Assessing Integrity

Sarah Repucci, 12th IACC, Workshop report, Sustainability


Charles Sampford presented his ideas on visualizing integrity. He first gave a historical overview. In the 1980s, based on the model in Hong Kong, integrity was considered to rest in a single institution. In the 1990s, the idea was developed of an ethics infrastructure, which was elaborated by Transparency International as the National Integrity System (NIS). TI drew the NIS as a Greek temple.

Integrity in this case is linked to the definition of corruption. Corruption is the abuse of power for private gain, and integrity is its correct use. In other words, integrity is the use of power for publicly justified purposes.

Integrity is not only about a country free from corruption, but about good government. Thus, anarchy is not the answer to corruption.

The National Integrity System Assessment of Australia in 2000 emphasized the interrelatedness of the institutions that feed integrity. The Greek temple model implies a list of institutions that may not be culturally appropriate. A temple uses a single design made by a single architect at one point in time, as opposed to bring together many people over time in different ways. It also implies perfection; in fact, all of the elements may not be strong, but the relationships between them are important.

Thus, a new metaphor is a bird’s nest. A nest is made of weak reeds, but together they are strong. It does not have to be perfect in order to support the egg, which represents integrity. It also needs constant maintenance, which is also required to maintain integrity.

Marianne Camerer provided an introduction to the Global Integrity Index. Global Integrity was formed as an organisation to improve the transparency and quality of information used by decision makers. It is intended to scope the governance landscape and identify strengths and weaknesses that exists.

The index is intended to improve, enhance, and influence governance. It responds to a desire to quantify, as many people feel they can manage something better if they can measure it. The index provides traceable standards, so that stakeholders can see whether they agree or not. For example, Zimbabwe has shown a lack of political will, and therefore it scores badly. Context is provided, as institutions evolve within a context. The index is not intended to shame countries. It is produced in the countries under consideration

The approach is comprehensible and actionable. It provides a road map for change. The World Bank, for example, uses it for assessments and decision making. It provides some funds for the index. The Millennium Challenge Corporation has found it useful as a briefing tool for dialogue among stakeholders.

The index is not intended for aid conditionality. It provides both a carrot and a stick.

Rosa Ines Ospina presented on the Transparency Index in Colombia. This index was developed in response to the Corruption Perceptions Index, which was helping Colombia move forward but did not show changes over time or enough subtly about what was happening. It was originally called the Integrity Index, but government complaints stemming from the definition of integrity led to a name change for simplicity. It does, however, measure integrity rather than transparency.

The index is a practical, pragmatic tool. The resources were not available to do everything, so it had to be limited. The index looks at the national, state and municipal levels. Higher integrity is considered an indication of a lower risk of corruption.

Integrity is examined in three areas: visibility, sanctions for wrongdoing, and institutionality. The number of sanctions was the only hard data available, but this can be difficult to interpret—do more sanctions indicate more or less corruption?

The index has been useful for pushing institutions to prevent corruption. By the third year of the index, the government was very interested in the results. TI Colombia has worked with public institutions to help them improve. Civil society, however, has been reluctant to accept the index’s results because they are based on official data.

Daniel Kaufmann gave an overview of measurement. He began by saying that a demand for periodic, internationally comparable measures is incompatible with an in-depth expert analysis in a country—you can have one or the other.

Governance used to be considered unmeasurable. Now, it is examined but can only be approximated through proxies. Thus, there are weaknesses in every tool as no tool can do everything. We need to be aware of these weaknesses.

Evaluations can be done by experts, business representatives, or citizens. However, measurement of a broad notion like governance benefits enormously from aggregation. Increasing from one to four sources cuts the margin of error on average by half. It is best to aggregate responses from all three groups. More sources is always better, but you must be mindful of the pitfalls of each source. It is also imperative that you be transparent about the margin of error of your results.

Objective and subjective indicators both have benefits. Objective indicators must be interpreted, as they could mean different things. Subjective questioning has evolved from previously, when asking the public how corrupt the government is was considered sufficient.

In his belief, all indicators name and shame.


Question from moderator: Why should you measure integrity, versus corruption? Are they different?

Daniel Kaufmann responded that we could discuss semantics all night. What is most important is to consider which issues need to be tackled.

Charles Sampford responded that measuring anti-corruption and transparency are not mutually exclusive. Each evaluation needs to state which values it stands for and why and how these make people better off. Governments must serve the people, not simply be clean. Thus, a corrupt government that builds half a school is better than a clean government that builds nothing.

Marianne Camerer pointed out that more awareness of corruption generates more reports of corruption. Corruption is everywhere, but public institutions combined with civil society and the media can help combat it. The context in a country is important, and linked to political will to fight corruption.

Rosa Ines Ospina responded that results always need to be interpreted.

Questions and comments from audience:

  • How can you draw the link between expert views and public opinion? Between those inside institutions and those outside? Between perceptions and an in-depth study such as an NIS study?

  • What is the difference between state capture and simply a corrupt government? Can you advocate against state capture?

  • Can you define integrity without morals/values?

  • What is systemic corruption? How does it fit in a political or economic analysis? What is the role of culture in corruption?

  • Can an index forecast the future governance situation? For example, by looking at the opinions of youth?

  • The Greek temple is not meant literally. The bird’s nest is too complicated as a rhetorical devise. TI Bosnia also thought about the idea of a nest, but they believed that there are too many variations in the diagram.

Panelist responses:

Charles Sampford responded that integrity can be non-religious, but it is always value-based. However, while values are critical, they should not be imposed from the outside. Integrity systems are worked against by corruption systems. Corruption systems are highly adaptable, so integrity systems need to adapt in response. Corruption is not always illegal—legal corruption must be examined as well.

Culture is critical, but it is only part of the answer to how to instigate change.

The temple is flawed most of all because it is held up as being perfect. To evaluate, you can ask agencies what they deliver and then ask the public what they receive, and then compare: are expectations met?

Rosa Ines Ospina responded that it is important to consider a tool’s use: in their case, they wanted a clear signal to the government. They do not measure state capture, which is a form of legal corruption.

Working with youth is very important, but the focus cannot be on youth to the exclusion of adults—adults need to change as well.

Daniel Kaufmann pointed out that not everything that goes wrong has to be criminalized; there are other ways to create incentives to stop bad behaviour.

Indicators are good for investigation and research but not for detailed action programs, for conditionality, or for ranking.

Systemic corruption is when informality of rules applies throughout the system. The opposite is individualized corruption, for example in Botswana or Chile.

Summary (by moderator):

 The workshop discussed measuring integrity, corruption, and even trust. Integrity requires a definition, as discussed by the panelists. It is important to know what you are measuring and how you are going to measure it. The relationships between things are also important. Measurement is a tool for change. But we must be aware of its strengths and limitations.  

docAssessing Integrity

Brazil 2012

Brazil 2012

IACC Video

IACC Video