The baby and the bathwater: can we have a more constructive debate on performance-based financing?

By Godelieve van Heteren

Paul et al’s critique

In a recent paper in BMJ Global Health, 24 authors (Paul et al) – the majority of whom are French-speaking academics and consultants from northern countries – declared that performance-based financing (PBF) in low- and middle-income country (LMIC) health systems needs an urgent rethink. They claim it is a donor fad. They argue that PBF is gaining too much traction, too much money is being spent (and wasted) and the gaps between promise and practice of PBF often go unchallenged.

They state that PBF’s dissemination is indiscriminate, with too much simplistic copy–pasting of a ‘PBF model’ (which the authors do not further explicate); that the current communities of PBF practitioners and practice thinkers are complicit in it all and not ‘independent’ enough to carry out the proper research that would set things straight.

What is required, the authors contend, is a group of ‘independent’ researchers to keep a finger on the pulse of implementation and to recommend better strategies to countries, donors and others, so that country stakeholders would no longer blindly follow World Bank models but work instead on matters that are more sustainable and country specific.


Declaring my bias

Let me come clear right away. I am fully biased. I’m a 59-year old white, middle-class woman from one of the richer countries in Europe. And I am a PBF person. I am too old to believe in being ‘independent’ or 100% objective. We are all driven by subjective feelings, however much our work requires us to be objective.

I believe in building things together; and I have, with thousands of friends, colleagues, and constructive adversaries, been involved – as an academic, iNGO director, parliamentarian, lecturer, trainer, and evaluator – in struggles for forms of health (systems) strengthening that boost quality and equity and use public funds efficiently. I have seen fashions come and go in global health. I have seen gains but also the wastages that occur in health systems. I learnt in the school of life that things work best when defined, carried out and managed by the people most concerned. I have worked with idealists and ideologues, passionados and power freaks, opportunist-hoppers and people who with amazing commitment stay with their promises and beliefs through thick and thin.

I have experienced on my own continent and elsewhere various incarnations of insurance systems, national health systems, input-based models, output schemes, pilots, projects, and large-scale programmmes. I have weighed the pros and cons of input systems versus output systems, vertical thinking versus attempts to horizontalise. And I have had the pleasure to deal with the lobby and advocacy groups for all these approaches.

But in the end, I prefer PBF to any other health systems approach. Not because it is perfect, but because in principle and in its make-up, it is an open and dynamic approach. It is based on a sense of human nature and potential, an awareness of tough social processes and a commitment to good governance, which resounds with my own life’s experience.


Where I agree with Paul et al

Paul et al built strawmen instead of bridges, with the risk of throwing the baby of PBF (i.e. years of critical practical development) away with the academic bathwater. But there are many points on which PBF practitioners like myself could agree with them, including that:

  • PBF implementation should be scrutinized: PBF has gained traction and international interest, and much money is being spent on PBF, so it is justified to follow that money and PBF implementation up close.
  • PBF should not be copy-paste as a model: PBF implementers have long agreed that things are locally specific and require nuanced thinking. PBF implementers do adapt practice and constantly add to the acquired wisdom. PBF communities of practice have constantly shared knowledge and discuss differences in implementation.
  • PBF implementers should take local ownership and country specifics: Many of the PBF implementers I work with do emphasise local ownership in their local settings. Surely, where PBF approaches are badly designed, sloppily implemented or work without government ownership, sustainability is out of the window. So, there is nothing new here, and there is a lot of scope and willingness to work better.
  • The ‘evidence base’ for PBF is mixed: Here again, PBF implementers would agree, and in fact for years have been involved in efforts to broaden that very base: through impact evaluations, action research, and by adding different angles to the PBF systemic approach: regarding quality, equity, sustainability, scale up, etc.


Where I disagree with Paul et al 

If all the work of the various PBF implementers and communities of practice over the last two decades were taken more seriously into consideration by Paul et al, their article could have been a ‘shared point of consideration’. But while they issued a justified appeal for critical thinking, they ignored four fundamental dimensions of PBF practice and already ongoing critical reflection in the field:

  • Dynamics: PBF implementation is dynamic, relying on communities of practitioners, policy makers and thinkers who are aware of the politics, the many deficiencies in implementation and the gaps in evidence. They are engaged in setting these straight through action research, internal and external evaluation, and constant adaptation; and by linking with other fields and stakeholders to improve quality, equity, efficiency, and sustainability. None of this is reflected in Paul et al’s critique, which instead seems based on a small set of cases, notably in Benin and Burkina Faso.
  • Package: PBF implementers are not naïve about the wider context in which health systems strengthening operates. They actively engage in discussions about UHC, equity, quality, community involvement, and local ownership, because they believe PBF offers a range of practices, governance mechanisms and tools that can help answer the question of ‘how’ to achieve these goals. Nowhere do Paul et al define the package of PBF practices and components – they treat PBF as a fixed monolithic model, leaving out a wealth of ongoing practical development.
  • Adaptation: PBF implementation follows the logic of practical development: trial and error, and adaptation; and it applies economic, systems and governance theory, to inform PBF set-up, such as separation of functions, verification, and other checks and balances. In practice, this means that things can go wrong. But it also means that critical reflection and adaptation are part of the PBF work-in-progress. The old-style randomised ‘impact evaluations’ did not sufficiently address such dynamics, so PBF communities of practice have been active in the journey towards more action research to redress adaptation. So, basing any critique of PBF on such impact evaluations alone is fundamentally flawed.
  • Politics: Because of its increasing prominence, PBF has entered into the arena of the political economy of health and academia with its current inequities and injustices, jealousies, and power and money games. In contrast to what Paul et al’s critique may suggest, many PBF practitioners are active and keen to be part of necessary changes in the political economy of health; and many are motivated to move much further than Paul et al’s recommendations, especially to make everybody (including northern academics and agencies!) part of the critical equation.


Moving on

PBF practitioners include many people who wish to free themselves from donor dependency and all the strings attached, fight the persistent remnants of colonial power framing in global health and build real equitable relationships. What often prevents this from happening is the complex – bread, butter, fame – struggle for survival!  But my hope for the future rests on the observation that the urge for change which many of my PBF colleagues express goes beyond northern academics telling other northern consultants what to do or not to do. This is no time for factionalism, ideological grandstanding, or turf wars. What we need is a more constructive debate on PBF, based on more honest, dynamic, practice-based, nuanced, and inclusive considerations than Paul et al introduced in their appeal for rethinking PBF.


About the author: Godelieve van Heteren, MB, MD, is an international health and development policy expert currently working as senior consultant, in PBF trainings and to assist the implementation of the new international Health Systems Governance Collaborative, hosted at WHO. She has previously been the foundational director of the Rotterdam Global Health Initiative (RGHI), a multi-institutional research network of global health innovators also based at Erasmus University, with collaborative research projects around the globe. She has also worked as the director of international development NGO Cordaid, as a Member of the Dutch Parliament, and as a full lecturer at the Radboud University, Nijmegen.

Competing interests: I have read and understood the BMJ Group policy on declaration of interests and confirm I have no conflicts of interests to declare.


(Visited 251 times, 1 visits today)

1 thought on “The baby and the bathwater: can we have a more constructive debate on performance-based financing?

  1. Paul et all’s critique is welcome and the authors may have hit the nail on the head regarding a few issues that plague Performance-Based Financing (PBF) based health reforms The lack of domestic ownership as manifest in the absence of Government funding in some places, or troubles with design and implementation of PBF approaches are real indeed and can be the main reasons for why some PBF programs in Low, -and Middle Income Countries (LMIC) may fail. However, in their attempt to take on PBF exclusively, the authors overlook several important factors:

    Firstly, they ignore the fundamental problems plaguing LMIC health sectors in general: remnants of dysfunctional and underfunded colonial centrally managed health systems, which are the root cause
    of unnecessary, high maternal and child mortality. The authors’ claims are based on a rather small set of contexts, of which one (Burkina Faso) is showing good results (Steenland, M., et al. (2017). “Performance-based financing to increase utilization of maternal health services: Evidence from Burkina Faso.” Soc Sci Med(3): 179-184.) and an enhanced and expanded version of the Burkina Faso PBF approach awaiting the results of an impact evaluation. Their second case, that of Benin, involves a PBF program which was rather successful at the outset but suffering from a general lack of autonomy in the health system for providers leading to lesser results and being confronted with a new president changing the direction of these reforms. In addition, the case of Benin also suffered the consequences from an unsuccessful process to improve the design, to enhance Ministry of Health ownership and to harmonize different PBF approaches. Their third example, Tanzania, was a rather small Pay-for-Performance pilot, which had multiple design faults.

    The authors do acknowledge, though, that where government ownership of PBF was present (as e.g. expressed in domestic funding) such as in Rwanda and Burundi, this contributed substantially to the documented success of PBF approaches in those countries.

    Secondly, they claim that PBF approaches are not efficient. This is untrue. In their assessment of whether ‘things work’, the authors bypass one crucial effect of PBF approaches: The efficiency of PBF programs or cost-effectiveness of PBF approaches is essentially through the impact on the quality of health services, for instance for women and children benefiting from such PBF programs. This increase, notably in settings coming from very low baselines, is so strong that even a moderately successful PBF approach in Zambia was shown to have been a very cost-effective intervention. Other examples: the Cameroon PBF program had strong impact on quality of care too. A chief driver of success in the Nigerian PBF approach, one of the two largest in the world, covering three states, is an impressive double difference of 49-percentage point increase in a composite measure of structural and process measures of quality of health services as compared to control districts (results pending). Sure, a country the size of Nigeria faces many challenges, but In Nigeria, not just structural quality improved significantly, but also content of care measures such as clinical quality for children visiting outpatient clinics and pregnant mothers visiting antenatal care clinics.

    So, what to conclude from the authors’ analysis?

    There is nothing wrong in looking critically at things. But should we abandon PBF? The authors argue PBF approaches in LMIC should be abandoned because of the presumed mixed results in three examples. Certainly. programs with no or detrimental effects on quality care should be abandoned. Such as is the case in many free health care policies in which no attention is given to compensating providers for their lost income. But well designed, implemented and locally owned PBF programs have shown substantial improvements, and can lead to real results in difficult circumstances.

Leave a Reply

Your email address will not be published. Required fields are marked *

Name *