GAB welcomes back Alan Doig, Visiting Professor at Newcastle Business School, Northumbria University, who contributes the following guest post:

Transparency International (TI) has developed a number of tools to assess corruption in different countries. The best-known is probably the Corruption Perceptions Index (CPI), which purports to reduce each country’s level of (perceived) corruption to a number, to facilitate international comparisons. But TI has also developed another tool, the “National Integrity System” (NIS) assessment, which evaluates an individual country’s governance system with respect to both internal corruption risks and its capacity to fight corruption in the society more broadly; the NIS evaluations typically focus on a number of governance “pillars,” such as the legislative, executive, and judicial branches, audit agencies, anticorruption agencies, media, civil society, etc. (TI maintains a number of more recent NIS assessments on its website.)

Recently, the NIS evaluations have been subjected to withering critiques. For example, last year on this blog Rick Messick summarized a critical article by Professors Paul Heywood and Elizabeth Johnson, which argued, on basis of the NIS for Cambodia, that the NIS reviews relied on a “narrowly conceived institutional approach,” displayed “insufficient appreciation of cultural distinctiveness,” failed to properly conceptualize the notion of “integrity,” and over-emphasized “compliance-based approaches to combating corruption at the expense of the positive promotion of integrity.” Alas, this critique largely misses the mark, and in fact goes a long way (as does Mr. Messick’s blog post) to perpetuating myths and incorrect assumptions about the NIS approach. It’s true that something has gone badly awry with TI’s NIS assessments—on that point, I agree with Heywood, Johnson, and Messick. But though these authors tell us what is wrong with the NIS, they never bother to ask themselves why; more importantly, they make the mistake of confusing errors in execution (on the basis of a single country!) with inherent problems with the concept itself.

As someone who worked on the first 40 NIS country studies, I can say that as originally conceived, the NIS evaluations had two main objectives. The first was to dig deep into the realities of corruption in each country—to look at what actually happens in practice, rather than to provide a superficial, surface-level description. The second objective was to develop a more holistic approach to evaluating anticorruption and government integrity, by considering the full range of domestic and external actors who play an important role in fighting corruption. It is for this reason that the NIS assessments surveyed these as “pillars,” assessing their strengths and weaknesses and identifying opportunities for reinforcing and utilizing them in the fight against corruption. (It’s also worth noting that originally only a minority of pillars identified a specific or single institution.) Moreover, the larger purpose of the enterprise was to provide a platform for additional research, a national plan, and further reviews to monitor progress. The first NIS was intended to be a first step in a longer process. The report would set a baseline for a set of elements (such as the identification of government activities that are most prone to corruption or the creation of a partnership between government and all elements of civil society) that would facilitate work toward a set of objectives (such as mechanisms supporting accountability and transparency in the democratic process or private sector self-regulation). The ultimate goal was not simply combating corruption as an end in itself, but rather the larger struggle against official misconduct, and the creation an efficient and effective government working in the public interest. Furthermore, in the early years the NIS concept was de facto “shareware,” with a number of academics adapting and developing the concept of “integrity systems,” recognizing specific issues relating to types and levels of developmental trajectories, and working toward second-generation NIS studies. Indeed, “corruption” was taken to be a flexible term, with an awareness of cultural and other issues in its country-specific application.

Alas, much of this has been lost. The NIS evaluations have become overly formalistic and formulaic, and this limits the utility of NIS country studies as an effective framework for analysis or platform for progression. As they stand, they are a snapshot in time, parading specific cases as evidence of corruption without unpacking the causes and thus the researchers’ ability to go behind the official position—to explore, for example, a country’s political economy or the role of organized crime—in order to define holistically the threat(s) to be addressed. Further, they provide an institutional narrative as the basis for a shopping list of undifferentiated recommendations for reform which offer nothing by way of a starting point for, or ownership of, a managed reform process, and often do not demonstrate an understanding of what is practical and realistic in the country in question. There is no focus on reform in terms of priorities, sequence, and cost within an action plan design and implementation program, along with an appropriate monitoring and reporting process.

Why did this happen? Part of the problem has to do with the internal politics and processes of NGOs, from which TI is not immune: the TI Secretariat in Berlin appeared to have wanted to establish the NIS system as part of the TI “brand,” which implied exclusive ownership and management of the NIS. And, partly to compete with other country assessment tools, which include simple (simplistic) checklists and scores, TI has played around with the pillars, and elevated the results of the questionnaire survey and the researcher’s own ranking over the more complex, nuanced evaluation of what really happens.

Yet this is a problem with the execution, not the concept. The NIS evaluations could have a significant role to play in anticorruption reform, but for this to happen, TI needs to stop treating the NIS studies as a standalone activity and look to its own holistic approach that draws on its various tools and techniques to shape the reform process. TI already has tools that take a more holistic approach, such as its Collective Action on Business Integrity: A Practitioner’s Guide for Civil Society Organizations, which outlines several key steps to engaging in collective action, underscoring the importance of stakeholder mapping, ongoing risk assessment, objective-setting, and real-time monitoring and evaluation.

A country study, such as an NIS review, should be the gateway or platform for reform initiatives, whether institutional, sectoral, or procedural. The NIS was originally conceived as a holistic starting point for developing practical, sustainable initiatives to deliver change. We may understand what is wrong with the NIS in current practice and we can discuss why, but a much more important discourse needs to take place on how to work toward an increase in the honesty or integrity of government as a whole, and how tools such as the NIS may contribute to that process.