The majority of the Malawi people is rural (85%) and lives primarily on subsistence farming (NSO 2001). More than 90% of households collect and use fuel wood as their main source of cooking energy (NEC 2001; NSO 2000). However, between 1990 and 2000, Malawi experienced an average annual deforestation rate of 2.4% that was significantly higher than both Africa's average deforestation rate of 0.78%, and the world's average deforestation rate of 0.22% (UN FAO 2001). Malawi is also reported as one of the countries that will experience water stress by 2025 (PRB 2002). The rapid depletion of natural resources can have significant consequences for the quality of people's lives. Since Malawian women and children are primarily responsible for rural transportation work including collection of fuel wood and water (Edmonds et al. 1995), environmental degradation is expected to affect them disproportionately. This study will investigate the effects of environmental degradation on women's and children's time allocation decisions and the implications on children's school performance and health. The importance of relationships between population, the environment, and poverty has been acknowledged at both international and regional levels since the 1970's (United Nations 1997). However, there is a serious dearth of empirical studies on the nature of such relationships (United Nations 1997). The lack of studies has been cited as one of the reasons frustrating policy makers in their attempt to adopt sustainable development efforts (Arizpe et al. 1994). Some recent studies in this area have looked at effects of environmental degradation on women's fertility (Filmer and Pritchett 1996; Aggarwal et al. 2001); women's time allocation decisions (Cooke 2000); farm productivity (Cooke 1998). Earlier work by the authors showed that environmental degradation was associated with lower school enrollment and high domestic child labor hours, especially for girls (Nankhuni and Findeis 2002). This analysis extends that research to investigate the impact of environmental degradation on school performance. Attempts to look at impact on child health were done by Aggarwal et al. (2001) in South Africa but no significant impact was found. This study will also contribute to the research on effects of environmental degradation on children's health. Methodology Data Data from an Integrated Household Survey of 10,698 households conducted in Malawi in 1997-98 by the Malawi National Statistics Office are used. The data contain information on demographic and socio-economic characteristics of individuals and households, including child levels of education and health. The data also include time allocation information on domestic activities that include time spent on fuel wood and water collection. Information on each household's access to different water sources and whether the household relies on purchasing rather than collecting fuel wood as their main source of cooking energy is also available. Supplementing these data are estimates of fuel wood availability (GOM 1987) and information on access and quality of schools and health facilities at the district level (Benson 2002). Theoretical framework Freeman (1993) proposes that the basis for measuring the economic value of changes in resource environmental systems is the effects on human welfare. Therefore, the starting point for analyzing the impacts of environmental degradation is utility theory. The household production-utility model based on Becker (1965, 1993) and as adapted to farming households by Singh et al. (1976) will be the theoretical approach used in this paper. In this model, households derive utility from consumption of household farm-produced goods and from having children. They also derive positive utility from children's quality normally reflected in the children's health and education. The household's utility is maximized subject to budget, farm and household technology, and time constraints. Assuming that an interior solution to the household's maximization problem exists, reduced-form demand equations for children's health and education can be derived. These demands will be functions of shadow wages, prices, individual and household socio-economic and demographic characteristics, and the state of the environment. As the environment degrades, more hours of work are spent on fuel wood and water collection. This results in increased price of education thereby decreasing demand for education, as children may be needed for domestic work. Similarly, if women in more deforested areas cannot spend enough time on farming, cooking, and childcare this will act as an increase in cost of children's health resulting in the children's poor health (Kumar and Hotchkiss 1988). Estimation strategy The effect of environmental degradation on children's quality will be estimated by adding environmental quality variables in school performance and child health models. The school performance indicators are children's attendance of senior primary school and being in a certain class at the appropriate age for that class. The child health variables are anthropometric measures of weight, height, and weight for height. Possible endogeneity of child schooling, child health, domestic child labor, and fertility decisions will be tested and corrected for, where appropriate. It is expected that environmental degradation will negatively affect children welfare.