Jul 26, 2016 | Review

The views expressed here are those of the authors and do not necessarily represent those of DFID or the UK Government. 

This peer reviewed report – Evidence Synthesis: Security Sector Reform and Organisational Capacity Building – is available open access here.

Many millions have been spent trying to reform the security sector in developing countries. Since the late-1990s, these programmes have become popular as the importance of security for development was recognized. In countries as diverse as Jamaica, Liberia, Malawi, Pakistan and Timor-Leste, donors and other parts of government have trained police and judges, built courts and police stations, renovated prisons, supplied uniforms and radios, introduced community policing, amongst a bevy of reforms. But have these investments paid off? Are security actors more accountable, responsive and able to deliver for their communities? Perhaps most importantly, are people safer as a result?

We set out to answer these questions in a recently published review of international experience for the UK’s Department for International Development. A pretty bleak picture emerges, pointing to serious flaws in capacity building approaches of international actors seeking to reform the security sector. But this bleak picture is obscured by the fact that only a limited literature within the wider research on SSR actually aims to address these questions – so we still have big knowledge gaps.

What research there is suggests assistance is often poorly tailored to the country context and is unsustainable, driven by the skills and knowledge of those implementing programmes, rather than by the needs of recipients. Assistance is criticized for assuming that injections of skills, knowledge or equipment will fill gaps in capacity that will enable improved security – failing to recognize that often insecurity is the result of political interests and incentives, not just a lack of ‘stuff’. Some of the worst examples are of ever-popular training programmes, as an example from Afghanistan captures:

 [Training] is largely developed and delivered by operational police … [who] are not experienced trainers … The achievement of training objectives, such as knowledge transfer and competency development, are thus sabotaged by ineffective training methods, poor learning materials, bad examples and sometimes out-dated information. It takes extraordinary motivation for an Afghan police officer to learn anything from sitting on a hard seat in a classroom that is too hot in the summer and too cold in the winter, listening for several hours a day to an instructor speaking an unintelligible language, which has to be translated by, often untrained, interpreters uncertain of police terminology, without the support of words written in the local language on the board, good training materials or any sort of audio-visual support. The inexperienced trainers make little use of proven adult learning techniques such as group discussions, scenarios, role playing and analysis of case histories, which promote participation, learning and retention of information.

The reliance on overwhelmingly technical skills in SSR is also criticized. This is not to suggest that there is no place for police, judges and lawyers in training their developing country counterparts – but technical skills are not enough. This point is strikingly made in relation to Timor-Leste: ‘no one would consider establishing a health service by putting 1,000 doctors on a plane and flying them to a postconflict situation.’ So why do we assume sending in a pile of police officers will transform policing?

Of course, good examples do exist. These include judicial training in Pakistan, and community policing programmes in Latin America. The more successful capacity building efforts have involved ongoing sustained training focused on a specific problem and built in ongoing support mechanisms to sustain the changes sought – rather than providing a one-off training and hoping newly acquired knowledge will lead to behavior change. There is also evidence of internationally sponsored SSR efforts helping to ensure stability in host countries at least in the short term (there aren’t longitudinal studies looking at longer term effects). However, this stability is often attributed to the presence of international forces and the sheer sums of money being injected into a country, rather than to the effectiveness of SSR programmes per se.

Progress requires change

The literature is more consistent on what ways of working tend to lead to more effective programming – and there are no great surprises here.

Programmes were seen to be more effective where they were politically aware. That is, where they recognized their work as being about fundamentally changing power relations, where they were tailored to the political realities of the context and where genuine local political support (not just lip service) for reforms was achieved.

Better results were attributed to interventions that adapted to context, needs and local capacities. This meant training on things that were relevant, like basic investigative capacities as opposed to forensic capacities in countries without forensic labs. Or making sure your programme isn’t delivering the fifth internationally-provided human rights or gender training in as many months when other critical skills remain overlooked. Critical here is ensuring that those undertaking capacity building have appropriate skills in terms of technical, country and language knowledge, as well as knowledge of training methodologies – of how to actually impart learning. Just because you’re a good lawyer doesn’t necessarily mean you know how to train others to be good lawyers.

And finally, donor approaches that were flexible, engaged for the long-term, devolved decision-making to those on the ground and worked on specific security challenges were argued to be more effective. These were able to adapt and be responsive to the dynamic environments in which SSR is often undertaken – where governments change rapidly, violence breaks out, and police transfers are constant.

What next?

Despite the reams of literature on SSR, it is astonishing how little addresses questions of effectiveness of different forms of support. Given the years of work and size of investments in SSR – and its importance for building safer environments for citizens – we need to be able to do better than that. Programmes that seek to improve security need to build in learning mechanisms that help answer these questions – getting beyond the pressures for purely positive reporting and actually helping to build an evidence base for future programming about what works in different contexts, and why. There’s not a lot of point spending money on capacity building if we know little about its effects.

Yet, while it is logical to call for further research and monitoring of practice to help fill knowledge gaps, this is only useful if such knowledge leads to improved SSR practice. Many of the factors highlighted as enabling or hindering progress are far from new. Considering the political economy of SSR providers – including their constraints, in part because of their own domestic interests – is thus also critical if we are to move towards more evidence-based and effective SSR.


Lisa Denney is a Research Associate with the Overseas Development Institute. She has worked for the last ten years as a researcher on issues of security, justice, governance and development in fragile and low income countries in sub-Saharan Africa and Southeast Asia. Lisa completed her PhD at Aberystwyth University and is the author of Justice and Security Reform: Development Agencies and Informal Institutions in Sierra Leone, published with Routledge in 2014.

Craig Valters is a researcher with the Overseas Development Institute. He has a specific interest in security and justice sector reform, community dispute resolution, and the role of both elites and local level ‘informal’ actors in shaping post-conflict transitions. His work is also concerned with learning and accountability in aid organizations.