Introduction
In South Africa, discussions of social cohesion and social impact tend to happen in silos. Civil society organisations and some government departments (Arts, Culture & Sports; Cooperative Governance; Justice; Home Affairs) talk about social cohesion as related to nation building or xenophobic violence, while social impact is a buzz-word in the corporate sector and other government departments and agencies (Minerals & Energy; Trade & Industry; Treasury; IDC). In fact, we should be bringing insights from social cohesion discussions into the social impact sphere. First, this is because social interventions aimed at improving wellbeing must be achieved in ways which are cohesive, or else they run the risk of increasing conflict potential and indeed doing harm. This is discussed elsewhere. Second, there are related insights in both the social cohesion and social impact fields about planning and implementation and the monitoring, evaluation and learning processes required. The latter is what I will discuss here.
In both arenas, there has been a major, but still unevenly applied, shift from thinking about the end (what does a cohesive community or a socially well-off community look like and how do we know when we have arrived) to the means (what is the quality of processes and institutional forms that help us take decisions and work together and how do we know that we are moving forward together in productive ways). An important component of this shift is integrating uncertainty, adaptive management & learning and continuous change management into how resources for social development are allocated and how development activities are designed, implemented and monitored.
I will unpack the implications of this nascent shift for monitoring, evaluation and learning (MEL) systems for social cohesion and social impact interventions. An MEL lens forces us to be explicit about our intended aims, the key characteristics of what we are trying to achieve, what matters and to/according to whom. I will use illustrative examples from the renewable energy sector in South Africa, which is currently considering new systems for cohesive social impact, having learned from the extractives and fossil fuel sectors.
The Shift to Adaptive Management and Systems Change
Before discussing the MEL implications, let’s establish when, where and why the shift to adaptive management has occurred in the social development field. Adaptive management is not particularly new or innovative as a concept. It is in many ways simply good project management and is taken for granted in the scientific, technology and corporate sectors as the core of scientific process (hypothesis testing), research & development, and agile management. Global scientific collaborations around climate, natural resource management and health (not least in relation to pandemics) have long brought together large and diverse groups of stakeholders, organisations and individuals for fast iterative learning and problem solving. Within South Africa, various methodologies have been used for some time to bring multiple stakeholders together around a common social problem to collaboratively and iteratively develop and test new solutions. Social Labs and Organisation Workshops are only some of the methods used.
Adaptive management approaches have become prevalent in both the corporate and the social development spheres due to the increased recognition of complexity and uncertainty as core context factors for any major social challenge. Simply defined, complexity means that no single actor or sector can solve the problem or impose a solution on their own, requiring a collaborative approach. Uncertainty means that either full information is not available to inform decision-making or that situations change so quickly that plans quickly become obsolete. Therefore, “adaptive management [is] a systematic response to ongoing uncertainty which avoids the two extreme options of either giving up trying to plan in conditions of uncertainty or rapid change, or trying to use tools and processes that are designed for more well understood, stable and predictable contexts.”[1] The adaptive principle has quickly become mainstream in international development practice, with USAID, UKAid, AusAid, and the World Bank, among others, all publishing guidance notes and policy papers. Various academic and practitioner networks around monitoring and evaluation have also developed extensive resources and case studies on adaptive management and MEL.
In the corporate sphere, the connection is also increasingly being made between social impact and systemic change management. In order to achieve the goal of sustained positive social impact, major universities such as the Bertha Centre for Social Innovation and Entrepreneurship are emphasising this link in their executive training programmes which go beyond technical matters like impact measurement to teaching process tools to “navigate complexity and strengthen … capacity for … collaboration.”
In summary, key components of adaptive approaches in both the private and social development sectors are being iterative (“rather than building a whole solution straight away, these approaches commonly encourage practitioners to start small and use structured cycles of testing and learning”)[2] and focusing on process (“adaptive approaches in development provide a wider range of options for what to create and facilitate – not only products or services, but also forms of collective action.”)[3] An additional focus for the social development sector is inclusive collaboration across the full diversity of affected actors, based on the complexity principle.
Implications for MEL of social cohesion and social impact
What are the implications of an iterative and collaborative process focus for monitoring, evaluation and learning systems around social cohesion and social impact interventions? The shift impacts on what is measured, when measurement happens, who does the measuring, how data is shared and used, and who takes decisions based on the data.
I will use examples from the renewable energy sector in South Africa to illustrate each point. The renewable energy companies included in the government’s ambitious Renewable Energy Independent Power Producer Procurement Programme (REIPPPP) are contractually obliged to contribute to local social and economic development in the areas where power generation projects are located. Over the 20-year duration of the Power Purchase Agreements concluded to date, the value of social commitments is R1.17bn. The challenge is how to utilise and distribute these resources in ways that result in sustained social development, within the context of rapid climate change and economic and political uncertainty, and without producing the kinds of social conflict and perverse social and environmental impacts generated by equivalent social investments in the mining sector.
The IPP sector in South Africa is currently at the stage where a number of sectoral lessons learning and thinking processes about social impact have started or are under way, but where the requisite institutional cultures, practices and policies are not yet embedded across the sector.
What is measured
Good MEL measures what is most important in a process’ theory of change – both what we are trying to achieve and how we intend to get there. As iterative processes and collaborative decision-making become central, and especially as the ends defined by these means are not yet clear, our MEL must focus on process indicators. Regularity of engagement, diversity of participants, collective conflict resolution capacity and quality of collaboration are all challenging but measurable indicators which can apply to the IPP sector. These indicators apply both to activities aimed at increasing social cohesion and activities aimed at improving social development in socially cohesive ways. As the iterative collaborative processes develop specific target outcomes for specific activities (such as health care projects or biodiversity protection agreements), measures for these are added to the MEL framework. The negotiation of these indicators becomes part of the iterative process rather than being an up-front unilateral design decision.
When does measurement and decision-making happen
The iterative logic of adaptive management changes MEL timelines from linear to circular. Rather than establishing indicators and targets up front and measuring change at the end of an intervention, often several years later, adaptive management both shortens and extends MEL timelines. Timelines for the cycle of indicator definition, data generation and reflective learning are shortened, so that decisions can be made to change intervention focus if necessary. At the same time, timelines for initial relationship building and planning, collaborative decision-making processes and learning activities are extended, since these are no longer expert-driven or desk-top exercises but are understood to be necessarily messy and drawn-out. As noted in a Practice Note from the REIPPPP lessons learning event on social impact “travelling fast means covering shallow ground only; to successfully travel slower and ‘deeper’, we must allow for more time to move from planning to implementation, and finally to reporting.”
The agility element of adaptive management is often associated with faster decision-making, which has been criticized as potentially endangering some of the community consultation processes required for large-scale development interventions. The argument here is in fact the opposite, namely that adaptive planning and monitoring should be ready to extend the time for institution-building, capacitating collaborative processes and building experiential trust through iterative small projects. As quoted above, adaptive management allows us to think about a ”wider range of options for what to create and facilitate, [including] forms of collective action” which in the IPP case can include interventions in collaborative leadership training for CBOs, municipal officials and IPP staff; participatory community-based asset mapping; building data sharing platforms and similar forms of process facilitation. This should not be thought of as delaying the ‘’actual work’’ of spending development funds. Rather, this is the actual work and should be invested in and evaluated as such.
Who does the measuring
Indicators which measure the quality of processes and relationships require reflective engagement by the participating stakeholders. The reflection process can be facilitated by professional MEL expertise, but is less an ‘’independent’’ external data collection exercise and more an integrated part of the relationship building process itself. Furthermore, as collective decision-making results in resource expenditure and project implementation, systems through which beneficiaries can report on their relationship with the projects on an ongoing basis (an integrated monitoring system) are central to learning, trust-building and accountability through the projects. The REIPPPP learning event included recommending that “it is beholden on IPPs to capacitate practitioners to engage meaningfully in monitoring and evaluation (M&E) methodologies” and indeed capacitating themselves, rather than considering monitoring to be predominantly an external technical function.
Data coordination, sharing and transparency
The goal of monitoring and evaluation is learning, which in an adaptive model should be collective learning across the various stakeholders. Therefore the monitoring data on which learning is based must be transparently shared. In the IPP sector, this includes data sharing across multiple energy producers working in the same region, with the local and district municipalities and with local NGOs and community-based organisations. Even as specific local interventions develop their own impact and monitoring indicators, the need for collaborative learning implies some level of coordinated data framework (at least through shared process indicators and high level impact indicators) as well as a generally accessible data platform. An industry-wide impact measurement framework, achieved in a “collaborative fashion” would “facilitate wider adoption of M&E, synergise indicators, and allow for reporting on an aggregate impact across the industry”, according to the REIPPPP learning event.
It is also valuable for the IPP sector to structure its data sharing in ways which allow for learning and comparison across sectors and regions, by, for example, linking with social impact measurement happening in parts of the mining sector. By integrating with existing lessons and frameworks for measuring social cohesion, such as those developed by the African Centre for Migration and Society, the IPP sector can share learnings with broader national discussions on social cohesion, as one of the core governmental and National Development Plan priorities.
Who takes decisions and on what basis
The purpose of MEL is to inform decision-making, which is especially the case in an adaptive management model. The key question, then, is who decides. When looking at corporate-resourced social development from a social cohesion perspective, as building a social licence to operate, then collaborative multi-stakeholder planning and monitoring must also include shared decision-making which includes the authority to change the direction of resource allocations. Collaborative decision-making is difficult, and personalised short term interests in controlling resources can easily undermine achieving broader and longer term collective benefit. However, the answer to this challenge is not to restrict decision-making but rather to expand access and transparency around decisions and the data on which they are based, linked to previously agreed goals and impact indicators, so that those pushing individual interests can be held accountable collectively.
In addition to transparent collaborative decision-making processes, adaptive management also requires an enabling policy environment in terms of IPP reporting requirements and incentives. If governmental reporting requirements for IPPs only allow for certain kinds of social expenditures (not including the funding of process facilitation and MEL activities, for example), insist on carrying through planned ‘’commitments’’ without space for adjustments or reallocations, and accept input and output indicators (amount spent, numbers trained, buildings erected) over process and impact indicators, then adaptive management and cohesive social impact is disincentivised.
Conclusions
Adaptive management processes are difficult to implement. They require time, funding, dedicated people with the right skills, and patience. They require an enabling policy environment. And they require the emotional and institutional willingness to expect and look clearly at failures, rather than choosing safe options or uncritical evaluation methods. For people in government and in companies used to thinking about social impact in terms of what is to be achieved rather than the quality of processes for achieving it, taking on board a social cohesion lens helps understand the importance of spending time on processes and relationships and the dangers of under-investing in process. As we argue here that taking process seriously is a necessary pre-requisite for cohesive social development, it is not on its own sufficient – the means need to point towards and finally achieve an end, which is cohesive social wellbeing.
[1] Rogers, P. and Macfarlan, A. (2020). What is adaptive management and how does it work? Monitoring and Evaluation for Adaptive Management Working Paper Series, Number 2, September. Retrieved from: www.betterevaluation.org/monitoring_and_evaluation_for_adaptive_management_ series
[2] Pett, J. (2020). Navigating adaptive approaches for development programmes. A guide for the uncertain. ODI Working Paper 589, Retrieved from: https://www.odi.org/publications/17367-navigating-adaptive-approaches-development-programmes-guide-uncertain
[3] ibid
