Finishing What Wolfowitz Started

By Sebastian Mallaby (THE WASHINGTON POST, 18/05/07):

Now that Paul Wolfowitz has agreed to resign, his successor at the World Bank will have to grapple with his signature issue: the fight against corruption. Wolfowitz mishandled this challenge so badly that it poisoned his tenure, and the bank's next president will be tempted to avoid it. But the challenge of corruption and, more broadly, of weak institutions in developing nations must not be neglected.

When Wolfowitz took the helm of the World Bank in 2005, everyone agreed that institution-building was essential. The "Washington Consensus" phase of development thinking, which held that pro-market policies would lift countries out of poverty, had given way in the mid-1990s to a new view: The policies you adopt can matter less than the skill with which you implement them. Russia had embraced big-bang market reforms but collapsed in a corrupt mess. China cleaved to an unorthodox mix of statism and markets, but skillful implementation produced an economic miracle.

Three prescriptions followed. If success depends on implementation, it's essential to get the implementers' enthusiastic commitment to whatever policy is chosen, so aid agencies began to restrain their tendency to dictate policies. Development loans ceased to be loaded up with hundreds of conditions, and the new buzzword was " ownership." Poor nations would henceforth choose their own development priorities. To cite another cliche of the time, they would be "in the driver's seat."

If poor countries get the keys to the car, some will drive toward the cliff; aid for these states will be wasted. So the next prescription to emerge in the late 1990s was "selectivity": Aid must be targeted at developing nations that are driving in a smart direction. The World Bank and other agencies began to concentrate their money more, and in 2002 President Bush unveiled the Millennium Challenge Account, which reserved all its cash for a handful of countries with good policies.

Ownership and selectivity are fine, but what about the billion or so poor people who live in countries with bad drivers? Abandoning them is unacceptable politically as well as morally. The public's support for aid is based on the expectation that it will reach the poorest of the poor -- and that it will help to stabilize misgoverned, fragile states that threaten Western interests.

Hence the third 1990s prescription: If good implementation is essential to development, the aid business has to strengthen the institutions in poor states that make good implementation possible. If education projects fail in Pakistan because the money is stolen, create a prosecutor's office to go after corruption. If the private sector is stifled in Zambia because government officials are on the take, roll up your sleeves and reform the civil service.

This cry for institution-building turned out to have big consequences. After the terrorist attacks of 2001, it crossed from the aid folk into the foreign policy community, stoking an appetite for nation-building that was at once liberal and neoconservative. Creating robust institutions was suddenly part of the battle against Third World poverty and part of the battle for First World security, since broken institutions created terrorist havens in poor countries.

Now this enthusiasm for creating institutions is gone. The failure of nation-building in Afghanistan and Iraq began this change; the failure of Wolfowitz at the World Bank threatens to complete it. Wolfowitz marched into the corruption battle without planning for the postwar; he declared the issue a priority without explaining what he would do about it. He never grappled with the verdict of the bank's Internal Evaluation Group: In a recent study of institution-building loans in 35 countries, the evaluators found that corruption receded in just one of them.

The lesson for the next World Bank president is that corruption must be attacked subtly. The bank's ambitious institution-building efforts have generally failed for the same reason old-style aid conditions didn't work: Donors don't have the clout to force change; reform must be "owned" locally. But the bank has succeeded better with indirect approaches. An African government is unlikely to come to the bank, confess its corruption and request help in fixing it. But it might ask for assistance with urban water, at which point the bank can explain that, along with extra reservoirs and pipes, success requires greater transparency in water management.

The bank can also do a lot to stoke demand for change in poor countries. Training investigative journalists gets corruption out into the open, and media revelations create more pressure to root out graft than hectoring from foreign donors. Getting facts into the hands of citizens can have a tremendous effect, too. If villagers know how much government money their school is meant to get, they'll pressure school administrators to steal less of it. This demand-side approach has worked wonders in Uganda, transforming the performance of both schools and clinics.

The neoconservative moment has passed. For good or possibly for ill, the appetite for ambitious nation-building has evaporated. But there are excellent reasons that development experts made fighting corruption a priority 10 years ago: If you want to make a difference in dysfunctional poor states, there is no way to avoid it. Wolfowitz's successor can make a difference by backing modest and practical corruption-fighting measures. He or she should not run from the agenda.