On the causality between energy efficiency and technological innovations: limitations and implications

ABSTRACT

Although evidence of technological efficiency effect on decreasing energy intensity is documented in a number of studies, there has been a very few detailed investigations of the causality between the variables. Thus, this paper examines the causal relationship between energy efficiency and technological innovations in nine EU countries during 1980–2013. Using the bootstrap panel Granger causality approach, the study finds evidence of a bidirectional causality in Germany, unidirectional causality running from technological innovations to energy efficiency in Denmark, Sweden, France and Austria, and from energy efficiency to technological innovation in Belgium. No causality at any direction is found in the Netherlands, the UK, and Italy. The lack of causality maybe due to a “rebound” effect”, reducing the energy savings from the efficiency improvement. To achieve the global goal of energy intensity, the policy implication of our results suggests that enhancing technological innovations and energy efficiency must be accompanied by taxation and regulations or followed by alternative energy sources such as solar or wind energy generated power in achieving optimal climate benefits.

KEYWORDS:

• Energy efficiency
• rebound effect
• panel causality
• technological innovation
• R&D
• patent

Acknowledgments

The author would like to thank two anonymous referees for their useful comments. Of course, any remaining errors are my responsibility.

Notes

¹ We use patent data as an indicator of innovation since economists have found that patents provide a good indicator of R&D activity, as patent applications are usually filed early in the research process (e.g., Griliches 1990). Furthermore, patents not only serve as a measure of innovative output, but are indicative of the level of innovative activity itself. That is, unlike R&D data, international patent data can be used to track patterns of diffusion (e.g., Lanjouw and Mody 1996). However, patents are positively correlated with innovations in a statistically significant manner (Igami and Subrahmanyam 2015). Although total factor productivity (TFP) is a comprehensive measure of technological innovation, there are some deficiencies when we use TFP. On the one hand, the growth of TFP contains other factors except innovation; on the other hand, the computational process of TFP is more complex and parameters are not unique, which will influence results.

² Real crude oil import prices were missing for France between 1980 and 1991. This was estimated by using electricity prices which was derived from the International Energy Agency: Energy Prices and Taxes.

³ The Akaike and Hannan-Quinn information criteria (HQIC) was also used to determine the number of lags but it did not change the outcome.

⁴ The procedure to generate bootstrap samples and country-specific critical values was carried out in TSP software.