Joseph E. Stiglitz

Competition is limited by the scale of the market, and as the scale of the market has changed, so has the effectiveness of competition.

In part, free trade has not worked because we have not tried it: trade agreements of the past have been neither free nor fair. They have been asymmetric, opening up markets in the developing countries to goods from the advanced industrial countries without full reciprocation.

Privatization is, at best, only a partial solution. Privatizations in which vouchers are given, and there is not a recapitalization, do not address this problem at all. Privatizations in which the firm is sold are likely, in the presence of limited competition for the firm, to provide an underestimate of the true value of the firm's assets. Accordingly good performance, based on this undervaluation of the firm's assets, does not provide a true measure of the firm's efficiency.

An essential commitment, which virtually all observers have emphasized, is not to subsidize enterprises that are making losses. While the standard remedy for this is "privatization," it should be recognized that this is neither necessary nor sufficient for imposing hard budget constraints. Governments in many countries have subsidized private producers (e.g., of steel), and governments in some countries have imposed hard budget constraints on government enterprises.

The theory of market socialism, for the most part, was not based on an analysis of these market failures, and of the reasons why government might be able to resolve them, but rather on the naive comparison of the actual performance of market economies and the hypothesized performance of a market socialist economy with an idealized view of government. This idealization not only failed to take into account the political realities, but more important from the perspective of this chapter, failed to take into account essential economic realities.

We need to distinguish between incentives at the organizational level and at the individual level, though some of the problems in designing incentive structures are common to both. We need to ask how, for the kinds of activities the public sector engages in, the organization is rewarded for "success" or punished for lack of it, and how individuals are similarly rewarded and punished.

As I argued earlier, I view competition as far more important than privatization.

The global financial system is not working well, and it is especially not working well for developing countries. Money is flowing uphill, from the poor to the rich. [...] With nearly two-thirds of reserves being held in dollars, the United States is, in this sense, the major recipient of these benefits. If the interest rate America has to pay is just one percentage point lower than it otherwise would be on these $3 trillion of loans from poor countries, what America received from the developing countries via the global reserve system is more than it gives to the developing countries in aid.

In short, whether for the obvious reason that in the absence of futures markets the price system cannot perform its essential coordinating role with respect to future-oriented activities, such as investments, or for the more subtle reasons just discussed, that in the absence of futures markets, extending infinitely far into the future, the market economy is likely to exhibit dynamic instabilities there is no reason to believe that even with rational expectations it will converge to the steady state; there is no presumption that markets, left to themselves, will be efficient. For advocates of market socialism, the implication of this analysis seems clear: There is a need for the kind of government control of the allocation of investment envisaged in market socialism.

The natural resource curse is not fate; it is choice. The exploitation of natural resources is an important part of globalization today, and in some ways the failures of the resource-rich developing countries are emblematic of globalization's failures.

At the core of the failure of the socialist experiment is not just the lack of property rights. Equally important were the problems arising from lack of incentives and competition, not only in the sphere of economics but also in politics. Even more important perhaps were problems of information. Hayek was right, of course, in emphasizing that the information problems facing a central planner were overwhelming. I am not sure that Hayek fully appreciated the range of information problems. If they were limited to the kinds of information problems that are at the center of the Arrow-Debreu model consumers conveying their preferences to firms, and scarcity values being communicated both to firms and consumers then market socialism would have worked. Lange would have been correct that by using prices, the socialist economy could "solve" the information problem just as well as the market could. But problems of information are broader.

Informational constraints not only limit the ability of shareholders to control rent-seeking behavior on the part of top managers, they also limit the ability of top managers to control rent-seeking behavior on the part of their subordinates. How much of the time spent by a middle-level manager to prepare a report was absolutely necessary? To what extent was it devoted to acquiring information, of marginal value to the firm, but which would make that manager look relatively good compared to other managers? To what extent are the efforts and resources spent by a manager to cultivate a client really being directed to enhance that manager's job opportunities? Private and organizational objectives are intricately intertwined, and in many cases they are not conflicting. But at the margin they frequently are, and there seems little reason to doubt that private objectives frequently, perhaps usually, win out.