Wednesday, May 21, 2008

Easterly's Reinventing Foreign Aid

I could not hold on reading the first chapter of William Easterly’s new book Reinventing Foreign Aid. As usual, he blasts Jeffrey Sachs, the MDG and similar projects, and gets irritated with the star-studded aid campaigns for their ignorance to a completely faulty aid bureaucracy. He also draws attention to lack of accountability, transparency, decentralization, and one-sided imposition of priorities on behalf of aid recipients by large aid donors like WB, IMF, PRSP programs, UNDP, OECD, G8, etc.

As in previous book, he explains the planners and searchers dichotomy in foreign aid and argues that the planning mentality of big aid agencies foreshadows the priorities of aid recipients, often leading to outcomes that are primary concerns of aid donors before releasing money. He also argues that arguments like aid works for economic development are nontestable and nonfalsifiable, leading to so much of inconclusive debate on the efficacy aid despite emerging evidences show that highest aid recipients are the ones that grew least.

He kind of cherry picks quotes from reports released by aid agencies like WB, IMF, DFID, and UN, among others and blasts them for contradictory statements and analysis. Easterly’s main arguments are like this: “Aid does not work. Either overhaul and restructure it or stop it. Let searchers find their own priorities and destinies. Enhance aid bureaucracy’s quality and efficacy based on the priorities laid out by the aid recipients.” He repeatedly (and rightly) blasts the aid agencies for being like planners (top-down approach), putting their priorities before the priorities of the aid recipients.

Everything sounds good and he more or less argues vociferously against planning mentality of the aid agencies. He also summarizes core of each chapters written by a wide range of authors. However, he explains more about those chapters that argue against the current aid structure and, more broadly, aid intervention. He explains less about chapters written by Banerjee and He and by Duflo and Kremer who argue that aid intervention can be made more effective by conducting randomized controlled trials (RCTs) of its efficacy.

Here are some of Easterly’s main points:

[…]“Nevertheless, the aid agencies often seem to have in mind the kind of engineering problem that a dam poses when designing Planning solutions to the problems of poverty. They seem to assume a Leontief production function between aid inputs and development outcomes that lends itself to detailed planning (and makes it possible to come up with precise estimates for costs of attaining plan targets): ‘‘The starting point is for donors and aid recipients to agree on a financial needs assessment that identifies the aid requirements for achieving the MDGs. Donors then need to provide predictable, multiyear funding to cover these requirements, and developing countries need to implement the reforms that will optimize returns to aid.”

[…]Because of the insistence on working through governments, aid funds get lost in patronage-swollen national health bureaucracies, not to mention international health bureaucracies. In countries where corruption is as endemic as any other disease, health officials often sell aid-financed drugs on the black market. Studies in Guinea, Cameroon, Uganda, and Tanzania estimated that 30 to 70 percent of government drugs disappeared before reaching any patients. In one low-income country, a crusading journalist accused the Ministry of Health of misappropriating $50 million in aid funds. The ministry issued an astonishing rebuttal: the journalist had irresponsibly implied the $50 million went AWOL in a single year, whereas they had actually misappropriated the $50 million over a three year period.

[…]In the domestic politics of democracies, the people who vote are the same ones who receive the services. In foreign aid, this feedback and accountability loop is broken: the rich people who give the money or vote for foreign aid are not the ones receiving aid services. The poor have no way of registering their satisfaction or dissatisfaction with aid services by how they spend or how they vote. The bottom line is that aid agencies have more of an incentive to please the rich than the poor.

[…]The comprehensive ambitions of the planners have misfired badly, crowding out more sensible and pragmatic approaches that are humble about their own limitations. The world’s poor will mostly determine their own fate by their own home-grown institutions and initiatives, as much historical and contemporary evidence suggests.

Here is Easterly about searchers and planners:

[…]The UN Millennium Project also suffers from the first problem: that planners do not really know the precise technology that translates inputs into outputs. The participants in the Millennium Project themselves know this obvious point—‘‘it is often difficult to precisely quantify the link between coverage of interventions and MDG outcomes’’—yet insist in the same sentence that somehow ‘‘national MDG planning involves mapping interventions to MDG outcomes.

I think there is something wrong with his extreme Hayekian views (see this as well) on “searchers” and “planners” because left to themselves, searchers would not even imagine to realize their own potential! Even if they do, it might take years to realize this (sadly, the poor people do not have the luxury of waiting for the right searchers for years!). Obviously, there is no doubt that the current aid structure is faulty but this does not mean that aid intervention does not work at all. Very selective aid intervention does work. For instance, Yunus (one of Easterly’s celebrated searchers) would not have been able to attend Vanderbilt University had he not received some form of financial aid and scholarship. Moreover, Yunus might not have been qualified to apply for higher education had he not attended primary and secondary school, funded by donors, in Bangladesh. Selective aid in education sector helps accelerate the emergency of searchers Easterly is talking about. Aid in education that is aimed at enhancing human capital surely works. What does not work is aid that is channeled through corrupt leaders, dictators, warlords, and those backed by special interest (especially business and corporate sector). Africa received this form of aid more than the workable form of aid. Selective intervention (not a wholesale one) in aid does work and it helps produce the searchers Easterly talks about.

Easterly also argues against “poverty traps” and doubts its very existence. He is going a bit far here because though there are doubts about countries embroiled in poverty traps, there should not be any doubt that individual households (a substantial number of them) in a poor, resource-stricken community could be in poverty traps arising from low education, lack of credit, poor healthcare, landlessness, and risk and vulnerability, among others.

Traps arising from different causes do exist but the question is whether it can be broken through aid and expert advice? Sometimes yes, sometimes no. Traps can be broken by selective intervention, a piecemeal intervention approach followed after diagnosing the causes of poverty and growth. Selective intervention has high feedback and accountability and is a piecemeal intervention approach based on recipient priorities and needs.

It can also be thought of as “little-P planning” that Easterly is talking about. Relying on searchers through repeated trial and error would be too expensive in terms of time and resources needed to create them. Through selective intervention, states can create an environment where Easterly’s searchers (“firms in private markets and democratically accountable politicians”) can emerge faster than it would take naturally (through trail-and-error experimentation). This is needed because private markets are always not efficient due to coordination failures and spillover effects, and democratically accountable politicians are hard to come by in the near future, at least in the politically messed up African continent. States can facilitate the rise of searchers by instituting reforms to create customized institutions that are consistent with history, culture, local capacity, resources, politics, socio-economic setting, and technology.

Easterly is, however, supportive of randomized trials to evaluate aid interventions but rues that this method is sidelined by aid agencies:

[…] Duflo and Kremer in chapter 3 discuss the methodology of randomized controlled trials (RCTs) to evaluate aid interventions. They argue, ‘‘There is scope for considerably expanding their use, although they must necessarily remain a small fraction of all evaluations.’’ The RCT is a welcome introduction of the scientific method into foreign aid and development, an area where wishful thinking, politically motivated conclusions, and pseudoscience have perhaps been more predominant than in other areas of economics. The RCTs are not a panacea, and they are not applicable to all areas of foreign aid and development, but they have already made a great contribution to the field of economic development.

Easterly’s suggestion: Easterly’s solution is to make the aid recipient masters and the aid agencies as quality service delivers:

[…] Having multiple searches for what works may sound like a lot to a planner, who thinks in terms of a top-down bureaucratic hierarchy. However, the great thing about searching is that it can be totally decentralized. A myriad of searchers are available in the field to look for what works for each piece of the puzzle. The aid system just has to be designed so that it rewards successful searches and scales them up to achieve widespread benefits for the poor.

[…] Another market-oriented step would be for the common pool [of aid money] to issue vouchers to poor individuals or communities, who could exchange them for development services at any aid agency, NGO, or domestic government agency. These service providers would in turn redeem the vouchers for cash out of the common pool. Aid agencies would be forced to compete to attract aid vouchers (and thus money) for their budgets.