(2020). What Is Adaptive Management and How Does It Work? Monitoring and Evaluation for Adaptive Management Working Paper Series, Number 2, September

Monitoring and Evaluation for Adaptive Management Working Paper Series Number 2

What is adaptive management and how does it work?

September 2020

Patricia Rogers and Alice Macfarlan BetterEvaluation

About This paper is part of the BetterEvaluation Monitoring and Evaluation for Adaptive Management working paper series, which focuses on using monitoring and evaluation to support adaptive management.

The series can be accessed at www.betterevaluation.org/evaluation_for_ada ptive_management_ series

While focused especially on international development, this series is relevant to wider areas of public good activity, especially in a time of global pandemic, uncertainty and an increasing need for adaptive management.

This publication has been funded by the Australian Government through the Department of Foreign Affairs and Trade. The views expressed in this publication are the author's alone and are not necessarily the views of the Australian Government.

This paper benefited from comments from reviewers Kirsten Hawke and Ally Bridges (DFAT).

Recommended citation: Rogers, P. and Macfarlan, A. (2020). What is adaptive management and how does it work? Monitoring and Evaluation for Adaptive Management Working Paper Series, Number 2, September. Retrieved from: www.betterevaluation.org/monitoring_and_evaluat ion_for_adaptive_management_ series

This way of thinking about ‘adaptive management’ Defining adaptive focuses on what the GLAM initiative (Global Learning on Adaptive Management) has referred management to as “complex problems that will always demand contextual learning, and … problems where the Implications of different challenges faced and/or the interventions are novel and untested, and where there is little definitions evidence for what will work in a particular context” The different ways that the term ‘adaptive (Ramalingam, Wild & Buffardi, 2019). management’ is understood and used reflect both While effective ongoing collaboration and learning the different sectors and contexts in which it has are essential features of successful adaptive been used and the widespread lack of management (as we explore later in this paper), understanding that there is a history of using it what distinguishes adaptive management from that refers to a specific type of practice. usual adaptions are the level of adaptation and the For example, USAID has used a very broad use of model-informed adaptation. Adaptive definition that focuses on the use of evidence to management can go beyond changing the details inform decisions: of how activities are implemented, to include changes to the types of activities, the strategies “an intentional approach to making decisions and even the intended outcomes and how the and adjustments in response to new information theory of change is understood. It uses an iterative and changes in context”. USAID, 2020. process of adaptation that is informed by indicative theories of change and contributes to By comparison the roots of adaptive management revising them. in natural resource management focused specifically on using evidence to inform action in This means that adaptive management differs conditions of ongoing uncertainty: from:

“a structured process that allows for taking action • Traditional single-loop learning use of data - under uncertain conditions based on the best Adaptive management goes beyond using available science, closely monitoring and data to detect and correct errors and gaps in evaluating outcomes, and re-evaluating and implementation. adjusting decisions as more information is learned.” California Department of Fish and • Learning from atheoretical trial and error – Wildlife, 2009. Adaptive management uses conceptual models to inform trials and revises models The broader framing of adaptive management is based on the results of these trials. useful in terms of focusing on actually using evidence from monitoring, evaluation and • Ad hoc decisions - The label should not be research to inform decisions and action, and in used as a way of excusing inadequate planning reminding us that, to some extent, all that has not made use of available evidence. management needs to be adaptive – not simply implementing plans but modifying them in Adaptive management will be appropriate in response to changes in circumstances or circumstances of uncertainty and ongoing understanding. However, using the term loosely to unpredictable change. It is therefore relevant for apply to all levels of adaptation obscures the very certain types of interventions and certain types of real changes needed in planning, management, implementation environments, including the monitoring and evaluation for high-level adaptive current global pandemic. It might be that, after a management. period of adaptive management and increasing knowledge and learning, there is less uncertainty Therefore this paper focuses on adaptive and more traditional approaches to planning, management as a systematic response to ongoing management, monitoring and evaluation can be uncertainty which avoids the two extreme options used. Or it might be that ongoing changes in of either giving up trying to plan in conditions of circumstances mean that it is never possible to be uncertainty or rapid change, or trying to use tools sure ‘what works’ in terms of certain types of and processes that are designed for more well interventions, and ongoing adaptive management understood, stable and predictable contexts. will be needed, at least for some elements of the intervention.

What can be learned parallel learning from simultaneous experiments, and processes for reflection and ‘double-loop’ from the origins of learning – not simply adding facts into identified gaps but rethinking models and assumptions. A adaptive management? key component of this was effective processes for The concept of adaptive management is not a new connecting the learning of scientists, managers one, nor did it begin in international development. and the public. Its origins can be traced back to adaptive environmental assessment and management and Thirdly, the process of adaptive management adaptive management of renewable resources in includes explicitly building, using and testing the late 1970s and early 1980s (Stankey et al, dynamic models of likely impact of alternative 2005) and iterative and incremental development policies: (IID) and agile software development from the … this step performs three key functions: 1950s and increasingly from the 1970s (Larman and Basili, 2003). • Problem clarification and enhanced communication among scientists, The key elements of adaptive management in managers, and other stakeholders. natural resource management include: • Policy screening to eliminate options • the importance of design and experimentation, unlikely of doing much good because of inadequate scale or type of impacts. • the crucial role of learning from policy experiments, • Identification of key knowledge gaps that make predictions suspect. (Stankey et al, • the iterative link between knowledge and 2005) action, The implications of this for adaptive management • the integration and legitimacy of knowledge in development are significant. It suggests a need from various sources, and for clear theories and models about how a • the need for responsive institutions. situation is understood to work – which would include a problem theory and a program theory of There are three important lessons from adaptive how one or more interventions might address the management in natural resource management. problem and/or capitalise on opportunities. It also requires structured engagement with different Firstly, while the term has been used in varied groups, and active reflection processes. ways, many of them refer explicitly to managing under conditions of ongoing uncertainty. The 2009 US Department of the Interior guide to Evidence is not just gathered and used in an ad hoc adaptive management, still widely used and cited, way but in a way to maximise its utility, while makes it clear that adaptive management is recognising its limitations. intended to both draw from and contribute to wider knowledge as well as make tangible, local Secondly, adaptive management involves learning improvements: from systematic tests, not from ad hoc trial and error. Adaptive management is different to “Adaptive management promotes flexible “muddling through”: decision making that can be adjusted in the face of uncertainties as outcomes from management “.. in which some learning inevitably results from actions and other events become better whatever management experience is undertaken. understood. Careful monitoring of these There is no purposeful direction to it and one outcomes both advances scientific understanding simply reaps whatever benefits derive from and helps adjust policies or operations as part of earlier experiences,” (Stankey et al., 2005). an iterative learning process. … This contrasts with some of the versions of It is not a ‘trial and error’ process, but rather adaptive management discussed in international emphasizes learning while doing. … Its true development – for example PDIA refers explicitly measure is in how well it helps meet to ‘muddling through’ (Andrews, 2013). environmental, social, and economic goals, increases scientific knowledge, and reduces Systematic adaptive management involves tensions among stakeholders,” (National sequential learning from experiments over time, Research Council, in Williams et al, 2009).

The other source of ideas about adaptive One version of IIR called ‘evolutionary project management is agile software development and management’ provided this rationale for the the associated iterative and incremental approach: development (IID) approach. The parallels between project planning and implementation for “’Evolution’ is a technique for producing the large technology projects and large development appearance of stability. A complex system will be projects are striking. most successful if it is implemented in small steps and if each step has a clear measure of successful The roots of this approach can be traced back to achievement as well as a ‘retreat’ possibility to a the use of a series of “plan-do-study-act’ (PDSA) previous successful step upon failure. You have cycles for quality improvement in Bell Labs in the the opportunity of receiving some feedback from 1930s, Deming’s promotion of this in the 1940s, the real world before throwing in all resources and the use of IID approaches in the development intended for a system, and you can correct of the X-15 hypersonic jet, and Project Mercury, possible design errors…” Gilb (1976) Software the NASA program that put the first American Metrics, cited in Larman & Basili, 2003. astronauts into space, and the command and control system for the first Trident submarine. The following barriers identified in successfully implementing the rational linear approach to By comparison the ‘waterfall’ approach, developed software development have broader relevance as in the 1970s, involved a linear and top-down well: cascade from requirements analysis, design and development phases. While the originator 1) In most cases the people who commission the expected there would be at least two cycles, building of a software system do not know involving an initial pilot phase, the waterfall exactly what they want and are unable to tell us approach was implemented widely on a ‘single all that they know. pass’ approach requiring completion of each stage before moving on. The waterfall approach was 2) Even if we knew the requirements, there are widely taken up in software development and many other facts that we need to know to design more broadly until its limitations became evident the software. Many of the details only become in analyses of failures in software development known to us as we progress in the (Larman & Basili, 2003). implementation. …

During the 1970s, 1980s, and 1990s development 3) Even if we knew all of the relevant facts before of various forms of IID continued, including we started, experience shows that human beings adaptive software development and scrum. In are unable to comprehend fully the plethora of 2001 a number of these practitioners developed details that must be taken into account in order the “Manifesto for Agile Software Development” to design and build a correct system. … based around 12 principles and four comparative values: 4) Even if we could master all of the detail needed, all but the most trivial projects are • Individuals and interactions over subject to change for external reasons ... processes and tools, • Working software over comprehensive 5) Human errors can only be avoided if one can documentation, avoid the use of humans… • Customer collaboration over contract (Parnas and Clemens, 1996) negotiation, and, • Responding to change over following a There are considerable lessons to be learned plan. from the long history of using adaptive management ideas and practices – in particular Agile software development, and its antecedents, being clear about the contexts in which is focused on developing early working versions and particularly appropriate and when a hybrid iterating them to improve them, with considerable approach might be used, the importance of end-user engagement throughout. iterations between planning, action and review, the use of conceptual models and the importance of wide stakeholder engagement throughout.

What is adapted and Adaptation in terms of changing intended actions (causal pathways) by whom? but not goals The focus of adaptive management can be an The USAID Discussion Note on Adaptive element of a project, an entire project, or a Management (2018b) was clear that adaptive program made up of a portfolio of projects (such management does not involve changing the goals: as a facility). Adaptation can refer to improving the quality of implementation of the same activities, “Adaptive management is not about changing or implementing different activities while keeping goals during implementation, it is about the same overall strategies, or changing strategies changing the path being used to achieve the or objectives. goals in response to changes.”

Adaptation at the level of changing objectives, or The accompanying diagram (Figure 1) does not the understanding of the problems and make it clear which decisions and actions are opportunities, might be appropriate in situations intended to be informed and influenced by where conditions are rapidly changing and new learning and adapting. types of interventions are being planned quickly in response to urgent needs.

In terms of who is involved in adaptive behaviour, this depends on the level of adaptation involved. Adaptive management in terms of improving implementation of existing activities or implementing different activities for the same overall strategy can be implemented by managers at all levels, including front-line delivery and project managers. But adaptation that involves changing strategies or objectives is likely to require involvement of higher levels of management, including those making decisions about which projects to invest in.

The following sections provide some examples of how adaptive management has been described in terms of: Figure 1 : USAID (2018b) Discussion Note on • Changing intended causal pathways (and Adaptive Management hence actions) but not goals

• Changing both intended causal pathways and goals

• Changing intended causal pathways, goals and the understanding of the problem

Adaptation in terms of changing Adaptation in terms of changing both actions and goals actions, plans and understanding Another representation of adaptive management, of the problem seen frequently in natural resource management, A further representation shows an even greater shows adaptive management as informing both range of adaptations, including changing the long-run changes to planning and shorter-run objectives and even the understanding of the changes to actions. problem or situation being addressed.

A representation from the Tasmanian Parks and The following diagram (Figure 3) from the Wildlife Service (Figure 2) spelled out the different Californian Department of Fish and Wildlife (no levels of adaptation that are intended – changes to date) illustrates this level of change and Figure 4 strategies and actions in addition to the longer from ecological management (Schreiber et al, cycle of making changes to plans. 2004) includes explicit development and use of a model.

Figure 2: The adaptive management cycle with iterative adjustments to implementation (DPIPWE 2014; Jones Figure 3: The adaptive management cycle with three levels of adaptation. Californian Department of Fish and Wildlife (no date) A Queensland Government report, which used the same representations, defined adaptive management as:

Adaptive management involves continually monitoring a process to evaluate its effectiveness, and improving the process based on this evaluation. It requires transparent planning systems and implementation strategies, and a strong emphasis on monitoring and reviewing to ensure emerging information is reflected in future planning. Qld Department of State Development, Manufacturing, Industry and Planning, 2015

Figure 4: The adaptive management process with an emphasis on development and use of modelling options Schreiber et al, 2004

Different approaches to While there are differences in emphasis in each of these, when their central elements are compared, adaptive management in they all address four components:

international • Attention to political analysis and development engagement at multiple levels • Adaptation to local conditions and local In the context of development, adaptive ownership management under a number of approaches and • Framing work around problem solving labels is applied to the work being done by • Adaptation to changing conditions and development practitioners. new information Thinking and Working Politically (TWP) originated from a Community of Practice formed in Delhi in 2013 that brings together a group of researchers and senior officials from leading development organisations with the aim of moving from simply Aspects Possible implications ‘thinking politically’ to ‘working politically’ (TWP for monitoring and CoP, 2013). evaluation Sidel and Faustino (2019) cite the emergence in the 2000s of new ideas about how development Attention to Who is involved in the process of should work as one of the trends which fed into political monitoring and evaluation – including the TWP CoP’s establishment, stating that these analysis and changes over time ideas “emphasized politics, institutions, leaders, engagement at multiple How to make the processes of elites, and coalitions in the making of effective evaluation, including decisions about levels development programs” (p.10). One example of the evaluation plan and sensemaking, TWP in practice which emphasises these facets is accessible to all those engaged the Pacific Leadership Program, which has a strong focus on developmental leadership and reform Processes of developing theories of coalitions to work towards its goals (Denny and change that engage with and challenge stakeholders’ existing mental models McLaren, 2016). Theories of change that incorporate Problem Driven Iterative Adaption (PDIA) is an network analysis and the actions of approach developed by the Building State independent actors in helping or Capability faculty at Harvard Business School hindering intended changes which aims to facilitate the emergence of local solutions to local problems through a process of Adaptation Theories of change which explicitly and problem construction and deconstruction, to local coherently incorporate context, change experimentation and adaption (Building State conditions theories and action theories Capability, no date.). and local ownership Explicit analytic extrapolation to Collaboration, Learning and Adaption is an answer predictive questions – what is approach introduced by USAID’s Bureau for Policy, likely to happen if we do this in a Planning and Learning (PPL) in 2012 to different time or place? operationalize adaptive management throughout Framing Framing evaluation around specific USAID’s Program Cycle (USAID 2017b). work around issues rather than generic reporting in problem terms of OECD-DAC evaluation criteria Doing Development Differently is a community solving built around the DDD Manifesto that arose from an initial meeting in 2014 that aimed to bring Adaptation Timely data that can be accessed and together practitioners who were engaged in to changing used to make changes conditions Iteratively revised theories of change development practices that were dynamic and that incorporate changes to needs and appeared to have impact – the DDD manifesto has and new information opportunities (including engagement of over 400 signatories from 60 countries (Doing additional stakeholders) and Development Differently Manifesto, 2014). understanding

The summary presented below compares the key and understanding, responsiveness to local features of these approaches under four context and flexibility and adaptability (Thinking, components, drawing on and adapting the three Working Politically CoP, no date), and adding a principles associated with TWP – political insight fourth dimension related to problem-framing.

Attention to Adaptation Framing of Adaptation to political analysis to local problem changing and engagement conditions conditions and new at multiple and local information levels ownership Doing “legitimised at all levels Important for “focus on solving “rapid cycles of planning, Development (political, managerial and initiatives to be local problems that action, reflection and revision Differently social)” “’locally owned’ are debated, defined (drawing on local knowledge, in reality (not and refined by local feedback and energy) to (Doing “work through local just on paper)” people in an ongoing foster learning from both Development conveners who mobilise process” success and failure.” Differently all those with a stake in Manifesto, progress (in both formal Focus on Use of “‘small bets’: pursuing 2014) and informal coalitions sustainability by activities with promise and and teams)” focusing on “real dropping others.” solutions to real problems” Thinking and Focus on “power “Ensure as far as “Focus on problems Allows “for sufficient Working dynamics, interests, possible that identified and flexibility and iteration in the Politically incentives and locally-defined articulated by local day-to-day efforts to make institutions” problems and actors, not progress towards goals.” (Thinking, proposed outsiders” Working “Understand the network solutions are “continue to assess the local Politically CoP, of stakeholders involved accepted as context, test original no date) and facilitate coalitions of legitimate by all assumptions, and adapt different interests, rather relevant programs based on new than relying on a stakeholders, information and ‘principal-agent’ thereby ensuring opportunities” relationship” ownership” “series of small ‘experimental’ or ‘incremental’ steps and monitor results.” Problem- “Engaging champions “Locally driven, Avoiding “Promoting active Driven across sectors and where local “predetermined experiential (and Iterative organisations who ensure actors solutions” to focus experimental) learning with Adaptation reforms are viable, define, debate on “local evidence-driven feedback legitimate and relevant.” and refine the nomination, built into regular (Andrews et al., problem articulation and management that allows for 2017) statement prioritisation of real-time adaptation.” through shared concrete problems” consensus.” “emphasis on the process of solving problems” Collaboration, Need to “think more Locally-led (no explicit focus on “Continue to build in time Learning and strategically about approach: problem-framing) and budget space for Adaptation collaboration: who should “Facilitate, adaptation through we be collaborating with, rather than pilot/inception phases of (USAID, 2017) why and what form create, projects that enable a range should that collaboration development” of strategies to be tested in take” ‘small bets.’”

they are based; triple loop learning refers to What is needed learning about learning itself, and which processes for adaptive are being effective. Adult learning refers to adults’ needs for management autonomy and to connect what they are learning to what they already know. It suggests a need for to work? more engaged processes for making sense of data and its implications rather than passively accepting Essential elements for findings from evaluation. adaptive management Factors which support or hinder Underpinning adaptive management approaches these behaviour changes to development are explicit or implicit theories of Michie et al. (2012)’s meta-model of behaviour change which explain why working in this way is change (COM-B model) can be useful here. This understood to contribute to better development considers three components which influence outcomes. changes in behaviour: • (human capital such as skills and These can be thought of in terms of particular capacity knowledge, social capital, organisational changes in behaviour, especially: capital) • More effective collaboration – working with other individuals, groups and organisations • motivation (including incentives and disincentives) and • More effective learning – finding, retaining, making sense of and using information to • opportunity (including identifying and think or act differently removing barriers).

More effective collaboration Explicit theories of change for interventions using adaptive management often revolve around bringing together the right people to work on a problem. Strategies for effective collaboration are therefore an important part of adaptive management.

More effective learning

It’s important to note that learning does not take These different elements can be brought together place in a vacuum – it must be guided and into the overall theory of change of an adaptive supported. Effective theories of change for program – as in this description of Coalitions for adaptive management should explicitly include Change program, a partnership between the theories of building, sharing and using knowledge. Australian Embassy and The Asia Foundation in the Philippines, which focused on key policy reforms to Theories of change for adaptive management can improve lives of Filipinos and promote their include explicit attention to different types of economic well-being: learning, including: CFC creates spaces for collaboration and brings • single-loop, double-loop and triple-loop together motivated groups, organisations and learning (Argyris and Schön, 1978) individuals – whether they be politicians,

scholars, reform advocates or civil society • (Knowles, 1984) adult learning members – to build the demand for better governance. With the right people, sound Single loop learning refers to identifying and evidence, and valuable connections, these correcting discrepancies with targets and plans; coalitions and networks are well-positioned to double-loop learning refers to questioning the successfully influence policies. The Asia rules and targets and the assumptions on which Foundation, no date.

The USAID theory of change about adaptive Capacity management refers to capability (developing improved knowledge through intentional Capacity can be understood to refer to different learning), motivation (through incentives for types of capital - human capital (knowledge and learning and managing adaptively) and skills), social capital (networks and relationships opportunities (through removing barriers that and norms of trust and reciprocity), and prevent program modification) organisational capacity (e.g. IT equipment and access to data sets). For example, the USAID program cycle operational policy includes attention to supporting adaptive Personality and skills management, referring to the need to be able to adapt in response to changes in context and to Staff involved in adaptive management need new information. strong technical skills, including critical thinking, problem-solving and political analysis skills (Teskey Manage Adaptively through Continuous Learning and Tyrrel, 2017; Mercy Corps, 2015; BOND, 2016). Facilitating international development inherently They also need an ‘adaptive mindset’ – which requires that USAID work in countries with involves a tolerance for measured risk-taking and evolving political and economic contexts. USAID adaption (Bain et al., 2016). is increasingly working in countries that are unstable or in transition and even in the most Other desirable traits include being inquisitive and stable environments, it is difficult to reliably being able to ask the right questions, think on their predict how events or circumstances will evolve feet, and spot opportunities, as well as being able and impact programs. Therefore, USAID must be to empathise with colleagues (Dexis Consulting able to readily adapt programs in response to Group, 2017). It’s also noted that a deliberate changes in context and new information. To do effort to create diverse teams with a variety of this, the Agency must create an enabling backgrounds and expertise is important. environment that encourages the design of Leaders need a strong interest and enthusiasm for more flexible programs, promotes intentional the task of adaptive programming, including learning, minimizes the obstacles to modifying willingness to accept failure and uncertainty programs, and creates incentives for learning (Goeldner, Byrne et al., 2016; BOND, 2016) and a and managing adaptively. Learning can take strong sense of direction, perseverance and place through a range of processes and use a determination (Faustino and Booth, 2014). They variety of sources such as monitoring data, also need to be willing to step back and let others evaluation findings, research findings, analyses, take the lead in decision making, such as local lessons from implementation, and observation. actors and frontline staff (Booth and Unsworth, (USAID (2018a): ADS Chapter 201 Program Cycle 2014). Leaders are also important to building a Operational Policy, pp 11-12. Emphasis added.) culture that celebrates learning, embraces failure, and encourages a positive risk culture of informed Monitoring and evaluation can contribute to, or risk-taking. fail to contribute to, each of these aspects, either directly or indirectly. The 3rd paper in this series, Networks and relationships ‘Monitoring and evaluation methods, processes The networks and relationships of individuals are and approaches for adaptive management’ provides more detail on these. key to the success of locally-led, adaptive approaches such as TWP, PDIA, DDD and CLA, and the ability to convene and facilitate external parties is important (Laws and Marquette, 2018; Bain, Booth, and Wild, 2016; USAID 2018b). This capacity is often equated with the role of a “political insider”, though it is often acknowledged how difficult it is to find true insiders (Teskey and Tyrrel, 2017; Wild, Booth, Cummings, Foresti, and Wales, 2015; Faustino and Booth, 2014).

Time and resources are needed to build and strengthen trusting relationships with external networks and partnerships. For managing

contractors, an effort should be made to ensure Knowledge of rules, systems and teams are made up of individuals with strong local processes networks, where possible. Donors and funders also have a role to play in terms of their ability to Knowledge of internal systems was also important, convene and facilitate linkages across groups. including staff knowledge of organisational rules and budget flexibility and procedures around Trust decision-making (Goeldner Byrne et al., 2016). This is important as while a program team might have a An important point in terms of both internal and solid understanding of what changes need to be external relationships is that it is not just the made to improve implementation, a lack of existence of the networks that is important, but understanding of organisational policy will very that trust is a vital element to cement these. This easily delay or even block this. In an includes trust between implementers and external implementation context that relies on rapid actors, as well as between external parties and experimentation and timely adjustments, this can groups (Laws and Marquette, 2018; USAID, 2015; be a major stumbling block to effective adaptive USAID, 2017c). Trust has been widely identified as management. important to facilitating adaptive behaviour (Honig and Gulrajani, 2017; Wild, Booth, and Valters, Knowledge management 2017) and it is important to have the ability and processes in place to build trust. Trust is important Quality knowledge management systems are on multiple levels – between the donor and highly important to effective learning and managing contractor; between the managing adaption. This is noted in USAID’s CLA literature contractor and staff; and between staff. review (USAID, 2017c) and is also found as a key Communication and early wins can help in building recommendation by the independent review of trust (Teskey and Tyrrel, 2017); as can informal KOMPAK (Whitelum, 2018), a facility funded by opportunities for information sharing and the Government of Australia (GoA) to support the encouraging social sensitivity (USAID, 2017c). The Government of Indonesia (GoI) in achieving its ability to convene and facilitate external parties is poverty reduction targets and addressing important here. inequality. Knowledge management systems can come in many forms. It’s important to think about Time and workload user needs, resources and motivations when designing these. Adaptive management practices take time to do well. This needs to be factored into workplans and Learning culture efforts made to reduce administrative burdens where possible in order to “free up managers to USAID’s literature review of the CLA literature manage” and allow program staff time for found the following features of organisational reflection, learning and planning. How this is culture to be important: collaborative teams, managed would be expected to change over time: encouragement of honest discourse and debate, encouraging a learning culture, group tolerance for Once there is a better balance of work risk taking and high levels of trust between team programmes and a ‘good enough’ understanding members (USAID, 2017c). An enabling of the organisation’s new, lighter bureaucratic environment for adaptive management includes burden, it should be possible have a candid attention to improving communication practices discussion about performance and and creating more learning opportunities, face to accountability, in light of the weight being given face and virtual, as well as opportunities for to new soft skills through the new core coaching and mentoring which can “build competencies. Bain et al. (2016) organizational culture and resilience for adaptive management” (Desai et al., 2018). It’s also Contextual knowledge important to build dedicated learning and reflection time into contracts and roles (USAID, Local and contextual knowledge of staff and 2017c). Formal roles can help support learning, managers is important for successful adaptive e.g. regional advisors, periodic reviewers, management (USAID, 2018b; Goeldner Byrne, communities of practice (Goeldner Byrne et al., Sparkman, and Fowler, 2016). This is often 2016) something that organisations will need to specifically hire for in order to create teams with diverse backgrounds and which include local staff.

Motivation This includes incentives to share ideas, encourage internal sharing, and maintain sustained Motivation can be understood in terms of both the engagement and communication (Andrews, positive and negative incentives to behave in Pritchett and Woolcock 2016a; USAID 2017; Desai particular ways, either because of observed or et al., 2018). assumed consequences. In practice, the actual BOND (2016) noted incentives need to be aligned incentive system might encourage behaviours that across the organisation: “adaptive management are different to the stated norms and desired requires multiple teams and parts of the behaviours. organisation to work in concert: programmes, Positive incentives operations, MEL, finance, HR, communications, etc. If those different teams and functions have Clear signals from leadership about the value of incentives and drivers that are in tension with one adaptive behaviours, and tying rewards and another, progress on adaptation will be painfully milestones to these, is an important element of slow.” incentivising adaptive management. This is important at the donor-implementor relationship, Negative incentives as well as within implementing teams (Laws and As important as providing positive incentives is the Marquette, 2018). removal of negative incentives that actively work Teskey and Tyrell (2017) suggested several ways in against staff displaying adaptive behaviour. These which donors can do this: disincentives can include:

focusing performance incentives at the hostile or high-profile political environments outcome level; – political environment, particularly focusing performance milestones in the first surrounding projects with high spend and 12-18 months of the program on high political visibility, can create operational system quality; additional pressures that disincentivise rewarding number of changes to result from adaptive behaviours (Goeldner Byrne et R&R exercises, rather than simply al., 2016; Teskey and Tyrrel, 2017; rewarding the number of R&R exercises; Mackenzie and Hearn, 2016). It’s been encouraging and rewarding reporting on noted that not all programs may be failure. suitable for adaptive management due to In addition, BOND (2016) suggested donors should their high-profile (Teskey and Tyrrel, adopt results frameworks that 2017); Mackenzie and Hearn, 2016).

“better recognise ‘success’ in complex, adaptive inflexible M&E systems and reporting programmes, for example through menus of requirements – e.g. Inherited log frames indicators, inclusion of good learning and that have limited room for flexibility and adaptation practice as a result in its own right, adaption (Goeldner Byrne et al., 2016) and seeking and expecting a contribution to and reporting requirements with the change, rather than results that can be potential to create incentives to “measure attributed solely to one intervention.” the measurable” and under-resource the unmeasurable (Honig and Gulrajani, Bain et al. (2016) in a paper for ODI note that 2017). incentives are needed at multiple levels, recommending rewards and promotions at the inflexible or inadequate finance and individual, team and manager level be linked to procurement systems – Budgetary and evidence of learning, adaption, critical thinking, accountability practices were cited widely partnership and debate. as a disincentive for adaptive behaviours. While flexibility in budgets is widely seen To encourage behaviour within and across as supporting adaptive practices, this program teams, managing contractors should frequently comes up against a need to introduce formal and informal mechanisms (Mercy spend money (Bain et al., 2016; Booth, Corps, 2015; Bain et al., 2016; USAID, 2017c; 2013; Goeldner Byrne et al., 2016). BOND, 2016).

inflexible or inadequate procurement Opportunity systems Procurement systems can also disincentivise adaptive behaviour by Having skills, knowledge and motivation to requiring a high level of specificity from manage adaptively is not enough. The systems the very outset, which can set the tone used to manage interventions and investments for how much adaption is possible need to provide opportunities for responsive, (Goeldner Byrne et al., 2016). flexible decision making.

isomorphic solutions – solutions designed to Flexible funding satisfy external funders but which lack Including a reserve fund for learning and adaption sustainability (Laws and Marquette, can be a valuable mechanism to support the ability 2018). to adapt when needed. Algoso (2017) suggests

flexible funding features might include dedicated evidence preferences – top-down preferences funding for innovation and risk-taking or triggering for certain types of evidence (such as mechanisms for funding increases or other budget quantitative information), which is noted changes. Desai et al. (2018) suggest sharing of by staff at both USAID and DFID staff as budget targets and centralised ‘rainy day’ funds for an issue (Goeldner Byrne et al., 2016). needs-based adjustments can be useful tools.

unaligned priorities – Such as donor priorities Flexibility in planning being seen as more important than those of partner governments (Teskey and Adaptive log frames and broad but defined Tyrrel, 2017), or having prescribed targets indicators, leaving room for adjustments, are more or partner organisations (Laws and useful tools for flexibility than traditional log Marquette, 2018). frames. Desai et al. (2018) note that communication and trust between donors and implementors is key to flexible planning, as is training staff to understand flexible systems and processes.

Innovative contract mechanisms Mechanisms such as payment by results, agile grants and fixed price mechanisms can improve adaptive behaviours by incorporating new incentive structures (Goeldner Byrne et al., 2016).

Inception periods Inception periods that allow time for strategic design and experimentation also help enable adaptivity (Dexis Consulting Group, 2017; Goeldner Byrne et al., 2016; Mercy Corps, 2015; Bain et al., 2016; BOND, 2016). Desai et al (2018) cite a DFID example of an inception period for the SWIFT program that allowed SWIFT’s stakeholders to innovate and refocus activities after trial and error, while Teskey and Tyrell (2017) suggest inception periods should be built into donor expectations for large facilities.

The other papers in this series ‘Monitoring and evaluation for adaptive management: An overview’ and ‘Monitoring and evaluation methods, processes and approaches for adaptive management’ provide further references and resources.

Cited References Booth, D., & Unsworth, S. (2014). Politically smart, locally led development, London: ODI. Algoso, D. (2017). How INGOS are Doing Retrieved from Development Differently. Retrieved from https://www.odi.org/sites/odi.org.uk/file https://www.wvi.org/sites/default/files/H s/odi-assets/publications-opinion- ow%20INGOs%20are%20DDD.pdf. files/9204.pdf. Andrews, Matt, Lant Pritchett, and Michael Californian Department of Fish and Wildlife (2009). Woolcock. 2012. “Escaping Capability Adaptive Management. [Website] Traps through Problem Driven Iterative Retrieved from Adaptation (PDIA)”. Retrieved from https://www.dfg.ca.gov/erp/adaptive_ma https://bsc.cid.harvard.edu/publications/ nagement.asp escaping-capability-traps-through- Denney, L. & McLaren, R. (2016). Thinking and problem-driven-iterative-adaptation-pdia. Working Politically to Support Andrews, M., Pritchett, L., & Woolcock, M. (2016) Developmental Leadership and Coalitions: Doing Iterative and Adaptive Work. In, CID The Pacific Leadership Program. Working Paper: Center for International (Research Paper.) Suva / Birmingham: Development at Harvard. Retrieved from Pacific Leadership Program / https://bsc.cid.harvard.edu/publications/ Developmental Leadership Program, doing-iterative-and-adaptive-work. University of Birmingham. Retrieved from https://www.dlprog.org/publications/res Andrews, M., Pritchett, L., and Woolcock, M. earch-papers/thinking-and-working- (2017). Building State Capability: politically-to-support-developmental- Evidence, Analysis, Action. Retrieved from leadership-and-coalitions-the-pacific- https://bsc.cid.harvard.edu/building- leadership-program. state-capability-evidence-analysis-action. Desai, H., Maneo, G., Pellfolk, E., & Schlingheider, Andrews, M., Pritchett, L., Samji, S., & Woolcock, A. (2018). Managing to adapt: Analysing M. (2017). Building capability by adaptive management for planning, delivering results: Putting Problem-Driven monitoring, evaluation, and learning, Iterative Adaptation (PDIA) principles into Oxfam. Retrieved from https://policy- practice. In Whaites et al. (2015), A practice.oxfam.org.uk/publications/mana Governance Practitioner’s Notebook: ging-to-adapt-analysing-adaptive- Alternative Ideas and Approaches, OECD, management-for-planning-monitoring- Paris, (pp. 123-133) evaluat-620446 https://www.oecd.org/dac/accountable- effective- Dexis Consulting Group. (2017). Evidence Base for institutions/Governance%20Notebook.pd Collaborating, Learning and Adapting: A f summary of the literature review update, November 2017,USAID. Retrieved from Argyris, C., & Schön, D. (1978). Organizational https://usaidlearninglab.org/sites/default learning: A theory of action perspective. /files/resource/files/eb4cla_summary_lite Reading, MA: Addison Wesley. raturereview_20171020_final.pdf. Bain, K. A., Booth, D., & Wild, L. (2016). Doing Faustino, J., & Booth, D. (2014). Development Development Differently at the World entrepreneurship - How donors and Bank: Updating the plumbing to fit the leaders can foster institutional change. architecture. The Asia Foundation/ODI. Retrieved from Beck et a, (2001) ‘Manifesto for Agile Sofware https://www.odi.org/sites/odi.org.uk/file Development’. Retrieved from s/odi-assets/publications-opinion- http://agilemanifesto.org/ files/9384.pdf. BOND. (2016). Adaptive management What it Goeldner Byrne, K. Sparkman, T. Fowler, B. (July means for CSOs. Retrieved from 2016) The road to adaptive management: https://www.bond.org.uk/sites/default/fil knowledge, leadership, culture and rules. es/resource- The BEAM Exchange, accessed from documents/adaptive_management_- www.beamexchange.org . _what_it_means_for_csos_0.pdf

Honig, D., & Gulrajani, N. (2017). Making good on Qld Department of State Development, donors’ desire to Do Development Manufacturing, Industry and Planning Differently. Third World Quarterly, 39(1), (2015). Great Barrier Reef Coastal Zone 68-84. Strategic Assessment Reef. Retrieved doi:10.1080/01436597.2017.1369030 from https://www.statedevelopment.qld.gov.a Larman, C., & Basili, V. R. (2003). Iterative and u/resources/report/gbr/chapter-9- incremental developments. a brief adaptive-management.pdf history. Computer, 36(6), 47-56. Retrieved from Schreiber et al (2004) Adaptive management: a https://www.craiglarman.com/wiki/down synthesis of current understanding and loads/misc/history-of-iterative-larman- effective application, Ecological and-basili-ieee-computer.pdf. Management & Restoration, Vol 5 No. 3 December 2004, 177-182. Laws, E., & Marquette, H. (2018). Thinking and working politically: Reviewing the Sidel, J. T., and Faustino, J. (2019). Thinking and evidence on the integration of politics into Working Politically in Development: development practice over the past Coalitions for Change in the Philippines. decade. ODI. Retrieved from Pasig City: The Asia Foundation. Retrieved https://www.odi.org/publications/11115- from thinking-and-working-politically- https://asiafoundation.org/publication/th reviewing-evidence-integration-politics- inking-and-working-politically-in- development-practice-over. development-coalitions-for-change-in- the-philippines/. Mackenzie, J., & Hearn, S. (2016). Impact evaluation for portfolio programmes on Stankey, George H.; Clark, Roger N.; Bormann, policy influence: Reflections on the Bernard T. 2005. Adaptive management Indonesian poverty reduction support of natural resources: theory, concepts, facility. London: Overseas Development and management institutions. Gen. Tech. Institute, Methods Lab series. Retrieved Rep. PNW-GTR-654. Portland, OR: U.S. from Department of Agriculture, Forest Service, https://www.odi.org/publications/10397- Pacific Northwest Research Station. impact-evaluation-portfolio-programmes- https://www.fs.fed.us/pnw/pubs/pnw_gt policy-influence r654.pdf Mercy Corps. (2015). Managing complexity: Teskey, G., & Tyrrel, L. (2017). Thinking and Adaptive management at Mercy Corps. working politically in large, multi-sector Retrieved from Facilities: lessons to date. In: Abt https://usaidlearninglab.org/sites/default Associates Governance Working Series. /files/resource/files/managing_complexit Retrieved from y_- https://abtassocgovernancesoapbox.files. _adaptive_management_at_mercy_corps wordpress.com/2017/11/abt-associates- .pdf governance-working-paper-series-issue- no-2-final-171120.pdf. Michie, S., Van Stralen, M. M., & West, R. (2011). The behaviour change wheel: a new Thinking Working Politically (TWP) Community of method for characterising and designing Practice (no date) What is TWP? [Website] behaviour change interventions. Retrieved from https://twpcommunity.org/about- Implementation science, 6(1), 42. us/what-is-twp Retrieved from https://implementationscience.biomedce The DDD Manifesto Community. (2014). The DDD Manifesto. Retrieved from ntral.com/articles/10.1186/1748-5908-6- https://buildingstatecapability.com/the- 42. ddd-manifesto/ Parnas, D. L., & Clements, P. C. (1986). A rational USAID (no date). CLA Toolkit. [Website]. Retrieved design process: How and why to fake it. from https://usaidlearninglab.org/cla-toolkit IEEE transactions on software engineering, (2), 251-257. USAID. (2015). The Facilitation Approach at USAID: A Discussion Paper. Retrieved from https://usaidlearninglab.org/sites/default

/files/resource/files/the_facilitation_appr Williams, L., & Cockburn, A. (2003). Agile software oach_at_usaid_07.28.2015.pdf. development: it’s about feedback and change. IEEE computer, 36(6), 39-43. USAID. (2017a). ADS 201 Additional Help - Drafting Retrieved from a Collaborating, Learning and Action plan https://cs.anu.edu.au/courses/comp3120 (Version 2/ October 2017). Retrieved /local_docs/readings/WilliamsAndCockbu from rn_AgileSoftwareDevelopment_ItsAboutF https://usaidlearninglab.org/sites/default eedbackAndChange.pdf. /files/resource/files/drafting_a_cla_plan_ -_update_october_2017_r1.pdf. USAID. (2017b). Collaborating, Learning, and Adapting (CLA) AN ANALYSIS OF WHAT CLA LOOKS LIKE IN DEVELOPMENT PROGRAMMING. Retrieved from https://usaidlearninglab.org/library/colla borating-learning-adapting-analysis-what- cla-looks-development-programming. USAID. (2017c). What difference does CLA make to development? - Key findings from a recent literature review. Retrieved from https://usaidlearninglab.org/library/what- difference-does-cla-make-key-findings. USAID. (2018a/2020). ADS Chapter 201 - Program Cycle Operational Policy. Retrieved from https://www.usaid.gov/sites/default/files /documents/1870/201.pdf USAID. (2018b). Collaborating, Learning & Adapting Case Analysis: Deep Dive Summary - Global Communities’ Ebola Response in Liberia. Retrieved from https://usaidlearninglab.org/library/globa l-communities%E2%80%99-ebola- response-liberia-cla-case-analysis-deep- dive. Whitelum, B. (2018). Independent Progress Review: KOMPAK Australia Indonesia - Partnership for Decentralized Basic Service Delivery. Retrieved from https://www.dfat.gov.au/sites/default/fil es/kompak-independent-progress- review.pdf. Wild, L., Booth, D., Cummings, C., Foresti, M., & Wales, J. (2015). Adapting development - Improving services to the poor. London: ODI. Retrieved from https://www.odi.org/sites/odi.org.uk/file s/odi-assets/publications-opinion- files/9437.pdf. Wild, L., Booth, D., & Valters, C. (2017). Putting theory into practice: How DFID is DDD. London: ODI. Retrieved from https://www.odi.org/sites/odi.org.uk/file s/resource-documents/11332.pdf.