Korea Productivity Association
논문검색
pISSN: 1225-3553
생산성논집, Vol.32 no.3 (2018)
pp.3~35
유가 충격 이질성을 고려한 에너지 다소비 사업장의 에너지 효율성 추정
To explore more reasonable GHG mitigation policies, more accurate information on the company's energy performance indicators are needed. Generally, energy intensity is frequently used as an indicator for energy performance. Although the indicator has the advantage of being intuitive and simple, it is also true that many other statistical factors are not taken into consideration. At this time, energy efficiency based on Stochastic Frontier Analysis (SFA) technique or Data Envelopment Analysis (DEA) technique can be used as an alternative indicator to energy intensity. The techniques can drive more accurate performance by considering more information than energy intensity. Despite the importance of measuring the energy efficiency of the enterprise unit, study on the energy efficiency measure at enterprise unit or business unit in Korea has rarely been studied so far. The reason for the lack of research on the issue is probably because the data of the enterprise energy consumption is not generally publicized. The purpose of this study is to measure the energy efficiency of the energy consumption industry in Korea based on the internal data of the Korea Environmental Industry & Technology Institute. We estimate energy efficiency of facilities in Energy Intensive Industries of Korea by Stochastic Frontier Model. Considering information and degree of freedom of the dataset, sophisticated model is not available. Therefore, this study benchmarked Stochastic Frontier Model proposed by Herrala & Goel (2012) which us one of the very simple model for greenhouse gas efficiency. The benchmarked part from Herrala & Goel (2012) is the functional form and variable selection of the regression equation. It is another matter to decide the technical way to estimation. In this study, time variable panel SFA modeling technique of Battese & Coelli (1992) is used for estimation. Energy Intensive Industries can be influenced by oil price shock; however, the time span of dataset of the study, 2013-2015, was a period in which the oil price fell sharply. If each firm's sensitivity to exogenous shocks such as oil price shocks, is different, time varying heterogeneity can be included in efficiency measure. Therefore, we control the oil shock heterogeneity using by oil price fluctuation as a proxy variable with facility specific coefficient based on the time varying panel SFA model of Battese & Coelli (1992). As a result, it was found that there were significant energy inefficiencies of firms in iron & steel, nonmetal, and chemical industries. However, there was no significant difference in energy efficiencies in firms in paper industry. The presence of inefficiency means that more energy is used under the same output. Although existence of a energy inefficiency is a negative sign, but it also means that there can be a chance to improve overall energy performance by improving inefficiency. The contribution of this study is in terms of statistical technique rather than finding a specific policy alternative. In particular, the time varying heterogeneity problem presented in this study is newly suggested in the efficiency related study. Policy implications can be found in that it serves as an important intermediate part of many analyzes, as is the case with energy intensity. The energy efficiency indicator proposed by us needs to be combined with a lot of information from individual companies in order to bring on a substantial policy implication; thus, a lot of follow-up research is needed.