I feel greatly honoured to have been invited to participate in the inaugural conference of Environment and Development Economics. I should like first of all to congratulate Karl-Göran Mäler, Partha Dasgupta and Charles Perrings on the inauguration of the new journal and wish it to be successful in opening an entire new vista in economics.

People in poor countries are for the most part agrarian and pastoral. In 1988 rural people accounted for about 65 per cent of the population of what the World Bank classifies as low-income countries. The proportion of the total labour force in agriculture was a bit in excess of this. The share of agriculture in gross domestic product in these countries was 30 per cent. These figures should be contrasted with those from industrial market economies, which are 6 per cent and 2 per cent for the latter two ratios, respectively.

This paper provides an overview of recent research on the interconnections between environmental and trade policies. It describes how the assumptions of these models can be important to the conclusions derived on the gains from coordination. The Harrison–Rutherford–Wooton CGE model of the European Union is used to illustrate the importance of these issues. This model was extended to include three air pollutants and their health effects as non-separable influences on household preferences in each region described by the model. The results suggest that the conventional assumption of separability in preferences between marketed and non-marketed goods is central to conclusions about the importance of coordination of these policies.

Economic models for assessing costs of greenhouse gas abatement can be classified as macroeconomic (top-down) models or micro (bottom-up) models. As yet little attention has been focused on the modelling of developing countries. This paper proposes a methodology for integrating the two approaches and illustrates it with the example of India. The results show that a carbon tax would have limited effectiveness in controlling carbon emissions. Investing carbon tax revenues in carbon-abating technologies can give substantial carbon savings. The costs of carbon abatement using the integrated model are higher than those predicted by other global models with India as a sub-region.

The sustainability of natural resource use is influenced by population pressure, but this exercises a much less critical impact than the overall policy framework. In Colombia, various agricultural and other policies whose effect is to constrain the poor's access to land encourage environmental degradation. A case is made in favour of the new land reform process that Colombia is launching.

The search for sustainability indicators should be guided by a theory of sustainable development (SD). In this paper we investigate two such theoretical frameworks and the indicators that they suggest. Indicators associated with weak sustainability are characterized by aggregative indicators such as green national income. We conclude, however, that a more promising offshoot of green accounting is measures of genuine savings (i.e. savings adjusted for loss of assets). To achieve SD, genuine savings rates must not be persistently negative. Strong sustainability indicators accord a more central role to the conservation of critical natural assets within the broader goal of prudently managing a nation's portfolio of assets over time. We discuss two approaches—carrying capacity and resilience—and conclude that, while measures of resilience are potentially attractive, more research is required regarding the resilience–SD link. However, an important conclusion that we can make is that, even in an economy operating under a strong sustainability regime, genuine savings are still key indicators of SD and are complementary to measures of changes in stocks of critical natural assets.