Take-up and Doubt: where have we got to on Thinking and Working Politically?

June 21, 2019

     By Duncan Green     

TWP Community of practice

Spent a day this week at a Washington workshop on ‘From Thinking Politically to Working Politically’, organized by Abt Associates, whose Graham Teskey is one of the TWP gurus. What struck me most was the combination of the spreading acceptance of TWP approaches within the aid sector, and serious questions being asked about important aspects of the whole enterprise by some of its leading thinkers. I banged on about my adaptive management research, but you’ve already heard loads about that, so for once, I’ll talk about what everyone else said.

In case you’re new to all this, TWP is one of a family of approaches (Doing Development Differently; Adaptive Management, PDIA) described by Teskey in a great paper as a ‘Second Orthodoxy’. Common features of the second orthodoxy are:

  • Context is everything: Political Economy Analysis (PEA) is central.
  • Best fit, not best practice: aid programmes need to ‘work with the grain’ of local institutions.
  • From blueprints to flexible, responsive, adaptive programming.
  • Programmes learn and adapt as they go, rather than simply evaluate at the end.
  • Long-term commitment: Most success stories take a decade or more to show significant impact at scale.

Brian Levy, author of the influential ‘Working With the Grain’ kicked off the second thoughts. He now thinks WWG is well suited to periods of fast inclusive growth, but sees this as just the first stage in a longer cycle. In that first stage, aid needs to ‘stay on the tightrope of forward movement through an incremental WWG approach. But over time power asymmetries lock in, hope turns to anger, and the broader political economy curdles’.

Meles Zenawi and Gordon Brown
Britain’s Prime Minister Gordon meets Meles Zenawi, Prime Minister of Ethiopia, at Number 10 Downing Street at the start of the London Summit on 01/04/2009. Crown Copyright/NTI

Then the response needs to be more about ‘inclusive renewal, rekindling hope by addressing the imbalances directly’. That means being more prepared to work against the grain of power and politics, if the entrenched elites and interests are part of the curdling. As an example he said working with the grain in Ethiopia under Meles made sense, despite the doubts over human rights, but now aid sectors should focus on ‘countries going through democratic recession, not developmental authoritarians like Meles’.

Clare Lockhart of the Institute for State Effectiveness and author of a great book, ‘Fixing Failed States’ asked a striking question: ‘who’s doing the ‘doing’’ in Doing Development Differently? Is it ‘us’ (aid donors, northern organizations, contractors) or ‘them’ (local decision makers, elites, civil society, political activists)? Is it top down (Ann Hudock of Counterpart International : ‘how to shove your pre-agreed agenda into a given context’) or bottom up? Even if it’s ‘them’, is it progressive or entrenching inequality? When does it become a tool of foreign policy for example in Australia using aid to make sure it is the ‘partner of choice’ in the Pacific (and not China)?

Everyone talked a lot about Political Economy Analysis (PEA) but Ann Hudock contrasted ‘the political economy of manipulation’ with ‘stakeholder driven PEA’. How PEAs are created shapes what they see and ignore, and how the impact they will have.

As always, back to Robert Chambers: ‘Whose Reality Counts?’ These questions need to be asked at several points: who sets the agenda, defines ‘the problem’ in PDIA; who implements; who judges what is success or failure? I think we’ve been fudging this issue for far too long. If we don’t ask these questions, the risk is that TWP becomes just another (not particularly progressive) toolkit, and loses its soul.

Other impressions from a day of animated conversations:

Some aid agencies are doing TWP, but not calling it that. For example, in chaotic conflict situations like Syria or

The Searchframe

Libya, USAID reportedly accepts that not everything can be planned in advance. Big aid grants are run with no pre-set indicators, on the basis of 6 monthly discussions between contractor and donor to assess progress and agree the direction of travel for the next 6 months –Matt Andrews’ ‘searchframe’ (right) alternative to the logframe is already in operation.

An Advocacy Strategy for TWP?

If we can resolve the dilemmas above, it feels like a TWP advocacy strategy within the aid sector would be relatively easy to design, and could get some real results. Elements would include:

  1. Get the stories right: we need more case studies that are not post-dated fairy tales, but record in real time the messy reality of TWP and some of its undoubtedly impressive results. Example: John Sidel’s case studies of the ‘Coalitions for Change’ programme in the Philippines (example here). Others on Cambodia on solid waste and Bangladesh leather sector.
  2. Get the analogies right: to convince doubters, a good analogy works wonders. Here are two for starters: for private sector types, TWP is like venture capitalism – a way of operating in situations of uncertainty and ambiguity, funding a dozen start-ups on the basis of judgement and partial knowledge, knowing that many of them will fail, but others will deliver. Or for peace and conflict people, TWP is like operating in wartime (like the Syria/Lebanon examples) – everyone knows it’s disastrous in warfare to blindly pursue the plan without feedback loops and adaptation (Charge of the Light Brigade, anyone?)
  3. Examples of good donorship. Scour the aid sector for examples of TWP, including (especially?) where they are not recognized as such. Analyse the factors that allowed them to be agreed, and what they achieved. Sing their praises over and over again until the message gets through.
  4. Develop a package of monitoring, evaluation and learning (MEL) specifically designed for TWP and ‘counting what counts’. Abt’s Lavinia Tyrrel suggests ‘Something that brings together the qualitative methods already out there designed specifically for TWPers (i.e. and detail the complex grubby reality – e.g. strategy testing) and the simpler metrics that suit the ‘bean counters’ (e.g. Jaime Faustino’s ‘measures that matter’). The point is we need to tell both stories (the simpler and the complex). As we have multiple masters/ naysayers.’
  5. Where donors have agreed elements of TWP, monitor whether they are being implemented on the ground, and if they aren’t (eg junior officials are blocking changes to plans) kick up a huge fuss until they get on message.

A good day – thanks to all the speakers and Abt.