The productive structure and consumption of final energy in Spain: an input–output analysis from a production perspective

ABSTRACT

Using an alternative input–output analysis based on the responsibility of the production of the total output of each sector, we determine the relevant sectors in energy consumption in Spain. Our approach offers a different and complementary vision to the usual analyses based on the responsibility of the final demand to help guide environmental and energy policies. The analysis also enables us to observe the importance of final energy consumption by type for the different sectors. We use the input–output table for 2015 from INE and data on energy consumption from EUROSTAT. The most relevant sectors in terms of the energy consumption that their total production induces in other sectors are “Food, beverages and tobacco products” and “Construction,” while various service sectors also stand out for inducing energy consumption in other sectors. In these sectors, policy measures aimed at improving the efficiency in the use of energy-intensive inputs and the substitution of these inputs by others less energy intensive would be effective. The most relevant sectors for being induced to consume energy by the production of other sectors are “Electricity and gas” and “Land Transport.” The incorporation of technological improvements aimed at improving energy efficiency, as well as the adoption of best practices, would be more effective in these sectors. Finally, the most important sectors due to their own final demand are “Electricity and gas,” and “Coke and refined.” In these sectors, policies aimed at reducing their final demand would be effective.

KEYWORDS:

• Energy policy
• final energy consumption
• input–output analysis
• productive structure
• responsibility of production
• total sectoral output

Acknowledgements

We are grateful to three anonymous reviewers and the editor for their helpful comments. We acknowledge support from project PID2021-126295OB-I00 (Spanish Ministry of Science Innovation).

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ In the present work (^) expresses the diagonalization or expression in the form of a diagonal matrix of a vector. Vectors are written in lowercase and bold. Matrices are written with capital letters and in bold. (’) expresses the transposition of both matrices and vectors. Scalars are written in italics.

² In the case of a conventional demand analysis, a structural path analysis could be more operative, in the sense of Lenzen (2003), although this is not our case, as explained above.

³ If A is a productive matrix, a_ss<1 and, therefore, (1 – a_ss)⁻¹ exists (Dorfman, Samuelson, and Solow 1958).

⁴ For a simple and rigorous analysis of Schur’s complement and its use in determining the block inverse see Lu and Shiou (2002).

⁵ Although in a different framework, and from a conventional analytical perspective, Dietzenbacher (2005) suggests an interpretation of this type.

⁶ The average is used as a reference magnitude (as in Alcántara and Padilla 2020, in an analysis of the key sectors in CO₂ emissions). However, other references, such as the median or the quartiles (as in Alcántara and Padilla 2003), could also be used.

Additional information

Funding

The work was supported by the Ministerio de Ciencia e Innovación [PID2021-126295OB-I00].