Public Policy Sources #35:
Interpreting the economic evidence in the Canadian context

[Previous] [Contents] [Next]

Before applying the evidence on the economic consequences of allowing private hospital competition, it is useful to summarize its conclusions. In general, the literature indicates that there are no distinct efficiency advantages or disadvantages to for-profit vis-à-vis non-profit hospitals, that there are distinct efficiency advantages to for-profit vis-à-vis government hospitals, and that enhanced hospital competition has been of clear social benefit in the last 10 years in the US, although only potentially beneficial on net prior to that. Because these findings are US-specific, however, it is crucial to assess their relevance and application to the Canadian situation in general, and the Alberta plan in particular. This assessment yields several policy conclusions.

For-profits are efficient

Undoubtedly the most germane information is the substantial empirical support for the proposition that for-profit hospitals are lower-cost than government hospitals. It provides prima facie evidence that Alberta and other provinces could save money by contracting out to private facilities. This lesson has not been lost on other budget-conscious but service-oriented governments. A recent report provided prominent examples in the US; for instance, Orange County in California has discontinued ownership and operation of hospitals, opting instead to contract with individual hospitals (Tradewell, 1998). Furthermore, these innovations extend beyond the developed world to Africa. Mills, Hongoro, and Broomberg (1997) describe recent successes in hospital privatization in South Africa and Zimbabwe, where quality was maintained but average cost fell. Privatization also indirectly lowers costs even when it is not implemented; McDavid reports the cost-reducing effect which the threat of privatization had with regard to municipal service provision in Vancouver (McDavid, 1988).

Critics of the Alberta plan might, nevertheless, raise the fact that the empirical research comparing non-profits and for-profits does not in general discern efficiency differences. Perhaps, the argument might go, this research is the more relevant, with current Canadian public hospitals the operational equivalent of the private non-profits in those US studies. This argument, however, is lacking in foundation. As noted earlier, some economists have explained the lack of performance difference between for-profits and non-profits as arising from the effective control of non-profits by their physicians (e.g., Pauly and Redisch, 1973). This theory, while plausible in the US context, is not sensibly applied to the Canadian setting.

This theory of the physicians' 'cooperative' depends on the ability of doctors to receive meaningful rewards from operating hospitals more efficiently, an ability that is severely attenuated if not nonexistent in Canada. One practical limitation to this motive is the uniform fee schedules applied in each province, whereby individual physician reimbursement does not vary according to the physician's characteristics (unless the physician opts out of the public system entirely). Consequently, the income from performing additional procedures is modest. Moreover, provincial ceilings on physician earnings further discourage the energetic practitioner, who finds her income reclaimed at an alarming rate (75 percent in some provinces) once the earnings cap is reached. More fundamentally, physicians are constrained by the availability of other factors, including operating facilities and nurses.

Apparent government efficiencies are probably illusory

But suppose, for the sake of argument, that the Alberta proposal is implemented, and private hospitals providing services are found to have average costs identical to or perhaps higher than those in existing government hospitals. Such evidence might still reflect the greater efficiency of private providers. This seemingly counterintuitive claim stems from the insight of Lindsay (1976), described earlier.

Lindsay proposed a theory of government enterprise in which government managers (i.e., bureaucrats) only produce those things that are visible, and thus measurable to their parliamentary funders. In the hospital context, this might be length of stay. In contrast, government managers skimp on those aspects of output (e.g., bedside manner) which are not measurable by bureaucrats. For-profit managers, however, do not have this luxury, as these less tangible aspects are effectively measurable by their customers, who can vote with their feet. The implication of this is that for-profit firms will typically efficiently provide these aspects while government firms will not. Consequently, total and average costs will appear higher for for-profits, relative to those costs in government firms, because government firms are not providing the same services. This tendency, then, will create inherent bias in any public-private comparison against the private for-profit appearing to have lower costs, even when it does so in actuality.

Thus, because private hospitals will only be paid the same fees for their services as public hospitals, Alberta residents will receive at least the same output or perhaps higher output without any increase in cost to the taxpayer.

Privatization would cut wage costs

Given the substantial benefits associated with privatizing hospital services, resistance to the Alberta plan might appear puzzling. This puzzle is at least partially resolved by a study of factors affecting the propensity of US city and county governments to contract out hospital services (Ferris and Graddy 1987). The study discovered that a significant deterrent to contracting out was the percentage of local public employees who were unionized. This opposition undoubtedly stems from the wage premiums which public union employees earn.

Across industries, this wage premium (earned by public union workers relative to their non-union counterparts) has often been estimated to fall in the 5 to 15 percent region (Ehrenberg and Schwarz, 1986). But Clarkson (1972) found that unionized government hospital employees earned, on average, 23 percent more than their unionized private-sector equivalents. Even more striking is the wage difference between public-union non-medical hospital workers (i.e., cooks, cleaners, painters, etc.) and corresponding private-union hotel workers in British Columbia (Ramsay, 1995). Ramsay found that the hospital workers in her sample earned 25 to 63 percent more than the equivalent unionized hotel workers. This evidence suggests that an additional benefit to the Alberta proposal is the wage savings it would provide. Moreover, the financial advantages might well be accompanied by productivity gains; according to Vitaliano and Toren (1996), "Unionized hospitals are less efficient" than nonunionized ones.

Competition will work particularly well in Canada

The empirical findings contained in the literature on hospital competition, while generally favourable, contain aspects which imply that competition will be particularly beneficial in the Canadian setting. Among the results in Kessler and McClellan (1999) is that the beneficial effects of competition - reduced costs and higher quality - are much larger for the markets in their sample which were the least competitive to begin with. Given the fundamental restrictions on competition currently present in the Canadian hospital market, this conclusion is a promising one for the success of the Alberta proposal.

In addition, one of the major concerns regarding hospital competition - the possibility of a "medical arms race" - is minimal in the Canadian context. The gist of the "arms race" argument is that if firms face regulated prices, limiting their ability to compete along that dimension, they will instead compete by ratcheting quality upwards. In some models, this escalation of quality is inefficient (Robinson and Luft, 1985). The possibility of excessive quality, however, is a modest one in Canada, a country which ranks among the bottom third of OECD countries in availability of technologies such as CT scanners, MRIs, and lithotripters (Harriman, McArthur, and Zelder, 1999).

[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.