Common Approaches to Understanding and Combatting Corruption

-the first in our corruption, criminal justice and legitimacy mini-series; part of the series on corruption in fragile states

Corruption is a complex, dynamic, and often contradictory phenomenon. It is hidden yet widely acknowledged, harmful yet at times beneficial, where a victim today can be a perpetrator or beneficiary tomorrow. It distorts the rule of law, keeps essential basic services out of reach of the poor, exacerbates grievances, and buttresses illicit power structures within the state apparatus, all the more so in fragile and conflict-affected states. The international community has made combating corruption a global priority in Sustainable Development Goal (SDG) 16; a key element of promoting just, peaceful and inclusive societies – is to substantially reduce corruption and bribery in all its forms.

Despite the significant investment made by international donors to combat corruption through development assistance, there is sparse evidence that their current strategies are effective. Evaluations of this assistance is not consistently of high quality, making it difficult to draw conclusions with confidence.  Even where evaluations and research are sufficiently robust, the evidence of effectiveness is mixed at best. In fact, there is strong evidence of the ineffectiveness of some very popular interventions—such as creating anti-corruption authorities, civil service reform, and aid conditionality. The emergence of research showing anti-corruption efforts actually increasing corruption is of even greater concern.

As the first in a short series of posts, we will share lessons from the Corruption, Justice, and Legitimacy (CJL) project about current approaches to understanding corruption, how international donors combat corruption, and weaknesses in their operative logic.

The Corruption, Justice, and Legitimacy (CJL) project advances innovate approaches to corruption analysis in fragile and conflict-affected contexts and supports holistic efforts to diminish corruption in core state activities related to human security. It is a Carnegie Corporation funded research project, housed within the Institute for Human Security at The Fletcher School of Law & Diplomacy, Tufts University.

What are the current approaches to understanding corruption?

Two theoretical models dominate the anti-corruption field’s thinking about how corruption happens and how it can be addressed: principal-agent and collective action. 


In this model corruption is seen as a breakdown of the principal-agent relationship. The “principal” delegates an “agent” with tasks and the responsibility of ensuring others comply with regulations. This agent is empowered to make decisions on the principal’s behalf. This is the basis on which all governments function.

Problems arise when the principal’s interest differ from the agents’.   As agents have more information than the principal about their tasks and transactions, they can choose not to disclose key information to the principal. Without this information, the principal has limited ability to control what the agents do.

According to this model, corruption arises when agents have the opportunity to make use of asymmetric information, and pursue their private interests at the expense of the principal’s interests (and, by extension, the public good).  This assumes however that the principal’s interest is in the public good.  In situations where the principal’s interest is for personal benefit, then the model falls apart.

Collective Action

This model is based on the idea that non-corrupt behavior is in the group’s collective interest.  However, in fragile and conflict-affected states, people often participate in corruption because they believe others do and that if they act in a non-corrupt manner they would lose out. In such societies, termed “particularistic” societies, corruption is not an exception to a norm of integrity and equal treatment, but is the norm itself.

How are international donors combatting corruption?

In a review of anti-corruption programming we found that the majority of programs that combat corruption with foreign aid are based, implicitly or explicitly, on the “principal-agent” theory.   Despite the common refrain found in the development community that context matters, this (often implicit) theory of corruption is used across, and regardless of differences in, contexts, stakeholders, and types of corruption.

In our review of corruption program designs, we identified two broad approaches that focus on different parts of the “chain” linking principals and agents:

Bureaucratic Principal-Agent Programs

These programs seek to change individual transactions within governments, where elected officials and senior bureaucrats are the principals and civil servants the agents. All of these programs seek to reinforce a (potentially false) model of the principal-agent relationship, which assumes that the public interest lies at the heart of every principal’s actions.

Examples of these programs are: civil servant capacity-building programs that seek to develop better disciplinary procedures, merit-based promotion systems, or secondment of international mentors to departments.

“Citizen” Principal-Agent Programs

These programs were often inspired implicitly or explicitly by “collective action” theories, but in practice are a variation on a principal-agent approach. In these cases, the “principal” is not an elected official or bureaucrat, but citizens or representative NGOs attempting to hold authorities accountable. Elected officials essentially become agents who are meant to take decisions on behalf of ‘the people’.

These programs typically include: advocacy campaigns demanding action from their elected officials, education and public awareness raising, media campaigns, non-violent resistance or electoral processes.

These classic principal-agent responses are argued to be ineffective, because they hinge on the assumption that there is political will to enact change.  For instance, in order for increased accountability to reduce corrupt behavior, enforcing the rules in a fair manner must be in the principal’s interest.

Click here to read the next post in the series, where we unpack the mismatch between corruption and anti-corruption programming.

Cheyanne Scharbatke-Church

Cheyanne Scharbatke-Church is Principal at Besa: Catalyzing Strategic Change, a social enterprise committed to catalysing significant change on strategic issues in places experiencing conflict and structural or overt physical violence. She has significant experience working on anti-corruption and state legitimacy in the DRC and Uganda.  As a Professor of Practice she teaches and consults on program design, monitoring, evaluation and learning.

15 thoughts on “Common Approaches to Understanding and Combatting Corruption

  1. Trond K. Botnen

    Thank you for an interesting blog post. I would just like to challenge you to defend and elaborate on the assumption that citizen principal-agent programmes depend on the political will to enact change.

    As I understand these collective action programmes, they do not depend on this. Quite on the contrary, they often seek to create political will where none previously existed, by changing the incentive structure of politicians and other decision-makers, basically by raising the cost of defending the corrupt status quo (and sometime increasing the benefits of fighting corruption, if collective action can credibly link this to e.g. the probability of getting re-elected).

    I think this assumption also simplifies the political game, as collective action may not only change the incentive structures for individuals, but also impact on coalition and systems at the political level, as politicans seeking to maximise votes may withdraw their support for leaders singled out by citizen’s campaigns as being corrupt (Brazil and Guatemala are interesting cases that may illustrate this, although they are of course also very complicated).

    I would contend that successful collective actions against corruption are exactly those that do not assume the existence of political will, but rather seek seek to create/strengthen this politicall will.

    Any thoughts on this?

  2. Sophia Dawkins

    Thank you for this important point. Please allow me to join the debate by sharing some observations.

    Yes, collective action can build political will, and we should not discount significant achievements in this realm. Consider, for example, the role collective action played in ending Apartheid, or how in the Second Liberian Civil War the “Women of Liberia Mass Action for Peace” put pressure on the belligerents to sign a peace agreement in 2003.

    However, systems of corruption are special – perverse, pernicious and particularly resistant to change. Where political will is truly absent, collective action can have contradictory effects. It can force the “unprincipled principal” and his/her agents to establish a veneer of good behaviour while sustaining corruption in more complex and repressive ways, out of clear sight. One view is that this is what has happened as civil society condemnation of corruption has gained momentum in places like Kenya – rather than extinguish corruption, exposure has altered the way it is concealed and denied.

    The hope in these bleak situations lies in the theory of “acculturation”. This holds that phony, surface-level commitments to a principle – such as “corruption is bad” – progressively generates domestic and international pressure on principals and their agents to behave in conformity with that principle, even if they don’t really mean to. The problem is that we don’t yet have good evidence on whether acculturation alters corrupt behaviour. Perhaps this should be next for the research agenda.

    1. Kiely Barnard-Webster

      Very thought-provoking observations from both. Taking a step back, though, I find myself dwelling on how we define “success”. I agree that some collective action initiatives operate under the assumption that political will must be created/strengthened. To this effect, research does suggest that effective collective action has the potential to create incentives at the socio-political level (Guatemala as a possible example), leading to positive changes in the system of corruption.

      However, when collective action is able to create political will, how then does this start to “add up” to a larger systemic change (so, that, for example, we don’t see a re-lapse of corrupt behavior like we did one year after the Orange Revolution in Ukraine)? Wouldn’t it prove true that when small pockets of political will do exist, collective action might be all the more able to change a complex system of corruption and then sustain it? To me, successful collective actions are those that have determined their strategy based on which approach is most likely to “add up” to a sustainable change. Something else to add to the agenda, perhaps.

  3. Hank Nelson

    Thank you for this interesting post and thought-provoking conversation. To add to your argument, Sophia, I would argue that programs that embrace a citizen principal-agent model of anti-corruption could have additional deleterious effects when there is little to no political will for reform. Programs in these contexts might identify and raise awareness of the seriousness of the corruption problem, while failing to engender any real political change to address it. In these cases, civil society and ordinary citizens might feel increasingly helpless to address corruption, increasing malaise and a begrudging acceptance of the status quo.

  4. Trond K. Botnen

    Thank you, Sophia, Kiely and Hank, for interesting responses. In response to Sophia, I would like to ask the question if paying lip-service to good govenance while continuing covert corrupt behaviour is really. a contradictory effect? Or is it a step in the right direction? We could ask the same about developments in human rights. Man parties that signed the Universal Declaration (and later various conventions) have paid lip-service to HR while continuing to violate them. But would it have been better if they had done so even more openly and without even paying lip-service? I don’t think so, because it is important that we have these standards. It might be the same with corruption. Paying lip-service to good governance, may, over time, elevate good governance standards and make it easier to hold corrupters accountable.

    As for systemic change, I think it is right that isolated collective action may not be able to change entrenches and complex systems of corruption. I think, in order to do so, one must build a “counter-system” of accountabiliy, checks and balance, both top-down and bottom-up. Such a system, consisting of legislation, institutions, mechanisms, CSOs and active citizens is what it would take to counter the systems of corruption. An interesting area of research would be to look at this in countries that have succeeded to see how a “system of good governance” can hold the “system of corruption” in check. Maybe there is research I am not aware of, but my point is that the systems approach would also be useful in looking at how we combat corruption. (and the collective action / citizen engagment mus be an integral part of such a system).

    Finally, top Hank, I do not believe that a heightened awareness is deleterious in any way. It is a necessary first step toward change. And it is not as if citizens today, in almost any country, are not aware of corruption as a problem. The lack of awareness is in the area of how to fight. it – together.

  5. John Fairfield

    Cheyanne, thank you. Yet another perspective is afforded by asking about the nature of government. If it is funded by the population (whether by taxation with representation, or by “fees” for services rendered) then the population has some control over it. An overwhelming pressure for corruption arises when the state is funded mainly independently from its population–the governments of many less developed countries are mostly funded by selling extraction rights to international corporations. The government exists to guarantee “stability” for the extraction process, i.e. to police its population, which can’t even vote with its pocketbook. Internal corruption is preferable to such externally funded corruption which delegitimizes government.

    1. Cheyanne Scharbatke-Church

      Hi John
      John, I think you bring up a really important point about the fiscal basis of a government which in turn creates incentives other than good governance as per a Western notion of the social contract. I want to think about that more in terms of how to incorporate that into a systems analysis.

      I am not sure I am on the same page with you about citizenry ability to vote with their pocket-books. In a state that did collect taxes but also was plagued by grand corruption, what would the average citizen do – refuse to pay taxes?

      Finally can you say expand on your point about internal vs external corruption and delegitimizing the government – I’m not sure I am following your argument.

      1. John Fairfield

        By internal versus external corruption, I was trying to make labels to distinguish corruption in governments funded internally (injustice in the levels of fees collected or of services provided) from corruption in governments largely funded externally (via contracts for extraction of minerals, woods, fuels etc). Methods for dealing with the former are very different from methods for dealing with the latter, so we need to distinguish them in our analyses.

        You’re right, it is very difficult to be a tax resister. But even absent tax resistance, a government funded by taxation needs to keep its cows contented, i.e. keep the level of public discontent from so affecting productivity that taxes collected drop. Such a government needs its citizens to be reasonably happy. Think China.

        And by the word “delegitimizing” I meant that a government that is not funded by the productivity of its citizens (via taxes or fees), but that is largely funded externally, has no interest in, nor need for, the productivity or happiness of its citizenry. Such a government needs only to keep its population in check.

        1. Cheyanne Scharbatke-Church

          Thanks for teasing out your argument more John.

          I’m particularly interested in your point on delegitimizing the government as there are a lot of beliefs in the international community about what creates or diminishes internal legitimacy (meaning that the citizenry feels a state or government is legitimate). Our current research tries to understand the relationship between corruption in the policy and courts in Northern Uganda and the legitimacy of those institutions. It is not proving as clear cut as many think e.g. systemic corruption in those institutions, that is actively disliked, yet the legitimacy of the institutions as defined by their ‘right to rule’ is still intact. We are still processing our data, so take this as tentative conclusion at this time. We will post more on this research on this site after the summer.

  6. Pingback: Making Power Analysis Useful To Anti-Corruption Programming – CDA Perspectives

  7. Pingback: What Can We Learn About Corruption in Fragile States? – CDA Perspectives

  8. Pingback: Final Blog of the Corruption, Criminal Justice and Legitimacy Mini-Series | CDA Perspectives

  9. Pingback: Why the Lid Doesn’t Fit the Pot: The Mismatch Between Corruption and Anti-corruption Programming | CDA Perspectives

  10. Pingback: What makes corruption complex? | SDTF

  11. Pingback: What Makes Corruption Complex? | CDA Perspectives

Comments are closed.