Quite a few market systems development projects I have come across in my practice have a goal in their logframe to achieve systemic change. In most cases this is spelled out around some or other market function that is supposed to be improved (e.g. improved access of poor farmers to seed). But in some cases, the log frame simply asks for a number of unspecified systemic changes to be achieved. Both cases are interesting in their own right, but particularly in the latter case evaluators need to be able to answer the question “is it systemic change or is it not?”. There has not been a clear way to answer the question.
In this post, I want to introduce two concepts that can be helpful to answer this question. Firstly, the idea of ‘depth of change’ taken from the systems thinking literature, which helps us understand how fundamental a change is with regards to a system’s architecture. Secondly, the idea of resilience and the question if development interventions build the resilience of the market system or economy.
How is systemic change interpreted now?
Before introducing the two new concepts, I’m quickly reviewing the current practice of assessing systemic change. There are various examples of projects that talk in their reports about how they have achieved systemic change. Often, systemic change is thereby understood as an innovation introduced by the development project that is then taken up by market actors – especially if these innovations spread beyond actors the project has directly interacted with and lead to some reaction of other actors in the system, e.g. in terms of changing rules or norms.
The thinking that systemic change is about an innovation that a development project introduces and how it spreads in the system is embodied in the so called AAER Framework , which represents systemic changes as innovations that go through four phases – adopt, adapt, expand, respond. The design of the innovations is generally based on a thorough problem analysis by the project, searching for the ‘root causes’ of underperformance. These analyses generally suggest a number of changes in actors’ behaviours that are needed to achieve a desired poverty reduction impact. The adopt phase represents the adoption of an innovation, introduced by a project in partnership with a number of actors in the system, by these partners. Adapt refers to sustained behaviour change among relevant players in the system – they need to gain ownership of the innovation and embed it in their standard operations. This phase looks at the institutionalisation of the change. Expand refers to a spread of the innovation beyond initial project partners into new geographies or market segments, bringing the benefits of the innovation to more people. Finally, respond looks at whether other functions or rules in a market are changing as a response to the spreading of the initial innovation. Adopt, adapt, expand and respond are not seen as subsequent phases but can happen in parallel. For example third party actors that take up an innovation (expand) will also need to go through an institutionalisation of the change (adapt) to make it stick.
Example 1: GEMS1 in Nigeria
The GEMS1 project in Nigeria worked with feed companies as their partners . The innovation that was developed based on an analysis of the problems in the production of red meat in the country was to introduce a balanced feed supplement and to create a new business model for the feed companies. In the new business model,
… feed companies contract paravets as their agents to promote and sell feed supplement to farmers. The paravets earn a commission on the sales of feed supplement. They mobilize communities for farmer training on the use of balanced feed (as well as complimentary animal husbandry practices) and provide them with veterinary services. The farmers buy feed supplement from the paravets and pay them for their veterinary services [2:4].
The aim of the intervention was for farmers to increase their profits from animal husbandry.
Systemic change was seen to happen through two pathways. Firstly, it was expected that if feed companies other than the ones the project directly partnered with would see that the business model is profitable, they would “crowd in” and also adopt a similar business model. Secondly, it was expected that farmers who observe their neighbours using the new feeding supplements and veterinary services would copy them and also benefit from higher profits. Both pathways correspond to the idea of expand in AAER-terms.
Example 2: Katalyst in Bangladesh
Katalyst has worked in a wide variety of sectors and piloted a number of different ways to assess systemic change (full disclosure: I have worked with Katalyst to pilot a new way to measure systemic change). The example I want to use here is from the vegetable sector, written up by Ben Taylor and colleagues . The Katalyst project engaged in the vegetable sector as it saw considerable opportunities for poor producers to improve their livelihoods. Taylor et al.’s case study focuses particularly on the marketing and distribution of agricultural inputs, particularly seed. Problems in the seed sub-sector were identified as lack of access, lack of quality, and lack of use. The underlying problem was identified as that
… the risks on both sides of the seed transaction were perceived to be too great. The costs of investment in changing behaviour to new business models – as growers of high value vegetables or as distributors and marketers of high value seeds to new markets – were perceived to be too high. [3:27]
Katalyst employed a number of strategies to improve the situation. Firstly, it worked together with five seed companies to establish demonstration plots that show the effectiveness of new improved seed varieties. Secondly, it engaged in the formalisation of a network mobile seed vendors that had existed before and directly connect them to the seed companies, who would train them and establish continuous commercial relationships (similar to the paravets in the GEMS1 example). This increased the access of farmers in remote areas to seed and embedded information that was spread through the mobile seed vendors. Thirdly, Katalyst worked together with seed companies on improved marketing material that were more accessible to farmers, particularly in remote areas. Finally, together with its partners, Katalyst sought to introduce a smaller, more affordable packet size of quality seeds to the market which was more appropriate to poor consumers. All perspectives of the AAER framework are represented in this case.
The results of these interventions are impressive. More farmers can now access better seeds and better information about how to use them. The innovations were taken up by companies beyond the direct partners of the project. It is estimated that at the time of the case study, 4,500 mobile seed vendors operated in Bangladesh, supplying an average of 125 farmers, who now have access to seed an information. Mini-packs are now the predominant form of vegetable seed retail by seed companies.
As seen in these two examples (and there are many more that are similar), systemic change is seen as a mix of changes in relationships, combined with new or improved products and business models. These are usually happening between companies that provide inputs or buy products and interlocutors like input retailers, smaller traders who aggregate products, etc.
The AAER framework is a useful tool to understand how innovations introduced by a project can lead to large-scale change in a system. However, instead of introducing an innovation, some projects help to build or improve institutions that help markets actors to function better in the economy, e.g. the establishment of publicly accessible vocational training centres or of quality infrastructure like standards and certification agencies. These changes are more difficult to assess using the AAER framework as they do not represent innovations that can diffuse; benefits do not spread by crowding-in or other actors. There the question remains: does the establishment of such an institution constitute a systemic change?
‘Depth of change’
The idea that systemic change in market systems needs to go ‘deep’ is not new. ‘Depth of change’ was one of seven principles that a group of market development practitioners has come up with back in 2012/2013 when discussing monitoring and measuring change in market systems . The report states that
… [f]ocusing the attention on these superficial changes creates a false illusion of success that can easily mislead donors when deciding where, how much and when to invest, and practitioners when deciding how and when to act, what and how much to subsidize, and when to exit. Systems thinking offer [sic!] us clues to assess the relative depth of impact of our interventions and of the actions of market actors themselves.
To better understand the notion of depth I will use a metaphor that I have introduced already in an earlier post and that is quite popular amongst systems thinkers: the systems iceberg – or pyramid, as seen in the illustration in that earlier post (I prefer the iceberg metaphor because of the water surface that let’s you play with the notion of ‘over’ or ‘under’ the surface). The original source of the iceberg idea is hard to determine, but I assume it goes back to an article by Donnella Meadows, in which she defines a hierarchy of generic leverage points in systems .
The metaphor of the iceberg (as opposed to the pyramid) also allows us to make another statement: an observed change is systemic if it is the consequence of a change that happened under the surface. And maybe even: the deeper the more systemic.
In principle, that sounds reasonable – only that I would probably put the water surface a bit deeper, between patterns and underlying structures, or somewhere in the middle of patterns. I would not call a change in patterns or trend systemic in every case as it is most probably only temporary if the underlying structures do not change. Hence, I would probably see a change to be systemic if it changes the underlying structures. So maybe we just need to redraw the iceberg and say that patterns and trends are, at least partly, still above the surface.
Now to come back to our two examples and look at them from the perspective of the iceberg.
In example 1, the changes were in the relationship between the feed company and the paravets as well as the paravets and the farmers. Also, there was a change in the business model of the feed company, as they now have a layer of retailers in between them and the farmers. What changes are we describing here? We definitely see changes in patterns and trends, as the way these people interact on a daily basis changes through the new business model. We also see some changes in the underlying structures, particularly in the relationships. But how far reaching are these changes? The case study does unfortunately not talk much about the crowding-in of other companies into the new market so we do not know if the structural changes are widespread or not. Hence, we do not know if this change is significant (I introduced ‘significance’ as an important marker of systemic change earlier).
Similarly, in the Katalyst case we can observe changes in every-day events, patterns and underlying structures – particularly relationships. The changes are significant in as far as they seem to lead to quite some crowding-in of other companies into these markets. We could speculate that there are changes in mental models. The realisation of many input companies that poor farmers in remote areas could be a potentially profitable market for them if they move in early could indicate such a change. But this could also be a short-term strategy as it might not make sense for many companies to crowd into this market that is marked by high transaction costs. So in that sense, we are not sure if there are actually changes on the level of mental models – the case study did also not specifically look for them.
We can, hence, confirm the findings of the AAER framework that these two projects seem to achieve systemic change as they affect changes on the level of the underlying structure in the market systems. However, one could argue that the underlying structure is again layered and relationships might be on top of these layers. Changes in institutions like rules, laws or social norms or changes in practices of actors that take up more systemic functions like organisations involved in education or public R&D are more significant indications of sustainable long-term change.
The iceberg model also allows us to look at the projects mentioned above that focus on changing the institutional set up of an economy rather than introduce innovations into the market system. These projects directly target the structural level of a system. An example would be a project that improves the quality infrastructure of a country, allowing companies to certify their products and get easier access to international markets. While these projects are directly targeting systemic change as an output rather than as a consequence of their interventions, the question about significance of the change remains.
Let’s look at the second concept that I want to introduce to make sense of systemic change.
Folke defines resilience as [6:xx] …
… about having the ability to live with change, and develop with it. It is about cultivating the capacity to sustain development in the face of change, incremental and abrupt, expected and surprising. Resilience is about persisting with change on the current path of development, improving and innovating on that path.
Sometimes, though, a chosen path can become a trap and development needs to shift to a different one. Here, the difference between adaptation and transformation is important, again Folke [6:xx]:
… adaptation refers to human actions that sustain development on the current pathway, while transformation is about shifting development into new pathways and even creating new pathways. Deliberate transformation is about shifting development into new pathways and even creating new pathways.
Hence, the question with regards to systemic change would be: has the intervention built resilience of the market system or economy? Has it improved the system’s ability to live with change, and develop with it?
Let’s look at our two examples. Neither of them has, in my view, drastically increased the resilience of the markets they work in. Yes, they have introduced innovations, but the introduction of these was mainly driven by the projects, who did the problem analysis and might even have incentivised the actors to take up the new ways of doing things. There arguably was some element of co-creation and experimentation and refinement done, but enabling system actors to experiment and become improving self-discovery was not the main aim. There is no significantly improved capacity of the system actors to do the problem analyses themselves, experiment and try things that help them adapt or even transform in the future.
This comes down again to the question of institution-building. Resilience can only be enhanced by having the right institutions in place that can coordinate and guide a deliberation process in a system on how to adapt or transform. Establishing new commercial relationships between seed or seed producers, retailers and farmers can help, but might not do the trick. On the other hand, establishing a strong quality infrastructure with a standard and accreditation body that is able to pull together various actors and discuss and debate about changes in the international market and how to react to these will enhance the system’s resilience.
This is again in line with the definition of systemic change Shawn and I used in our research paper we wrote for BEAM/DFID [7:47]:
Systemic change is most likely to be achieved when influential actors or networks of actors become aware of how change happens, and their role in realising the evolutionary potential of the economy. These influential actors need to develop the capability to engage in, collectively discover and continuously shape their institutional landscape – a process that is most effective when it is done in a transparent and participatory way.
Resilience thinking is a wide and still developing field which is currently focusing on social-ecological systems. I want to dig quite a bit deeper her to better understand its implications for understanding social-economic systems and systemic change in economies. Watch this space.
I don’t want to draw too strong a conclusion here as this post is more about me thinking out loud. But my current take on systemic change is that there should be a strong focus on working on or influencing the institutional level to improve resilience to achieve meaningful systemic change. This can either happen through directly working on an institutional level such as with the quality infrastructure example I used above. Or it could mean starting from a level of events and patterns and ensure that these changes are strong enough the elicit institutional adaptations. This would mean that the ‘respond’ part of the AAER framework explicitly needs to look at responses on the institutional level.
Institutional change is inherently incremental and slow. So there is a natural tension between results donors want to see (often linked to short-term property reduction), which pushes projects to stong-handedly introduce innovations, and long-term strengthening of system resilience, which does not deliver easily visible and measurable results in the short term. This tension will influence the choices made by donors and projects.
 Taylor, B. (2016). Systems and Systemic Change – Clarity in Concept. The Springfield Centre for Business in Development. [accessed 06.02.2018]
 Sen, N. and Hafiz, W. (2015). Measuring Systemic Change – The case of GEMS1 in Nigeria. Donor Committee for Enterprise Development. [accessed 06.02.2018]
 Taylor, B., Lomax, J. and Smith, K. (2016). Katalyst’s contribution to systemic change. Katalyst and The Springfield Centre for Business in Development. [accessed 06.02.2018]
 Osorio-Cortes, L. and Jenal, M. (2013). Monitoring and Measuring Change in Market Systems – Rethinking the Current Paradigm. The SEEP Network, FHI360, USAID. [accessed 06.02.2018]
 Meadows, D.H. 1999. Leverage Points. Places to Intervene in a System. Sustainability Institute. [accessed 06.02.2018]
 Folke, C. (2015). Foreword. In Biggs, R., Schlüter, M. & Schoon, M.L. Principles for Building Resilience. Cambridge University Press.
 Cunningham, S. & Jenal, M. (2016). Rethinking systemic change: economic evolution and institution. Technical paper. The BEAM Exchange. [accessed 06.02.2018]