September 8, 2017
The UK’s Department for International Development (DFID) has been a leader in the use of PbR in international development. DFID was a founding donor to the Global Alliance for Vaccines and Immunisation (GAVI) in 2000, the Global Fund in 2002, the Global Partnership for Output Based Aid in 2003, and the Health Results Innovation Trust Fund in 2007, which each trialed and tested innovative PbR mechanisms.
Since 2010 there has been an increased drive to get UK government departments using PbR contracting. In DFID, we’ve also accelerated our work on PbR, commissioning new research and producing new guidance to drive further scoping and piloting. For example, two Cash on Delivery Aid pilots – in the Ethiopian and Rwandan education sector – were developed in collaboration with the Centre for Global Development (CGD) and partner governments. The UK has also been a leader in developing Social Impact Bonds (SIBs) for the delivery of public services domestically. In DFID we are also at the early stages of developing programmes to test and learn what their international equivalents, Development Impact Bonds (DIBs), can achieve. At the same time, PbR has also gained momentum with others working in international development, with several aid organisations starting to pilot or scale up the approach.
In terms of value for money, we’re really interested in the potential that PbR offers as an opportunity to transform the delivery of development assistance by paying once results are achieved. In 2014, DFID published a strategy on Payment by Results, including a commitment to build the evidence base of what works, and build capacity for doing PbR well.
Emerging evidence suggests PbR may be helping deliver promising or high-performing programming in health, education, water supply and sanitation, private sector market-led development, agriculture, and in accelerating the take-up of new technology that improves poor peoples’ lives. DFID’s PbR portfolio continues to grow.
Our use of PbR in appropriate contexts is part of broader reform to improve commercial awareness and capability across the organisation and to make sure we get good value for money from the development budget. PbR is a valuable tool, which when designed well can be used to change incentives. In doing so, it gives us new means which we can bring to bear in delivering against UK priority areas.
By piloting and testing performance based payment approaches, we and other funders are looking to realise a number of benefits. In traditional aid, by paying upfront donors accepts the bulk of the risk of programme failure. PbR can alter this balance by sharing some risk for delivery with partners. Where designed well and in appropriate contexts, this can in turn sharpen incentives for implementers to perform. Particularly relevant in the context of agriculture technology markets, it can also increase innovation and flexibility in delivery. By not specifying how results should be achieved, payment by results frameworks can, if designed appropriately, provide implementing organisations with space to innovate to improve outcomes. Payment by results may also increase transparency and accountability for, and focus on, results. Finally, it can create a strong focus on performance in service providers. By being paid on results, partners are strongly encouraged to examine what is and isn’t, working, driving up performance standards, management and measurement.
Our wider experience with PbR suggests there are some limitations and tensions which affect when PbR might be appropriately used.
- Transferring additional risk to suppliers may make them less willing to try innovative approaches, instead relying on more predictable tried-and-tested options. They are more likely to charge a premium to manage the additional risks or changes in the operating environment.
- Incentives affect different delivery partners differently. For instance, smaller businesses may be more able to nimbly respond to incentives with new, innovative approaches, but less able to manage the associated financial risk than can large, well-financed businesses.
- Pre-defining the price and volume of results may not be the best way to manage ambition in poorly understood, uncertain, rapidly changing or volatile environments. In such circumstances, adaptive programme management might offer advantages over pre-determined results targets.
- Emerging evidence shows that developing robust direct measures of a supplier’s effect on outcomes – which may not be fully under their control – is the single most difficult thing to get right in a PbR contract. But rewarding indirect or proxy results can create perverse incentives and gaming. For instance targeting sales of livestock vaccines, rather than directly measuring improvements in animal health, might encourage vaccine delivery agents to game the system.
- The difficulty of designing the right results measures, verifying performance against these and the overall complexity of designing PbR contracts tend to be underestimated at the outset by both programme staff and implementing partners. However, given we’re still at an early stage with PbR, this isn’t altogether surprising.
At a donor portfolio level, significant increases in PbR programming present an additional challenge for predictable financial management. Certainly these challenges resonate with our own early experience with AgResults “solvers” – a very diverse set of private sector players operating in agriculture technology markets. Some of these will be picked up and discussed further in future AgResults lesson sharing pieces and blogs.
If we want to extend the use of PbR in agriculture, then we need to identify situations where the results we want to achieve are clear and measurable, where we can confidently fix the price or results-reward in advance, where the benefits of using PbR to create particular incentives clearly outweigh the various additional financial costs and other potential limitations, and where the scope for creating perverse incentives and unintended consequences are low or manageable. As we are learning in AgResults, designing pilots is challenging process, requiring a strong understanding of local context and markets.
Evidence about how PbR can be most effective is still at an early stage. This is particularly the case in agricultural and other technology markets, primarily because these are relatively new mechanisms. We need to develop strong evidence about when and how these incentives work in practice.AgResults is unique in its approach in testing and piloting the use of results-based incentives or “pull mechanisms” to competing private sector players for the uptake and development of new agriculture technologies. This is why we have championed the need for a robust, well-designed evaluation of AgResults built in from the outset. AgResults is therefore also uniquely well placed in its approach to testing the effectiveness and efficiency of pull mechanisms in comparison with traditional approaches. Whilst the evaluation is still ongoing, there are already valuable insights and lessons emerging from the External Evaluator (Abt Associates), which they will be sharing via the website and through future blogs. We will also reflect on some of these lessons from a donor perspective in a future blog.