Launched in 2003, the Extractive Industries Transparency Initiative (EITI) is a major international effort to disclose information about extractive activities. Initially focused on payments between companies and governments, the EITI evolved into a broader instrument seeking to improve transparency and accountability to contribute to more developmentally effective extractive resources exploitation.

EITI has been most successful in reaching its institutional goals, and fairly successful with some operational goals, but it is not clear whether the EITI has had any impact on its development goals. This, we suggest, calls for an explicit Theory of Change (ToC) that is able to clearly explain how the EITI is expected to improve natural resource governance and developmental outcomes. Our study confirms that many EITI stakeholders expect a ToC to be useful and we note that the EITI Board and International Secretariat are now embracing that view.

Based on previous literature, a survey, three country case studies, and a decade of engagement with the EITI, we thus introduce three simplified and stylised ToC models for the EITI:

1. Naming and shaming, through which the EITI helps identify individuals or institutions mismanaging or embezzling revenues to curtail revenue loss and increase integrity.
2. Public debate, with the EITI improving natural resource governance through increased knowledge among the public and demand for better resource governance.
3. Technical reforms, with the EITI increasing the effectiveness of bureaucracies in handling resources revenues collection and allocation.

Survey and interview results suggest that the Public debate model is deemed as the most important, the Technical reforms model as the most efficient, and the Name-and-shame model as the most relevant in the early phase of implementation.

The models are not mutually exclusive and can help structure discussions around the design and implementation of the EITI.

Each of these models needs to be further elaborated, including at the subnational level, by identifying intermediary steps, implementing agencies, and mobilising civil society organisations (CSOs). A ToC should reflect the evolution of national contexts, objectives, and capabilities, so that it suits local conditions and evolves in response to changes – both in terms of the EITI itself and the requirements of the implementing country. We note in this regard that:

The Naming and shaming model requires close links with effective investigations and prosecution – either at the national or international level – in order to curtail revenue loss and increase integrity in resource management.

The Public debate model relies in part on civil society having access to information and being capable of interpreting and mobilising it politically – through social movements or the political system – so as to effect change. A lack of freedom of expression for CSOs in repressive political regimes can drastically undermine the viability of this model.

A relatively strong bureaucracy can help the Technical reforms model deliver incremental changes, even in the absence of high-level political leadership. This opens the way to make the EITI effective, even within politically challenging contexts.

Despite the importance of nationally contextualising and designing ToCs, the ToC also needs be grounded at EITI central level through common goals and the 2019 EITI Standard, which includes among others the EITI Principles, the EITI Requirements, and a protocol for civil society participation. The EITI International Secretariat has an important role to play. This can include the creation of ToC templates, support for national-level ToC design, as well as acting as a clearing house to exchange examples of ToCs and implementation experiences.

In summary, there is still a need for the EITI to go from transparency to accountability, and to demonstrate effective improvements. Clear ToC models suited to national contexts and objectives could help in this regard, not only to identify and implement pathways towards improvements, but also as a clearer way to assess progress. We make the following recommendations:

• The EITI Standard must not prevent countries from innovating and adapting to reflect their own requirements.
• The EITI should constantly look to national- or local-level demands and initiatives as a source of inspiration.
• Implementing countries can produce their own country-specific ToC with assistance from the EITI International Secretariat.
• The promotion by EITI donors of efficient and relatively independent bureaucracies can deliver technical reforms, even in the absence of a strong political commitment for such changes.
• The complex interplay between information dissemination processes, public awareness, and decision-making by companies and governments is a fundamental element for the EITI to consider.
• The EITI should pursue efforts to include social and environmental dimensions within its framework.
• A ToC should integrate ‘rights-based’ approaches, to improve transparency and accountability on consent rights, as well as rights to a clean and healthy environment.
• The EITI should foster a systematic monitoring of evidence of impact, including uptake of EITI findings and recommendations by relevant organisations.