Publications

2006
Mullainathan, Sendhil, Joshua Schwartzstein, and Andrei Shleifer. 2006. “Coarse Thinking and Persuasion”. Abstract
We present a model of coarse thinking, in which individuals group situations into categories, and transfer the informational content of a given message from situations in a category where it is useful to those where it is not. The model explains how uninformative messages can be persuasive, particularly in low involvement situations, and how objectively informative messages can be dropped by the persuader without the audience assuming the worst. The model sheds light on product branding, the structure of product attributes, and several puzzling aspects of mutual fund advertising.
Mullainathan, Sendhil, Joshua Schwartzstein, and Andrei Shleifer. 2006. “Coarse Thinking and Persuasion”. Abstract
We present a model of coarse thinking, in which individuals group situations into categories, and transfer the informational content of a given message from situations in a category where it is useful to those where it is not. The model explains how uninformative messages can be persuasive, particularly in low involvement situations, and how objectively informative messages can be dropped by the persuader without the audience assuming the worst. The model sheds light on product branding, the structure of product attributes, and several puzzling aspects of mutual fund advertising.
Mullainathan, Sendhil, Joshua Schwartzstein, and Andrei Shleifer. 2006. “Coarse Thinking and Persuasion”. Abstract
We present a model of coarse thinking, in which individuals group situations into categories, and transfer the informational content of a given message from situations in a category where it is useful to those where it is not. The model explains how uninformative messages can be persuasive, particularly in low involvement situations, and how objectively informative messages can be dropped by the persuader without the audience assuming the worst. The model sheds light on product branding, the structure of product attributes, and several puzzling aspects of mutual fund advertising.
Kremer, Michael, and Eric Maskin. 2006. “Globalization and Inequality”. Abstract
Supporters of the anti-globalization movement argue that "globalization has dramatically increased inequality between and within nations" (Mazur, 2000), and in particular that it has marginalized the poor in developing countries and left behind the poorest countries. Meanwhile, more moderate mainstream politicians argue that the poor must invest in education to take advantage of globalization (Clinton, 2000). Such views are difficult to reconcile with a standard Heckscher-Ohlin trade model with two countries, two goods, and two factors (skilled and unskilled labor, or alternatively capital and labor). Under a simple model, globalization should benefit the poor in poor countries and reduce inequality in poor countries, and within the developing world the poorest countries and least educated workers should have the greatest opportunity to benefit from globalization.
Kremer, Michael, and Eric Maskin. 2006. “Globalization and Inequality”. Abstract
Supporters of the anti-globalization movement argue that "globalization has dramatically increased inequality between and within nations" (Mazur, 2000), and in particular that it has marginalized the poor in developing countries and left behind the poorest countries. Meanwhile, more moderate mainstream politicians argue that the poor must invest in education to take advantage of globalization (Clinton, 2000). Such views are difficult to reconcile with a standard Heckscher-Ohlin trade model with two countries, two goods, and two factors (skilled and unskilled labor, or alternatively capital and labor). Under a simple model, globalization should benefit the poor in poor countries and reduce inequality in poor countries, and within the developing world the poorest countries and least educated workers should have the greatest opportunity to benefit from globalization.
Kremer, Michael, and Stanley Watt. 2006. “The Globalization of Household Production”. Abstract
Restrictions on migration of low-skilled workers to richer countries are arguably the largest distortion in the world economy and the most costly to the world’s poor. Yet rich countries seem unlikely to eliminate these restrictions due to concerns about the impact of migration on inequality among natives, public finances, and native culture. A rapidly growing new type of migration may not be subject to these concerns. Many "new rich" countries issue special visas for foreigners, women in particular, to work as private household workers. "Old rich" countries often choose low levels of enforcement against illegal immigrants working in this sector. We argue that by allowing high-skilled native women to increase market labor supply, this type of immigration increases the wages of low-skilled natives and provides a fiscal benefit by correcting tax distortions toward home production. Calibration suggests welfare gains to natives from a program, such as Hong Kong’s or Singapore’s, under which roughly 7% of the labor force are foreign private household workers, may increase the ratio of native low-skilled workers by 3.9% and increase native welfare by 1.2% of income, roughly 100 times the level estimated by Borjas and increases the relative wages of native low-skilled to high-skilled by 3.9%. Paradoxically, however, even if these programs are pareto improving, they may conflict with ethical norms requiring stronger social obligations to long-term residents than to other foreigners. Short-term programs may be more acceptable.
Kremer, Michael, and Stanley Watt. 2006. “The Globalization of Household Production”. Abstract
Restrictions on migration of low-skilled workers to richer countries are arguably the largest distortion in the world economy and the most costly to the world’s poor. Yet rich countries seem unlikely to eliminate these restrictions due to concerns about the impact of migration on inequality among natives, public finances, and native culture. A rapidly growing new type of migration may not be subject to these concerns. Many "new rich" countries issue special visas for foreigners, women in particular, to work as private household workers. "Old rich" countries often choose low levels of enforcement against illegal immigrants working in this sector. We argue that by allowing high-skilled native women to increase market labor supply, this type of immigration increases the wages of low-skilled natives and provides a fiscal benefit by correcting tax distortions toward home production. Calibration suggests welfare gains to natives from a program, such as Hong Kong’s or Singapore’s, under which roughly 7% of the labor force are foreign private household workers, may increase the ratio of native low-skilled workers by 3.9% and increase native welfare by 1.2% of income, roughly 100 times the level estimated by Borjas and increases the relative wages of native low-skilled to high-skilled by 3.9%. Paradoxically, however, even if these programs are pareto improving, they may conflict with ethical norms requiring stronger social obligations to long-term residents than to other foreigners. Short-term programs may be more acceptable.
Mullainathan, Sendhil, Marianne Bertrand, and Simeon Djankov. 2006. “Obtaining a Driving License in India: An Experimental Approach to Studying Corruption”. Abstract
We conduct two experiments to understand the process of obtaining a driver’s license in India. In the first experiment, we randomly assign license candidates to one of three groups: bonus (offered a financial reward if they could obtain their license fast), lesson (offered free driving lessons upfront), and control. The control group alone illustrates bureaucratic failures: 71% of the license getters in that group avoided the mandated driving test and 62% failed a surprise driving test. The system responds to private needs—there are more license getters in the bonus group—but at a social cost: there are more license getters who cannot drive in the bonus group. The system however also appears to respond to social considerations, as there are more license getters in the lesson group. Large extra-legal payments are made by license getters: those in the control group pay 2.5 times the official fee. More of these extra-legal payments take place in the bonus group. Surprisingly, these extra-legal payments are not direct bribes to bureaucrats but instead payments to agents. In the second experiment, we perform an audit study to better understand the role of agents. The audit shows that agents can provide licenses to individuals even if they cannot drive; but the audit also shows that agents cannot as easily circumvent all other rules. We argue that our findings are most consistent with agents being the channel for corruption in this system. We also report some suggestive evidence that bureaucrats create red tape, possibly to induce more license candidates to use agents.
Mullainathan, Sendhil, Marianne Bertrand, and Simeon Djankov. 2006. “Obtaining a Driving License in India: An Experimental Approach to Studying Corruption”. Abstract
We conduct two experiments to understand the process of obtaining a driver’s license in India. In the first experiment, we randomly assign license candidates to one of three groups: bonus (offered a financial reward if they could obtain their license fast), lesson (offered free driving lessons upfront), and control. The control group alone illustrates bureaucratic failures: 71% of the license getters in that group avoided the mandated driving test and 62% failed a surprise driving test. The system responds to private needs—there are more license getters in the bonus group—but at a social cost: there are more license getters who cannot drive in the bonus group. The system however also appears to respond to social considerations, as there are more license getters in the lesson group. Large extra-legal payments are made by license getters: those in the control group pay 2.5 times the official fee. More of these extra-legal payments take place in the bonus group. Surprisingly, these extra-legal payments are not direct bribes to bureaucrats but instead payments to agents. In the second experiment, we perform an audit study to better understand the role of agents. The audit shows that agents can provide licenses to individuals even if they cannot drive; but the audit also shows that agents cannot as easily circumvent all other rules. We argue that our findings are most consistent with agents being the channel for corruption in this system. We also report some suggestive evidence that bureaucrats create red tape, possibly to induce more license candidates to use agents.
Mullainathan, Sendhil, Marianne Bertrand, and Simeon Djankov. 2006. “Obtaining a Driving License in India: An Experimental Approach to Studying Corruption”. Abstract
We conduct two experiments to understand the process of obtaining a driver’s license in India. In the first experiment, we randomly assign license candidates to one of three groups: bonus (offered a financial reward if they could obtain their license fast), lesson (offered free driving lessons upfront), and control. The control group alone illustrates bureaucratic failures: 71% of the license getters in that group avoided the mandated driving test and 62% failed a surprise driving test. The system responds to private needs—there are more license getters in the bonus group—but at a social cost: there are more license getters who cannot drive in the bonus group. The system however also appears to respond to social considerations, as there are more license getters in the lesson group. Large extra-legal payments are made by license getters: those in the control group pay 2.5 times the official fee. More of these extra-legal payments take place in the bonus group. Surprisingly, these extra-legal payments are not direct bribes to bureaucrats but instead payments to agents. In the second experiment, we perform an audit study to better understand the role of agents. The audit shows that agents can provide licenses to individuals even if they cannot drive; but the audit also shows that agents cannot as easily circumvent all other rules. We argue that our findings are most consistent with agents being the channel for corruption in this system. We also report some suggestive evidence that bureaucrats create red tape, possibly to induce more license candidates to use agents.
Kremer, Michael, Jessica Leino, Edward Miguel, and Alix Peterson Zwane. 2006. “Spring Cleaning: A Randomized Evaluation of Source Water Improvement”. Abstract
Water-related diseases, particularly diarrhea in young children, kill two million people annually. To address this problem, donors and governments often provide infrastructure such as communal standpipes, wells, and protected springs in rural areas, where piping water into homes is infeasible. We study the impact of source water quality improvements achieved via spring protection in rural Kenya using a randomized evaluation. Spring protection leads to large improvements in source water quality as measured by the fecal indicator bacteria E. coli. Water quality gains at the home are smaller on average, but this finding depends critically on households’ water source choices. At households that only used the sample spring at baseline, 71% of the spring water quality benefits are translated into home water gains, suggesting that recontamination in transport and storage may be less of a concern than is sometimes claimed. Consistent with this view, the home water quality gains from spring protection are no larger for households with better baseline sanitation or hygiene knowledge. Changes in household water source choices after spring protection are used to derive revealed preference estimates of the willingness to pay for improved water quality using a travel cost approach. The average willingness to pay for the moderate gains in home water quality due to spring protection is at least US$3.27 per household per year. We find no significant child health effects of spring protection.
Kremer, Michael, Jessica Leino, Edward Miguel, and Alix Peterson Zwane. 2006. “Spring Cleaning: A Randomized Evaluation of Source Water Improvement”. Abstract
Water-related diseases, particularly diarrhea in young children, kill two million people annually. To address this problem, donors and governments often provide infrastructure such as communal standpipes, wells, and protected springs in rural areas, where piping water into homes is infeasible. We study the impact of source water quality improvements achieved via spring protection in rural Kenya using a randomized evaluation. Spring protection leads to large improvements in source water quality as measured by the fecal indicator bacteria E. coli. Water quality gains at the home are smaller on average, but this finding depends critically on households’ water source choices. At households that only used the sample spring at baseline, 71% of the spring water quality benefits are translated into home water gains, suggesting that recontamination in transport and storage may be less of a concern than is sometimes claimed. Consistent with this view, the home water quality gains from spring protection are no larger for households with better baseline sanitation or hygiene knowledge. Changes in household water source choices after spring protection are used to derive revealed preference estimates of the willingness to pay for improved water quality using a travel cost approach. The average willingness to pay for the moderate gains in home water quality due to spring protection is at least US$3.27 per household per year. We find no significant child health effects of spring protection.
Kremer, Michael, Jessica Leino, Edward Miguel, and Alix Peterson Zwane. 2006. “Spring Cleaning: A Randomized Evaluation of Source Water Improvement”. Abstract
Water-related diseases, particularly diarrhea in young children, kill two million people annually. To address this problem, donors and governments often provide infrastructure such as communal standpipes, wells, and protected springs in rural areas, where piping water into homes is infeasible. We study the impact of source water quality improvements achieved via spring protection in rural Kenya using a randomized evaluation. Spring protection leads to large improvements in source water quality as measured by the fecal indicator bacteria E. coli. Water quality gains at the home are smaller on average, but this finding depends critically on households’ water source choices. At households that only used the sample spring at baseline, 71% of the spring water quality benefits are translated into home water gains, suggesting that recontamination in transport and storage may be less of a concern than is sometimes claimed. Consistent with this view, the home water quality gains from spring protection are no larger for households with better baseline sanitation or hygiene knowledge. Changes in household water source choices after spring protection are used to derive revealed preference estimates of the willingness to pay for improved water quality using a travel cost approach. The average willingness to pay for the moderate gains in home water quality due to spring protection is at least US$3.27 per household per year. We find no significant child health effects of spring protection.
Kremer, Michael, Jessica Leino, Edward Miguel, and Alix Peterson Zwane. 2006. “Spring Cleaning: A Randomized Evaluation of Source Water Improvement”. Abstract
Water-related diseases, particularly diarrhea in young children, kill two million people annually. To address this problem, donors and governments often provide infrastructure such as communal standpipes, wells, and protected springs in rural areas, where piping water into homes is infeasible. We study the impact of source water quality improvements achieved via spring protection in rural Kenya using a randomized evaluation. Spring protection leads to large improvements in source water quality as measured by the fecal indicator bacteria E. coli. Water quality gains at the home are smaller on average, but this finding depends critically on households’ water source choices. At households that only used the sample spring at baseline, 71% of the spring water quality benefits are translated into home water gains, suggesting that recontamination in transport and storage may be less of a concern than is sometimes claimed. Consistent with this view, the home water quality gains from spring protection are no larger for households with better baseline sanitation or hygiene knowledge. Changes in household water source choices after spring protection are used to derive revealed preference estimates of the willingness to pay for improved water quality using a travel cost approach. The average willingness to pay for the moderate gains in home water quality due to spring protection is at least US$3.27 per household per year. We find no significant child health effects of spring protection.
Johnston, Alastair Iain, and Robert S Ross. 2006. New Directions in the Study of China's Foreign Policy. Stanford University Press. Publisher's Version Abstract

This book brings together several generations of specialists in Chinese foreign policy to present readers with current research on both new and traditional topics. The authors draw on a wide range of new materials—archives, documents, memoirs, opinion polls, and interviews—to examine traditional issues such as China's use of force from 1959 to the present, and new issues such as China's response to globalization, its participation in several international economic institutions, and the role of domestic opinion in its foreign policy.

The book also offers a number of suggestions about the topics, methods, and sources that the Chinese foreign policy field needs to examine and address if it is to grow in richness, rigor, and relevance.

Johnston, Alastair Iain, and Robert S Ross. 2006. New Directions in the Study of China's Foreign Policy. Stanford University Press. Publisher's Version Abstract

This book brings together several generations of specialists in Chinese foreign policy to present readers with current research on both new and traditional topics. The authors draw on a wide range of new materials—archives, documents, memoirs, opinion polls, and interviews—to examine traditional issues such as China's use of force from 1959 to the present, and new issues such as China's response to globalization, its participation in several international economic institutions, and the role of domestic opinion in its foreign policy.

The book also offers a number of suggestions about the topics, methods, and sources that the Chinese foreign policy field needs to examine and address if it is to grow in richness, rigor, and relevance.

Co-Winner of the 2007 Best Book Award, European Politics and Society Section, American Political Science Association Winner of the 2004 Gabriel Almond Award for Best Dissertation in Comparative Politics, American Political Science Association Winner of the 2003 Ernst B. Haas Prize for Best Dissertation in European Politics, American Political Science Association.

Daniel Ziblatt begins his analysis with a striking puzzle: Upon national unification, why was Germany formed as a federal nation-state and Italy as a unitary nation-state? He traces the diplomatic maneuverings and high political drama of national unification in nineteenth–century Germany and Italy to refute the widely accepted notion that the two states' structure stemmed exclusively from Machiavellian farsightedness on the part of militarily powerful political leaders. Instead, he demonstrates that Germany's and Italy's "founding fathers" were constrained by two very different pre-unification patterns of institutional development. In Germany, a legacy of well-developed sub-national institutions provided the key building blocks of federalism. In Italy, these institutions' absence doomed federalism. This crucial difference in the organization of local power still shapes debates about federalism in Italy and Germany today. By exposing the source of this enduring contrast, Structuring the State offers a broader theory of federalism's origins that will interest scholars and students of comparative politics, state-building, international relations, and European political history.

BASED ON his constitutional powers and the authorization for the use of military force granted by congressional resolution after the events of Sept. 11, 2001, President Bush has declared himself free to ignore any law that he thinks limits his ability to fight terrorism. This is an extraordinary claim for any president in a country that prides itself on a rule of law binding government officials as well as ordinary citizens.

In signing the McCain amendment outlawing cruel, inhuman, and degrading treatment of detainees this month, Bush announced that he might ignore the amendment in order to fight terrorism, the very field that the amendment, adopted by overwhelming majorities in both Houses, had specifically addressed. The statute forbids the president only to do anything that, in the circumstances, “shocks the conscience,” thus violating the due process clause of the Fifth Amendment. This leaves him broad discretion and little reason to claim powers Congress has specifically denied him. But that is what he has done.

This is at least the fourth occasion Bush has announced that he is not bound by statutes or treaties. He has said he is also free to ignore statutes prohibiting torture, detention of Americans without legislative authority, and electronic surveillance for intelligence purposes without compliance with laws set up to regulate that activity. These claims could be consistent with obedience to statutory law only if either the Constitution had given him exclusive powers (a contention that few accept), or the situations in which he claims authority were so unusual as not to have ever been contemplated by Congress. Certainly the general words of the congressional authorization to use force to deal with Al Qaeda were not meant to overrule every statute the president felt was a hindrance in fighting terrorism.

In each of these cases, Congress plainly addressed the very situations in which Bush now claims an exemption from law. The statute regulating electronic surveillance for intelligence purposes includes emergency and wartime exceptions. Congress had in mind the wartime detention of Japanese-Americans when it forbade detention of an American seized far from a war zone without a specific statute. The McCain amendment was intended to leave the president with discretion to apply the vague constitutional standard of “shocking the conscience,” but only that much discretion. Only the prohibition of torture is absolute and without exception, and Congress wanted it that way.

Indeed, the president's defiance of statutory law is even bolder than this suggests. Each of these executive actions, taken in violation of specific statutory prohibitions, has been treated as a matter of national security secrecy, and therefore anyone who reveals the fact that the president is violating statutes passed by Congress is subject to the immediate threat of prosecution under the espionage statutes. The result in the recent case of wiretaps of Americans without judicial warrants is particularly bizarre. There was nothing secret about our technical capacity to monitor phone calls coming to or from the United States. Nor was there anything secret about our desire to do so to prevent terrorism. No one has, finally, revealed whose calls or e-mail messages were the subject of surveillance. All that could have been secret about the activities described in the New York Times was that the president was defying a law that most thought he had to obey.

It is a fundamental mistake to think that the central domestic conflict about fighting terrorism is only between supporters of national security and supporters of civil liberties of Americans. The prior question is about the effect of law in the form of duly enacted statutes, negotiated between Congress and the president, reconciling these competing claims. The president is claiming that his powers to deal with terrorism as commander-in-chief override a negotiated compromise with the Congress, embodied in a statute signed by the president. He is saying, simply and flatly, that no law can stand in his way. We should not accept that claim.

If the threat of terrorism is to be with us for decades, will our children and grandchildren remember a time when our president's actions were ruled by law?

Philip B. Heymann, former US deputy attorney general, is a professor at Harvard Law School and a faculty associate of the Weatherhead Center for International Affairs.

The map tells the story. Latin Americans are voting for leftist or center-left governments in record numbers. Only El Salvador and Colombia seem certain to end 2006 with conservative governments still in power.

While Latin America is moving to the left, the US government has moved to the right. The Bush administration is far more likely to oppose and far less interested in accommodating Latin American hopes for change than its predecessors.

Can a collision between the United States and Latin America be avoided? Why is Latin America moving to the left? Should (North) Americans worry or cheer?

Latin America freed itself from Spanish and Portuguese rule in the 1820s, but it took another half-century of civil strife to secure legal equality for indigenous citizens, end slavery, modernize institutions, and get economic growth started. But economic growth also made Latin America more unequal than ever before. With plenty of unskilled labor, wages stagnated while rents and profits rose rapidly. Worse yet, those at the top excluded others. Literacy and property requirements deprived most Latin American men and all women of the right to vote until well after World War II.

In the 1930s and 1940s, Latin America began to change. Old dictatorships collapsed and democratic elections brought new, reformist governments to power. The new democracies nurtured vibrant civil societies by legalizing labor unions and encouraging organizations representing farmers and farm workers, rural communities and city neighborhoods, women, and minorities. Among the organizers and participants in this democratic upsurge were socialists, anarchists, liberals, and progressive-minded church people—along with communists and other radicals. In Brazil, Chile, Cuba, Guatemala, Mexico, and Peru, communists won wide support because of their role in supporting democratic movements.

Many in Latin America saw this as dangerous—especially wealthy elites, conservative politicians and military officers, and the influential hierarchy of the Catholic Church. When the Cold War began, the United States allied itself with these reliably anticommunist groups. Governments of the left and center-left that failed to confront and exclude communists and their followers soon became anathema in Washington. The United States encouraged and even directly aided its Latin American allies to overthrow such governments, even when they were freely elected. By the early 1950s, the dictators were back.

The Cuban Revolution formed part of a much larger wave of revolts against military dictatorships that brought democratic governments back to power in 10 Latin American countries between 1956 and 1960. Rather than work with this trend, the United States came to see it as potentially dangerous. The Castro regime fulfilled the prophecy when it sought Soviet aid to resist US efforts to destroy it. To prevent another Cuba, the United States called back the dictators again, despite the democratic rhetoric of the Alliance for Progress. By 1970, only Venezuela and Costa Rica were still holding open, competitive elections.

For more than 40 years, from 1948 to the early 1990s, the United States used its power and resources to make sure that Latin Americans had governments more conservative (and thus anticommunist) than Latin American voters were willing to elect. The human cost of this effort was immense. Between 1960, after the Soviets had dismantled most of their gulags, and the Soviet collapse in 1990, the number of political prisoners, torture victims, and executions of nonviolent dissenters in Latin America vastly exceeded that of the Soviet Union and its East European satellites.

Even when economic problems and protests drove the military from power in the 1980s, voters tended at first to elect conservative administrations to make sure the military would have no excuse to return to power.

By the 1990s, however, voters had lost their fear. Civilian governments had succeeded in downsizing and purging military establishments and with the collapse of the Soviet Union, the United States was no longer so obsessed with keeping the left out of power.

The conservative, mostly military governments of the Cold War had little interest in fixing inequality. Politicians and citizens who worried about inequality were the people these regimes jailed, tortured, and executed. Instead of overcoming inequality as their economies grew, the Latin American countries became more unequal. And then they stopped growing.

The 1982 debt crisis helped to push military regimes from power but undermined the new democratic governments that replaced them. The centrist and center-right political parties and politicians that won the first democratic elections usually lost re-election bids. Second-time elections often benefited centrist and populist parties anxious to jump start economic growth. Many opted for a package of reforms that came to be called the “Washington Consensus”—balanced budgets, freer trade, deregulation, and privatization.

With a few exceptions, such as Chile, these orthodox economic reforms did not generate sustained economic growth. Most countries experienced short spurts of growth followed by reversals, some linked to volatile capital outflows as investment fund managers in the United States and Europe panicked when slow growth undermined currencies and governments failed to raise the revenues they needed to pay off their debts. In the quarter century from 1980 to 2005, Latin America has barely grown at all—the worst quarter century in Latin American economic history since the civil wars after independence from the 1820s to the 1850s.

It is important not to forget, however, that despite its recent travails, most of Latin America ended the 20th century much better off than at the beginning. The per capita gross domestic product of the region's major economies averaged about 1.5 percent, roughly equal to the US economy. The fastest growing economy in the western hemisphere for the past hundred years was not the United States but Brazil. Living standards also improved.

Latin America is not the poorest region in the world, though some Latin American countries, including Bolivia, Haiti, and Honduras, rival the poverty of Africa and South Asia. In fact, Latin America has a large number of middle class voters impatient with the failure of the governments they elect to improve conditions for their poverty-stricken fellow citizens.

Political systems that seem rigged for wealthy elites, persistent and highly visible inequalities, and economies where living standards have not risen for 20 years have pushed Latin American voters back to the left. The only surprise is how long it has taken.

Some leftist and center-left governments have performed brilliantly. The best example is Chile, where a coalition of Socialists and Christian Democrats has raised economic growth from less than 1.5 percent under the Pinochet dictatorship to over 5 percent since the restoration of democracy in 1990—and managed to reduce poverty dramatically at the same time. In fact, economic growth in countries with leftist or center-left governments was more than twice as fast in 2005 as in countries with conservative or center-right governments.

For US policy makers accustomed to having their way in the hemisphere, the temptation to intervene, overtly or covertly, to reverse Latin America's leftward drift will likely grow harder to resist in 2006. The likely strategy will be to look for ways to drive a wedge between the noisier anti-American, pro-Cuban regimes in Venezuela and Bolivia and the more moderate administrations elsewhere. This strategy will probably fail, because the United States has so little to offer even its conservative friends after essentially giving up on new free trade agreements.

Faced with few peaceful options, the United States may drift back to the Cold War policies of the 1960s and 1970s—or to the dollar diplomacy that preceded it. Between 1898 and 1994, the US government succeeded in overthrowing, or inducing friends to overthrow, no fewer than 41 Latin American governments (a dozen of them freely elected). That's one every 28 months. Before it goes that route again, the United States should probably consider modifying its ideologically driven hostility to such regimes. A little pragmatism, coupled with a ban on overthrowing elected governments, would make a good start.

John H. Coatsworth is a professor of Latin American affairs, director of the David Rockefeller Center for Latin American Studies, and a member of the Executive Committee of the Weatherhead Center for International Affairs at Harvard University.

Pages