The German Eco Tax and Its Impact on CO2 Emissions

Published: 13 September 2021| Version 1 | DOI: 10.17632/t3wxp4xpjf.1
Petrik Runst,


The data can be used to study the introduction of quasi-carbon-taxes in the German transportation sector and to evaluate the effectiveness of environmental taxation. The data permits the use of synthetic control methods. The results indicate that the carbon price increase by about 66 €/t CO 2 led to a considerable decline of transport emissions by 0.2 to 0.35 t per person and year. It can be shown that the sales share of diesel cars quickly increases after the reform, whereas the fuel efficiency of non-diesel cars increases with a three year time lag.


Steps to reproduce

Data set 1 (European data) was largely obtained from Eurostat and the European Commission, supplemented by data from the OECD, the International Energy Agency (IAE) and the Ameco database. It contains country-level data for 31 European entities, from which we remove Sweden because it has implemented and raised a carbon tax during the period under consideration. We are left with 23 countries with non-missing information. The panel data set covers the period from 1993 to 2005. The limited (pre-treatment) timeframe represents the main disadvantage of data set one, while the wider availability of relevant covariates such as fuel prices and road length represents its main appeal. Data set 2 (global data) is based on CDIAC (Carbon Dioxide Information Analysis Center ) data. It enables us to expand the analysis back to 1970. This allows for an improved fit in the pre-treatment period. The dataset is available from Boden et al. (2017) and their methods are outlined in Boden et al. (1995). As sources of CO2 emissions, they consider the consumption of several fuels and cement production. Data on fuel consumptions can be can be obtained from the UN energy statistics. Fuel consumption is then translated into CO2 emissions using a simple equation considering the properties of different fuel types. Bunker fuels for international transport are excluded. Data on cement production are obtained from the US Department of Interior’s Bureau of Mines. The resulting emissions can be calculated using the tons of cement production and the average calcium oxide content of the cement produced. This dataset has some noteworthy features: until 1990, the data for Germany was disaggregated into the western Federal Republic of Germany and the eastern German Democratic Republic. Afterwards, the data is only available for reunited Germany. The unification process was associated with a de-industrialization process in the east. Consequently, there is a CO2 emissions decline around the year of reunification, especially from 1990 to 1991. As the overall trend around this year is consistent, the data can be used for the SCM analysis, although we expect a somewhat lower pre-treatment fit for 1990 to 1992. The amount of CO2 emissions from traffic is not directly available from the CDIAC dataset, although it can be easily calculated in combination with the World Bank data on the share of traffic emissions from fuel combustion. As additional predictors, GDP per capita from the World Bank and the density of the road network are included. The control variable of road density was obtained from the Global Roads Inventory Project (GRIP, see Meijer et al., 2018). The raster file map contains worldwide data on road density. Each raster cell has a size of 5 arcminutes (Or roughly 9 km). Within each cell, road density is measured in meters per square km. We use the QGIS software to generate an average road density by country. Data set two contains 25 developed countries.


Georg-August-Universitat Gottingen


Transport, Emissions, Carbon, Taxation