Links I liked

September 23, 2014

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

September 23, 2014

Thinking and Working Politically update: where have aid agencies, consultants etc got to?

September 23, 2014
empty image
empty image

Spent an engrossing couple of days last week at a ‘Thinking and Working Politically’ (TWP) seminar, organized by a group of donors, thinktanks and consultants (sorry, Chatham House Rules, so that’s as much as I can say about them). Their common ground is that aid needs to get beyond its technocratic comfort zone, and take politics and power more seriously. It’s a new initiative, and as with all such efforts, is pretty messy and confusing at first, as people try to agree on problems, definitions, language etc before deciding what to do. But this was the third such meeting, and I think we’re getting somewhere.

TWP uptake spectrum

First a bit more clarity on ‘the spectrum’ of what constitutes TWP (see graphic). At one end is what was termed an ‘evolutionary’ approach – getting more politically savvy in the way donors do their normal aid activities (building stuff, offering technical support, financing public services). It’s a random number, but people typified this as ‘adding 15% to the impact of aid programmes’ by designing them with a fuller understanding of institutions, incentives and interests (both material and political).

At the other end is a more transformative ‘revolutionary’ approach, for example where donors do not claim to know the answer, and either respond flexibly to events and political opportunities, or concentrate on bringing together local players to solve a problem (what Matt Andrews calls Problem-Driven Iterative Adaptation, or I call ‘convening and brokering’).

A shout out to readers of this blog whose ‘where the fxxx is gender’ comment stream after the first TWP meeting was credited by various speakers who raised gender and the wider question of ‘are we talking about power, or just formal politics?’ this time around. See? Commenting on blogs can make a difference. Sort of.

We also had a good discussion on what TWP is not. This is partly a response to grizzled old aid types who sniff ‘huh, that’s not new – we’ve been TWP for decades.’ One speaker proposed the following list:

  • Formal Political Economy Analysis (PEA) without changing approach
  • Standard donor policy dialogue with government
  • Country ownership/partnership narrowly defined as government-led
  • Demand side of governance that solely focuses on “more and louder NGOs”
  • Working through the same old local partners that speak the donor language
  • Supporting “reform champions” without deeper understanding of political dynamics
  • Assumption that no one will lose from reforms
  • Conditionalities attached to a loan

We got a bit clearer on ‘why change doesn’t happen’ – the sources of inertia within the aid system that mean that even when the country director says

Not a great theory of change

One alternative to TWP

‘yes, this sounds like the world I know’, TWP has minimal impact on the country programme. There is a dispiritingly long list of blockers, including the pressure to spend aid budgets: ‘In Afghanistan, people talk constantly about maintaining the burn rate [spending the budget within the financial year]. We can’t do anything that slows the burn rate – we have to address anything going wrong, while continuing to spend.’

More broadly, the technocratic approach of logframes and roll-outs has created a system of staff, contractors, partners and evaluators, who even though they recognize that the system is often based on false assumptions, are either unable or unwilling to do everything differently. It’s an interesting question whether the best tactic is to try and get them to unlearn decades of the old ways, or merely help those who recognize its flaws subvert the current system in a more politically informed way (which Ros Eyben has documented beautifully) – ‘how to bypass a logframe’ guidance notes?

So if promulgating TWP approaches is uphill work, where are the best prospects? Some smart advice here:

–          Areas of the aid business where failure is rife and/or risks are high, so people are willing to try new things (fragile and conflict states, oil revenue management, anti-corruption)

–          TWP needs to get out of the ‘governance silo’, and show how the approach is relevant to bigger spending areas of the aid business – infrastructure, natural resource management, service delivery etc. Some high profile non-governance champions would help.

–          New or rapidly changing contexts, which have not yet had time to entrench standard approaches and so people are more ready to experiment (eg Myanmar).

–          Countries with senior champions within aid agency offices – eg ‘Heads of Mission’, or their equivalent

And a few concerns:

Are we designing TWP for gurus or newbies? Rejecting rigid guidelines in favour of ‘every context is different, just cross the river by feeling the stones’ is fine if you’re a veteran of dozens of previous river crossings. It’s not so encouraging if you’re in your late 20s and you’re panicking in your first aid job. My own attempt to square this circle is that a combination of case studies, sample questions to ask, and mentoring via HQ or peer networks, can provide the support people need, without destroying the ability to be flexible and adapt to context. But you probably need some element of written guidance too.

It’s also still very top down. As Craig Valters recently noted, TWP approaches and their accompanying theories of change tend to be dreamt up by the donors and their consultants, not arrived at through anything like a participatory process involving the actual people concerned. That’s pretty worrying.

There is still a tendency to default to ‘if in doubt, commission more research’, even though people accept that the political economy of aid is probably a much greater barrier to TWP than a lack of research. Probably worth getting some professional lobbyists in to design a TWP advocacy programme that covers windows of opportunity, champions, reform coalitions, tactics etc as a counterweight to all those academics seeking yet more research contracts.

It’s still a bit linear: the donor people there are practical types, with very little time for all that stuff about complexity and systems thinking, even though they are highly relevant to making sense of TWP. That pushes us towards the ‘evolutionary’ end of the spectrum – add a dollop of TWP secret sauce to your standard linear programme and voila!

I need a toolkit for the toolkit

I need a toolkit for the toolkit

Finally, how to communicate all this? For a start, change the name – TWP is unnecessarily and off-puttingly shrill in my view. Call it ‘what works’ or even (not my favourite), ‘politically smart, locally led development’ – the title of a new ODI paper which I will review tomorrow.

But also, there’s tension on how to frame it – consultants and evangelists want a shiny new product, to which aid staff already overwhelmed by endless restructurings and management processes wail ‘OMG, not another toolkit’! We need a sharper narrative and a lot more case studies on what is actually distinct about a TWP approach if we are going to convince the sceptics. Some horror stories and ridicule of bad aid projects that fail by ignoring power and politics would also help open minds. That sounds fun.


  1. Apologies in advance if this will come off as unnecessarily grumpy but I believe the spectrum is misleading and will bury some important aspects of “understanding of institutions, incentives and interests.”
    The ‘evolutionary’ end seems to be -a somewhat smarter but still- yet another instrumental go at forcing some premium but essentially ill-fitted spare parts in place, whereas the ‘revolutionary’ end is brought about in acknowledgement of development being political in itself. There’s a fundamental difference in what is means and ends and how they’re defined.

  2. I appreciate your sharing the main points of the TWP discussion. Perhaps I am being overly optimistic about the future of politically smart development programming, but I do see more donors including requests for PEAs in their solicitations. I have been focusing on the use of TWP approaches at the problem or project level. However, I also have limited expectations that the tyranny of the technical approach will be downplayed anytime soon, if ever. Procurement regulations aside, this is because the political institutions which authorize funding for overseas development are of many minds about the need and purpose of development aid. Bureaucracies which manage aid funds are often loathe to be seen as meddling in local politics or stepping on the pol/econ turf of embassy diplomats. Some development practitioners also feel uncomfortable paying too close attention to local politics because they often don’t see the direct relevance to their projects, consider their own context analysis as sufficient, and/or fear being viewed as foreign spies. That said, I have seen an upsurge of interest from colleagues in health, livelihoods and NRM in what TWP approaches can offer to improve project outcomes.

    About what to name this tool that will motivate more development practitioners to think and work in a politically savvy way, “what works” works, but perhaps we should be more honest and call it “Intelligence for Development”; and, because we are development practitioners, immediately give it an acronym: I4D or “Development Intelligence” DI, well, that may be rather ominous. Actually, adopting the private sector use of “political risk analysis.” At Pact, we call it Applied Political Economy Analysis (APEA); we realize that it that may not mean much to many of our colleagues at first glance. Our hope is that while APEA may seem murky now, its utility for the design and implementation of development projects will make it very obvious to most in the near future.

  3. Interesting, thanks for sharing. In my view, one’s position in the “TWP” spectrum depends very much on how accustomed one is to the various existing tools to ‘think politically’. I think few development practitioners would disagree with the importance of taking politics into account or even using it to ‘their’ advantage in programming but are put off by the jargon and acronyms that go with it. I am personally not a fan of the term “TWP” even if I advocated for it (using different expressions) for a long time at my own organization, and I can see why it does not resonate well with colleagues from other areas. It is a bit like telling a governance person to ‘think and work healthily or ‘think and work environmentally’. Interestingly, after we launched the Institutional and Context Analysis approach to programming at UNDP, it was more widely used by colleagues working on energy and environment (who recognized that they were missing something to get their projects to work better) than the governance people. Perhaps because the latter thought they were already doing it, even if many project results showed otherwise.

  4. Maybe I don’t get what you’re talking about, but it seems to me that until the people for whom aid is intended actually control the resources, then they can’t lead their own development, so what happens, while it may be “good” isn’t actually development at all. THAT is the politics that we need to be discussing, no?

  5. Claudia, you make a valid point about how the imposition of “political thinking” on non governance colleagues can have undesired consequences, while use of more neutral nomenclature achieves the goal.
    Nora, If I understood your point correctly, I’d say political economy analysis or institutional context analysis can help practitioners better understand how best to ensure that intended beneficiaries/participants have a say, if not complete control over the resources intended for them.

  6. I want to call attention to what Jonathan Fisher and Heather Marquette refer to as the “third generation” of PEA thinking. In their DLP paper “Donors Doing Political Economy Analysis™: From Process to Product (and Back Again?,” the authors state that donors need to discard the conventional use of PEA because it fosters resentment in the host government and runs against the Paris aid effectiveness principals of country ownership. The authors argue, inter alia, that what is needed is more “joined-up” analysis with government and other local stakeholders that is shared and debated to solve development challenges. The World Bank is using this approach and it seems to be yielding positive results.

    At the implementer level, our [Pact] contention is that PEA process can be very helpful in ensuring the projects and programs use a more systematic approach to understanding the political context in which the project is operating. Many project managers already do this in one form or another, however, some do not. The purpose of adapting PEA to the project level is to ensure the assumptions inherent in the project design (which was designed by donors) are still valid; and, if not, use PEA to justify changes in work plan activities and expected outputs and outcomes. While project level PEA need not be a secretive process, some of the findings emanating from the analysis at times are better left out of report and discussed first with key stakeholders. Moreover, the need for PEA reports at the project level is debatable. Given the bureaucratic nature of the donor-implementer relationship, the need to show evidence of how public monies are spent is often depicted in the form of a report. I am not stating that PEA reports should be avoided, they are useful, but my point is that the report is just one product of the PEA process that helps field staff, local partners and donor staff who may not pay adequate attention to political factors (whether they emanate from foreign donors or local actors) that can affect the project’s likelihood to achieve its intended outcomes. This takes us back to the way to create incentives for implementers to “think and act politically” or as Claudia hints above to use a more robust form of “contextual analysis” throughout the project cycle.

Leave a comment

Translate »