RT Dissertation/Thesis
T1 Domestic and agricultural water use by rural households in the Oueme River Basin (Benin): an economic analysis using recent econometric approaches
A1 Arouna,Aminou
WP 2009/11/09
AB Improving the management of water resources as well as an efficient use of available water are particularly important to address the increasing scarcity of water and the low level of water accessibility in many developing countries. However, better water management requires an understanding of the existing pattern of water use for domestic and agricultural activities. With a view towards contributing to such knowledge, this dissertation analyzes domestic and agricultural water use by rural households in the Oueme river basin of Benin. This is done within the scope of three research articles. The specific objectives of the dissertation were: 1) to analyze determinants of domestic water use in the rainy and dry seasons; 2) to estimate households? willingness to pay for water supply improvements and analyze its determinants; and 3) to quantify the efficiency of water use for agricultural production and identify factors explaining the differences in water use efficiency among households.
The analyses are built on primary data collected from a household survey administrated to a sample of 325 households in the Oueme river basin, in 2007. To analyze domestic water demand, we identified three types of households: those that use only free water sources, those that use only purchased sources and those that combine both free and purchased sources. A system of two demand equations (one equation for free water and another for purchased water) was estimated using a Seemingly Unrelated Tobit (SURT) approach. The advantage of using the SURT approach is that it is appropriate to account simultaneously for the censored nature of water demand and the correlation between the error terms of two equations. In the analysis of households? willingness to pay (WTP) for water supply improvements, particular attention was given to the distribution of WTP, which has been addressed using (arbitrary) parametric assumptions in many previous studies. To avoid distributional assumptions, the dissertation introduced a semi-nonparametric bivariate probit approach to estimate WTP. To analyze water use efficiency, the dissertation combined an input-specific Data Envelopment Analysis (DEA) with a bootstrapped Tobit model. Bootstrapped Tobit takes care of the dependency problem between efficiency estimates. The analysis of water use efficiency focused on vegetable production in the dry season when water is scarce.
Results showed that the average daily domestic water consumption per household during the rainy season (252 liters) is significantly higher than in the dry season (216 liters). SURT estimation results showed that water demand from purchased sources is perfectly price inelastic in the rainy season; indicating that rural households in Benin are very insensitive to changes in water price. This suggests that households are willing to pay more for water supply improvements, due not only to the necessity nature of water but also to its scarcity. Factors affecting domestic water use in the rainy season are household size and composition, education, time for fetching and accessibility to water sources. In the dry season, econometric analysis revealed that there is a positive relationship between wealth and the use of water from free and purchased sources. This result suggests that poverty reduces water use. Purchased water demand in the dry season is also perfectly price inelastic. However, a comparison of determinants of water use between seasons revealed that variables such as time for fetching water, access to water sources and wealth have differential influence on water use during the rainy and dry seasons. These results imply that policy makers must consider among other factors seasonal variation of the determinants of water use. 
The results of this dissertation provided the first evidence that, in rural Benin, households wanting to improve water supplies are willing to pay more than existing water prices. Households are willing to pay over one and a half times the present average water price. Furthermore, results revealed that estimated WTP would generate substantial revenue from the community, which can lead to significant reductions in subsidies. The supply of safe and adequate water based on estimated WTP will reinforce both the participation of the rural population in water supply management and the sustainability of water facilities. A related policy is that a demand-side management approach can be successfully implemented in rural areas for water supply improvements and sustainability. The important determinants of WTP for water supply improvements were education, age of household head, wealth, queue time at existing water sources and preferred improvements. The policy implication of these findings is that a combination of socio-economic factors affecting WTP, and a demand-side management approach, are likely to improve the sustainability of water projects in rural areas of Benin.
Average water use efficiencies were 0.38 and 0.50 under constant and variable returns to scale specification, respectively. This implies that if vegetable farmers in the study area become more efficient in water use, significant amounts of water could be saved and made available for dry season farming land expansion. In addition, many farmers operated at an increasing return to scale (average scale efficiency is 0.70), revealing that most farms should be larger than they currently are to produce efficiently. Water use efficiency in vegetable production was determined by market access, land fragmentation, extension service, ratio of children to adults, water expenditure, water sources, off-farm income and wealth. Results suggest that policy makers should focus on improving farmers? access to input and output markets as well as their access to technical information and training through extension service or NGOs. The findings also showed that households paying for irrigation water or systems are more efficient in water use. However, any price policy should be combined with other policy options such as training and development of improved irrigation techniques adapted to socio-economic conditions of farmers. 
Overall, various socio-economic characteristic of households and institutional factors are found to explain water use for both domestic and agricultural activities. These factors must be carefully considered for the design and implementation of water management programs that can lead to sustainable accessibility to water. Although the research focuses on Benin, most of the conclusions and policy implications are relevant and could be applicable to many developing countries with similar socio-economics conditions. The dissertation also applies and extends recent econometric approaches that may be used for empirical studies on water management policy in developing countries.
K1 Wasserverbrauch
K1 Effizienzanalyse
K1 Gemüseproduktion
K1 Ländlicher Haushalt
K1 Benin 
PP Hohenheim
PB Kommunikations-, Informations- und Medienzentrum der Universität Hohenheim
UL http://opus.uni-hohenheim.de/volltexte/2009/393