Public Policy Sources #35:
The theory: why is private contracting beneficial

[Previous] [Contents] [Next]

Three areas of economic research address the desirability of the Alberta plan: comparisons of for-profit and non-profit enterprises, comparisons of public and private enterprises, and assessments of the effect of competition in the hospital market. All three strands of inquiry provide compelling reasons to examine the social benefits from private contracting of hospital services.

Enterprise comparisons

While the conventional profit-maximizing firm has been a long-standing and enduring subject of analysis in economics, attention has only shifted much more recently to the study of non-profit private and government enterprises. One of the early foundations of this literature contrasted the incentives faced by decision makers in for-profit and non-profit firms (Alchian and Demsetz, 1972). Their fundamental point was that because a decisionmaker in a non-profit firm was unable to be rewarded in the form of profits for making good (i.e., profit-enhancing) decisions, she would tend to make different choices about the firm's operation than a comparable decisionmaker in a for-profit firm. This point has been elaborated and particularized in subsequent literature.

Non-profit versus for-profit

While there are a number of meaningful distinctions between the economic decision environments facing non-profit and for-profit ("proprietary") firms, the salient one for this study is that because "'decision makers' ... are unable to extract residual income in the form of cash ... [they] will choose to take it in other forms" (Pauly 1987). Among these "other forms" (in the hospital context) are "better office facilities, more congenial colleagues, more relaxed personnel policies, or any other personally rewarding activity even if it is more costly to the nonproprietary hospital than its proprietary counterpart" (Clarkson, 1972, emphasis added). In other words, rather than solely maximizing profits, managers in the non-profit setting may be willing to sacrifice profits in order to enhance their own pecuniary and nonpecuniary income.

It is not, however, a theoretical fait accompli that the non-profit environment requires deviation from profit-maximizing behaviour. Danzon has pointed out that "Although the rights to residual profit in a non-profit hospital are not well defined, profit maximization is nevertheless an appropriate model provided the various claimants can agree on maximizing their joint gain" (Danzon, 1982). In the hospital setting, Pauly and Redisch have modeled hospitals as physicians' "cooperatives" with the objective of maximizing the combined incomes of the staff physicians (Pauly and Redisch, 1973). A binding agreement to do so would enable a non-profit hospital's decision making to resemble that of a for-profit hospital's. Specifically, a hospital governed by such a binding agreement would perform as many procedures in a given year as a for-profit hospital, and would use the same mix of inputs (nurses, pharmaceuticals, technology, etc.) as a for-profit. Furthermore, as Newhouse notes, even without the internal quasi-for-profit constraint imposed by income-maximizing physicians, free entry will force surviving non-profits to produce efficiently (Newhouse, 1970). Consequently, it is reasonable to test whether the observable operating characteristics of for-profit and non-profit hospitals are identical, or whether non-profits operate at higher cost (lower efficiency) and have worse patient outcomes than for-profit firms.

Public versus private

Similar considerations exist in the comparison of public (i.e., government) and private enterprise. Like a private non-profit, a government enterprise has no inherent mechanism to induce managers to act solely in the best interest of the enterprise. Furthermore, the income-maximizing arrangement potentially found in a non-profit entity (e.g., a physicians' cooperative) is legally precluded in most government enterprises. Consequently, the "manager" (bureaucrat) running a government enterprise does not trade off profit for nonpecuniary income; rather, he maximizes his budget, which enables the acquisition of greater pecuniary (salary) and nonpecuniary income (Niskanen, 1971). Therefore, the level of output of the public enterprise is higher than would be found in an otherwise equal private enterprise, and the input combination used also differs from that employed in a comparable private firm. Both disparities reflect inefficiency.

A later elaboration of this model of bureaucratic behaviour was proposed by Lindsay (1976). Lindsay criticized the budget-maximizing model as too simple, in that "it cannot be the explicit intent of Congress to reward managers for simply spending money. Congressmen are interested presumably in pleasing their constituents, and this involves getting as much bureau output as they can for as little as they can" (Lindsay, 1976). Naturally, however, the only output which can be monitored by a legislative body is that which is measurable. In contrast, even unmeasurable output can be effectively monitored by a private firm, argues Lindsay, because its customers can observe such output and react in ways measurable by the firm's managers.

Thus, parliamentary pressure exists for efficiency in public enterprises, but only with regard to output measurable by politicians. This motivates the manager-bureaucrat to "divert resources from the production of attributes which will not be monitored to those which will" (Lindsay, 1976). Thus, in a public hospital setting, a disproportionately large amount of (measurable) expensive computer equipment might be supplied along with a disproportionately small amount of (unmeasurable) politeness or clean floors.

This bias towards measurable attributes creates a corresponding bias in cost comparisons between private and public firms. Because unmeasurable attributes cannot be monitored effectively in a government enterprise but can be so monitored in a private enterprise, legislatures will refuse to fund the provision of such unmeasurable attributes, while private firms will fund them. As a result, the average cost of a unit of output produced by a government enterprise will be biased downward relative to the average cost of the same type of output produced by a private firm; the private firm bears costs per unit which the government firm (inefficiently) avoids (Lindsay, 1976). Thus, inefficiency in government enterprise can exist even when the government provider's average costs are lower than a private firm's.

Effect of competition

In most markets, it is an unquestionable prediction of economics that increased competition produces socially-beneficial outcomes - lower prices, greater access, enhanced quality. In health care markets, however, it has been correctly concluded that, even as a matter of theory, intensified competition is not unambiguously beneficial. Thus, it is worth considering whether one aspect of the Alberta plan - its enhancement of competition - should be expected to benefit society.

There are at least three reasons why competition among hospitals might not be efficiency-enhancing: insurance effects on patient demand, price regulations, and information costs (Kessler and McClellan, 1999; Frech, 1996). Because of insurance, consumers only pay a fraction of out-of-pocket costs; in Canada, of course, that fraction is zero. Consequently, consumers less often reward those firms with lower costs (and thus lower prices) by shifting demand to those firms. Thus, the rewards to firms from lowering costs, and entering markets where high costs prevail, are attenuated. Price regulations, as exist in Canada, may further constrain the potential beneficial effects of competition. Indeed, economists have predicted that where price regulations exist, a possible outcome is a "medical arms race" in which competition occurs on the basis of escalating quality, perhaps to the point where hospitals provide higher quality than is efficient (Robinson and Luft, 1985). Finally, increased competition may reduce efficiency if it leads to higher information costs to consumers. This phenomenon occurs if an increase in the number of providers makes it harder for consumers to compare one provider with another (Stiglitz, 1987).

In each case, however, these undesirable effects of competition are only possibilities and not logical necessities. Indeed, McClellan (1994) has demonstrated that even with regulated prices and insurance, hospital competition can theoretically lead to the best outcome from society's point-of-view. Furthermore, the prediction that increased competition can increase information costs depends crucially on particular assumptions regarding the cost of information (see Stiglitz, 1987). Therefore, while some models used to analyze the effect of competition in the hospital industry offer pessimistic predictions, many others support the standard analysis of competition as it is applied to markets in general: competition leads to lower prices, greater access, and higher quality.

[Previous] [Contents] [Next]

info@fraserinstitute.ca You can contact us at the above email address for any comments or information requests. Please report any dead links or technical problems.

If you know someone who would be interested in this web page, please enter their email address below, and we will forward this URL to them: Email Address:

Last Modified: Thursday, August 5, 1999.