Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty, by Abhijit Banerjee and Esther Duflo (Public Affairs, 320 pp., $26.99)
This book’s title and subtitle sound like preemptive strikes against criticism: after all, who would stand against the poor or the need to fight global poverty? Moreover, Poor Economics comes with laudatory appreciations by Nobel Prize–winning economists, not to mention Bill Gates. Its authors, MIT economists Abhijit Banerjee and Esther Duflo, founded the Abdul Latif Jameel Poverty Action Lab (J-Pal), which has pioneered scientific evaluation of poverty-alleviation programs around the world. Admiring media profiles of the French-born Duflo usually show her dressed in fatigues in some remote African or Indian village, not teaching at MIT, where she is nonetheless regarded as the economics department’s rising star.
If the book’s prepublication hype had been less excessive, Poor Economics would suffice as a remarkable collection of data about the efficiency of poverty-alleviation programs. But instead of providing a comprehensive reevaluation of global poverty issues, as promised, the book is mostly a collection of anecdotes about well-intentioned micro-actions that governments, international organizations, and above all NGOs implement in destitute tropical villages. The authors’ intention is to bypass grand ideological debates about the origins of poverty or the rationale for international aid. Such debates, they argue, do nothing to improve the present living conditions of the poor. Instead, they prefer to instruct us how to act now, for the benefit of approximately 800 million poor people globally—those living on less than the equivalent of $1 daily. Immediate improvement is possible, the authors maintain, if we select programs that actually work.
Floating above controversies between those who believe in market solutions and those who rely on government intervention, Banerjee and Duflo recommend what they characterize as rational options, based on hundreds of randomized control trials. Should free food be distributed to the poor? No, they answer; most of the world’s poor already have access to food supplies. What they really need is iodized salt to prevent vitamin deficiencies and other maladies. Should poor children be compelled to attend primary schools? Yes, but it’s more efficient to give education money directly to parents (as is done in Mexico and Brazil) as an incentive to send their children to school, instead of keeping them at work on the farm or in the streets. Should anti-malaria mosquito nets be distributed free, or offered for sale? The debate has to do with whether people will make use of something that they didn’t purchase themselves; the authors argue that, in this case anyway, the evidence seems to say yes—so the nets should be distributed free.
In this way, Banerjee and Duflo promote hundreds of poverty solutions, all based on random trials. Some of their conclusions sound counterintuitive: they’re not sold on microcredit, for instance, which has been heavily promoted for years as the best way to generate individual enterprise in the developing world (what I have called “barefoot capitalism”). They observe that microcredit goes not to the poorest, but more often to already-active entrepreneurs. Further, the authors say, most poor people don’t want to become entrepreneurs, anyway: they look for safe jobs, preferably in the public sector.
“To promote deworming children in Africa as an effective way to improve future income may be less exciting than a debate on aid,” Banerjee and Duflo admit, “but it is more effective.” They have a point. Their book will likely serve as an effective guide for honest NGOs (not all of them are) and other “on-the-ground” organizations. It’s true that today, more often than not, poverty-alleviation programs are rarely subjected to evaluation.
Banerjee and Duflo don’t hesitate to promote what they call a “paternalistic” stance (what is often called “soft paternalism” has become trendy among liberal economists, a rebuttal to the rational-action theory promoted by Gary Becker). The poor, they maintain, make bad choices—like foregoing vaccination, failing to send their children to school, buying DVD players instead of food, or spending lavishly on weddings or funerals instead of piping clean water into their homes. Duflo and Banerjee cast themselves as righteous economists: they know what’s good for the poor better than the poor do. And not only for the poor. In rich countries, “we are ignorant as well but with no consequence,” they claim. “When we drink tap water, we suppose it is safe because the decision has been taken in our name by our well-intentioned government.” The authors, against the dominant trend in contemporary development economics, dismiss the argument that stable governments and institutions are the preconditions for progress. Even within a corrupt government, they argue, incremental improvements are achievable. They cite positive results of poverty-alleviation programs they conducted in Uganda and Indonesia (though these are not the worst of the world’s governments).
It’s difficult not to admire Banerjee and Duflo for their fieldwork and dedication to the poor. The data they provide are indisputable and useful. The problem with their book is what is not in it. The geographical sample is far from exhaustive. China, for instance, is not included—even though public data show that at least 100 million Chinese live under the $1-a-day poverty line. Banerjee and Duflo may not have been authorized to work in China, which prohibits foreign NGOs. The Arab world is absent from the survey as well. With these huge omissions in mind, is it legitimate to draw universal conclusions from the comparison of a Kenyan village with a Calcutta slum? The authors mix disparate conditions as if poverty is the same everywhere, disregarding culture and social norms. The Anglo-Bengali economist Amartya Sen wrote decades ago that to be poor in a democracy—where citizens generally have access to free media, voting rights, and religious freedom—cannot be compared with being poor in a repressive regime. Regrettably, Duflo and Banerjee don’t take noneconomic factors into consideration.
In addition, incremental programs against poverty—or “removing poverty traps,” in economic parlance—are less efficient than more dramatic shifts in economic policy. The authors observe that 800 million people remain poor, despite development policies. If they had written their book 20 years ago, though, that figure would have been 2 billion. The massive shift of China, India, and Brazil, among other nations, toward free markets and entrepreneurship, has pulled over 1 billion people out of poverty in just 20 years. Duflo and Banerjee wouldn’t deny this unprecedented success, but they also don’t mention it. The poverty-alleviation programs they rightly support are welcome, but limited; they have achieved far less than market policies.
Poverty-alleviation programs, such as the Gates Foundation’s well-intended actions, don’t necessarily contradict other development policies. However, many bad governments, mostly in Africa, pretend to fight poverty with mosquito nets and deworming efforts in lieu of an actual development policy. Duflo and Banerjee advocate incremental changes within ruling oligarchies, which may bring some immediate alleviation for the poor. But some ruling oligarchies will also use the NGOs for their own purposes. NGOs and international-aid organizations, protecting their own legitimacy and market space, are too often prone to tolerating bad governments as long as those governments let them continue. This devilish contract between well-intentioned NGOs and oppressive regimes may facilitate smaller initiatives, but it will never bring the poor out of poverty.
Almost incidentally, Duflo and Banerjee concede that the shortest way out of poverty is to obtain a job in an industry. They’re right: industrialization is the path out of mass poverty, but only better economic policies can bring it about. Duflo and Banerjee’s feel-good, small-is-beautiful philosophy may prove, from the poor’s perspective, somewhat counterproductive.