Why Apply Systems Thinking in International Assistance?
By Abid Mallick, Monitoring, Evaluation and Systems Strengthening Consultant
Never has the imperative for using systems thinking in international assistance been more important as the COVID-19 pandemic has pushed millions of people into poverty, disrupted schooling and placed severe strain on public services in developing countries. It has exposed a dense and inescapable web of interdependencies that extend from local to global. Incremental and isolated responses with limited impact will fail to change the status quo or help us reach the sustainable development goals. A greater focus on systemic change will contribute to sustainable and wider population-based impact. For this to happen, we need increased application of systems thinking in international assistance to better understand the issues and to design interventions that lead to better performing and inclusive systems – public, social, and economic. The four recommendations to promote systems thinking presented at the end may help its uptake across the international development sector. We also need to revisit results-based management (RBM), a universally accepted planning and management approach in the sector, to find ways to enhance RBM’s effectiveness, as argued below.
Over the last three decades, the international assistance sector has witnessed widespread adoption of results-based management and its tools - theory of change (ToC), logic models, assumptions and risks, and performance measurement frameworks – to improve effectiveness of interventions throughout the project life cycle. These tools were designed to assist practitioners navigate the messy and non-linear pathways of development by breaking complex change into manageable building blocks for planning, managing, and monitoring purposes. However, in applying these tools, compromises are made in simplifying and modelling non-linearity and complex interdependences in development, which could lead to weakly designed projects and undermine aid effectiveness. The RBM approach also implicitly pushes funders, program designers and implementors to focus on achievement of higher-level results or outcomes (related to income, education or health status of people), often at the cost of paying adequate attention to strengthening processes and the broader system that would sustain the results.
As these are known issues, RBM experts advise on using results frameworks iteratively. They caution on viewing development issues simply as linear cause and effect relationships and emphasize on the need to understand the context or conditions that influence the project. In the last couple of decades there has been an increasing body of literature and a growing voice within the international development sector on understanding problems as part of wider systems within which these exist and finding solutions that result in systemic change to address root causes of problems. Donella Meadows, lead author of The Limits to Growth and one of the best communicators of systems thinking argues in Thinking in Systems: A Primer, that hunger, poverty, unemployment and other similar problems “are intrinsically systems problems – undesirable behaviors characteristic of the system structures that produce them. They will yield only as we reclaim our intuition, stop casting blame, see the system as the source of its own problem, and find courage and wisdom to restructure it”. The market systems approach developed by Springfield Centre in 2008 and the World Health Organisation’s Systems Thinking for Health Systems Strengthening provide important contributions to introducing systems thinking in international development within economic inclusion and health sectors. Building on this and other similar work, this note calls for increased use of systems thinking across the entire international assistance sector and proposes to couple it with RBM.
Why couple systems thinking and RBM approach
As documented in the Global Affairs Canada’s Results-based Management Guide, 2016, the application of RBM requires an initiative or program to be based on a theory of change that explains how and why the program is expected to produce its results and what are the assumptions, risks and context that support or hinder the achievement of expected results. This theory connects the program's activities with its expected outcomes and is the most fundamental part of results-based management. The Guide further clarifies that it is based on the knowledge and experience of the team, research and analysis, evaluations, best practices, and lessons. In practice, it is packaged in a theory of change narrative, a logic model that shows the connections from outputs to outcomes, an activity and output matrix, and risk and assumptions matrix.
But the effectiveness of this approach to RBM is compromised at two stages - problem analysis and intervention design. Although RBM calls for project design to be based on a thorough analysis of the problem and the context in which it exists in order to arrive at an evidence-based solution of the problem, the RBM repository does not include an approach or tool for issue analysis. Secondly, the process and tools for developing a theory of change through mapping the logical sequence from inputs to outcomes in a project with assumptions and risks identified, and its visual depiction in a results chain is likely to induce a linear thinking in the design – and later in the implementation stage. This is largely because in RBM, we tend to focus on finding causal links (cause and effect) but do not consider the feedback loops that may influence the cause itself or affect other parts of the system. As we know causal relationships are not always hierarchical and the pathway from input to outcomes includes complex interplay of interdependencies and feedback loops among various elements of the wider system within which the problem exists. For example, training of health workers may improve their knowledge, skills, and performance which may lead to increased utilisation of services by women and children but could adversely affect staff motivation and performance if the workload becomes excessive or medical supplies are limited. These shortcomings in issue analysis and consideration of feedback loops are blind spots (not weaknesses) in the RBM approach which if left unattended can compromise the design and effectiveness of interventions. Application of systems thinking at the analysis and design stages can help address these shortcomings.
Complex development contexts require considerable flexibility in applying RBM. When applied with flexibility along with systems thinking and systems analysis tools, it should lead to a better design that seeks to address root causes of the problem. While RBM helps in keeping the focus on ‘results’ throughout the project cycle, systems thinking enriches it in better ‘design’ of intervention/project to achieve those results. Both are important and coupling of the two approaches into RBM plus is likely to enhance aid effectiveness.
Applying systems thinking in international development
Systems thinking is a well-established approach with defined concepts, methods, tools, and skills but more importantly it is another way of viewing things (See box).
In international development, projects are designed to address certain problems or weaknesses. For example, a problem could be poor learning outcomes of primary school students, maternal and newborn mortality, malnutrition, gender-based violence, youth unemployment, or high teenage pregnancy in a community, district, province, or country. Each of these are complex and dynamic problems which exist within dynamic systems and sub-systems characterised by interdependence, multiple actors, resource flows, power play, and feedback loops. Poor learning outcomes of students may be viewed as a problem existing within the schooling system but is also influenced by the family system. Unless the problem is well understood, effective solutions cannot be designed. It is here that coupling of systems thinking with RBM may come into play by defining the problem dynamically and then proceeding to modelling i.e., results chain development, and preferably a process chain development as well.
Defining a problem dynamically starts with viewing it as part of a wider dynamic system and goes on to identify the elements, actors and stakeholders in the system, understanding how elements of the system are structured, mapping out the linkages among elements and the environment, understanding how the behavior of elements effect the overall system behavior, identifying high leverage points in the system, and understanding how these can be influenced with available resources to make the system behave differently and produce better results. Systems thinking tools such as concept mapping, causal loop diagram, stock and flow, and behavior-over-time graphs illustrate the relationship between system elements and show patterns of behavior over time (e.g., short, medium, long term) and space (village, district, province, national, global). They can be used to develop a deeper understanding of the causes of a complex problem (such as youth unemployment or gender-based violence) and to anticipate the behavior of the system – education and economic systems, family or community system - in which these problems exist. As Meadows stated in her writings (p1), “once we see the relationship between structure and behavior, we can begin to understand how systems work, what makes them produce poor results, and how to shift into better behavior patterns”.
But system analysis can become a complex exercise if system boundaries are too broadly defined, and the focus is on reducing system into parts. Such a process can continue almost endlessly as one divides an element into sub-element and sub-sub-element until one loses sight of the system. Before going too far in that direction, Meadows suggested to stop dissecting the elements and start looking at the relationships that hold the elements together. For example, in a primary education system, some of the key relationships that hold teachers, students, and other elements of the system together are the pedagogical practice by teachers, the curricula, school culture, motivation of teacher, relationship between teacher, head-teacher and district education managers, incentives for performance of various actors, relationship among students and between school and community and so on. Understanding these relationships leads to better insights on behavior of the education system and why its functioning is resulting in poor learning outcomes of students. It will also help in identifying maximum leverage points which can be used to shift the behavior of primary education system to produce better results.
The design of intervention commences once the problem analysis is done and high leverage points for intervening in the system/sub-system are identified. This is the second stage where a coupled approach may come into play to bring systems perspective to the core of the design by questioning ‘whether and how is the intervention leading to systemic change ’. This guiding question helps focus the efforts of design teams toward developing interventions that strengthen or improve the system to achieve the desired higher-level and longer-term outcomes.
There is, however, a huge need to demystify systems thinking and present its concept, methods and tools in ways that are easy-to-apply in untangling development issues, in identifying high leverage points to intervene for systemic change, and for designing interventions. In 2008, the Swiss Agency for Development and Cooperation (SDC) and the UK Department for International Development (DFID) commissioned a series of documents produced by the Springfield Centre on market systems development, ‘ Making the Markets Work for the Poor (M4P)’ approach. Since then, market systems development (or M4P) approach has witnessed increasing uptake by funders and practitioners in the economic inclusion sector. This uptake was possible because of the push by many donors such as SDC, DFID and USAID for a market systems approach in their economic sector programs. It was also possible because of operational guides and well-designed capacity building programs by the Springfield Centre and creation of platforms such as Beam Exchange dedicated to promoting market systems approach. Such efforts to promote systems thinking across international assistance and within specific sectors must continue. The following recommendations may contribute to this objective:
- Greater emphasis by international assistance financiers on systemic change and strengthening systems through development interventions. This could be an explicit objective of programs and used as a criterion for awarding grants and evaluating programs. But this should also be accompanied by recalibrating the timeframes of development programs. Systemic change and systems strengthening, especially at national level, happen over time horizons that are longer than the typical project duration.
- Strengthen RBM by coupling it with systems thinking. Funding agencies (bilateral, multi-lateral, and philanthropic foundations) can support this by adopting it in their programs and advising grantees to apply systems thinking in problem analysis and program design while following the RBM approach. For example, RBM guides can be revised to include explicit guidance on using systems thinking in situational/problem analysis and in design of interventions when developing the results chain.
- Develop operational guides for applying systems thinking in international assistance programming in general as well as for specific sectors and issues. Following the SDC and DFID’s initiative on promoting market systems development approach, other agencies may step in to develop similar guides for other sectors and issues.
- Promote systems thinking in designing and evaluating programs across international assistance sector through webinars, conferences, case studies, community of practice, training, and platforms.
Resources on systems thinking
As the purpose here is mainly to promote the idea of applying systems thinking in international assistance, the concept, methods, skills, and tools of systems approach have not been discussed in detail. But following sources, which are by no means exhaustive, are useful reading to develop a richer understanding of this field. Meadows’s Thinking in Systems is one of the most readable sources on the subject. The US National Cancer Institute’s Greater than the sum: Systems thinking in tobacco control is perhaps one of the most comprehensive documentation of theory and application of this approach on a multi-dimensional issue. The World Health Organisation’s Systems Thinking for Health Systems Strengthening discusses concept, methods, skills and tools for systems thinking and its application in addressing health sector issues. An article by Zazueta et al, on Development Trajectories and Complex Systems-Informed Theories of Change presents in detail a methodology based on socio-ecological systems (SES) concepts to construct theory of change as an evaluation tool for a fishery sector project in Indonesia but clarifies that ToC are more effective when developed during design of interventions. The Springfield Centre’s M4P Operational Guide is an excellent practitioners guide for market systems development.
Abid Mallick is an international development consultant with a focus on systems and institutional strengthening, evaluation and monitoring, and program design and management. He is a systems engineer by training and founder and head of DevSYS Consulting, based in Ottawa. He can be reached at email@example.com