Getting results from our aid: Rethinking what we measure
26 May 2009
In the midst of an economic crisis, US policy makers are more prone to ask, "Are we actually getting results from our foreign aid?"
In the midst of an economic crisis, US policy makers are more prone to ask, "Are we actually getting results from our foreign aid?" Taxpayers and policy makers have a right to know how tax dollars are being spent, and on the receiving end, poor people have a right to know about aid spent in their name. The public perception that our development aid is not producing concrete results is actually a result of not measuring our aid for the results that matter.
What is aid supposed to accomplish?
To measure the success of any program, you need an idea of what it was intended to accomplish. One reason for the perception that our development aid is not producing concrete results is the tension that exists between two key purposes of our aid:
1. Is US development aid about delivering goods and services?
2. Or is US foreign aid about helping citizens and countries get to a place where they can provide those services without our help?
What's the unit of measurement?
If the purpose of US development aid is to deliver goods and services, the way to measure our progress is to provide the inputs and then count the outputs (for example, the number of farmers trained, the number of bed nets disbursed).
But if the purpose of US development aid is to create lasting solutions to poverty, then what we measure looks very different: did the farmers who received the training consistently produce enough food to feed their families year after year? Did the families who received the bed nets use them properly, and did the bed nets keep mothers and their children from getting malaria not just this year, but five years from now? More fundamentally, are citizens and the government better equipped to provide these services on their own as a result of US aid?
Room for long-term goals, too
To alleviate suffering and produce immediate results in a pandemic or crisis, US aid agencies must engage in treating the symptoms of poverty. But in order to help prevent the next pandemic or crisis from taking such a high human toll, US aid agencies must also work toward treating the root causes of poverty. In recent years, pressure on the agencies to produce immediate and tangible results has led to a higher percentage of programs that deliver goods and services that can be easily measured. For US development aid to be most effective in putting itself out of business, balance must be restored in favor of programs with longer time horizons but more potential to help create long-lasting solutions to global poverty.
Why aren't we measuring for results now?
1. Not everything that counts can be counted easily.
Problem: Helping to build effective states and support active citizens takes time, and U.S. aid agencies can't claim sole responsibility for progress; they can only provide homegrown groups with the resources they need to do it themselves.
Solution: Congress and the Administration should provide more multiyear funding of development programs (as they do through the Millennium Challenge Account) and in countries elsewhere where governments and citizen watchdogs continue to meet yearly benchmarks. Progress should be measured in improvements in social indicators, and government performance in providing for its citizens.
2. Implementers are stuck reporting the wrong stuff.
Problem: Strenuous reporting requirements from a host of different US government agencies can actually impede getting results from our aid, because implementers in the field spend their time filing reports and engaging in complicated and not always useful accounting exercises, which leaves them less time to assess whether the project is actually making a difference in the long term.
Solution: Congress and the Administration should work to reduce excessive and duplicative reporting requirements, and they should hold aid agencies accountable not just for numbers of goods and services provided, but for whether those projects actually reduce poverty.
3. Implementers have a fear of failure.
Problem: Perhaps the biggest obstacle to stronger evaluation is implementers' fear that if they prove that their program did not produce the intended results, their funding will be cut off midstream or they will be penalized without getting a chance to correct early mistakes.
Solution: Congress and the Administration should reward innovation and accept the risks it entails. Give aid agencies incentives that reward creativity, and give them the flexibility to throw out approaches that fail without facing penalties.
How can policy makers get better results from US aid?
Measuring results is important, and equally important is measuring the right kind of results. Measuring only inputs and outputs fails to deliver on the real bottom line: did US foreign aid succeed in helping families escape poverty? Instead of spending their time bean-counting and filling out reports, US aid agencies should be given the space to measure whether they are producing long-term reductions in poverty. Mandating that US aid agencies focus on long-term development, and providing incentives to aid agencies to create a culture of learning and innovation, will go a long way toward getting the kind of results that both the American taxpayers and the rice farmer in Cambodia deserve.
