| |
 See also
|
| |
|
|
|
Press Info item.
22/06/2007
An economic model to manage natural resources in a risk situation
The risky nature of managing natural resources, and particularly water, is a crucial element which must be taken into account in any economic analysis. However, the problem of estimating the risk preferences adopted by decision-makers has been little studied. An INRA researcher, in collaboration with the University of California, has developed an economic model which can estimate the risk aversion of a decision-maker, as well as his desire to smooth consumption over time. In 2006, this work(1) received the Quality of Research Discovery Award from the American Agricultural Economics Association. This rewards a research project for its significant contribution to advancing knowledge in agricultural economics. It is the first time that a French scientist has been distinguished by this award.
|
| |
The dynamic nature of the problems encountered when managing natural resources (water, forests, etc.), and the uncertainties faced by managers, create complexities which must be integrated in any economic analysis so as to ensure the credibility of any recommendations which result, in terms of public policies.
The recursive intertemporal utility function hypothesis* concerning economic agents, an hypothesis frequently accepted regarding the economics of natural resources, constitutes a major limitation to understanding decision-making by agents in an uncertain context. In particular, this hypothesis cannot distinguish preferences regarding a risk from those regarding time. This limitation is particularly problematic in the management of natural resources which, by its very nature, is a long-term process.
This question formed the starting point for a joint project by researchers at INRA and the University of California, Davis, aimed first of all at integrating non-recursive utility functions in natural resource management models in an uncertain context, and then proposing novel methods to estimate the preferences of decision-makers in these models. From a more normative point of view, these approaches enable characterisation of the optimum strategies for natural resource use in an uncertain context. The methodological contribution consisted in integrating a maximum likelihood estimate of preference parameters in the programming model.
An application of the model: management of the Oroville storage dam in California
This model was first applied to management of the Oroville storage dam in Northern California. The results showed that recourse to this new model considerably improved predictions of the management behaviour of the actual decision-makers. Indeed, over a period of more than 20 years, the researchers compared the records of decisions taken regarding management of the Oroville storage dam (water releases) with several simulation models based on different optimisation criteria. Use of the new model then made it possible to reproduce the history of water releases with a predictive error reduced by nearly 50% when compared with previous models.
A second project currently under way aims to use observations of the water release records from four of the main storage dams in Northern California to estimate risk preferences and intertemporal substitutions by decision-makers.
Another field of application: the management of forests faced with catastrophic risk
Forest management is a long-term problem for which non-recursive utility criteria find a second, natural field of application. The storms in December 1999 clearly served to remind us that climatic risks could have a major impact on this resource.
In collaboration with INRA scientists in Nancy, a study was thus performed on the management of Lorraine forests in the presence of a risk of storms(2). The cutting of timber generates income that the manager decides either to consume or to save. The methodological value of this research was to integrate consumption and saving decisions in the optimisation programme of the forest manager when faced with a risk of storms. The authors of this study thus characterised the optimum use of several forest stocks when a risk of storms could affect them heterogeneously, and when the forest owner could save. Introduction of an arbitration between consumption and saving opened an opportunity for the intertemporal smoothing of consumption (which enabled a carry-forward in terms of income consumption) which had hitherto not been analysed in forestry economics.
An application of this model to the management of pine forests in Aquitaine is currently under way. The question under study is that of optimum cutting dates in the presence of an increased risk of fire.
*This hypothesis means that to evaluate the utility to an agent of consuming an asset at different periods, it is sufficient to calculate the updated sum of utilities at each period.
(1) Reference of the article which received the Quality of Research Discovery Award 2006 from the American Agricultural Economics Association (AAEA):
HOWITT R., S. MSANGI, A. REYNAUD, et K. KNAPP (2005). "Estimating Intertemporal Preferences for Natural Resource Allocation". American Journal of Agricultural Economics. 87(4). 969-983.
(2) COUTURE S. et A. REYNAUD (2007). "Recursive Utility and Saving in Forest Management Modeling". Environmental Modeling and Assessment. Awaiting publication.
|
| |
|
Written by :
INRA press service, phone: +33 (0)1 42 75 91 69
Contacts :
Arnaud REYNAUD
tél. : 05 61 12 85 12
areynaud@toulouse.inra.fr
INRA-Université Toulouse I Joint Research Unit for Environmental and Natural Resource Economics
Social Sciences, Agriculture and Food, Rural Development and Environment Division,
Toulouse Research Centre.
|
|
|
|
|
Search:
