Are pay-for-performance aid programs such as Cash on Delivery Aid more vulnerable to corruption than traditional input-focused programs? My guests this week, senior fellows William Savedoff and Charles Kenny, argue in a new new working paper and brief that the opposite is true.
Bill says that input-tracking programs –those that attempt to track how the money was spent rather than paying based on previously agreed outcomes—favor the dishonest.
“It’s much easier for the dishonest person to divert money and not do any work, than for the honest person who’s trying get their job done and then also report on all this,” he says.
Bill explains with a simple story that I like to call the parable of the bean counter.
“You give somebody beans, and say, ‘We want you to account for all these.’ The person produces records showing that they planted all the beans. “
“When nothing grows, they have plenty of excuses for why inputs didn’t turn into outputs. They can say, ‘It was a bad year for rain,’ or ‘the birds pulled them up.’ The point is, all they had to do was show the records and they got the money.”
An honest bean grower would have to plant and tend the beans as well as keeping records. Since results don’t matter for payment, it’s easier to be dishonest than honest.
However, with a results-based approach, Bill explains, the honest bean grower has only one task (grow beans!) while the dishonest bean grower has a double job, growing beans even while finding ways to skim something off the top.
It all depends on which beans the bean counter counts: inputs or outputs.
Charles offers a real-life example citing Justin Sandefur’s essay on a little-known but highly successful USAID health program in Afghanistan. The program, he says “has saved thousands upon thousands of people’s lives…at an incredibly low cost. But because we don’t have all the receipts in a nice little book, SIGAR [the Special Inspector General for Afghanistan Reconstruction] is shutting it down.”
Charles’ example shows just how badly things can go wrong when a system prioritizes receipts over results. Payment for performance, he says, can put things right.
“Get the bureaucracy to care about what the taxpayer cares about. Measure what they care about, and make payment based on the outcome you are trying to achieve. Then corruption cannot stop your programs working,” he says.
Charles is adamant that we still need bean counters. “We just need them counting the beans that we want delivered, not the beans we put into the process in the first place.”
Persuaded? Good! Not so much? Listen to the Wonkcast!
My thanks to Kristina Wilson for recording and editing the Wonkcast and for a draft of this blog post.