New CIDA Reforms Risk Excluding the Poorest of the Poor

September 1, 2001

In June 2001, the Canadian International Development Agency (CIDA) released a policy document proposing distinct new terms for development aid. The report, called "Strengthening Aid Effectiveness," outlines a new direction for CIDA and new terms of reference for distributing aid.[1] On the one hand, it ties aid to global security, with the aim that CIDA should work together with other aid agencies to contribute to a more stable and affluent world. On the other, it proposes new conditions for receiving aid that may leave a number of developing countries off the map for development assistance. Countries that are affected by conflict, political instability, and "poor governance" would not qualify for aid from CIDA, in my interpretation of the new policy, and yet the civilian populations in these countries are often the poorest of poor. Without help from aid agencies, they risk becoming further marginalized within the category of 'developing countries', generating greater poverty and socio-spatial stratification across the world.

In early September 2001, the Minster Responsible for International Cooperation, Maria Minna, led cross-country consultations on the proposals made in 'Strengthening Aid Effectiveness.' My brief to the Minister highlighted three points: 1) the policy could contribute to the emergence of a two-tier aid program in which countries with 'sound management' receive funds and those assessed as lacking in this area receive only ad hoc humanitarian assistance; 2) the report is conspicuously silent on the gender implications of the proposed policy, despite CIDA's longstanding commitment to gender equality; and 3) the policy suggests a shift from managing individual projects to working at a broader scale of sectoral planning in which projects are part of larger programs. What will happen to the innovation and success of current CIDA programming? Allow me to elaborate briefly.

First, CIDA's "Strengthening Aid Effectiveness" has antecedents in World Bank programming that has been implemented elsewhere (i.e. UK, Netherlands, US). While it purports to increase the efficiency of development assistance, it also enables governments in donor countries to exert influence on the policies and practices of governments in recipient countries, effectively instantiating an element of governance from a distance. CIDA's desire to make big changes is politically rational, given the severe cuts to its budget and the decline of overseas development assistance (ODA) generally.[2] I understand the political importance of proving CIDA's relevance, impact, and accountability, so that its legitimacy is strengthened and its good works may continue. Nonetheless, analyses of this approach have questioned the two-tier structure of aid delivery that such a policy generates (Macrae and Leader, 2000; Duffield, 2001). On the one hand, developing countries that meet requirements of sound economic policies, political stability, and good governance will receive development assistance; on the other, developing countries that fail to meet these criteria, including countries at war and/or those affected by military coups or despotic regimes will not (World Bank, 1998).[3]

According to the World Bank:

"Donors in 1996 gave less assistance to countries with very poor policies than to ones with mediocre policies. However, they also tended to give small amounts of assistance to countries with truly good policies. (The figure is drawn controlling for poverty level and population, so it answers the question: how have donors treated countries with similar poverty and population but different qualities of policy?) Strong reform countries such as Ethiopia or Uganda received less aid per capita than poor countries with weak policies. To maximize poverty reduction, aid should ideally taper in with reform, when in fact it tends to taper out." (ibid.)

While clear definitions of poor, mediocre, and good policies are not completely clear in this document, the World Development Report (World Bank, 2001: 198) does outline a measure of policy and institutional soundness. Called the 'Country Policy and Institutional Assessment", this tool gives equal weight to twenty components that fall under four rubrics: economic management, structural policies, policies for social inclusion and equity, and public sector management and institutions. Sound familiar? Criteria such as gender equality, human rights records, infant mortality, life expectancy, and other indicators of need, broadly defined, are not included. Yet even in countries deemed to do poorly in this assessment, poverty is no less grating. Dispossession and need are likely to be higher in countries affected by conflict, economic instability, and poor governance. Certainly, aid does not have to be (nor is it) delivered exclusively via governments. The work of non-state actors, such as non-governmental organizations, has proven effective in CIDA's own assessments and evaluations. While this policy aims to untie aid in the conventional sense, a move that I applaud, it does tie aid to new and different conditions.[4]

Second, the gender implications of this strategy have yet to be articulated. The "Strengthening Aid Effectiveness" policy points to a shift from small-scale community development projects to a focus on macroeconomic policy, governance structures and institutions, and stronger partnerships with recipient governments. Historically, government has been the centre of power relations, and yet its membership has excluded women to varying degrees. CIDA has long been a leader in raising awareness of such power relations and insisting on more equitable participation and decision-making by underrepresented groups, including women (see Gender equality is one of CIDA's six program priorities, and yet it is not integrated into the proposed modality of aid allocation.[5] The policy paper does refer to gender in two ways: 1) by noting that gender analysis and environmental assessments are process requirements for CIDA projects; and 2) that poverty, experienced more by women than men, will be alleviated by the implementation of this comprehensive policy. Nonetheless, there is little if any analysis of power relations, and the shifts in decision-making and participation that will occur under the proposed model. Gender is relegated to the domain of 'soft' policy, in a climate of 'hard' economic and political reform. A gender audit is in order.

Third, according to the World Bank (1998), "there has been more progress with poverty reduction in the past fifty years than in any comparable period in human history." If this is true, then CIDA and other development agencies, in conjunction with private sector developments, are doing some things very well. In my view, CIDA's successful programming should not be displaced by the provisions of this policy. That is to say, the new approach should not be a choice of 'either/or.' Rather, where CIDA is doing a demonstrably effective job in a country characterized by conflict and instability, such difficult development work should be continued, documented, and disseminated for others to learn from. To eliminate aid to countries characterized by war, instability, or poor governance until they meet the specified conditions is to ignore the significant gap between humanitarian, or relief, assistance - that often consists mostly of food aid - and development assistance that strengthens the capacity of institutions, governments, and civil society organizations.

CIDA is clear and correct in noting this in the policy paper:

Another issue that CIDA needs to come to grips with is the increasingly serious gap between the provision of emergency assistance as a conflict situation comes to an end, and when conditions are right for return to traditional longer term development assistance....there remain numerous countries...[that] find themselves in a kind of limbo- beyond the need for large amounts of humanitarian assistance, but not yet stable enough to allow traditional long-term programming to take place. It is not clear that existing CIDA mechanisms are adequate to confront this task (CIDA, 2001: 27)

Nor does the proposed modality of delivering development assistance prove adequate to fill this gap. The initiatives supported by CIDA's Peacebuilding Fund are certainly modest but significant steps to meeting this requirement, but CIDA's work on conflict reduction and peace-building as development strategies must be strengthened and fortified, so as not to exclude countries that will not otherwise qualify for assistance. Because conflict reduction strategies are difficult to measure in quantitative terms, they too risk being relegated to the 'soft' policy domain, clearly subordinate to the 'hard' policy domain of calculable development as macro-economic growth.

One example I can speak to from experience is that of Sri Lanka. This country has hosted war on and off for 18 years. CIDA's operations in Sri Lanka are unusual. First, "Canada is the only donor in Sri Lanka which does not fund line government ministries or programs. CIDA's total program, albeit a modest one of about $5 million, is delivered through non-governmental and community-based organizations" (CIDA briefing paper, no date).This means that Canada's assistance to Sri Lanka is channelled exclusively through NGOs and civil institutions that can address the political, social and economic causes of the conflict (ibid.). Canadian aid targets conflict reduction, increased tolerance in a country fraught by nationalist chauvinisms, and respect for human rights. CIDA's involvement has a stabilizing affect on power relations in the country.

The training projects that CIDA funds in Sri Lanka tend to invest more in human development in the current context of conflict. This is critical to its success, as resources for training people are much less likely to be diverted by warring parties and used for other ends, potentially fuelling further conflict. CIDA has an enviable reputation among donor agencies and among Sri Lankans on all sides of the conflict.

A concluding remark

While political pressures force CIDA to prove its relevance, effectiveness, and accountability, CIDA should not abandon its successes and institutional memory. While increasing the effectiveness of aid is important, it should be measured in terms that go beyond mainly macro-economic indicators. A gender audit of this policy - to determine its implications for women and men, in rural and urban contexts, and from different class backgrounds - is critical. Without an audit, the agency risks undermining its gender equality objectives and social development priorities. The policy indicates a shift from working with community organizations, local NGOs, and educational groups towards working in cooperation with partner governments and other donor agencies, a move that would have gender, class, and geographical implications given the gender composition of these various institutions.

Finally, CIDA should assess the success of such policies in the UK, US, and Netherlands. Have these efforts proven effective through evaluation and assessment? Does this policy create a two-tier system of development? And if so, what are the implications for the numerous conflict-ridden and fiscally unstable countries that will be precluded from CIDA funding?

References Cited

CIDA. (2001) "Strengthening Aid Effectiveness", policy paper for consultation, June.

_____ (no date) "CIDA's Program in Sri Lanka", briefing paper.

Duffield, M. (2001) Global Governance and the New Wars (London: Zed Press).

Macrae, J. and Leader, N. (2000) "Shifting Sands: The search for 'coherence' between political and humanitarian responses to complex emergencies", Humanitarian Policy Group Report 8, (London: Overseas Development Institute), August.

Valpy, M. (2001) "Our hearts as big as all outdoors", Globe and Mail, September 3.

World Bank (2001) "World Development Report 2000/2001: Attacking Poverty", found at, accessed August 30, 2001, New York: Oxford University Press.

_____ (1998) "Assessing Aid: What works, what doesn't, and why?", New York: Oxford University.


1. See CIDA's web-page for the most recent version:

2. ODA is half of what it was 15 years ago (Valpy, 2001); Canada's development aid was at its peak at this time at 0.5% of GNP; today it is 0.25% of GNP.

3. The World Bank is mentioned 15 times in the February 16th draft of the report.

4. Tied aid initially referred to the tying of bilateral assistance to goods and services produced by vendors in the donor country. It did not refer to partnerships with NGOs and other non-state actors involved in development.

5. CIDA's 6 program priorities include: basic human needs, gender equality, infrastructure services, human rights, democracy and good governance, private sector development, and the environment CIDA's 4 social development areas are health and nutrition, basic education, HIV/AIDs and child protection.

Jennifer Hyndman is an assistant professor in the department of geography at Simon Fraser University, and a research associate with the CCPA. This article is based on a brief presented to Maria Minna, the federal Minister Responsible for International Cooperation, in September 2001.