Epsilon Archive for Student Projects

Abstract

The Swedish Reduction Mandate (Reduktionsplikten) was implemented in July of 2018 and requires gasoline and diesel suppliers to mix in a certain percentage of biofuel into their products. This paper evaluates the effect of this mandate in terms of emissions from Sweden’s transport sector. We use the difference-in-differences strategy, which is well adapted to policy impact evaluation, to estimate the effect of the mandate on emissions. The treatment group of the model is composed of gasoline and diesel which are subject to the policy whilst the control group consists of a fuel that is not affected by the mandate (such as ethanol). Biogas and natural gas are additional control fuels that are incorporated to gain a wider understanding of the relationship between treated and untreated observations. The results from the main regression suggest that there seems to be a significant and negative relationship between the emissions from regulated fuels in the transport sector and the intervention. However, as the parallel trends assumption wasn’t met by any of the treatment and control group combinations, the results of the main regression cannot be deemed valid. Incorporating a regression controlling for fuel specific linear trends was therefore needed to provide valid results. When controlling for linear trends, the results suggests that the mandate led to an emission reduction but that a zero effect cannot be ruled out as the treatment effect is not significant. Since the mandated percentage of biofuel blended into the fuels increases, the observed effect is expected to increase significantly over time.