Going private

Is privatization the ticket to improving production?



In The People’s IPO, the government of Kazakhstan plans to sell off part of the ownership of such state enterprises as the airline Air Astana and the railway Temir Zholy. How could this affect their performance?

Radically. The incentives facing the private firm differ dramatically from those for the public enterprise.

The manager controls what the private firm does, although stockholders are the owners, since they cannot monitor him continuously – and lack reason to do so, given the small stake of the typical shareholder in a large corporation. The manager works less – and provides more nonmonetary benefits – than owners would wish. Consider that nine-hole golf course on the corporate grounds, which the manager is prone to investigate each weekday at 2pm.

The manager’s incentives for boosting profits might include a bonus as well as a growing reputation that attracts tempting job offers. Those are the carrots. The stick is in the New York Stock Exchange: If the manager does poorly, then the firm’s stock will lose value. A raider might buy up the shares, boot the manager, hire a new team to improve profits, and then sell the firm at a healthy price.

Those incentives don’t apply to public enterprises, as Cotton Lindsay pointed out in 1976. Private and public enterprises behave differently, because they are controlled by different groups.


Gargantuan groves of academe


Consider a private university. Ultimately, it is controlled by the customers – students and their parents. To satisfy them, the school will offer them the desired courses, taught in the desired style.

But a public university is controlled by legislators. They don’t take classes, so they judge the school instead by variables that they do observe: The number of courses, students and professors. Rather than improve the quality of education, the school will increase those observable factors. Prediction: The public university will be larger, and more diverse in offerings, than the private college.

To test his theory, Lindsay compared public hospitals in the United States, run by the Veterans Administration (VA), to private hospitals. The latter will improve care through such attributes as the doctor’s bedside manner, since this attracts patients. But in a VA hospital, bureaucrats and legislators do not observe patient care, so the hospital will increase instead observable characteristics such as the rate of survival. Since this requires only a minimal amount of care, the staff-to-patient ratio should be lower for a VA hospital than for a private hospital where additional staff improves unobservable care. Lindsay did find lower staff-to-patient ratios for VA hospitals.

Similarly, in Kazakhstan, Air Astana might accrete political support by offering sumptuous meals (at least in theory), albeit at a high cost and price.


The price isn’t right


Other economists regard the price itself as an observable factor of the public enterprise. In the early 1970s, Sam Peltzman proposed a model in which the enterprise sets the price below the profit-maximizing one, because voters dislike subsidies. It would not prefer a price that was higher than the profit-maximizing one, since it would lose both profits and votes. Privatizing the airline and railway in Kazakhstan might lead to higher ticket prices.

Also, because it wants political support, the public enterprise is less likely to price discriminate – to charge different prices to different groups of consumers -- than the private firm is. If privatized, Temir Zholy would become more likely to offer cheap tickets to youths – and expensive ones to businessmen.

Why won’t public enterprises discriminate? Suppose that as the government keeps raising the price of, say, airline services, the resulting loss of additional votes among passengers enlarges. The government might mitigate this loss by spreading the cost of services over non-passengers as well.

In general, the government will consider both taxes and demand for a good before providing it. Consider the demand for operas, which is largest among the rich. Since they pay the same tax rate on personal income as do the poor – roughly 10% -- a sharp reduction in the cost of an opera ticket could be politically costly. The rich will respond to the discount by demanding many more operas, increasing the total subsidy. But since the income tax rate is flat, the rich will pay only a proportional share of the tax revenues that finance the operas, although they receive a disproportional share of the benefits. The politically savvy government will cut the price of an opera ticket only slightly.

Peltzman’s main point: Since the public enterprise is political, financial markets cannot discipline it as easily as they can a private firm. When the firm is inefficient, its stock prices will fall, inviting a raid. But for a public enterprise, the “owners” are the voters, who are unlikely to leave the jurisdiction just because they dislike one state-owned enterprise.

In Lindsay’s model, the public enterprise may charge a higher price than private firms do, so that it can afford to offer observable quality. In Peltzman’s model, the public enterprise charges a lower price, in order to attract political support.

What are the facts? A 1970 study in the U.S. by Thomas Moore found that regulated electric utilities charged a price that was 5% or less below that of an unregulated monopoly. But electric utilities that were public enterprises charged prices 10% to 22% below the monopoly price.

In 1971, Peltzman found that among electric utilities, the number of rate schedules for public enterprises – a form of price discrimination -- was much smaller than the one for private firms. The difference exceeded six standard errors, which is a measure of dispersion. A difference of just two standard errors would have been quite unlikely to have occurred by chance; a difference of six almost certainly means that the public utilities differed inherently from private ones. These results raise the possibility that in Kazakhstan privatization may increase both profits and practices, like price discrimination, that many citizens regard as unfair. – Leon Taylor, tayloralmaty@gmail.com



Good reading

Cotton Lindsay. A theory of government enterprise. Journal of Political Economy. October 1976.

Thomas Gale Moore. The effectiveness of regulation of electric utility prices. Southern Economic Journal. April 1970.

Sam Peltzman, Pricing in public and private enterprises: Electric utilities in the United States. Journal of Law and Economics. April 1971.

W. Kip Viscusi, Joseph E. Harrington, Jr., and John M. Vernon. Economics of regulation and antitrust. Fourth edition. Cambridge, Mass.: The MIT Press.2005. Chapter 14 surveys work on private and public enterprises.