$15bn is spent every year on training, with disappointing results. Why the aid industry needs to rethink ‘capacity building’.

Guest post from Lisa Denney of ODILisa Denney

Every year a quarter of international aid – approximately US$15 billion globally – is spent on capacity development. That is, on sending technical assistants to work in ministries or civil society, running training programmes, conducting study tours or exchanges, or supplying resources and equipment to help organisations function better. This is often referred to as ‘teaching men to fish.’ Rather than giving men (or, one might add, women) fish, teaching men to fish is seen to provide sustainable capabilities that will empower people and eventually negate the need for external support. Yet this task, far from being the technical transfer of skills, is fundamentally about social and political change.

In countries affected by conflict or fragility, this assistance is not just about improving development outcomes but is also expected to strengthen the state itself – because its very weakness is framed as both a cause and consequence of violence and underdevelopment. By defining state fragility as a problem of weak capacity, capacity development becomes the primary solution. It is also, conveniently, a problem that donors and NGOs can do something about – by providing capacity development programs.

teach_a_man_to_fishBut despite the dominance of this idea, results in practice are frequently disappointing. This is borne out by six-years of research by the Secure Livelihoods Research Consortium (SLRC) on state capacity and how it tends to get built in eight countries (Afghanistan, the Democratic Republic of the Congo, Nepal, Pakistan, Sierra Leone, Sri Lanka, South Sudan and Uganda). While diverse, the studies point to a range of familiar challenges, including:

  • the limited toolkit of capacity development approaches (as one respondent in Sierra Leone noted ‘training, training, training – how much training does one person need?’);
  • the focus on technical aspects of service delivery, neglecting how power and politics are at the root of many problems that appear to be about capacity;
  • the neglect of alternative capacities outside of the formal realm or what Western notions of ‘capacity’ look like; and
  • a focus on tangible ‘units’ of a delivery systems (that is, individuals and organisations) and less on the system as a whole and how its parts interact.

Given that capacity development activities stretch back to at least the 1950s, why the frequent lament of poor results? (The SLRC synthesis report is far from the first study to point to these limitations).

The SLRC research finds many of the reasons are to do with the political economy of the aid industry itself – short timeframes, quantitatively-driven results reporting, risk aversion and an emphasis on technical rather than political and contextual skills and knowledge. But perhaps most fundamentally, and what all these factors derive from, is the fact that the aid industry turns fundamentally political processes of social and institutional change into projects – which are in turn cloaked in value-neutral technocracy. The problems become ones of limited equipment or resources, insufficient knowledge, weak coordination, poor monitoring, and so on, rather than about power, incentives and interests.

So, we see aid programmes attempting to improve service delivery in Afghanistan by building the technical capacity of villagecapacity building thanjavur development councils, overlooking the incentives of local power brokers who hold the real sway in how services are delivered. Or programmes imply that malnutrition in Sierra Leone can be reduced by training mothers in infant and young child feeding, with little regard for the gendered power relations that mean women have little control over household finances and serve the most nutritious meal portions to older men. In South Sudan, there are examples of statebuilding being boiled down to workshops – often conducted in English – to address complex social transformations, such as local justice and peacebuilding. In all of these cases, the complexity, power and politics of the changes sought are sidestepped and turned into technical, more easily-delivered projects.

We would never think to use logframes and value for money calculations to manage the civil rights movement, for instance (although such programme management tools are used for all kinds of domestic social policy reforms). So why do we think that processes of social change can be projectized in countries affected by conflict or fragility? Is it because, while we recognise the politics inherent in some social change in our own countries, we strip it out from social change processes elsewhere and treat problems as ones of capacity alone?

To address these limitations, the SLRC argues that we need a re-politicisation of capacity development, acknowledging that we are ultimately interested in fostering social and political change. That might be by making service delivery more equitable, changing power relations in the household, or making governance arrangements more accountable. None of these are technical endeavours. Rather than shying away from this politics, we must put it front and centre and engage with it.

gandhi and metrics h-t aid thoughtsTo do so, we might start from building a nuanced understanding of how people use services in practice – recognising what drives people’s decision-making and the breadth of providers people rely on, often extending beyond the state. We must move beyond thinking about capacity as the tangible assets of individuals (eg: teachers) and organisations (eg: schools) and begin to think much more holistically about the capacity of systems. Ensuring health workers have good technical knowledge does not make a well-functioning health system on its own. Ensuring that staff also have good bedside manner, the trust of the community, access to reliable drug supply chains, are paid on time and so do not charge bribes or sell drugs on the black market, and so on also matter.

Finally, we must be prepared to change our ways of working and thinking about the challenges faced in countries affected by conflict and fragility and move beyond the idea that capacity is the key constraint and therefore capacity development the natural answer. Rather, we need to bring to bear a much wider toolkit to support what are social and political processes of change. Ironically, this may begin by improving the capacity of development practitioners themselves to support such change processes.


Subscribe to our Newsletter

You can unsubscribe at any time by clicking the link in the footer of our emails. For information about our privacy practices, please see our .

We use MailChimp as our marketing platform. By subscribing, you acknowledge that your information will be transferred to MailChimp for processing. Learn more about MailChimp's privacy practices here.


13 Responses to “$15bn is spent every year on training, with disappointing results. Why the aid industry needs to rethink ‘capacity building’.”
  1. Rick Messick

    Thought this would be a useful footnote to a very useful post —

    “In relative terms, the highest direct public losses are encountered in training projects (67% of budget lost in projects affected) . . . .” PwC & Ecorys [for EU], Identifying and Reducing Corruption in Public Procurement in the EU, 2013, p. 175.

  2. Robin

    Wonderful to see this study and post. I’m so careful to assess any possible “capacity building” projects now (in part because I view so many as probably ineffective or counter-productive), that I’m not working on any such projects at all – even though I love engaging in properly formulated training, coaching, and problem solving!

  3. lilych

    Capacity building is essential esp as starter to a development activity, to define/level off on the roles/participation/linkages (direct or indirectly) of stakeholders, the introduction of technology, legal/institutional mandates, responsibilities & accountabilities etc etc . The training design should identify specific outputs in direct correlation with the activities, as well as undertake a thorough assessment on the implementation (how doable) of the knowledge/skills to be acquired. When capacity building is a tool to introduce an approach or process that impacts on the structure/systems/work culture etc, then capacity building of individuals is not an appropriate stand-alone intervention.
    And without providing for means to measure or assess relationship of capacity building to desired outcomes… capacity building becomes simply an added value to the trainees’ stock of knowledge and a lucrative profession for trainors.

  4. Geoff

    Good blog Lisa.

    I am not as wedded to systems thinking as the silver bullet though. Prefer learning by doing approach focussed on specific problems rather than shooting for the holy grail of systemic change. In the latter, I find that you never fully understand the system as it is always changing (mainly in the informal side), and even if you could it does not mean you will know what to do to successfully chart a course leading to reform. Paralysis by analysis.

    There is more thinking that could be done on why donor’s stay in the safe space of technical interventions. Partner national sovereignty and donor national interest often make strange bedfellows.

  5. Lisa Denney

    Many thanks for the comments – and to Duncan for hosting the discussion.
    Geoff, I think you raise an interesting question about why donors stay in the safe space of technical interventions. For me, this goes back, first of all, to their own domestic politics – what they have the political space to do, what is easily sold to politicians and home publics, what attracts donations (if you’re an NGO), etc. But it also speaks to a deeper bureaucratic rationality that informs development, as well as public sector programming, more broadly – that turns political projects (whether it be universal healthcare, public education, a more accessible justice system, etc.) into compartmentalised projects. Kind of the Fordism assembly line approach to social change.

    Lilych, I wonder whether starting with defining roles and responsibilities, introducing new approaches/technologies, etc. sets us up to get caught in the same project thinking that is ultimately unhelpful. Spending more time understanding how things actually work can help us see what is actually working (and by what logics), what’s not working and why, etc. and can help us move beyond that deficit lens that assumes ‘we’ are bringing something others lack. Similarly, matching outputs to outcomes becomes tricky. Of course, we should have a theory of how we anticipate change/outcomes might be possible in a given context, but we should accept that we often don’t know what will work. Here, I found useful a paper by Teles and Schmitt called ‘The elusive craft of evaluating advocacy,’ which demonstrates how difficult it is to match inputs and outcomes in relation to advocacy/public policy/social change. It’s a literature others might already be much more familiar with but I found it offered some interesting ideas.

    Robin, as you say, avoiding the pitfalls of conventional CD in our own day-to-day is astonishingly difficult! I also find it very easy to revert to trainings or substituting capacity that, when I think about it, I know don’t really work. Maybe we need some kind of reminder/prompt on the wall above our desks that push us to sense-check this sort of defaulting thinking and make sure we’re all applying what we actually know. But I also know from my own experience that the very tangible constraints of time, budget, competing priorities, etc. can also encourage a reversion to conventional CD. Those things are harder to budge.

    And Rick, thank you for that reference – I am going to go and check out that report now!

  6. Arjun Tasker

    Hi Lisa,

    I very much like your post. I think our training often lack any context awareness and rarely are designed to teach a specific skill. Training that just introduces concepts are not actually training one IN anything, particularly when we have no idea if the concept is actually relevant to the audience’s context.

    Your post also brings to mind some of the work colleagues in USAID are doing. The focus is on improving an organizations fit to the local system, assuming that improving adaptiveness and connections between organizations is more useful than trying to teach best practices.

    This document is illustrative: https://usaidlearninglab.org/sites/default/files/resource/files/ads_additional_help_lcd_1.13.2017.pdf
    But I would just highlight this list from the Table of Contents:

    Capacity Development Considerations
    a. System-dependent
    b. Complexity
    c. Interrelationships
    d. Timeframe
    e. Responsiveness to change
    f. Local ownership and capacity development
    g. Timeframe

  7. lilych

    Just an afterthought …again, i am thinking (as you noted) in a project context (same as in contract or a cooperative agreement). “How things work,what is working and what’s not working” would be in the realm of research (not capacity building), specifically needs assessment or an evaluation by whatever methodology one pleases. I would think “what things work” is always contextual and lessons may be replicable, however considering that development environment everywhere is dynamic, the factors that come together in one case that predisposed for success in an instance may not be the same in another/other contexts….therefore, time and space are independent variables to consider. Matching outputs to expected outcomes is inevitable (again from a project/contract/ cooperative agreement context) — esp if funding is outputs-based with a logframe that indicate a timeline on expected deliverables (outcomes). On advocacy as well as with all other capacity building/development activities … prior to the exercise, parties agree on the basic parameters including expected outcomes they all agree are doable, achievable within a defined timeframe. Again.. we differ in perspectives which is good — as it broadens the space for discussion. Thank you and appreciate the reference.

  8. I agree mostly with the idea that emphasis is not on the system and the interplay of the the components and the power perspective is overlooked. One or two points I like to raise: a) We over expect from the training – it cannot solve all the problems rooted in the society, b) The trainee is laden with values that may not be compatible with what we wish to convey. For example, you may be delivering a good method of making compost but its not a prestigious of jobs to work with hands in the society. c) And many a times, the delivery of training is good but not applied because they come for the trophy of per diem that most development agencies should commit for this I would call it a sin. Prime objective is not learning. Unless there is collective commitment it is hard for a single agency to respond to this problem.

  9. Vincent

    Capacity building that has become (to an extent) the goal instead of the means to a goal is fraught with danger; and the proberb used as an example is a good example of my point. that there is already preconceived notion that the challenge is about capacity… give a man a fish and he will eat for a day (well maybe not, maybe he’s like my Indian friend who is vegetarian) but teach him how to fish… (again, maybe not if he is from an arid area far away from fishinggrounds) an analogy of course.

  10. Erin

    The timing of your blog post is excellent for me Lisa as I’m grappling with these issues right now in implementation. Recently I was told ‘we need a capacity development specialist’ and I realized I hadn’t been asked that for many years. Thinking more on this I realized that previously I had been in a setting which worked politically in a middle income country (one you might be working with now) and we had never talked about capacity development (while it was implicit in everything we did) it wasn’t the means to our ends. Instead we focused on the right actors both within and outside the program, local ownership and getting in behind partners and letting them lead on reform agendas. Most of what we did was TA, supporting development of evidence bases for evidence based policy advocacy and capacity development but it was never called that because it wasn’t capacity development for capacities sake.
    Now I’m in a post-conflict high aid dependancy low/middle income country (sorry for the jargon typology) where the capacity development wheel has been spinning for a while and has come back around again, and I can’t ‘buy’ that capacity development will solve what are political issues. Some of the questions we are asking ourselves and grappling with are – How do you shift thinking in this context? Should we be looking at teams of local and regional actors who could work on these issues politically? How do you work on political issues in a context with limited actors and fragility? How do you improve capacity development practitioners? Is it about hiring? Or training? Or ways of working? How do we create the flexibility in programs for this way of working?
    My overwhelming experience of capacity development as an approach is that its often necessary, but never sufficient (alone) in development.

  11. A few of my essential ‘aide-memoires’ to optimizing chances for success of any potential capacity building exercise:
    1. Understanding the metacontext: country, culture, society; dynamics with other countries, especially neighbours; history.
    2. Understanding the mesocontext: specific host organization, government bureaucracy; who are the decision-makers; how is the interaction between NGOs and government; what regulations, if any, may have an effect on the success of those who complete the programme etc.; details on any similar previously implemented programme(s)—content, shortcomings, successes; receptivity of host(s).
    3. Determining the actual issues and needs: via participatory investigation with key stakeholders including peripheral and ‘weaker’ individuals and groups.
    4. Process of planning the ‘remedy’: jointly designing the scope of planned intervention; realistic expectations.
    5. Process of intended programme delivery: joint facilitation, not just by outsiders; use of locally relevant modalities; exploration of alternative capacity building modalities
    6. Knowledge of the probability of those whose capacity has been ‘upgraded’, ‘refreshed’ or ‘enhanced’ to actually get opportunities to use newly acquired knowledge and skills
    7. Ensuring that those positioned above the intended participants (e.g. supervisors, superintendents, directors, etc.) would also participate, at least in per-identified critical sessions
    8. Immediate post-programme review and documentation of suggested changes for future such programmes
    9. Understanding and agreement on what –if any– follow-up there would be to optimize retention and evolve locally planned follow-ups.
    10. Be reasonable: even in ‘developed’ societies we can overdo the amount of capacity building that is instituted—often to the detriment of the principal work to be accomplished.

  12. Donna

    Hi Lisa, thanks for stimulating this discussion. I feel both heartened and dismayed reading the post. Firstly, the positive – it is great development folk are talking about issues of ‘capacity development’, how it conceptualised and practiced and seeking ways to improve approaches to development to achieve better outcomes. Secondly, the less positive – the issues discussed here have been raised in many forums for more than a decade. Capacity development was central to my own doctoral research and this drew on research from the likes of: UNDP who critiqued technical cooperation in the early 2000s; ECDPM’s series on capacity development from nearly 10 years ago (http://ecdpm.org/publications/capacity-change-performance-study-report/); and Nils Boesen and Ole Therkildsen’s work (Danish Institute for International Studies, one report was called ‘A pragmatic approach to CD’) that examined different types of CD, the ‘mix’ that was needed, and the context and politics of CD. I haven’t been following the debates on capacity so closely in the last 3-4 years but the discussion here feels very familiar. So I wonder if or how the thinking about capacity development has advanced in recent years, what has been learned and is being applied, or if it hasn’t changed much then why not.