Agency theory assumes that the best organizational form is one where the leader (CEO or president) owns 100 per cent of the company; in this case, the top executive is also the principal (owner). When the top executive is not the sole owner, then that individual becomes an agent (employee) of the firm, at which point agency problems begin to ensue. Agency problems are said to occur when agents pursue individual goals, which are not necessarily consistent with those of the organization. In addition, risk preferences of agents are different from those of the principal, resulting in decision making that is less than optimal. Agency theory differentiates principals from agents in a very simplistic manner because it treats ownership as a dichotomous variable, and it also considers ownership at the individual level. This had led to fairly basic notions about goal congruence and risk taking behavior. Owners pursue organizational goals and are risk neutral, while agents pursue personal goals and are risk averse.

In reality, ownership should be measured as a continuous variable because in many cases, particularly in entrepreneurial firms, management continues to retain partial ownership in the firm (through stock grants, options). Agency theory does specifically recognize stocks for executives because it views these programs as a form of monitoring (incentive alignment) that helps to mitigate the agency problem. However, the concept of a continuous ownership variable is important because it also allows for consideration of ownership level within the organization. Ownership is often extended beyond the CEO to the top management team, and in many cases, to all employees. When this occurs, the phenomenon is not how much one person is encouraged to act like an owner but how many people within the firm are behaving like owners. Even though ownership is dispersed, a large percentage of firm ownership might be retained within the organization rather than with outside shareholders or with one executive. Thus, not only can the extent of CEO or top management team ownership be studied, but proliferation of ownership throughout the organization can also be investigated. In this case, the entire organization moves toward the "principal" point on the continuum, and the focus of study is firm performance, which is affected by ownership proliferation.

This paper extends agency theory research by considering the effect of three different forms of ownership on firm performance. Rather than studying the degree of ownership of only one person (such as the CEO), this research considers the proliferation of ownership throughout the organization and how different levels of ownership affect short and long-term firm performance. In addition, the moderating effect of firm level risk is investigated.


Agency theory has been used to understand situations where an individual delegates responsibility for a task to other persons (Fama, 1980). Agency theory has been classically applied to study the relationship between owners of an organization and the managers who run those firms (Fama & Jensen, 1983). In practice, it has been most often employed in research on the mechanisms used by owners to align CEO interests with those of the organization (Gomez-Mejia, 1994). The exception has been a few studies that have extended agency theory to other positions such as university faculty (Gomez-Mejia & Balkin, 1992), sales representatives (Eisenhardt, 1985), and production workers (Welbourne, Balkin & Gomez-Mejia, 1995).

In all cases, agency theory was used to explicate alternative ways of controlling behavior of individuals who have been delegated work by someone. The person delegating the work is called the principal, and the individual to whom tasks are assigned is referred to as the agent. Since most agency theory research has involved top management, much of the work has dealt with the study of incentive systems used to align the interests of executives (agents) with those of the owners (shareholders). Executives have been the focal point of study because their behavior is difficult to monitor through more traditional mechanisms (i.e. direct supervision, close monitoring of behaviors). Agency theory assumes that the best way of aligning employees’ interests with those of the principal is through formal monitoring, and only when the cost of monitoring is high should a company consider alternatives to formal monitoring. In the case of executives, formal monitoring is assumed to be impossible, therefore the study of executive compensation, as a form of monitoring, has been pursued by a number of researchers from a variety of academic fields (e.g. organization behavior, accounting, finance, human resource management) (for a review see Gomez-Mejia, 1994).

One form of incentive is firm ownership, which serves to create a situation where the goals of the agent are identical to those of the principal (Tosi & Gomez-Mejia, 1994). By aligning goals, stock plans help mitigate problems associated with risk taking propensities of the agents. Agency theory assumes that principals are risk neutral because their portfolios are not 100 per cent tied to one firm, however, agents, who cannot diversify their employment portfolios, are considered to be risk averse. Being risk averse, agents will make decisions that minimize risk in order to assure continued employment. According to Jones and Butler (1992: 73), "risk aversion encourages managers to select safe projects that provide normal rates of return." By providing executives with some form of ownership in the firm, it is assumed that they might be more willing to take risks that optimize long-term performance of the organization. Assuring an adequate level of risk taking is important in entrepreneurial firms where risk taking is necessary in order to pursue opportunities.

Agency theory has been applied to the study of top executives because their behavior is not easily monitored, but the theory should be equally applicable to any situation where work is delegated, and particularly applicable when jobs are not easily monitored. In fact, agency theory has been used to understand organizational control for the overall employee population. According to Becker and Olson (1989: 247): "Two management strategies are possible. First, managers can attempt to allocate some of the firm’s business risk to labor, with the aim of increasing workers incentive to act as owners. The current support for profit sharing and employee stock ownership plans by firms is due, in part, to a belief that these plans will reduce agency costs by aligning the interests of the workers with the current and future profitability of the firm. A second strategy is to closely supervise and control employees, allocating the greater share of the firm’s business risk (and associated returns) to the shareholders." These two forms of control parallel the types of control strategies that are suggested by organizational theorists (Ouchi, 1979; 1980; Thompson, 1967). Organizational theorists argue that control mechanisms can be described as focusing on behaviors versus outcomes, where behavioral control results when policies and procedures (or compliance) are used as the dominant method of operating, and outcome control ensues through the use of incentives (alignment). Recently Eisenhardt (1985) combined the agency theory and organizational theory approaches to organizational control and noted that two underlying control strategies emerge from both theoretical perspectives.

As noted by Becker and Olson (1989), stock option programs, which provide a form of ownership in the firm, can increase alignment among all employees within an organization. These programs provide individual employees with incentives to work toward the organization’s goals in the same way that CEO stock plans are incentives for executives to make decisions that will support the interests of the shareholders or owners. These systems attempt to create an environment where employees are part owners of the business.

It has been suggested that in entrepreneurial firms, where rapid change is occurring and bureaucratic structures have not been established, basically all jobs are difficult to monitor (Jones & Butler, 1992). Entrepreneurism has been defined as the process by which firms notice opportunities and act on those opportunities (Kirzner, 1993). If a firm (rather than an individual) is to remain entrepreneurial, then all employees in the company need to pursue and act upon opportunity, and this requires all personnel to be somewhat willing to take risks and pursue opportunities that enhance the organization’s goals. Thus, entrepreneurial firms should be more successful if ownership is widely dispersed throughout the organization. The argument is not that the CEO have an enhanced stake in the firm but that more employees share in the ownership of the company. Thus, this paper suggests that as ownership is expanded to others in the firm (top management and all employees), firm performance will increase.

Hypothesis 1: Proliferation of ownership throughout the workforce positively affects firm performance in entrepreneurial organizations.

Ownership can only be dispersed if the CEO has less ownership in the firm, therefore, it is expected that high levels of CEO ownership will have a negative impact on firm performance.

Hypothesis 2: High levels of CEO ownership will have a negative effect on firm performance in entrepreneurial organizations.

The advantage of ownership is that goal alignment will ensue and agents will be more likely to take risks. Therefore, it seems that the "gain" from ownership should be more significant in higher risk firms. In fact, research on strategic human resource management suggests that newer, dynamic, growth firms are more successful when they use compensation packages that have lower base pay and a higher incentive component (Jackson, Schuler & Rivero, 1980; Miles & Snow, 1984).

Hypothesis 3: Proliferation of ownership will have a greater impact on firm performance for higher risk firms.

Because agency theory focuses on problems that arise when ownership is separate from management, most of the research done has been conducted within large, established organizations where data on the top executives can be easily obtained and studied. However, agency theory has unique and particularly interesting applications for the study of smaller, entrepreneurial firms (Jones & Butler, 1992). In addition, smaller, newer organizations present a unique opportunity in which to study the issue of ownership and more adequately address causation. As Romanelli (1989: 369) notes, "new organizations are notoriously poor at surviving their early years." Thus, samples of smaller and younger firms have more variation in performance, including firm survival.

IPO firms are organizations that offer their stock to the public market for the first time; they are moving from being a privately owned firm to a publicly owned company. The move is not an easy one, and it requires at least one year of the company’s time (particularly that of the top management team) in preparing and "marketing" the company. In addition, the firm undergoes numerous internal changes as it prepares to become "professional" and submit to the scrutiny of shareholders, investment bankers, and the Securities and Exchange Commission. IPO firms might be considered more successful than other small organizations because they have survived to this stage. However, they still face a high risk of failure that is comparable to that encountered by other start-up organizations. Of the 3,186 companies that went public in the 1980s, with stock listed on the NYSE, AMEX, or NASDAQ, only 58% were still listed by 12/31/89 (Zeune, 1993). Comparably, a Dunn and Bradstreet study showed that 53% of all failures and bankruptcies of firms in 1980 occurred less than five years after founding, and 80% failed in less than ten years (Romanelli, 1989).


Previous Page | Main Menu | Next Page



1997 Babson College All Rights Reserved
Last Updated 4/27/97 by Germaine Wong

To sign-up for the Center for Entrepreneurial Studies' publication lists,
please register with the
Entrepreneurship WebTeam.