Through our rapacious exploitation, ecosystems and the benefits they bring to us are disappearing at an unprecedented and alarming rate. Qualitative evidence suggests that the rich world is profiting from this process, whereas poorer countries are bearing the brunt of the resulting environmental degradation. Srinivasan et al.1, writing in Proceedings of the National Academy of Sciences, provide a quantitative basis for that claim, calculating the distribution of costs and benefits over a range of indicators of ecosystem change. The results might, in an ideal world, lead to a radical reassessment of who is in debt to whom.

Measured in terms of gross domestic product (GDP), the size of the world's economy has doubled almost three times since 1950. But the proposition that aggregate economic growth alone is the most important and powerful force for human progress and poverty reduction has increasingly been questioned2,3. Economic growth fuelled by international trade relies on the consumption of heavily advertised and marketed goods and services. Poorer people and the natural world, marginalized by the market economy, lose out.

Besides basic economic needs, a high quality of human life must include satisfactory human relationships, communities, freedoms and opportunities4,5. Well-functioning ecosystems are even more fundamental. The buzz-word is 'ecosystem services': the favours such as food and fuel; protection from storms, disease and solar radiation; regulation of water and climate; creation of soils; and inspiration for art, literature, religion and culture that the natural world bestows on us. Current economic models undervalue this provision, because many ecosystem services are 'public goods', historically provided for free. This leaves them prone to overexploitation, a trend exacerbated by global population growth: 15 of the 24 ecosystem services identified in the seminal Millennium Ecosystem Assessment initiated by the United Nations in 2001 were found to be in decline at the global scale6.

The current political focus on climate change has brought this pressure on our ecological support systems into sharper relief. Sustaining ecosystem processes is all the more difficult because many change abruptly and nonlinearly when pressed beyond a certain threshold. Qualitative or context- and case-specific evidence indicates that poorer countries bear most of this pressure7. An example is the widespread conversion of tropical mangrove forests to shrimp aquaculture. These farms supply Europe and North America with cheap shrimp, but nearby residents must pay the costs: the loss of the storm regulation, fish nurseries, and fuel and fibre sources that the mangrove forests provided8.

This type of unequal exchange has been termed an 'ecological debt' owed by rich countries to the poor7. As raw materials — whether they be shrimp, palm oil, or crude oil and gas — become scarcer, and exploitation of them advances into new territories, this debt will probably increase. The prices at which exports are sold do not include compensation for the goods' local — or sometimes global — environmental costs. That problem is exacerbated by rich countries' disproportionate use of environmental sinks, such as the atmosphere, to take up their higher carbon emissions.

Srinivasan et al.1 present for the first time a global-scale quantitative analysis of the distribution of major environmental costs across nations in three income groups: low (representing 32% of the world's population), middle (50%) and high (18%). The timescale of the study is 1961–2000, and it covers six categories of environmental change: climate change; stratospheric ozone depletion; agricultural intensification and expansion; deforestation; overfishing; and mangrove conversion. In the case of climate change, the authors calculate that nations in the low, middle and high income groups were responsible for 13%, 45% and 42% of greenhouse-gas emissions, respectively. The resulting climate damages were estimated to be distributed 29%, 45% and 25%. The same pattern of cost transfer to the poorest nations is observed for ozone depletion, overfishing and aquaculture. In the case of agricultural change and deforestation, the authors find that arguably the costs were largely generated and borne within nations.

When the figures for the six categories are aggregated and adjusted for different currency purchasing power, the low-income group of nations bears 20% of the total costs, the middle-income group 60% and the high-income group 20%. Taking into account equity weighting (the fact that a unit of income is not the same to a poor as to a rich person, making costs a heavier burden to shoulder) the percentages are 45%, 52% and 3%. A significant proportion of the cost burden of the low-income group is caused by the activities of the other groups: looking at climate-change damage alone, rich countries might already have imposed costs on poor countries greater than the poor countries' existing foreign debt. The people bearing these costs include the one billion or so who already lack daily access to safe drinking water, electricity, secure food supplies and basic education.

Several caveats are needed to put these striking results into context. First, any study is only as good as the available data. Globally robust data sets on environmental externalities are incomplete, and therefore the results are indicative, not absolute. Some significant environmental changes have been omitted: the destruction of coral reefs, the dispersal of persistent pollutants and the costs of biodiversity loss. The authors are up-front about these limitations. But they justly suggest that the omissions indicate that their estimates of ecological debts are conservative.

Furthermore, the study does not look at the benefits of increased trade to material wealth and human health. But it is unlikely that this would change the balance of the results: the available, albeit partial, evidence indicates that, despite some gains by poorer countries through the liberalization of trade and finance, the number of people living in poverty has increased or stayed the same during the past 25 years, and the gains of economic globalization have been heavily skewed towards wealthy nations2 (Fig. 1). One of the most detailed studies on global inequality concludes9 that income divergence between rich and poor nations has “at best ... decelerated after 1950, but [has] not reversed”. Large areas of Central and South America, and almost all of sub-Saharan Africa, have been left behind.

Figure 1: Inequality in economic growth.
figure 1

a, Although gross world product and the value of exports throughout the world rose steadily during the period 1981–2001 (values relative to 1990 = 100), the number living below the US$2 per day poverty line has increased slightly10. b, Whereas in 1820 the difference between the mean incomes of the poorest 60% of countries and of the richest 10% was fourfold, by 1992 the disparity was tenfold (value of mean income of poorest countries in 1820 = 1)9.

Srinivasan and colleagues' work1 should raise the scientific profile of issues vital to human well-being, including research into ecosystem services and the effects of large-scale ecosystem conversion, and how such changes both create and alleviate poverty. We must better understand the complex interactions between our economic, social and ecological systems, and the biological diversity that supports them. Scientists and society as a whole need to ask of our current economic paradigms in an era of globalization: why do they produce such inequities; who pays the costs; and are they ecologically and socially sustainable?