Although the theory of property rights would suggest that differing structures of rights will influence economic performance, the size of any effects is still open to debate. Further, comparisons require careful interpretation because there are many differing criteria of ‘performance’ and the relative performance may vary according to the market and regulatory environment in which particular firms have to operate.
Economic performance may be judged by a number of criteria:
The technical efficiency of operations (achieving the greatest output from given inputs in a purely engineering sense).
The overall productive efficiency of operations (achieving the lowest cost for a given output at the prevailing factor prices).
The allocativeefficiency of the enterprise (ensuring that the prices to consumers reflect the social costs of production and that enterprise output and quality maximise net social benefits (see Section 8.4.1).
The choice of institutional structure will be complex if there is not a single form which dominates all others in every category of performance. An unregulated private joint-stock enterprise, for example, may, for the reasons outlined in Section 8.5, show a higher standard of performance in categories (i) and (ii) but a lower standard in category (iii) than a public enterprise. A regulated joint-stock enterprise on the other hand may achieve a better performance in category (iii) relative to the public enterprise depending upon the type of regulatory regime in force (see Module 3 and review question 3.4). This better performance in category (iii) may be at the price of a rather poorer performance in category (ii) if the regulatory regime results in distortions to factor inputs (as already discussed in Section 3.5.1). On the other hand a tighter monitoring environment in the private sector might produce a better performance with respect to category (i). The situation is complicated still further when product market conditions are considered. A public enterprise operating in a market subject to competition or the threat of new entry would behave differently from one operating in a protected natural monopoly situation. However, the same observation could be applied to most other institutional forms. In addition, it is not unlikely that a public enterprise facing competition from other public enterprises will behave differently from one facing competition from joint-stock enterprises. This raises the possibility that the ‘performance’ of a particular type of enterprise may depend upon the existence of a different type of enterprise with which it competes. Measures of comparative performance which do not allow for interdependence then become very misleading. It is the combination of institutions as much as the structure of institutions in isolation which determines overall performance.
With the above reservations in mind we can now consider a series of examples of studies which have attempted to investigate the impact of property rights on performance. In the example that follows two regulated firms are compared, one private and the other public.
An Early Study of Airlines in Australia
An attempt to compare firms with different property rights arrangements but in otherwise similar circumstances is represented by Davies' (1971, 1977) celebrated study of two Australian airlines. Two firms, one private (Ansett Australian National Airways (ANA)) and one public (Trans Australian Airlines (TAA)) operated internal flights in Australia. Davies noted that during the period of his study the private firm was a ‘creature of the state’ in that the industry was protected from new entrants and was highly regulated. The Federal Government controlled routes, ports of call and frequency of stops. Further the government ‘required that not only must the two firms provide the same amount of capacity and the same number of aircraft, but for all practical purposes government policy required that each enterprise acquire the same kind of aircraft’ (1971, p. 156). So similar were the two airlines that Harley (1974, quoted in Davies 1977) dubbed them ‘Tweedledum and Tweedledee Airlines’ and commented that ‘their symbiotic relationship is so pervasive it even reaches into their separate restaurants in Tullamarine Airport with each having the same number of tables and chairs’. Flights were often scheduled to take-off and land almost simultaneously. Various performance measures were constructed. As measures of productivity for the years 1958–9 to 1973–4, Davies used tons of freight per employee, passengers carried per employee and revenue earned per employee. The average figures are reproduced in Table 8.4.
| Tons of freight | |||
| Company | and mail | Passengers | Revenue |
| TAA | 5.14 | 326 | 10 740 |
| Ansett ANA | 10.25 | 373 | 12 009 |
In spite of the similarity of the two companies, Davies concludes that ‘the private firm is economically more efficient than the public firm’ (p. 165). The chosen indices of productivity all favour the private firm. Later, however, Davies' results were questioned by Forsyth and Hocking (1980). If output is redefined as ton kilometres or passenger kilometres, they argue, the productivity advantage of Ansett ANA disappears. Apparently the private firm, although competing with TAA on longer-haul flights, operated more extensively than TAA on the shorter routes and this gave it an advantage when output is expressed simply as passenger numbers or freight tons. On the other hand of course, administering shorter flights may require more employees which would be a disadvantage. This example illustrates well the general problem of productivity comparisons. Even in the case of Tweedledum and Tweedledee airlines it is possible to argue about the appropriate measurement of output.
In contrast to the Australian airlines case a similar study of public and private enterprise from Canada arrives at different conclusions albeit in a subtly different regulatory environment.
A Study of Canadian Railroads
Caves and Christensen (1980) consider the Canadian National railroad (CN) a public company, and the Canadian Pacific railroad (CP) a private company. These are large railroads of approximately equal size which compete directly over many routes. Competition from road and air has grown since the World War II, and although heavily regulated for many years, most rate regulation had gone by 1967. Restrictions remain only on the abandonment of track and on rates for the hauling of grain and flour. Caves and Christensen calculate a measure of total factor productivity for the two railroads and compare their growth over time and their level between firms. The study uses freight ton miles and passenger miles as measures of output. Inputs are divided into labour, structures, equipment, fuel and materials. The study concludes that CN had a lower level of productivity in the early 1960s but that by 1967 it had caught up with CP and that productivity growth was similar in the two firms thereafter. From this evidence Caves and Christensen conclude that ‘public ownership is not inherently less efficient than private ownership’ (p. 974). They speculate, however, that the existence of competition might help to explain the performance of CN, although clearly their study is not capable of disentangling the impact of property rights from competition.
The possibility that publicly owned firms will operate at a higher level of productive efficiency when subject to product market competition is obviously important. The next example considers the evidence from the electricity utilities in the United States.
The Influence of Competition on Performance
Primeaux (1977) compares a cross section of municipally owned electric utilities in the United States which sell in monopoly markets with another cross section of municipally owned firms which face competition. Where competition existed, the municipally owned firm usually competed with a private utility. The nature of this competition would vary depending on location. In some places a customer could switch supplier at will, in others only new customers had an effective choice. Competing and non-competing municipally owned utilities were then matched together. A non-competing firm had to be in the same US state as the firm it was paired with, be of approximately equal size and have an identical power source. Primeaux found that competition reduced average cost and that this competitive effect was more pronounced for small firms than for large.
The electricity industry in the United States has been one of the most intensively studied by economists interested in the effects of property rights. One result has been to highlight the significance of the regulatory environment faced by private companies in determining their productive efficiency relative to public enterprises.
Notwithstanding the dissatisfaction expressed with the performance of the nationalised industries, the literature on the relative efficiency of public and private enterprise was inconclusive in the 1970s and 1980s. Millward and Parker (1983), for example, took the view that ‘there is no systematic evidence that public enterprises are less cost-effective than private firms’ (p.258). In contrast De Alessi (1974) and Borcherding (1977) interpreted the econometric evidence as showing a generally higher productive efficiency in private firms. Picot and Kaulmann (1989) used a sample of firms taken from Fortune's ‘The Foreign 500’ to compare the performance of private and government owned corporations from six western countries in the years 1975–84. Their ‘performance variables’ included productivity indicators such as sales per employee as well as measures of profitability. They confined their study to firms operating in unregulated markets and exposed to national and international competition. Their results confirmed ‘the differences in performance between government-owned and privately-owned large industrial corporations that property rights theory would predict’ (p.312).
In the 1990s the privatization programmes pursued across the world have given rise to more opportunities for statistical work. D'Souza and Megginson (1999) use data ftom 85 companies privatized in the early 1990s in 28 countries and compare performance for the three years before and the three years after privatization. They find significant increases in profitability and operating efficiency, increases that are not explained by the exploitation of monopoly power. Similarly, in a review of the empirical evidence Shirley and Walsh (2000) argue that the literature ‘strongly favors private ownership in competitive markets over the state-owned counterfactual’.
