What is adaptive management?

Dr. Gregory J. Wilson Dr. Gregory Wilson, Governance and Public Sector Management

wide image

Adaptive management is not a new concept, particularly in international development.  As long ago as 1983, Dennis Rondinelli, had alluded to many of the problems with the uncertainty in development work and advocated for approaches that are more iterative and adaptive, that allows space for incremental learning[i]. However, these prescriptions were gradually scuttled by, among other things, the development community’s increasing appetite for performance measurement as a result of the influence of new public management, the increasing need to demonstrate value for money to key constituents back home, their general inflexibility, and for some their narrow political agendas.

Recently the aid community has come to grips with two ideas.  Firstly, there has been growing recognition that most development challenges are in actual fact ‘wicked’ problems[ii] that requires development actors to recognize that the best management approaches will be iterative not definitive, and inclusive not technocratic” [iii].  Secondly, the Taylorist approach to performance management[iv], or what Andrew Natsios, visiting fellow at the Centre for Global Development and a former Administrator of USAID calls “the rise of the counter-bureaucracy and obsessive measurement disorder”, has led to an culture of aid that prefers interventions where results are easy to quantify and measurable after a short period.[v]

A selection of our Adaptive Management case studies

So it is unsurprising that Rondinelli’s ideas are being revisited and experimentation is back on the table. Despite his insights over 30 years ago, it is only recently that the notion of ‘wicked’ problems and the Adaptive Management (AM) agenda have found traction within the aid policy circles[vi].  Both USAID and DFID are now jointly pursuing research into Adaptive Management approaches.  And the proposed Global Learning for Adaptive Management (GLAM) programme contracted through DFID proposes to establish a centre for learning[vii].  Some other notable contributions include the application of AM in the humanitarian field[viii], using AM to give theory of change and learning a much needed makeover[ix], and our own research[x] in Afghanistan that concluded there was no choice but to be adaptive and actively learn on the job when dealing with wicked public sector governance problems.

Framing fragile States and conflict as ‘wicked problems’

If most development challenges are considered ‘wicked’ problems, then countries dealing with fragility or conflict, let us call them Fragile and Conflict Affected Situations (FCAS) for now, most certainly embody the essence of the definition.  In these ‘wicked’ situations, more than anywhere else, delivering tangible and timely results to citizens is critical. However, this is made difficult by an array of constraints imposed by security, national capacity, economic performance, the threat from non-State actors, the impact of geopolitical and regional political dynamics as well as the limitations on field presence of international partners (to name just a few).  Indeed, the only constant in development programming in FCAS is rapid and unpredictable change itself.

Practically speaking, improvements to security and the field presence of international staff are things that the donors and their partners can influence in a number of ways, even if most ways imply additional costs.  More complicated is identifying what to do, in what sequence and most importantly how.  In common with most fragile States, the evidence base for what works is limited. Most conflict countries Antylles has worked in are characterised by poorly organised information and hard to access evidence. Typically, the government’s capacity to undertake its own research, data gathering and analysis is also limited. Major surveys take a long time to prepare and implement in FCAS and usually experience problems with both fieldwork and analysis. Getting the humanitarian actors to work closely with development actors on joint analysis has proved hard to get going.

Wicked problems naturally resist ‘best practice’ solutions

The rationale for Adaptive Management is rooted in the realities faced by trying to do development in complex FCAS settings[xi].  The emerging body of evidence on capacity-development interventions advocates for the adoption of ‘problem driven iterative adaptation’ (PDIA) in programmes[xii]. Recent research shows that the drive by donor governments and IFIs to achieve ‘effective accountable institutions’ and ‘good governance’ in developing countries has often led to the externally-driven imposition of “best practice” solutions. It was found this approach is flawed and leaves States in a perpetual ‘capability trap’[xiii], which are failures caused by “big development (that) encourages progress through importing standard responses to predetermined problems” [xiv]. This in turn incentivizes State organisations – who are desperate for aid funds – to outwardly adopt the forms of a functional State or organisation by papering over a persistent lack of any real function or capacity, sometimes referred to as an ‘institutional façade’.  The ‘trap’ is finally ‘set’ when the international community withdraws or downscales its support under the false assumption the ‘mission has been accomplished’.  This structural myopia creates “premature load bearing” that results in State institutions being overwhelmed when the institution is faced with either a typical workload, or worse, a critical situation such as a humanitarian crisis or the resumption of hostilities.

However, the thinking above only addresses part of the problem. The compounding factor is caused by social complexity. FCAS involve multiple actors and unique implementation challenges. The research shows, that the greater the complexity of relationships, hierarchies and stakeholders, the more the managerial competence and professionalism of public servants in practice will count[xv] . In places like Somalia, Afghanistan, South Sudan the local workforce simply cannot supply the level of competence required to manage this level of relationship complexity.  On the other side of the table, the conceptual and operational structures of donors (the ‘rules of the game’) constrain their own professional staff to patterns of recognition and intervention that are not appropriate for the wicked problems they are being asked to address. Worse still, in conflict situations like Afghanistan and Somalia where stabilisation activities are prevalent, the military and their technical advisers in the face of complexity often simply fall back onto tried and trusted ‘command and control’ management and a bewildering array of ‘metrics’. As if we can somehow project manage and ‘surge’ our way out of complexity!

We need to re-tool

The literature is replete with case studies to support the notion that FCAS and other wicked development problems resist standardised responses.  The growing movement of researchers and practitioners who now argue that development programmes need to be more sensitive to local needs and more adaptive to the hyper-dynamic environment is welcomed.

In order to evolve this argument into concrete action, new adaptive tools and approaches are required. A starting point is that interventions should be designed with these four core principles in mind[xvi].

  1. Local solutions for local problems
  2. Pushing problem driven positive deviance
  3. Try, learn, iterate, adapt
  4. Scale through diffusion

Adaptive management seeks to better achieve desired outcomes and impacts through the systematic, iterative, and planned use of emergent knowledge and learning throughout the programme lifecycle. It involves reacting and responding quickly to changes in the political and socio-economic operating environment. In FCAS, this may be particularly relevant where overall programme goals or aims may be clear, but the pathway to achieving them might be uncertain or unknown. Adaptive programmes therefore adopt deliberate processes of testing, learning and experimentation, drawing on monitoring, evaluation and other data and evidence strategies[xvii]. In military parlance…pause, think, adapt.

However, the clear implication is that becoming more adaptive will require shifts in aid culture and increased capacity to design and manage systems that respond to this way of working.  The intention is that both the government and development partners can reduce the impact of uncertainty and improve the alignment with outcomes through deliberate processes of testing, experimenting, evidence gathering and learning and allocating the resources required to generate the learning. But critically, this is also contingent upon organisational and behavioral changes in the development community.  For example, centralized command and control systems, ‘risk averse’ incentives for promotion linked to the culture of ‘fear of making mistakes’, delegations of authority, consultant acquisition processes, grant and contract management approaches, and of course the “bench strength” of senior and mid-level leadership, are just some of the many organizational “levers” that need to be adjusted to give donors and their implementing partners the capacity for AM and ultimately make them accountable for change and getting maximum value out of every dollar spent.

Practical ways forward

Returning to ‘wicked’ problems, the weaknesses of the rational comprehensive approach are well known. But in practice, the complications associated with multiple stakeholders on complex international missions (such as Afghanistan, Iraq, Somalia etc.), the geopolitical and regional dimensions, and the dominance of the military presence, have inadvertently caused the International Community – through simple institutional inertia – to fall back on the rational planning model as if it offers succour and a safe place.

Improving the chances of learning by increasing interaction between consultants, researchers, diplomats, government stakeholders, and the military undoubtedly could improve the responsiveness of the IC to some of these complex problems, as would increasing dialogue around what analytical work and other research is required. Frankly and openly recognising the complexity of the issues might make policymakers more humble and open to alternative approaches. At the project level, this means encouraging donors to be less prescriptive, to be more open to change and experimentation, more prepared to learn and have the bureaucratic flexibility to adapt projects quickly through speedy feedback into the project cycle (and the log frame if used) rather than wait for the inevitable pre-planned evaluation cycle that involves lengthy procurement cycles to bring in external agents. In such experimental situations, the value of carefully planned ‘in-built’ Monitoring, Evaluation and Learning should be heightened. This will help donors solve their natural inclination to ‘hold back’ until they are more certain of what they can achieve.

Recommendations – How could Adaptive Management be encouraged and addressed by Donors and Host Governments?

  • Encourage implementing partners to embrace AM and the ways and means to tackle complexity.
  • Strongly support the host Government’s drive for more information and data on their current situation and development needs, not just at the project level but improvements to national statistics, and poverty data
  • Share and encourage open access policies; free and open access to publicly-funded research offers significant social and economic benefits. Development bodies must not be allowed to hoard data
  • Encourage adaptive management with innovative funding mechanisms, encouragement for innovative research, and development of new technology solutions to managing aid and improving VFM. Support local universities to develop their capacity, encourage partnerships, north-south, south-south
  • Ensure all programme designs and implementation plans draw on the perspectives of local partners and those with whom they work. Gateways in the programme and project approval process will need to demand that provision is made for systematic MEL.
  • In situations where Development Trust Funds operate ensure that all proposed projects incorporate a clear statement on MEL and how the programme/project will provide a steady stream of information that is used to understand the context and programme performance, and how the information will be stored and made accessible. In short, a closer focus on accountability for learning
  • Establish more joint programmes for both DPs and other partners to learn about Adaptive Management and PDIA type approaches; adopt a joint commitment to collaborative learning. In complex emergencies where humanitarian and development action co-exists, demand joint approaches to understanding and responding.

Later in this series of blogs on AM we will look at some case studies of AM in practice.


[i] Rondinell, D. A. (1993) Development Projects as Policy Experiments: an adaptive approach to development administration, 2nd ed, Routledge, London and New York.
[ii] Webber, M. and Rittel, H. (1973), “Dilemmas in a General Theory of Planning”, Policy Sciences, vol. 4, no. 2, pp. 155-169
[iii] Head, B. and Alford, J. (2008) “Wicked Problems: The Implications for Public Management”, 12th Annual Conference International Research Society for Public Management, Vol. Panel on Public Management in Practice, 26-28 March, 2008, Brisbane.
[iv] Taylor, F. (1911). The Principles of Scientific Management. New York and London, Harper & brothers
[v] http://www.cgdev.org/publication/clash-counter-bureaucracy-and-development.
[vi] See for example Ramalingam, B., Laric, M. and Primrose, J. (2014) ‘From Best Practice to Best Fit: Understanding and Navigating Wicked Problems in International Development’. Working Paper. London: ODI
[vii] The main programme activities are likely to be: Supporting the design and use of monitoring, evaluation and learning (MEL) systems for adaptive management; increasing donor and partner (both international and local) capacities to use MEL systems and implement adaptive management practices; building consensus, encouraging coordination and establishing standards for MEL practices in adaptive programmes; Evaluating and learning about adaptive management
[viii] Mercy Corps. (2016) ‘Managing complexity: Adaptive management at Mercy Corps’. Portland, OR: Mercy Corps.
[ix] Valters, C. (2015) Theories of change: Time for a radical approach to learning in development. London: Overseas Development Institute.
[x] Wilson. G (2013) Limits to Civil Service and Administrative Reform in a Fragile and Conflict Affected State – a Case Study of Afghanistan 2002–2012. PhD Thesis, Cranfield University
[xi] The Overseas Development Institute (ODI) has also joined in the debate and written a research paper in this area. ODI (2016) Putting Learning at the Centre.
[xii] https://usaidlearninglab.org/lab-notes/how-do-you-do-adaptive-programming-two-examples-practical-experience-help-some-answers
[xiii] Pritchett, L. and de Meijer. F (2011) Fragile States: Stuck in a Capability Trap, WDR 2011 Working Paper, World Bank, Washington.
[xiv] Pritchett, L., Woolcock, M. and Andrews, M. (2010) Capability Traps? Mechanism of Persistent Implementation Failure, 234, Center for Global Development, Washington.
[xv] Hupe, P. (2011) “The Thesis of Incongruent Implementation: Revisiting Pressman and Wildavsky”, Public Policy and Administration, vol. 26, no. 1, pp. 63-80.
[xvi] See also Andrews, M., Pritchett, L. and Woolcock, M. (2015) Doing problem driven work. Working Paper 30. Cambridge, MA: Centre for International Development, Harvard University; See also Booth, D. and Unsworth, S. (2014) Politically smart, locally led development. ODI discussion paper London: Overseas Development Institute; Fritz, V., Levy, B. and Ort, R. (2014) Problem-driven political economy analysis: The World Bank’s experience. Washington DC: World Bank.; Pritchett, L., Samji, S. and Hammer, J. S. (2013) It’s all about MeE: Using structured experiential learning (’e’) to crawl the design space. Working Paper 322. Washington, DC: Center for Global Development.
[xvii] ODI Has been at the forefront of recent discussions on Adaptive Programming Research: Joint work between DFID and ODI. Includes a definition paper, case studies of best programming examples and drawdown support to country offices. Laying out “top tips” to procuring and managing adaptive programmes. USAID’s Global Development Lab/Digital Development is developing a toolkit intended to promote adaptive management based on the use of real-time data.

For more information on this article please contact our expert: Dr. Gregory Wilson.
Photo: A chameleon painted by the 17th century indian painter, Ustad Mansur.
Photo credit: Art Gallery ErgsArt