Beyond Remedies: How Accountability Mechanisms Can Protect Communities from Irreversible Harm

In Rethinking “What Counts” as Accountability, an essay published as part of the  Accountability Perspectives Project marking the first thirty years of Independent Accountability Mechanisms (IAMs), Jonathan Fox takes a close look at the different purposes accountability is expected to serve. He suggests that one major hindrance to delivering answerability, deterring bad behavior, and incentivizing positive change within an institution has to do with accountability being conceived fundamentally as way to address past acts or omissions, and not as a means of preventing harm. However, in our experience working directly with impacted communities, accountability is sought not only to repair damage, but also to prevent it. Their ability to achieve these goals depends on the extent that IAMs have been equipped to help development finance institutions facilitate remedy and manage risk. Accordingly, this article aims to explore the ways that IAMs can be effective in underpinning the “do no harm” principle as development finance institutions pursue their respective mandates.

      A major way that IAMs have been equipped to prevent and mitigate harm is by broadening complaint eligibility criteria to allow communities to effectively raise concerns about risks and not only harm realized. Afterall, why should communities wait until harm occurs first before holding institutions to account for their environmental and social commitments? This best practice is fortunately widespread, but other key harm prevention features are less so. For example, of the 25+ active IAMs, only a handful are explicitly permitted to recommend preemptive measures like suspension or reconsideration of financing in the face of imminent or irreversible harm. Further, IAMs have yet to be granted the degree of deferential independence that would allow them to issue something akin to an injunctive order-- although there is a compelling argument for doing so ; therefore, preventive outcomes often hinge on management/board uptake and reputational pressure rather than formal enforceability.

Fox argues that transparency is an essential ingredient to support preventative and reactive accountability strategies, but equally important is the independence--and respect for the independence--of the mechanisms designed to provide institutional accountability to the public. Case examples reveal pragmatic ways that IAMs have been and can be effective tools of harm prevention, even considering the current conceptual limitations of accountability in development finance.

Prevention in Practice: How IAM Processes Have Helped to Delay, Suspend, or Terminate Potentially Harmful Projects

Although IAMs have yet to be empowered to issue binding orders, several cases demonstrate that preventive impact is possible. Data from the Complaint Dashboard on the Accountability Toolkit reveals that a handful of IAMs have demonstrated an ability to act preventively by prompting institutions to halt or suspend projects before irreversible harm has occurred. Preventive authority rarely operates as a single formal mechanism; instead, it emerges through the interaction of compliance findings, governance conditions, political context, and community mobilization. The cases below illustrate four distinct pathways through which IAM processes can slow or stop harmful activities before irreversible harm occurs.

1️⃣ Equipping IAMs to Facilitate Dialogue on the Design of Proposed Projects

More than half of the IAMs of major development finance institutions are enabled by policy to receive complaints concerning projects under active consideration by the institution’s Board or management. Of the IAMs that are not provided discretion over projects not yet approved, many are still empowered to at least receive and refer complaints to appropriate management teams to facilitate proactive and early resolution of concerns. Enabling IAMs to address concerns early in the project approval process helps to support environmental and social due diligence and stakeholder engagement efforts, thus proactively integrating accountability into decision-making processes.  

When communities living along Lake Turkana in Kenya became aware that the African Development Bank (AfDB) was considering whether to help finance what would be the third largest hydroelectric plant in Africa– the Gilgel Gibe III Dam on the Omo River in Ethiopia– they filed a complaint to the AfDB’s IAM known as the Independent Redress Mechanism (previously named the Independent Review Mechanism). The communities were concerned that the dam would threaten the viability of the world’s largest permanent desert lake in northwestern Kenya and southwestern Ethiopia, an ecosystem they relied on for  

“A grievance or complaint can be submitted to the IRM by a person or group of persons or community who has/have been or who may be affected by adverse impacts of a GCF funded project or programme.1 [fn 1] GCF funded project or programme includes a project or programme being actively considered for funding by the GCF.”

2️⃣ Relying on leverage available in the financial relationship

In a case related to a project intended to reduce deforestation and bolster climate resilience in Nicaragua, known as the “Bio-CLIMA” project, the compliance review report of Independent Redress Mechanism (IRM) of the Green Climate Fund (GCF) reveals how GCF safeguard architecture and Board-imposed conditions prior to fund disbursement gave the institution multiple opportunities to consider and address compliance issues that presented certain risks of harm.  In this instance, the compliance issues pertained to the rights of Indigenous People and the safety of rights defenders, among other things. In apparent appreciation of the sensitivities around this particular project, the GCF Board had required tranche disbursements and independent oversight of the project through the appointment of a third-party monitor. This foresight enabled necessary leverage to address non-compliance should it manifest. Accordingly, when communities sought accountability to protect their rights after the first disbursement for the project, the IRM compliance review process revealed a need to better appreciate conflict sensitivities in the area and to ensure that the free, prior, and informed consent of Indigenous Peoples was actually achieved before any on-the-ground activities began. After issues were identified in the course of implementation, the GCF Secretariat ultimately terminated the project. The GCF’s conditional-approval, combined with the IRM process, slowed roll-out so that compliance gaps could be identified and further harm prevented. 

3️⃣ Integrating Preventive Possibilities into IAM Policies

The necessary leverage to prevent and mitigate harm caused by non-compliance can even be baked into IAM policies themselves. For example, a section on “temporary and pre-emptive measures” within the guidelines of the IAM of the United Nations Development Programme (UNDP), the Social and Environmental and Compliance Unit (SECU), provide that SECU may call for temporary suspension of a project where there is a risk of “imminent, significant and irreversible harm” to project affected communities. This provision was preempted by UNDP Myanmar In practice, this provision helped to stave off the infringement of the rights and traditional stewardship of Indigenous Karen communities with respect to lands included in a large-scale “Ridge to Reef” conservation project in Myanmar. In 2018, on behalf of 612 community members who signed the complaint, UNDP suspended project activities pending SECU’s compliance review, consistent with SECU’s guidelines. Amid project suspension, a community-aligned conservation group known as Conservation Alliance Tanawthari was able to help communities advance an alternative Indigenous-led conservation plan, called Landscape of Life, which offered to both protect the rich biodiversity of the region, as well as respect traditional peoples’ knowledge. The Ridge to Reef project was formally cancelled in 2024 amid ongoing conflict in Myanmar, and UNDP has since committed to applying lessons from the SECU process to strengthen its future programming in Myanmar.

Another means of integrating preventative possibilities is by enabling IAMs to intervene in apparent compliance issues challenged early in the process. For example, risks of harm presented by the Eco Oro Angostura gold mining project in Colombia, supported by an early equity mining approach of the World Bank Group’s International Finance Corporation (IFC). Under the approach, the IFC’s mining strategy entailed providing equity and loan financing for mining companies to support several phases of mine development, from exploration and feasibility to construction and production, and eventually decommissioning and reclamation. The Eco Oro Angostura project was still in the early stage of feasibility, and its design was not complete when the IFC’s Compliance Advisor/Ombudsman (CAO) found eligible a complaint raising alarm to several serious issues implicating meaningful stakeholder engagement, environmental and social assessment, and appreciation of biodiversity impacts to the sensitive Santurbán páramo ecosystem. Under these contexts, the CAO process was able to facilitate early intervention in the design phase of the project, and it ultimately revealed compliance-based risks that would need to be addressed to avoid undue harm to the protected ecosystem of the region. Ultimately, project development was suspended by Eco Oro’s management, and the IFC divested from it entirely in 2016, marking a preventive outcome that withdrew its support before irreversible harm could occur.

4️⃣ Mustering Political Leverage Through IAM Processes

Even when financial leverage is tenuous and IAMs are not expressly equipped to prevent harm, grassroots advocacy campaigns have been able to effectively use IAMs to stave off harm until meaningful dialogue on the issues is had. Take for example the proposed Cerro de Oro hydropower project in Oaxaca, Mexico. The Office of Accountability (OA) of the former  Overseas Private Investment Corporation (OPIC) (now the U.S. International Development Finance Corporation) intervened in 2010 in response to demands from communities concerned about impacts to their lives and livelihoods. During the OA’s review, project construction and financing were voluntarily suspended, illustrating how the filing of a complaint can help to create a political moment that opens space for dialogue even before a formal IAM review commences. The OA was able  to encourage parties to the table, alongside the voluntary suspension of the project, so that the project proponents could engage in mediated dialogue with aggrieved communities to address the feared adverse impacts. When communities signalled that revised project designs still carried an unacceptable level of risk, the project was suspended, and remains so to this day.

      Taken together, these cases illustrate that preventive impact often arises through different channels, be it through the presence of political or financial leverage, or through deference and responsiveness to IAMs empowered to recommend suspension. Each route emphasizes that preventive action is possible even though IAMs presently lack formal unilateral suspension power.

Early-Resolution Pathways and Consent: Ensuring Community Agency During Ongoing Project Implementation

Building on the examples of preventive action, a complementary frontier for strengthening accountability lies in ensuring that agreements between companies and communities are not merely aspirational but enforceable. As highlighted by Inclusive Development International’s Perspectives Project, community-company agreements should be legally binding and enforceable by the communities themselves supported by independent legal counsel and enforceable through courts or human-rights–compliant arbitration such as the Hague Rules on Business and Human Rights Arbitration. Moreover, development finance institutions, commercial lenders, and investors should provide resources for independent legal and technical assistance to communities and use their leverage to ensure that these obligations are honored throughout the project cycle. Embedding such enforceability under an architecture of preventive accountability would help bridge the current gap between voluntary commitments and actual remedies, ensuring that communities have accessible, effective means to prevent harm before it occurs.

Another challenge that needs to be addressed is that approximately 8–12% of all closed IAM cases are categorized as either “addressed outside process” or “not desired by the complainant”, confirming the existence of referral pathways and opt-out possibilities. However, there is often not any indication that complainants were provided informed, voluntary, and time-bound consent to such referrals, nor whether they retained the right to re-enter the IAM process if early efforts failed. In most cases, the outcomes of these referrals are not publicly documented, leaving uncertain whether community concerns were substantively addressed or merely redirected. Here, as Jonathon Fox advocates, is where transparency is direly needed. As stated by our recent report exploring IAM outcomes, titled Accountability in action or inaction?, there is simply too much opacity to understand the outcomes achieved when IAMs redirect complaints outside of process.

When Harm Has Already Occurred: Halting as a Preventive Accountability Tool

Prevention in accountability should not be understood narrowly as action taken only before any harm occurs. In practice, some of the most consequential preventive impacts of IAMs emerge after harm has already taken place, when decisive institutional responses can halt ongoing abuses, prevent their recurrence, and signal that certain forms of conduct will not be tolerated going forward. In this sense, suspension and cancellation function not merely as remedial responses, but as forward-looking safeguards that prevent the entrenchment and replication of harm.

The World Bank Inspection Panel’s investigation into the Tanzanian “Resilient Natural Resource Management for Tourism and Growth” (REGROW) project illustrates this dynamic. Although serious human rights abuses like forced evictions, beatings, killings, and the seizure of livestock by park rangers in Ruaha National Park had already occurred, the Panel’s findings had the potential to transform the accountability process into a mechanism for halting further harm. The Panel’s  investigation uncovered non-compliance with World Bank’s policies on Involuntary Resettlement, Investment Project Financing, and Environmental Assessments to the extent that management failed to recognize inevitable involuntary physical and economic resettlement enabled by existing law in Tanzania and its support of a recognized law enforcement agency. To halt the proliferation of harm caused by the non-compliance, World Bank management responded by invoking contractual remedies and suspending disbursements to the project. Shortly after, the Tanzanian government formally requested cancellation of the project, and the Bank confirmed its termination.

 Unfortunately, the World Bank’s management action plan intended to address noncompliance issues failed to integrate community demands for justice and reparations, and ultimately promises made by the government within the MAP were broken. As an advocate for the communities, the Oakland Institute has reported that even after project closure violence against the community persisted, including the murder two young villagers, the seizure of over 1,000 cattle, and other efforts to prevent farmers from cultivating their lands.

This cautionary tale reminds us that preventive accountability is not a guaranteed outcome of IAM processes, but rather something that has to be intentionally pursued. Preventive accountability depends on IAMs, and the boards and management they advise, being prepared to not only recommend halting projects when continuation would exacerbate harm, but also to follow-up with what leverage may still exist after exiting a project. 

Conclusion

While IAMs can help communities achieve meaningful remedies, prevention remains one of the most urgent and underdeveloped frontiers for accountability. If environmental and social due diligence is primary preventive tools relied on development finance institutions, then consider proactive accountability and transparency to be the safety gear needed to accompany those tools.