More Than Good Intentions: How a New Economics is Helping to Solve Global Poverty
Dean Karlan and Jacob Appel
Plume, $16 (paper)

Pillars of Prosperity: The Political Economics of Development Clusters
Timothy Besley and Torsten Persson
Princeton University Press, $29.95 (paper)

Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty
Abhijit V. Banerjee and Esther Duflo
PublicAffairs, $15.99 (paper)

Why Nations Fail: The Origins of Power, Prosperity, and Poverty
Daron Acemoglu and James A. Robinson
Crown Business, $30 (cloth)

The classical economists, from William Petty and Adam Smith to John Stuart Mill, were all development economists. They offered general theories of markets and growth and wrote about a particular developing country (typically Britain) going through a process of industrial transformation.

Then, for more than a century, development shifted from the economic and intellectual core to the periphery. Debates about development focused on controversies over Soviet industrialization and on the need for protectionist policies to defend infant industries in places—such as the United States, Germany, Eastern Europe, Australia, and India—trying to “catch up” with England.

After the Second World War, as a large number of countries became independent (or “liberated,” as in China), development economics blossomed but remained at the intellectual margins of the economics discipline. Underdevelopment was conventionally understood as the product of market and institutional failures. In the prevailing paradigm, such failures were regarded as exotic exceptions or as the remediable results of bad policies instituted by new governments.

The challenges of persistent poverty and underdevelopment eventually undercut this intellectual marginalization. In the last three decades of the 20th century, economists recognized that standard market-equilibrium models—with their smoothly functioning invisible hands—break down in the context of information failures and dysfunctional institutions in all economies. So economists started paying more attention to rumblings from the periphery. As Joseph Stiglitz put the problem in a 1989 essay:

A study of less developed countries is to economics what the study of pathology is to medicine; by understanding what happens when things do not work well, we gain insight into how they work when they do function as designed. The difference is that in economics, pathology is the rule: less than a quarter of mankind lives in the developed economies.

In the past decade, development economics has grown to extraordinary prominence, not just in academia but also in the public arena. This new development economics has moved in two strikingly different directions. The first focuses on micro-level policy interventions. It uses the tools of field experiments and randomized controlled trials (RCT) to evaluate focused strategies for alleviating persistent poverty. The second trend focuses on macro institutions: the structures of democracy, autocracy, centralized and diffused power, and legal protections of property and contracts that organize politics and markets. Drawing ideas from the burgeoning field of institutional economics, this second stream of work has addressed grand, old historical questions such as why some countries have succeeded in the march to prosperity while others have languished.

In the past two years, some of the brightest minds of the new development economics have published books exploring these two trends. A critical appreciation of these works may give us some insight into the high points and pitfalls of the emerging micro and macro approaches.

Field research has long been part of development economics, but Abhijit Banerjee and Esther Duflo, as well as Dean Karlan and Jacob Appel, have taken this work in novel directions. With striking singularity of purpose, they evaluate micro-level policy interventions that might, if scaled up, reduce global poverty. They use randomized controlled trials, a field experimental method relatively new to economics, to correct for biases in earlier statistical inquiries. They also rely on team effort and an almost missionary zeal to carry out experiments and find out what works.

In Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty (2011), Banerjee and Duflo, two of the best economists around, aim to disabuse readers of the notion that there is a magic bullet for poverty. Extreme poverty is such a crushing burden on so many lives that we are often tempted by sweeping solutions. But instead of offering grandiose principles, Banerjee and Duflo direct attention to the lives of poor people in all their richness and complexity.

The methodological innovation does not fully justify the holier-than-thou attitude that many users of randomized controlled trials adopt vis-à-vis the results of earlier research.

Drawing on the results of field experiments, Banerjee and Duflo isolate concrete deficits that the poor face. For example, they find that the poor often lack basic information about the benefits of children’s immunization and nutrition, the dangers of over-medication, the risks of HIV infection, how much fertilizer to use, and the quality of politicians vying for election. Moreover, the poor suffer from the usual human weaknesses of will, intensified by the special pressures their oppressive lives put on their decision-making abilities.

In response, Banerjee and Duflo suggest nudging poor people in the right direction by establishing various default options: savings accounts in which money is easy to deposit but somewhat difficult to take out, simple chlorine dispensers at drinking water collection points, and easy availability of salt fortified with iron and iodine. Of the various RCT findings, the two most cost-effective programs seem to be de-worming children in areas where intestinal worms are rampant and providing remedial education for poor children who fall behind in class.

Karlan and Appel cover similar ground in More Than Good Intentions: How a New Economics is Helping to Solve Global Poverty (2011). While Banerjee and Duflo are more wide-ranging and substantive in their analysis, Karlan and Appel offer more detailed human-interest stories and a special focus on microcredit.

The provision of small loans to the poor has captured public imagination and generated a great deal of hype. To evaluate the enthusiasm, Karlan and Appel carefully sift evidence from different parts of the world. The results are decidedly mixed. Microcredit is not a panacea, though it generates real benefits in some cases. But even when rates of return on investment run high, there are not many borrowers, and so the impact of the loans is small (Poor Economics and More Than Good Intentions both explain why). In many cases the poor may be helped more by micro-saving products, and nudges that facilitate them, than by microcredit products.

RCTs face serious challenges to their “external validity.” Because an intervention is examined in a microcosm of a purposively selected population, the results do not generalize beyond the boundaries of the study.

Banerjee, Duflo, Karlan, and Appel all brim with enthusiasm, as the subtitles of their books suggest. Much of this enthusiasm for a good cause is justified and heart-warming. The RCT approach is modeled on clinical drug trials in medicine, where RCT is an old technique. In a field experiment, people are randomly assigned to treatment and control groups. The treatment group might, for example, have access to a savings plan, while the control group does not. This approach provides a relatively clean way of deciphering the average impact of a policy intervention. In older studies of savings, a new savings plan might have looked effective because people with strong predispositions to save decided to use the plan. But if the plan had not been available, they might have saved just as much in some other way. RCTs thus appear to avoid some of the issues of selection and bias that afflict the older statistical studies in development economics.

The methodological innovation does not, however, fully justify the holier-than-thou attitude that many users of RCTs adopt vis-à-vis the results of earlier research. RCTs face five problems that limit the force of findings based on them.

First, it is very hard to ensure true randomness in setting up treatment and control groups. So even within the domain of an RCT, impurities emanate from design, participation, and implementation problems.

Second, RCTs face serious challenges to their generalizability or “external validity.” Because an intervention is examined in a microcosm of a purposively selected population, and not usually in a randomly sampled population for any region, the results do not generalize beyond the boundaries of the study. For all their internal flaws, the older statistical studies, often based on regional samples, permitted more generalizable conclusions. Neither method has a monopoly on correctness.

Third, for many important policy issues, RCTs are not very useful. You cannot run experiments in order to decide where to put power plants or ports. You cannot do a controlled test on the advisability of tight money, fiscal austerity, or deregulation. Moreover, even if you can show convincingly that a policy intervention works in a small-scale trial, policymakers still have to worry about the economic and political spillover effects of a policy when it is implemented regionally or nationally. What will be its impact on other markets and the macro economy? And what happens when a policy once handled experimentally by a local NGO is taken up for large-scale implementation by a national bureaucracy, even a well-functioning one?

For many important policy issues, RCTs are not very useful. You cannot run experiments in order to decide where to put power plants or ports.

Fourth, RCTs show only the average impact: a policy intervention may be very helpful for some people and not at all for others, just as a clinical drug trial may show that a particular drug works well for the average person, but it may not work at all for you. One of the standard questions of political economy, however, concerns who gains and who loses from a given policy. RCTs cannot answer that distributional question.

Finally, even when an RCT shows quite cleanly that A causes B, we do not quite know the mechanism through which it works. In interpreting many experimental results, Banerjee, Duflo, Karlan, and Appel give plausible accounts of the processes that may be at work, but these are at best their informed guesses. They are usually not rigorous derivations from the experiments themselves. In understanding alternative mechanisms through which A may have caused B, theory has to play a more important role in empirical economics than the experimentalists have assigned to it.

Banerjee and Duflo show commendable humility in their admission that they do not have ready-made answers to policy questions. This humility is strangely absent, however, when it comes to their method. The criticisms of RCTs have been known for some time (see, for example, a forum featuring Banerjee in the July/August 2006 issue of this magazine), but Banerjee and Duflo do not respond or even refer to them in their book. They also do not tell us that they may be missing some of the most important issues in fighting poverty, simply because RCTs or experimental methods may not be appropriate for studying them. Non-experimental studies do not pass the experimental purists’ standard. But that cannot mean that the questions such studies explore are less important than the ones investigated with RCTs.

Non-experimental studies do not pass the experimental purists’ standard. But that cannot mean that the questions such studies explore are less important than the ones investigated with RCTs.

An unhappy irony is at work here. Banerjee and Duflo lament the power of fads, simple formulas, and instant miracles in development economics. Unfortunately, and partially under their leadership, another fad now dominates the subject. Many graduate students are obsessed with carrying out experiments or finding some other clean econometric tools, so that their micro results can all contribute to the good fight against global poverty.

As in some types of anthropology, a “narcissism of the small” now pervades parts of the field, which is overwhelmed by researchers focused only on immediate policy usefulness. Undoubtedly that is an important goal, but, correspondingly, much less interest is focused on larger, structural issues or the contribution the subject could make to social thinking in general. It’s as though campus physics departments have been taken over by teams of frightfully useful engineers. I have no doubt that lending institutions such as the World Bank should appoint people specialized in experimental methods in order to get good evaluations of their loan projects in education and health, though obviously not for infrastructure. But I am not sure of the wisdom of turning economics departments into project evaluation agencies that primarily serve the World Bank and train its recruits.

Another fad now dominates the subject. Many graduate students are obsessed with carrying out experiments or finding some other clean econometric tools.

Even at the micro level favored by Banerjee, Duflo, Karlan, and Appel, there have been many studies of structural-institutional factors that are difficult to randomly vary and that affect poverty at the ground level. Because these studies used old-school statistical methods, they are not kosher, and hardly any of them register in Banerjee, Duflo, Karlan, and Appel’s books. Those studies, largely performed in the 1970s and ’80s, used data to discuss how land tenure (say, sharecropping) affects farm productivity, how different types of labor relations in agriculture affect wages and employment of poor workers, how interlocking land-labor-credit relations with the landlord-employer-creditor trap workers in poverty, and how security of ownership rights affects productivity and investment. These studies also explored the role of credit arrangements in risk pooling and how land inequality hinders resolution of conflicts in the management of village commons—forests, fisheries, irrigation water, grazing lands, etc. All of this research concerned the livelihoods of the majority of poor people working in agriculture and related occupations. Yet the new development economics fighting global poverty finds no use for it. Nor is there much attempt to reopen earlier questions using the new methods, maybe because they are not as amenable to NGO-mediated RCTs as education, health, or microfinance interventions are.

These studies of micro-level institutions are also largely ignored in the second stream of new development economics: macro-institutional political economy. Since the 1990s, when Nobel laureate Douglass North and others turned the profession’s attention to the historical importance of large-scale institutions, cross-country statistical regressions exploring the relationship between institutions and development became the vogue. Daron Acemoglu, James Robinson, Timothy Besly, and Torsten Persson work in this style—although Acemoglu and Robinson’s popular-level book,Why Nations Fail: The Origins of Power, Prosperity, and Poverty (2012), avoids the regressions.

Why Nations Failis a big book. It includes dozens of historical examples in a breathtaking sweep over millennia and continents. Still, the main argument is simple. Nations succeed or fail in development according to how “inclusive” their political and economic institutions are. Among the inclusive political institutions are political centralization (a well-functioning state establishing cohesive order) and a pluralistic distribution of political power. Inclusive economic institutions allow broad-based participation in economic activities, secure property rights and entry of new business, and provide a level playing field and unbiased system of law. Though the theory is very general, the history—and there is a great deal of it in the book—is not deterministic: contingencies, small differences, and critical junctures (such as the Black Death and the opening of Atlantic trade) mattered a great deal in determining when and where inclusive institutions emerged.

Acemoglu and Robinson belong to a tradition of institutional economic history—including Marxist ideas about class struggle—that locates distributive conflicts at the root of institutional change, or lack of it. The violent and tortuous history of land reform has repeatedly shown us how landowning elites will act to preserve their own political power even if that power is a hurdle to economic growth and political inclusion. Thus, students of the political economy of development are familiar with the role of autocracy—from Czarist Russia to Imperial China—in preserving the power and property of landlords from the resistance of land-poor peasants. To those, such as myself, who take institutional-historical analysis seriously, Acemoglu and Robinson’s findings are thus important but not particularly earthshaking, though some of the historical examples drawn from Africa and Latin America are fascinating. But to mainstream economists, long insulated from classical political economy, the book serves as a welcome and instructive reminder, especially because Acemoglu and Robinson are widely respected for their theoretical and empirical virtuosity and breadth.

Simplicity is a great theoretical virtue. But I wonder if Acemoglu and Robinson have taken this virtue to excess. They tend, for example, to emphasize synergies among the different dimensions of political and economic inclusion. If we unpack inclusiveness, however, we can decipher some interesting tensions within the set. Exploring those tensions—between economic and political inclusion and among different aspects of each—makes for a less tidy package. But reality may be complicated and truth comes before tidiness.

For example, economic inclusion requires secure property rights. But political inclusion, with its pluralistic distribution of political power and broad participation, may not always secure the property rights of the few against encroachers and squatters or high taxes. Or, to take another example, the rule of law—part of political inclusion—is often an instrument used to protect the propertied from the propertyless, thus enforcing economic exclusion. English enclosure laws famously turned the poor users of the village commons into poachers. Still, the rule of law may be, on balance, a very good thing, even if it is sometimes at odds with economic inclusion. In short, political inclusion may not work smoothly with economic inclusion. The tension involved in the rule of law is captured well in the nuanced conclusion of Whigs and Hunters (1975),by the Marxist historian E. P. Thompson:

We reach, then, not a simple conclusion (law = class power) but a complex and contradictory one. On the one hand, it is true that the law did mediate existent class relations to the advantage of the rulers . . . . On the other hand, the law mediated these class relations through legal forms, which imposed, again and again, inhibitions upon the actions of the rulers. . . . In a context of gross class inequalities, the equity of the law must always be in some part sham. . . . We ought to expose the shams and inequities which may be concealed beneath this law. But the rule of law itself, the imposing of effective inhibitions upon power and the defense of the citizen from power’s all-intrusive claims, seems to me to be an unqualified human good.

Perhaps in order to acknowledge such important nuances, Acemoglu and Robinson say they will “refer to political institutions that are sufficiently centralized and pluralistic as inclusive political institutions” (emphasis added). This is unsatisfactory. “Sufficiently” preserves the appearance of theoretical simplicity by leaving the door open for circularity. We call institutions “politically inclusive” when they are inclusive enough to yield development. But then how much substance is left in the idea that inclusive institutions foster development?

More fundamentally, the idea that political centralization is an element of political inclusiveness is rather puzzling. Most historical instances of political centralization, either in the empire states of the past or in the modern nation states (such as Meiji Japan, Ataturk’s Turkey, and Mao’s China), have been associated with less inclusiveness in important respects. To be sure, a certain degree of political unification is necessary to build a coherent institutional framework for long-term development policies. But centralization and pluralism can be in contradiction with one another. Pluralism makes the collective-action problem for attaining a cohesive governance structure more difficult. (More on this later in the context of India.)

The case of England may suggest that centralization and pluralism can sometimes fit together and foster economic development. A distinctive feature of English history over many centuries has been a balance between political centralization and individual rights. Like much of the recent institutional economics literature, Why Nations Fail follows the lead of North and Barry Weingast in emphasizing the role of the Glorious Revolution in the story of England’s economic development. North and Weingast trace the success story of England to the king giving up royal prerogatives and increasing the powers of Parliament in 1688. The resulting political pluralism, against a background of centralization, helped to secure private property rights against state predation and allowed private enterprise and capital markets to flourish.

But other historians have proposed a very different story, with pluralism playing a significantly diminished role. S. R. Epstein has shown convincingly that the sharp decline in the cost of capital—which North and Weingast ascribe to the pluralism established by the Glorious Revolution—was instead a matter of England catching up with the Continental norm. England had lagged behind the Continent, some of it ruled by absolutist monarchs, since at least 1500. Epstein argues that the English advance over its Continental neighbors was not so much due to the liberties wrested from the monarch, but to the fact that early modern England had been able to build a centralized government, partly helped by its relative isolation and lack of domestic war. The strong state of England was thus in a better position later to invest in the expansion of empire and overseas markets. Economic historians such as Robert C. Allen and Gregory Clark have also expressed doubts about whether England’s success can be attributed to the superior institutions that came with the Glorious Revolution.

More recently, it has been easy to associate East Asian growth with political centralization. But Korea, for example, has long had a coherent bureaucratic state structure, and its take-off into self-sustained, state-guided capitalist development was largely achieved by the 1980s, before political liberalization. Acemoglu and Robinson find this consistent with their story about inclusiveness because they think South Korean economic institutions were already essentially inclusive. Land reforms prior to liberalization had indeed removed the hurdles to development posed by a landed oligarchy. But when General Park arrested the leaders of the industrial conglomerates (chaebol)—only the head of Samsung escaped because he happened to be abroad at the time—and made them conform to his long-term plans for Korea, no one bothered about property rights or rule of law, the inclusive economic institutions. Even when the chaebol played ball, helped by the carrot of large credit subsidies, their profits were ensured by long periods of severe labor repression under the thumb of the KCIA. In this respect South Korea was similar to many of the extractive political regimes of the time supported by the American CIA.

To reach the heart of the East Asian story, we need to get past Acemoglu and Robinson’s sweeping categories. Growth was not simply a product of political centralization plus some market-supporting institutions and economic inclusion. Masahiko Aoki and his colleagues have described the deal between the state and large business conglomerates in Korea and Japan as “cooperation-contingent rent.” In their view, economic development in these countries was not founded just on institutions that secure property rights and enforce contracts—no doubt very important for long-term investment—but on a state with a more affirmative role. The states helped to foster coordination, particularly in financial markets in early stages of industrialization. For example, the state would underwrite investment risks, facilitate interdependent investment decisions in orchestrated networks of producers and suppliers (in steel, shipbuilding, automobiles), establish public development banks and other institutions for long-term industrial finance, and nudge firms to upgrade their technology and move into sectors that fit with a national vision of development goals. Enabling and encouraging coordination is fundamentally different from protecting property rights. When private financial and other related institutions were underdeveloped and coordination was not self-enforcing, East Asian states created opportunities for extracting rents in return for performance and cooperation among producers and investors. The performance criteria in East Asia often included export success, which in a world of international competition kept the subsidized firms on their toes and encouraged cost and quality consciousness.

Acemoglu and Robinson are wedded to the idea that extractive institutions can generate some growth but cannot sustain it. Sustained growth needs innovation with its attendant “creative destruction,” allowing for the entry of new innovative firms and the exit of old, inefficient ones. Applying this to China, they ask us to trust their prediction that the country’s recent high growth will not last for long. They may be right, but I am not so sure. In general I find their statements about China pretty simplistic and largely “faith-based.” And their analysis of India is surprisingly scanty. In a sweeping book about economic development, they offer somewhat superficial and thin views about the remarkable economic performance of these two countries—which happen to be the world’s largest—in the last three decades.

Of course, Chinese growth rates are unlikely to be sustained, simply because China’s population is aging quickly (Japan and South Korea faced a similar fate). Still, for some aspects of human development—including mass education and health—the highly extractive institutions of Mao’s China delivered much better than those of most developing countries and provided a permanent foundation for long-term growth. Those institutions also provided mass rural electrification, which, along with some minimum human development, formed the basis of the phenomenal rural industrialization of the post-reform period.

Chinese growth has also been helped by a particular (pluralistic?) feature of the system dating back to its imperial history, which makes China different from other cases of political centralization. China has what economist Chenggang Xu calls “regionally decentralized authoritarianism,” in which powerful centralization in political and personnel control is combined with considerable decentralization of administrative and economic matters. The system gives substantial autonomy to local officials, encouraging regional competition and experimentation. But the central government tries to keep them under control through career-promotion incentives. (It has not been entirely successful. In recent years, as in many other cases of decentralization, the system has been afflicted by local elite capture, as evident in arbitrary land acquisition and toxic pollution.)

What about continuing to innovate once the current catch-up process ends? Like Acemoglu and Robinson, I believe that the too-big-to-fail Chinese state-owned enterprises and the iron grip of the Party nomenklatura on them may be a drag on innovation, particularly disruptive innovation. But the Korean and Japanese cases suggest that big entrenched companies can be successful in some kinds of innovation, particularly the steady, incremental kind.

Moreover, China now has a thriving private entrepreneurial sector, providing the largest share of industrial value-added in the economy: 5 million registered private enterprises have $1.3 trillion in registered capital. Even an authoritarian Chinese state probably can’t (or won’t) re-bottle the already-unleashed entrepreneurial spirit of such vast numbers for long.

Furthermore, using firm-level productivity data for 1998–2007 economist Loren Brandt and his colleagues have shown that creative destruction is alive and kicking in Chinese manufacturing. Thus far these innovations may still be catching up with those of high-income countries. But large rises in spending on research and development, in the Chinese share of patents taken out in other countries (particularly in chemical engineering, computers, and telephone and data transmission), and in the number of articles by Chinese researchers in top international science and technology journals all point beyond catch-up. Parts of the economy now show signs of commercially active technology innovation. In a recent study of China’s integrated circuit design industry, economists Dieter Ernst and Barry Naughton find that the country seems to be successfully innovating in global technology sourcing and rapid responses to changes in global production. China has achieved far more than the autarchic Soviet Union ever did in terms of innovation. On balance, I am less inclined to pronounce a definite opinion about China’s future than Acemoglu and Robinson are.

India deserves more attention from Acemoglu and Robinson, at least from the point of view of assessing their theory of inclusive institutions. India is a vigorous electoral democracy with intense political competition. The once-predominant Congress Party has been in serious decay for the better part of four decades, and for the last two of them India has been ruled by shaky coalition governments. Indeed, India arguably has too much political competition—370 parties contested in the last national election—with some elements of a race to the bottom as a consequence. There is competitive populism and clientelism, with parties announcing all kinds of subsidies and handouts that wreck the budget. For instance, provincial politicians running for election have been known to promise free water and electricity, decimating the chances for long-term investments in them.

And democracy has brought about a veritable social revolution in India, mobilizing the hitherto subordinate castes. While caste today has much less impact on social status and occupational distribution than it once did, competitive politics has rejuvenated caste as a political mobilization device. Intense pluralism, participatory politics, political fragmentation, and what V. S. Naipaul has called “a million mutinies” have made collective action on long-term decisions immensely difficult in India.

India shares with Italy weak political centralization as well as endemic corruption, mafia-controlled localities, sleazy politicians, and shaky coalition governments. But, also like Italy, India has entrepreneurial capitalism, a vigorous civil society, and an active and independent judiciary. If Acemoglu and Robinson are right, weak centralization spells trouble for India’s development. But if Italy, with per capita income about two-thirds that of the United States, has made it to the comity of developed countries, India can have hope for the future. In fact, compared with Italy, India has a much younger population, a larger domestic market, and higher saving rates. Its future may be brighter than one would be led to believe by Acemoglu and Robinson’s emphasis on the negative effects of a lack of political centralization.

Political institutions, particularly those that enhance or weaken state capacity, are also the central focus of Besley and Persson’s Pillars of Prosperity: The Political Economics of Development Clusters (2011). This book is much more technical than the other three. For the sake of comparison, I shall skip the technical parts and confine myself to some general aspects of the book that extend the macro-political analysis of Acemoglu and Robinson.

Besley and Persson start with a quotation from Adam Smith suggesting that “peace, easy taxes, and a tolerable administration of justice” are the pillars of the wealth of nations. (By the way, more than two thousand years before Adam Smith, the Indian philosopher Kautilya had made broadly similar statements in Arthashastra, a classical treatise on the political economy of statecraft). This leads Besley and Persson to focus on the legal and fiscal capacities of states and their ability to curb internal violence. The authors find that these capacities vary together because of common determinants, inherent complementarities, and the various feedback loops they generate. For example, if you are good at keeping internal peace, then subjects are more willing to pay taxes; more taxes can pay for a better legal system, and a better legal system helps to keep the peace.

Besley and Persson’s theoretical models explore leaders’ incentives to build state capacities and the ways in which opponents react. Besley and Persson then carry out a grand empirical exercise, constructing a Pillars of Prosperity Index (PPI) for each country. PPI reflects state fiscal and legal capacity, social peace (lack of civil war or repression), and per capita income. The authors—who are careful not to suggest causation—find that across 150 countries, PPI positively correlates with variations in conditions such as constraints on the power of the executive and ethnic homogeneity. This suggests that social homogeneity and an executive limited by democratic checks and balances are good for, or go together with, prosperity and peace. Somewhat to the surprise of Besley and Persson, India comes out as a drastic outlier. China does too, but somewhat less so.

Like many social scientists, I am not a fan of cross-country statistical exercises. I am more impressed by their theoretical models, which elegantly unify a disparate literature on institutions that perpetuate poverty and political violence.

What Besley and Persson call “cohesive political institutions” have affinity with Acemoglu and Robinson’s inclusive ones. Both books emphasize that the incumbent elite may want to block generally beneficial institutional change. Some of my criticisms of Acemoglu and Robinson’s framework apply to Besley and Persson’s as well. Just as Acemoglu and Robinson ignore the tension between political centralization and pluralism, Besley and Persson do not consider how an expansion of democratic rights (one part of cohesive institutions) may not always go well with the security of property rights (another element of cohesion), thus producing interference in the all-good–things-go-together scheme. And like Acemoglu and Robinson, Besley and Persson overstate the importance of enforcing property rights and contracts and understate the importance of resolving coordination failures—a key factor in East Asian state capacity.

The Indian case, which seems to spell difficulty for Besley and Persson, as it does for Acemoglu and Robinson, illustrates, in my judgment, an aspect of collective-action problems that requires more analysis. Some decades back I wrote a short book that suggested a unifying theoretical framework for understanding Indian macro-politics. India is one of the world’s most heterogeneous large societies. When there is such a diversity of social groups none of which is individually powerful enough to hijack the system for itself, each of the contending groups then may invest in checks and balances to curb the excesses of the others. This facilitates democracy. But the same social diversity that helps Indian democratic institutions survive also makes collective action on long-term development decisions more difficult, often encouraging dissipation of the social surplus in populist subsidies and handouts. Whatever the right story about India may be, it is an important case for institutional analysis.

Besley and Persson, somewhat implausibly, find their macro-political models consistent with recent micro-experimental studies, like those of Banerjee, Duflo, Karlan, and Appel. Acemoglu and Robinson take a more adversarial position with regard to those experiments:

Many of the micro-market failures that are apparently easy to fix may be illusory: the institutional structure that creates market failures will also prevent implementation of interventions to improve incentives at the micro level.

Banerjee and Duflo do not share Acemoglu and Robinson’s institutional pessimism. Their view is that

the political constraints are real, and they make it difficult to find big solutions to big problems. But there is considerable slack to improve institutions and policy at the margin. . . . These changes will be incremental, but they will sustain and build on themselves. They can be the start of a quiet revolution.

While I find that some RCT practitioners bubbling with policy advice are not as sensitive to the issue of political constraints, on balance my position on these matters is closer to Banerjee and Duflo’s. Of course, the lower the state capacity, the narrower the scope for the incremental changes.

All four books agree that there are no general recipes or quick fixes for poverty. The macro-political books point us to the grueling task of building inclusive or cohesive institutions and remind us of historical contingencies. The micro experimentalists are a bit more cheerful. In their judgment, small transfers and nudges can make a difference.

Neither group gives enough attention to much of the pre-existing micro-institutional work in development economics or to many of the regional policy issues crucial to fighting poverty. And none of the authors speak to the need for in-depth country or regional studies of political and economic processes, which provide deeper insights than those gleaned from cross-country standardized data, case studies, or micro experiments. Meticulous country and regional investigations do not answer “hot” global questions with over-arching judgments, which seems to preoccupy many prominent development economists today. Having languished for years in the periphery of economics, this profession now has larger aspirations, including hobnobbing with the likes of Angelina Jolie.