Skip to main content
SLU publication database (SLUpub)
Doctoral thesis, 2018

Essays on energy demand and efficiency analysis

Tafesse Tirkaso, Wondmagegn


This thesis consists of four papers focusing on the issues of energy demand and efficiency analysis in the agricultural and environmental sectors. The first paper examined the long-run relationship among rainfall distribution, electricity consumption and economic growth (GDP) in Ethiopia using time series data between 1981 and 2014. Results indicated the existence of unidirectional Granger causality from rainfall to real GDP, and bidirectional Granger causality between capital stock and real GDP. There was also evidence of unidirectional Granger causality from real GDP to electricity consumption but not vice versa. The second paper estimated national and regional-level gasoline and diesel demand elasticities in Sweden using county-level panel data for the period 2001-2015. National-level elasticity estimates from a partial adjustment model indicated that per capita income, own and substitute prices, and per capita vehicle stocks were statistically significant in determining gasoline demand, and per capita income and own price were statistically significant in diesel demand case. Furthermore, regional elasticity estimates were computed from the national estimates, revealing a variation between counties, with the highest being approximately 40% greater than the lowest in absolute terms. The third paper evaluated cost efficiency associated with hydropower-related biodiversity restoration projects in Sweden based on a survey data from 245 projects. The results provided evidence of cost inefficiency with an average efficiency score of 55%, suggesting the potential to minimise cost efficiency loss by up to 45%. Factors such as project duration, project owner and restoration measure type were statistically significant determinants of cost inefficiency score. The fourth paper estimated farmers’ level of technical efficiency score and examined its connection with commercialisation using a survey data in rural Ethiopia. Estimates from the stochastic frontier production function showed that farmers had a mean technical efficiency score of 51%. The variables related to educational level and radio and mobile telephone access were positively linked to farmers’ technical efficiency. Furthermore, farmers with a higher level of commercialisation were technically more efficient than those with lower market participation.


rainfall, electricity consumption, economic growth, fuel demand, elasticity, technical efficiency, stochastic frontier, biodiversity restoration, Ethiopia, Sweden

Published in

Acta Universitatis Agriculturae Sueciae
2018, number: 2018:60
ISBN: 978-91-7760-260-6, eISBN: 978-91-7760-261-3
Publisher: Department of Economics, Swedish University of Agricultural Sciences

    UKÄ Subject classification


    Permanent link to this page (URI)