A decade or so ago, virtually no one in the main aid agencies was talking or thinking about politics. There was concern that widespread corruption in aid receiving countries was impeding development progress, and donors were supporting interventions to strengthen the capacity of formal institutions and promote better governance. But with some limited exceptions (for example conflict analysis in Nepal, or early drivers of change studies in Nigeria) there was little attempt to understand the underlying causes of weak public institutions and lack of support for policy change.
Ten years later, virtually all the main agencies have developed frameworks for political analysis, at country and sector level. They go under different names (drivers of change, Strategic Governance and Corruption Analysis, power analysis, problem-driven analysis, institutional and context analysis, political economy analysis) and are tailored to the preoccupations of individual agencies. But they are underpinned by a set of shared concepts, prompting analysts to investigate the way historical trajectories, geopolitical factors, deeply embedded social and economic structures, and formal and informal institutions all shape the political system and the incentives and behaviour of different groups and individuals within it. There is a particular focus on the relationships between political and economic institutions and actors. These frameworks overlap with others developed for conflict analysis or different aspects of social analysis, and can be used alongside them.
Today most development practitioners pay at least lip service to the idea that politics plays a central role in supporting or impeding development. Crude assumptions about the ability of aid and conditionality to stimulate good governance have given way to the concept of good enough governance; discussion has shifted from pursuit of best practice to a search for good fit. The 2004 World Development Report put the spotlight on the politics of service delivery and stimulated an explosion of donor initiatives to strengthen social accountability and civil society demand for better services. The OECD has issued detailed guidance on working in fragile and conflict affected states, emphasising the need to understand the underlying political settlement and the role of legitimacy in state building. The conversation about corruption has become more nuanced, recognising the links to levels of political competition and the role that patronage plays in maintaining political and social stability. A growing body of research  has drawn attention to the importance of informal institutions and local political processes in facilitating or impeding collective action needed to support developmental change.
In short, compared to ten years ago, a much better informed, politically aware conversation is taking place among development practitioners. The striking thing, however, is how limited the impact has been on what the main aid agencies actually do . Donors have found it hard to move from better analysis to better practice.
A common pattern is for staff of donor agencies – especially those based in developing countries – initially to embrace political economy analysis as a welcome opportunity to legitimise and structure the tacit knowledge they already have about the impact of politics on local development processes. This is often followed by disillusion as the gulf between what the analysis is telling them and the political economy demands of the aid business become apparent. The analysis reveals the depth of the challenges involved in building effective states and public institutions, and the limited influence of outsiders seeking to promote progressive policy change in clientelist systems where there are few incentives for politicians to deliver public goods. This reality undermines the optimistic – and simplistic - narrative that sustains the aid business, including assumptions of a shared agenda with development “partners”, and the scope for aid and policy dialogue to contribute directly to transformational change.
Donors have also encountered problems at a practical level: it can be hard, especially in conflict affected contexts, to assemble evidence about very informal relationships and practices; and many of the findings are politically sensitive, raising awkward questions about whether or not to make them public. The quality of studies varies considerably, reflecting inadequate investment of time and resources; and analysis is too often seen as a one-off exercise. The nub of the problem is that donor agencies tend to see political economy analysis as a toolkit to improve aid programming rather than a starting point for investigating more honestly how development processes are unfolding over time in a given regional, country or sector context. A preoccupation with aid and the role of donors gets in the way of recognising that most progressive change is cumulative, unpredictable, highly political and needs to be locally led.
Where donors have made good use of insights from political analysis this is often because they have adopted quite unconventional ways of working.  They have stood back and allowed local actors to take the lead, provided flexible funding to support iterative problem solving, invested intensively in building relationships that rely on shared interests not the prospect of large amounts of aid money, and made long term commitments. These cases show that donors can work in ways that are politically smart and locally led if there is a supportive environment within their own agencies. But they are the exception rather than the rule, and recent trends in donor practice including an emphasis on demonstrating short term results, meeting ambitious spending targets and reducing administrative costs all work against them.
There is mounting evidence from research and practice that aid which is not politically informed is likely to be ineffective, and may well be harmful – especially in countries struggling to contain violent conflict. However the problem for donors is that political economy analysis lays bare the flaws in technocratic, aid centric approaches to development that have long characterised mainstream practice. The tension this creates helps explain the ebb and flow of interest in political analysis within donor agencies. The good news is that when donors do take time to understand the complex political reality underlying current development challenges, they can contribute to finding solutions that are both technically sound (if not optimal), and politically feasible.
 For example, Centre for the Future State (2010) An Upside Down View of Governance, Brighton: Institute of Development Studies; Booth and Cammack (2013) Governance for Development in Africa, London: Zed Books.
 Carothers and de Gramont (2013) Development Aid Confronts Politics: the Almost Revolution, Washington D.C.: Carnegie Endowment for International Peace.
 Booth and Unsworth (2014) Politically Smart, Locally Led Development, London: Overseas Development Institute.