email icon Email this citation


Women, the Environment, and Development Assistance   *

Matthew Auer

International Studies Association

March 18-21, 1998

Abstract

During the 1990s, many international declarations and action plans proclaimed that low income women's welfare and the health of the environment were interdependent and that international development agencies must devote more attention to the gender-environment nexus. This study examines whether and to what extent legal and rhetorical commitments to gender and the environment have been translated into official development assistance (ODA). Based on project abstracts authored by bilateral aid agencies, and with special attention to Canadian and U.S. aid activities, geographical flows of gender and environmental aid were identified and the frequency of this aid by sector was measured. The number of new Canadian and U.S. aid projects containing gender and environmental objectives increased during the 1980s and early 1990s. However, in the vast majority of project descriptions, gender objectives and environmental objectives were not formally linked to one another.

Introduction

Increasingly, economic progress, environmental protection, and women's rights and welfare have become intertwined in the sustainable development discourse. The professional development community has recognized low-income women not only as the main natural resource managers at the household level, but as the principal resource custodians for low-income agrarian economies, generally. Declarations and action plans signed at major United Nations-sponsored conferences on the environment (1992), population and development (1994) and women (1995) proclaim that sustainable development hinges on women's full participation in economic, social, and political institutions at local, regional, and national levels.

This study considers to what extent the development community has responded to its own exhortations about gender and the environment. First, theories and philosophies equating women's welfare and environmental quality are examined. The second section describes the convergence of gender and environmental concerns in the proceedings of major international conferences and in international agreements. Remaining sections present the findings of research on donor assistance for gender and environment activities. Of particular interest is the relationship between the donors' rhetorical commitments to gender and environmental concerns and the physical aid rendered for these concerns. This study also examines whether donors have integrated gender objectives and environmental objectives in aid projects, over time. Canadian and U.S. aid forms the core of the analysis, as women's welfare and environmental protection are key mission objectives of bilateral aid agencies in these two nations, and Canada and the United States have published data on virtually all of their official development assistance (ODA) activities since the 1960s.

Women, the Environment, and Development: Theory

Beginning in the early-1970s, critics observed, and some donors conceded, that development assistance was biased toward low-income men (O'Kelly, 1973, pp. 92-96) or that low-income female beneficiaries were "passive recipients" of aid, seldom participating in aid planning and implementation (World Bank, 1975; OECD, 1985, p. 216). Around the same time, feminist and ecofeminist theorists began writing about the special bond between women and the natural environment. According to these authors, the alliance between women and nature is forged by their shared experience of exploitation at the hands of men and male-dominated society (d'Eaubonne, 1974; Gray, 1981). Others asserted that the women-environment connection is determined by women's primary resource responsibilities and the fact that low-income women are more often dependent than are men on natural resources for subsistence and income (Murphy & Murphy, 1974; French, 1985). An extension of this argument proposes that versus men, women are more acutely affected by resource scarcity and environmental degradation (e.g., Davidson et al., 1992, pp. 23-25; Rodda, 1991, pp. 879-891).

Some have challenged the validity of the supposed women-and-environment imperative. For example, Shah & Shah (1995) observed that environmental decay in Gujarat, India was injurious to women, men, and children alike. Jackson (1995) found that Zimbabwean women living in a region with relatively scarce resources faired better than female counterparts living in a region with relatively abundant resources. In this case, market relations, women's disassociation from household-based farming units, and male out-migration were more important than resource scarcity in explaining differences in women's welfare. Still other commentators urge that women's resource-based roles may render them more susceptible to environmental risks, but reject the notion that women are inherently more environmentally conscious than are men (Venkateswaran, 1995, p. 13). In addition, some feminists and ecofeminists have parted ways on philosophical grounds, with feminists disavowing a tendency among ecofeminists to equate the rights of non-human species with women's rights. Moreover, some feminists disagree with ecofeminists' apparent willingness to use gendered conceptions of work and familial roles in the development discourse (Bretherton, 1996).

Development experts and advocates of women's rights have evaluated the development community's efforts to assist low-income female natural resource managers. Critiques of externally financed gender and environment projects are numerous, though largely descriptive (see, e.g., Dankelman & Davidson, 1988; INSTRAW, 1991, pp. 119-132). Some studies reinforce dominant themes in the development and feminist literatures, namely that development philosophies promoted by bilateral and multilateral aid agencies have not improved the lot of women (Wilson and Whitmore, 1994, pp. 55-56). Discussions of project failures range from tempered self-criticisms by donors to strong indictments of aid policies and outcomes (see, e.g., St. Hilaire, 1993). Other commentators and organizations identify gender and environment projects that satisfy one or more criteria for success (e.g., Hannan-Andersson, 1995; Dankelman & Davidson, 1988, p. 20; 38). In spite of the numerous case studies on aid to women and the environment, there is relatively little comparative research on the development community's efforts to integrate gender and environmental objectives in policies, programs, and projects over time. The remainder of this paper is a modest first step to address this research need.

Origins of Gender and Environmental Aid

For more than 100 years, nations have signed, ratified, and complied with international conventions governing the development and conservation of natural resources. International legal conventions on women's rights also claim a relatively old pedigree: the first such agreement predates the founding of the United Nations. However, gender and environmental concerns have become integrated in the international legal discourse only relatively recently. The United Nations' gender-related activities focused on women's human rights in the 1950s; its agenda expanded in the 1960s to include women's economic welfare and women's participation in development (United Nations, 1996, pp. 8-25). In the 1970s, women's health movements in the United States and Europe drew attention not only to reproductive health issues but ancillary concerns like women's access to sanitation services and potable water (Sen, 1994). Growing international interest in gender issues, buoyed in part by women's political movements in industrialized nations, culminated in the declaration of the International Women's Year in 1975, convocation of the first UN-sponsored World Conference on Women (1975), and promulgation of the Decade for Women (1976-1985).

Of the first three UN conferences on women, only at the third meeting in Nairobi (1985) were women's welfare concerns and environmental issues formally linked. The conference's Forward Looking Strategies declared that both natural and man-made disasters forced women into marginal environments (1985, |P 224).

Several new international organizations and events dedicated to the gender and environment nexus were inaugurated in the 1980s. 1 Over the last 15 years, most bilateral aid agencies have created an internal office or program dedicated exclusively to women and development issues. At the United States Agency for International Development (USAID), "gender analysis," the systematic appraisal of a prospective aid activity's impact on women, became a standard planning tool in all new aid projects during the early 1990s. Around that same time, "gender" began to supplant "women" as an organizing concept in aid planning. The appellation, Gender and Development (GAD) replaced Women in Development (WID) in program and project mission statements, consistent with a new attention to how men and  women participate in "production and reproduction" in poor countries (Blumberg, 1995, p. 9).

The 1992 Rio Declaration signed at the UN-sponsored Conference on Environment and Development (Earth Summit) marked a watershed in the international community's rhetorical commitment to and legal recognition of the gender and environment relationship. Signatories proclaimed that sustainable development depended on women's full participation in environmental management (Rio Declaration, 1992, Principle 20). Member delegations also approved a chapter in Agenda 21 dealing with gender and environment issues, and calling for governments to formulate and implement policies that would provide women with "equality in all aspects of society" including access to resources, credit and property rights (Agenda 21, 1992, Chapter 24, |P f). Agenda 21 also urged governments, international organizations, and aid agencies to ensure adequate funding for gender-related concerns in on-going and future activities (|P 24.9-24.12 and |P 24.12 passim ).

Some women's advocates were skeptical about the resolutions adopted at Rio. Were Agenda 21 to be implemented, one critic observed, it would burden already overworked low-income women, while offering uncertain returns (Häusler, 1994). These doubts notwithstanding, subsequent international conventions and action plans reiterated Rio's call to increase women's access to and control over natural resources. For example, both the Platform for Action developed at the Fourth World Conference on Women (1995, |P 246-258) and the Conference on Human Settlement's Habitat II Agenda {1996, 3|P 46(d); 4|P 78(f); 116(b); 124(a); and 141(l)} reinforced Agenda 21's demands for gender-sensitive sustainable development.

Examining Gender and Environmental Aid

With the advent of the Rio Declaration most members of the United Nations agreed not only that gender and environmental priorities were linked, but that sustainable development depended on women's secure rights to and control over environmental resources. The analysis turns to whether development assistance for gender and the environment kept pace with these legal and rhetorical commitments. Among OECD members countries, Canada and the United States sponsor large numbers of aid activities containing both gender and environmental objectives. In addition, Canada and the United States have assembled and made available a complete compendia of official development assistance project abstracts, backdated to the first ODA activities sponsored by extant aid agencies in each nation. 2 The analysis below identifies the major sectoral and regional interests of the bilateral sponsors of gender and environmental assistance, and it weighs North American donors' efforts to integrate gender and environmental objectives in their project descriptions. With all due respect to the putative distinctions between "GAD" and "WID", to be consistent, the acronym "WAE" is used below as an abbreviation both for "women and environment" and "gender and environment."

The principal data source for this study was a commercially-available database containing over 167,000 ODA project records contributed by over 30 bilateral and multilateral aid agencies and international organizations. 3 The database, called INDIX, is produced by the International Development Research Centre (IDRC), a Canadian public corporation. Agencies contributing information to INDIX use a standardized reporting format to increase data comparability (IDRC, 1994). This study relied on data from the eighth edition of INDIX. The eighth edition catalogues donors activities through 1995 and part of 1996. Projects commencing after 1995 were not included in this analysis.

The vast majority of records in the database indicate the name of the contributing organization; an activity identification number; an activity title; the funding source; the recipient country/region; an activity status; an executing entity; and an estimated, planned or actual activity start date and completion date. Virtually all records contain some combination of single or multiple word descriptor terms. Most records include an activity abstract and a range of supplementary information on the type of assistance rendered (e.g., grant; loan; technical assistance; etc.); an activity budget; budget currency; and contact address. The database does not contain a complete collection of ODA sponsored by bilateral and multilateral aid agencies and international organizations. Some multilateral aid agencies have not contributed to the database (e.g., the European Bank for Reconstruction and Development) and some bilateral aid agencies from member countries of the Organization for Economic Cooperation and Development (OECD) as well as non-OECD member country aid agencies are not represented. However, INDIX contains a comprehensive set of records from USAID, the Canadian International Development Agency (CIDA), and the quasi-governmental International Development Research Centre (IDRC) (personal communication, Jim Harold, CD-DIS Coordinator, USAID, Rosslyn, Virginia, April 17, 1996; personal communication, Maureen Ahern-Blais, Corporate Memory System, Canadian International Development Agency, Ottawa, April 21, 1997). Hence, the data gleaned from Canadian and U.S. records are drawn from the entire population of Canadian and U.S. ODA activities since the early 1960s.

To identify all records containing gender and environmental objectives, multiple key word searches were conducted. Each search identified records containing references to "women"; "gender"; "mother"; or permutations of these words, and one or more terms from a list of over 50 different natural resource/environmental key words. Searches were conducted in English, Spanish, French and German, as INDIX contains unique records in all four languages. Duplicate records across the multiple searches were detected and eliminated from the analysis. Content analysis was performed on 1,371 records from an original pool of over 2,200 records. Of these records, 214 were eliminated because the records were contributed by multilateral donors or international organizations; the analysis here examines bilateral aid activities of OECD member countries only. Another 581 records were removed because they contained no natural resource management or environmental references, despite containing at least one natural resource/environmental search term. For example, a record containing the word, "health" was retained for further analysis when it was associated with an environmental media, such as water, or an environmental concern, such as sanitation. Not all records dealing with health, nutrition, population, and agriculture were retained for data analysis, despite reasonable arguments that agriculture is a type of natural resource management and that health, family planning, and nutrition are synonymous with environmental health. Rather than identifying all activities that might be broadly defined as environmental or having an impact on the environment, record selection relied on the contributors' articulation of project objectives, inputs, and intended outputs and outcomes. When key words had a clear environmental and/or natural resource management connotation, the record was retained for data analysis. A collection of 576 records remained from the pool of 1,371 unique records. Of these 576 records, 453 or 78.6 percent were contributed by Canadian and U.S. aid agencies. 4

Information gleaned from records included classification of activity by: title; sector; environmental/natural resource management program area; donor agency; recipient country/region; life of project; and budget. In addition, each record was assigned three separate ranks to differentiate activities by their gender content; environmental content; and integration of gender and environmental content. Criteria for the three rankings are discussed immediately below.

All records retained for data analysis portrayed women as beneficiaries of aid or participants in aid. In this study, no attempt was made to determine whether women were active agents in aid planning and implementation. 5 Information contained in the database did not permit this type of inquiry. However, there was sufficient information to assign ranks to records based on their gender and environmental content. A three-point scale was adopted for both a gender ranking and an environmental ranking; a two-point scale was devised for an integrated gender-environmental ranking. The gender ranking indicated whether gender-related references were, according to the contributor's project description, primary activities (a rank of "3"); one among a group of two or more activities (a rank of "2"); or incidental or unspecified activities (a rank of "1"). Often, records assigned a gender rank of "1" mentioned women as generic beneficiaries, e.g., "the project is expected to improve women's lives" or the term, "women" was mentioned in a list of project descriptor terms but specific benefits to women or women's role in the project was not specified. A record received a rank of "2" when gender concerns were mentioned as project objectives or activities, but other project objectives or activities made no reference to gender. For example, a record received a gender rank of "2" when "educational training for women" was one among two activities, but the other activity did not specify women as participants or beneficiaries. A record dedicated primarily or entirely to a gender-related activity received the highest ranking ("3"), e.g., a microenterprise project that provided training to women, exclusively.

Like the gender ranking, an environmental ranking was assigned to each record based on relevant content. A score of "3" indicated that environmental/natural resource management components were primary activities, while a score of "1" indicated that environmental and/or natural resource aspects were incidental or undefined.

Finally, records received a rank based on whether or not gender and environmental project objectives were linked. A multisectoral project dealing with, say, literacy, forestry and agriculture might contain a literacy activity for women and a forestry management activity, but the contributor may not formally link the two activities. In this case, the project receives a rank of "1" indicating no linkage between gender and environmental components. To receive a "2," the link between the gender activity and the environmental/natural resource management activity must be stated unambiguously e.g., "women will receive training in forestry management." The strictness of the criteria for a "2" means that many records with substantial gender content or substantial environmental content receive a "1" because the records do not specify how women participate in or benefit from the environmental or resource management component of the project.

Frequency Of Gender And Environmental Aid

Many donors began reporting their WID and GAD spending in the last five to ten years. However, donors have not published data on the portion of their gender aid that also deals with environmental concerns and vice versa. The quality and specificity of budgetary data in the INDIX database is no improvement over the indicators submitted by the donors to OECD. Among INDIX records where gender or environmental concerns are two among multiple sub-activities, expenditures are not disaggregated by sub-activity and hence it is impossible to tease out the portion of the budget that is dedicated to WAE. INDIX's budgetary data on gender and environmental aid is useful only when the entire activity is devoted to gender and environmental concerns.

As an indicator of donors' commitment to WAE, (and as a substitute for data on donors' spending on WAE), donors' new aid activities containing WAE objectives were tallied each year. A count of new projects is an imperfect indicator of donors' commitments because it provides no information about capital and human resource investments. It is conceivable that expenditures during a year when few aid activities were initiated may actually exceed expenditures during a year when many activities were initiated. To help establish whether a count of new WAE projects is a salient measure of donors' commitment to WAE, the sum of new projects initiated each year was compared against donors' annual WID expenditures. Figure 1 exhibits annual OECD expenditures on WID (stippled line) against a count of new WAE activities initiated each year by OECD member countries (solid line). The first year when OECD published donors' expenditures on WID projects was 1988.

Figure 1 exhibits WAE activities beginning after 1982. (In the INDIX database, 78.5 percent of OECD members' WAE activities were initiated between 1982 and 1995). A period of rapid growth in new WAE activities commenced in the late 1980s and peaked with over 50 new activities in 1991 and 1992. New WAE activities declined sharply in 1993. 6 This drop-off in new activities mirrors a decline in OECD spending on WID projects. Numbers of new WAE projects recovers the following year and WID expenditures jump sharply. In 1995, new WAE projects and OECD spending on WID decline again. Between 1992 and 1993, OECD spending on WID fell 27 percent while the number of new gender and environmental aid projects fell 37 percent. OECD expenditures on WID fell 20 percent between 1994 and 1995. During that same period, new WAE activities fell by 30 percent. The drop-off in new WAE projects in 1993 and 1995 reflects cut backs in overall aid expenditures by OECD countries. Between 1992 and 1993, real aid disbursements fell among 18 of 21 OECD members (OECD, 1995, p. 2). While real spending levels increased for 13 members in 1994 (OECD, 1996, p. 89), disbursements slumped for 15 members in 1995 (OECD, 1997a, p. 95). Trends in OECD members' WID disbursements as well as indicators of members' support for WAE mirrors the donors' overall ODA disbursement trend. As revealed in Figure 1, WID aid (measured in current dollars) and WAE (measured as the number of new projects sponsored) decreased in 1993, recovered in 1994, and fell again in 1995.

OECD's data on members' WID disbursements must be interpreted with caution because the overall upward trend in members' WID spending probably reflects the gradual rise in the number of countries reporting their WID expenditures (OECD, 1996, p. 97). The upward trend in new WAE projects may be affected by the same phenomenon: over time, more donors contribute records to the INDIX database, but donors are not necessarily submitting complete sets of records from earlier years. Ambiguity in interpreting donors' support for WAE is mitigated by examining only Canadian and U.S. WAE activities because both nations' have contributed historically complete ODA records to INDIX.

Canadian And U.S. Gender And Environmental Aid

Among bilateral aid agencies contributing to the INDIX database, between 1969 (the first year when a WAE activity is registered in the database) and 1995, the United States and Canada sponsored around 79 percent of all new bilateral donors' WAE activities (Figure 2). 7 The remaining 21 percent was sponsored by bilateral aid agencies in over one dozen countries. In this latter group, northern European donors were the most active WAE sponsors.

Perhaps because of the comprehensiveness of North American aid agencies' contribution to INDIX, Canada and the United States emerge as the leading donors in terms of new WAE projects initiated between 1969 and 1995. Since Canadian and U.S. records constitute a relatively large fraction of the overall population of records in INDIX (over 21,000 records or 12.7 percent of the total) one should conclude that these two countries are the principal bilateral sponsors of gender and environmental assistance in the context of this database, only. This precaution notwithstanding, the United States is among the world's four most generous sponsors of ODA in real dollar terms, hence capable of drawing from a relatively large resource base to support WAE activities. 8 Support for women in development and for environmental protection are two among six priority programs of the Canadian aid mission (OECD, 1996, p. 106). The sheer number of activities sponsored by Canada reflects the fact that both Canada's official aid agency, CIDA, as well as the quasi-governmental IDRC are WAE sponsors.

Canada's and the United States' sponsorship of new WAE activities (stippled line) is exhibited against OECD members' new WAE activities (solid line) in Figure 3.

From 1975 to 1985, Canada and the United States account for the vast majority of OECD members' support for WAE. However, this trend may reflect the fact that other donors did not contribute complete historical records to INDIX. Both the OECD and North American trend lines assume the same steep upward slope in the late 1980s, peaking in the early 1990s, and falling, recovering, and falling again in the mid-1990s. In recent years, the drop-off in new Canadian and U.S. WAE activities reflects the overall decline in aid spending by these nations. U.S. ODA disbursements fell 19 percent in real terms from 1992 to 1993 and another 27.6 percent between 1994 and 1995 (OECD, 1995, p. 104; OECD, 1997a, pp. 172, A11-A12). Canadian ODA remained stable in real terms between 1992 and 1993 (OECD, 1995, pp. 92-93) but fell 9.2 percent in real terms between 1994 and 1995 (OECD, 1997a, p. 132).

Between 1969 and 1995, countries in the Africa-Near East region, especially sub-Saharan African countries, hosted the largest number of Canadian and U.S.-sponsored WAE activities -- around 43 percent of all such projects (Figure 4).

Asian and Latin American countries hosted approximately 26.5 and 24.5 percent, respectively. The countries hosting the largest number of Canadian and U.S. WAE projects were India and Indonesia, with 19 and 18 activities respectively, followed by China with 14 activities.

Sixty-four percent of all Canadian and U.S. activities containing gender and environmental objectives were dedicated to four sectors: agriculture, health, forestry, and energy (Figure 5).

The vast majority of projects in the health sector were water and sanitation improvement activities. Donor support for water and sanitation, strong in the 1980s, remained vigorous in the 1990s -- a reminder of the development community's unfinished agenda from the International Drinking Water and Sanitation Decade (1981-1990) (see also, OECD, 1996, p. 98). Water and sanitation activities were the dominant form of WAE assistance directed to urban and peri-urban areas. But, rural areas were the setting for over 75 percent of all WAE assistance. This finding was expected, because historically, rural areas have been the most popular destinations for environmental ODA (see, e.g., USAID, 1989, p. 27) and the largest regional recipient of WAE assistance, Africa, is also the least urbanized. Figure 5 understates the role of forests and forestry as key constituents of gender and environmental aid projects. When forestry-related components of projects in other sectors (e.g., agroforestry activities in the agriculture sector and fuelwood activities in the energy sector) are added to the forest management sector, 57 Canadian and U.S. WAE projects (12.6 percent of all Canadian and U.S. WAE projects) contain a forest management dimension. 8 The total number of new projects dealing with water, sanitation, and forestry increased every year through 1991, but the surge in support for these activities is tempered by a lack of integration between projects' gender and environmental objectives.

Gender And Environmental Objectives In Canadian And U.S. Aid

Canadian and U.S. aid agencies do not disaggregate gender-related aid data by project sector nor sectoral aid data by its gender content. Using the INDIX database, it is easier to gain insight into the former than the latter because searching and screening all records with gender content is less resource intensive than a comparable review of all records with environmental content. Human and financial resource constraints precluded a review of all Canadian and U.S. records containing environmental/natural resource terms (in excess of 7,000). However, the over 2,100 Canadian and U.S. gender-related records were reviewed for their environmental content. A relatively large portion (around 19 percent) of North American WID projects contained references to environmental protection or natural resources management (Figure 6).

That roughly one in five WID projects also contains environmental objectives is logical since the donors have disbursed relatively large volumes of WID to women in agrarian-based societies and to women in natural resource management occupations. Of the two North American donors, USAID has stated explicitly that improving environmental management is a central objective of its WID program (USAID, 1993).

Despite the relatively large fraction of WID projects that also mentioned an environmental objective or activity, relatively few of these projects were dedicated principally to either gender or environmental concerns. Recall, to receive the highest environmental ranking, an activity must be primarily or entirely dedicated to environmental or natural resource management objectives. Similarly, only projects primarily or entirely dedicated to gender-related objectives receive the highest gender ranking. While almost one in five Canadian and U.S. WAE activities was primarily or entirely dedicated to environmental activities, fewer than one in ten activities was dedicated primarily or entirely to gender concerns. The latter finding is consistent with OECD indicators of members' sponsorship of aid dedicated entirely to gender objectives (so-called, "WID-specific" aid) versus aid that includes gender as only one among two or more project concerns. Each year between 1990 and 1995, an average of 12.6 percent of OECD members' WID spending was for WID-specific projects (OECD, 1997b). In the remainder of projects, WID was one among two or more project objectives. Donors' tendency to support WID or GAD as a project component is consistent with the spirit of international declarations like Agenda 21 and the Beijing Platform for Action. Both statements summon donors to integrate gender considerations into aid activities, but do not demand an increase in the number of activities or amount of spending dedicated exclusively to gender concerns.

Linkage of Gender and Environmental Aid Objectives

Integrated WAE activities explicitly link gender and environmental objectives. Their records state plainly whether women are beneficiaries or participants in the environmental component of the project. The number of new Canadian and U.S. WAE activities containing integrated gender and environmental objectives increased after 1985 and climbed even higher in the mid-1990s. These two nations sponsored an average of six new activities each year between 1986 and 1995 versus an average of three projects per year in the preceding decade. However, during the period 1986-1995, integrated gender and environmental activities did not increase substantially as a percentage of all activities containing gender and environmental objectives. Between 1986 and 1995, 18.7 percent of all new Canadian and U.S. WAE activities were of the integrated variety. In the preceding ten year period (1976-1985), 18.8 percent of all new activities were integrated (Figure 7).

Despite increased international awareness of gender and environmental dependencies, integrated gender and environmental aid as a percentage of all aid containing gender and environmental objectives remained virtually constant between the mid-1970s and the mid-1990s. It is conceivable that funding for integrated gender and environmental activities increased as a percentage of Canadian and U.S. aid expenditures. However, since very few activities in the INDIX database were dedicated exclusively to gender and environmental objectives and records do not contain data on expenditures disaggregated by individual project components, it is impractical to monitor funding levels for integrated gender and environmental projects. In all likelihood, total  expenditures by Canadian and U.S. aid agencies on integrated WAE did  increase over time, as the number of new WAE projects with integrated gender and environmental objectives doubled from the 1976-1985 period to the 1986-1995 period. However, the increasing numbers of integrated WAE projects in the late 1980s and early 1990s are occurring against a background of increasing numbers of new activities with non-integrated gender and environmental objectives.

Analysis

The late 1980s and early 1990s witnessed an increase in the number of new aid activities containing gender and environmental objectives. However, growth in support for these activities reflected donors' elevated environmental concerns, generally, rather than donors' enlightened understanding of gender-environmental dependencies. The gender and environmental imperative, so powerfully enunciated at the diplomatic level did not create a sea change in the design of development assistance. In its evaluation of member countries' WID policies and programs, OECD found that gender-related objectives were often "added-on" to projects in order to pass political muster but often WID objectives and activities were not integrated into project inputs nor well-documented in project outcomes (OECD, 1994, appendix I-5). OECD's assessment was corroborated here in the context of WAE activities. While gender and environmental objectives were increasingly common in Canadian and U.S. aid project descriptions, the percentage of WAE abstracts that contained integrated gender and environmental objectives remained virtually flat from the mid-1970s through the mid-1990s.

The findings presented here must be treated with caution because of the imperfect quality of the data published by bilateral donors and limitations of the INDIX database. Among other shortcomings, donors' spending on WAE is difficult to gauge. While Canadian and U.S. contributions to INDIX are comprehensive, contributions from other donors may be incomplete. Also, project records, which relate information about project intentions, may deviate from project implementation and outcomes. A more complete understanding of the aims, administration, and impact of gender and environmental aid demands evaluation of, among other things, project reports, independent program audits, and information from interviews with beneficiaries, implementors, and aid officials. The challenge posed by this research agenda partly explains why much of the literature on WID/GAD is of the case study variety. The problems encountered in this analysis are similar to those hampering OECD's assessments of member countries' commitments to WID/GAD. OECD has been precluded from publishing data on individual members' disbursements for WID/GAD because some members' data are missing or their data sets are incomparable (OECD, 1994, p. 24).

In its 1994 assessment of OECD members' WID policies and programs, OECD found that WID assistance "may be overburdening women without increasing their ability to make decisions and that projects 'give the appearance' of benefiting women when, in fact, the reverse may be true" (OECD, 1994, p. III-2). The assessment offered here does not speculate about WAE outcomes or whether or not WAE assistance serves women's interests. Rather, the objective is to determine whether donors' political commitments to the gender-environment nexus are translated into policy and program action. Until donors develop better indicators of WID and WAE project inputs, outputs, and outcomes, and incorporate these indicators into their regular project assessment criteria, most assessments of WAE will depend on case analyses, with their limited insights to general trends. A richer understanding of gender and environmental aid must also include an assessment of aid that is not explicitly gender and environmental in content. It is conceivable that aid not expressly designed to promote gender or environmental values may nevertheless greatly affect women's welfare and/or environmental quality. For example, a structural adjustment loan with conditionality may have more profound effects on women's lives or on the environment than any number of smaller, targeted gender and environmental aid activities.

Conclusion

This article examines trends in gender and environmental aid based on donors' articulation of aid project objectives and inputs. Some commentators have observed a gap in the development community's rhetorical support for WID and programmatic support for WID. Focusing on Canadian and U.S. gender and environmental assistance, this study identifies an early moment in the development planning process when legal and normative commitments become dislocated from project design. At the project description stage, donors do not integrate gender and environmental objectives. From the mid-1970s to the mid-1990s, Canada and the United States sponsored ever increasing numbers of integrated gender and environmental activities. But during this period, there was no appreciable change in integrated activities as a percentage of all activities containing gender and environmental objectives. Additional research is required to determine whether and to what extent donors integrate gender and environmental aims in project implementation, outputs, and outcomes. Donors must participate in this research endeavor because comparable, multivariate, longitudinal data sets on aid activities, especially at the subsectoral level are not widely available. As a first step, donors must develop performance indicators to track gender-related objectives of aid activities. Gender and environmental concerns crosscut a range of development objectives. Hence, monitoring the performance of gender and environmental aid requires a departure from sector-specific data collection. Donors are beginning to respond to these needs. For example, in cooperation with USAID African missions, USAID's regional bureau for Africa is developing a new set of indicators to track women's participation in natural resource management projects (personal communication, Michael McGahuey, Natural Resources and Sustainable Agriculture Advisor, USAID, Washington, D.C., July 24, 1996). At the bilateral level, the infrastructure is in place for donors to coordinate data collection and dissemination: OECD members issue a compendium of data on their ODA activities, every year. The donors are obliged to advance this research agenda. As specified in Agenda 21, donors have already agreed to monitor their support for gender and the environment. In addition, they have declared gender empowerment and environmental protection to be interdependent concerns and vital to sustainable development.

References


Note *: This study was supported by a grant from the Indiana University School of Public and Environmental Affairs' Workshop on Sustainable Development. Research assistance provided by Kathleen Dowd and Morgan Jencius is gratefully acknowledged. The author is indebted to Constance Becker for her helpful inputs. Don McCubbin provided useful recommendations for revisions to an earlier draft of this paper. Back.

Note 1: International organizations whose missions deal specifically with the cross-section of gender and environmental concerns include the Committee of Senior Women Advisers on Sustainable Development (created by UNEP in 1986) and the Working Group on Women, the Environment and Sustainable Development (created by IUCN in 1987). Of more recent vintage is the Women's Environment and Development Organization (WEDO), founded in 1990. WEDO has participated in major international conferences to promote gender and environmental concerns, including the 1992 Earth Summit and 1995 World Conference on Women. Back.

Note 2: Canada and the United States have contributed an historically complete set of records to the International Network for Development Information Exchange (INDIX). The INDIX database is representative of the dollar value of all U.S. official development assistance, beginning with the first ODA activities sponsored by the United States Agency for International Development in 1961 (personal communication, Jim Harold, CD-DIS Coordinator, USAID, Rosslyn, Virginia, April 17, 1996). Canada's contribution to INDIX is comprehensive with the exception of activities sponsored under the Canada Fund. The budgets of Canada Fund projects seldom exceed CAN$ 1,000 or CAN$ 2,000 per project (personal communication, Maureen Ahern-Blais, Corporate Memory System, Canadian International Development Agency, Ottawa, April 21, 1997). Back.

Note 3: The earliest record in the database describes an aid activity initiated in 1930. However, only three records in the INDIX database concern activities commencing before 1952. Back.

Note 4: When WAE activities of both bilateral and multilateral donors are extracted from the INDIX database, Canada and the United States account for around 58 percent of all new projects sponsored during the 1969-1995 period. Back.

Note 5: From a feminist perspective, a trend indicating women's increasing involvement in aid decision-making does not necessarily connote progress. Some might argue that poor women who participate in aid planning and delivery are merely advancing the aims of the dominant, patriarchal development model. Back.

Note 6: While the number of new WAE activities declined sharply between 1992 and 1993, there were many on-going projects in 1993 initiated in previous years (155 such projects). Back.

Note 7: Until the late 1980s, the United States sponsored more ODA (in real dollars) on an annual basis than any other OECD member nation. The United States' number one ranking slipped in 1989; by 1995, each of three nations' annual ODA volume exceeded that of the United States': Japan, Germany and France (OECD, 1996, p. 97). The United States regained its number one ranking in 1996. That year, its ODA, which does not include food aid, Peace Corps funding, aid to the Newly Independent States, or military aid approached US$ 9.1 billion. Some Japanese commentators were skeptical about the jump in U.S. foreign aid in 1996 since the fiscal year 1996 U.S. ODA budget included behind-schedule payments to multilateral development organizations for 1995 and two years worth of grant disbursements to Israel (Altbach, 1997, pp. 5-9). Back.

Note 8: Not all experts agree that the development community's keen interest in gender and forestry is appropriate. Paolisso (1995), for one, argues that researchers' preoccupation with gender, deforestation, and fuelwood scarcity has detracted attention from gender and environmental interactions in the agriculture, non-agriculture income-earning, and health sectors. Back.