From the Executive Director 

Free Market Reforms in Latin America: Let’s Pause for a Second

Latin American living standards were supposed to be catching up with the developed world’s by now. That, at least, was the promise behind the ambitious economic reforms many countries in the region introduced over the last two decades—more or less consistently with the so-called Washington consensus.[1]

The failure to fulfill that promise has triggered a wave of dissatisfaction with market reforms, beginning with Venezuela in 1999 and continuing with Argentina in 2001–02. Since then several countries in the region have elected governments—or are reportedly about to—that have vowed to undo the Washington-consensus “neoliberal policies” of the last 20 years. The neoliberal label often used in Latin America to deride free market reforms seems to mistakenly identify them with granting monopoly rights to vested interests, when the intention is precisely the opposite. Free markets are supposed to liberate societies not only from government monopolies but also from their private-sector counterparts.

Whatever its underlying motivations, the reaction against free market reforms cannot be easily dismissed as idiosyncratic to Latin America. India’s voters have just rejected the BJP (Bharatiya Janata Party)—which had implemented ambitious market reforms—and restored to power the Congress Party, whose anti-liberalization electoral campaign slogans captivated the four-fifths of the populace still living in what some analysts call the “bullock cart economy.”

The geographical breadth of a backlash that encompasses Argentina, Venezuela, India, and places in between suggests that something has gone wrong with market reforms. And that is where economists could step in with some analytics if they only paused for a second—a “second theorem,” that is. Indeed, economists have produced two important theorems that suggest logical motives behind what perhaps are also ideologically driven criticisms of market reforms: the Theorem of the Second Best and the Second Welfare Theorem.

The Theorem of the Second Best states that reforms must be comprehensive or an economy could wind up worse than before. Well-meaning policymakers have, for example, implemented free trade reforms without corresponding labor market reforms. If rigid labor legislation impedes the reallocation of workers from the inefficient industries swept away by free trade to those it bolsters, more jobs could be destroyed than created.

The Second Welfare Theorem says that free market reforms improve everyone’s standard of living provided the losers are compensated with lump-sum transfers from the winners. The unpopularity of free market reforms in many countries suggests that even when society is better off overall, large fractions of the population have been hurt by or left out of the benefits.

There is no doubt that the backlash against market reforms and globalization results in part from serious misunderstandings exploited by the eternally discontented. But it may also be an indication that the important caveats of the second theorems have been neglected in the drive to implement those reforms.

Empirically speaking, much remains to be discovered about which liberalizations are crucial. Even in industrial nations it is not clear which labor regulations and rigidities are destructively binding and which are not really deal breakers for an economy. If policymakers have to pick their fights, they had better learn which ones to pick.

But making such judgments will not be easy in the current state of the economics profession, whose reform recommendations have come largely from models that assume the paradigmatic “representative household”—that all households and economic agents are the same. This analytical shortcut has proved valuable to much research, but is inadequate for identifying the potential losers and winners of alternative liberalization programs. That kind of identification requires a new generation of quantitatively implementable models that can account for differences among economic individuals—a technical problem several orders of magnitude beyond what we have solved so far.[2]

The increased power of computers now lets us address some of the computational difficulties posed by heterogeneous agent models, but we still must bring those models to the data. For example, identifying the winners and losers of a particular trade liberalization would require tracking displaced workers to determine whether they are absorbed into the dynamic sectors and under what conditions. This chore requires data at a level of detail that is not easy to come by even in countries with the best statistics.

Once these hurdles are jumped, the job of implementing the Second Welfare Theorem’s transfers from winners to losers remains—and the revenues required must come from nondistortionary lump-sum or poll taxes. Unfortunately, poll taxes are politically problematic. (Recall the unrest triggered by UK Prime Minister Margaret Thatcher’s attempt to implement a poll tax in the 1980s.) Only distortionary fiscal instruments are typically available, and basing transfers on them might well undo the benefits of the reforms. In any case, Latin American countries have nothing like the U.S. programs that pay for workers displaced by international competition to learn new skills.

The challenges ahead have been eloquently summarized by Manmohan Singh, recently designated prime minister of India by the victorious until-not-long-ago opposition Congress Party: “Nobody today is against reform. The question is, how do you package reform so it is not seen as merely an elitist exercise?”

The second theorems may hold the answer to that question—if the profession pauses long enough to think about what they mean for the success or failure of free market reforms. To understand what has gone wrong with previous market reforms and what can go wrong with future ones—and to design corrective and preemptive measures—it may take decades of protracted and frustrating effort at the very frontier of research.

Encouragingly, conference papers and journal articles are paying more attention to these issues—with their important consequences for the well-being of people across the globe.[3] We hope that the Center for Latin American Economics—with its focus on long-term policy-linked research and on promoting dialogue between scholars and policymakers—will prove to be a significant part of that effort.

Notes This expression was first coined by John Williamson in his account of a conference on the topic of market reforms organized by the Institute for International Economics in Washington, D.C., in 1990. Nobel Laureate Robert Lucas Jr. emphasized the importance of heterogeneous agents models for the study of business cycles as well in his 2004 Presidential address to the American Economic Association. Evidence of these recent efforts can be found in the 2002 IMF conference on Macroeconomic Policies and Poverty Reduction. For example, the paper “Evaluation of financial liberalization: a general equilibrium model with constrained occupation choice,” by Xavier Giné and Robert M. Townsend, presented there explicitly identified winners and losers of the financial liberalization program implemented in Thailand in the period 1976–96. (This and related papers in that conference have been published in the August 2004 issue of the Journal of Development Economics.)

  |