The Unhelpful Nature of Anti-Corruption Research; As seen by people trying to develop solutions

January 24, 2017 | Mark Pyman

Share this article

Dr. Mark Pyman, who is deeply involved in efforts to reduce corruption in Afghanistan, pinpoints what frustrates him about anti-corruption research, in this latest post from the corruption in fragile states blog series. Few sector-specific approaches or insightful typologies of the multiple sorts of corruption within each sector, few analyses of sub-national variations, and an anti-corruption research community that is slow to move beyond diagnosis into treatment and politics. Continue reading to find some helpful treasures Pyman has found, and let us know if you share Pyman’s frustration, or if you have any great resources to add to the forum.

A colleague asked me some years ago “Why is it that people working and researching in corruption seem only to enjoy showing how bad it all is? They seem to like nothing more than to ‘admire the problem’.” Sadly, this is also my experience, and I’d like to vent my frustration here: because this is deeply unhelpful to everyone who is trying to address the problem in each of their countries. Surely, researchers know more than just how to explain the problem?

Needed: Sector-Specific tools and approaches to Help Shape Reforms

I am one of three international Committee members in the ‘Afghanistan Joint Independent Anti-Corruption Monitoring and Evaluation Committee’ (Corruption people like snappy titles, too). This Committee, known by its acronym as ‘MEC’,  was set up in 2011 by the Government of President Karzai and the international donors, to act as a counter-weight to the official anti-corruption agency, which was well known to be highly corrupt. We are six Commissioners – three international, three Afghan – plus a secretariat in Kabul of some 25 professional staff led by an Executive Director. We watch over what the government does, or does not do; we review the performance of the Ministries and Agencies in relation to corruption; we do analyses to identify the principal corruption vulnerabilities and how they might be addressed; and we speak out publicly in areas where there needs to be more progress.

So, I am intensely interested in what research there is that can tell us about addressing corruption in conflict and fragile environments. There must be so much more that we can do in practice to implement reforms, learning from successes and failures equally. We need to know much more about tackling corruption in health, in education, in the police, in the judiciary, in mining, in commerce and industry, in revenue-generating Ministries, in the civil service, in local government, and so on. Although a country index is helpful, we need more granular detail. I need to see inside the health sector, say, and see reform results in technical areas like forensics, or the health management system; in the political areas of senior appointments, in local corrupt elite eco-systems, or in patronage for major elements like hospital construction or medical service out-sourcing. Equally, I need insights in to how civil services can be gradually – or rapidly – reformed, at a level that shows how the situation varies according to different parameters: such as between different Ministries, or between agencies, or between revenue-generating agencies and service delivery agencies, or between agencies that belong to rival political factions.

Found Treasure

I have been looking hard over the past 18 months at what could be helpful to me and my colleagues in relation to Afghanistan. There are some articles or reports that really do help me. For example, the work of Merilee Grindle on civil service reform against corruption in 5 countries in Latin America is a tremendously insightful and research-rich book. It really helps in seeing all the ways that civil service reform was contested, and the multiple
different strategies used by various groups of reformers. Similarly, the research from the University of Gothenburg by Nicholas Charron and his colleagues on differences in civil-service corruption between different Ministries within the same country, () and how these differences are much larger than the differences between countries in the region, is really helpful. (here, and here) Similarly, Francesca Recanatini at the World Bank has useful observations about using disaggregated data to identify corruption distinctions across agencies and regions within a single country. There have been a few that develop decent-looking typologies of the specific corruption types in the sector, and collate actions that have been taken by others, such as by U4 in the education sector. There is the useful evidence paper from DFID. I am very happy to take these and use them.

But such treasures are few and far between. I have recently trawled through the research in the education and health sectors, and been deeply disappointed. Hundreds of articles admiring the corruption problems from an ethnographic point of view, from a political economy perspective, and above all from a shopping list point of view “Let me describe the many ways that education is corrupted in this country… “

Lesson Learned After Tackling Corruption in the Defence Sector

Let me declare something of a bias here. For twelve years I led the Transparency International programme on tackling corruption in the defence sector. I found very early on that there was almost nothing in the research other than ‘the defence sector is rife with corruption. Defence corruption was lazily equated with defence procurement corruption. My team and I got on and developed a whole system of sector-specific tools; a typology, an index, technical guidance on how to tackle each of the problems, a technical index for the relative performance of defence ministries and militaries, a technical index for the relative strength of the anti-corruption systems of the defence companies, corruption-testing exercises for the military forces, etc, etc.  We worked with about 30 countries in some depth, with considerable success, and evaluated about 130 countries in all.

But, in my ignorance, I thought that the defence sector was lacking in sector-specific tools because it was special – it is a tricky sector, it is shrouded in secrecy, it is not a natural target for researchers or NGOs, it is highly technical; and so on. Only now do I realise that almost all sectors seem to be similarly un-researched in relation to tools and analysis that can help shape reforms.

I know there are some good things going on in response to this lack, for example the EU Anticorpp programme led by Professor Mungiu-Pippidi, and the ‘Anti-Corruption Evidence Partnership (BA/DFID ACE Partnership)’ being run by the British Academy on behalf of DFID. But these are drops in the ocean; they hardly get into individual sectors, and mostly do not cover fragile states.

From Diagnosis towards Treatment

Why isn’t the whole anti-corruption research community moving away from diagnosis towards treatment? Surely the political economy community can do more on how powerful elites and corrupt patronage networks can be countered, based on examples from the many countries where such efforts are underway? Surely we have got to the point where researchers should be trying out approaches, such as new technical indexes at the sector level, to see whether they can help diagnosis and reform? Surely we can have more of the analysis pointed up by such as Grindle and Mungiu-Pippidi, who show the impact of committed groups of individuals and how we can foster and support such groups in hostile political climates? Surely we can get beyond yet more analyses of context, into analyses of the success or failure of specific reform efforts?

Is there someone out there who can tell me I am hopelessly wrong, and that there is a library of research somewhere that I have just totally missed? Please, someone?

About this article

This post is part of the corruption in fragile states series. The series provides a space for conversation about corruption in fragile states. Since its inception in 2016 as part of the CDA Perspectives Blog, the series has sought to challenge status quo thinking with a particular emphasis on exploring systems-based approaches to understanding and acting on corruption dynamics. Topics in the series range from new research findings in Uganda, Iraq or the DRC to provocative thought pieces intended to contest dominant paradigms or practices.

Now hosted by the Institute for Human Security at the Fletcher School of Law and Diplomacy, series contributions are inspired by, but not limited to, the Corruption, Justice and Legitimacy project as well as the, now concluded, Central Africa Accountable Service Delivery Initiative. All blog posts published after March 1, 2018, information about submitting guest posts, and subscribing to future series updates is available here.

To receive blog posts on other topics from CDA subscribe here. You may contact [email protected] if you are interested in submitting a guest post on the latest work in the fields of accountability and feedback loops, conflict sensitivity, peacebuilding effectiveness, and responsible business.

Image: Treasure map with x storage auction gold 900 x 444. Uploaded by Steven Johnson

About the author(s)

Dr. Mark Pyman is an experienced advocate, scholar and practitioner at the forefront of untangling the nexus between corruption and insecurity worldwide. He is currently one of three International Committee Members on the Afghanistan independent Anti-Corruption Committee. For the previous eleven years he led the global Security and Defence Programme at the NGO Transparency International. This groundbreaking programme worked on the ways that corruption undermines countries, especially in relation to security and conflict. He has led the team’s field work in over 30 countries, including Afghanistan, Bulgaria, Burundi, Colombia, Georgia, India, Kenya, Latvia, Lebanon, Norway, Palestine, Poland, Saudi Arabia, South Korea, Taiwan, Turkey, Uganda, Ukraine, UK and the USA.

His work has been instrumental in shaping the United Nations Arms Trade Treaty (2013), in influencing NATO policy and operations in respect of counter-corruption, in shaping the military doctrine of several countries, and in policy forums such as the Munich Security Conference. He has also led the development of techniques for evaluating the corruption vulnerabilities of all the world’s defence companies. His last publication for TI in 2015 reviewed the 165 largest defence companies worldwide in an analysis that is viewed by the industry as the most authoritative analysis available.

He has authored or supervised some sixty publications. He has also supervised  detailed reports analysing the defence corruption vulnerabilities of 120 countries.


  1. Anne-Christine Wegener

    Mark, I agree. For me, there are three audiences for corruption literature: the interested public at large; the academic community; and practicioners; for the first two groups, the amount of good research is increasing, and that’s important (I would say that, having written a book targeted at the first group mainly). But as a practicioner I, too, find too few examples of success stories, let alone stories/examples of failure (I’d be just as interested in learning why things did not work out in specific cases). Like you, I work on sectoral reforms, and here the air becomes even thinner. U4 does great work, as do a few others, but it appears that we are not taking nearly enough time to share and write down what we have learned and the tools we have assembled – the good, the bad, and the ugly to use your words 🙂

    • Ivan G. Somlai

      There is a fourth audience: the corruptors who get a heads up from what has been revealed and then plan new modalities to keep circumventing appropriate behaviour. This of course cannot be avoided unless we never publish; I am noting it just for the record:)

  2. Andrew

    On implementation, take a look at continuous improvement systems, one of which is reflected in the team-based performance management approach. That one has been used relatively successfully in two countries on detect and deter parts of fiscal management reform. However, it remains under constant threat.

  3. Liz David-Barrett

    Mark, I see your point and sympathise. From the academic side I think the problem is that the results of case studies are not necessarily generalisable, at least not easily so, and hence it can be difficult to get case studies and more qualitative research published in high-reputation journals. The system rather discourages academics from spending time on this kind of research. It is therefore important that practitioners keep making the point that this is valuable!

    Having said that, I think attitudes are changing and there is much more attention to political dynamics and context in the current wave of research. This is certainly true of the ACE programme, which I’m involved with, and of the other projects you mention.

    In terms of success stories, do you know the Princeton programme on innovations for successful societies – see here: They have some really useful case studies, not all about corruption per se, but quite relevant to your concerns

    • Anne-Christine Wegener

      Dear Liz,
      I agree that case studies and project experience is very difficult to compare and generalize – that is exactly the problem we are facing as practicioners: what has worked in one circumstance / region might be a failure somewhere else. However, the optimist in me says that there must be a waynto at least categorize these examples in a more systematic way to at least attempt making them replicable. Perhaps in a similar way as we did with the defence index, ie by thinking about the relevant factors and questions that might result in different outcomes. Sounds like a big project but at some point we’ll need to think about this more seriously. There are not enough a-c resources for trial and error and reinventing the wheel in practice, and the areas of concern are just too many. (I’m feeling overwhelmed now and then, can you tell? 🙂 )
      The projects that are mentioned here might be a good start for this sort of endeavour, I will certainly have a thorough read.
      To end on a positive note: I do think that academics can bring exactly the kind of rigour and methodology to the table that is missing sometimes in the wide array of projects, and it would be such a great leap forward if academics and practicioners could come up with a joint categorization of some sort.
      All the best, AC

  4. Jennifer Widner

    Dear Mark and Anne-Christine, Our Innovations for Successful Societies program would love to hear more from you about what is useful. We share your perspective, but our work may not yet be finely grained enough for practitioner-experts. ISS will shortly publish a series on implementing UNCAC anti-corruption strategies–five stories with varying degrees of success, along with work on strengthening land registries and improving supply chain monitoring.

    You can find published, operationally detailed cases about anti-corruption agencies, introducing standard operating procedures (Indonesia), integrity pacts, judicial monitoring and judicial sector anti-corruption reform, police services, and other subjects on our website. at We would greatly value specific suggestions about how we could make these materials more useful.

    Liberia National Port cleanup

    • Anne-Christine Wegener

      Dear Jennifer, many thanks, I will have a look, sounds great.

  5. Ivan G. Somlai

    One of the insidious aspects of corruption research and response, at least in international development, is that some research groups themselves accept the corruption. By research groups I include academic, corporate, NGO, INGO, IFI and other civil society “executing agencies” partaking in investigating any corrupt activities. Most development projects, whether executed by a civil society consortium or government agency, is that the project/group makes a profit from the work. Cancellation of a significant portion–if not all–of a project because of suspected or even proven corruption affects the bottom line of the executing agencies. Hence,, known corrupt practices may not be revealed; unknown may not be sufficiently investigated; proven may not be published.

  6. Manfredo Marroquín

    Many thanks Mark for making this much needed reflection in these times when corruption is raging in many parts of the world. From my point of view the problem is not in the usefulness or not of the research, but in that to bring about real changes it is necessary that many factors, knowledge, will of change, leadership and available resources need to come together. In Guatemala we are living a unique experience with the creation of the International Commission Against Impunity in Guatemala, which has succeeded in bringing to justice numerous networks of corruption in all sectors of the State. However, institutional and systems changes still do not find enough knowledge and leadership for this story to have a happy ending.

  7. Pierre Landell-Mills

    Mark: You are right to focus on studies that lead to real results in fighting and reducing corruption. Partnership for Transparency Fund was established in 2000 to assist local civil society organisations to take initiatives that would result in reduced corruption based on the conviction that durable change depended on organised local citizen pressure opposing corruption. PTF supported over 200 projects in the period 2001-2013 many of which were remarkably successful. This story has been written up in a book entitled Citizens Against Corruption published with World Bank support in 2013. The book analyses why and how these projects produced identifiable reductions in corruption which has very important lessons for those agencies wishing to promote anti-corruption initiatives. But none of the official aid agencies seem willing to listen and learn. See

  8. Mushtaq Khan

    Mark I couldn’t agree with you more about the thrust of anti-corruption research so far. The SOAS-ACE programme supported by DFID directly addresses your questions. We take a country and sector-specific approach to anti-corruption and work out how public resources are actually being captured in particular sectors to identify feasible anti-corruption strategies that are likely to have a development impact. Feasibility is determined by the local political settlement and the possibility of using existing interests or organizing new ones to address particularly damaging corruption outcomes. We are beginning with Nigeria, Tanzania and Bangladesh with a focus on a large number of sectors important to each country (including health!) and aim to extend to other sectors and countries eventually. Here is a background paper with references to the types of solutions we are looking for

  9. Alison Fahey

    Mark, thanks for these insights. With academic co-chairs Ben Olken and Rohini Pande, I manage governance sector work at the Abdul Latif Jameel Poverty Action Lab (J-PAL), a research center at MIT with regional offices around the world. J-PAL’s 140+ affiliated researchers conduct randomized evaluations of government and NGO programs and policies, aiming to identify effective solutions–or to borrow your expression, moving beyond diagnosing the existence of a problem towards testing the impact of different possible treatments. Over the past several years, we have funded RCTs to help fill gaps in the evidence and generate useful insights for practitioners.

    For an overview of the state of rigorous research on topics in governance, I’d suggest J-PAL’s 2013 Governance Review Paper ( The paper reviews evidence on political participation, the impact of leakages, and determinants of corruption. An update is forthcoming in the next few months which will include an expanded focus on state capacity and highlight emerging lessons from evaluations completed in the last few years. You may also be interested in specific studies that address some of the issues you raise: On anti-corruption efforts in education, a study is ongoing in Afghanistan to test the impact of mobile salary payments for public school teachers (the abstract for that evaluation is fourth on this list: Studies in several countries have focused on strategies to reduce education and health provider absenteeism–see for instance research on health provider attendance in Pakistan ( As you point out, corruption among public officials involved in procurement or taxation requires different approaches; there are a few ongoing studies focused on public procurement and completed studies focusing on taxation–for instance, in Pakistan, performance pay for tax collectors increased tax revenue and expanded the tax base ( On reforms to major social safety net programs, rollout of ID cards to eligible households in Indonesia increased benefits received by those households (; in India, biometrically-authenticated smartcards linked to electronic transfer of benefits reduced leakage (; and an evaluation of IT reform to the funds flow system in India’s cash for work program reduced program expenditures without reducing wages or employment ( On regulation of industry, third party pollution audit reforms improved audit accuracy and reduced measures of pollution ( On elections, researchers tested the impact of photographic monitoring on aggregation fraud in Afghanistan (

    Many more such studies are searchable here: While what works in one context should not necessarily be replicated in another without careful tweaking, piloting, and testing, we hope the lessons from these evaluations can help inform better design of anti-corruption programs. We’d be interested to hear more about what evidence would be most useful to you in your work, and/or if there’s perhaps scope to partner on testing some of the innovative programs and policies you are helping roll out.

  10. Don Bowser

    Mark great points. Here are some points to ponder based on my experience:
    1) Finding actual solutions is a lot of hard work, takes time and have a low-rate of success. Considering most development programs are aimed at quick fixes – who wants to put that in their log-frame?
    2) Many of people who write about corruption are doing it from a non-practitioner view in some not-so-corrupt countries. Those of us in the field do a lot of doing and thinking on the issues but rarely have the leisure to write about it but this is slowly changing as the next generation of corruption fighters who are social media savvy get involved.
    3) There is a continual search for the ‘new’ silver bullet that kill all corruption and it is always onto the new thing. The reality is what Klitgaard wrote in his 1988 seminal work “Controlling Corruption’ had it right. Fry a few big fish and chip away at corruption with sectoral work. However, few have used it as a programming guide and the community has been led down a lot of false paths.
    That’s my rant from the conflict zone in eastern Ukraine.

  11. Larry K

    Mark, See if you can get your hands on the report titled “Strong Anti-Corruption Safeguards Needed as Condition for Increased International Support to Justice Sector” (Kabul: UNDP, Accountability and Transparency Project (ACT), 2009)

    It is research-based and entirely focused.on solutions.

    • Don Bowser

      So Afghan judiciary is less corrupt now? Was the ACT project a catalyst for that? Having spent my time there I am somehow in doubt.

  12. Paul M Heywood

    Hi Mark,

    Many thanks for your blog post, much of which many academic researchers would fully agree with (as indicated in some of the responses above). As the responses also show, there is actually a lot of good work going on out there that seeks to address some of the concerns you highlight – notably via the ANTICORRP programme, the DFID ACE Programme, Innovations for Successful Societies, and the J-PAL Political Economy and Governance Sector. Also, as you know, the U4 team is very focused on this space just now. There’s a pretty widespread consensus that we need to understand better what actually works in practice, and why.

    A special issue of Slavonic and East European Studies, linked to the ANTICORRP project, has just been published ( I’ve got an article in it that seeks to explain some of the reason why the research you refer to has been so focused on (mis-)diagnosis: it’s not all the fault of hapless researchers!

    • Don Bowser

      Hey Paul any chance you send me the article? My email is one the post. This is a good help to the National Agency to Prevent Corruption here in Ukraine.

  13. Merle Lefkoff

    This was an interesting discussion. Frankly, I’d like to suggest that all of you folks are simply irrelevant. The way corruption is being addressed successfully (see Romania) and will continue to be addressed successfully in the future is with the scaling up of mass resistance led by civil society groups, in the hope that a space then opens up for better leadership to emerge. It’s hard to make reform happen inside civil service since bureaucracies are structured to be resistant to social change as well as becoming incubators of crime and corruption. My suggestion if you’re frustrated, is to design your banner and hit the streets!

    • Don Bowser

      Merle the Romanian experience in combating corruption is driven by the domestic intelligence service SRI who conducts the covert ops of the anti-corruption agency the DNA. While it may be effective it circumvents the rule of law. The current mass demonstrations are about not curtailing the DNA but there are many Romanians cautious of the role of the SRI and the applicability of using it in other countries. Here in Ukraine the a-c agency the NABU looked at Romania but won’t replicate it. In short who will police a powerful intelligence agency?

  14. Pierre Landell-Mills

    Merle: I agree with you in part. The drive against corruption has to be led by organised locally based civil society groups but focused on specific areas of corruption and scaled up where and as feasible. Mass resistance on the streets will lead only to violence followed by either successful repression (viz: Russia and China) or regime change which brings a new bunch of thieves to power (viz: Egypt). The only strategy likely to be successful is the patient long term struggle which makes progress gradually but does not threaten those in power too directly and overtly. That is the experience of Partnership for Transparency Fund supported projects described in the book “Citizens Against Corruption”.

    • Trond K. Botnen

      I think the notion that mass resistance in the streets will only lead to violence or a new corrupt regime is an oversimplification and not in accordance with historical evidence. That is, these to end games are obvious possibilities, especially if mass demonstrations are not strategically connected to other tactics and methods, but they are not the only possible outcomes. For examples of successful and partially successful citizen actions against corruption I recommend the book “Curtailing Corruption: People Power for Accountability and Justice” by Shaazka Beyerle.


    While discussing some of the difficulties involved in addressing morals, the article argues that without reconsidering its conception of human nature, it will be difficult for TI to re-orient and improve its approach.


Submit a Comment

Your email address will not be published.