Title : Eco-Efficiency Analysis of Processing Industry in Indonesia: Case Study on Energy-Intensive Industry
Author : Deni Kusumawardani
Universitas Airlangga
Item Type : Thesis (Dissertation)
Abstract
The main objective of this study is to analyze the eco-efficiency of the processing industry in Indonesia with a case study of energy-intensive industries. Specifically, these objectives are divided into four studies: (1) measuring and comparing eco-efficiency; (2) measuring and comparing technology gaps; (3) decomposing sources of eco-inefficiency; and (4) identifying determinants of eco-inefficiency. In this study, energy-intensive industries include eight types of industries, namely palm oil processing, textiles, pulp & paper, chemicals, fertilizers, glass & ceramics, cement, and metal & steel. The analysis is conducted both in aggregate across energy-intensive industries and by industry type using annual firm-level micro data from 2010 to 2015. Eco-efficiency is measured using a meta-frontier analysis framework that is able to accommodate heterogeneity in production technology across industry types. Data Envelopment Analysis (DEA) is applied as a frontier estimation model and the radial Directional Distance Function (DDF) as an eco-efficiency measurement method. Meta-frontier produces two types of eco-efficiency measurements, namely eco-efficiency to meta-frontier (MEE) and group-frontier (GEE). The technology gap is measured by the meta-technology ratio (MTR) which is defined as the ratio of MEE to GEE. Differences in eco-efficiency (MEE) and technology gaps between industry types are statistically tested using the Kruskal-Wallis Test. Furthermore, within the meta-frontier framework, eco-inefficiency (MTI) can be decomposed into two parts, namely eco-inefficiency caused by technological gaps (TGI) and by managerial failure (GMI). Meanwhile, the determinants of eco-efficiency are identified using the Tobit Panel Model. This study produces several conclusions. First, the average eco-efficiency score of energy-intensive industries against the meta-frontier (MEE) in the 2010-2015 period is 0.687. According to industry type, the highest average MEE score is achieved by the cement industry (0.802), while the lowest is achieved by the fertilizer industry (0.681). In the same period, the average eco-efficiency score of energy-intensive industries against the group-frontier (GEE) was 0.687. The highest average GEE score was achieved by the cement industry (0.962) and the lowest by the textile industry (0.709). The results of the Kruskal-Wallis test indicate that there is a significant difference in eco-efficiency scores between types of industries within the energy-intensive industry group. Second, the average technology gap (MTR) of energy-intensive industries in the 2010-2015 period was 0.940. According to industry type, the highest average MTR was achieved by the textile industry (0.974) and the lowest by the pulp & paper industry (0.738). The results of the Kruskal-Wallis test indicate that there is a significant difference in technology gaps between types of industries. Third, 82.7% of the eco-inefficiency of energy-intensive industries in the 2010-2015 period stemmed from managerial failures in the form of underproduction and excess CO2 emissions, while the remaining 17.3% stemmed from technology gaps. Meanwhile, the sources of eco-inefficiency vary by industry type. In the palm oil, textile, chemical, glass & ceramic processing industries, managerial failures are more dominant, contributing 68.3%, 92.4%, 64.4%, and 84.2%, respectively. Conversely, in the pulp & paper, fertilizer, cement, and metal & steel industries, technological gaps are more dominant, contributing 85.0%, 62.3%, 80.5%, and 73.8%, respectively. Fourth, the results of the Tobit Panel Model regression show that the eco-efficiency of energy-intensive industries is influenced by foreign investment, scale effects, composition effects, energy intensity, and carbon intensity. According to industry type, the factors influencing eco-efficiency are as follows: (1) palm oil processing (foreign investment, scale effects, and composition effects); (2) textile (scale effects, composition effects, energy intensity, and carbon intensity); (3) pulp & paper (energy intensity); (4) chemical (scale effects, composition effects, and energy intensity); (5) fertilizer (composition effect, and energy intensity); (6) glass & ceramics (foreign investment, scale effect, composition effect, energy intensity, and carbon intensity); (7) cement (none); and (8) metal & steel (composition effect and energy intensity). The estimation and statistical testing results indicate that the Environmental Kuznets Curve (EKC) Hypothesis is not accepted in both energy-intensive industries and all types of industries. Based on these conclusions, this study recommends the development of long-term industrial areas located outside Java to reduce the technology gap, especially in the pulp & paper, fertilizer, cement, and metal & steel industries. In addition, investment in production technologies that produce energy efficiency, reducing dependence on fossil fuels, and developing alternative energy are needed to improve eco-efficiency. For further research, it is recommended to use balanced panel data, use other environmental indicators, and apply different approaches, models, and measurement techniques.
Keywords: Manufacturing, Industries, Technology