Do energy efficiency subsidies help in reducing energy consumption?
In Europe the households sector accounts for the 36.4% of the total European energy consumption and the percentage of energy efficiency building in Europe is around the 35%, this means that most of the energy produced goes to waste. Among all the measures that can be implemented in order to reduce energy consumption, subsidies seem to be a good alternative, but is it true?
This article is meant to investigate and add a contribution to the existent literature regarding this specific topic. To do so it is proposed as a systematic literature review that becomes then a meta-analysis of the collected studies.
To overcome the barriers regarding the reduction of greenhouse gas emissions, both the EU and national governments started to put in force policy measures to increase the energy efficiency level in the household sector. The EU set ambitious targets and policies to orient Member States towards energy efficiency in buildings. Among them, the Energy Performance of Buildings Directive (EPBD) 2010/31/EU and the Energy Efficiency Directive (EED) 2012/27/EU, both revised in 2018. Therefore, a huge number of policy measures have been implemented at the Member States’ level to promote energy efficiency. The MURE database collect and divide them between: legislative/performance, legislative/informative labelling, information/education, financial/fiscal measures which include subsidies or tax deduction.
As the main goal of this article is verifying if energy efficiency subsidies help in reducing energy consumption, the papers collected to conduct the research are: energy efficiency and households oriented, focused on subsidies or financial measures, they represent the last 11 years’ findings about the topic, and they are econometric or empirical. After having created the query and having done the research, 10 papers were selected. The majority of the selected papers (80%) compare the effect of policy instruments (mostly subsidies or financial measures) with the reduction of energy consumption. The other 20% refers to energy savings due to energy efficiency subsidies. The 70% of the database is composed by articles providing a strictly within country analysis of the effect of public funding on energy. The difference in difference method (DiD) represents one of the most used (30%) regression approach to investigate the ex-post impact of public financing on renewable energy sources along with the Panel data regression. The independent variables are classified in four groups: environmental and geographic variables, energy efficiency’s performance variables, socioeconomic variables and macroeconomics variables.
Subsequently, a dataset on STATA has been created to examine the data collected in the literature review and to conduct the meta-analysis. The type of meta-analysis that has been chosen to conduct this dissertation is the random effect (RE) one. The model assumes that the study effect sizes are different and that the collected studies represent a random sample from a larger population of studies. The goal of RE meta-analysis is to provide inference for the population of studies based on the sample of studies used in the meta-analysis. The RE model may be described as 𝜃̂ 𝑗 = 𝜃𝑗 + 𝜖𝑗 = 𝜃 + 𝑢𝑗 + 𝜖𝑗 where 𝑢𝑗∼ N (0, τ2 ) and, as before, 𝜖𝑗 ∼ N (0, 𝜎̂ 2 j). Parameter τ2 represents the between-study variability and is often referred to as the heterogeneity parameter. The restricted maximum-likelihood is the elected method to compute the between study variability, because it produces an unbiased, nonnegative estimate of τ2.
The output from the standard meta-analysis summary contains the heterogeneity statistics, the individual and overall effect sizes and other information. The overall effect size is presented at the bottom of the table and labelled as theta. It is computed as the weighted average of study’ specific effect sizes. For these data, the overall estimate is -0.368 with a 95% CI of [−0.630, - 0.106. The heterogeneity statistics reported in the header of the study and for instance, I2 = 99.98 indicates that 99.98% of the variability in the effect-size estimates is due to the differences between studies. Assessing and addressing heterogeneity was the last step of the research. The techniques used were the subgroup meta-analysis, the forest plot and the exclusion of studies. The heterogeneity identified was the statistical one, that means a mix between clinical (differences between studies) and methodological (differences between methodology used to investigate the problem).
In conclusion, the primary meta-analysis gave a results of energy consumption reduction of 0.368 kWh/year, but the results should be take carefully due to the presence of high heterogeneity.