Programme conditions, project safeguards: Quo vadis World Bank?

8 October 2007 | At Issue

A fully formatted PDF version of this briefing is also available.

This briefing will clarify the landscape of programme conditions and project safeguards. Our goal is to see what can be learned from experiences with both, and what it implies for a move towards responsible lending standards. While most directly relevant for the World Bank, this debate will have far-reaching implications for the international aid system.

With the introduction of structural adjustment loans in the early 80s, the World Bank’s portfolio shifted from support for discrete projects towards programme financing. Rather than finance the opening of a new mine, the Bank would back reform of the mining sector, or more broadly still, reform of the business climate in an attempt to encourage foreign mining companies to invest. With this shift, the portion of the Bank’s portfolio directly governed by do-no-harm policies has declined, whilst that which involves active interventions into state sovereignty and policy-making has grown.

'good governance' is an arbitrary political construct and lacks an international legal basis

Whilst some civil society activists have been appealing for project safeguards to incorporate international standards on human rights, environment and labour, others have been calling for the reduction or even outright elimination of the use of programme conditions. Some officials label this incoherent – ‘more conditions’ on the one hand, ‘less conditions’ on the other. This confuses two distinct concepts – minimum standards and commitments that states have signed up to in ratifying international treaties as against creditor-imposed programme conditions.

I. Project safeguards

Originally drafted as internal operational policies to guide staff, World Bank safeguard policies evolved in the 1980s into public commitments to avoid doing harm. High pressure campaigns by environmental NGOs and social movements attacked the Bank for sponsoring projects that caused large-scale environmental and social damage. In response, the Bank’s safeguard policies were developed to offer certain standards of environmental and social protection, even if these protections are not provided for in the applicable national law. In 1993, the Inspection Panel, which can respond to complaints of violations of operational policies for IBRD/IDA financing, was established.

While safeguards apply de jure to both project and programme lending (with the exception of emergency lending), in practice they are most closely associated with projects. Importantly, they make scant reference to binding international principles particularly in the case of human rights.

The Bank’s private sector arm, the International Finance Corporation (IFC), did not adopt the safeguard policies until the late 1990s. In 2000, the Compliance Advisor Ombudsman (CAO), the recourse mechanism for private sector financing was created. In a controversial move, the IFC broke away from the safeguard policies in 2006, launching its own eight ‘performance standards’ to guide clients and the Social and Environmental Sustainability Policy to govern IFC actions (see Update 50). By doing this it claimed it was modernising the process and addressing private sector risk. Critics say too much discretion is now allowed to the investor to determine and monitor a project’s social and environmental requirements. Positive gains include a requirement for social assessment and reference to ILO core labour standards.

Safeguards present a dilemma to NGOs. They are narrow in their scope (limited to violations of the Bank’s own policies), inconsistently applied, and have limited mechanisms for monitoring and enforcement. The Inspection Panel and the CAO have no powers of enforcement, and many affected communities feel that these bodies are ineffective mechanisms for redress, providing for a forum for discussion but few concrete solutions (see Update 49). In the absence of any other mechanism recognised by the Bank, NGOs find themselves uncomfortably defending policies that do not reflect their positions strongly enough and have also defended them as the model to be followed by private banks and export credit agencies.

Calls for stronger and more binding safeguards are countered by the argument that this makes them too cumbersome and costly for many borrowers, who will take their custom to less scrupulous lenders. Partly in response to this the Bank’s new ‘country systems’ approach will rely on a borrower government’s relevant institutions and laws, regulations and procedures. The use of country systems refers both to safeguard policies and to fiduciary policies governing financial management and procurement.

Country systems are being discussed for use in operations primarily in middle-income borrowers. National systems will be employed where they have been judged ‘equivalent’ to a simplified set of Bank safeguards and/or fiduciary policies. An “acceptability determination” indicates that borrower countries have the capacity and willingness to implement these systems. After a stillborn pilot in 2005, a detailed proposal to ramp up the initiative is expected by year end. Increasing the use of country systems is one part of the Bank’s commitments to meeting the Paris declaration on aid harmonisation.

Some NGOs fear that this could further weaken existing safeguards since many governments lack the monitoring and enforcement capacity or legal frameworks to provide adequate protections to affected communities. In theory Bank projects have always been required to comply with national laws and regulations. There is also concern that the use of country systems in procurement and financial management is simply a new way to force countries to reform domestic systems to meet international standards designed to reduce the space for state intervention and improve conditions for foreign investment.

II. Programme conditions

Programme conditions refers to those conditions attached to budget support or sector financing. There is vigorous debate over what type of conditions are either appropriate or effective:

Policy conditions are the conditions of the infamous structural adjustment loans. They stipulate that an aid recipient implement reforms in domestic policies such as public sector wage levels or subsidies to industry. They are commonly sub-divided into economic and governance conditions, though in practice the distinction is difficult to maintain. Extensive research suggests that they do not work primarily because, right or wrong, governments who don’t want to make the reforms will find a way around them.

Process conditions relate to transparency, participation and accountability. They tend to be seen as less intrusive. However, there are concerns that conditions requiring transparency in procurement, for example, may prevent governments’ ability to use procurement as a tool for economic development, something they have rejected in WTO negotiations. Evidence on the effectiveness of process conditions is mixed.

Outcome conditions require that a government reach a certain goal to keep the aid money flowing. This may be an internationally-agreed target (such as the Millennium Development Goals), a donor priority, or, ideally, a priority of the citizens of the recipient country. Outcome conditions are meant to allow the government to determine the policies and institutions which it believes are appropriate to reach the specified outcome. There is little experience with outcome conditions, and the difficulty lies in establishing causality between government actions and achievement of certain goals.

In July 2005 the World Bank published an extensive review of its use of programme conditionality, which committed the Bank to five ‘good practice principles’ (GPPs) – ownership, harmonisation, customisation, criticality, and transparency and predictability (see Update 47). The Bank’s GPP implementation review in September 2006, found that the use of conditionality was falling and that the Bank was ‘broadly following’ its new principles. Various NGO shadow reports and the British government contradicted both these assertions, citing specific cases where the GPPs had been violated (see Update 52). A second review, released at the end of 2006, still upheld that “satisfactory progress” had been made but admitted that there was room for improvement.

A number of factors are at play behind the Bank’s manoeuvrings on the use of programme conditions. The power of the Bank over its ‘client’ countries is on the wane owing to a strong global economy and competition from other IFIs, sovereign and private lenders. The Bank is also under pressure to help developing countries reach the Millennium Development Goals, requiring a massive scaling-up of aid. Balanced against this is the increasing importance of the Performance Assessment Framework (PAF). With more donors moving towards budget support to deliver their aid, there is increasing alignment behind conditions included in the PAF. In contrast to the stated intention of budget support, this may actually increase the intrusion of donors into the policy-setting arena.

III. Governance: where programme and project meets

Although civil society groups have decried the social and environmental impacts of economic policy conditions requiring liberalisation, privatisation and deregulation, their position on governance conditions is less united. While some groups advocate an end to all programme conditions, others support the use of World Bank leverage to force the adoption of human rights, environmental protections or increased transparency in otherwise recalcitrant regimes.

Should the Bank require the promotion of human rights and environmental protections by borrower countries as a condition of its support? There is a danger that the Bank could turn the human rights approach into yet another layer of good governance conditionality, framed in their own terms (see Update 52). ‘Good governance’, upon which many Bank and other loans are based is an arbitrary political construct and lacks an international legal basis. The notion of conditions derived from human rights norms (and increasingly, environmental and labour principles) is further mired by the World Bank’s governance issues. It is not a court, is undemocratic and does not allow equal decision-making on its board. Its potential role as an arbiter could never be taken seriously until it addresses its complicity in past violations.

In practice the issue has been muddied. In Congo-Brazzaville, for instance, steady progress was being made by transparency activists who had called for concrete conditions on oil revenue transparency. However, the Bank’s approval of debt relief, under pressure from debt campaigners, removed the motivation for the government to improve its transparency. Local transparency activists were subsequently harassed and detained. Though the Bank then protested over their arrest, its action in cancelling the debt was viewed by many (both within and outside the country) as an endorsement of a corrupt regime.

IV. Responsible financing standards: The way forward?

Three principles emerge which should guide the Bank’s ‘direction of travel’: country leadership, universality and accountability.

We have chosen to use country leadership rather than ownership because in Bank parlance the latter means countries taking ownership of the ‘right’ economic policies. Leadership in programme lending implies continual pressure to move from policy conditions (both economic and governance) to outcome conditions to process conditions with the goal of no conditions or perhaps one condition (transparent use of funds for the purpose specified). Leadership in project lending means continual pressure to move towards country-based safeguard and fiduciary systems.

In practice, country leadership necessitates building the capacity of recipient countries to conduct the ex-ante impact assessment necessary to make informed policy decisions. The use of country safeguards must be introduced cautiously to ensure that it does not result in the de facto elimination of protections for affected peoples, and the shirking of responsibility by the Bank. The use of country fiduciary systems must not be cover for undermining policy space.

Universality refers to the need to move away from creditor-specific conditions and safeguards towards those which apply across all bilateral and multilateral financial transactions. In programme lending this implies moving beyond good practice principles for the Bank towards good practice principles for development finance perhaps developed through the UN Financing for Development process. In project lending this means moving from inadequate institution-specific operational policies and safeguards towards international standards. This shift would also imply a move towards universalised, independent and effective complaint and redress mechanisms to cover the development finance sector as a whole.

Universality must however allow sufficient flexibility to consider the difference in degree and pace of application of these shifts in, for example, low-income versus middle-income countries, or in post-conflict countries or countries where predatory corruption is rampant.

Accountability encompasses transparency and participation of stakeholders in the setting, monitoring and enforcement of programme conditions and project safeguards. This means the revision of policy frameworks to ensure that stakeholders consistently implement their human rights obligations. In programme lending this means that any conditions must be based on transparent participatory processes – this would point us towards minimal conditions encompassing transparent use of resources. Compliance should be monitored by peer review or citizen’s mechanisms rather than officials of donor agencies. In project lending, this similarly means that citizens should be involved in monitoring and compliance mechanisms.

Key references:

  • ActionAid International, 2006, What progress? A shadow review of World Bank conditionality
  • Bank Information Centre and Centre for International Environmental Law, 2004, Country systems approach to World Bank social and environmental safeguards: concerns and challenges
  • Eurodad, forthcoming, October 2007 Charter on responsible lending
  • Halifax Initiative, 2006 One step forward, one step back
  • McBeth, A, 2006, Breaching the vacuum: A consideration of the role of international human rights law in the operations of the international financial institutions, International Journal of Human Rights, Vol. 10, No. 4, 385-404
  • World Bank, 2006 Good practice principles for the application of conditionality, a progress report