Committed to health for all? How the G7/G8 rate.

The G7/G8 group of nations dominate the world political and economic order. This article reports selected results from an investigation of the health implications of commitments made at the 1999, 2000 and 2001 Summits of the G7/G8, with special reference to the developing world. We emphasize commitments that relate to the socioeconomic determinants of health (primarily to reducing poverty and economic insecurity) and to the ability of national governments to make necessary basic investments in health systems, education and nutrition. We conclude that without a stronger commitment to redistributive policy measures on the part of the G7/G8, historic commitments on the part of the international community to providing health for all are likely not to be fulfilled.

Introduction and rationale

In 1978, building on the 1948 Universal Declaration of Human Rights, a United Nations conference proposed the goal of health for all by the year 2000 (WHO, 1978). In 2003, only limited progress has been made toward that goal. This article assesses the reasons for that lack of progress, with specific reference to the dominant role played by the G8 (Group of 8) nations in the international economic and political order. In other words, it provides a “report card” on key health impacts and implications of G8 policies, with particular reference to effects in countries outside the industrialized world that account for roughly five-sixths of the world's population.

The G8 was formed in 1975 after the “oil crisis” provided an early warning of the dangers of economic interconnectedness. The six countries originally included were France, the United States, Britain, Germany, Italy and Japan. Canada joined in 1976; the European Community (now the European Union) joined in 1977, but does not have the same status as national governments. Russia achieved partial membership in the group in 1998, and full membership as of 2003; thus, the G7 is now the G8. The G8 account for 46.6 percent of global GDP and 46.8 percent of global exports (International Monetary Fund, 2003, Table A). Perhaps more importantly, the G7 countries dominate World Bank and IMF decision making, and wield considerable power in the World Trade Organization (WTO) because the size of their markets and access to specialized expertise provide them with formidable bargaining advantages with respect to countries of the developing world.

Access to health care is only one factor amongst many affecting the health status of a population (Evans and Stoddart, 1990; Diderichsen, Evans, & Whitehead, 2001). For much of the world's population, ability to lead a healthy life is limited by direct and indirect effects of poverty. Almost half the world's people live on an income of $2 per day or less (World Bank, 2001, pp. 36–38). This figure has been criticized on methodological grounds as a substantial underestimate of the extent of absolute poverty (Reddy & Pogge, 2003), but clearly it describes complex and health-destructive vulnerabilities (Narayan, Chambers, Shah, & Petesch, 2000; Diderichsen, Evans, & Whitehead, 2001). Ill health not only results from poverty, but also can limit the ability of individuals, households and entire societies to escape from poverty. “[F]or the poor their body is often their only asset, and when the body is weakened through hunger, illness and accidents, an entire family can plunge into destitution” (Narayan, 2001, p. 15; see generally Narayan, Chambers, Shah, & Petesch, 2000). The potential contribution to economic development of low-cost interventions to improve health was a central theme of the work of the World Health Organization's Commission on Macroeconomics and Health (2001). Conversely, the impacts of HIV/AIDS and malaria provide especially dramatic, large-scale illustrations of the economic damage that can result from poor health (Haacker, 2002; Sachs & Malaney, 2002).

“Globalization” adds a further dimension to the challenge of providing health for all. The term is a convenient way of describing the growing interconnectedness of the world's economies and societies. Some observers regard globalization as a “process of closer interaction of human activities across a range of spheres including economic, political, social and cultural…[and] occurring along three dimensions: spatial, temporal and cognitive” (Lee, 2000, p. 30). Such broad descriptive definitions, while accurate, fail to take into account the fact that the primary influences of globalization on the social determinants of health are changes in patterns of international trade and investment, along with the underlying technological developments (Labonte & Torgerson, 2003). The economic manifestations of globalization, defined in this way, affect health by changing exposures to health risks, by changing the characteristics of health systems, and by affecting the structure of household, community and national economies (Zielinski Gutiérrez, & Kendall, 2000; Butler, Douglas, & McMichael, 2001; Woodward, Drager, Beaglehole, & Lipson, 2001; Labonte & Torgerson, 2003). Perhaps most dramatically, financial crises arising from the rapid flow of capital across national borders can plunge millions of people into poverty, while health and social service spending decreases (Hotchkiss & Jacobalis, 1999; Chavez & Cordero, 2001; O’Brien, 2002; Kim et al., 2003). Over a longer time scale, technological and institutional change have resulted in the emergence of a genuinely global labour market, within which there are clear winners and losers (World Bank, 1995).

Methodology

In preparing the “report card”, we pursued two lines of inquiry. First, we considered the extent to which G7 countries have lived up to their Summit commitments. Second, we considered the adequacy of those commitments when measured against relevant population health challenges. In other words, we were concerned both with the policy effectiveness of the G7, when assessed with reference to their stated intentions, and with the substantive impact of their policies, when assessed with reference to a large and growing body of research on the determinants of health in the developing world (for overviews see Evans, Whitehead, Diderichsen, Bhuiya, & Wirth, 2001; WHO, 2002c).

These lines of inquiry are analytically distinct, but also related. Notably, in 2000 the G7 committed themselves “to the agreed international development goals (IDGs), including the overarching objective of reducing the share of the world's population living in extreme poverty to half its 1990 level by 2015” (G8, 2000, p. 13). These IDGs were published in 2000 as a joint effort of the UN, the OECD, the World Bank and the International Monetary Fund (2001), with the comment that: “Each of the seven goals addresses an aspect of poverty. They should be viewed together because they are mutually reinforcing” (International Monetary Fund, OECD, United Nations & World Bank Group, 2000, p. 4). A resolution (A/RES/55/2) of the UN General Assembly in 2000 incorporated several of the IDGs, as well as other objectives that are equally ambitious, and also directly related to health (Table 1 ), to generate a list that is now widely referred to as the Millennium Development Goals (MDGs). Because the G7 countries, both individually (the MDGs) and collectively (the IDGs), have committed themselves to support a range of goals that are related to improving global health, it is appropriate to assess their Summit undertakings in light of that position.

Table 1

The international development goals and the millennium development goals compared

	International development goals	Millennium development goals (Goals 1–7)
1	Reduce the proportion of people living in extreme poverty (less than US $1/day) by 2015	Goal 1: Eradicate extreme poverty and hunger
		Target 1: Halve, between 1990 and 2015, the proportion of people whose income is less than 1$ a day
		Target 2: Halve, between 1990 and 2015, the proportion of people who suffer from hunger
2	Enrol all children in primary school by 2015	Goal 2: Achieve universal primary education
		Target 3: Ensure that, by 2015, children everywhere, boys and girls alike, will be able to complete a full course of primary education
3	Eliminate gender disparities in primary and secondary education by 2005	Goal 3: Promote gender equality and empower women
		Target 4: Eliminate gender disparity in primary and secondary education preferably by 2005 and to all levels of education no later than 2015
4	Reduce infant and child (under-5) mortality rates by two-thirds between 1990 and 2015	Goal 4: Reduce child mortality
		Target 5: Reduce by two-thirds, between 1990 and 2015, the under-5 mortality rate
5	Reduce maternal mortality ratios by three-quarters between 1990 and 2015	Goal 5: Improve maternal health
		Target 6: Reduce by three-quarters, between 1990 and 2015, the maternal mortality ratio
6	Provide access for all who need reproductive health services by 2015	Goal 6: Combat HIV/AIDS, malaria and other diseases
		Target 7: Have halted by 2015, and begun to reverse, the spread of HIV/AIDS
		Target 8: Have halted by 2015, and begun to reverse, the incidence of malaria and other major diseases
7	Implement national strategies for sustainable development by 2005 so as to reverse the loss of environmental resources by 2015	Goal 7: Ensure environmental sustainability
		Target 9: Integrate the principles of sustainable development into country policies and programs and reverse the loss of environmental resources
		Target 10: Halve, by 2015, the proportion of people without sustainable access to safe drinking water
		Target 11: By 2020, to have achieved a significant improvement in the lives of at least 100 million slum dwellers

Sources: International Monetary Fund, OECD, United Nations and World Bank Group (2000) for International Development Goals; Devarajan, Miller and Swanson (2002, pp. 34-35) for Millennium Development Goals.

We analysed commitments made at the three Summits preceding the start of our research: Cologne (1999), Okinawa (2000) and Genoa (2001). In addition, we limited our focus to the G7 countries, given Russia's newer membership and transitional situation. Early on, we confronted a fundamental choice. We could restrict our assessment of G7 performance to a few narrowly specific commitments, ideally involving dichotomous end points. Alternatively, we could err on the side of inclusiveness, starting from an inventory of statements many of which were not readily amenable to quantitative assessment of subsequent performance. We followed the latter course, for two reasons. First, the complexity of the determinants of health and the long period of time that sometimes elapses between policy change and health impact mean that it is necessary to assess patterns of policy commitment and implementation over time. Second, as we have noted, the fact that the G7 countries have lived up to the specific terms of a commitment made at the Summits says nothing about the adequacy of the response described in that commitment, or about its consistency with other policy objectives such as those embodied in the IDGs/MDGs.

We began with key texts from the 1999–2001 Summits, primarily the formal statements issued at the start of Summits and the Communiqués issued at their conclusion.1 Three individuals, each familiar with population health determinants (the two social scientist authors of this article and a post-doctoral researcher with training in health sociology) read these texts and independently identified statements with potential significance for population health using 13 subject matter headings. Summit commitments were then classified into one (or, sometimes, more) of three columns in a matrix:

Commitments that could be assessed in quantitative or dichotomous terms (e.g. expenditure figures, actions taken or not).

Commitments about which data exist, but where assessment would be primarily qualitative or narrative (e.g. commitments using language such as “improve” or “increase”).

Commitments reflecting a pre-existing, but contestable or problematic position on appropriate social and economic policies (e.g. the presumption that integrating developing countries into the global economy represents the only appropriate development strategy).

The matrix is available in full on the Internet (http://www.spheru.ca/www/html/Research/Research_globalization.htm). Many commitments spanned more than one column; some also related to more than one subject matter heading. This article summarizes our findings with respect to health systems and to three other areas that are especially important influences on determinants of health: macroeconomic policy, nutrition, and education.

We then surveyed what turned out to be a massive literature in order to assess G7 performance with respect to Summit commitments, and the health implications of the policies reflected by those commitments. The literature comprised: quantitative data assembled by organizations including the World Bank, OECD, and several United Nations agencies; an extensive body of research by civil society organizations (CSOs) such as Oxfam and Jubilee Research; and an expanding research literature on determinants of population health in the developing world. These categories tend to overlap, in particular as the work of key CSO-affiliated researchers is published by “mainstream” agencies (Third World Network, 2001; Pettifor & Greenhill, 2002; Watkins, 2003). We carried out our own calculations and policy evaluations using these data, but did not check on their accuracy beyond the identification of clear inadequacies in the data as published. It must be noted that at least as many questions have been raised about the accuracy of data generated by agencies such as the World Bank and the World Health Organization (e.g. Musgrove, 2003; Reddy & Pogge, 2003) as about the research and policy recommendations of CSOs.

Table 2 shows in extremely condensed form our inventory of “promises kept, promises broken” with respect to the subject areas included in this article, along with a one-sentence commentary. We explain these findings in the sections that follow, after which we discuss the links between the policies that they document and a more general conception of the development process that appears to underpin and unify the positions taken by the G7. That development model incorporates a powerful presumption against substantial international redistribution of resources, but our findings indicate that genuinely redistributive policies are imperative in order to improve the health of populations in the developing world.

Table 2

‘Promises kept (✓), promises broken (×)

Commitment		Assessment
Support for international development goals, “including the overarching objective of reducing the share of the world's population living in extreme poverty to half its 1990 level by 2015”	×	Many IDG targets for 2015 will almost certainly not be achieved
Provision of debt relief under Heavily Indebted Poor Countries (HIPC) initiative	✓	Debt relief now being provided, but amount is often inadequate; Poverty Reduction Strategy Paper process seriously flawed; many heavily indebted countries not covered
Create the Global Fund to Fight AIDS, Tuberculosis and Malaria	✓	Current financial pledges far below need identified by Commission on Macroeconomics and Health
By 2010: reducing the number of HIV/AIDS-infected young people by 25 percent, reducing TB deaths and prevalence of the disease by 50 percent, and reducing the burden of disease associated with malaria by 50 percent	×	Resources almost certainly inadequate
Non-specific commitment to strong national health systems	×	Official development assistance (ODA) for health from all industrialized countries: $6 billion/year (less than one-quarter the needed amount as identified by Commission on Macroeconomics and Health); during three Summit years of study, ODA from G7 countries actually declined slightly
Recognize need for “flexibility” with respect to intellectual property protection in order to ensure availability of essential drugs	✓	Agreement now reached on interpretation of intellectual property provisions of WTO Agreement, but its significance remains uncertain
Non-specific commitments to supporting agriculture through ODA as an element of poverty reduction, to “target the most food-insecure regions, particularly Sub-Saharan Africa and South Asia”	×	Few specifics, and no clear commitment to IDG of reducing underweight among children; recent slow progress in reducing undernutrition now reversed
Heavy emphasis on promoting biotechnology to increase agricultural productivity	✓	Appropriateness of such ‘solutions’ questionable
Clear support for Dakar Framework goals re: improving access to education by 2015	×	Strong evidence that these goals will not be achieved

Macroeconomic policy, structural adjustment and debt relief

Because of the numerous causal pathways and feedback loops linking poverty and ill health (Narayan, Chambers, Shah, & Petesch, 2000; Commission on Macroeconomics and Health, 2001; Diderichsen, Evans, & Whitehead, 2001), we emphasize in this article the impacts of G7 commitments that operate on domestic macroeconomic and social policy. Those commitments, in turn, must be analysed with reference to the issue of developing country debt because, “despite repeated rescheduling of debt by creditor countries, developing countries continue[d] to pay out more each year in debt service than the actual amounts they receive in official development assistance—ODA” between 1986 and 1996 (Cheru, 1999, ¶ 10). The net outflow of funds became even more significant in the years that followed, as a result of the financial crisis in south Asia (Pettifor & Greenhill, 2002; United Nations, 2002). Debt service obligations represent the most fundamental constraint on many developing countries’ ability to meet basic health-related needs—a constraint the significance of which has been recognized at least since the mid-1980s (World Commission on Environment and Development, 1987, pp. 67–75; Ramphal, 1999).

The Heavily Indebted Poor Countries (HIPC) initiative, announced by the World Bank and IMF in 1996 and “enhanced” in 1999, has become the centrepiece of G7 debt relief efforts (G8, 2001, ¶ 7,15). This is true even though the 41 HIPC-eligible countries, 33 of which are in sub-Saharan Africa, account for only 10 percent of the developing world's debt (UNRISD, 2000, p. 22), and HIPC's value in terms of poverty reduction is limited because a clear majority of the world's poor people live in countries that are not eligible for HIPC (Table 3 ). As of January 2003, 26 countries had reached their “decision point”—the point at which debt relief is approved—and were receiving debt service relief that will amount to $40.4 billion (World Bank, 2003a, Table 2). This is more than 70 percent of the total debt relief anticipated under the Initiative (World Bank, 2002b, p. 1). The G7, in other words, have lived up to their debt relief commitments as stated at the Summits.

Table 3

Poverty in non-HIPC countries

	Population (million)	Number of people living on <$1/day (million)	Number of people living on <$2/day (million)
Bangladesh	133.4	38.8	103.8
Brazil	172.6	20.0	45.7
China	1271.9	239.1	669.0
India	1033.4	456.8	890.7
Indonesia	213.6	27.6	139.9
Mexico	99.4	15.8	37.5
Nigeria	129.9	91.2	117.9
Pakistan	141.5	43.9	119.9
These 8 (non-HIPC) countries		933.2	2124.4
Entire world		1198.9	2801.0

Sources: World Bank, 2001; World Bank, 2003b.

However, such commitments may not be adequate when measured against the resources that will be required to achieve such objectives as the IDGs. Oxfam (2001, Fig. 1) has calculated that in 14 HIPC countries, annual debt servicing costs will exceed combined public spending on health and primary education even after the maximum debt relief available under HIPC is obtained. This is because the value of debt relief available under HIPC is currently determined based on a ratio of debt service costs to anticipated future export revenues; a country's debt load is considered “sustainable” if its net present value is less than 150 percent of annual export revenues. A more appropriate criterion for assessing sustainability would ensure that debt service costs did not compromise a country's ability to meet such objectives as the IDGs/MDGs (Greenhill, 2002; Greenhill & Sisti, 2003). Hanlon (2000), working backward from estimates of the expenditure that would be required to meet a list of targets similar to the MDGs, estimated that approximately US$600 billion (at current value) in debt cancellation would be necessary to ensure that debt repayment did not occur at the expense of essential social spending. This is an order of magnitude greater than the value of all debt relief to be provided under enhanced HIPC. Hanlon's estimates consider not only the HIPC countries, which he estimates will require debt relief worth $180 billion, but also many others. His calculations imply, for instance, debt relief of $24 billion for now-beleaguered Argentina (for which national poverty data are not even available), $116 billion for Indonesia, and $98 billion for India. A more cautious set of calculations, restricted to the HIPC countries, nevertheless reached the conclusion that meeting the MDGs in many countries would require not only complete cancellation of external debt but also substantial increases in revenues from ODA (Greenhill & Sisti, 2003).

Further problems arise because eligibility for HIPC is contingent on the recipient government's completion of a Poverty Reduction Strategy Paper (PRSP). PRSPs were launched by the World Bank and IMF in December 1999, as “a new approach to the challenge of reducing poverty in low-income countries based on country-owned poverty reduction strategies that would serve as a framework for development assistance” (International Development Association/IMF, 2002, p. 5). Although PRSPs ostensibly place poverty reduction at the centre of their analysis, direct parallels exist between the process of qualifying for debt relief through the preparation of a PRSP and earlier forms of conditionality (Cheru, 2001; International Monetary Fund, 2001, pp. 50–52; UNCTAD, 2002a, p. 191). In order to understand the significance of these parallels, some historical background is needed. In 1980, the World Bank initiated structural adjustment loans to help heavily indebted poor countries cope with the impact of the 1979–80 recession on their ability to service external debt. Structural adjustment became far more important after 1982, when the government of Mexico announced that it was prepared to default on billions of dollars in loans, primarily made by major US banks. The result was the first of a series of “debt crises”. Apprehensions about the stability of major banks in the industrialized world in the event of coordinated default led industrialized country governments, bilaterally and through the World Bank and the IMF, to provide new money for debt rescheduling.

However, the new money came with strings attached (conditionality): funds were made available only if the debtor country agreed to a relatively standard package of macroeconomic policies including reduced subsidies for basic items of consumption, the removal of trade and investment controls, and privatization of state-owned enterprises (Sparr, 1994; Dixon, Simon, & Närman, 1995; Milward, 2000). As early as 1987, a UNICEF-sponsored study indicated that a combination of global recession and the austerity measures adopted by national governments as the price of debt relief had the effect of reducing such basic indicators of child welfare as nutrition, immunization levels and education (Cornia, Jolly & Stewart, eds. (1987), Cornia, Jolly & Stewart, eds. (1988); see also Stewart, 1991). By the end of the 1980s, “cross-conditionality” that involved both the World Bank and the IMF (Walton, Sedden et al., 1994, p. 19) further ensured subordination of domestic policy goals to the imperative of fiscal restraint and the generation of export revenues sufficient to meet debt obligations. Among the consequences was reduced access to such services as health care and education as public expenditures were cut and user charges introduced (see e.g. Cheru, 1999; Cornia, Jolly & Stewart, eds. (1987), Cornia, Jolly & Stewart, eds. (1988); Schoepf, Schoepf, & Millen, 2000; Walton, Sedden et al., 1994; Yong Kim, Shakow, Bayona, Rhatigan, & Rubín de Celis, 2000).

The United Nations Development Program, in its assessment of the PRSP process, notes that advice on the requirement for a macroeconomic framework identifying fiscal and financing policies for poverty reduction is weak, contains many unexamined assumptions and does not adequately emphasize distributional impacts of macroeconomic policies (UNDP, 2002). The United Nations Conference on Trade and Development (UNCTAD, 2002b, p. 197) links the PRSP process with the inadequacy of overall levels of debt relief, noting that in order to ensure that a PRSP is perceived as “realistic”, countries like Uganda and Tanzania are still investing far less than the minimum amounts required for health and social programs. The World Health Organization goes further in analysing serious gaps in existing PRSPs with respect to health (WHO, 2002a). Among its major criticisms: PRSPs deal with ill health as a consequence of poverty, but do not reflect an understanding of its role as a cause of poverty, and thus are too willing to recommend cost recovery as a way of financing health care services for the poor. In addition, PRSPs do not deal with such important health system issues as expenditure levels well below the minimum needed to provide basic primary health care. As the next section of the article shows, this has been another neglected area in terms of G7 commitments.

Health and health systems

In 2000, the G7 committed themselves to an “ambitious agenda” of “deliver[ing] three critical UN targets” by 2010: reducing the number of HIV/AIDS-infected young people by 25 percent, reducing TB deaths and prevalence of the disease by 50 percent, and reducing the burden of disease associated with malaria by 50 percent (G8, 2000, ¶ 29). However, without major increases in the resources available for health care expenditure, it is unlikely that these targets can be met. Much the same is true for the health-related components of the IDGs and MDGs, even though major improvements in health could be achieved by way of relatively low-cost, low-technology interventions to prevent the spread of infectious disease and reduce the toll from diarrheal disease and childbirth (Spinaci & Heymann, 2001). The World Bank recently concluded, based on a scenario of 3.6 percent annual per capita income growth in the developing countries between 2005 and 2015 (which may well be optimistic) that South Asia was the only region likely to achieve the infant and child mortality reduction target specified in the IDGs (World Bank, 2002a, pp. 31–33).

The world's Least Developed Countries (LDCs)2 spend an average of just $11 per capita annually on health, including both public and private expenditures. For other low-income countries, average per capita expenditure on health is $25 (Global Forum on Health, 2002, p. 5). The Commission on Macroeconomics and Health (2001, p. 11) estimated the cost of a “set of essential interventions”, which would not need to be the same for each country, at $34 per capita per year. The report warned that: “If anything, we are on the low end of the range of estimates of the cost of such essential interventions.” As if to corroborate this observation, according to the former Director-General of the World Health Organization: “It is becoming clear that health systems which spend less than $60 or so per capita are not able to even deliver a reasonable minimum of services, even through extensive internal reform” (Brundtland, 2000).

The Commission on Macroeconomics and Health identified the need for “an additional $22 billion per year by 2007 and $31 billion per year by 2015” in grant financing for country-specific interventions against infectious diseases and nutritional deficiencies. Above and beyond these country-specific interventions, it called for additional grant funding of $5 billion by 2007 and $7 billion by 2015 for research and development on diseases of the poor and other public goods like epidemiological surveillance, for a total of $27 billion in 2007, rising to $38 billion in 2015. This estimate must be compared with total ODA for health that is now “on the order of $6 billion” (Commission on Macroeconomics and Health, 2001, p. 11).

In 2001, the G7 addressed three infectious diseases that are major killers in the developing world by establishing the Global Fund to Fight AIDS, Tuberculosis and Malaria (GFATM). They described its creation as fulfilling a pledge from the preceding year “to make a quantum leap in the fight against infectious diseases and to break the vicious cycle between disease and poverty” (G8, 2001, ¶ 15). Financial commitments from governments to date amount to $4.68 billion, with $1.6 billion paid to date and the balance payable at various dates as far away as 2008 (http://www.theglobalfund.org/en/funds_raised/pledges/, accessed November 27, 2003). Since pledges do not all cover the same period, direct comparisons must be made with caution. However, the gap between the lowest per capita contribution among the G7 ($1.57, from Japan) and the highest ($10.80 from France) indicates varying levels of enthusiasm for the Fund's activities.3 More importantly, financial commitments made to date are far below the amounts recommended by the Commission on Macroeconomics and Health (2001), which argued that GFATM will require $8 billion per year by 2007, and $12 billion per year by 2015, in order to provide adequate support for prevention and treatment. To put these amounts into perspective, $8 billion is about as much as Americans spend per year on cosmetics or bathroom renovations, and about one-sixth as much as Europeans spend on cigarettes (Scott, 2002; UNDP, 1998, p. 37). It is easy to dismiss such comparisons as polemical, but they serve a critically important purpose in comparing the discretionary consumption of the global few with the low cost of health improvements for the many.

The Commission on Macroeconomics and Health estimates assume that developing countries have functioning health care systems. However, in many countries, the more immediate problem is how to avoid collapse of existing health infrastructure because of such factors as constraints on government expenditure, the impact of HIV/AIDS and the emigration of health professionals (see e.g. Sanders, Dovlo, Meeus, & Lehmann, 2003) . Although the G8 stated in 2001 that “[s]trong national health systems will continue to play a key role in the delivery of effective prevention, treatment and care and in improving access to essential health services and commodities without discrimination” ([24]G8, 2001, ¶ 17), aid for all aspects of health system development accounts for just over 4 percent of total G7 ODA expenditure. To the extent that available data permit the calculation of trends, this proportion was stable or even declined during the 1990s (Table 4 ), albeit with wide variations between years and countries.4 Declining child vaccination coverage in all developing areas during the 1990s may illustrate the consequences. The decline in Africa is particularly troubling since almost 50 percent of African children are now not adequately vaccinated (Simms, Rowson, & Peattie, 2001; UNICEF, 2001, p. 89; WHO, 2002b). In addition, when governments lack the funds for minimally adequate health infrastructure, privatization of health services and the adoption of cost recovery measures tend to emerge as a superficially attractive, but highly inequitable alternative (Arhin-Tenkorang, 2000; Melgar, 1999; Schoepf, Schoepf, & Millen, 2000; Whitehead, Dahlgren, & Evans, 2001; Yong Kim, Shakow, Bayona, Rhatigan, & Rubín de Celis, 2000). The resulting “medical poverty trap” (Whitehead, Dahlgren, & Evans, 2001) may actually undermine the potential for future economic growth.

Table 4

Trends in aid to health as percentage of total G7 ODA

	1990/92 averagea	1996/98 averagea	1999b	2000b	2001b
Canada	3	3	1.8	2.6	4.3
France	3	4	4.5	4.4	6.0
Germany	1	5	4.0	3.2	3.3
Italy	5	4	7.2	7.7	4.7
Japan	1	2	2.7	2.9	2.8
UK	9	10	6.8	9.6	5.9
US	5	17	4.4	4.1	4.7
G7 average				4.3	4.3

Source: OECD, 2001, Table 2 (1990/92 and 1996/98 data); OECD, 2002, Tables 14 and 19 (2000 data); OECD, 2003, Tables 13, 15 and 19 (2001 data).

Because of data limitations, includes only bilateral aid.

Includes both bilateral and multilateral aid (contributions made by donor countries to the European Commission, the World Bank and regional development banks). Published data on the sectoral distribution of individual countries’ multilateral aid contributions are not available. Instead, we attributed multilateral aid contributions to specific sectors based on the following calculation: country specific percentage of total aid contributed through each of the three multilateral agencies (Regional Development Banks, World Bank, European Commission)×the percentage of aid provided to the specific sector by each of the multilateral agencies. The sum of these calculations was then added to that country's sector-specific bilateral contribution. There may be small margins of error; the OECD report from which our data were drawn (OECD, 2002) itself cautions that figures for the European Commission are “approximate.” Total 2001 ODA contributions are based on the same calculations, using data from OECD (2003). An even greater note of caution is expressed for 1999 multilateral estimates. We applied the same formula as for 2000 and 2001, but the percentage of European Commission aid contributions by sector is not available for 1999. We therefore used the percentages for 2000 as a rough approximation, but calculated G7 averages only in years for which data are more reliable (2000, 2001).

By 2001, controversy over the pricing of antiretroviral drugs for HIV/AIDS in southern Africa had demonstrated the potential constraint on health services in developing countries created by harmonized patent protection under the Trade-Related Intellectual Property Rights (TRIPs) component of the WTO Agreement (’t Hoen, 1999; Médecins sans Frontières, 2001). The authors of the year 2000 United Nations Human Development Report took the problem seriously enough to warn that TRIPs may conflict with international human rights agreements that recognize the right to share in scientific progress, because it “dramatically reduces the possibilities for local companies to produce cheaper versions of important life-saving drugs” (UNDP, 2000, p. 84; see also Mayne & Bailey, 2002; Médecins sans Frontières, 2001; Watkins 2002, pp. 208–224). Although patents are only part of the problem, since effective administration of antiretrovirals is among many therapeutic interventions that require adequate health care infrastructure (Attaran and Gillespie-White, 2001; see also Attaran & Sachs, 2001), they are not irrelevant.

In 2001, the G7 stated: “We welcome ongoing discussion in the WTO on the use of relevant provisions in the TRIPs Rights agreement. We recognize the appropriateness of affected countries using the flexibility afforded by that agreement to ensure that drugs are available to their citizens who need them, particularly those who are unable to afford basic medical care.” (G8, 2001, ¶ 17). Subsequently, the November 2001 WTO Ministerial Conference at Doha acknowledged the need for “flexibility” when public health is at issue (WTO, 2001), and stated that TRIPs “does not and should not prevent Members from taking measures to protect public health”. It took until August, 2003, however, for the General Council of the WTO to agree on an interpretation of TRIPs that reflected this position (WTO, 2003),5 and concern persists about its true effectiveness because of the limited circumstances under which a “public health emergency” can be invoked by developing country governments (Pollock & Price, 2003).

Nutrition and education

The pairing of nutrition and education may at first seem surprising, but is logical because each is an indispensable prerequisite for protecting and enhancing health; access to each is closely related to economic variables, and in particular adversely affected by poverty; and each has been the focus of commitments by the G7 nations either as members of the Group or as part of the broader international community. Nutritional deficiencies represent an adverse health outcome in themselves, and increase vulnerability to other stressors such as infectious disease (see e.g. Rice, Sacco, Hyder, & Black, 2000). The World Health Organization (2002c, pp. 49–56) has estimated that 15.8 percent of the global burden of disease (GBD) is attributable to childhood and maternal undernutrition—an underestimate of the full significance of nutritional factors, since it does not take into account, e.g. the relation between adult nonmaternal undernutrition and infectious disease. A strong correlation exists between poverty and childhood underweight, which alone accounts for 9.5 percent of the GBD.

At the 2001 Summit, G7 leaders made vague commitments to supporting agriculture through ODA as an element of poverty reduction (G8, 2001, ¶ 20) and to “target the most food-insecure regions, particularly Sub-Saharan Africa and South Asia”—apparently, given the context of the statement, for food relief. In addition, one of the key IDGs involves reducing the proportion of children under five who are underweight, but the G7 made no specific commitments related to achieving this goal. Notably absent at the Summits we studied was any commitment to the goal endorsed by the World Food Summit (WFS) in 19966 of halving the number of undernourished people in developing countries by 2015, with “a mid-term review to ascertain whether it is possible to achieve this target by 2010” (World Food Summit, 1996).

Instead, the 1999–2001 Summits addressed issues of nutrition primarily by emphasizing the need to promote applications of biotechnology. The biotechnology industry is actively supported by some G7 governments as an element of their strategies for the knowledge economy, but its relevance to nutrition and food security is highly controversial (Crouch, 2001; Persley & Lantin, 2000; Serageldin, 1999; Tilman et al., 2001). Part of the dispute is about whether the problem should be defined with reference to resource scarcity (with the corollary being that it can be “solved” by improving agricultural productivity through, e.g. the diffusion of genetically modified crops) or resource distribution. Amartya Sen (1981), Sen (1982), Sen (1989)'s path-breaking work on the political economy of famine showed that famines are not “natural” phenomena, and that access to nutrition and food security are directly related either to purchasing power or to the availability of some other entitlement to food. This may explain the absence of specific G7 commitments on the topic. It could be argued that they are addressing the issues instead by way of economic development and poverty reduction, but the adequacy of their commitments in this area is itself open to question. So, too—as we note later in the article—is the appropriateness of the underlying presumptions about economic development.

What is beyond dispute is the slow pace of worldwide progress toward improving nutrition, perhaps because of the marginal political status of food security issues and the associated international institutions (Amalric et al., 2001; UNFAO, 2001). According to the UN Food and Agriculture Organization (UNFAO, 2003, p. 30), “[t]he number of undernourished people in the developing world decreased by less than 20 million since the 1990–1992 period used as the baseline at the WFS. Worse yet, over the most recent 4 years for which data are available, the number of chronically hungry people actually increased at a rate of almost 5 million a year.” These figures actually understate the extent of undernutrition, and its potential health consequences, since they refer only to insufficient caloric intake and not to micronutrient deficiencies that affect much larger numbers of people.

Connections between education and health are harder to quantify than those involving nutrition, but it is known that education operates to reduce health risk both directly and through such intervening variables as economic growth and gender equity. Income and health gains are more dramatic as education levels for women rise, and “societies that limit girls’ access to education pay a price in poorer health, and thereby in poorer economic growth” (WHO, 2001, p. 75). Education also reduces HIV risk (World Bank, 2002c), particularly for girls and women. Those countries showing the greatest lack of knowledge about HIV/AIDS (primarily in Sub-Saharan Africa and several of the former Soviet republics) are also ones with very low and in some cases rapidly declining rates of education spending and participation (Canadian International Development Agency, 2002; World Bank, 2002c).

In contrast to the situation with respect to nutrition, the G7 have clearly stated support for numerical targets in the field of education. The Dakar Framework for Action, which emerged from multilateral meetings in 2000, identified several goals for the developing world, including “ensuring that by 2015 all children, particularly girls, children in difficult circumstances and those belonging to ethnic minorities, have access to and complete free and compulsory primary education of good quality” and “eliminating gender disparities in primary and secondary education by 2005, and achieving gender equality in education by 2015” (UNESCO 2000). Support for the Dakar Framework was clearly expressed at the 2000 Summit, and restated in 2001 (G8, 2000, ¶ 33–34; G8, 2001, ¶ 18), although without identifying the resources that would be made available. Enrolling all children in primary school by 2015 and eliminating gender disparities in primary and secondary education by 2005 are also among the IDGs.

UNESCO's 2002 Monitoring Report on progress toward the Dakar goals warned that 37 countries will probably not meet the universal primary education (UPE) target by 2015, with another 20 countries noted as requiring “renewed efforts” (UNESCO, 2002b, p. 17). Only 21 countries remained on target. Estimating progress toward the Dakar goals using school completion figures, rather than enrolment figures, the World Bank has arrived at an even more pessimistic assessment: this technique “raises the number [unlikely to meet the UPE goal] to 88 countries, out of the total 155 for which data were established. Some 35 countries are unlikely to meet the goal of eliminating gender disparities at the primary level by 2005, even when the goal is simply universal primary education and not universal primary completion” (World Bank, 2002d, p. 3).

As with health systems, the amounts of additional financing that would be needed to achieve major improvements are small in the global scheme of things. UNESCO (2002b, p. 75), noting that documentation from G7 bilateral aid agencies makes it difficult to sum up their new education commitments, estimates them at about US $1billion annually, of which US $0.3 billion will probably go to basic education—less than 10 percent of UNESCO's estimate of the new annual contributions that will be needed to meet the goals of universal primary education and eliminating gender disparity. Further, although in 2001 the G7 committed its members to “support UNESCO in its key role for universal education” (G8, 2001, ¶ 18), UNESCO's Director-General subsequently warned that budget constraints mean “the Organization cannot afford to remain on such a path of continuous belt-tightening lest it be depleted of its vitality and ability to respond to new challenges” (UNESCO, 2002a, p. ix). Here, again, we see the theme of rhetoric unmatched by necessary financial commitments.

Discussion: health, development and redistribution

Other things being equal, richer is healthier. Over the long term, the evidence for this proposition is overwhelming, both within and among nations (World Bank, 1992, pp. 10–12, 50–55; World Bank, 1993, pp. 7, 34, 39–42; Sieswerda, Soskolne, Newman, Schopflocher, & Smoyer, 2001), but how long is the long term? And how much longer may the poor be asked to wait for improvements in access to health care and in the basic determinants of health? The question is not just a rhetorical one, because contemporary development policy appears implicitly to accept a trade-off of short-term health deterioration for the prospect of eventual improvement. That acceptance was made explicit by a team of World Bank researchers studying dramatic declines in health status in Central Europe and the former Soviet Union: “In the long run, the transition towards a market economy and adoption of democratic forms of government should ultimately lead to improvements in health status…. In the short run, however, one could expect that health status would deteriorate” as incomes drop, inequalities widen, stress increases, basic health services break down and already inadequate regulation of environmental and workplace hazards deteriorate (Adeyi, Chellaraj, Goldstein, Preker, & Ringold, 1997, p. 133). The G7 have not directly addressed this issue, but analysis of the health implications of commitments made at the 1999–2001 G7 Summits, informed by the history of the last few decades of development policy and by an expanding literature on the connections between social and economic policy and population health, leads us to question the seriousness of the G7 commitment “to make globalization work for all [their] citizens and especially the world's poor” (G8, 2001, ¶ 3).

This conclusion is strengthened by observations of G7 policies in two other policy fields: ODA and trade. In Agenda 21, the document that emerged from the Earth Summit, developed countries as a whole “reaffirm[ed] their commitments to reach the accepted United Nations target of 0.7 percent of GDP for ODA”, first proposed in 1969, and “to augment their aid programmes in order to reach that target as soon as possible” (United Nations, 1992, p. 33). In 1999, the G7 committed themselves gradually to increase the volume of ODA, and to put special emphasis on countries best positioned to use it effectively (G8, 1999, ¶ 27). At the 2000 and 2001 Summits, emphasis shifted instead to the “effectiveness” of ODA (G8, 2000, ¶ 20; G8, 2001, ¶ 14). Today, none of the G7 countries approaches the 0.7 percent target and, in contrast to the performance of some industrialized countries outside the G7, the trend has been one of declining G7 commitments to ODA over the past 15 years (Table 5 ), during the very period of growth that has produced, for those countries, unprecedented prosperity. Table 5 shows that some other industrialized countries have met and exceeded the target, so it is not inherently implausible. Table 6 breaks down the costs for each G7 country of moving to the 0.7 percent figure, in terms of one of the most familiar international commodities: the Big Mac.

Table 5

Trends in G7 ODA as a percentage of gross national income (GNI)

	1984–85	1989–90	2001
Canada	0.50	0.44	0.22
France	0.62	0.60	0.32
Germany	0.46	0.42	0.27
Italy	0.27	0.36	0.15
Japan	0.31	0.31	0.23
United Kingdom	0.33	0.29	0.32
United States	0.24	0.18	0.11
And for comparison …
Denmark	0.83	0.94	1.03
Netherlands	0.97	0.93	0.82
Norway	1.02	1.11	0.83
Sweden	0.83	0.93	0.81

Includes both bilateral aid and commitments to multilateral institutions.

Source: OECD, 2002, Table 4; OECD, 2003, Table 4.

Table 6

G7 aid commitments, 2001

Country	Value of ODA 2001 $ million	ODA as percentage of GNI 2001	Additional resources that would be made available by meeting the 0.7 percent target $ million	Population million	Value per capita of additional resources needed to meet the 0.7 percent target $	Cost of a Big Mac 2001, $	Additional annual cost of meeting the 0.7 percent target, in Big Macs per capita
Canada	1533	0.22	3345	31.08	107.63	2.14	49
France	4198	0.32	4985	59.19	84.22	2.49	34
Germany	4990	0.27	7947	82.31	96.55	2.30	42
Italy	1627	0.15	5965	57.35	104.01	1.96	54
Japan	9847	0.23	20,122	127.21	158.18	2.38	65
United Kingdom	4579	0.32	5438	58.79	92.50	2.85	33
United States	11,429	0.11	61,301	285.02	215.08	2.54	85
Total		0.18	109,103

Source: OECD, 2003, Tables 4, 19, 37 except Big Macs/capita calculation, based on national cost figures (for the Big Mac) from “Big Mac Currencies Index,” The Economist, April 19, 2001.

As for market access, Oxfam (Watkins, 2002) and the World Bank (2002a) alike have noted that the industrialized world continues to restrict access to its markets to the products of the developing world, even though unrestricted market access might generate more substantial benefits to developing economies than the current combined value of ODA and debt relief. Agricultural subsidies in the industrialized world, which simultaneously limit developing countries’ market access and offer domestic producers an incentive to generate surpluses that are dumped on international markets, represent an especially intractable problem. It is difficult to disagree with economist Ha-Joon Chang (2002) description of the growth strategy now being urged on the developing world by the G7, the World Bank and the IMF as “kicking the ladder away”: the strategy is one that no G7 country followed on its own path to industrialization and wealth creation—with the partial exception of England after the repeal of the Corn Laws, which had the advantage of empire as a captive market as well as a source of raw materials. Thus, not only the ethical defensibility of contemporary neoliberal prescriptions for health through wealth through growth, but also their empirical plausibility, is called into question.

Population health measures, including infection control and some forms of environmental protection, may represent genuine “public goods” that the rich world cannot feasibly purchase for itself while excluding others (Chen, Evans, & Cash, 1999). On this line of reasoning, population health represents an investment in global security, and the economic costs associated with the recent outbreak of Severe Acute Respiratory Syndrome show the value of infection control as a public good. A more expansive economic argument is exemplified by the work of the Commission on Macroeconomics and Health, for which “investment in health” represented an investment in future development, because it can initiate virtuous cycles of human capital formation and growth. The approach is empirically well grounded, yet without further elaboration it invites a form of triage in which the countries, regions and populations that receive investments in health will primarily be those where “development” offers the greatest promise of economic returns, e.g. because of the availability of expanding consumer markets or the availability of healthy and relatively skilled, yet low-cost labour.

John Williamson, who coined the term “Washington consensus” to describe official wisdom on development policy circa 1989 (Williamson, 1990), has noted that in codifying the consensus he “deliberately excluded from the list anything which was primarily redistributive, as opposed to having equitable consequences as a by-product of seeking efficiency objectives, because [he] felt the Washington of the 1980s to be a city that was essentially contemptuous of equity concerns” (Williamson, 1993, p. 1329). Our analysis suggests that with some important exceptions, the attitude to which Williamson referred appears to have diffused throughout the official levels of the G7. It is stronger than ever in today's Washington. Our analysis also suggests that in the absence of more extensive redistribution of wealth across national borders, progress toward improving health for all will be slow at best, and may not be possible at all in some situations. Future research must not only investigate in greater detail the health consequences of contemporary development policy as promoted by the G7 and the international financial institutions, but also undertake explicit ethical analysis of the health consequences of current G7 development policies.