Alan Hudson, Executive Director
December 5, 2017
Learning and adaptation are high on the governance and development agenda (see for instance the World Bank’s World Development Reports for 2017 and this video for the 2018 Report). Our strategy at Global Integrity is based around the hypothesis that learning-centered and adaptive approaches can play an important role in addressing governance-related challenges that are complex, political and context-dependent. So we very much welcome the attention that is currently being given to learning and adaptation.
However – despite careful consideration of the value of collaborating, learning and adapting, and welcome attention to issues around “the locus of learning” from USAID – we are concerned about a lack of clarity in many discussions about learning, and find ourselves always wanting to ask:
- Whose learning are we talking about?; and, most importantly,
- How are the processes and products of learning expected to have an impact on the incentives and power dynamics that are at the heart of governance-related challenges?
A recent event on “supporting politically smart and adaptive USAID programs” (agenda here) gave me an opportunity to address – or at least to air – some of these concerns. The event, organized by Palladium and DAI, aimed to explore the challenges and opportunities in designing and implementing more adaptive and politically smart programs. It was a great event, with the overview provided by Sarah Swift of USAID’s Center of Excellence on Democracy, Human Rights and Governance, and the rich country experience provided by Adiya Ode, Austin Ndiokewelu (both working with the Partnership to Engage, Reform and Learn (PERL) Nigeria), and Julie Lostumbo (Local Enterprise Value chain Enhancement (LEVE) program, Haiti), about the facilitative and politically smart approaches taken in those programs particular highlights.
My contribution was on a panel on monitoring, evaluation and learning for politically-smart and adaptive programming, alongside Monalisa Salib (USAID LEARN), Imara Crooms (IRI) and Drew Koleros (Palladium). I kicked off by explaining that while Global Integrity is not an implementer of big governance and democracy programs, we are in the business of supporting the locally-led innovation, learning and adaptation that we think is key to address complex political and context-dependent challenges.
We do that through combining two strands of work. First, our country-level work with local partners, where we bring our expertise around data, learning and politics – and our position in broader discussions about governance and development – to facilitate collaborative efforts to understand and address governance-related challenges. And second, our engagement with actors such as the World Bank, the Millennium Challenge Corporation, the Open Government Partnership, and USAID, where we deploy insights and evidence from our country-level work, to support and encourage their efforts to work in ways that put locally-led innovation, learning and adaptation center-stage.
Our strategy and associated learning plan give attention to three interconnected levels of learning (giving me the opportunity to use an image of Star-Trek Three-Dimensional chess!): learning by country level actors; learning by external actors; and, our own organizational learning. All three levels and their interconnections matter, but in our view, the primary focus needs to be on learning processes that can enable country-level actors – the actors who really understand the context and are best-placed to drive effective and sustainable reforms – to navigate and shape the political dynamics and incentives around locally-prioritized problems. What flows from this is a number of characteristics, or principles, that learning has to have if it is to have any hope of shaping these dynamics and incentives.
A first pass at these principles is as follows:
- Learning should accelerate, and focus on, progress towards particular outcomes. Learning that lacks a purpose does little to drive results;
- Learning should inform decisions and influence actions. Learning that does not inform decisions or influence actions has little impact;
- Learning should go hand in hand with implementation, on a rolling basis. Learning and analysis that takes place only before a program starts, or after a program ends, does little to support effective implementation;
- Learning should be fueled by data, including data generated through monitoring and evaluation, so that it’s informed by and aligned with local realities. Learning that is not informed by data about facts on the ground, is unlikely to lead to effective approaches;
- Learning should involve reflection on the political challenges and opportunities for reform, and the incentives faced by different stakeholders, in order to inform the design and implementation of effective reforms. Learning that neglects politics, power and incentives will not address the root causes of social problems; and
- Learning should be inclusive, participatory and empowering, so that it has the potential to strengthen and broaden the local ownership of reforms, and to shift the power dynamics and incentives of the system. Learning processes that are restricted to powerful players are unlikely to lead those players to change the rules of a game that they are already winning.
We endeavour to put these principles into practice, in different ways, across everything we do, including: our support for the learning journeys of civil society organizations in Africa and Asia; our work with Mexican partners to understand and improve the fiscal governance landscape; our ongoing engagement with the Open Government Partnership about the potential of data-driven learning cycles; our work on the use of data – including governance data – to support collaborative learning around governance-related challenges; and, our leadership of the space for learning and collaboration that is the Open Gov Hub.
Putting these M/E/L principles into practice, in order to support change in complex political systems, entails a number of challenges. These include the following:
- First, it’s a challenge to persuade funders to invest in approaches that, by putting locally-led innovation, learning and adaptation center-stage, relinquish some of their (sometimes illusory, sometimes unhelpful) control, with the promise, but not the guarantee, of that leading to better outcomes. Relatedly, it’s a challenge – and one that we should not and can not duck – to persuade funders that investing in the capacity of a DC-based organization to support others’ learning makes sense;
- Second, while we’re determined to link and sync our levels of learning and to make the case that collaborative learning and action is the path to sustainable results, it can be a challenge to balance the need to meet our own organizational learning needs and our need to report to funders about what difference their investments are making, with our imperative to support and prioritize local actors’ learning;
- Third, and remembering I was on an M/E/L panel, it’s a challenge to pick indicators that are relevant to outcomes AND that we can make a plausible claim to have affected AND that can be measured; not an insurmountable challenge if you’re not hung up on attribution and are comfortable with qualitative indicators, but a challenge nonetheless; and
- Fourth, the most important challenge is that of designing and facilitating learning that – by getting the process, the participants, the content, and the location right – really enables local actors to navigate and shape the political dynamics around governance-related challenges.
Finally, I concluded by making the point that adaptive programming has two related but distinct aspects. One way in which programming can be adaptive is when it involves funders and implementing organizations such as DAI and Palladium adapting their approaches in response to emerging challenges and opportunities. A second way in which programming can be adaptive is when it supports learning and adaptation by the local actors who – if they have the autonomy – are best placed to make the decisions, and drive the action, that is needed to address complex, political and context-dependent challenges.
The first aspect of adaptive programming is important. The second is key if we really want to support and leverage the power of local actors to shift the dynamics of complex political systems and reshape the landscape of incentives. Otherwise, lots of talk about learning and adaptation will deliver little more than better managed programs that have little purchase on the political dynamics and incentives that lie behind complex governance and development challenges.
The question, to quote Robert Chambers on “learning to adapt”, is: “whose adaptation really counts, and how can we empower actors on the ground to put their adaptation first?” If learning-centered and adaptive approaches are to make a difference, it seems to us that they have to empower the relatively powerless and/or shift the incentives of the relatively powerful. We look forward to continued collaboration as we work with local partners, grappling with the complexities of power and learning, in order to address governance-related challenges, in ways that drive progress towards sustainable results.