Academic journal article Public Administration Review

A Framework for Evaluating the Government Contracting-Out Decision with an Application to Information Technology

Academic journal article Public Administration Review

A Framework for Evaluating the Government Contracting-Out Decision with an Application to Information Technology

Article excerpt

How can government do a better job at contracting out? Steven Globerman and Aidan Vining argue that contracting out is justified only when one can expect to lower the sum of production costs and the costs of managing the relationship [between government and the contractee (transaction costs). While contracting out can usually reduce production costs, its value depends on whether it can reduce combined transaction costs and production costs. However, most transaction costs can be anticipated before contracting out. The article provides a logical taxonomy that lays out the main issues which determine the feasibility and likely success of contracting out. The key issues in assessing contract design are task complexity, contestability and asset specificity. Potential strategies are available to modify each of these determinants. The authors offer some illustrations from information technology.

Contracting out has been recommended as a way to reduce governmental costs (Saves, 1982; Butler, 1985; Osborne and Gaebler, 1992), although the recommendation has its critics (e.g., Sullivan, 1987).(1) There are certainly contracting-out "horror stories" in which governments appear to have been victimized by opportunistic or incompetent contractees. High-profile examples can be found in information technology (IT) contracting out. One such recent example involves the federal government of Canada's contract with Hughes Aircraft of Canada to modernize Canada's air traffic control system. The project ran years late and hundreds of millions of dollars in excess of original cost estimates with original technical specifications still failing to be met. Although complex contracting out will always present significant risks to a government purchaser, the thesis of this article is that such risks can be effectively mitigated by identifying the precise nature of the risks and implementing appropriate strategies to cope with them.

Information Technology is a good test of the viability of effective contracting-out mechanisms, because it often involves a highly complex transacting environment. It is also worthy of analysis in itself given the growth of IT expenditures in the public sector. However, all of the important issues in contracting out are also generic to simpler activities such as office cleaning or food preparation. Therefore, the prescriptions presented here should be useful to all public sector managers engaged in contracting out.

Contracting out is now fairly common at the state and local government levels (Touche Ross, 1987, 1989). However, for contracting out to save societal resources, several issues must be confronted. First, the objective of governmental contracting out must be clean Second, an appropriate framework must be developed. Third, it must be shown that such a framework can be applied to real contracting-out problems. The objective of the public sector should be to minimize the costs of delivering any given quantity and quality of service(s).(2) Costs, in turn, consist of resource expenditures for inputs (production costs) and the costs associated with "governing" the production process. The evidence suggests that contracting out has a potential for lowering the first set of costs (to be discussed below), but these savings might be more than offset by increases in governance costs.(3)

Our major purpose of this article is to suggest a framework to address two fundamental issues: How can government assess before the event, or ex ante, the potential governance costs that arise with contracting out? How, and under what circumstances, can governance costs be reduced? The framework proposed does not provide a detailed recipe for mitigating all governance costs. However, it does identify alternative instruments that are more or less effective depending upon the relevant circumstances, and it provides a basis for meaningfully categorizing them (Figure 1).(4)


The framework is illustrated with reference to contracting for IT services, that is, any service relating to the design, development, implementation, operation, and maintenance of information technology systems, including service bureau services and facilities management. …

Author Advanced search


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.