My first job after college, in 1984, taught me to ask a question about foreign aid that I am still asking forty-two years later: Who really benefits? I was a young MBA working for Bangladesh's Master Plan Organisation (MPO), which was preparing a twenty-year national water plan. The project was undertaken with assistance from the United Nations Development Programme and the World Bank, with Harza Engineering Company International among the principal consultants. MPO was an attractive employer for young Bangladeshi professionals. I earned about 5,000 taka monthly when, as I recall, a newly minted MBA elsewhere might earn 1,500 to 2,000. Although MPO positions were temporary contracts without retirement benefits or job security, the project attracted capable local professionals at substantially lower cost than expatriate consultants.
Watching the work, I saw many tasks that Bangladeshi professionals could perform. Specialized foreign expertise certainly had value, but did that value justify its additional cost? My recollection is that foreign expertise was built into the assistance arrangements. Such restrictions were common. Under what became known as "tied aid," donors often required procurement from their own countries. A subsequent World Bank study identified restrictions affecting the use of domestic consultants and recommended relaxing them. The arrangement created a peculiar incentive: a government official accepting development assistance was not spending his own money. If millions of dollars arrived on condition that expensive foreign consultants be employed, accepting the package could be rational for the official even when local professionals could perform much of the work more cheaply. The donor financed a project, the recipient government obtained resources, and consultants received contracts. Taxpayers financing the arrangement had little influence over its design.
Years later, Texas Senator Phil Gramm gave me a framework for understanding this problem. Gramm frequently invoked Dicky Flatt, a printer and small businessman from Mexia, Texas. Before spending taxpayers' money, Gramm asked whether the benefits justified taking that money from someone like Flatt. During a 2000 Senate Banking Committee hearing, Gramm applied this test to the World Bank, IMF, and foreign assistance. Global poverty could justify substantial expenditures if the programs worked. The question was whether they worked well enough to justify their costs.
Apply that test to Bangladesh. The National Water Plan was not worthless. It assembled information, developed models, and provided a framework for addressing complicated water problems. But useful does not necessarily mean cost-effective. The relevant comparison was not foreign consultants versus no plan. It was extensive reliance on expatriates versus capable Bangladeshi professionals, supplemented by foreign specialists wherever their expertise justified the premium. World Bank studies subsequently identified a broader problem: reliance on foreign consultants could hinder the development of domestic professional capabilities. Bangladesh's experience raises precisely that concern. Development assistance should not merely complete today's project; ideally, it should leave behind professionals capable of undertaking tomorrow's.
This is not primarily an argument about corruption. Nobody has to steal anything for a development project to be economically wasteful. Every consultant can be competent, every invoice legitimate, and every procurement rule followed. Inefficiency can be embedded in the agreement itself. Development agencies frequently report how much money they spend, but spending is an input, not an accomplishment. What matters is the incremental benefit reaching intended recipients, its cost, and whether the same outcome could have been achieved more efficiently. I didn't know Dicky Flatt when I sat in that MPO office in 1984, but I was already asking his question: Was the additional benefit received by people in a poor country worth the additional cost taxpayers like Dicky Flatt were asked to bear?
