Why don’t foreign aid programs simply pay recipients for attaining agreed-upon results? The idea has been around for decades, but it continues to meet resistance. Some donors worry that programs that pay for outputs or outcomes would not be able to control how funds are used and would thus be vulnerable to corruption. This brief explains why results-based payment systems are actually likely to be less vulnerable to corruption than traditional input-tracking approaches by making the effects of corruption—the failure of programs to deliver results—more visible.
This brief is based on CGD working paper 345, “Can Results-Based Payments Reduce Corruption?,” by Charles Kenny and William Savedoff.
What Are the Real Costs of Corruption in Foreign Aid?
Some donors are hesitant to implement aid programs that pay for results out of concerns over corruption. The cost of corruption in foreign aid is often understood in terms of diverted funds, the amount of money taken from a program through theft or bribes or otherwise. But measuring corruption this way ignores the much larger failure costs, which result when programs fail to generate the promised benefits. Failure costs can exceed diverted funds by large amounts. For example, Olken (2005) estimated that each dollar’s worth of construction materials lost to corruption led to substandard construction that reduced the benefit of the program by $3.41. In addition to diverted funds and failure costs, an accurate estimate of the costs of corruption should also account for the costs added to programs to prevent and detect corruption, the cost of inefficient designs from accommodating fiduciary controls, and dynamic costs from distorting institutions to facilitate corruption.
Traditional Corruption Controls Are Less Effective Than You Might Think
The primary claim in favor of traditional programs is that by tracking how money is used, funders will know if money is diverted. If the process works as designed, inputs are delivered at a competitive price. In theory, if fund “leakage” is prevented by such tracking, then all the money will be used to buy the intended inputs and the program will succeed.
Of course, in practice many things can and do go wrong. There are errors in design, implementation, and oversight. So it should come as no surprise that studies have shown that tendering, procurement, and audit procedures are often weak controls on corruption. A survey of firms that bid on international contracts found that only 15 percent of respondents thought that tender rules were an obstacle to corruption (Søreide, 2006). A World Bank study found that despite the use of similar corruption controls in all its road programs, average costs were substantially higher in countries where bribes were more common. Rehabilitating a two-lane highway cost an average of $30 per square meter in countries where the average bribe paid for a government contract (reported by firms in surveys) was less than 2 percent of contract value, but more than 50 percent higher in countries where the average bribe was reportedly more than 2 percent of contract value.
Conventional corruption control measures—tracking spending on inputs—are not only often ineffective; they can be quite costly. The World Bank annually spends at least $30 million on internal audit and institutional integrity departments and suspension and sanctions boards. This does not include the cost of 417 full-time procurement and financial staff members and 200 procurement-accredited staff members who monitor compliance with financial-management and procurement standards. Nor, of course, does it account for the staff in recipient-country governments who actually implement the procurement and financial procedures.
Most importantly, the traditional input-tracking model has no built-in mechanism to detect and respond when outcomes
fail to materialize. Thus the failure costs of corruption are never revealed. Accurate information on outcomes is uncommon, and legitimate problems are easily blamed for failures. So it is relatively easy for people to authorize and disburse money almost without regard to its impact.
Results-Based Payments Are Less Vulnerable to Corruption Than You Might Think
Results-based programs disburse funds in relation to outputs or outcomes rather than inputs and activities. In a pure results-based model, like Cash-on-Delivery Aid, the funder and recipient government agree on a common goal and transfer a fixed amount for each unit of progress toward that goal. When they are designed well, such programs identify good indicators for measuring results. Funders disburse funds ex post, once progress reports are independently verified.
For many reasons, results-based programs often diverge from this model. Programs sometimes provide some funds up front, independent from results achieved. Concerns over misuse of funds have led many results-based programs to preserve the procedures associated with traditional procurement and financial management. Results-based programs can end up with so many indicators and measure them so poorly that disbursements become contingent on negotiating skills rather than the outcome measurements. In practice, results-based programs are not perfectly implemented. Errors in choosing indicators, measuring progress or verifying results can lead programs to over- or underpay. When conditions are favorable to achieving results without much effort, opportunities emerge for individuals to divert funds to other purposes.
While traditional input-tracking programs try to control corruption by specifying what can and cannot be purchased with funds and then monitoring the how money is used, results-based programs manage corruption by only paying when results are achieved. As a result of this fundamental difference in design, results-based programs do not have direct costs associated with monitoring and controlling corruption: there is no equivalent of the fiduciary, procurement, and investigatory apparatus linked to input processes. The costs of administering the results-based program (associated with measuring and verifying the results) are essentially the same with and without corruption, although more exacting approaches to verification may be required when corruption is of greater concern than usual.
The real-world evidence on corruption in results-based programs is still scanty. To our knowledge, no cases of fraud have been proven in results-based programs. Some programs may overpay for results because they have relied on self-reported performance, as appears to have happened with the GAVI Alliance’s Immunization Support Services program. According to one study (Lim et al., 2008), 39 countries overreported and 8 countries underreported vaccination achievements, suggesting that some of the variation may have been due to real errors in reporting rather than systematic fraud. Allegations of corruption emerged in an output-based agreement between the World Bank and a Southeast Asian country, but the project was halted before outputs were fully delivered and verified. Thus only initial funding was potentially diverted; the ex post funds were never disbursed (Mumssen and Kenny, 2007).
It Is Harder to Steal from a Results-Based Program
Corruption is certainly possible with both kinds of programs, but diverting funds from an input-tracking program interferes with progress while diverting funds from a results-based program requires making progress. Someone who wants to defraud a results-based modality program can only do so by improving efficiency, that is, by generating more outcomes so that more aid money is disbursed and thus available to be stolen.
Under input-based programs, honest agents face the same or higher transaction costs than dishonest ones. After all, honest agents have to do their jobs (achieve results) and comply with all reporting and auditing requirements. Dishonest agents do not necessarily do their jobs and invent reports and invoices. By contrast, in a results-based program, honest agents face lower costs than dishonest agents. Honest agents can focus on doing their jobs to generate results that trigger disbursements. It is the dishonest agents who have to find a way to achieve results and then, in addition, find ways to divert funds.
Will Results-Based Programs Always Work Best?
Results-based programs have numerous advantages over input-tracking programs, especially in regard to corruption. However, they are not always feasible or always best. When outcomes are easy to measure (e.g., children who are literate, children who survive to age five), results-based approaches are more effective at limiting the effects of corruption and delivering results (see table 1). In cases where inputs are easy to measure and outcomes are hard to observe (e.g., technical assistance), input-based approaches may have an edge—but they still cannot guarantee progress. In short, if outcomes are hard to measure, neither results-based nor traditional approaches will work very well.
When outcomes are difficult to measure, good programs have to find other mechanisms, such as independent professional reviews and expert standards, to try to control abuses and ensure performance. When neither inputs nor outcomes are easily measured, results-based approaches are likely to be preferable. Calibrating payments to the actual cost of delivery will be difficult in such cases, but the implication of underestimating costs is to have fewer outcomes, not more corruption. This suggests that in poor-information contexts, the best strategy may be to create a results-based program with a payment that is deliberately underestimated to avoid the risk of corruption. Complementing this with professional reviews of performance would allow funders to adjust the results payment up or down depending on the level of response.
Table 1: Preferred Modalities by Ease of Measuring Inputs and Outputs
Easy to Measure
Hard to Measure
Easy to Measure
Results-based COD Aid
Results-based COD Aid
Hard to Measure
Input-based with professional
review of performance?
Results-based with price
iteration and professional
review of performance?
Results-based aid programs are criticized for being more vulnerable to corruption than input-based programs that monitor inputs and impose specific procedures for procurement and financial accounting. In fact, corruption-control strategies applied to input-based aid modalities are often ineffective, expensive, and—despite their widespread use—largely unproven. Ironically, the input-tracking approach leads to strategies that undermine rather than improve the effectiveness of foreign aid. The critical factor in choosing among aid modalities is to recognize that failure costs, the foregone benefits of a program that has been defrauded, are the true costs of corruption. By focusing on whether or not a program is achieving results, it is possible to make programs achieve more and limit the impact of corruption on development.
Lim, Stephen S. et al. 2008. “Tracking Progress Towards Universal Childhood Immunisation and the Impact of Global Initiatives: A Systematic Analysis of Three-Dose Diphtheria, Tetanus, and Pertussis Immunisation Coverage.” The Lancet 372(9655): 2031–2046.