The marginal price elasticities estimated by Martinez-Espineira conforms to expectation. The price specification that accounts for the changing proportion of water users in each block yields a higher elasticity (-0. 47) compared to the spec ification ignoring this feature of the data. However, this difference is not found to be statistically significant, a result attributed to the low power of the test (small sample size limiting the accuracy of estimates). In conclusion, the paper provides a theoretically correct price specification for demand functions under block pricing and aggregate data. The empirical findings in the paper, however, are not conclusive and further empirical work using more data and alternative (nonlinear) demand functions, is needed to show the practical implications of the arguments put forward by the Martinez-Espineira's paper. Static empirical consumer demand functions estimated with aggregate data are well known to suffer form serial correlation and other statistical problems asso ciated with misspecified dynamics. These dynamics arise because consumers do not react immediately to a change in prices due to their largely predetermined lifestyle. In the case of demand for water, for example, current purchases can be largely predetermined due to commitments arising from past purchases such as swimming pools, bathtubs, dishwashing machines, etc. Muellbauer and Pashardes (1992) show that the autoregressive nature of consumer demand data can be cap tured in a theoretically consistent manner by incorporating intertemporal aspects of consumer behaviour in the model through habit formation and durability."

The marginal price elasticities estimated by Martinez-Espineira conforms to expectation. The price specification that accounts for the changing proportion of water users in each block yields a higher elasticity (-0. 47) compared to the spec ification ignoring this feature of the data. However, this difference is not found to be statistically significant, a result attributed to the low power of the test (small sample size limiting the accuracy of estimates). In conclusion, the paper provides a theoretically correct price specification for demand functions under block pricing and aggregate data. The empirical findings in the paper, however, are not conclusive and further empirical work using more data and alternative (nonlinear) demand functions, is needed to show the practical implications of the arguments put forward by the Martinez-Espineira's paper. Static empirical consumer demand functions estimated with aggregate data are well known to suffer form serial correlation and other statistical problems asso ciated with misspecified dynamics. These dynamics arise because consumers do not react immediately to a change in prices due to their largely predetermined lifestyle. In the case of demand for water, for example, current purchases can be largely predetermined due to commitments arising from past purchases such as swimming pools, bathtubs, dishwashing machines, etc. Muellbauer and Pashardes (1992) show that the autoregressive nature of consumer demand data can be cap tured in a theoretically consistent manner by incorporating intertemporal aspects of consumer behaviour in the model through habit formation and durability."

One of the central tenets of this book is that governmental policies must be designed to take into account market characteristics and environmental phenomena - simultaneously. This volume contains a new research effort of the Fondazione Eni Enrico Mattei' and explores the theoretical underpinnings of environmental policy in a sub-optimal world. Topics considered link economic issues (oligopolistic market structures, firm heterogeneity, and the strategic behavior of governments) to environmental issues (emission abatements, cleaner technologies, and environmental taxation). The articles in this volume were chosen to achieve a balance between breadth and depth and were written by leading experts in the field. In short, this book is rich in policy implications and raises new issues and questions for future research.