Step 2 – Review different reform approaches & experience

Add to reading list


Read the summary and the guidance to reform approaches below. Navigation is via the Contents list on the sidebar to the left. Or you can download a complete copy of this Guidance to Step 2 as a pdf document to read separately.

Summary of Step 2

We recommend you consider each of these ten reform approaches:

  1. Functional approaches: improving institutions, public financial management, systems and controls 
  2. People-centred approaches: working directly with affected communities; building networks and coalitions of supporters
  3. Monitoring approaches: strengthen oversight groups and their independence
  4. Justice & rule of law approaches: prosecuting, raising confidence, improving laws
  5. Transparency approaches: making visible what others wish to keep hidden
  6. Integrity approaches: motivating, instilling pride and commitment
  7. Whistleblowing approaches: finding safe ways for people to speak up
  8. Civil society and media: creating space for external voices
  9. Incentives and economic theory approaches: aligning stakeholders and economics
  10. Nudge approaches: new science show how small changes can make a big difference

Talking with colleagues and stakeholders about how each of them might work in your environment enables you to ‘circle around’ the problem, looking at different ways and combinations to tackle it. A reform strategy might, for example, consist of institutional improvement projects, plus strengthening integrity among staff, plus strengthened sanctions and discipline.

Reform measures will always be specific to the particular circumstances….context is all important. For example, the initiative called I-paid-a-bribe had a significant beneficial impact in India, but it made no headway at all in China. We provide some ways later in Step 3 to think about whether your context is likely to assist or hinder any particular reform.


The originating author of ‘Reform approaches and experience’ is Mark Pyman, managing editor of CurbingCorrruption. Additional contributions have been made by Tehmina Abbas, Scott Guggenheim, Paul Heywood, Tom Shipley, Agata Slota (who contributed the detail of the tenth approach ‘Nudge Approaches’).

1. Functional approaches

Improving institutions, processes, public financial management, systems and controls

Most institutions, processes and systems have numerous weaknesses,  relating usually to poor efficiency and effectiveness. Sometimes these processes may have actually been designed to enable corruption. But, more commonly, they are simply weak processes, or overly complex processes, that are corruptly exploited.  This is the reform area that tends to get the most attention, fixing the technical problems. Improving the professional processes, for example, such as teacher training in the education sector, treatment protocols in the health sector, patients having to pay for minor treatment that should be free, or evidence tampering rules in policing, is a vital part of institutional improvement.

This makes sense, especially when it involves systematising something that previously involved many human-to-human interactions. However, an exclusive attention to technical measures only is a high-risk strategy – technical solutions often fail, they can be slow and are sometimes deliberately slowed down, and sometimes they are used as an impressive-sounding excuse for what the leadership knows will be little real change. And, at the analysis stage, you may also find that reform areas which seem easy often involve complex social interactions and require multiple stakeholders to make changes: a narrow scope is usually better than a broad scope.

So, do not be too carried away by expectations that functional reforms alone will solve your corruption problems. Often the reforms will encounter the same problems – such as complex social habits and relationships of long-standing – that the previous inefficient processes also fell prey to.  Do use functional reforms, but use them in combination with other approaches, and with sufficient prior analysis.

Functional approaches: Seven measures you can consider

2. People-centred approaches

Working directly with affected communities; building networks and coalitions of supporters

Never doubt that a small group of thoughtful, committed citizens can change the world; indeed, it’s the only thing that ever has. The truth of this famous quote, by the American anthropologist Margaret Mead, is confirmed in the anti-corruption sphere by recent research.

The most recent ANTICORRP research investigating success against corruption identifies that all the successful country cases involve ‘human agency’ to a major degree. They were all effected by committed groups of people, sometimes public officials, sometimes politicians, sometimes civil society actors, often a combination of all three. Furthermore, the country successes were all achieved by the country’s own human agency efforts rather than by the efforts of outsiders (Mungiu-Pippidi and Johnston 2017 Transitions to good governance),in countries as diverse as Botswana, Georgia, South Korea and Taiwan.

Another analysis, on how good governance evolved in Oman, identified the importance of a well-educated group of Omani citizens, who had been recently violently expelled from the island of Zanzibar. These citizens had skills that allowed them to administer core technical functions within a nascent state bureaucracy. This group was not at the centre of political or economic power hierarchies, but was dominant at the middle and upper-middle level positions of the bureaucracy (Hunt and Phillips 2017).

So, one of the most important steps you can take is to supplement your own commitment by aligning with – or helping to develop – other committed people and groups. This might be as simple as building up a team of committed people around you in your area of responsibility. Or it could be on a larger scale, collaborating with groups in the private sector or professional associations, such as doctors or engineers. Or it could be working with civil society or in collaboration with relevant international initiatives.

People-centred approaches: Eight measures you can consider

3. Monitoring approaches

Strengthening oversight groups and their independence

Monitoring and scrutiny are central elements in curbing corruption. Monitoring by official, permanent parts of the government system – such as internal auditors, external auditors and regulatory agencies – may not be in place or, more commonly, are in place but are ineffective, for diverse reasons such as lack of budget, institutional neglect or deliberate marginalisation. Much faith is therefore placed in external bodies, such as independent agencies and NGOs, but these bodies too often perform poorly, or are subverted in the political environment through being under-resourced, or denied access to key people and records, bribed or threatened, and otherwise marginalised This is an area with much scope for experimentation, such as trying out hybrid organisations of government people plus outsiders, or unusual mandates, or new forms of citizen oversight/participation through the opportunities afforded by mobile phones and social media.

Finding ways to get multiple forms of independent scrutiny into action is therefore a core part of  anti-corruption strategies.

Monitoring approaches: Ten measures you can consider

4. Justice and Rule of Law (ROL)

Prosecuting, raising confidence, improving laws, using civil discipline options

Routine prosecution of corruption cases is one of the hallmarks of a state that has control of corruption, and  some experts describe routine prosecution as the definition of success against corruption: ‘Reducing corruption to the status of exception in a sustainable way thus defines a successful evolution’ (Mungiu-Pippidi and Johnston 2017 p9).

There is some evidence that prosecutions for corruption against top politicians and business people is becoming more common. For example, there seems to be a surge in corruption prosecutions of former presidents in Latin America. As  Manuel Balan wrote in a blog in September 2018: In the last year,  sitting or former presidents have been prosecuted for corruption in Brazil, Guatemala, El Salvador, Honduras, Colombia, Costa Rica, Ecuador, and Panama. In Peru, Pedro Pablo Kuczynski resigned from the presidency amid corruption probes, and the last three former presidents are either facing trial or serving time for corruption. Argentina may soon join this list as a result of the so-called “Notebook Scandal,” which has triggered a fast-moving investigation that has already snared 11 businessmen and one public official, and is getting closer to former President, Cristina Fernández de Kirchner. (Argentina’s former vice-president Amado Boudou was also sentenced to almost six years in prison for corruption in a separate case.)

However, prosecutions are high risk: they can be very slow, often many years, to come to court; they can be unpredictable, as powerful individuals find ways to escape prosecution or conviction. Worse, in countries where the judiciary are corrupt, this tends to mean that corruption reform will take decades (World Bank Development Report 2011, p109). Perhaps only Sicily, in Italy, has had success against corruption by a prosecution-led strategy.

Whilst prosecution may be a tactical response to public pressure, it is unlikely to be a major part of the strategy.

Justice and Rule of Law approaches: Eight measures you can consider


5. Transparency approaches

Making visible what others wish to keep hidden

Along with independent review and monitoring, transparency is one of the important tools in reducing corruption. Corruption problems naturally thrive when the relevant data is not going to be made public.  This reform approach has now advanced to the status of there being an international collaboration, the Open Government Partnership (OGP). OGP was launched in 2011 to provide an international platform for domestic reformers committed to making their governments more open, accountable, and responsive to citizens. It now has about 100 countries as members. As the table opposite shows, some of them had made significant progress between 2012 and 2015 (as measured by the Open Budget Index), whilst the rest have made little progress so far.

However, though transparency is a valuable tool, it also has its limits and counterproductive implications. It has been easy for many countries and sectors to seem to comply, putting a lot of data online, whilst in reality continuing to obstruct. So much so that some people now refer to ‘transparency theatre’. The OGP initiative points out how much of the information that is put into the public domain by governments and official bodies is not actually usable because of the way it is presented, formatted etc.  So, there is more to this than just transparency, or open data, alone. (If you want to read a detailed analysis of the limitations of transparency, see Alloa and Thoma (2018)).

Transparency approaches: Five measures you can consider

6. Integrity approaches

Instilling pride is the most motivating way to curb corruption

‘Traditional approaches based on the creation of more rules, stricter compliance and tougher enforcement have been of limited effectiveness. A strategic and sustainable response to corruption is public integrity. Integrity is one of the key pillars of political, economic and social structures and thus essential to the economic and social well-being and prosperity of individuals and societies as a whole’   (OECD 2018 Behavioural insights for public integrity: harnessing the human factor to counter corruption).

Integrity is a magical word. Coming from Latin (integritas) and then French (integrité) in the Middle Ages, its meaning combines good moral character of the individual and wholeness of a larger entity in one single word. The Oxford Dictionary definition gives this in full. Strengthening integrity is a positive emotion and one that strikes a strong chord of pride in people.

However, note that ‘integrity’ does not always translate well. For example, there is no equivalent word for integrity in Slavic languages. Some countries and sectors naturally focus on ‘building integrity’ as a base for building a more trusting and less corrupt society. Rwanda is one such country (See Heywood et al 2017), and the Military is one such sector. For examples, see the Defence Integrity Framework of the Canadian military approach, or a critical analysis of the decline of US army integrity in Wong and Gerras 2015, or NATO’s initiative Building Integrity and the NATO publication ‘Building integrity and reducing corruption in defence’, available in multiple languages.

Integrity approaches in government are seen to have developed from the Scandinavian civil service and the UK Civil Service in the 19thcentury, as they moved towards meritocracy-based recruitment and developed norms of professional service. See for example Mungiu-Pippidi (2015) or Johnston (2008). Since then, most national public servants receive training and guidance in good ethical behaviour, and Colleges of Government teach the values.

In similar vein, companies have developed their management systems to control corruption and comply with anti-corruption laws, many of them have simultaneously embraced a values approach with their workforce. They too recognise that compliance-only approaches are limited and insufficient. There is plenty of guidance on developing programmes on both compliance and ethics approaches in companies, and how to develop the two in tandem. More generally than any of the above, the norms of most societies and of all religions include acting with integrity; religious leaders and their institutions, of whatever faith, exert the principal force in promoting and maintaining good personal conduct in many countries.

You can find informative articles discussing integrity as a mechanism for tackling corruption. See, for example, Heywood et al (2017), Pyman and Wegener (2010), or Lamboo et al (2016).  Below is OECD’s model of the three ‘pillars’ of a national integrity system – accountability, culture and system.

More on OECD's approach

Integrity approaches: Fifteen measures you can consider

7. Whistleblowing approaches

Finding safe ways for people to speak up is vital: it may be worth setting up a custom mechanism

Whistleblowing is now recognised as an important mechanism to support integrity in organisations. However, in reality these mechanisms are usually weak, or may exist on paper only, and the whistleblowers usually end up suffering.

Yet they are vital in identifying and calling out corruption: the challenge is to find a way in your context to make them effective, and to publicise this.

Interestingly, the oldest whistleblower defence organisation is based in the USA. The Government Accountability Project has advised thousands of whistleblowers in every sector and industry since it was established in 1978 after the release of the Pentagon Papers.  They provide tools and resources for whistleblowers and the wider public.  Their website is hereand or contact them at There is also an NGO called the National Whistleblowers Centre’ who can give you information on which organisations in your country work with whistleblowing. You can find their website hereor contact them at

The good news:In a few countries, the whistleblowing law is seen to work well

The usual reality: For most people in most sectors, the reality is that the government-wide system is poor and/or mistrusted and is anyway outside of their control. So, what can you do that will be helpful to your situation?

Here are four things you can do:

7.1 Examine if there is a part of your organisation/directorate/ministry is already doing something good, and build on it

7.2 Consider contracting a civil society whistleblowing group to operate an independent sector-specific whistleblowing system

7.3 Call the NGOs PROTECT and Whistleblowers International Network (WIN) to ask for advice

7.4 Use the UNODC best practice guide to improve your current system


8. Civil society and media approaches

Creating space for external voices

Both civil society and media are seen to be important to reducing corruption. In the case of civil society, this has been the orthodox view since it was proclaimed by the World Bank back in 1997.  However, civil society projects often have no discernible impact. As recorded by Mungiu-Pippidi in her 2015 book, p174, out of some 471 civil society projects in Eastern Europe in the period 2000-2010, only about a third of these projects had any traceable impact. These projects were the ones that targeted corruption directly and concretely, they were done in collaboration with journalists, and they were set up by grass roots organisations rather than by donors.

A similar analysis by the Partnership for Transparency Fund (PTF) of 200 projects that they sponsored came to similar conclusions. Successful projects needed to i) Focus on a single set of problems and not be ambitious; ii) Do not start until you well understand the problem; iii) be non-confrontational and patient; iv) work with key champions of reform within the power structure; and v) seek to reform the official accountability system with the evidence you have gained from the project (Landell Mills (2013) p230-231).

Media are clearly important, as the Eastern Europe example quoted above shows. In the case of investigative journalism, their role and value are self-evident. The problem in recent years has been twofold:  the increasing antagonism towards them in many countries, and the declining appetite or ability for news corporations to employ them.

The lesson is to bring civil society and media on board as collaborators, NOT as after-thoughts or add-ons. You can consider using a civil society organisation to monitor your overall progress. You could create a policy advisory body, or a progress discussion body into which you could bring civil society and media.

9. Incentives & economic theory approaches

Aligning stakeholders and economics

Economists have long argued that successful implementation of any policy requires that the preferences of all those involved be appropriately aligned with achieving the goals of that scheme. In tackling corruption, for example, experts consider the criminal law measures that can act as incentives not to be corrupt, such as legalisation of payments, reform of public programs or procurement systems or privatisation. See Rose-Ackerman and Palifka (2016), p126.

9.1 Incentive approaches. However, in its application to tackling corruption, the incentive approach has had a mixed reception, as it is often unclear who is the agent and who is the principal. Furthermore, monitoring agent behaviour and holding agents accountable is particularly difficult in the public sector. This has led to other sorts of approaches being preferred, such as ‘collective action’ by all the involved stakeholders. Nonetheless, once we are down at sector level – as opposed to national-level measures like criminal law reform – it becomes easier to see how specific incentives can help to re-align policies so as to promote less corruption.

There is also some analysis that is helpful in distinguishing between sectors where it may be easier to improve performance through incentives, and sectors where it will always be difficult. In an analysis by Fukayama 2004 he suggests looking at public sectors on a matrix that plots ‘transaction intensity’ against ‘specificity’. Transaction intensity is a simple concept – it is simply the number of decisions that need to be made by organisations. Thus, for example, a central bank makes very few decisions, such as making changes to interest rates, whilst a health system makes millions of decisions every day. Specificity refers to the ability to monitor a service output. For example, jet engine maintenance is a highly specific service output: it is very hard to fake. By contrast, early years school education classes are very easy to fake and are hard to measure reliably. If you set up this matrix of Specificity versus Transaction intensity for public sector services, Fukuyama suggest that it looks like the chart opposite.

One current example of the application of incentives to reduce corruption in particular sectors is from the work of Khan et al 2016, 2017 at the School or Oriental and African Studies (SOAS) at the University of London. They have examined in detail mechanisms for reducing corruption in the Bangladesh energy sector, the Bangladesh garments sector, and the Nigerian power sector. Another example is from corruption reduction efforts in the forestry sector, for example in Ecuador, where they combined a new government decree with the ability to monitor logging activities on the basis of verifiable indicators. See Kishor and Damania (2004) Improving governance and reducing corruption in the Forestry Sector. In these cases, the objective of the incentives is not to ‘reduce corruption’, but to make the sector more productive: This is in line with guidance elsewhere in this website that your objective is rarely to ‘reduce corruption’, but more likely to make the sector more productive (See here).

9.2 Other economic approaches. There is a raft of economic measures that have a place in reducing corruption. Examples include more competition, more consumer choice, better information, fewer layers, firm incentives, better banking rules, the role for technology, spillover and demonstration effects, and so on. These measures can all have significant impact on corruption, but there is little written up about their specific anti-corruption impact.

If you have examples and reports, do let us know. This is an area where more knowledge of the anti-corruption impact of economic reforms is needed.

10. Nudge approaches

New science shows how small changes can make a big difference

Policy makers are increasingly using ‘nudges’ as a tool for changing behaviour. Instead of controlling how people behave by taking away their freedom to choose, they are changing the decision-making context in a way that leads people to make different choices, thereby ‘nudging’ them towards a desired behaviour.

A ‘nudge’ is a means of encouraging or guiding behaviour, but without mandating or instructing, and ideally without the need for heavy financial incentives or sanctions. We know what it means in everyday life: it’s a gentle hint; a suggestion; a conspicuous glance at a heap of clothes that we’re hoping our kids might clear away. Research in social and cognitive psychology has led to new thinking on how people make decisions. We are not ‘rational choice beings’, as the economists would have it. Rather, we are influenced by our context and our emotions, and our decision-making  ability is prone to errors in remembering, predicting and deciding between options. Further, the research has shown that these errors are often not random, but predictable.

This in turn has led to a rapidly growing new discipline, called ‘behavioural science’. Amongst the insights are ways in which small changes in people’s decision-making context can have large impacts on behaviour.

In the UK, for example, in order to increase savings for retirement, the government has instituted a policy where employers automatically enroll their employees in pension schemes. Employees can opt out of the scheme if they want to – that is, they are under no obligation to remain enrolled. However, simply by making having a pension scheme the ‘default’ option, employers have increased uptake of pension schemes significantly.

Nudging is appealing for a number of reasons. First, nudges are often relatively cheap as compared to other policy tools. Changing the default option, for example, may simply entail changing one sentence and one tick box on a form.

Relatedly, nudges can be easy to implement. Implementing them often does not require legislative changes or multiple approvals. In contexts where the bureaucratic processes and inefficient governance slow down change, this can be particularly beneficial. 

Third, nudges may work in contexts where increasing the level of control over people’s behaviour can actually backfire. Higher levels of regulation and monitoring do not necessarily lead to reduced corruption. In fact, restricting people further may signal to them that they are not trustworthy or to make them feel resentful, driving them to behave more unethically. Control and punishment may also reduce people’s internal motivation to act ethically.

Finally, the design of nudges is based on an understanding of human psychology. As a result, nudges may be more effective than policies that forget to take the ‘human’ element into consideration. Nudge theory recognises that people are not always rational and that they won’t necessarily act logically, and that information or awareness alone won’t necessarily change behaviour. People do not always carefully weigh up the costs and benefits of acting unethically, factoring in the risk of getting caught and the potential punishment.

The theory also accepts that multiple drivers – the physical context, individual psychology, and those around us, to name a few – impact on behaviour. For example, there are many reasons why people may be dishonest or corrupt, such as feeling social pressure to conform, losing self-control due to exhaustion, or a sense that they deserve the spoils of a corrupt behaviour because they are being treated unfairly. Even well-meaning individuals who are informed about corruption and its consequences may fail to make the right ethical choices sometimes. A nudge-based approach acknowledges these aspects of human psychology and responds to them.

Below are examples of nudges that can be used to reduce corruption – either on their own or as applied alongside other more traditional methods. While based on behavioural science research, nudges by their very nature are highly context specific and should be tested when used in new environments and situations.

If you want to read more generally about this whole new topic, we recommend the original book on the theory by Thaler and Sunstein (2008) and a book about its application in political practice in the UK by Halpern (2015). The application to tackling corruption is well covered in the new publication from the OECD (2018).

Nudge approaches: Ten measures you can consider


11. Bibliography

Full bibliography of sources


[i] In academic writing and research, you will find that this effect of peoples’ actions is rather clumsily called ‘Human agency’. It is used by sociologists to describe the capacity of a human being to act and to effect change personally, arising out of their individual behaviour. It is in contrast to change happening due to social ‘structures’, by which sociologists mean the circumstances that the individual finds themselves in, such as social class, religion, gender, ethnicity, subculture, etc., which seem to limit or influence the opportunities that individuals have.

Contribute your experience. Connect with us on Twitter

Connect with us on LinkedIn

Send the Editor an email