This study represents the first econometric study that has attempted to model energy demand exclusively for Jordan. In so doing, recent developments in time series econometrics modeling techniques are adopted to estimate total energy demand functions at the aggregate and sectoral levels together with demand functions for individual fuels for Jordan using a new data set covering the period (1968-2000). Different econometric approaches were employed to estimate the elasticities including OLS, Johansen's ML, DOLS and ARDL to enable a comparison of the statistical results and the estimated price, income, urbanization elasticities. The different econometric techniques therefore act as a check of the robustness of the results obtained. All estimates of the demand functions whether at the aggregate level or at the disaggregate levels are robust not only in terms of statistical competence but also in terms of economic intuition. At the aggregate level, the various econometric techniques yield almost identical elasticity estimates. The estimates indicate a long run elasticity of around unity with respect to per capita GDP, 0.35 with respect to per capita area constructed, -0.30 with respect to real energy prices, a coefficient around 0.1 for the dummy variable representing the level of conflict in the region and a coefficient around 0.7% p.a. for the time trend as a proxy to the Underlying Energy Demand Trend (UEDT). The income elasticity implies that economic growth is likely to be accompanied by proportional increases in energy consumption. The price elasticity suggests that taxes on their own are unlikely to achieve government goals for energy conservation or environmental improvement, although they may well be efficient for revenue raising. Furthermore, evaluating the impact of per capita area constructed on energy consumption levels helps provide guidance for the future need of power generation and refining capacities. The study present forecasts for aggregate energy, aggregate electricity and aggregate petroleum over the period 2001-2015 using 2000 forecasts from the Jordanian governments and international organizations for the exogenous variables. The forecasts are constructed using three different scenarios of the GDP and area constructed growth, with constant real energy price at the 2000 level and with real energy prices increasing by 2% annually. The low growth scenario assumes 4% annual growth of GDP and area constructed, the medium growth scenario assumes 6% for both annual growth of GDP and area constructed, and the high growth scenario assumes 8% annual growth of GDP and 6% annual growth of area constructed. With constant real energy price, the low growth scenario suggests that total energy demand will increase by an average annual growth rate of 3.9 % over the forecasted period. The medium growth scenario indicates that the total energy demand will increase by an average annual growth rate of 6.37 % over the forecasted period, implying that aggregate energy demand will double by the year 2012. The high growth scenario suggests that the total energy demand will increase by an average annual growth rate of 8.24 % over the forecast period and double by the year 2010.

Le monde est actuellement confronté à une transition du marché de l'énergie qui est influencée notamment par la dynamique de la croissance économique globale, les négociations relatives aux changements climatiques et des prix de plus en plus volatils. Cette évolution rapide des réglementations et de la macro-économie transformera les conditions de la demande d'énergie, obligeant les gouvernements à acquérir un ensemble croissant d'outils quantitatifs pour mieux évaluer les résultats de leurs politiques fiscales. Cette thèse aborde cette problématique en analysant, par une approche basée sur les élasticités, les différentes facettes de la demande d'énergie dans le but d'achever une consommation énergétique durable. Cette approche est complémentée par l'analyse par grappes, la décomposition structurelle ainsi que par diverses outils économétriques appliques conjointement à l'échelle mondiale et nationale. Une attention particulière est faite sur la modélisation de la demande des marchés subsidiés notamment des pays du Conseil de Coopération du Golfe Arabique / The world is currently witnessing a transition in the energy scene that is significantly characterized by global economic growth dynamics, climate change negotiations and volatile energy prices. Rapidly evolving regulatory and macro-economic environments heavily impact on the demand-side of energy, forcing governments to acquire an ever-increasing set of quantitative tools to better assess the results of their taxation policies.This thesis addresses some of these issues by analyzing various facets of energy demand in order to generate sensible demand and price elasticities with real-life applications in sustainable energy management. For that purpose, a combination of cluster, decomposition and multiple econometric analysis is undertaken at global, regional and country-specific levels for households complemented by a policy analysis. A special focus is made on modeling consumer demand behavior for resource-rich economies of the Gulf Cooperation Countries, and the potential impact of removing residential electricity subsidies on the net societal welfare of Saudi Arabia.