The Opportunistic Principal

This article sheds light on a surprisingly persistent gap in economic theory: Economists have extensively studied agency relationships between principals and agents, but economic analysis has traditionally been preoccupied with problems created by agents rather than principals. Principal&#8208;agent theory traditionally neglects opportunism on part of the principals. As a result, economic theory fails to fully explain gaps in economic performance. Even worse, the for decades biased concept of opportunism has shaped generations of students in economics and business and thus has contributed to deficient managerial and economic outcomes. Under the presumption that not only agents but also principals demonstrate opportunistic behavior, the economic definition of opportunism as established by Oliver Williamson is further developed. This contribution extends the concept of opportunism to principals. The principal&#8217;s illegitimate interference with the autonomy of the agent is discussed which shows that whilst agent opportunism largely rests on information asymmetry, principal opportunism feeds itself off power asymmetries. In an interdisciplinary effort and building on recent research in the area of bullying and mobbing, the economic rationale of principal opportunism is explained with the help of a simple game&#8208;theoretic model. On this basis, it is demonstrated how a more balanced perception of principal agent relationships can contribute to an updated theory of the firm.

Paraphrasing Ronald Coase's first sentence in The Problem of Social Cost (1960,1), this paper is concerned with those actions of individuals which have harmful and undesired effects on others in the same organization. The standard example is that of a boss bullying subordinates which has deleterious effects on the employees concerned and leads to an inefficient allocation of resources on the level of the organization.
Economists have extensively studied agency relationships between principals and agents but economic analysis has traditionally been preoccupied with problems created by agents rather than principals. Organization economics (Ouchi and Barney, 1986), and more specifically agency theory (Eisenhardt, 1989) recognize that agents do not always act in the best interest of their principal. Regularly, agents have more information about a situation than their principal. In such a situation, agents have incentives to act opportunistically which leads to issues that can be subsumed under the headlines of adverse selection and moral hazard. This appears to be particularly relevant where hierarchy and employment relationships replace market transactions. Here, agents are perceived to misrepresent their qualifications, loaf on the job or otherwise not put in the effort that can be expected of them. Opportunism has been identified as a necessary condition for the emergence of transaction costs. "If humans could be relied upon to always behave honestly, then uncertainty, asset specificity, and bounded rationality would pose no extra problems. Transaction costs would be zero" (Moschandreas, 1997, 41).
Again, the context is different, but there is another interesting parallel to Coase's problem statement here which can be found in the reciprocal nature of the problem (Coase, 1960, 2): If agents are described as opportunistic, why would principals not be? Especially, since in organizations except for those in roles at the very bottom and at the top of a hierarchy, individuals act as both, agent and principal (Eggertsson, 2008, 41).
The asymmetric use of the opportunism assumption has attracted critics from outside (Perrow, 1986;Shapiro, 2005) but also from within the profession of economics (Dow, 1987;Moschandreas, 1997). In considering both, "shirking" by agents and "sharking" by principals, many areas of the social sciences including legal studies (Orts, 1998), organization theory (Robinson and Bennett, 1995), sociology (Perrow, 1986) and political sciences (Sobol, 2016) seem to have arrived at a more balanced view. In economic agency theory however, primarily scholars who study cooperative firms (e.g. Albanese et al., 2015) are appreciative of a symmetric and holistic perception of opportunism. A recent further exception is Zardkoohi et al. (2015) who consider opportunistic short-term oriented behavior of shareholders opposite company CEOs. In general, economic theory has been consistent in not taking into account that the principal may exploit both, the situation and the agent. In the tradition of Alchian and Demsetz (1972) economic theory "rigorously reject[s] the image of a firm as a kind of private Leviathan" (Miller, 1992, 65).
The conceptual ignorance indicates theoretical relevance. The issue becomes more pressing once it is acknowledged that economic theory is not limited to explanation and accurate prediction of outcomes as claimed by Friedman (1953, 7ff) but that including the normative implications of its assumptions it has the propensity to shape institutions and economic outcomes: "The ideas of economists and political philosophers, both when they are right and when they are wrong, are more powerful than is commonly understood" (Keynes, 1953, 306). In following Gergen (1973), Sumantra Ghoshal claims that, "unlike theories in the physical sciences, theories in the social sciences tend to be self-fulfilling" and proposes that "a management theoryif it gains sufficient currencychanges the behaviors of managers who start acting in accordance with the theory" (Ghoshal, 2005,77). Today, it is widely agreed that agency theory and the theory of the firm along with transaction cost economics have had and still do have a significant influence on how organizations are structured and managed (Donaldson, 1990;Zajac and Westphal, 2004;Shapiro, 2005). The fundamental concepts of Alchian and Demsetz (1972), Jensen and Meckling (1976), Williamson (1985) and others have for many years been an integral part of economic and business education (see for example Douma and Schreuder, 2013;Morschett et al., 2015;Jones, 2013). In this sense, for decades a biased concept of opportunism can be expected to have shaped generations of students in economics and business and thus to have translated into managerial practice.
Business and organizational practice does indeed face substantial challenges: According to empirical studies like the Gallup Engagement Index (Nink, 2014) or the Global Workforce Study (Towers Watson, 2014), the vast majority of employees in the private sector are in a state of resignation or burn-out, or work-to-rule. They leave their organizations because of their bosses (Abbasi and Hollmann, 2000;Harter and Adkins, 2015). This implies a massive cost problem asking for timely explanations, also by the economic profession. Opportunism by principals can no longer be afforded to be a blind-spot in economic agency theory. The visible hand of management (Chandler, 1977) may sometimes hurt too much.
It is proposed here to take into account the concept of the opportunistic principal in order to fully capture the reciprocal nature of the problems arising in agency relationships. In the tradition of Simon (1951), Coleman (1990), andMiller (1992) the focus will be put on the employment relationship, notwithstanding that the concept can be of equal importance in other principal-agent relationships. To achieve this, the cornerstones of economic agency theory will briefly be summarized before relevant criticism will be outlined. This leads to a conceptualization of the opportunistic principal which rests on an extended definition of opportunism. The principal's illegitimate interference with the autonomy of the agent is discussed which shows that whilst agent opportunism largely rests on information asymmetry, principal opportunism feeds itself off power asymmetries. Building on recent research in the area of bullying and mobbing and with the help of a simple game-theoretic model, the economic rationale of principal opportunism is explained: short term, individual benefits incentivize opportunistic behavior and explain why often neither behavior-oriented, nor outcome-oriented contractual solutions proposed by traditional economic agency theory seem to work. Instead, principal and agents spiral into an opportunism game. Finally, insights from organization theory are presented and combined with findings from agency theory to explore the multiple causes for these unproductive dynamics and to search for potentially game-changing institutional arrangements from an economic point of view.
Methodologically, the paper can be considered as an integrative literature review (Torraco, 2016). It is interdisciplinary in the sense that it integrates economic agency theory with findings from organization theory and sociology. In particular, research on bullying is taken into consideration. The focus will be on those aspects of bullying research that are relevant for the portrayed gap in economic agency theory. This will lead to a symmetric conceptual framework of the principal-agent relationship which acknowledges both, information asymmetries and power asymmetries.

II. AGENCY THEORY
Agency relationships, in which one party (the principal) delegates work to another (agent), are the golden-thread of economic life. A so-called agency problem arises when the agent does not act in the best interest of the principal (Jensen & Meckling, 1976, 308). Agency scholars distinguish two approaches: positivist agency theory and formal principal-agent theory (Jensen, 1983, 340). Along both lines of research, the contract between the principal and the agent is the unit of analysis. And as they also share common assumptions about people, organizations, and information (Eisenhardt, 1989, 59), the approaches will not be further distinguished for the purpose of this paper.
Agency theory rests on specific behavioral assumptions, namely bounded rationality and opportunism. Williamson (1996, 50) explains that the assumption of opportunism serves the purpose of avoiding contractual naiveté and it invites the identification, explication, and mitigation of hazards. In a principal-agent relationship the principal has limited information about the behavior of the agent. This can lead to situations where the agent is not putting in the agreed effort and is shirking. In a multi-agent environment one agent may be free-riding and let others do the agreed work. Such situations of ex-post opportunism are referred to as moral hazard. Ex-ante opportunism may occur when an agent misrepresents skills or abilities before entering into a desired principal-agent relationship. The resulting problem is regularly referred to as adverse selection (Furubotn and Richter, 2005, 148ff).
The focus of agency theory is to determine contracts that efficiently govern the principal-agent relationship in the light of moral hazard and adverse selection problems. To achieve this, agency theory generally distinguishes two options the principal can pursue: behavior-oriented contracts and outcome-oriented contracts (Eisenhardt, 1989, 61). In other words, the principal can invest in information systems to control agent opportunism, or she can try to align the interests of the agent with her own by providing suitable incentives.

III. AND ITS CRITICS
Agency theory's focus on the contract between principal and agent has raised questions. Economists like Stiglitz (1988), Akerlof and Yellen (1986) and Leibenstein (1988) observed that there are inconsistencies between typical labor contracts and contractual solutions predicted by agency theory. Baker, Jensen and Murphy (1988, 614) found that "managers have few incentives to structure and enforce value-maximizing contracts with subordinates." Building on this, Miller (1992, 6f) criticizes the narrow economic view that the principal-agent problem in organizations is about monitoring and rewarding subordinate efforts. As any formal incentive system leaves room for self-interested behavior leading to efficiency losses, Miller (1992, 11ff) discusses the political economy of hierarchy and analyzes how political authority in organizations is created, used and deployed. In doing so, he implicitly raises another criticism of agency theory brought forward by Shapiro (2005), which is that it portrays the relationship as acontextual, ahistorical, and static: "Agency relationships are enacted in a broader social context and buffeted by outside forcesother agency relationships, competitors, interest groups, regulators, legal rules, and the likethat sometimes right informational imbalances, offer or constrain incentives, exacerbate the risk of adverse selection or moral hazard, provide cover or opportunity for opportunism, and so forth" (Shapiro, 2005. 269).
Like his fellow sociologist Susan Shapiro, Charles Perrow (1986, 15) accuses agency theory of a "celebration of self-interest". In line with this, stewardship theory tries to show that rather than pursuing their self-interest, agents may aim to be effective stewards of their organizations (Davis, Schoormann and Donaldson, 1997). Miller and Sardais (2011) even analyze the phenomenon of "angel agents". Efforts of angel agents who have considerable positive impact on their organizations are in line with the need to question another assumption of agency theory: principals are not always the ones and not the only ones to direct the organization, making contracts that agents must follow (Sharma, 1997, 775ff). This is obviously related to the question of power which appears to be Perrow's main argument as he is of the view that agency theory fails to recognize the asymmetry of power in organizations. In particular, he notes that agency theory is extremely biased as principal-agent models almost invariably assume that the agent is opportunistic, even to the point of cheating, rather than the principal (Perrow, 1986,14). Similarly, Dow (1987, 33) criticized that "the literature has not examined the possibility that authority opens up new avenues for opportunism." In summary, there is substantial critique from other disciplines but little theoretical progress. And the economic profession has been very reluctant to accommodate the concept of power asymmetries in general and the idea of opportunistic principals in particular. One can only speculate over the reasons. This might be ideologically motivated and on purpose (Perrow, 1986, 15), it might be out of a fear that principal-agents models could lose their elegance, or it could be that the issues were not considered to be relevant from an economic(s) perspective. Whatever the reasons, it will be shown here that an elegant design of agency theory that takes opportunism of principals into account is possible and that such an approach gives access to highly relevant problems in and for organizations.

IV. THE OPPORTUNISTIC PRINCIPAL -SELF-INTEREST SEEKING WITH DOMINANCE
When reviewing the theoretical basis for principal-agent analysis, it seems reasonable to assume that behavioral assumptions for principals should be the same as those for agents. This has always been the case for bounded-rationality but not for opportunism. Here, agency theory has traditionally been biased and opportunism has always been an 'agent's thing'. This already becomes apparent when reviewing Oliver Williamson's omnipresent definition of the term opportunism which stands for "self-interest seeking with guile". "This includes but is scarcely limited to more blatant forms, such as lying, stealing, and cheating. Opportunism more often involves subtle forms of deceit" (Williamson, 1985, 47). Such dishonest behavior is something that is typically shown if an economic actor is in the role of an agent. The economic literature has extensively studied situations where asymmetric distribution of information or transaction specific investments, give rise to opportunistic agent behavior.
As his reference to Niccolò Machiavelli's 'principe' shows, Williamson (1985,48) is well aware of reciprocal or preemptive opportunism. Nevertheless, his prevalent definition has distorted the view on what is actually a wider phenomenon. In plain English, opportunism means trying to exploit a situation to your own advantage. It is "the policy or practice, as in politics or business, of adapting actions, decisions, etc., to expediency or effectiveness without regard to principles or consequences" (American Heritage Dictionary, 2011). As such, opportunism may well go beyond self-interest seeking with guile.
To achieve a more balanced understanding of opportunism in principal-agent relationships, I propose to perceive opportunism on the side of the principal as self-interest seeking with dominance. This includes but is scarcely limited to more blatant forms such as bullying, insulting, and intimidating. Opportunism more often involves subtle forms of dominance like excessive monitoring or manipulative political practices. Such dishonest behavior is something that is typically shown if an economic actor is in the role of a principal. Borrowing a term from Orts (1998), self-interest seeking with dominance will below in short also be described a "sharking". As will be shown, self-interest seeking with dominance is facilitated by the authority relationship between the principal and the agent. It is an asymmetric distribution of power and transaction specific investments which give rise to opportunistic principal behavior, leading to situations where an abuse of authority can be observed, resulting in distorted economic performance.

V. ILLEGITIMATE INTERFERENCE WITH THE AUTONOMY OF THE AGENT
In a principal-agent relationship in the workplace, the agent acts under the authority of the principal (Simon, 1951, 294). Coleman defines the right to control another's actions as authority. These actions are resources which were originally controlled by the agent. Organizations therefore can be perceived as authority structures (Coleman, 1990, 66). In a liberal market economy, organizations come into existence, develop and grow when individuals see it in their best interest to give up the right to control certain of their actions (agents) whilst others assume the authority over these actions (principals). "Yet an actor cannot physically transfer actions to another actor in a way that he can physically transfer a pound of sugar or a pair of shoes. He can only transfer to others such intangibles as a promise to act in a certain way or the right to control his actions within certain specified limits" (Coleman, 1990, 66). If the principal deliberately does not respect the specified limits, he acts opportunistically with dominance. By violating the specified limits, the principal interferes in illegitimate ways with the autonomy of the agent. According to Coleman (1990, 68) "authority that is not voluntarily vested, must be backed by coercive power if it is to be enforced". This is what makes dominance the key characteristic of opportunistic principals. The possibility for opportunism with dominance arises as "the rights to control actions gained by the superordinate [principal] give the superordinate [principal] the opportunity to extend that control" (Coleman, 1990, 81). In general, the principal can take two approaches to extend control and interfere with the autonomy of the agent: the abuse of authority and the abuse of power. The abuse of authority occurs when instructions and control activities of principals undermine or inflict with the contract with the agent. Whereas the abuse of authority stays within the dimension of the underlying contract (instruction, outcome and control), the abuse of power opens up a new dimension and other forms of ascendancy that principals have over their agents (Breton, 1995cit. in Vafaї 2010. More specifically and as exemplified in table 1, opportunistic behavior of the principal can unfold on the subject level as well as on the personal level. As can be seen from the inventory of behaviors represented in table 1, economic agency theory can build on insights from organization research, namely on bullying and mobbing (see Bartlett and Bartlett, 2011;Salin, 2003; Branch Table 1 Options for principal opportunism (partly based on Bartlett and Bartlett, 2011) Power Personal level (direct) to verbally attack to (sexually) harass to yell to make belittling remarks to insult to expose to intimidate to threaten to humiliate to manipulate to be intentionally demeaning to interrupt to persistently criticize Personal level (indirect) to isolate to lie to ignore to make false accusations to exclude to undermine to gossip Authority Subject level to set too demanding/too many goals to withhold information to set antinomic goals to overrule decisions to reign in between to set unreasonable deadlines to shift opinions to overload with work to judge work wrongly to set up for failure to avoid variable pay/bonuses to delegate menial tasks to monitor excessively not to assign any tasks to micromanage to withhold resources to disproportionately sanction et al., 2013 for reviews) to understand principal opportunism. Whilst bullying and mobbing are considered to be targeted at particular individuals only (Einarsen et al., 2011, 10), principal opportunism may be applied more widely.
And it may be intended by the principal that bilateral acts opposite one agent send signals to other agents. In this sense, the behaviors listed in table 1 are not always part of bullying activity, but they are considered to be relatively common in the workplace (Leymann, 1996). Two defined features of bullying relevant for principal opportunism are that the behaviors are intended to be hostile or perceived to be hostile and that they occur repeatedly and consistently. "Hence, the concept of bullying [or principal opportunism] at work relates to persistent exposure to negative and aggressive behaviors of a primarily psychological nature" (Einarsen et al., 2011, 9). All of these behaviors fall into the category of ex-post opportunism. Whilst postcontractual opportunism of agents is due to both, asymmetric information and specific investments, postcontractual opportunism of principals is due to both, asymmetric power and specific investments. Given that employment contracts regularly require comparatively high transaction specific investments not only for the principal but also for the agent (Zardkoohi et al., 2015, 4) the problem of moral hazard is raised. Quitting a job can be costly to the agent, he therefore may accept if the principal shows dominant behavior and exceeds the (implicit) limits of the employment contract.
Ex-ante opportunism can also occur. It typically rests on informationasymmetries and leads to adverse selection. As Perrow (1986, 14) describes, the principal may misrepresent his "type" and/or the contractual situation, e.g. in terms of hazardous working conditions, production pressures, adequacy of equipment, fairness of supervision, advancement possibilities, amount of compulsory overtime etc." Opportunism by the principal at this stage is merely a mirror image of the widely studied agent opportunism in the same situation (Spence, 1974;Furubotn and Richter, 2005, 149). Therefore, ex-ante the principal does indeed seek self-interest with guile in order to enable, facilitate or even necessitate a seeking of self-interest with dominance at an ex-post stage.
As widely discussed in the academic literature on bullying, opportunistic behaviors may not be carried out with the (primary) intention to inflict harm on agents (Einarsen et al., 2011, 19). But from an agency-theoretic point of view, opportunism is in the self-interest of a principal who aims to efficiently manage the contracts with his agents.

VI. PRINCIPALS AND AGENTS IN INTERACTION -AN OPPORTUNISM GAME
As stated above, agency theory expects the principal to pursue behavior-oriented contracts and outcome oriented-contracts to deal with agent opportunism. But already Baker, Jensen and Murphy (1988, 594) found that traditional economic agency theory is often inconsistent with empirical observations from business and organizational reality and that principals can have insufficient incentives to structure and enforce value maximizing contracts (Ibid. 614). From an economic perspective, the problem is that the contractual approach is complex and costly (see also Eisenhardt, 1989;Edmans, Gabaix and Jenter, 2017). This is particularly relevant, if, in contrast to the traditional principal-agent literature, not only owners but more generally managers and supervisors in hierarchical organizations are perceived as principals (Eggertsson, 2008, 41). These principals do not own the residual claims and thus may not sufficiently benefit from value maximizing contractual solutions. On this basis, it can be concluded that principals have reasons to reach out to substitute mechanisms and behaviors to pursue their goals. This is where the concept of the opportunistic principal offers a perspective to widen the narrow view which, in agency theory, limits the principal's role to monitoring and rewarding agent efforts within the strict limits of the contract: An alternative option for the principal is to counter (anticipated) opportunism with opportunism. In other words: instead of working on contractual solutions, the principal can bend or breach the contract. This also reflects the reciprocal nature of the problem of opportunism by acknowledging the relevance of the asymmetry of power.
The principal may choose to act opportunistically opposite all of his agents, for example by pursuing manipulative practices, setting unrealistic goals or by micromanaging. Or, the principal may directly target his opportunism to selected agents only, for example in form of verbal attacks and insults. But even the latter behavior has effects on the behavior of agents who are not directly affected by the abuse of power and authority when those actions function as deterrents (Alberts et al., 2005;Jennifer et al., 2003;Vartia, 2001Vartia, , 2003. In the economy, such behaviors occur on different levels. For example, stockholders of public companies regularly act aggressively opposite company CEOs to pursue their dismission, even if this may not be rational (Allaire andDauphin, 2016, Clayton et al., 2005;Denes et al., 2017). Zaardkoohi et al. (2015, 5) suggest that such processes of executive succession may be "one of the manifestations of the principal problem." Whilst building on the same approach of self-interest seeking with dominance, principal opportunism within organizations is less easy to detect and less transparent. This will be the focal-point for the discussion below.
Game-theory can be used to describe the situation as 'opportunism game' with one principal and several agents. This can be modeled as a prisoners' dilemma as depicted in figure 1.
Here, the joint incentive of cooperation is in conflict with individual incentives to act opportunistically. If principal and agents cooperate, they can achieve a payoff of (4, 4). However, regardless of what the other party does, from an individual point of view it is better to act opportunistically. Opportunism is the dominant strategy for both, principal and agents and thus leads to an outcome of only (2, 2).
The stylized model can help to integrate insights from economic agencytheory and from organization research. It highlights particular aspects of the principal-agent relationship that blend-in well with findings in both areas: In principal-agent relationships, the principal has incomplete information about the behavior of the agents. This information asymmetry incentivizes agents to exploit the situation to their advantage. And it can lead the principal to anticipate opportunistic behavior on part of the agents. In order to counter potential or real opportunism by the agents, the principal has the opportunity to exploit his authority and power over the agents: the power asymmetry is to his benefit. A situation where the principal chooses to be opportunistic and the agents remain cooperative or vice versa is unlikely to be stable. Making direct reference to the prisoners' dilemma, Andersson and Pearson (1999) conjecture a self-reinforcing, spiraling process which is built on vicious circles. Research on workplace bullying has shown that agents experience lower job satisfaction (Ayoko et al., 2003;Thompson, 2003) and react with decreased commitment and poor morale (Gardner and Johnson, 2001;Namie, 2003). Individual reactions also include a cut back on hours worked (Gardner and Johnson, 2001) and increased absenteeism (Kivimaki et al., 2000;MacIntosh, 2005). The agent's reactions are not limited to opportunistic behaviors but include physical and emotional health impacts. Various clinical pictures reflect an overall decrease of physical health (Kivimaki et al., 2000;Randle et al., 2007). Emotional consequences include fear, sadness and anger (Ayoko et al., 2003;Namie, 2003) and lead for example to loss of concentration or feelings of powerlessness (Ballien et al., 2009).  Ballien et al. (2009) andYildirim (2009) report lower productivity. Related to this, Paice and Smith (2009) identified increased workplace errors. Absenteeism due to shirking or poor health increase inefficiency and cost (Namie, 2007). Increased turnover (Gardner and Johnson, 2001;Namie, 2003Namie, , 2007 affects productivity and induces additional costs for rehiring and training.

VII. REASONS FOR OPPORTUNISM GAMES
Based on the references made, it appears that there is a certain practical relevance of opportunism games between a principal and his agents which lead into prisoners' dilemma situations. To avoid or resolve such outcomes, knowledge about the underlying causes is essential. Again, the integration of research on bullying with established knowledge from agency theory proves to be insightful. The literature suggests that the phenomenon has multiple causes (Zapf, 1999;Einarsen et al., 2011;Salin, 2003). Put in the context of the stylized model of an opportunism game (figure 1) the interplay of these causes can be captured: • Economic analysis typically focuses on information asymmetries to the benefit of the agent. These are particularly relevant in poorly organized work environments or organizations that undergo change. But even rather static hierarchical and bureaucratic organizations have long been found to be fertile grounds for agent opportunism (Downs, 1965;Niskanen, 1971). In general, information asymmetries enable shirking and require efforts on side of the principal to detect and reduce deviant behaviors by the agents (Alchian and Demsetz, 1972;Pratt and Zeckhauser, 1985;Eisenhardt, 1989). Agent opportunism can lead to outcomes C and D. Quadrant D of the opportunism game is the traditional focal point of economic agency theory. Consequently, economic education in universities and business schools has largely limited itself to the "small world" (Savage, 1954, 8) of D, ignoring the game as a whole. Such a partial view nurtures suboptimal and potentially bad management practices (Ghoshal 2005, 77). • In an effort to capture the game as a whole, it is necessary to consider that the personalities of both, the agents and the principal, may lead to opportunistic behavior. As Oliver Williamson noted repeatedly, "the propensity for opportunism varies among individuals and cultures" (Williamson, 1996, 50). Kiazad et al. (2010) found that principals with a Machiavellian leadership style are more adept to abusive supervision. Machiavellianism, psychopathy and narcissim are known as "dark triad" traits (Paulhus and Williams, 2002). Individuals with these traits can often be found in principal roles (Wille et al., 2013) and the traits have been found to be linked with opportunistic behavior at work (Linton and Power, 2013). This implies that sometimes outcome A may be maintained by principals. • Organizational deficiencies can incentivize opportunism by both, principal and agents. Leymann (1996, 177) claims that poor organization of work is a primary cause for bullying (outcome A followed by C). This in turn can be a result of laissez-faire style of leadership at the top of the organization (Einarsen et al. 1994;Hoel and Cooper, 2000) which would be represented by an initial situation B, again followed by A by C lower in the hierarchy of the organization. According to Salin (2003 precipitating processes can also be restructuring, cost-cutting and other organizational changes. • Sometimes but not always related to organizational deficiencies are power asymmetries. Perceived power imbalances have been identified as a cause for opportunistic behavior by principals (Salin, 2003(Salin, , 1219, and subject to agent reactions result in outcomes A or C. Hierarchical organizations with rank structures appear to be particularly vulnerable (Ashforth, 1994, Archer, 1999. Large and bureaucratic organizations show principal opportunism more frequently (Einarsen and Skogstad, 1996). This is consistent with the findings of Fehr et al. (2013) who study the motivation and incentive effects of authority and conclude that principals often retain authority even when its delegation is in their material interest. They also observe that authority leads to an overprovision of effort by the principal and to an underprovision of effort by agents which means that sharking is complemented by shirking. • Low cost for agents to raise their voice or to exit the organization (Hirschman, 1970) and to find adequate alternative employment, are likely to discourage opportunistic behaviors by the principal. However, agents are usually required to invest in firm-specific human capital. This firmspecific investment exposes the agent to principal opportunism (Zardkoohi et al., 2015, 4). • Low moral standards in the work environment allow opportunism to gain ground. Deviant behaviors prevail in unilateral fashion (outcomes A or D) or in mutually reinforcing ways (outcome C) as and when these are tolerated or even rewarded by organizational cultures (Rayner, 1997;Collinson, 1988;Brodsky, 1976). This is also influenced by technological change as information technology increases work dynamics leading to higher pressure on all involved actors (Wagner, 2015). It provides the principal with new tools to shark, for example by micromanaging (Serrat, 2010;Sutton, 2011). As and when face-to-face interactions can be avoided, the cost to engage in such behaviors come down, even when the absence of physical proximity dampens the effect on the affected agent (Johnson, 2016) and it makes it easier for the agent to shirk (Lim et al., 2002).

VIII. IN SEARCH FOR BENEFICIAL INSTITUTIONAL ARRANGEMENTS
The multi-causality outlined above suggests that opportunism is mutually reinforcing, or, in other words, that non-cooperative behavior by one party can lead into an opportunistic tit-for-tat, a self-reinforcing, spiraling process which is built on vicious circles (Andersson and Pearson, 1999). Whether the principal or agents are the first to defect, will depend on the specific circumstances, and may here be classified as a chicken or egg problem. Economic terminology can help to summarize the causes: poor organization means that property rights are poorly defined and insufficiently enforced. In organizational structures and cultures which tolerate opportunistic behaviors, property rights are not mutually supported and enforced by the members of the organization. This means that the cost for opportunistic behavior in form of both, shirking and sharking are comparatively low. Power imbalances or asymmetries due to personal traits or organizational and cultural circumstances reduce the cost for the principal to act opportunistically. And, the more likely it is that the principal will alter property rights to his or her own benefits, the lower the commitment that can be expected from the agents (North and Weingast, 1989, 803). Information-advantages incentivize agents to act opportunistically. Opportunistic tit-for-tat can lead to undesirable outcomes for both, involved individuals as well as the organization.
Ultimately, it will depend on the actual payoff structure of the game whether or not C will indeed be the outcome. In addition to personal traits, information and power asymmetries, specific investments by principal and agents will be of relevance. If one party to a contract has made a much larger transaction-specific investment than the other, outcomes A and D become more likely, and the more vulnerable side will incur a welfare loss (Furubotn and Richter, 2005, 150). The so-called efficiency wage (Akerlof and Yellen, 1986;Katz, 1986) appears to be of particular relevance in this context, as it equips the principal with "an asymmetric ability to dictate the behavior of employees" (Miller, 1992, 67).
If (potential) opportunism problems evoked by either party can be overcome and principal and agents can convince each other of credible commitments (Williamson, 1985, 48, 163ff), outcome B is within reach. In business and organizational practice this often does not happen (Nink, 2014;Towers Watson, 2014;Abbasi and Hollmann, 2000;Harter and Adkins, 2015). Nevertheless, the search for suitable managerial and organizational practices which take into account the possibility of principal opportunism is on. The discipline of economics can contribute here with the identification, study and review of institutional arrangements that have the potential of solving opportunism games in which also the principal defects by shaping the payoff structure in suitable ways. These include but are not limited to, institutions that fulfill one or more of the following purposes: • Enhance the transparency to lower transactions costs of using the market mechanism and to optimize firm-specific investments of human capital, like for example online rating platforms for employers (DeKay, 2013;Symitsi et al., 2018), • enhance the transparency to lower transaction costs and reduce power asymmetries within hierarchies like for example human resource management systems that include 360-degree feedback processes (Bracken et al., 2016;Morgan et al., 2005) or agile management approaches (Abrahamsson et al., 2002;Jalali, and Wohlin, 2012;Sutherland, 2015), • enforce and develop property rights not only of principals but also of agents, for example with a corresponding corporate culture (Collins, 2001;Schein, 2016), or specific approaches to leadership and management like autonomy enhancing paternalism (Binder and Lades, 2015). • redefine property rights within the organization to reduce power asymmetries and increase participation of agents in decision-making processes, for example with participatory approaches in the tradition of "New Work" (Bergmann, 2017;Laloux, 2014), with approaches to democratic corporate design (Boes et al., 2015), or with maturing institutions of traditional codetermination (Jirjahn et al., 2011), • redefine property rights within and from outside the organization to protect agents from principal opportunism, for example through policies and legislation addressing workplace abuse (Duffy, 2009, Wewers, 2017. Some of these institutions have been around for some time with limited success. Others are relatively recent phenomena. Future institutional analysis would have to take these into account in conjunction with established knowledge of behavior-oriented and outcome-oriented contracts dealing with agent opportunism (Eisenhardt, 1989;Furubotn and Richter, 2005). Whilst overall effects of different institutional arrangements on the opportunism game and on economic performance still need to be better understood, it seems evident that the decisive role comes to the principal "to shape expectations among subordinates about cooperation among employees, and between employees and their hierarchical superiors.
[…] Most important perhaps, the leader has a central role in committing the organization to what is in effect the "constitution" of the hierarchythe allocation of generally accepted responsibilities, rules of the game, and property rights that provide the long-run incentives for investment in the firm" (Miller, 1992, 217). With respect to the central role of the principal the reverse is of course also relevant and remains a key challenge: Dominant opportunistic principals have a propensity to undermine institutions that were designed to limit opportunism.

IX. CONCLUSION AND OUTLOOK
Along the lines of the introduction, it seems reasonable to, for a last time, paraphrase Ronald Coase (1960, 2) and state that the traditional approach in economics has tended to obscure the principal-agent problem. The question is commonly thought of as one in which the agent inflicts harm on the principal and what has to be decided is: how should we restrain or incentivize the agent? Whilst this is not wrong, it is certainly incomplete. We are dealing with a problem reciprocal in its nature. We also need to ask how to restrain or incentivize the principal not to inflict harm on the agent. "… there is a great deal of unhappiness with hierarchy, centered around doubts concerning both its efficiency and its effect on individual autonomy or liberty" (Miller, 1992, 19).
In essence, the concept of the opportunistic principal can help to unravel this "unhappiness". It sheds light on a blind-spot within economic agency theory and it can be a trigger to leave the "small world" (Savage, 1954, 8) to which agency theory has confined itself and those it has been educating (Ghoshal, 2005). Consideration of the opportunistic principal provides a more differentiated perception of opportunism as it complements opportunism as self-interest seeking with guile with opportunism as self-interest seeking with dominance. The first builds on asymmetric distribution of information whereas the second relies on asymmetric distribution of power. By taking both asymmetries into consideration, a symmetric perception of principal-agent problems is enabled. Or in other words, the reciprocal nature of principal-agent problems is acknowledged.
Empirical as well as theoretical evidence suggest that principal and agent opportunism interact in form of opportunism games in rather unpleasant ways leading to undesirable social and economic outcomes. It was shown that the integration of findings from agency theory and organization theory can enhance the understanding of such opportunism games. The review provided indicates that the origins of opportunistic tit-for-tat are in need of further explanation.
As exemplified here, insights from organization theory can help to specify and properly capture principal opportunism, so far neglected by agency theory. Agency theory, on the other hand, points towards an explanation for principal opportunism which can fill research gaps in economic and in organization theory: In trying to establish or rather enforce cooperative agent behavior, principals may perceive sharking as a cost-effective alternative to contractually compliant monitoring and incentivizing of agents.
One might say that with such a proposition, economics once more emphasizes its antiorganizational and antisocial view of human actors (Donaldson, 1990, 373). However, a clear cut (economic) understanding of undesired behaviors can pave the way to institutional arrangements that lead to preferable behaviors and outcomes. If "power is to organization as oxygen is to breathing" (Clegg et al., 2006, 3), then it needs to have a place in agency theory. The concept of the opportunistic principal problematizes the harmful aspects of power and authority abuse. Other than a blind-spot, a well understood weak-spot can be a useful point of reference and criteria for future oriented organizational designs. As shown above, the concept opens-up a research agenda that embraces a new concept of work in economics (Spencer, 2014) and that is in search of institutional arrangements that lead to mutually cooperative behaviors and ultimately to economically superior outcomes. Following many years of research on how to control agent opportunism by monitoring and incentivizing, such effort could be extended to principal opportunism. It will be instructive to connect with the growing interest in having less hierarchy, less principals and more self-organization that can be observed in organizational practice.