Structural Contingency Theory Research Paper

1. Bureaucratic Structure And Its Contingencies

The structure of an organization includes how the work of the organization is divided up and allocated among people, in terms of job specialization, sections, and departments. It also includes the number of levels in hierarchy and how many subordinates report directly to each manager, that is, his or her span of control. Another aspect of organizational structure is the degree of formalization: how far rules, standard operating procedures, and manuals govern the way people work. A further aspect of structure is the degree of decentralization, the degree to which decision taking authority is distributed down the hierarchy (Jones 1998).

Many of these structural variables tend to be correlated so that they compose a single structural dimension of the degree of bureaucratic structure (Donaldson 2001). An organization that is low on bureaucratization has low specialization, few departments and sections, narrow spans of control, low formalization, few hierarchical levels, and high centralization. Conversely, an organization that is high on bureaucratization has high specialization, many departments and sections, wide spans of control, high formalization, many hierarchical levels, and low centralization. Organizations are distributed widely along this continuum of bureaucratic structure.

2. Bureaucratic Structure And The Size Contingency

Size is the number of people who must be organized, that is the number of employees or members of the organization. For high performance, the organization needs to have the degree of bureaucratic structure that fits its size (Child 1984).

An unbureaucratic structure fits an organization of small size, while a highly bureaucratic structure fits a large size. For small size, the required, unbureaucratic, structure is a short hierarchy, with most of the decisions being taken by the top manager. The top manager controls the organization directly by giving instructions (Child 1984). There is little formalization because the top manager is making decisions on each case, rather than bothering to identify abstract rules to anticipate future cases. There is low specialization and few departments and sections, because there are few people among whom the work could be divided. The span of control is narrow, because each manager must supervise subordinates who are doing a broad range of tasks.

Conversely, large size entails a bureaucratic structure that has a tall hierarchy because an additional level of management has to be introduced each time the number of employees exceeds the feasible span of control of their supervisor. Size also brings increased complexity and this combines with the tall hierarchy to force top management to delegate decisions down the hierarchy and so decentralization increases. With so many people to be organized in the large organization, the work is finely divided up among them, producing an elaborate division of labor. There is high specialization and many departments and sections. The span of control is wide, because each manager can supervise more subordinates because the subordinates are doing a limited range of tasks. There is high formalization because, with larger size, many decisions recur about similar cases, leading to codification of best practice into rules and standard operating procedures (e.g., a hiring procedure). Such formalization relieves managers from having to attend to each decision, so that administration can be delegated to clerks who follow rules, which speeds decision-making, cuts costs, and is another reason why managerial spans increase with size. Overall, top management has relinquished much direct control and instead controls the organization indirectly and impersonally, through formalization (Child 1984). As the organization grows in size over time, so it needs to increase progressively the degree of bureaucratization of its structure in order to maintain the fit that sustains high performance.

Information technologies, whether used in operations or administration, tend to reinforce the effects of size (Donaldson 2001). Information technologies increase bureaucratization, such as by increasing formalization, because they reduces uncertainty, though their effects are weaker than size. Moreover, computerization itself is really a form of bureaucratization. Work is executed according to computer programs that are rules (e.g., ‘If the checking account balance is less than $500, then apply a $10 fee to every transaction’). Knowledge workers and other employees working with computers are constrained in their actions by the programs in their computers. Thus computerization is electronic formalization that allows top management more indirect, impersonal control over lower-level employees, so that top management can further increase decentralization.

Bureaucratic structure is sometimes criticized for having too many managers and administrators relative to operating personnel (e.g., workers). However, the ratio of managers and administrators to operating personnel tends to fall as size increases. While larger organizations have more levels in their hierarchy, their spans of control increase so as to more than compensate (Jones 1998). Thus bureaucratic structure brings economies of scale in administration (Donaldson 1996).

3. Bureaucratic Structure And The Task Uncertainty Contingency

While a bureaucratic structure fits low task uncertainty, a less bureaucratic structure is required to fit high task uncertainty.

The creation of effective rules relies on all the cases of the same occurrence being similar. In routine tasks this holds, allowing rules to be formulated. However, some tasks are uncertain, so that one case varies markedly from another, and future cases differ from past ones. Some uncertainty in organizations comes from the environment, for example, competitor pricing is hard to predict. Other task uncertainty comes from innovation inside the organization, such as designing a new product or developing a new technology to distribute a service.

Whatever the source of the uncertainty, tasks high in uncertainty cannot be handled effectively by rules. Tasks of medium uncertainty can be handled effectively by the hierarchy. However, the greater the uncertainty, the more information that has to be processed by the organization. Because of the pyramidal shape of the hierarchy, it soon becomes inadequate for dealing with more than moderate uncertainty. High task uncertainty, such as involved in radically new products or services, or making many innovations at the same time, requires reliance upon experts. The high uncertainty in their tasks prevents the organization from governing their work effectively by rules. Some decision-making authority needs to be delegated to these experts, even if they are at low hierarchical levels. Similarly, managers of innovative programs need to be able to act autonomously. Thus task uncertainty is fitted by a structure that is lower in formalization and centralization (Jones 1998).

The effect of the task uncertainty on bureaucratic structure is secondary to that of size. High task uncertainty partially offsets the high formalization that large size causes by lowering formalization. Conversely, high task uncertainty reinforces the high decentralization that large size causes by further increasing decentralization. Thus a large organization with highly uncertain tasks is less formalized and more decentralized than a large organization with highly certain tasks. The reduction in formalization decreases bureaucratic structure, while the increase in decentralization increases that aspect of bureaucratic structure.

Moreover, these effects of task uncertainty are localized to specific parts of the organization. For a firm that is developing new products, the uncertainty of its tasks is greatest in some parts, such as the research and development department that is inventing the new product. In contrast, the manufacturing department of the firm focuses mainly on the routine production of existing products, so that the uncertainty of its tasks is lower. Thus the required structure of the research and development department is low in formalization, while the manufacturing department is high. Similarly, the research and development department employs qualified scientists and technologists and devolves some decision-making to them so that they exercise autonomy and creativity. Conversely, the manufacturing department employs mainly less educated personnel who lack professional qualifications and trains them to conform to the operating procedures of the firm, so the department is run in a more formalized and centrally planned way. Nevertheless, innovation entails solving some novel problems that involve all the functional departments. For example, research needs to know whether manufacturing can make their new design on its existing machines. The integration required between the functions necessitates the adoption of cross-functional project teams. These contain representatives from each of the functional departments who are brought together for collaboration (Jones 1998). The teams provide lateral linkages that run across the hierarchy and rely upon ad hoc internal coordination rather than planning or formalization. Therefore they are a further way task uncertainty reduces bureaucratic structure.

4. Divisional And Matrix Structure And The Diversification Contingency

4.1 Functional And Divisional Structures

Another aspect of structure is how activities and responsibilities are grouped in the organization. In a functional structure, the managers reporting to the top manager (i.e., the CEO; chief executive officer) are specialized by function, including operational functions (e.g., manufacturing). In contrast, in the divisional structure, the senior managers are specialized by their products or services, customers, or geographical areas, and each has a broad range of operational functions within their division. A divisional structure is more specialized, formalized, and decentralized than a functional structure and so divisionalization is an extension of bureaucratization (Donaldson 2001). However, the contingency causing divisionalization is primarily diversification, with size playing a minor role (Donaldson 2001) so that the contingency determination of divisionalization differs from that of bureaucratization.

Organizations vary in their levels of diversification of their products, services, customers, or geographical areas. For brevity, product diversification will be discussed here but not service and customer diversification because similar considerations apply. Increasing diversification reduces the interdependence between the tasks involved in one product and the others. This requires less coordination between products, so each can be organized as a separate subunit acting autonomously from the other subunits and from the corporate center. Consequently, the corporate center needs to contain fewer operational functions. The greater the diversification, the greater the decentralization needs to be and thus the smaller the head office. As diversification varies from low to high, the fitting structure goes from being a pure function to a pure divisional structure. At intermediary levels of diversification, the role of the central functions decreases with each level and the role of the product heads increases. Where a firm is undiversified, such as producing only one product, then specialization by function fosters economies of scale and avoids duplication. Each function is highly interdependent on the next, so that operational coordination is needed from the CEO, producing centralization. Also, each function is a cost center, with profitability only being assessed for the firm overall.

The next level of product diversification is where a firm offers products that are vertically integrated with each other. For example, an aluminum-refining firm might vertically integrate backward by acquiring the bauxite mine that supplies its feed material. It might also vertically integrate forward by creating a plant that fabricates its refined aluminum into window frames. Each stage along the value-added chain, that is, mining, refining, and fabricating, becomes an organizational subunit, which may be termed a division. Some decisions will be taken autonomously by each division, such as the types of production equipment to purchase, so that the structure is more decentralized than a functional structure. However, a large flow of materials must be coordinated between divisions. This necessitates some restrictions on divisional autonomy and the presence of corporate-level staff with expertise in transportation and the planing of production and inventory. Thus the corporate center contains many staff additional to the administrative functions such as accounting. Whereas the fabrication division is a profit center, the mining and the refining divisions should be cost centers because their sales are to other divisions making their profitability a contentious issue. Divisional general managers need to have their bonuses tied to corporate profitability to encourage them to cooperate.

A medium level of diversification is where a firm has two or more related products. The relatedness between the products lies in similarities among their materials, production technologies, skills, brands, sales force, or distribution systems. Each product is organized as a division and is run quite autonomously, except where the divisions are related and so synergies may be extracted through central coordination. For example, an electronics firm has three product divisions: avionics, computers, and domestic appliances, with each focusing on their product market. But there are commonalties in the underlying technologies so that there is a large central research and development laboratory that pursues major innovations beyond those being researched in the divisions. Therefore, the corporate center contains functions beyond the merely administrative and is quite large in size. Each division is a profit center, but the transfer pricing of shared components moderates its accountability. This is also moderated by large corporate overhead charges for the central laboratories and by corporate decisions about which division receives the profitable, new products developed in the central laboratory. Divisional general managers need to have their bonuses partly tied to that of their own division and partly to corporate profitability to encourage them to cooperate with the other divisions and the corporate staff. The divisional structure boosts innovation by having functions for each product which facilitates functional interaction. However, this is at the cost of some duplication of assets across divisions. Thus, where cost pressures are paramount, the organization instead needs to use a functional structure.

A high level of diversification is where a firm has two or more unrelated products, e.g., a conglomerate. There are no operating synergies to be extracted, so little central coordination of operation is required and the divisions are highly autonomous. Therefore the corporate level needs no functions beyond the merely administrative and so can be small. Each division can be straightforwardly held profit accountable. Divisional general managers need to have their bonuses tied solely to the profitability of their own division.

Area diversification has similar effects. Area diversification is how far the organization contains an area that operates separately from other areas, that is, it designs, manufactures, and sells the product all within its own area. For example, area divisions are frequently found in the food industry, where low price relative to weight makes intercontinental transportation too costly. Low area diversification fits with the functional structure, while high area diversification fits with the area divisional structure.

4.2 Matrix Structures

In a matrix structure, a subordinate reports to more than one boss. Despite the inherent complexity and resulting conflict potential of matrices, they fit some contingencies. A related product organization attains both some product innovation and some cost reduction by using a project-functional matrix structure. Each project manager heads a temporary team, which fosters close interaction between different functional specialists, to facilitate product innovation. Additionally, the heads of the functions coordinate the deployment of resources across projects to reduce costs. The relatedness of the products (e.g., at Lockheed-Gelac all the products are transport airplanes) allows the same resources to be used by the projects. This resource sharing is facilitated by the projects each being at different stages so that the same resources are not required simultaneously. The project-functional structure fits medium product diversity and low area diversity so that operations are geographically proximate and there is no area diversity that needs managing.

In contrast, an area type of matrix is appropriate for an organization that requires coordination along the area dimension in addition to coordination by either function or product. For example, a multinational corporation has offices and plants scattered around the world, each reporting up to functional or product heads. However, the corporation also needs to have a local manager coordinating each geographical area to deal with local issues, e.g., the host government. The increase in environmental diversity, caused by cultural and political heterogeneity, that is, medium area diversification, requires that the organization mirrors and manages this diversity through having a manager for each area. Thus, area type matrices fit medium area diversification. There interdependencies exist between the areas, requiring coordination through high-level functional or product heads, at the same level as the area managers, thereby creating the matrix. Where area diversification is medium and product diversification is low, then core operations need to be structured functionally and so the fitting structure is the area-functional matrix. However, where area diversification is medium and product diversification is also medium, then core operations need to be structured divisionally and so the fitting structure is the areaproduct matrix (e.g., Asea Brown Boveri). Medium product diversification obtains not only for related, but also for unrelated, product diversification. This is because the medium area diversification introduces links across the separate products due to their geographical co-location (e.g., being in France and so having to deal jointly with the government there). Thus, the level of product diversification has to be determined jointly with the level of area diversification, that is, interdependence is established by examining together products and areas.

The heart of structural contingency theory is the concept of fit of organizational structure to the contingency. Many of the more promising developments in structural contingency theory delve into the concept of fit, using it to illuminate organizational performance and organizational change. A major conceptual statement about fit that has influenced subsequent research is by Van de Ven and Drazin (1985). Some researchers have extended the analysis by considering the fit of an organizational structural variable to multiple contingencies (e.g., Keller 1994). Other researchers have analyzed the fit of multiple organizational structural variables to a contingency (e.g., Drazin and Van de Ven 1985). Still other researchers have analyzed the fit of multiple organizational structural variables to multiple contingencies simultaneously (e.g., Pennings 1987). Other researchers have analyzed the differential effects of fit on different aspects of organizational performance such as financial performance vs. social costs (e.g., Kraft et al. 1995). Structural contingency theory has sometimes been criticized for being static, however, on the contrary, it contains a theory of change dynamics in that fit and misfit play crucial roles in organizational structural change (Donaldson 2001). An organization in fit tends thereby to have high performance, which leads it to increase its contingencies, such as size and diversification, so that it moves into misfit and so has lower performance. An ensuing crisis of low performance causes the organization to adopt a new structure that fits the new levels of its contingencies, thereby restoring fit and performance (Donaldson 2001). Whether the performance level of a firm in misfit becomes low enough to trigger structural change depends on other causes of performance that interact with misfit. These other causes include debt and divestment and can reinforce misfit or offset its depressive effect on performance. The set of causes of organizational performance and their effects on organizational growth and adaptation have been formalized into a new organizational theory, organizational portfolio theory (Donaldson 1999).

Fit is traditionally conceptualized as a line of iso-performance, meaning that a fit of organizational structure to a low level of the contingency (e.g., small size) produces the same performance as a fit of organizational structure to a high level of the contingency (e.g., large size). Such a model fails to provide an incentive for an organization to increase its level of the contingency (e.g., size), so that there is no explanation of why an organization changes its level of the contingency, such as by growing larger. A reconceptualization of fit has been proposed in which fit is a line of hetero-performance, in which fits to higher values of the contingency variable produce greater performance than fits to lower levels (Donaldson 2001). Therefore there is an incentive for the organization to change its level of contingency such as by growing larger over time. In this way, reconsidering the effect of fit on performance allows the explanation of changes in the contingency (e.g., size) of an organization in fit, thereby rendering structural contingency theory more comprehensive and dynamic. Thus, deeper analyses of fit, both empirically and conceptually, provide a major route for the ongoing development of structural contingency theory (for developments synthesizing structural contingency theory with other organizational theories see Donaldson 1995).

Bibliography:

1. Child J 1984 Organization: A Guide to Problems and Practice, 2nd edn. Harper and Row, London; Hagerstown, MD
2. Donaldson L 1995 American Anti-Management Theories of Organization: A Critique of Paradigm Proliferation. Cambridge University Press, Cambridge, UK; New York
3. Donaldson L 1996 For Positivist Organization Theory: Proving the Hard Core. Sage, London; Thousand Oaks, CA
4. Donaldson L 1999 Performance-Driven Organizational Change: The Organizational Portfolio. Sage, Thousand Oaks, CA
5. Donaldson L 2001 The Contingency Theory of Organizations. Sage, Thousand Oaks, CA
6. Drazin R, Van de Ven A H 1985 Alternative forms of fit in contingency theory. Administrative Science Quarterly 30: 514–39
7. Jones G R 1998 Organizational Theory: Text and Cases, 2nd edn. Addison-Wesley Publishing Company, Reading, MA
8. Keller R T 1994 Technology–information processing fit and the performance of R&D project groups: a test of contingency theory. Academy of Management Journal 37: 167–79
9. Kraft K L, Puia G M, Hage J 1995 Structural contingency theory revisited: main effects versus interactions in Child’s National Study of Manufacturing and Service Sectors. Revue Canadienne des Sciences de l’Administration—Canadian Journal of Administrati e Sciences 12: 182–94
10. Pennings J M 1987 Structural contingency theory: a multivariate test. Organization Studies 8: 223–40
11. Van de Ven A H, Drazin R 1985 The concept of fit in contingency theory. In: Staw B M, Cummings L L (eds.) Research in Organizational Behaviour. JAI Press, Greenwich, CT, Vol. 7, pp. 333–65