How Change Happens: Great new case studies + analysis on ‘Politically Smart, Locally Led Development’

September 24, 2014

     By Duncan Green     

The research star of the show at last week’s Thinking and Working Politically event was a great new ODI paper from David Booth and Sue Unsworth.

Credit: Bioversity International/Ronnie Vernooy

Credit: Bioversity International/Ronnie Vernooy

Politically smart, locally led development seeks to identify the secret sauce behind 7 large and successful aid programmes: a rural livelihoods programme in India; land titling and tax reform in the Philippines; disarmament, demobilization and reintegration in the Eastern Congo; the EU’s global plan of action to reduce illegal logging; civil society advocacy on rice, education and HIV in Burma and inclusive governance in Nepal.

The paper identifies a number of common elements:

‘Purposive muddling’: Project teams experimented, hit dead ends, tried something else, meaning that often spending and results built up over time. In the Philippines ‘It took four years to find effective ways forward. [This] involved exploring and abandoning several different avenues, and eventually hitting on a way of sidestepping major opposition by restricting the reform to residential land titling (so avoiding vested interests opposed to changes in agricultural land titling).’ There was a lot of learning from previous failures, which required having experienced staff who knew where the institutional bodies were buried.

Brokering relationships: a hugely time-consuming effort to establish relationships and build trust with partners and institutions. In India, this included hiring recently retired, senior government officials to support design and help build trust and credibility. In Myanmar, ‘the Land Core Group brought together for the first time civil society and government officials in a national dialogue on land tenure and land use rights reform. [Elsewhere] support has facilitated networks involving different ethnic and faith groups to improve HIV prevention and treatment for marginalized groups in remote areas.’

Politically smart: the leaders of the interventions were politically well informed and had the skills to use that knowledge effectively. They acquired their knowledge and skills in a variety of ways (personal experience, commissioned political economy analysis, well-connected intermediaries).

Local leadership: ‘In all cases, the interventions addressed issues with real local salience and the solutions were locally negotiated and delivered because project managers allowed local actors to take the lead. Across the cases there was a common willingness of the ultimate funding body to take a back seat. Donors provided external stimulus and had their own vision of the kind of change they were seeking to support, but they avoided dominating either the agenda (in the sense of specifying what to do) or the process (specifying how to do it). This was critical in freeing the front-line personnel to explore pathways towards changes that were both worthwhile and tractable.’

Gandhi v logframe cartoon‘There are two other factors that were important in all the cases. First, flexible funding arrangements supported iterative approaches to design and implementation and allowed people to respond to opportunities as they arose. None of the programmes was under pressure to meet particular spending targets or timetables. Secondly, in all cases funding agencies were prepared to make long-term commitments, and in many cases there was an unusual degree of staff continuity.’

What is the ‘so what’ for donors seeking to work more along these lines? If success is down to so many unpredictable factors – finding development entrepreneurs, champions in government, the right issue etc etc, trying to pick winners and then simply applying a new ‘politically smart, locally led’ toolkit is unlikely to work. A better option is creating an broader ‘enabling environment’ where those entrepreneurs who happen to emerge at the right time and place can get stuff done.

Unfortunately, the aid business currently has something closer to a disabling environment: high staff turnover, pressure to deliver short term results and spend a lot of money and risk aversion, along with the imbalance of power between donor staff wielding chequebooks, and ‘partners’ seeking cheques. By contrast, ‘Iterative, adaptive problem-solving requires an underlying relationship of trust between the funder and front line operators: the funding agency must show some willingness to let go.’

In some of the 7 cases, donors achieved this by operating at arm’s length, handing over responsibility to an intermediary organization. In others, aid agency staff had the personal and professional status, self confidence and/or political backing to defy the institutional culture.

The authors wonder ‘what kind of political and bureaucratic environment would have killed off our seven programmes or arrested their development early on? Among the obvious, killer conditions would be:

  • a requirement at the outset to write a ‘business case’ setting out options and assessing their relative value for money, based on a theory of change red tapethat fails adequately to capture complexity and unpredictability;
  • conducting a diagnostic analysis to define the conditionalities attached to a large loan;
  • setting spending targets rather than allowing funding requirements to emerge;
  • requiring regular progress reports against predetermined targets;
  • banning funding to politically connected individuals and organisations;
  • placing ceilings on the share of administrative costs in project budgets; and
  • tolerating high staff turnover.’

Which unfortunately looks a lot like the typical aid agency rules and procedures.

What to do? The paper has some helpful tips and tactics, but also discusses broader strategy. The key is for donors to nurture their assets beyond the chequebook, especially political knowledge, connections and skills. Those mean they can put the chequebook aside, and start to build relations built on mutual respect and genuine trust.

‘Donor agencies could do more to make it clear through recruitment and promotion practices that they value in-depth country experience as well as technical knowledge; could keep staff in post longer; encourage them to invest in understanding local context (through formal analysis but also by building broad relationships and networks); give them flexibility to (re)define outputs and outcomes as their knowledge of specific challenges increases; support the development of more systematic learning mechanisms and less prescriptive monitoring practices; and avoid setting unrealistic spending targets or requirements to achieve short-term, pre-defined results.’


‘The good news is that there is nothing inherently new or esoteric about politically smart, locally led approaches that support iterative problem-solving: they have much in common with good policy-making anywhere. Indeed it is a measure of how detached the aid business has become from everyday reality that we should consider any of the seven cases remarkable. They show that donors can facilitate developmental change in very challenging contexts, but only if they are prepared to align their own thinking and practices with the uncomfortable reality that processes of developmental change are complex, unpredictable, mainly endogenous, and pervaded by politics.’

My only criticism? Why do political scientists insist on coming up with such clunky, instantly forgettable slogans? Politically Smart, Locally Led Development; Problem-Driven Iterative Adaptation; Problem-Driven Political Economy Analysis. Not exactly ‘Getting the Prices Right’ are they? As usual, the devil has the all the best tunes. Anyone got a better suggestion?