through relational and formal (contractual) mechanisms. He also stresses the role of cultural issues and trust in the successful implementation of risk-sharing strategies, suggesting that trust plays a crucial role in increasing the suppliers’ compliance intentions and motivations. This is more likely to happen “when risk and reward appear to be equally balanced” (Fayezi et al. 2012: p. 563).

Cheng and Kam (2008) adopt agency theory to evaluate how different supply chain actors respond to unpredictable and unexpected risks. Agency theory assumptions can help actors to analyse the contextual factors responsible for supply chain integration and collaboration (Danese 2011). They suggest that network collaboration is highly linked to several factors such as network structure, supply performance assessment, actors’ functional position and mostly the principal–agent contract arrangement (including the definition of incentives, trust building processes and long-term orientation). In particular, the contract arrangement regarding suppliers’ performance evaluation processes is of crucial importance in a tiered, multiple supply chain network (Prosman et al. 2016; Ritchie et al. 2008).

A consistent part of the selected articles extensively analyses the supply chain relationship variables, such as incentive alignment, information sharing and reward/sanction development, which can be considered as pillars of agency theory (e.g. Agrell et al. 2004; Manatsa and McLaren 2008; Norrman 2008; Simatupang and Sridharan 2002; Zsidisin and Ellram 2003). Goal incongruence and conflict among supply chain actors are also recurring topics in studies addressing the barriers to supply chain integration and synchronisation (e.g. Morgan et al. 2007; Tate et al. 2010). In the same vein, Melek et al. (2015) investigate inter-organisational integration patterns between the buying firm, design agency and component supplier in NPD projects in order to identify the determinants of these approaches. Asymmetric information, power and incentive structure in supply chain relationships represent relevant topics in the selected literature. For example, Kudla and Klaas-Wissing (2012) analyse dyadic buyer–supplier relationships of logistics services in order to address how the actors send stimuli to each other and respond to such stimuli in conditions of asymmetric information. Prosman et al. (2016) address the power differences among supply chain actors and explore factors influencing the effectiveness of buyer-initiated behavioural-based governance methods.

1.4 Agency Theory Contribution in Supply Chain Management

From the literature review on the contribution of agency theory in the supply chain management literature, a number of considerations can be summarised. Firstly, agency theory provides a useful basis for understanding the multiple aspects of supply chain relationship management: it can explain the impact of a principal’s decisions on the agent’s propensity to cooperate and comply, as well as to inform contractual responses to (outcome/behavioural) the uncertainty of both actors in the relation (Fayezi et al. 2012). The tenets of agency theory can also contribute to the design of contractual arrangements where mutual information, risk and reward sharing (Cooper et al. 1997; Lee and Whang 2000) and goal congruence (La Londe and Masters 1994) have more chance of being effective, and the establishment and maintenance of long-term business relationships are more easily assured (Cousins 2002).

Secondly, the positivistic agency theory perspective can be usefully applied in the supply chain context in order to understand the potential causes of abnormal behaviours (non-rational or opportunistic) of multiple actors. “Agency theory identifies behavioural change by supply chain actors and sheds light on activities involving principal and agent, self-interest, risk aversion, lack of trust, goal conflict and imperfect policy implementation” (Fayezi et al. 2012: p. 564; Simatupang and Sridharan 2002). Furthermore, the theory suggests how to design contractual arrangements able to moderate or eliminate tensions among supply chain actors by means of information sharing, incentive structure and goal alignment (Knoppen and Christiaanse 2007). Zomorrodi and Fayezi (2011) and Wilson (2006) assert that zones of tolerance and relational elasticity exist in principal–agent relationships. In such situations, behavioural abnormalities are less likely to occur. Such arguments can inform the relationship development in order to reduce the uncertainty typical in supply chain decisions and increase the efficiency of the economic exchange (Prosman et al. 2016).

Finally, agency theory can be adopted to compensate for aspects widely neglected by the supply chain management literature adopting different theoretical perspectives (such as TCE and social exchange theory). As suggested by Fayezi et al. (2012: p. 564), agency theory—mostly from a positivist approach—“extends views centred on task and transaction by attending to specific attributed of agents (or principals) operating within the supply chain”. In doing so, agency theory is able to deal with some of the most complex elements of supply chain relationships, such as actors’ motivations to cheat, risk preferences and communication. In particular, communication, as an aspect of supply chain relationship development, has received marginal attention in the selected body of literature. This is in contrast to the central role played by communication in orienting actors’ behaviours and in contributing to the reduction of supply chain uncertainty.

This literature review—as the others previously published on similar topics (e.g. Fayezi et al. 2012; Bergen et al. 1992)—clearly shows the explanatory power of agency theory in the supply chain management context. However, the adoption of this theory has its limitations. For example, an overemphasis on economic drivers can limit the theory applicability (Heracleous and Lan 2011). According to Fayezi et al. (2012: p. 565), this is in part “an historical legacy, and also occurs because many [supply chain relationships] have traditionally been conceptualised as economic exchange mechanisms, rather than being comprised of complex social and economic relationships”.

The agency theory’s assumption of “perfect” principal and “imperfect” agent is also questionable and reductive. As Perrow (1986) suggested, the adverse selection and moral hazard agency problems are not restricted solely to the agent sphere of behaviour but also exist on the principal side. This consideration is essential to understand supply chain dynamics where the interdependence and interconnection among the actors are high. In fact, a supply chain is a system where the same actor may act as the principal in a specific relationship and, at the same time, act as the agent in a different one. For instance, a particular distributor who is acting as the principal of a manufacturer (the agent) while at the same time plays the role of agent for the wholesaler (the principal).

Increased organisational complexity—as in the case of an extended network of multiple principals and agents—is another important extension to the agency theory that is useful to improve its contribution to supply chain management literature. The existence of complex arrangements in which multiple actors interact (either multiple principals and/or multiple agents) makes the detection of non-rational behaviours more difficult and challenges the information sharing and monitoring activities. This becomes even more critical when multiple agents have competing interests over a similar (or identical) delegated task, as in the case of code of conduct implementation in a global supply chain discussed later in the study.