Analyse this: getting better at understanding how market systems work

Are our market systems analyses good enough to identify the underlying causes of poor performance? A recent critical review of the ILO Value Chain Analysis reports suggests room for improvement.

The International Labour Organisation (ILO) follows a diagnostic process to understand value chains. As set out in guidance on Value Chain Development for Decent Work ‒ now in a recently-released second edition ‒ the ILO seeks to go beyond traditional value chain analysis (VCA) to take on a systemic lens: one that sees value chains at the core of a market system, where each link in the value chain (from raw materials through production, processing and distribution to retail) is part of a wider network of influencing functions and rules that should be the integral focus of analysis. A good market systems analysis can help interventions get beyond addressing narrow, firm-based issues and instead tackle those problems that are considered underlying and systemic.

In 2014, we assessed a sample of VCA reports in the ILO. Our suspicion was that many projects were failing to achieve intended outcomes because they did not understand the causes of the problems they set out to address. We were right. Over 90 percent of the VCA reports reviewed contained only a superficial level of analysis. These reports successfully mapped the value chain and identified some symptoms of under-performance, but just one report out of eighteen looked at system-level constraints in supporting markets to determine reasons why the value chain might be underperforming. Not a single report managed to provide a springboard from analysis to action by prioritizing the most pressing constraints which could be tackled by the project.

This failure to distinguish between symptoms and causes has important implications. As projects move from understanding the core value chain to supporting systems, the potential for larger and more lasting impact increases. And if VCAs provide an inappropriate entry point for interventions, then superficial analysis risks leading to superficial action. Analysis, of course, is also just a means to an end: a systemic VCA needs to not only identify constraints, but also to say which of them are the most feasible and impactful to address. This, in turn, feeds into the project’s vision for how the market system needs to change to become more inclusive for target groups. As noted in the BEAM Exchange guidance, not doing this means projects risk putting energy in the wrong places.

Fast forward two years later, where are we now? There has been a marked improvement in the diagnostic work, even if we have not yet quantified the change. This has been helped by having dedicated market systems development projects in Zambia, Afghanistan and Timor-Leste – the ILO’s next wave of systemic projects after ‘Enter-Growth’ in Sri Lanka in the mid-2000s, which was one of the first initiatives to adopt a market systems approach. Through our Swiss SECO-supported innovation lab, we have also run dedicated market systems analyses for projects looking at construction in Zambia and Mozambique, tourism in Mexico and grain crops in Peru. But there is still a way to go. We need to:

• Decrease outsourcing analysis to consultants. Finding good consultants who understand the approach and the sector is a challenge; and anyway ‒ it risks outsourcing project understanding of a sector to third parties.
• Increase staff capacity to carry out market systems analysis. To this end, we plan to re-vamp the ILO’s value chain training and get even more staff on the frontline conducting research themselves, not just writing terms of reference for others.
• Ring-fence time to do some initial research before rushing in to intervene. Dedicated inception phases can allow ample time to set up initial diagnostic work, which is often quite time-consuming (though, of course, we have to be careful not to fall into the paralysis by analysis trap!).
• Keep testing different ways to do this kind of analysis. A more action research-oriented approach, as we found in Peru’s wooden furniture sector, can help generate faster insights based on real-world experience, rather than just theory.
• Recognise the incentives for conducting a good market systems analysis: if a project is tool-led, it will not be analysis-led. The lessons from Enter-Growth still ring true a decade later: the process of designing and implementing a project needs to start with the market system and addressing the underlying causes of market weaknesses; it should not start with or be led by the tools themselves.
• And, as the M4P Operational Guide says, keep stressing that diagnosis is not a one-off task. We need to stop seeing market systems analysis as a deliverable, and more as the beginning of a journey towards better understanding the complex systems we work in.

To find out more about our work, download the Executive Summary of the VCA review and read some of the new market systems analysis, visit: www.ilo.org/thelab.

Matt runs the ILO’s ‘markets systems development for decent work – the lab’ project, and Daniela is a junior technical officer at the lab project.