Systems Innovation Blog Series - Part Three: Systems Innovation – Some Lessons, Some More Caution
By Benjamin Kumpf (OECD), Nina Strandberg (Swedish International Development Cooperation Agency) and Robbie Barkell (UK Foreign, Commonwealth and Development Office). Illustrations by Vidushi Yadav.
*Disclaimer: The views expressed in this blog series are the authors’ own and do not reflect the official positions of their home organizations, and not of the International Development Innovation Alliance.
In this five-part blog post series we share some of our lessons on advancing systems innovation practices in bilateral development organizations. With this, we seek to advance conversations on improving innovation efforts in the international development sector. In the first post, we describe what systems innovation is. In the second post, we make the case for systems innovation. This third post discusses some lessons from international development cooperation that inform our approach. In the fourth post, we present promising practice and building blocks of systems innovation. The final post shares five questions to help advance systems innovation practices in international development organizations.
“The world is a complex, interconnected, finite, ecological – social – psychological – economic system. We treat it as if it were not, as if it were divisible, separable, simple, and infinite. Our persistent problems arise directly from this mismatch.”
(Donella Meadows)
The year 1997 marks somewhat of a turning point for international development. The Progresa cash transfer programme was launched, providing cash to low-income Mexican households that provide evidence on children’s enrolment in schools, regular health check-ups at selected clinics, and attending nutritional education meetings. Monetary social assistance programmes were introduced in West and North European welfare states as early as in the first half of the 20th century. These turned into welfare schemes after the end of the Second World War. In the global South, cash transfers were only tested and scaled from 1997 onwards. The concept of conditional cash transfers was pioneered and tested in countries in Latin America and is now coined as a ‘Southern innovation’.
Within a few years after the rollout of the first conditional cash transfer programme, there was a growing evidence base demonstrating positive impact of the transfer programmes. Accordingly, many development organizations pivoted from providing goods and services to managing the transfer of cash to households and individuals, often led by national ministries. Across regions and countries, the introduction and scaling of cash transfers has followed a unique course, depending on the systemic dynamics between governments, civil society and international development partners, and the capacities of the respective players.
There is an extensive body of empirical research outlining the impacts of conditional and unconditional cash transfers on poverty alleviation, and effects in domains such as education, health, and labour activity. Many of the studies are based on experimental designs, with the majority focusing on the short-term success of cash transfers.
Evidence from recent years on what might contribute to positive long-term effects shows: for cash transfers to be successful, development providers and partners need to move from fragmented programmes to a systems-approach to social protection. This includes collaboration across organizations and administrative boundaries to provide multi-sector services to the poorest households. A coordinated approach across public sector entities, civil society organizations and service providers that reflects the multi-dimensional nature of poverty in practice is necessary to develop and sustain effective social protection schemes.
So for a technical innovation that emerges on the micro-level, in this case conditional cash transfers, to unfold its impact, institutions on the meso-level such as public sector entities, NGOs and service providers need to be engaged. These institutions are coined as the engine room of a system by Charlie Leadbeater and Jennie Winhall in the Green Paper ‘Building Better Systems’. Both argue that promising ideas and interesting experiments are needed for systems innovation to happen, “but without a new regime there is no new system.”
In our agencies and together as a working group of the International Development Innovation Alliance, we are currently pursuing systems innovation, the practice of how we can support the transformation from one social and technological system to another. We see this as necessary becuase energy and transport systems, as well as global food and welfare systems need to be transformed if we are to adequately address the challenges of the 21st century, notably climate change, biodiversity loss and widening social inequalities.
The premise of our work is that single innovations do not suffice. Technological innovations and single point solutions are merely components for needed transformations. Systems innovation is as a set of interconnected interventions and innovations in social, cultural, technological, economic and political spheres.
In the pursuit of supporting the transformation of systems, we found promising systems practice in the international development sector. With lessons for systems innovation from efforts to strengthen health, welfare and education systems.
Good practice is establishing a shared understanding of the system that produces the negative outcomes, its components and boundaries together with people living and working in this system. This can be done on the basis of pre-established conceptual framings, for example the building blocks of health systems, proposed by the World Health Organization. These facilitate a common language and approach in the sector. The blocks include service delivery, health workforce, information, medical products, financing and governance. Evidently, such building blocks alone do not constitute a system. Rather, the multiple, sometimes conflicting relationships within and among the blocks. The way how one affects and influences the other is what turns these into a system that produces outcomes. The experiences in national health system strengthening and the use of building blocks, inspired us to propose building blocks for systems innovation. We present these in part four of this series.
Lessons from the health and welfare sectors underscore that strengthening or transforming a system is based on context, and as much a political as a technical issue. That successful efforts factor in investments in relationship- and coalition-building, and in establishing shared intent across diverse stakeholders.
Good practice in programming in this context also include the establishment of management configurations that enable decision-making closest to where the information is. Importantly, to construct funding, contractual and results frameworks around learning. In fact, we found that learning is treated as a dedicated outcome in multiple systems-strengthening programmes. For example MUVA, a multimillion pound programme of the UK Foreign, Commonwealth and Development Office to help advance women’s economic empowerment in urban Mozambique. MUVA is essentially a social incubator to challenge social norms and make systems work better for women’s economic empowerment. Since its start in 2015, MUVA has designed and implemented 17 projects. All of these were developed with local partners, based in the joint identification of barriers that limit women from accessing decent work opportunities. The MUVA case study in the book ‘Human Learning Systems’, emphasizes the programme’s approach to learning and to defining outcomes. “These outcomes were viewed from the perspective of the participants of the programme, rather than from pre-defined and standardised metrics. Each achievement was celebrated, often led by the participants themselves. For example, participants of the MUVA programme designed their own graduation and leaving ceremonies telling the stories of how they experienced the programme and what changes it had impacted in their lives. This approach was mirrored in MUVA’s formal evaluation processes via qualitative interviews with participants to understand the impact of the programme on the young people who participated. MUVA employs the Ripple Effect Mapping technique which allows the programme team to understand how various groups understand change - participants in themselves, their families and their wider communities. Interviews are also held with the parents, guardians, peers and the wider community that they worked with. When undertaking evaluation with participants MUVA never used the programme’s theory of change or framework, but simply asked them "How do you see change?"” The managers of the MUVA programme also created participatory environments for learning and adapted decision-making for the responsible parties within the FCDO in London, and ‘downwards’ for delivery teams.
MUVA and other development programmes that set out to strengthen or transform systems tend to provide resources for dedicated investments in relationship-building, acknowledging the importance of trust between partners. They also factor in diverse forms of evidence, including lived-experiences by people living in the system.
These are a few of the lessons that inform our work on systems innovation within our agencies. We are cognisant that such a concept might add to the conceptual overload in international development and risks to only add new jargon, new highfalutin prose and think-pieces such as this one, not better practice.
Systems thinking is a well-established field with a plethora of different and often competing definitions. It overlaps with fields such as complexity science and systems. Our interest here is not to present a definition of systems, nor of complexity. We rather refer to the abundant literature that has been produced over the last half century and more.
A significant contribution to systems thinking stems from militaries of high-income nations that pioneered tools to better understand choices of actors to mitigate the risk of nuclear conflicts. Among the multiple schools of thoughts, two rather distinct approaches can be differentiated: one focusing on quantitative models, and another one interested in processes and mental models that are hard or impossible to quantify. Accordingly, there are multiple practical approaches to systems thinking and over the last decade or so we witnessed a new wave of enthusiasm for systems thinking and systems change practices across the international development sector.
The renewed enthusiasm has created a paradox: on the one hand, a wealth of academic models, a rapidly growing industry of consultancies and toolkit-developers and an ever-growing number of development funders that are positioning themselves as system leaders. They bring along a host of implementing organizations that align around calls for ‘systems change.’ But observers note that the results are, so far, mixed: “These [efforts] have had at best uneven impact on how decisions are made or large organisations are run”, as Geoff Mulgan points out.
On the other hand, we have engaged with the Doing Development Differently movement, worked with adaptive management approaches and importantly collaborated with implementing agencies, with local partners and evaluators that have been telling us, development funders, how crucial complexity appreciation is. Back in 2013, one of our collaborators, Ben Ramalingam, published the book ‘Aid on the Edge of Chaos’ to advocate for a step-change in the sector. To amplify the voices that state that politics and power determine the success of technical solutions and linear thinking and programming are not likely to result in sustainable change. We believe that, as a sector, we haven’t sufficiently changed.
Our intent here is not to position our agencies as ‘leaders’ in the field of systems change or systems innovation. Rather we seek to identify and address some of the shortcomings of innovation and of systems change practices we have witnessed. These include frequent jargon that has exclusionary effects, insufficient considerations of power dynamics, along with the difficulties of balancing the often opposing objectives of a system’s many stakeholders. In some systems-practices there is an overemphasis of modelling and a lack of reflection on one’s own function in systems that are influenced by Western mental models and practices.
Overall, we see a gap between making sense of a system and making an impact in the world, or affecting the organization’s decision-making processes. A further constraint of putting systems-thinking into practice in the development sector is the mental bias and structural push for well-defined and well-planned projects with clear theories of change and detailed execution plans, as opposed to more loosely defined, spacious, and adaptive forms of organizing work around complex challenges, as Dominic Hofstetter argues.
We particularly set out to challenge binary thinking that frames reductionist, linear thinking as bad, and complexity thinking as good. In international development, there is a clear need and place for mechanistic thinking and practice, for single-point solutions that go to scale. COVID-19 vaccines being a case in point. But to unfold equitable positive impact, such single-point solutions need to be embedded in a portfolio of interconnected interventions and innovations across social, technological, economic and political systems – pursuing a clear intent to either strengthen a system, or transform it.
We would like to thank Ben Ramalingam for his kind and critical support throughout the drafting process of this blog. Huge thanks also to Aarathi Krishnan (United Nations Development Programme), Alan AtKisson (Swedish International Development Cooperation Agency), Ammaarah Martinus (Western Cape Government), Anne Germain (Global Affairs Canada), Asha Meagher (OECD), Angela Hanson (OECD), Antony Herrmann (UK Foreign, Commonwealth and Development Office), Constance Agyeman (Nesta), Dave Milestone (McKinsey) Dominik Hofstetter (Climate KIC), Emma Foster (UK Foreign, Commonwealth and Development Office), Karlee Silver (Grand Challenges Canada), Morag Neill-Johnson (Results for Development), Peer Priewich (German Corporation for International Cooperation GmbH), Rahul Malhotra (OECD), Will Spencer (USAID Centre for Innovation and Impact) and Yara van Heugten (Dutch Ministry for Foreign Affairs) for critical and constructive feedback on earlier versions of this blog series.
