The Benefits of Privatization: Evidence From Mexico
Why do some countries cling to state-owned industries, despite compelling evidence that privatization works? The answer often lies in political leaders' fears that the higher profitability of private companies comes at the expense of the rest of society, especially during the difficult transition period.
In The Benefits of Privatization: Evidence From Mexico (NBER Working Paper No. 6215), Rafael La Porta and Florencio Lopez-de-Silanes investigate whether companies become more profitable following privatization, whether privatization leads to significant social losses, and if so by which channels. They conclude that the positive changes in performance of privatized firms are the result of significant restructuring efforts and not of exploitation of market power, or of massive layoffs and lower wages. In other words, firms undergo a harsh restructuring process following privatization and do not simply markup prices and lower wages, as many economists predicted. Deregulation, particularly the removal of price/quantity controls and trade barriers, is associated with faster convergence to industry benchmarks. The authors suggest that the additional revenues the government receives as a result of the privatization auctions, and the increased tax revenues, are probably enough to offset the cost to society of job losses.
Mexico provides the stage for one of the most massive privatization programs in the world. Using data from all 218 nonfinancial privatizations which occurred in Mexico from 1983 to 1991, La Porta and Lopez-de-Silanes document the effects of privatization in seven general areas: profitability, operating efficiency, employment and wages, capital investment, taxes, output, and prices. By including both privately held and public companies, spanning a wide variety of sectors, and using detailed wage data and product-level quantities and prices, they can test competing theories on the effects of privatization.
The authors show that newly privatized firms come up to industry performance standards in the first few years. For example, the companies averaged a 24 percent gain in their operating-income-to-sales ratios, which is one of the four measures of profitability the authors use, each of which were positive. Large increases in operating efficiency accompanied the gains in profitability. As an example, the average cost-per-unit plummeted 21.5 percent, while the average sales-per-employee nearly doubled.
As one would expect, newly privatized firms cut employment, with the rolls of white and blue collar workers nearly halved. These numbers may actually underestimate the effects of privatization, since in the years preceding privatization most companies already had trimmed payrolls to prepare for divestiture. These findings suggest that transfers from workers to shareholders play a role in the success of privatization. However, productivity gains resulted in large increases in real wages in the post-privatization period: real wages increased 114 percent.
La Porta and Lopez-de-Silanes also show that privatized firms increased sales 54.3 percent, despite workforce reductions and only modest increases in capital. Surprisingly, prices rose only 2.9 percent relative to the Producer Price Index.
The authors find that monopoly power does not play an important role in increased profitability. Nor do firms in the noncompetitive sector increase prices in real terms any more than those in the competitive sector. In fact, in virtually every observed category, firms in competitive and non-competitive sectors acted similarly.
Finally, the authors decompose the reported increases in profitability. Approximately 10 percent of that gain was attributable to higher prices and 33 percent to worker layoffs, while productivity gains accounted for the remaining 57 percent. Some of the social effects of higher prices and layoffs were offset by corporate taxes, which absorbed slightly more than half of the gains in operating income.
The Digest is not copyrighted and may be reproduced freely with appropriate attribution of source.