Politicians and agency officials are always morally indignant when it comes to corruption in foreign aid, pointing to elaborate procedures and investigative offices to prove that they are “tough” and calling for zero tolerance (most recently here and here). However, for most governments and agencies, corruption is only a problem when it is discovered. That is when it becomes an obstacle to disbursing funds and keeping business moving.
The most common strategy could be called “scandal-driven management” - which involves sporadic overreaction to publicly exposed abuses (without regard to their scale or importance) and little systemic attention to how to achieve development outcomes in corrupt environments. Ironically, one of the best systemic ways to limit the negative effects of fraud and abuse in foreign aid programs – paying for verified results – is criticized for being vulnerable to corruption when such an approach is probably less problematic than more traditional ones, as Charles Kenny and I have recently explained in this paper, policy brief, and wonkcast .
I think of scandal-driven management as responding to three different kinds of stories: the luxury jet scandal, the broken incubator scandal and the unaccounted funds scandal.
The luxury jet scandal occurs when a president or high official from a country that receives substantial foreign assistance purchases something expensive and unnecessary – a luxury jet, a gold-plated bed, a mansion in Paris, a massive statue.
The broken incubator story (also known as the road to nowhere or stolen drugs story) breaks when a journalist or investigator shows that goods and services that were supposed to be delivered by a foreign aid program are not, with dire consequences for the population.
The unaccounted funds stories are usually prompted by the release of audits or other reports that cannot vouch for whether funds were applied properly (whether or not the program achieved its goals).
The typical foreign aid response in almost every case is to tighten fiduciary controls and layer on more procedures. In other words, look tougher regardless of the impact on efficiency and outcomes. Even so, there is little evidence that these controls make a systematic difference.
In fact, the typical input-tracking approach with all its procedures and rules has no real impact on avoiding such scandals at all. Perfect input-tracking cannot address fungibility – President Wade can still build his statues with borrowed funds or domestic revenues freed up when donors pay for education and roads. Perfect input-tracking only proves that procurement was done “properly” and that valid-looking receipts were provided – not that incubators were delivered and are functioning. Input-tracking programs are good at reporting on unaccounted funds but will do so even when no corruption has occurred. And when corruption does occur, impunity remains common.
By contrast, performance programs can make a dent in this problem in at least two ways. First, results-based programs will rarely face broken incubator scandals if they have independently verified deliverables before they disburse. Second, results programs never have unaccounted funds because they don’t prespecify what money has to be spent on. It is true that results-based programs can’t stop luxury jet scandals from happening, but then, no foreign aid program can ever avoid luxury jet scandals without controlling all public spending.
So next time a scandal erupts, watch the pattern. It isn’t hard to see. And if you work in an aid agency, know somebody who does, are a legislator, legislative aide, or a mere tax payer, start pushing for structural changes that might really address corruption. Start pushing for aid disbursement arrangements that pay for independently audited outcomes, not dubious receipts.
Thanks to Charles Kenny for holding me accountable to logic and Albert Alwang for finding examples.