Estimating and Analyzing Energy Efficiency in German and Colombian Manufacturing Industries Using DEA and Data Panel Analysis. Part I: Energy-intensive Sectors

Abstract

In this article, data envelopment analysis (DEA) is employed to study the comparative performance of German and Colombian energy-intensive sectors between 1998 and 2005. The results of the DEA indicate that the great majority of energy-intensive sectors improved on this index during the sample period, demonstrating that energy input is an important variable within the production structure and a key element in technology development. At a second stage, regression analysis using panel data analysis reveals that several factors, including labor productivity, the share of electricity, investments and enterprise size can be considered determinants of differences in energy efficiency among German and Colombian energy-intensive sectors. Our results also show that different energy policies should apply, and that they should encourage the importance of energy efficiency in order to achieve a sustainable economic development and climate stabilization today and in the near future.

Keywords:

• data envelopment analysis
• data panel analysis
• energy efficiency
• energy intensive sectors

ACKNOWLEDGMENT

The author is grateful for the support provided by Universidad del Rosario and comments and suggest of Professor Wolfgang Irrek of the Hochschule Ruhr West. Any remaining errors are the responsibility of the author.

Notes

¹ This concept could be expressed as a formula: energy = progress = civilization (Basalla, 1980).

² German energy tax law defines EISs as sectors where the cost of energy is above 3% of total costs.

³ DEA produces relative efficiency, rather than absolute efficiency for each DMU under consideration. DEA evaluates a DMU as technically efficient if it has the best ratio of any output to any input and this shows the significance of the outputs/inputs taken under consideration (Halkos and Tzeremes, 2005).

⁴ Input package (x₀): capital, labor, material and energy; output package (y₀): gross production.

⁵ This is the concept of technical efficiency of the firm according to Debreau (1951) and Farrel (1957).

⁶ The ISEC classifies data according to kinds of economic activity in the fields of production, employment, gross domestic product, and other statistical areas.

⁷ For Germany, we use the Statisches Bundesamt Deutschland (German Bureau of Statistics) and for Colombia, we use the Departamento Nacional de Estadística (Colombian Department of Statistics, DANE).

⁸ This variable is calculated taking into account the categories established by the German and Colombian statistical offices based on the number of workers and output levels for every manufacturing industrial sector.

⁹ Technology innovations play a central role by enabling reductions in energy use, and these innovations change the amounts of various inputs (energy, material, labor) in the production function required to produce a given level of satisfaction (utility). Typically, technology innovations create opportunities to save energy, save other inputs, or increase utility (IAC, 2007).

¹⁰ Estimations with tobit model and maximum likelihood estimation provide similar results.