The case for thinking & working politically

Evidence tells us that domestic political factors are usually much more important in determining developmental impact than the scale of aid funding or the technical quality of programming.
Although international development organisations have made extensive efforts to improve the technical quality of programs, in many cases, these improvements have not led to greater impact during implementation. Successful implementation usually happens when programs are aligned with a domestic support base that is influential enough to generate reform momentum, and overcome the resistance of those benefitting from the status quo. Too many times over the past few decades, we have seen projects fail because they demand changes that are not politically feasible.

These findings demonstrate that an understanding of political dynamics is frequently the critical missing ingredient in project design and implementation. Many influential thinkers[2] have looked at the difference between success and failure in development, and all point to the centrality of domestic politics. Admittedly, this conclusion does not necessarily help to predict how developmental change will unfold in different contexts, and it directly confronts the notion that some institutional models will always work better than others. However, we have learned that progressive change usually involves local political processes of contestation and bargaining among interest groups, and that development programs can significantly improve their impact by understanding and responding to these dynamics. Recent evidence indicates the importance of reform-oriented leaders, who find ways to make progress by facilitating local problem-solving and collaboration among wide-ranging interest groups.

History teaches us that politics is intimately tied to inclusive economic growth, and as such, a major factor in poverty reduction. Meaningful and sustainable poverty reduction requires changes in social structures and in political institutions – changes that will be contested at every step. Every country has to find its own way to translate political power into change for the public good. This is true of all polities. A critical part of this process is the routine, daily struggle over ‘the rules of the game’, which are shaped by emerging coalitions, political mediation, negotiation and compromise, and innumerable calculations of political risk and opportunity.

Meanwhile, traditional ‘gap-filling’ Official Development Assistance (ODA) is fast becoming out-dated. With the emergence of significant new resources from non-OECD donors, the private sector and philanthropists, aid ‘recipient’ countries have many more options than hitherto available for development financing and technical assistance. The influence that donors once sought to wield through conditionality and policy dialogue has largely diminished. Many partner governments are now far more assertive and sophisticated than in the past, in-part bolstered by the Paris, Accra and Busan agendas. Furthermore, with more than 50 per cent of the world’s poor now living in lower middle-income countries, the critical development challenges for poverty alleviation are more than ever a result of domestic policy change and institutional reform rather than small sums of money to fill perceived capital ‘gaps’. While technical knowledge and financing for development are rarely the key bottlenecks to development progress, these are precisely what traditional aid programs are designed to provide.

Over the past decade, development donors have increasingly acknowledged the role of politics, but mainstream operations are only now beginning to change. In the past few years there has been a step-change in the number of donor agencies undertaking analysis of political context and processes, and some adjustment to aid practices to reflect the need to be more responsive to local political economy dynamics. A growing number of donor policy statements clearly situate politics as a critical factor in developmental progress, and commit to programs that are more politically aware. However, the process of translating these insights and commitments into changes in mainstream development practice has been slow and contested. Despite the growing accumulation of evidence and bitter lessons, the majority of development programs continue to use traditional approaches.

However, there are now several efforts underway within the international development community advocating for fundamental changes to the way development assistance is conceived and implemented. Since 2013, there have been a number of new initiatives involving many of the leading thinkers, influential policy-makers, donors and practitioners. In October 2014, Harvard University hosted a meeting to consider ways of ‘doing development differently’. The consensus document produced at this meeting – the DDD consensus or more affectionately, the ‘Harvard Manifesto’ – has been widely circulated, and many development leaders have publically endorsed it.[3] Similarly, since November 2013, a group of senior officials from major donors, along with a few leading thinkers and researchers, have been working together to promote thinking and working politically (TWP) in development, with a particular focus on what donors can do to allow this to happen.[4]

Notwithstanding this progress, changing aid practices has proven much more difficult than raising levels of knowledge and awareness among donor staff, undertaking ‘set-piece’ political-economy analysis, and drafting more nuanced policy statements. The dramatic expansion of political-economy analysis over the past decade has not transformed the delivery of development programs, and has had a limited effect on development impact. This is probably due to the fact that much aid remains predominantly technocratic, inflexible, and averse to the types of operating approaches that could translate political-economy findings into more effective development practice.[5] In-country front-line program staff are obliged to follow the (legitimate) rules and regulations of their parent departments – which rarely admit flexible and responsive disbursement of funds. Logical frameworks (the predominant management tool for program implementation) generally incentivise rigid, linear program logic, which does not reflect reality in developing countries and makes it difficult for program managers to adapt to changing circumstances. Collectively, these factors reinforce traditional development approaches, and create obstacles for development professionals attempting to do development differently.

So what does a ‘doing development differently’ agenda look like? The outline of what such approaches may look like is now becoming clearer. The aim of the Harvard meeting and the TWP initiative has been, in some ways, to formalise the progress being made, incrementally, donor by donor, country by country, project by project. The TWP and DDD agendas are driven by three core principles:

  • strong political analysis, insight and understanding;
  • detailed appreciation of, and response to, the local context; and,
  • flexibility and adaptability in program design and implementation.
Principle Characteristics
1.     ANALYSIS: Political insight and understanding ·       Interrogate the project, and the sector with a relentless focus on power dynamics, interests, incentives, and institutions.·       Be frank about where power resides and on whose behalf it is being used.

·       Move away from idealised models of development change, and start with contextual realities.

·       Recognise the multiple (and potentially contradictory) nature of interests at play.

·       Focus on problems identified and articulated by local actors, not outsiders.

·       Ensure (as far as possible) that locally-defined problems and proposed solutions are accepted as legitimate by all relevant stakeholders, thereby ensuring ownership.

2.     CONTEXT: Responsiveness to domestic environment ·       Work with and through domestic stakeholders, convenors and power-brokers (also referred to as ‘arm’s length’ aid).·       Understand the network of stakeholders involved and facilitate coalitions of different interests, rather than relying on a ‘principal-agent’ relationship with one Ministry / Minister.
3.     DESIGN: Flexibility and adaptability in design and implementation  ·       Be guided by the program goal, and do not be overly prescriptive in how to achieve it. Strategy should set a clear goal, allowing for significant flexibility and iteration in the day-to-day efforts to make progress towards these goals. Clear goals should not translate into rigid project frameworks – they represent an understanding of what changes you are hoping to promote.·       Recognise that politics are not static – continue to assess the local context, test original assumptions, and adapt programs based on new information and opportunities.

·       Merge design and implementation with a focus on a series of small ‘experimental’ or ‘incremental’ steps and monitor results. In this way, implementation and monitoring & evaluation become one concurrent process.

·       Periodically engage in ‘review and reflection’ exercises to critique and understand what is working and what is not – and stop doing what does not work.

·       Understand your own agency’s political-economy – which issues can be negotiated and which ones cannot.

‘Politically smart’ development assistance combines political-economy knowledge with more responsive, adaptable and contextually relevant operations. These approaches are grounded in a growing body of research and experience (see page 1). There is less reliance on aid conditionality and comprehensive institutional reform, and more emphasis on the need to build on local motivation and capacity, responding flexibly to events and opportunities as they arise. This includes removing any design `straight-jacket’ stemming from program design tools that encourage prescriptive approaches.

Thinking and working politically’  is neither a silver bullet nor a passing fad; it reflects a new resolve to learn from years of well-intentioned but often unsatisfactory aid practice, grounded in mistaken assumptions about the ability of external actors to drive complex processes of change by supplying finance and technical advice. The ambition should be to tailor aid programs to the growing body of evidence about how change happens and what kind of approaches work, and to strengthen the evidence base through better piloting, monitoring and evaluation.

Progress is needed across the broad spectrum of aid programs – from large ‘traditional’ sector programs, to small and nimble reform initiatives. The next critical challenge is to influence the practice of larger-scale programs that necessarily require greater structure and planning. This means integrating a political lens, allowing greater room for manoeuvre during implementation, and consideration of governance constraints in all development assistance programs – from health and education, to infrastructure and climate change.

Our goal should be to encourage political awareness in all aid programs, while creating space for a significant expansion of explicitly TWP (and DDD) programs. Indeed, it is probable that only a modest percentage of ODA funded initiatives will be fully iterative, adaptive and flexible – and these initiatives will be mainly in areas of policy, institutional or governance reform. However, TWP is not a ‘governance’ solution to be applied only to a narrow set of institutional issues (public financial management or civil service reform for example). On the contrary, TWP is an approach to improve delivery of any aid program that involves reform and behavioural change – it is as relevant to better delivery of health services or economic policy reform as it is to an anti-corruption initiative. TWP takes the naivety out of institutional relationships by understanding that change happens as a result of decisions that invariably have a political dimension.



[1] Contributors to this note include (in alphabetical order): Sakuntala Akmeemana, David Booth, Deryck Brown, Diana Cammack, Marta Foresti, Lawrence Garber, Duncan Green, David Hudson, Stefan Kossof, Heather Marquette, Neil McCulloch, Alina Rocha Menocal, Michael O’Keefe, Thomas Parks, Graham Teskey, Sue Unsworth, Alan Whaites, Lisa Williams.

[2] See for example: Daron Acemoglu and James Robinson (2012) ‘Why Nations Fail’. New York: Crown Books; Matt Andrews (2013) ‘The Limits of Institutional Reform in Development’, New York, Cambridge University Press; Thomas Carothers and Diane de Gramont (2013) ‘Development Aid Confronts Politics: The Almost Revolution’, Washington DC, Carnegie Endowment; Francis Fukuyama (2012) ‘The Origins of Political Order’, New York, Farrar, Straus and Giroux; Francis Fukuyama (2014) ‘Political Order and Political Decay’, New York. Farrar, Straus and Giroux; Douglas North, John Wallis and Barry Weingast (2009) ‘Violence and Social Orders’, New York, Cambridge University Press; and Dani Rodrik (2007) ‘One Economics, Many Recipes’, Princeton, NJ, Princeton University Press.



[5] Carothers and de Gramont, cited above, make a particularly strong argument on this point.