Ma06_Article 3 2/24/06 4:17 PM Page 96

Journal of Economic Literature Vol. XLIV (March 2006), pp. 96–105

The Big Push Déjà Vu: A Review of ’s The End of Poverty: Economic Possibilities for Our Time

∗ William Easterly

Jeffrey Sachs’s new book (The End of Poverty: Economic Possibilities for Our Time, Penguin Press: New York, 2005) advocates a “Big Push” featuring large increases in to finance a package of complementary investments in order to end world pover- ty. These recommendations are remarkably similar to those first made in the 1950s and 1960s in . Today, as then, the Big Push recommendation overlooks the unsolvable information and incentive problems facing any large-scale planning exercise. A more promising approach would be to design incentives for aid agents to implement interventions piecemeal whenever they deliver large benefits for the poor relative to costs.

1. Introduction fifteen years ago about shock therapy versus effrey Sachs is the economics profession’s gradualism in the ex-Communist countries. Jleading advocate of megareform. Whether It taps into a long-standing intellectual it is stabilization of hyperinflation in Bolivia, debate about social change in general, dat- shock therapy to leap from Communism to ing at least as far back as Edmund Burke’s capitalism in Poland and Russia, or a “Big critique of the French Revolution. Burke Push” to end world poverty, Sachs’s recom- preferred gradual reform to revolution, just mendation throughout his career has been to as in the twentieth century do it fast, do it big, do it comprehensively, argued for “piecemeal democratic reform” and do it with lots of Western money. as opposed to “utopian social engineering.” Sachs’s approach, as set out in vivid detail Earlier in the twentieth century, there was a 1 in his fascinating new book, raises an inter- debate about planning versus markets. esting debate for about the mer- This debate is particularly relevant for its of the “Big Push” versus incremental , where “utopian reform. This carries on the debate begun 1 A recent treatment of economic reform, in light of this intellectual tradition is John McMillan (2004), who discuss- es “shock therapy” in particular. McMillan is one of those ∗ Easterly: . who inspired me to go back and read Popper’s writings.

96 Ma06_Article 3 2/24/06 4:17 PM Page 97

Easterly: The Big Push Déjà Vu 97

social engineering” has attracted more intel- sound plausible: insecticide-treated bed lectual support than in other fields of eco- nets to protect children from malarial mos- nomics. Because the tragedy of the world’s quitoes, fertilizer for crops, and health clin- poorest peoples is so heartbreaking, an ics. Sachs believes you must do everything appropriately large response is more appeal- to accomplish anything: “These interven- ing in development economics than in more tions need to be applied systematically, dili- emotionally neutral areas like, say, the effi- gently, and jointly since they strongly ciency of the stock market. The Big Push reinforce one another” (p. 208) and “success appeals because it seems to promise a quick in any single area, whether in health, or end to the tragedy. Perhaps it is also seduc- education, or farm productivity, depends on tive to us as economists because it suggests investments across the board” (p. 256). our analysis and recommendations can have Who would actually implement this pack- a large effect. As Sachs says: “I have also age and how? This is really the critical part of gradually come to understand through my the book, and here Sachs seems to back social scientific research and on the ground adviso- engineering against Popper’s piecemeal dem- ry work the awesome power in our genera- ocratic reforms. The book says “We need tion’s hands to end the massive suffering of plans” (p. 227). Sachs many times seems to the extreme poor . . .” (p. 2). suggest that aid agency officials and national To unleash this power, Sachs proposes the government leaders should do planning from Big Push for poor countries. The principle the center to make everything happen. element in the Big Push would be a doubling Professor Sachs has subsequently objected to of foreign aid, to about $100 billion a year, the characterization of his arguments as top- then nearly doubling again by 2015. Foreign down planning, so I will give some quotes aid would fill a “financing gap” between what from his book that led many readers (includ- a country needs and what it can afford on its ing this reviewer) to this interpretation. In a own, making it possible for each of the poor- section titled: “Planning for Success”: est countries “to break out of the poverty trap and begin growing on its own” (p. 250). “The UN Secretary General…should oversee Another big element in ending the pover- the entire effort…the secretary general…should ensure that the global compact is put into oper- ty trap is a comprehensive package of inter- ation” (p. 269) ventions covering virtually all of the needs of poor people. This is because Sachs sees spe- Each country should have 5 plans, including an cific areas of underinvestment as particular- “Investment Plan, which shows the size, timing, ly critical, like insufficient spending on and costs of the required investments” (p. 273) health and soil fertility. Sachs identified these interventions as part of his work lead- “Getting from here to there is a matter of rou- tine planning, not heroics” (p. 274) ing the UN Millennium Project, which com- missioned ten task forces that mobilized 250 “Each low-income country should have the ben- experts. Sachs presents a fifty-four-item efit of a united and effective UN country team, checklist of barriers to development (p. 54) which coordinates in one place the work of the that must be overcome with the compre- UN specialized agencies, the IMF, and the hensive package. In a separate report, the . In each country, the UN country team should be led by a single UN resident UN Millennium Project listed 449 interven- coordinator, who reports to the United Nations tions that it recommended to end world Development Program, who in turn reports to poverty (UN Millennium Project, the UN secretary-general” (p. 285). Millennium Development Goals Needs Assessment). The book discusses many of The book clearly does not advocate that these interventions in detail, many of which any economy switch over to central planning Ma06_Article 3 2/24/06 4:17 PM Page 98

98 Journal of Economic Literature, Vol. XLIV (March 2006)

as an economic system; Sachs is obviously a campaigner as well as an , perhaps believer in free markets in general. Yet he we can indulge a slogan that “we” will be argues in the book that free markets do not able to “to heal a divided planet, to end the offer much hope for the poorest countries, suffering of those still trapped by poverty, and that a comprehensively planned set of and to forge a common bond of humanity, interventions is needed to get the poorest security, and shared purpose across cultures one-sixth of humanity out of poverty. and regions” (p. 3). However, Sachs reiterates the promises of 2. Sachs and Development Economics great results so frequently that it seems to be more than political rhetoric. “The end of How are we to place Sachs’s prescriptions poverty is at hand—within our generation— for ending world poverty into the develop- but only if we grasp the historic opportunity in ment economics literature? The first thing front of us” (p. 25). The “success in ending the that stands out is how familiar Sachs’s argu- poverty trap will be much easier than it ment sounds. He is restating almost exactly appears” (p. 289). Referring to the old the arguments of development economists Enlightenment dream of world peace, democ- made in the 1950s and 1960s, even using the racy, and prosperity, Sachs says “many of its same words. Then, as now, there were econ- sweetest fruits are within our reach” (p. 352). omists who advocated a Big Push to get countries out of a “poverty trap,” foreign aid 3. Big Solutions in Other Contexts to fill the “financing gap,” and action on all fronts through comprehensive “planning.” The book includes Sachs’s successful expe- To be sure, development economists at that riences in Bolivia and Poland in stabilizing time advocated far more state intervention high inflation through “shock therapy.” Here in the poor economies than Sachs does now. perhaps is a clue to the book’s confidence in Yet otherwise the concepts are remarkably the power of policymakers. You can solve similar between then and now. The 1960 high inflation with some top-down actions by counterpart to Sachs’s The End of Poverty policymakers—fix the exchange rate and was Walt Rostow’s best-selling The Stages of stop printing money. Unfortunately, the top- Economic Growth, which argued that coun- down solutions do not translate well into fix- tries could emerge out of stagnation into ing whole societies that are at one-sixtieth of self-sustained growth thanks to an aid- U.S. per capita income and with very differ- financed increase in investment. Over the ent institutions, social norms, and economic intervening forty-five years, many academic arrangements. development economists abandoned these The promise of a big solution to a very big ideas as simplistic, realizing that economic problem is an outlier in the practice of eco- development is a complicated interplay of nomics, where usually economists study markets (with many imperfections), politics, marginal changes to existing systems or poli- social norms, institutions, and government cies to generate marginal improvements. No policies, social services, and microeconomic serious economist that I know of is proposing interventions. Nevertheless, the idea of an a big plan to triple U.S. per capita income or aid-financed takeoff into growth has main- to end poverty in the . tained its appeal in the development policy However, there are arguments in econom- community. ics in favor of more comprehensive pack- The second thing Sachs has in common ages. The theorem of the second best says with the generation of fifty years ago is to that we cannot know whether a partial promise great results. Sachs certainly does movement toward a free-market equilibrium promise a lot. Since Sachs is a gifted political will be beneficial. Coordination problems Ma06_Article 3 2/24/06 4:17 PM Page 99

Easterly: The Big Push Déjà Vu 99

and models with multiple equilibria have ideas on how the poverty trap happens. I attracted a lot of interest from economists highlight three that are familiar, because over the past decades, although there has development economists first highlighted been much more theory than empirics. them almost half a century ago. The first is It does make sense that one policy change that poor people do not save enough: will only be effective if other policies are sat- isfactory—for example, privatization without When people are . . . utterly destitute, they need laws to enforce good corporate governance their entire income, or more, just to survive. and property rights could backfire. These There is no margin of income above survival that can be invested for the future. This is the main arguments were particularly popular after reason why the poorest of the poor are most the Berlin Wall fell and economists were prone to becoming trapped with low or negative called in to advise governments how to move economic growth rates. They are too poor to from Communism to capitalism, which was save for the future and thereby accumulate the when Sachs’s shock therapy became famous. capital that could pull them out of their current misery (pp. 56–57). Why was shock therapy in the ex- Communist countries so disappointing, with output collapses, high inflation, and political Sachs’s second reason for a poverty trap “is backlash? It turned out in the ex-Communist a demographic trap, when impoverished countries not to be feasible to do all the pos- families choose to have lots of children” (p. sible complementary reforms at once. There 65). Population growth is so high that it out- was a similar failure in developing countries paces saving (which was already too low, with the attempt by the IMF and World Bank according to the first reason). to implement comprehensive “structural The third element in the poverty trap is a adjustment.” nonconvexity in the production function. The relevant choice turned out to be Sachs suggests there are increasing returns between a big partial reform and a small par- to capital at low initial capital per person tial reform. The big partial reforms failed on (and low income per person): an equally big scale. Gradualism and incre- mental reform has more appeal now after An economy with twice the capital stock per per- son means an economy with roads that work the the chastening experience of “transition eco- year round, rather than roads that are washed out nomics” and “structural adjustment,” as well each rainy season; electrical power that is reliable as the consistent failure of Big Push eco- twenty-four hours each day, rather than electric nomics in foreign aid to poor countries. power that is sporadic and unpredictable; work- Why are comprehensive big reforms not ers who are healthy and at their jobs, rather than workers who are chronically absent with disease. advocated in rich countries? In rich coun- The likelihood is that doubling the human and tries, the small partial reforms have been the physical capital stock will actually more than dou- source of progress over time, and econo- ble the income level, at least at very low levels of mists studying rich countries sensibly stick to capital per person (p. 250). such marginal improvements rather than big reforms. Sachs gives the example of a road with half of the road paved and half impassable due to 4. Sachs’s Theory of Aid and the missing bridges or washed out sections. Poverty Trap Repairing the impassable sections would double the length of road, but would much Nevertheless, Sachs argues forcefully that more than double the output from the road. the poor countries badly need a Big Push. “This is an example of a threshold effect, in The intellectual foundation for this is the which the capital stock becomes useful only idea of the “poverty trap.” Sachs has many when it meets a minimum standard” (p. 250). Ma06_Article 3 2/24/06 4:17 PM Page 100

100 Journal of Economic Literature, Vol. XLIV (March 2006)

These three mechanisms have been countries, but their growth rate has fallen prominent in the development policy com- over time. munity because they seem to create a simple The evidence that Jeffrey Sachs adduces way in which aid can break the poverty trap. for the poverty trap in The End of Poverty is The role of foreign aid is to increase the cap- from the period since 1980. This is indeed ital stock enough to cross the threshold level: the period where the stylized facts are most “if the foreign assistance is substantial consistent with a poverty trap, as the poorest enough, and lasts long enough, the capital countries had per capita growth around zero, stock rises sufficiently to lift households in contrast to positive growth in richer coun- above subsistence . . . . Growth becomes tries. However, the poorest countries did self-sustaining through household savings have significant positive growth 1950–80. To and public investments supported by taxa- make things worse, the poorest countries tion of households” (p. 246). were getting more in foreign aid as a percent The poverty trap story is theoretically of their income in the last twenty years, com- intriguing and may well give insights into pared to the previous period. Foreign aid is poor development outcomes. There is a supposed to be helping the poor countries large literature discussing evidence for escape from the poverty trap; hence the poverty traps (see the excellent survey by poorest countries in the recent decade Costas Azariadis and John Stachurski 2004, should have been LESS likely to be stuck in who point out the difficulty of drawing poverty than the previous decades with unambiguous conclusions about the exis- lower foreign aid. In sum, the evidence for tence of poverty traps).2 I concentrate more an aid-induced Great Escape from poverty is narrowly on whether aid has had the effects less than overwhelming. that the Sachs poverty trap model would predict. 5. Poverty Traps versus Bad Government The main prediction is that aid would have large and positive growth effects in Sachs argues that it is the poverty trap poor countries. Decades of research on aid rather than bad government that explains and growth has failed to generate evidence poor growth of low income countries over for this prediction (I summarize some of this recent periods. For example, Sachs says “the research in William Easterly 2003). The big claim that Africa’s corruption is the basic stylized facts certainly do not support the source of the problem {the poverty trap} prediction that aid has big growth effects, does not withstand practical experience or helping countries to escape from poverty serious scrutiny”(p. 191). traps: (1) aid has been highest as a percent Yetalarge literature argues that bad insti- of income in Africa, but African growth is tutions and policies ARE a part of the prob- lowest of any continent, (2) aid has risen lem of low growth. Let us do a quick test of over time as a percent of income in poor bad government against the poverty trap as a story for poor economic growth. The earli- est rating we have on corruption is from 2 Aart Kraay and Claudio Raddatz (2005) have recently tested directly whether the savings and increasing returns 1984, from the International Country Risk mechanisms hold in the data. They point out that saving Guide. We have a rating on democracy for would have to follow an S-curve to generate a poverty the same year from a research project at the trap, first increasing little with income, then increasing steeply, then flattening out again. They reject the S-curve University of Maryland called Polity IV. Let’s in the data on saving and per capita income. They also fail take countries that have the worst ratings on to find evidence of the technological nonconvexities that both corruption and democracy, and call also are necessary to create the poverty trap. They con- clude there is little evidence for a poverty trap based on these countries “bad governments.” While these mechanisms. poor countries did worse, it’s also true that Ma06_Article 3 2/24/06 4:17 PM Page 101

Easterly: The Big Push Déjà Vu 101

the twenty-four countries with bad govern- causality between corruption and income is ments in 1984 had significantly lower still hotly debated, but it seems foolhardy to growth 1985 to the present: 1.3 percentage simply dismiss corruption as a factor points slower than the rest. There is some inhibiting development. overlap between these two stories, as poor If corruption and other forms of toxic pol- countries are much more likely to have bad itics are a big part of the problem, then the government. So which is it, bad government solutions are going to be a lot more compli- or the poverty trap? When we control for cated than the top-down Big Push. We need both initial poverty and bad government, it to do serious to get at the is bad government that explains the slower roots of poverty. We need to understand the growth. The coefficient on initial per capita actions and incentives facing very many is not significantly different from zero, once actors. The response of aid donors is also of we control for bad government. This is still interest, as and Beatrice true if we limit the definition of bad govern- Weder (2002) show that aid does not dis- ment to corruption alone. The recent stag- criminate against more corrupt countries. nation of the poorest countries appears to Jakob Svensson (2003) finds that aid actually have more to do with awful government increases corruption in ethnically divided than with a poverty trap, contrary to the societies. Sachs hypothesis. Sachs is very insistent that bad govern- 6. Development as a Technical Problem? ment is not the problem in poor countries. Another striking thing about Sachs’s pres- This is an important part of his case for entation of the causes of poverty and the increased aid, in that he imagines skeptics possible solutions is how little economics asking “If the poor are poor because . . . their seems to be invoked in his analysis. We hear governments are corrupt, how could global very little about incentives to invest and cooperation help?” (p. 226). He wants to innovate, or to honor contracts and property reassure potential aid donors that “the rights, or to do political and economic poor . . . are also ready to govern themselves reforms, for example, while we hear a lot responsibly, ensuring that any help that they about geography, agronomy, and medicine. receive is used for the benefits of the group Poverty to Sachs seems to be a technical rather than pocketed by powerful individu- problem that can be fixed with interventions als” (p. 242). Even where he sees some cor- from the natural sciences. He gives medi- ruption, he thinks more aid can solve the cine as not just a solution in itself to many problem “Part of the corruption {in Kenya} problems of the poor, but a model for all of is new and completely avoidable, but only if development economics: donors help Kenya to improve the function- ing of the public administration . . . by the Development economics today is not like mod- installation of computer systems, published ern medicine, but it should strive to be so. It can accounts, job training and upgrading, higher improve dramatically if development econo- pay for senior managers (p. 237). mists take on some of the key lessons of modern medicine, both in the development of the Unfortunately, there is an extremely underlying science and in the systematization of strong correlation between per capita clinical practice, the point where science is income and corruption. Sachs classified brought to bear on a particular patient (p. 75). countries as relatively well-governed if their corruption is low for their level of income, This emphasis on scientific and technical but this does not address the possibility that solutions to problems runs throughout the their income is low BECAUSE they are cor- book: “Household spraying, insecticide- rupt, as a large literature has argued. The treated bed nets, and antimalaria medicines Ma06_Article 3 2/24/06 4:17 PM Page 102

102 Journal of Economic Literature, Vol. XLIV (March 2006)

all work in Africa just as they do in other incentives the international administrators parts of the world” (p. 199). “Africa’s prob- and government civil servants will face to lems . . . are . . . solvable with practical and deliver and implement the foreign aid inter- proven technologies” (p. 208). ventions that Sachs proposes. Decades of This portrayal of poverty as just a techni- field experience amongst development prac- cal problem is at odds with the rich litera- titioners suggest that implementation is very ture in economics and political science that difficult and incentives can be very much stresses the social causes of poverty—bad askew. Dean Filmer, Jeffrey S. Hammer, and institutions, bad politics, misguided policies, Lant H. Pritchett (2000) cite studies in trading networks that exclude the poor, high Guinea, Cameroon, Uganda, and Tanzania, transaction costs in markets, and ineffectual that estimated that 30 to 70 percent of gov- aid donors. Sachs does not clearly link his ernment drugs disappeared before reaching case for public action to either unleashing the patients. In one low-income country, a incentives to produce private goods or the crusading journalist accused the Ministry of need to provide public goods. He seems just Health of misappropriating $50 million in as eager to intervene when the problem is a aid funds. The Ministry issued a rebuttal: the missing private good as when there is a journalist had irresponsibly implied the $50 missing public good. million went AWOL in a single year, where- The view that poverty is just a technical as they had actually misappropriated the $50 problem leads Sachs to administrative solu- million over a three-year period. In a survey tions to poverty. Much literature over the at government health centers in the Mutasa years have pointed out the problems with district of Tanzania, new mothers reported administrative plans to solve economic prob- what they least liked about their birthing lems. There are at least three major classes experiences assisted by government nurses. of problems: The poor mothers-to-be were “ridiculed by (1) Plans can be decreed at the top, but nurses for not having baby clothes (22 per- must be implemented at the bottom— cent) . . . and nurses hit mothers during how to design principal–agent contracts delivery (13 percent).” More systematically, to give good incentives for aid agency Filmer, Hammer, and Pritchett find no cor- officials and government civil servants relation between public health spending and to do the interventions right and get health outcomes. results for the poor? This problem is Actually, we can turn to Sachs’s own UN particularly severe in foreign aid, when Millennium Project report for reports of the actions at the bottom are often weak incentives to deliver results. unobservable to the administrators at Cambodian teachers “supplement their the top. income by soliciting bribes from students, (2) Administrators at the top do not have including the sale of examination questions enough information about the realities and answers . . . the end result is a high at the bottom to design the right inter- dropout rate.” Even among less venal civil ventions in the right place at the right servants, we hear that in Uganda “sanitation time. This problem is particularly has been highlighted as a high priority in the severe in foreign aid, which gets little sector plans of water, health, and education. or no feedback from the voiceless poor. Practically however, it is yet to clearly come (3) Multiple goals and multiple agents out as a funding priority for any of them. In weaken incentives for agents to deliver fact, by spreading out everywhere, it has on goals. received no attention anywhere!” As far as Sachs does not address these problems. international and rich country aid agencies, We hear nothing in the book about what accountability is so weak that foreign aid “is Ma06_Article 3 2/24/06 4:17 PM Page 103

Easterly: The Big Push Déjà Vu 103

not evaluated or documented systematically for the agent, since each principal try to get for results.” the agent to work on her goal and not on the The Millennium Project is also a source of others. Together, the principals tend to cancel information about the lack of information fac- each other out, so the agent is at best con- ing planners. In the project’s document fused and demoralized, or at worst feels he where it tries to estimate the costs of all the can get away with doing little on any of the 449 interventions that it recommends doing goals. When there are multiple agents, we to end world poverty (UN Millennium have all the usual problems of collective Project 2005), it ran into some gaps: action. Each agent hopes the others will • “it is extremely difficult to predict the expend effort, lessening the need for him to direction, let alone the magnitude, of undertake costly effort. There is no way to the net impact of extending the cov- hold any one agent accountable for action erage of interventions on marginal when everyone is collectively responsible. costs . . . there are only very few suc- One solution to this problem is to have a cessful examples of large-scale scaling very powerful principal at the center assign- up of interventions in low-income ing goals to actors. Sachs seems to lean in countries.” this direction when he calls for the UN • In Uganda, “current coverage levels for Secretary General to coordinate the whole antenatal care are difficult to estimate effort to end poverty, but the United Nations due to a lack of data and definitional does not have political power over rich coun- problems.” try governments, poor country governments, • In Cambodia, lack of data meant esti- or even the World Bank and IMF, and there mates of energy interventions were is no prospect of the United Nations getting based on Bangladesh data. There was such power (thank goodness). Sachs does no data on Cambodian power generat- not link this recommendation to any analysis ing capacity or usage, so all incremen- of incentives facing aid actors. Anyway, a tal demand was assumed to require powerful principal at the center would run new capacity. into problems (1) and (2). • In Tanzania, “our results show that Tanzania needs at least $1.1 billion 7. The Piecemeal Approach over 2005–15 to attempt to meet the gender goal. We recognize that gender What is the alternative to the Big Push? data is it a large extent nonexistent, and Instead of setting utopian goals of ending that therefore these numbers probably world poverty, foreign aid could just con- underestimate the true cost of meeting centrate on finding particular interventions the gender MDG.” that work and keeping those interventions • There are forty target areas where going. Anecdotal evidence suggests piece- action is desirable but the project was meal approaches to aid are more successful. unable to come up with a policy inter- A vaccination campaign in southern Africa vention. The intervention is either list- virtually eliminated measles as a threat to ed as “to be specified further” or there kill children. Routine childhood immuniza- is no intervention listed at all and it is tion combined with measles vaccination in “to be specified” in the future. seven southern African nations cut measles The third problem with administrative cases from 60,000 in 1996 to 117 in 2000. action is when there are multiple goals and Donors collaborated on a program to wipe multiple actors. Principal–agent theory says out river blindness in West Africa starting in that having multiple principals (equivalent 1974, virtually halting the transmission of here to multiple goals) weakens the incentives the disease. Eighteen million children in the Ma06_Article 3 2/24/06 4:17 PM Page 104

104 Journal of Economic Literature, Vol. XLIV (March 2006)

twenty-country area of the program have continued disappointing economic growth), been kept safe from river blindness since not to mention the successes in East Asia, give the program began. Another partnership signs of hope for homegrown development. amongst aid donors contributed to the near eradication of guinea worm in twenty 8. The Dangers of Hubris African and Asian countries where it was Sachs writes eloquently about the needs of endemic (Levine et al. 2004). the poor, and that is enough by itself to read More formally, Abhijit V. Banerjee and this book. Some of his descriptions of sim- Ruimin He (2004) list numerous piecemeal ple interventions that could make the poor interventions that researchers have con- better off are plausible, and indeed could firmed through scientific evaluation to be form the basis for piecemeal reform. successful (at least in some times and However, Sachs’s proposed Big Push places): should set off alarm bells about the dangers • Subsidies to families for education and of hubris in economics. Indeed the descrip- health inputs for their children (such tion of how cheap and simple are the inter- as PROGRESA in Mexico) ventions that Sachs describes as not done •Reducing class size, remedial teach- yet is an indictment of the past perform- ing, uniforms and textbooks, school ance of Big Pushes in that they failed to vouchers find these interventions. The failures of • Deworming drugs and nutritional sup- past Big Pushes in foreign aid, the failure of plements shock therapy in the ex-Communist coun- •Vaccination, HIV prevention, STD tries, and the failure of IMF/World Bank management, Indoor Spraying for structural adjustment almost everywhere Malaria, Bednets suggests that megareforms usually do not •Fertilizer work out well. • Clean water I close with words of wisdom from two Of course, getting or keeping piecemeal scholars who wrote about the dangers of approaches that work well would require hubris in social change. Karl Popper wrote improving incentives for aid agencies. in 1957: Improved incentives might come from put- ting much more emphasis on independent The piecemeal engineer knows, like Socrates, evaluation of aid projects. Better incentives how little he knows. He knows that we can learn could come from devising means to get only from our mistakes. Accordingly, he will more feedback from the intended benefici- make his way, step by step, carefully comparing aries, holding aid agencies accountable the results expected with the results achieved, and always on the look-out for the unavoidable when the feedback is negative. It seems unwanted consequences of any reform; and he more productive to focus on such critical will avoid undertaking reforms of a complexity incentive problems in foreign aid rather than and scope which makes it impossible for him to promise the rich country public the end of disentangle causes and effects, and to know world poverty. what he is really doing . . . . Holistic or Utopian social engineering, as opposed to piecemeal If the Big Push will not generate develop- social engineering . . . aims at remodeling the ment, are things hopeless for poor coun- “whole of society” in accordance with a definite tries? Fortunately, the poor countries are plan or blueprint (Popper 2002, p.61). making much progress on their own, with- out waiting for the West to give them a Yale anthropologist and political scientist push. The market-driven growth of China James C. Scott has a brilliant review of top- and India, the movement toward democracy down plans in his 1998 classic Seeing Like a in Africa and Latin America (even amidst State. He sums up the implications of his Ma06_Article 3 2/24/06 4:17 PM Page 105

Easterly: The Big Push Déjà Vu 105

review for policy making: “In an experimen- Banerjee, Abhijit V., and Ruimin He. 2004. “Making Aid Work.” MIT. Mimeo. tal approach to social change, presume that Easterly, William. 2003. “Can Foreign Aid Buy we cannot know the consequences of our Growth?” Journal of Economic Perspectives, 17(3): interventions in advance. Given this postu- 23–48. Filmer, Deon, Jeffrey S. Hammer, and Lant H. late of ignorance, prefer wherever possible Pritchett. 2000. “Weak Links in the Chain: A to take a small step, stand back, observe, and Diagnosis of Health Policy in Poor Countries.” then plan the next small move” (p. 345). World Bank Research Observer, 15(2): 199–224. Levine, Ruth, and the What Works Working Group Small moves can cumulate into bigger with Molly Kinder. 2004. Millions Saved: Proven benefits. Perhaps the rich country public Successes in Global Health. Washington, D.C.: will be more willing to increase aid when Center for Global Development. Kraay, Aart, and Claudio Raddatz. 2005. Poverty Traps, they see aid projects working well according Aid, and Growth. World Bank. Mimeo. to independent evaluation and feedback McMillan, John. 2004. “Avoid Hubris And Other from the poor, as opposed to realizing once Lessons for Reformers.” Stanford University Graduate School of Business. Mimeo. again that Big Pushes do not deliver on their Popper, Karl. 2002. The Poverty of Historicism. New utopian goals. York: Routledge Classics. First published 1957. Scott, James C. 1998. Seeing Like a State: How Certain REFERENCES Schemes to Improve the Human Condition Have Failed. New Haven: Yale University Press. Alesina, Alberto, and Beatrice Weder. 2002. “Do Svensson, Jakob. 2000. “Foreign Aid and Rent- Corrupt Governments Receive Less Foreign Aid?” Seeking.” Journal of International Economics, 51(2): American Economic Review, 92(4): 1126–37. 437–61. Azariadis, Costas, and John Stachurski. Forthcoming. UN Millennium Project. 2005. Investing in “Poverty Traps?” in The Handbook of Economic Development: A Practical Plan to Achieve the Growth, Vol. 1A. Philippe Aghion and Steven Millennium Development Goals, Main Report and Durlauf, eds. MDG Needs Assessment. New York: United Nations.