As previous posts on the Impact Blog have highlighted, one aspect of the impact agenda that has until recently been relatively neglected has been that of negative impact, or ‘grimpact’. In this post Valeria Izzi and Becky Murray draw on examples from development practice and research to advance a more complex understanding of grimpact and argue that as development research moves increasingly from theory to practice, there is a need to re-engage with these debates in development studies.

Most discussions about impact are underpinned by the assumption that research uptake and use are inherently positive. Typically, impact case studies show how research has been picked up by policy-makers, assuming a positive (if slow and tortuous) trajectory from there. In other words, the worst-case scenario would be no impact at all. Failure to consider the possibility of unintended negative impact remains a blind spot in the debates so far, as discussed in recent posts by Andrew Crane and Gemma Derrick and Paul Benneworth.

From the perspective of international development practitioners, these concerns about ‘GrImpact’ certainly sound familiar. A large body of literature relates these potential negative effects. At the macro-level, these range from the ‘Dutch disease’ phenomenon (by which large volumes of aid stifle productivity, in a similar way to the sudden influx of income from natural resources) to the question of aid fungibility (with aid ‘freeing up’ government resources for other –non-development- aims) and its potential in fuelling corruption. The volatile nature of aid can make the budgeting process difficult, and hamper long-term planning. Even more fundamentally, aid can affect the state-citizen relation by shifting the channels of government’s accountability from the internal sphere (towards citizens) to the external sphere (towards donors).

At the micro-level, development projects can become a field of covert struggle between various actors. Well-meaning interventions can unintentionally raise unrealistic expectations, fuel inequalities, strengthen patrimonialism, and commodify social relations. Money brings power, and creates winners and losers: whether an intervention is a ‘success’ or a ‘failure’ is – at some level – in the eye of the beholder. When relying on catch-all labels such as ‘local communities’ or ‘the poor’, development actors remain oblivious to these local dynamics. In extreme cases, such artificially ‘apolitical’ approaches can even fuel violence, as Peter Uvin famously argued in the case of the Rwandan genocide.

A growing share of UK academic research is funded by Official Development Assistance (ODA), and is intended to benefit poor communities in developing countries.  As academic research ventures into the field of ‘doing development’ (as opposed to just ‘studying development’), or as in the case of other disciplines where increasingly engaged research methods are being adopted, researchers will inevitably be faced with the same dilemmas that have long confronted development practice. In other words, Research for Development finds itself at the crossroad between the well-established notion that ‘aid can do harm’ and the emerging awareness of Grimpact for academic research. This poses crucial dilemmas for both research teams and funders, in terms of the amount of ‘political risk’ that they are willing to tolerate. It also leads to questions about the extent to which ‘negative impact’ for one particular group can be justified if outweighed by longer-term common good.

In an apparent paradox, these risks can be even higher for research that is co-produced with stakeholders and intended research users, as has also been discussed here. While co-production is widely presented as a benchmark for impactful research, it also has hidden costs. In the absence of adequate time, flexibility and will for compromise, co-production may become just a fashionable way of rebranding (and legitimising) extractive research practices. It can also give disproportionate representation to certain individuals or groups, and be taken over by the ‘usual suspects’. Research outputs can be co-opted to serve particular agendas and the self-interest of powerful groups.

Furthermore, the search for ‘intermediaries’ and ‘knowledge translators’ at the local level can empower some individuals (and their agendas) in ways that it is not easy for external actors to grasps. Multiple iterations with a community can strengthen the quality and value of research findings, but can also lead to saturation and research fatigue on the part of local participants. In most cases, local communities are not likely to feel the benefits of research in the foreseeable future, or at least not at a scale that they might attribute to the project. As a minimum, returning to local communities to share research results can play an important role in promoting trust. However, both the budget and the timeline of many research programmes often do not allow for going back to remote locations for community feedback.

Community involvement does not lend itself to standardised tick-box procedures of ethical reviews. It blurs boundaries around intellectual property, and calls for careful consideration of how various contributions should be acknowledged, attributed and rewarded. The conventional principle of informed consent may be ill-suited to addressing community-level concerns. Similar considerations apply to the issues of anonymity and safety. For example, research on the use of natural resources deals with very sensitive questions, and can often unveil illegal activities, such as illegal entry into protected areas to collect firewood and water, or unsustainable agricultural practices that are against local regulations. Guaranteeing individual anonymity does not automatically mean that no harm will come to the community as a whole. There are clearly difficult decisions to be made: while research methods must be transparent and information about the location of the research is often important for the interpretation of the results, protection of participants should be a paramount consideration.

ODA-funded research, like all forms of aid, interacts with local dynamics in ways that are difficult to predict, let alone control. An overly risk-adverse attitude may jeopardise the value of critical analysis and open-ended scientific quest. While there are no off-the-shelf solutions about how to minimize harmful impact, the question needs to be put forward more forcefully than it has been the case so far. One place to start and a rich seem of insights stemming from two decades of learning is international development practice.


We are interested in developing and continuing this discussion in developments studies further; if you would like to share your experiences of unintended negative impact we would welcome you to get in touch.

About the authors

Valeria Izzi is an independent consultant working on policy-oriented research, impact and theory-based Monitoring, Evaluation and Learning for international development.

Becky Murray is a Senior Consultant with LTS International, where she leads on impact strategy, programme management and Monitoring, Evaluation and Learning for effective R4D investments.


Note: This article gives the views of the authors, and not the position of the LSE Impact Blog, nor of the London School of Economics. Please review our comments policy if you have any concerns on posting a comment below

Image Credit, Micaela Parente via Unsplash (Licensed under a CC0 1.0 licence)

Print Friendly, PDF & Email