Characteristics of SEM

The term structural equation modeling (SEM) does not designate a single statistical technique but instead refers to a family of related procedures. Other terms such as covariance structure analysis, covariance structure modeling, or analysis of covariance structures are also used in the literature to classify these techniques together under a single label. These terms are essentially interchangeable, but only the first will be used throughout this book. Another term that you may have heard is causal modeling, which is a somewhat dated expression first associated with the SEM technique of path analysis. For reasons elaborated later, the results of an SEM analysis cannot generally be taken as evidence for causation. Wilkinson and the Task Force on Statistical Inference (1999) were even more blunt when they noted that use of SEM computer tools rarely yields any results that have any interpretation as causal effects (p. 600). Some newcomers to SEM have unrealistic expectations in this regard. They may see SEM as a kind of magical technique that allows one to discern causal relations in the absence of experimental or even quasi-experimental designs. Unfortunately, no statistical technique, SEM or otherwise, can somehow prove causality in nonexperimental designs. The correct and realistic interpretation of results from SEM analyses is emphasized throughout this book. Covariances Always, but Means Can Be Analyzed, Too The basic statistic of SEM is the covariance, which is defined for two continuous observed variables X and Y as follows: covXY rXY SDX SDY (1.1) where rXY is the Pearson correlation and SDX and SDY are their standard deviations. A covariance represents the strength of the association between X and Y and their variabilities, albeit with a single number. Because the covariance is an unstandardized statistic, its value has no upper or lower bound. For example, covariances of, say, 1,003.26 or 13.58 are possible. In any event, covXY conveys more information than rXY, which says something about association in a standardized metric only. To say that the covariance is the basic statistic of SEM means that the analysis has two main goals: (1) to understand patterns of covariances among a set of observed variables and (2) to explain as much of their variance as possible with the researchers model. The part of a structural equation model that represents hypotheses about variances and covariances is the covariance structure. The next several chapters outline the rationale of analyzing covariance structures, but essentially all models in SEM have a covariance structure. Fami ly Hi story and a Remi nder about Context Because SEM is a collection of related techniques, it does not have a single source. Part of its origins date to the early years of the 20th century with the development of what we now call exploratory factor analysis, usually credited to Charles Spearman (1904). A few years later, the biogeneticist Sewell Wright (e.g., 1918) developed the basics of path analysis. Wright demonstrated how observed covariances could be related to the parameters of a model that represents both direct and indirect causal effects among a set of variables. In doing so, he also showed how these effects could be estimated from sample data. Wright also invented path diagrams, which are graphical representations of direct and indirect effects that we still use to this day. In hindsight, Wrights innovations are remarkable. The technique of path analysis was subsequently introduced to the behavioral sciences by various authors, including Blalock (1961) and O. Duncan (1966), among others (see the annotated bibliography by Wolfle, 2003). The measurement (factor analysis) and structural (path analysis) approaches were integrated in the early 1970s in the work of basically three authors: K. Jreskog, J. Keesling, and D. Wiley, into a framework that Bentler (1980) called the JWK model. One of the first widely available computer programs able to analyze models based on the JWK frameworknow called SEMwas LISREL, developed by K. Jreskog and D. Srbom in the 1970s and subsequently updated by them several times. The 1980s and 1990s witnessed the development of more computer programs and a rapid expansion of the use of SEM techniques in many different areas of the behavioral sciences. There have been many recent developments, too, many of which represent the extension of models about continuous latent variables to other kinds of analyses. For example, there are now many works in the SEM literature about the estimation of growth and change over time on latent variables (i.e., latent growth curve modeling; e.g., Duncan, Duncan, Strycker, Li, & Alpert, 1999) and also about the estimation of curvilinear and interactive effects of latent variables (e.g., Schumaker & Marcoulides, 1998). Work by Muthn (1984) concerning estimation methods for non-normal data, such as when the indicators are dichotomous or ordered-categorical (ordinal) variables, further extended the range of application of SEM. Another major recent development concerns the convergence of SEM and techniques for multilevel analysis, which are applied in data sets where scores (cases) are grouped into higher-order units, such as siblings within families (Muthn, 1994). Within each level, the scores may not be independent, and multilevel techniques take this dependency into account. Recent versions of some SEM computer tools, including EQS, LISREL, and Mplus, feature built-in syntax for multilevel analyses. The origin of the term causal modeling dates to Wrights pioneering work, but here is a critical point: Wright invented path analysis in order to estimate the magnitudes of effects when the basic causal pathways were already known (e.g., genetics). That is, given a true causal model, the technique of path analysis could be applied to estimate

it for observed variables. However, this is not how we generally use path analysis or related SEM techniques for analyzing latent variables today. In the behavioral sciences, we rarely know the true causal model. Instead, we usually hypothesize a causal model, and then we test that model using sample data. This context of use is vastly different from that of Wrights. Specifically, when the true causal model is unknown but our hypothesized model fits the data, about all we can say is that our model is consistent with the data, but we cannot claim that our model is proven. In this way, SEM can be seen as a discomfirmatory technique, one that can help us to reject false models (those with poor fit to the data), but it basically never confirms your particular model when the true model is unknown. Bollen (1989) put it this way (emphasis in original): If a model is consistent with reality, then the data should be
consistent with the model. But, if the data are consistent with the model, this does not imply that the model corresponds to reality.

Group Stewardship The rapid deployment of work groups into organizations to increase employee involvement and decision making and increased spans of control caused by organization delayering create an increasing need for organizations to rethink traditional top-down, command and control systems (Spreitzer & Mishra, 1999) used to direct and align the efforts of employees. Given their knowledge of the work being performed, employees close to the work should have the ability to potentially make better decisions than their supervisors (Lawler, 1992a). However, managers face the question of how they can be confident employees will act in the best interest of the organization as they relinquish traditional forms of monitoring and controlling. Numerous studies have been published relating to the factors that motivate employees and influence the basis and focus of their commitment. Agency theorists propose that employees act in their own self-interest to maximize their economic gains (Albanese, Dacin, & Harris, 1997; Davis, Schoorman, & Donaldson, 1997b). Graham and Organ (1993) propose a continuum of organizational relationships between employees and their organizati ons. At one end of the proposed continuum, the transactional relationship is similar to agency theory; employees enter into an agreement specifying what services are to be rendered in exchange for benefits to be received. At the other end of their continuum in the covenantal relationship parties commit to the welfare of each other. The covenantal relationship is based more upon commitment to values and relationships than upon reciprocity or fairness. Such relationships tend to be of long duration and would be expected to result in employees working to improve the organizational context as well as perform the specified duties of their jobs. Stewardship theory (Davis et al., 1997b; Donaldson & Davis, 1991), a form of the covenantal relationship, posits that employees aligned with their organization through common values, commitment to the organizations welfare, and identifying with the organization will act in the best interests of the organization. While some have written about individual stewardship, no work appears to have been published regarding group stewardship, a collectively held sense of responsibility to oversee and improve performance in the groups area of responsibility in accordance with the best interests of the organization. Such groups would be expected to learn, adapt to changing needs, and align themselves to work proactively. While Davis et al. (1997b) propose antecedents expected to promote development of stewardship, no empirical testing has been done to verify the proposed relationships. This research addresses the need to define and operationalize the concept of group stewardship and test a set of antecedents and consequences of group-level stewardship. GROUP STEWARDSHIP In this section the construct of group stewardship will be developed. First, the need for a group stewardship construct will be presented. Next, roots of the stewardship concept in the literature will be reviewed. The concept of stewardship will be further defined by contrasting it with agency theory. In the fourth subsection, recent research concepts including shared mental models and psychological ownership that lead to a formal definition of group stewardship will be presented. The fifth subsection defines and distinguishes group stewardship from related constructs. The final subsections develop sets of expected antecedents and consequences of group stewardship. The overall organization of this section is shown below. The Need for Group Stewardship

The Roots of Stewardship Stewardship Contrasted with Agency Theory Four Concepts Supporting the Development of Group Stewardship as a Group-level Construct Group-level Social-psychological Constructs Shared Mental Models Psychological Ownership Theory Two-way Covenantal Relationships Group Stewardship Definition and Differentiation from Related Constructs Antecedents of Group Stewardship Antecedents of Shared Mental Models Antecedents of Stewardship Consequences of Group Stewardship The Need for Group Stewardship A sense of trustworthiness and commitment to the organization are the two employee traits most valued by managers (Cappelli, 2000). When work requires development of relationships and working across organizational boundaries, employers need people who will stay in their jobs long enough to learn the responsibilities of their jobs and those with whom they work. Employers need employees they can trust when large spans of control and temporal or geographical separation make traditional hierarchical overseership difficult or impossible. Trust is also needed with employees who will be given responsibility to manage relationships with customers or make real-time decisions to control quality and safety. However, at least in corporate America, and increasingly throughout the world, shareholders are the stakeholders to whom leaders give their primary allegiance (Waddock, 2000). The concerns of other parties are subjugated when leaders too narrowly focus on shareholder wealth. As a result, employee loyalty, commitment, and morale have suffered through downsizing, management restructuring, relocation of production to countries with cheaper labor, and re-engineering. Given these changes, employees have increasingly developed a sense of loyalty to their profession or personal careers rather than to a specific employer. At least three forces have brought about this change (Cappelli, 2000; Waddock, 2000). First, many corporate actions have broken the sense of a psychological contract between employers and employees for lifetime employment. Employers who break the employer-employee psychological contract at will in search of better opportunities find employees also feel free to terminate the relationship in search of better opportunities. Secondly, portable pension plans and skills with value across many employers tend to unlock the golden handcuffs keeping employees with a single employer. Third, the prolonged economic expansion and resultant personnel shortages for many jobs have shifted the labor market in favor of the employee. Development of a supportive context in which groups work as stewards in behalf of their organizations may fill an important need for both employees and employers. Employers are increasingly involving employees in work groups hoping to improve decision making, innovation, and responsiveness to customers (Hunter, 2000; Lawler et al., 1998). If structured appropriately, work groups can provide many employees with what they seek: interesting work, open communications, and opportunities for development (Cappelli, 2000). Similarly, a job design and organizational context supporting development of a collective sense of partnership with the organization may provide organizations what they need: committed, trustworthy employees. Creation of a condition in which employees work with a sense of stewardship in behalf of the best interests of their organizations would avoid several of the problems potentially emerging from the use of work groups. For each of the following emerging work group problems noted by Quinn et al. (2000) the potential contribution of stewardship is noted. Groups acting as owners may lead to unintegrated efforts across the organization. Stewards work in the best interest of their organizations rather than self interest. While ownership is about whats mine, stewardship is based on service to the organization. It promotes business-focused solutions and maintains accountability to organization stakeholders who influence or set objectives.

Teamwork may promote an inward focus leading to collusion and a blockage of information flow. Group stewardship grows out of internalized organizational goals and values. It requires identification with both the organization and group values resulting in a balance of priorities across the organization and work group. The continuous challenge of new ideas within groups may lead to unresolved conflict. Stewardship is based on commitment to relationships and service that permits the development of collectively held mental models of teamwork and heedful interactions. The presence of shared mental models enhances the speed and efficiency of group decision making (Klimoski & Mohammed, 1994). The replacement of supervision by self-managed or coached work groups may lead to unproductive discussions and low expectations. The development of stewardship requires clear definition of performance expectations, roles, and authority within a relationship that develops intrinsic motivation. The Roots of Stewardship The concept of stewardship is several thousand years old. Ancient households of distinction or of sufficient wealth had stewards. The steward was empowered to act in behalf of his master during the masters absence. He was fully accountable to his master and rendered an account when called upon to do so. Several words from early languages are translated into the English word steward. For example the Greek words epitropos (trustee, guardian) and oikonomos (overseer) have a clear relationship to the term stewardship (Stavropoulos, 1988). A steward is one who manages anothers property, finances, or other affairs (Soukhanov, 1984). At least in Judeo-Christian tradition, stewardship includes two noteworthy responsibilities that may not be obvious from the words definition. First is the responsibility to manage the assets placed in the stewards trust for the good of the owner and not for personal gain. This responsibility includes the idea of service to another as opposed to mastery over that which is managed. Second, stewardship carries the responsibility to improve or increase the assets being managed. Caretaking or maintenance alone do not meet the requirements of stewardship (Genesis 1 & Matthew 25, 1979). Steven Covey has promoted the concept of stewardship delegation (Covey, 1989). According to Covey, stewardship delegation gives the steward a choice of method while holding him/her accountable for results. This style of delegation is based upon a common vision, mutual trust, and joint commitment in five areas: Clear mutual understanding of the desired results. The steward must be able to see, describe and make a quality statement of what results will look like and when they will be accomplished that matches the managers expectations. Guidelines or parameters within which the individual should operate. Human, financial, technical or organizational resources that may be used to accomplish the desired results. Standards of performance and reporting. Clear accountability requires specifying the criteria used to evaluate results as well as the times when reporting and evaluation will take place. Receipt of good and bad consequences that could result from the evaluation. Consequences could include financial rewards, psychic rewards, job assignments, and expected natural consequences tied to the overall mission of the organization. Covey contrasts stewardship with the me and mine perspective of ownership (Covey, 1998). Covey proposes an ownership attitude is typified by use of freedom to maximize personal wealth without concern for other stakeholders. In contrast, stewardship implies representing or being part of something else. It is a clear expression of interdependence, feeling were in this together, and is consistent with a social contract in which employees are guided by a shared vision, mission, and value system. Peter Block wrote a well-known practitioner book about stewardship (Block, 1996). Block says stewardship is holding something in trust for another. To Block stewardship is not only being accountable for ones specific area of responsibility, but includes willingness to be accountable for the well-being of the larger organization of which one is a part. Stewards act in service of rather than in control of those around them. Their objective is building the capacity of future leaders to govern themselves. Block argues stewards choose authentic service over selfinterest through four practices. Empowerment of partners versus control through patriarchy. Patriarchy embodies a parentchild

relationship between the leader and the governed. The key values of patriarchy are consistency, control and predictability. Partnership implies a balance of power between the leader and those within his responsibility. Partnership values placing control close to where the work is done and yielding consistency of centralized management to support for local units in creating policies and practices that fit local needs. Commitment to a larger community. Block believes focusing attention on the individual or small work group breeds self-centeredness and feelings of entitlement. Shared purpose and meaning. Employees join leaders to determine purpose and culture. Balanced and equitable distribution of rewards. As every level in the organization shares in wealth creation, the results of success in the marketplace are equitably distributed. The distribution of rewards demonstrates integrity with respect to the expressed value of contributors. Block writes What is difficult is that we need commitment from people when we can no longer offer them much security. . . . The environment is too unstable to promise a future. . . . We need to create a workplace that evokes commitment that is not based on false promise (Block, 1996, p. 21). However, leaders can commit themselves to create stewardship-based organizations to pursue purposes that transcend short-term self-interest. Those working in such organizations can discover what it means to commit themselves to acts of service in a part of an organization that is theirs to create. With the element of service at its core, stewardship creates a form of governance that offers choice and spirit to core workers so they, in turn, can offer the same to their marketplace (Block, 1996, p. 22). Stewardship Contrasted with Agency Theory Organizational researchers have studied what motivates managers and how to achieve alignment between managers behaviors and the interests of principals. Within this field of research, agency theory is frequently used. Agency theory deals with situations in which principals and managers have differing goals and risk preferences. It analyzes the contract between principal and agent. Agency theory operates under the assumptions that humans will behave in ways that maximize their selfinterest and will tend to be more risk averse than their principals (Eisenhardt, 1989). Stewardship theory has been proposed to explain what causes manager-principal interests to be aligned. Managers are said to act as stewards when they are motivated not by individual goals, but rather by motives that are aligned with the objectives of their principals. Managers are said to act as stewards when they believe that personal needs are best met by working toward the best interests of the organizations collective needs (Davis et al., 1997b). Agency theory can be contrasted with stewardship theory. Some have argued that agency theory is broad enough to accommodate the conditions described by stewardship theory as proposed by Davis et al. (Albanese et al., 1997). However, given the difference in the core motivations assumed to drive manager behaviors (Preston, 1998) as well as the difference in governance structures used, contrasting agency with stewardship theory is useful in defining stewardship. Six key factors which Davis et al. (1997b) propose differentiate stewardship from agents who seek to maximize their personal gain are described in the following paragraphs. Motivation is the primary difference between stewardship and agency. Extrinsic rewards are used in agency theory: tangible, exchangeable commodities that have market value. In contrast, stewardship theory relies more heavily on intrinsic motivation. In terms of the job characteristics model, work designed to provide meaningfulness, responsibility for outcomes and knowledge of results should mediate the relationship between tasks and work motivation (Hackman & Oldham, 1976). Thomas and Velthouse (1990) developed the four-factor model for intrinsic motivation that is the basis for most current individual and group-level empowerment work. Their model posits that meaningfulness, autonomy, impact and self-efficacy are keys to intrinsic motivation. More recently the intrinsic motivation model has been revised by replacing impact with sense of progress through feedback systems (Thomas & Jansen, 1996; Thomas & Tymon, 1997). The expected outcomes of increased intrinsic motivation are expected to be higher levels of effort and attention to doing the task well (Thomas & Tymon, 1997). Identification with the organization is posited to be a key factor in stewardship. Organizational identity includes those characteristics which are central, unique, and enduring (Pratt & Dutton, 2000). Identification occurs when a person or group employs elements of the organizations

identity to define oneself (Pierce, Kostova, & Dirks, 2001). Managers who identify with their organizations take comments about the organization as also referring to themselves and are more likely to engage in spontaneous, cooperative, unrewarded behaviors (Mowday et al., 1982). In contrast, managers who externalize organizational problems to avoid blame would be expected to have lower levels of identification. Hence, individuals who identify with their organizations would be more likely to become stewards while those who externalize blame are more likely to behave as agents. Internalization of organization values is also expected to accompany the development of stewardship. Mayer and Schoorman (1992) found that organizational commitment is multidimensional consisting of continuance commitment (e.g., the desire to remain with the organization) and value commitment (e.g., belief in and acceptance of the organizations goals and values). Value commitment is associated with levels of effort on core work tasks, improving the context within which one works (Pierce et al., 2001), and intent to remain with an organization (Mayer & Schoorman, 1992). Value commitment is not an outcome with economic value and would not be expected to be associated with agency theory. The source or use of power is another important factor in differentiating management stewardship from agency theory. One source of power found in organizations is institutional power vested in the holder by virtue of position within the organization (Gibson, Ivancevich, & Donnelly, 1991). In principalagent relations, control is likely to be maintained through use of institutional power to establish the desired levels of coercion, hierarchical control, and influence over rewards. In principal steward relationships personal power is more likely to be used (Davis et al., 1997b). Personal power utilizes influence derived from perceived expertise or affective relationships where individuals identify with each other. Management philosophy is a key factor differentiating the nature of principal steward and principal agent relationships. The unit of analysis in principalagent relationships is the contract. Agency theory assumes agents and principals are likely to have conflicting goals. The actions of agents are controlled through use of an economic or employment contracts. Performance is ensured or monitored through the use of information systems (Eisenhardt, 1989). The principalsteward relationship is based upon trust and interpersonal relationships where trust is defined as a willingness to be vulnerable in the context of a relationship based upon the belief that another party is competent, honest, reliable and concerned about [ones] own interests (Spreitzer & Mishra, 1999, p. 159). The differences in management philosophies between agency and stewardship are reflected in the sources of commitment which are assumed to develop. Agency theory is consistent with the exchange approach that views commitment primarily as an outcome of the inducements offered for contributions given. Stewardship theory is consistent with psychological approaches that depict commitment as a positive, high-intensity orientation toward the organization that develops as individuals goals and values become integrated or congruent with the goals and values of the organization (Mayer & Schoorman, 1992). Individualism and collectivism are aspects of culture aligned respectively with agency and stewardship theory. In cultures stressing collectivism, the individual defines self through group or family membership; group membership is an important part of identity and achievement. While individualism and agency theory place personal goals over group goals, collectivism and stewardship define success in terms of the collectives success (Davis et al., 1997b). Stewardship Theory Stewardship theory, linked to Donaldson (1990) and Davis et al. (1997), is separated from the Agency theory because of the hypothesis that managers should be less individualistic, less opportunistic and less self-serving than usually happens. On the contrary, managers should be more collectivist, more pro-organization and more trustworthy. Stewardship theory implies that managers can better achieve their objectives by serving to the multiple interests of organization. "Homo economicus" is replaced by a "steward" whose behavior is pro-organization as well as more collectivistic than individualistic and self-serving (Robins, 2008).

A steward who successfully improves the performance of the organization generally satisfies most groups, because most stakeholder groups have interests that are well served by increasing rganizational wealth. (Vaisanen, 2006 in Robins, 2008, pgs. 333) There is, however, the same problem as in the theory of stakeholders. Unless clearly defined who is a stakeholder, it is unclear whose interests should be supported.
Stewardship theory
In contrast to agency theory, stewardship theory posits that managers are essentially trustworthy individuals and so are good stewards of the resources entrusted to them (Donaldson, 1990; Donaldson and Davis, 1991, 1994). Since inside (or executive) directors spend their working lives in the company they govern, they understand the businesses better than outside directors and so can make superior decisions (Donaldson, 1990; Donaldson and Davis, 1991, 1994). As a result, proponents of stewardship theory contend that superior corporate performance will be linked to a majority of inside directors as they naturally work to maximise profit for shareholders. In the well-known language of motivation (McGregor, 1960), stewardship theory plays a Theory Y view of managers to agencys Theory X perspective, arguing that an overemphasis on monitoring is unnecessary for the board to impact on corporate performance. Stewardship theory is based on two premises; namely, that managers are naturally trustworthy (Donaldson, 1990; Donaldson and Preston, 1995) and/or that agency costs will be minimised as a matter of course, as senior executives are unlikely to disadvantage shareholders for fear of jeopardising their reputations (Donaldson and Davis, 1994). Further, even if agency costs are a significant concern to a company and monitoring is necessary, stewardship theorists also hypothesise that outside or independent directors will lack the knowledge, time and resources to monitor management effectively (Donaldson and Davis, 1994). As with agency theory, however, there is no clear empirical evidence to support any claim that a preponderance of inside directors provides superior corporate performance. Since stewardship theory is a mirror of agency theory, it is worth reiterating that the overwhelming evidence both from individual studies (e.g. Kesner et al., 1986; Daily and Dalton, 1992a, 1992b, 1993) and meta analyses (Dalton et al., 1998, 1999; Rhoades et al., 2000) fails to establish any clear relationship between board composition and/or leadership structure and corporate performance or behaviours. The processes that link the board of directors to superior firm performance are not made explicit in the stewardship literature, although making superior decisions (that in turn positively affect corporate performance) is regarded as a key issue (Baysinger and Hoskisson, 1990). Access to information and the ability to take a long-term view are seen as key aspects of the decision-making process (Donaldson and Davis, 1994). For example, studies have examined the superior amount and quality of information possessed by inside directors (Baysinger and Hoskisson, 1990), the apparent relationship between investing in the long-term (R&D spending) and inside directors (Baysinger et al., 1991) and a more balanced approach to CEO compensation taken by inside directors (Boyd, 1994). The implication from these findings is that, because inside directors know the company intimately, they have superior access to information and are therefore able to make more informed decisions. If stewardship theory holds, we would expect to find that: Pattern 2(a): High levels of inside directorships are associated with high access to information, which leads to high quality decision-making and, consequently, high corporate performance. Alternatively, we would expect that if there were few inside directors on the board, the board would not be in a position to fully understand the company. It would only have access to information provided by management and would lack the contextual nature to make more informed decisions. Similarly, outside directors would not have the same access to informal knowledge sources within the firm. As a result, decisions made by a board dominated by outsiders would be of a lower quality and this would in turn lead to low firm performance. Therefore, we would expect the following pattern: Pattern 2(b): Low levels of inside directorships are associated with low access to information, which leads to poor quality decision-making and, consequently, poor corporate performance.

3.2 Stewardship Theory

Contrary to agency theory, stewardship theory views the managers as stewards of the organization who derive more utility from organizational and collective achievements than individual achievements (Davis et al, 1997). Stewards prefer organizational goals to individual goals, however, this preference is not irrational as they gain more utility by pursuing the organizational goals and hence, the principals goals. Therefore, stewards are trustworthy and they work hard to achieve the organizational goals (Donaldson & Davis, 1994). Stewardship theory suggests that, provided the steward is getting a salary/ compensation necessary for his survival, the steward will always choose to pursue organizational goals over personal goals. This happens because the steward strongly believes that the achievement of organizational goals will also lead to the achievement of his personal goals. He links the success or failure of the organization to his personal success or failure. He will rejoice and take pride in an organizational achievement in the same manner as he would do in case of a personal achievement. The steward not only takes credit, but also takes a personal blame for an

organizational failure. Therefore, the steward will not increase his personal utility at the cost of organizational goals or principals utility. Hence, in such a setup, the principals need to trust and empower the stewards in order to maximize their own utility, rather than trying to monitor and control them. Therefore, under a stewardship regime, the job of board of directors is support and enable rather than monitor and control. Davis et al. (1997) suggest that rather than questioning whether stewardship theory or agency theory is applicable in a given situation, one should understand that there are external and internal factors affecting the managers that will determine whether they act as agents or stewards. Internal factors like motivation, identification and use of power, along with external factors like management philosophy, power distance and culture, are important determinants of whether stewardship theory will apply. Managers motivated by intrinsic factors are much more likely to act as stewards rather than those who need external motivation. A person who has a strong need for achievement and recognition, and aspires to reach a stage of self-actualization, is a likely candidate for being a steward. An agent on the other hand would need tangible and measurable extrinsic rewards to be motivated to perform well. Hence, stewards aspire for achievement of higher order needs, whereas agents aspire for fulfillment of lower order needs. Managers who are stewards identify themselves with the organization, treat organizations success and failures as their own, and take comments directed towards the organization as personal. However, a manager whose selfimage and concept of self is not affected by the organization is more likely to act as an agent. Stewards derive their power through their personal relations and expertise, whereas agents depend upon organization position to derive power. Organizational power is easy to obtain, however, it is not easy to maintain it in the long run. Personal power on the other hand, which is the preferred source of power for stewards, requires managers to invest time in building relations and earning respect by demonstrating expertise. Along with the personal characteristics of the managers, the management philosophy prevalent in the organization also plays an important role in determining whether mangers act as stewards or agents. If the management believes in keeping a tight control and not providing a lot of scope for flexibility, agency behavior will dominate. On the other hand, empowered managers, who are given flexibility to make independent decisions based on their judgment, are more likely to become trusted stewards. Also, if the organization is based in a low power distance culture, there will be more trust and clarity in communications, leading to a prevalence of stewardship. Finally, managers in a collectivist culture are more likely to be stewards than the managers working in a culture where individualistic goals and achievements are given preference.

Stewardship Perspective Stewardship theory (e.g., Davis, Schoorman, & Donaldson 1991, 1997a) adds another dimension to studying the inter-relationships between incubators and entrepreneurs. According to this theory, individuals are motivated to behave in a manner that is in the best interest of their organization (Donaldson and Davis 1991). They attempt to align their goals with the goals of their organization. Performance variations, according to the theory, often have to do with the structural situation in which the actor works within, thus there must be sufficient resources and means to facilitate good action (Donaldson & Davis 1991). Incubator directors, especially those of the not-for-profit and public incubators often consider themselves as stewards of their fledging tenants. In addition, the best interest of the incubator and the community immediately surrounding it is linked with the overall success of the tenant-entrepreneurs. Therefore, both incubator and tenant would like to see the startup grow and prosper. One important measure of an incubators success is the number of its tenants-graduates and their respective success rates. Graduates success also contributes to the local economy and job creation in the surrounding community. The incubator can therefore be credited for that positive economic impact and value to the local geographic area. Thus, goal alignment seems to be critical in order for all stakeholders to obtain their goal. In dealing with multiple stakeholder groups with competing objectives, stewards are motivated to make decisions they perceive in the best general interest of the group in order to provide the most overall benefit (Davis, Schoorman, & Donaldson 1997a). Thus, according to stewardship theory, the manager of the incubator and the entrepreneur benefit most when self-serving behaviors are put aside (Davis, Schoorman, & Donaldson 1997a) and the two parties collaborate. Agency and stewardship theory
Cribbs (2006) contribution to a recent Policy Quarterly, framing the current contracting system in the voluntary

sector in terms of agency and stewardship theory, provides a useful benchmark in this discussion. Agency theory is the underlying philosophy of the contracting model that developed in New Zealand in the late 1980s, as expressed in the Public Finance Act 1989. Agency theory is based on an assumption that people are self-interested and motivated to maximise their own advantage. Principals use contracts to delegate tasks to agents, who must be carefully monitored to ensure that they undertake the task with maximum effi ciency and do not abuse the system. Inevitable aspects of the system are goal confl ict between principal and agent; information asymmetry, or different levels of knowledge; and moral hazard, whereby the agent will try to outwit the principal, leading to adverse selection when principals contract with agents who are not up to the task. Guarding against these requires a great deal of enforced legal compliance. However, in her research with community organisations Cribb found that reality defi ed the model. She saw that people in the organisations placed a strong emphasis on their relationships with their clients, rather than on the compliance requirements set out in the contract. Cribb argued that while these requirements are there to ensure high standards, in reality contracting and funding agreements with government agencies were seen to be driving down standards of care (Cribb, 2006, p.12). Cribb instead offers the theory of stewardship, which relies on goal alignment and a relationship of trust, and recognises altruism, whereby people work for the good of the organisation rather than their own fi nancial gain. The attention to shared goals means less emphasis is required on auditing and monitoring. The agency-theory foundations of the contractual model which replaced the previous grants-based system have been summarised as follows: A central policy department advised the responsible Minister on the services required by the Crown. The Minister purchased the required services as outputs from the department itself, or from a third party ... The policy department monitored the delivery of the required services under a purchase contract. (Buchanan and Pilgrim, 2004, p.4) The 1999 election of the LabourAlliance coalition government signalled a change from agency theory but not from the contracting process itself, which was by then entrenched. Government has developed various strategies since 1999 to improve relationships and capacity-building. The 2004 amendment to the Public Finance Act refl ected the desire for a stronger governance relationship between government and crown entities; public service standards for public resources; and managing for outcomes rather than outputs. However, in a June 2006 paper, Robert Buchanan of the Offi ce of the Controller and Auditor-General replied to Cribbs call for a move from agency to stewardship theory. He advised that the Offi ce of the Controller and Auditor-General has advocated a riskbased approach to procurers, as one means of reducing transaction and compliance costs and ensuring that the available monitoring resources are used effectively and effi ciently (Buchanan, 2006, p.11). He suggested that promotion of stewardship was simplistic since the notion of stewardship already underlies not only public service ethics but also the policy of having legally enforceable obligations in respect to resources that pass from one sector to another ... the fact that a contract is legally enforceable does not mean that the relationship as a whole should be characterised by a mentality of enforcement and compliance (Buchanan, 2006, p.11). It should also be noted that contestable funding, a policy which sits more easily with agency theory than with stewardship theory, is now compulsory. This follows a 2005 complaint by an opposition MP about the Ministry of Health procurement and contract management processes, under which former employees were winning non-contested contracts. This led to another report from the Controller and Auditor-General and the end of sole-provider emphasis in procurement (Offi ce of the Controller and Auditor General, 2005). The problem remains: how to fi t compulsory contestability into a model consistent with the governments frameworks?