| COICOP |
Indirect emissions
|
||||
|---|---|---|---|---|---|
| Austria | EU | ROW | Total | ||
| Bread and cereals | 0.12 | 0.16 | 0.13 | ||
| Meat | 0.15 | 0.11 | 0.09 | ||
| Fish and seafood | 0.21 | 0.20 | 0.12 | ||
| Milk, cheese and eggs | 0.13 | 0.14 | 0.10 | ||
| Oils and fats | 0.13 | 0.12 | 0.09 | ||
| Fruit | 0.10 | 0.11 | 0.16 | ||
| Vegetables | 0.06 | 0.12 | 0.20 | ||
| Sugar, jam, honey, etc. | 0.11 | 0.14 | 0.12 | ||
| Food products n.e.c. | 0.10 | 0.14 | 0.18 | ||
| Coffee, tea and cocoa | 0.10 | 0.11 | 0.16 | ||
| Source: EXIOBASE3 | |||||
Emissions inequality
Hendrik Theine 
Stefan Humer
Mathias Moser
Matthias Schnetzer
What is this paper all about?
The social consequences of carbon taxation are closely linked to the income and expenditure patterns of private households. We combine expenditure data from the Household Budget Survey and emissions multipliers from EXIOBASE3 to analyse the distribution of the carbon footprint in Austria. The results indicate a strong variation in greenhouse gas (CO2e) emissions along the income distribution, with the top income decile emitting 4.1 times more than the bottom income decile. Moreover, we assess the implications of carbon taxation and find socio-demographic factors that explain why some households have higher tax-to-income ratios. Socially balanced carbon mitigation policies should focus on these excessive emitters as they might be particularly exposed to CO2 taxation.
How to estimate the carbon footprint of households?
To calculate the consumption-based emissions of private households, data on expenditure and the carbon content of the consumed goods and services is needed. The latter is called CO2 multipliers which are factors used to estimate the total carbon dioxide emissions associated with the production and consumption of goods and services. These multipliers take into account not only the direct emissions from the production process but also the indirect emissions from the entire supply chain. By applying CO2 multipliers, researchers can better understand the full environmental impact of different economic activities and identify opportunities for reducing emissions. In the context of this paper, CO2 multipliers help to quantify the carbon footprint of households based on their consumption patterns, providing insights into how different income groups contribute to overall emissions.
First, direct and indirect emissions are obtained from the multi-regional input-output database EXIOBASE3, which maps global economic relationships and their environmental impacts for 44 countries and five rest-of-world (RoW) regions. It provides detailed information on 163 industries and 200 products, making it suitable for analyzing GHG emissions.
The allocation of CO2e emissions to households is based on the Austrian Household Budget Survey (HBS) of 2009/10, a representative expenditure survey by Statistics Austria. It covers 6,534 households with detailed consumption expenditure data, disposable income, and socio-economic characteristics, enabling analysis of the distributional aspects of GHG emissions.
The distribution of income, expenditure, and emissions
Inequality in incomes is typically higher than in expenditure, which is in turn higher than emissions inequality. The figure below shows the distribution of income, expenditure and emissions with respect to the deciles of disposable household income. The top decile of the income distribution receives 22% of income, spends 18% of expenditures, and causes 17% of emissions. This observation corresponds to the expectation of a decreasing marginal propensity to consume along income deciles. The bottom decile of income distribution, in contrast, accounts for just 3% of income, 4% of expenditure and 4% of emissions. The top income decile is thus responsible for roughly 4.1 times more emissions than the bottom decile. The bottom half of the income distribution in Austria emits 34% of total emissions and the top half around 66%.
For the ease of analysis, we aggregate emissions from all products and services to a total of six expense groups: food, housing, energy, mobility, goods, and services. The allocation of individual consumption goods to the broader expense groups is primarily determined by the hierarchy of the COICOP classification (12 main groups) and further assigned manually to the six remaining groups. Next, we analyse the CO2e emissions by expenditure group and income decile.
It is important to distinguish between domestic and external emissions when analyzing carbon footprints. With the EXIOBASE3 dataset we calculate that approximately 55% of emissions in Austria are domestic, with the remaining 45% split between other EU countries (20%) and the rest of the world (25%). This distribution remains consistent across different household income levels.
From a policy perspective, focusing solely on domestic emissions would overlook a significant portion of the total emissions. For example, the national carbon tax in Austria only addresses a quarter of the emissions (i.e. direct emissions), while a carbon border tax and the EU ETS would cover additional portions. Therefore, comprehensive measures that consider both domestic and external emissions are essential for effectively addressing the climate crisis.
Excessive emitters
Departing from the perspective of average emissions, there is a large dispersion of emissions within each income decile. While median and mean emissions are clearly rising with household income, the variation within each decile is substantial. Some households emit considerably more CO2e emissions than their income peers and might be seen as excessive emitters.
The median annual emissions across all households is roughly 20 metric tons. Households from all income deciles are found in the upper half of emitters.
Some households at the bottom of income distribution even emit more than households with very high incomes.
In contrast, a few households in the top income decile emit less than the overall median. However, the highest share of excessive emitters are found at the top.
Distributional impact of a carbon tax
In a static exercise, we introduce a comprehensive carbon tax and identify households with the highest tax-to-income ratio as potentially overburdened. The literature suggests a CO2e price of 70 to 140 Euro per metric ton by 2030. We consider a global CO2e price of 100 Euro, fully passed to consumers, and apply this tax to all households’ carbon-related expenditure. We calculate the potential tax burden relative to household income to identify high tax-to-income ratios. The average ratio across deciles ranges between 9.2% in the bottom and 4.9% in the top income decile.
Finally, we examine the relationship between socio-demographic variables and the likelihood of being overburdened by taxation. Households owning multiple cars or using coal, oil, or gas for heating are more likely to be overburdened by carbon taxes due to higher emissions. Conversely, renewable heating reduces emissions and tax burdens. Urban living, apartment blocks, and newer buildings are linked to lower emissions and tax burdens. Larger households and older buildings tend to have higher emissions. These findings align with studies showing suburban and rural living is more energy-intensive than urban centers, highlighting “urban economies of scale” in housing and mobility.
What do we learn from the paper?
This paper examines the relationship between income distribution and the carbon footprint in Austria. The findings show that emissions increase with income, driven mainly by mobility and energy consumption. The top income decile emits 4.1 times more than the bottom decile.
The study also reveals that emissions are evenly distributed between direct, local, EU-based, and rest-of-world sources. A national CO2e tax would cover only 25% of emissions, while a carbon border tax and the EU ETS would cover additional portions.
The main takeaway is that carbon taxation needs to be universal to capture all consumption-induced emissions. It should also consider the social impact, as lower-income households with high emissions could be adversely affected. A balanced approach would include financial incentives for cleaner alternatives and regulatory measures to prevent companies from passing carbon taxes to consumers.
This document was created with Quarto, closeread, R, and ggiraph.
The webpage is based on an article written by Hendrik Theine, Stefan Humer, Mathias Moser and Matthias Schnetzer and published in Ecological Economics.
matthias.schnetzer@akwien.at mschnetzer.github.io matschnetzer