Abstract

The carbon tax is an economic instrument strongly recommended to address the climate crisis. Still, its effectiveness is based mainly on ex-ante evaluation methodologies, and evidence of its causal effect using impact evaluation techniques is still embryonic. In the case of Chile, a carbon tax on large-fixed sources was introduced in 2017. So, this research aims to evaluate the effect of this tax on CO2 emissions, fuel consumption, and electricity generation through the difference-in-differences (DiD) method. For the above, databases were built at the thermoelectric generation unit level monthly, quarterly, and annually in the period 2014-2022. Specifically, the DiD method compares changes over time between a treatment group and a control group to estimate causal effects in observational studies, isolating other confounding influences. The results indicate that the application of the carbon tax (5 USD/tCO2) significantly reduced CO2 emissions (15.3 %-18.2 %), fossil fuel consumption (12.9 %-15.2 %), and electricity generation (15.1 %-15.8 %) in the treated units relative to the year before to this climate policy. The previous findings imply that annual national CO2 emissions are decreased between 4.5 % and 5.4 %. All the above confirms the environmental effectiveness of this instrument and the existence of some indirect effects on energy markets. However, no impacts on the efficiency and intensity of CO2 emissions are observed. Finally, it is concluded that a higher tax rate and expanded tax coverage are required to contribute decisively to the country's climate goals; for example, around 30 USD/tCO2 could substantially reduce emissions from thermoelectric power plants if linear impacts are assumed.