A top Toolkit on Adaptive Management. But is that a good idea?

April 20, 2021

     By Duncan Green     

In recent years, I’ve been one of a crowd of people thinking and pontificating about ‘adaptive management’. The debate has been rather dominated by academics and thinktankers, fond of hand-waving generalizations and rather better at taking down the bad stuff that suggesting what might replace it.

In those conversations, Graham Teskey has played the role of the stroppy practitioner, demanding more practical guides to how to ‘do development differently’, things that might help a harassed donor or programme manager (or him) get it right. He’s obviously got tired of waiting because he (with Lavinia Tyrrel) has written his own. And I think it is going to be really significant.

The paper, Implementing adaptive management: A front-line effort. Is there an emerging practice?, spends just 16 pages summarizing the main issues – what’s wrong with standard, aid practices such as the logframe; the principles behind Adaptive Management (or the overlapping approaches to Doing Development Differently, or Thinking and Working Politically or – for the really nerdy – Problem-Driven Iterative Adaptation – lots of common features, but see Heather Marquette’s great summary of the differences). They come up with a catchy new acronym – PILLAR: Politically Informed, Locally-Led and Adaptive Responses (wonder how many beers that took) – to summarize their recipe.

But then they get onto the important stuff – 15 ‘how to’ notes for development practitioners on everything from how to ‘screen’ potential recipients of donor grants to setting out practical guidelines on the 3 elements I developed with Angela Christie a few years ago (adaptive governance, programming and delivery, above) and a final note on how to measure all this. I’ll cover those tomorrow.

Some highlights from the initial paper (italics are mine):

‘The current state of play: Among the many principles that currently inform donor-funded development initiatives, three appear to stand out: first, that they should be politically informed; second that they should be locally led; and third, that they should be adaptive. The purpose of this paper is not to interrogate these principles or summarise different views expressed in the literature. Rather, it is to grapple with the question of how to operationalise them. It is written from the point of view of two practitioners who are operating at the ‘front-line’ and who are struck by the predominance of papers that explain the ‘what’ and the why’, but rarely the ‘how’.

What was wrong with what went before: These three principles arise not only from a recognition that development initiatives often do not go according to plan, but also from concerns regarding the nature and functioning of the logframe, and results frameworks more generally. As far as the logframe is concerned, Chris Mellor, then in the Evaluation Department of what was the Department of International Development, identified three intrinsic problems: (i) they assume that causal pathways can be known in advance; (ii) they condense and simplify ‘messy realities’ and assume they can be treated in reductionist ways; and (iii) they establish incentives to deliver, rather than to adapt. While the logframe undoubtedly requires donors to think more rigorously about their investments, it can sometimes act as something of a ‘straightjacket’, preventing the emergence of more flexible and appropriate ways to design and deliver investments. Adaptive management is part of the response.

Their theory of change: What will it take to shift practice away from linear and planned approaches (top left quadrant of figure), towards models which foster local leadership and -suggests that the answer is not to throw out the discipline of the logical framework, results frameworks, or theories of change. Rather they need to be handled rather more reflectively and ‘elastically’. We need tools for planning, monitoring and reporting which can be actioned within the dominant aid paradigm.

Conclusion: Operationalising adaptive management – what we have called PILLAR – is conceptually difficult, politically risky and organisationally challenging. However, given what we have learned about complexity, systems theory, competing interests and incentives, changing policy contexts, and the sheer unpredictably of individual and collective human behaviour, it seems – to these authors at least – that there are two options. Either we accept the challenges and embrace PILLAR, or we resign ourselves to the likelihood of continued program underperformance or failure.

It is, by some distance, the most practical guide so far to this cluster of approaches, but I have to admit to some misgivings, which I’ve discussed before on this blog. The aid sector, along with most public institutions, seems only able to absorb new ideas when they are codified and turned into toolkits. Then people can be trained, measured and evaluated against an agreed set of criteria. Graham and Lavinia have taken on that challenge as the only way to drive the changes they see as urgently needed.

The trouble is that the essence of adaptive management is to some extent, ‘throw away your toolkits – study the system (even dance with it), adapt, respond, improvise’. And codification can easily take power away from people on the frontline and into the hands of ‘adaptive managers’ versed in the jargon of the new orthodoxy, when the exact opposite is needed. I only hope that when people use this guide (as they undoubtedly will), they are able to follow the spirit, as well as the letter, of the adaptive management message.

More on the actual guidance notes tomorrow.

April 20, 2021
Duncan Green