Comments

Luis F. López-Calva: The paper by Chong and López-de-Silanes is extremely valuable for understanding economic reform and its effects in Latin America. Although establishing a true record is an ambitious task, the paper does review a series of robust results from careful research carried out by several authors using Latin American data. Specifically, the authors review exercises carried out for seven Latin American countries following a methodology first used by La Porta and López-de-Silanes with Mexican data.¹ They describe the main data limitations in each case and then establish some empirical regularities, which can be summarized as follows:

—As a general result, profitability and operating efficiency improved, while both output and investment grew in privatized firms in most countries;

—Employment trends after privatization are mixed, though the average tends to go in the direction of a reduction in employment;

—High bureaucratic quality and absence of corruption increase privatization prices;

—Evidence for Mexico shows that restructuring is not worth pursuing, for it has a negligible net effect on final privatization prices, whereas it pays to sell via a fast process;

—The net effect for the government is positive, mainly because money-losing firms are sold and turned into profitable enterprises that pay taxes, although a number of bailouts have had a high fiscal effect;

—Profitability gains do not come from reduction in wages, but from an increase in efficiency;

—Profitability is higher in less competitive sectors, but the evidence does not support the hypothesis of market power exploitation;

—Welfare gains are progressive, and quality improvements are also important, as in the case of water privatization in Argentina; and

—Corporate governance and regulatory laws are key complementary policies in the process.

The positive net effect of privatization described in the paper strongly disqualifies the general disapproval found in public opinion polls and mass media. The disparity between the position of the paper and common belief has two possible explanations: either economists have failed at communicating the positive effects of these reforms, or the reforms have effects that go beyond what is shown in the paper, in which case more research is needed to understand the political economy of privatization. I tend to support both explanations, with a strong emphasis on the second. The microeconomics of privatization are relatively well established, and the evidence is now strong in this respect, partly as a result of the research carried out by López-de-Silanes and other economists.² That, however, is only one piece among many others for fully understanding the political economy of the process. I now discuss some of the issues that need attention.

Do Higher Proceeds Indicate a Better Process?

Evaluating the scope of a privatization process based on the amount of money received by the government is a common mistake. This mistake can be seen in the paper in the description of the process and the strong emphasis on privatization prices. In principle, if proceeds were the variable to be maximized, an optimal policy would be to sell firms under a noncompetitive environment. This is, of course, contrary to the real objective: achievement of microeconomic efficiency. Perhaps the bias comes from the fact that privatization has also played an important role in terms of strengthening public finances by bringing resources to the treasury. However, this is only a side effect—albeit an important one—and never the main purpose.

The focus on prices is linked with the discussion on restructuring. Labor restructuring is often a necessary condition for the process to be feasible, even if it does not have a net effect on prices. Labor liabilities are an important deterrent for participation of bidders for financial and political reasons. Nobody would suggest that other kinds of restructuring, like new investments, should be made before the sale.

Looking for a True Record

An understanding of public opinion and reactions against privatization must take into account not only net gains and losses, but also the nature and distribution of the losses...