Inflation and Government Response – Distributional Impact on Austrian Households

The measures targeted at households are a combination of income supporting and price reducing measures, of means-tested and non-means-tested measures, and of temporary and permanent measures. In the beginning of the energy and inflation crisis, the focus was on income support measures. Starting in autumn 2022, as inflation began affecting a wider range of products and services and turned out to be more persistent than initially expected, price-reducing measures became more relevant. Table 3 provides a list of measures included in the distributional analysis, along with their associated relief volumes.

The total relief volume of government measures to support households amounts to between 1.3% and 1.6% of GDP per year in the analysed period 2022 to 2024. In 2022 and 2024, the focus is on income support measures with a share of the total relief volume of more than 80%. In 2023, price reducing measures gain relevance due to temporary electricity price subsidies and the share of income support measures amounts to 63%. Details on the design of measures are provided in Appendix A.1.

Table 4 provides a summary on the type of measures (tax, transfer or subsidy), on means-testing (yes/no) and on the time span (temporary or permanent).

In 2022, government measures were largely characterised by temporary, non-means-tested transfer measures. The relevance of permanent measures increases from 2023 onwards, with the start of the annual indexation of income tax brackets and of further social transfers.⁶ By 2024, already 80% of the total relief volume announced so far stems from permanent measures. The share of means-tested measures is low throughout the entire period.⁷ In 2022 and 2023, roughly one quarter of the total relief volume is allocated to means-tested measures, dropping to only 5% in 2024. On the one hand, the low share of means-testing results in broader parts of the population benefitting from the measures. On the other hand, it increases the fiscal cost and higher incomes can result in higher demand and fuel inflation.

The relevance of tax measures increases from 29% of the total relief volume in 2022 to 73% in 2024. Subsidy measures primarily are of relevance in 2023 with a share of 31%. Transfers dominate in 2022 and lose relevance over time.

In this section, we analyse the distributional effects of the measures by deciles of equivalised disposable income. Table 5 shows the allocation of the total relief volume from 2022 to 2024 for each measure and decile.

Altogether, the relief volume increases with income. While 9% are allocated to the first decile, the top decile receives 12%. This is mostly driven by income support measures, which are more unevenly distributed⁸ – besides also making up a higher share of the total volume. Within income support measures, the indexation of income tax brackets, which accounts for 45% of the volume, has a steep gradient. The first decile receives 2% of its relief volume, while the tenth decile receives 19%.⁹ In contrast, lower-income deciles receive more than a proportional share of measures for families, of targeted one-off payments to low-income households, and of the indexation of further social transfers. Certain households in the upper part of the income distribution also benefit from these measures, as some targeted measures (e.g. one-off payments to unemployed) as well as some social transfers (e.g. child benefits) are not subject to means-testing based on household income. Broad-based one-off payments to all households are distributed almost equally across deciles.¹⁰

In a next step, we analyse the progressivity of the measures by calculating relative changes in average equivalised income across deciles of the income distribution.¹¹ Figure 3 shows the results for each year and measure. Overall, there is a clearly progressive effect in all years. The relief amounts to between 8.5% and 9.4% of income per year in the lowest decile and between 1.3% and 2.5% in the top decile.

Figure 3

Download asset Open asset

The one-off payments in 2022 led to a higher relative increase of incomes in the lower deciles as they were distributed almost equally across deciles in absolute terms. They increased the average equivalised disposable income by 4.6% in the lowest decile and by 0.7% in the highest. One-off payments targeted to low-income households are more progressive by design. They raise incomes in the first decile by 2.7% in 2022 and 3.0% in 2023.

Measures for families have a progressive effect in all years, but especially so in 2023 and 2024 when they contain transfers and tax credits for low-income earners. Their biggest effect on average equivalised disposable income is in the first decile in 2024 (+2.7%).

The temporary increase of commuter tax allowances and credits is a regressive measure, primarily because the share of commuters rises with income. Compared to the other measures, it has a smaller volume, and it increases income by rounded 0.0% at the bottom and by up to 0.2% in 2022 and 2023 in the higher deciles.

The indexation of income tax brackets starting in 2023 has an inverse U-shaped relative relief. The biggest effects on income occur in the seventh decile, with increases of 0.9% in 2023 and 2.5% in 2024. In contrast, the indexation of further social transfers is progressive, increasing average equivalised disposable incomes by up to 0.8% in 2023 and 2.1% in 2024 for the bottom decile.

The price reduction through energy tax measures and the electricity price subsidy both have a progressive effect, benefitting the bottom deciles most relative to income. The biggest effect is in 2023, when electricity prices before applying the subsidy were higher than in 2024. In addition, power grid costs were only subsidised in 2023. In total, the estimated relief due to price reduction amounts to 3.0% of disposable income in the bottom decile and to 0.5% in the top decile in 2023.

Analysing the effect on average disposable income of each decile masks the high degree of heterogeneity even across households within the same decile. Figure 4 shows the dispersion of relief due to the income support measures.¹² In general, this dispersion is greater in the lower deciles of the income distribution than in the higher deciles. First, differences between incomes across households within the same decile are relatively larger in the bottom deciles. Then, the relief from one-off payments expressed relative to income varies more. Second, not all households benefit from the targeted measures as they are tied to certain requirements (e.g. social assistance, unemployment, minimum pensions, children in the household). In the higher deciles, income tax-related measures are a more important component of relief, and their effect varies less across households with similar income.

Figure 4

Download asset Open asset

Figure 5 displays our results for income and expenditure in the years 2022 to 2024 compared to the year 2021. The left panels show the average change in income and the inflation burden per capita for the total population as well as for the deciles of the income distribution. The right panels show these numbers relative to the respective average equivalised disposable income.

Figure 5

Download asset Open asset

As shown in Section 4, income support from the government response to high inflation has a progressive effect and leads to a higher relative increase in disposable income in the lower deciles. In 2022, persons in the lower deciles received a higher inflation relief per capita, not only in relative but also in absolute terms. In 2023 and 2024, the indexation of income tax brackets favours high earners more in absolute terms. In those years, the higher relative relief in the lower deciles is due to the denominator effect arising from their lower disposable incomes. Similarly, the eco-social tax reform favours high-income earners more in absolute terms but leads to higher relative increases in disposable income for low-income households. In contrast, the estimated residual of other income changes tends to increase income in the upper deciles by more, even in a relative sense. This result is explained by income from employment being more important in the higher deciles and by expected wage growth per person being stronger than the growth of transfers. An additional factor especially in the top decile is investment income, which was comparably low in 2021 and grew strongly in 2022.¹⁴

The burden of inflation increases in absolute terms with income but decreases in relative terms. Hence, the excess inflation has a regressive effect. Even though we do take account of different consumption baskets, the burden is roughly proportional to consumption and heterogeneous inflation rates play a minor role (see discussion in the next subsection). As a result, the increase with income in absolute terms mostly reflects higher consumption levels of higher-income households. The decrease of the burden relative to disposable income in the higher deciles reflects average propensities to consume that are decreasing with income. In the lower parts of the income distribution, the average propensities to consume are larger than one (negative saving), which drives the result that income increases are not sufficient to cover expenditure increases in 2023 and 2024 in the bottom decile. Price reducing measures only played a minor role in 2022 and are most important in 2023. Their relief volume is distributed relatively equally across deciles, such that the relief per capita is similar. Relative to disposable income this results in a higher relief for low-income households, on average.

Comparing the absolute change in nominal income with the burden of inflation lets us answer the question, whether households on average receive enough income to cover their additional expenditures without changing the consumption behaviour. If the changes in income and expenditures were equal, nominal (dis)savings of households would remain unchanged. Real (dis)savings would go down in this case, however. If the average propensity to consume is larger than one, even a constant real disposable income is not sufficient to cover the increase in expenditures.

Note that this question compares households in a decile of the income distribution with those households that were in the same decile in 2021. Those are not necessarily the same households as there is income mobility. Furthermore, the results show averages within deciles. The affectedness of single households may deviate significantly from these averages (see for instance Figure 4 on the dispersion of income relief measures).

In 2022, nominal income increases were enough to exceed the inflation burden on average. Overall, the positive effect was €160 per person or 1.0% of equivalised disposable income. In terms of equivalised income the effect is positive on average in all deciles.¹⁵ While this result holds on average, it does not hold for all households, as there is substantial heterogeneity of nominal income changes and households’ inflation burden.¹⁶

In 2023, the estimated nominal income increase for the total population (9.8% in excess of trend growth of 2% per year) is about the same as the burden of inflation (9.7% of equivalised disposable income). Especially in the lower deciles, the burden significantly outweighs the relief. In the first decile, the gap is around €520 per person or 5.7% of equivalised disposable income. Partly, this is due to income support being smaller overall and less progressive in 2023 than in 2022. In addition, the regular indexation of pensions lags inflation, such that real pension incomes decrease. Wages are also adjusted with a lag, but from 2021 to 2023, the real decrease of average wages (-3.3%) is significantly smaller than that of pensions (-8.1%).

In 2024, the expected annual adjustments of nominal wages and pensions are significantly above the annual inflation rate. This means that average annual wages are expected to get close to the level of 2021 in real terms (+0.2%). The indexation of pensions from 2021 to 2024 will be 3.0% short of inflation, compared to 8.1% from 2021 to 2023. Together with the income relief from fiscal policy, the rise in nominal incomes is expected to significantly outweigh the inflation burden. On average, across all households, the surplus is around €1,230 per person or 5.5% of equivalised disposable income. As the income relief becomes less progressive in 2024 (see Figure 3) and wages grow faster than pensions, there is still an estimated gap in the first decile. But it is smaller than in 2023, amounting to around €300 per person or 1.6% of equivalised disposable income.

In this subsection, we analyse the differences in the inflation burden across deciles of equivalised disposable income. The overall effect of inflation is regressive as shown in Figure 5. In theory, this could be due to higher personal inflation rates in households with lower income. The dominant effect in the years 2022 to 2024, however, stems from variation in average propensities to consume. The same amount of inflation hurts households with higher propensities more relative to income. As propensities to consume on average decrease with income, so does the inflation burden. To show the importance of this effect compared to idiosyncratic inflation rates, we decompose the total burden into three components:

1. Effect of average inflation: The burden from this effect is calculated by applying the same average inflation rate to all households. In addition, we keep the population constant in all years. That is, we don’t adjust the weights in our sample to account for e.g. population growth and employment changes and we also keep the classification of households into deciles constant over time. The resulting burden is proportional to the average propensity to consume.

2. Effect of idiosyncratic inflation: In the second step, we use the idiosyncratic inflation rates resulting from the different consumption patterns to calculate the burden. But we still hold the population constant across time. The difference to the burden after the first step reflects heterogeneity across inflation rates.

3. Effect of changes in population: This is the residual burden not explained after step 2. It reflects changes in the weights of our sample households as they are calibrated for each year to reflect e.g. changes of employment (see Appendix A.4 for the methodology). In addition, even with constant weights, households can switch between deciles of the income distribution over the years due to the uprating of income and changes in taxes and transfers.¹⁷

Table 6 shows the results of this decomposition exercise for the years 2022 to 2024. For each of the three effects, we calculate the difference between the effects in the deciles and the effects in the total population. For example, in 2022 the total inflation burden is 3.6 percentage points higher in the first decile than in the total population. This primarily results from a deviation of +4.4 percentage points due to average inflation and the average propensity to consume (first effect). Below-average idiosyncratic inflation rates in the first decile reduce the burden by 0.8 percentage points (second effect) and changes in the population have a minor impact (+0.1 percentage points).

The effect of idiosyncratic inflation across deciles varies between -0.8 percentage points and +0.3 percentage points in the three years. Its mean absolute deviation is around 0.1 percentage points and the standard deviation around 0.2 percentage points. It is dominated by the first effect of the interaction between aggregate inflation and varying average propensities to consume. The dispersion due to the first effect increases over time as aggregate inflation (in excess of 2% per year compared to 2021) increases in years 2022 to 2024. Already in 2022, the deviations from the burden in the total population vary between -1.4 percentage points in the tenth decile and +4.4 percentage points in the first decile. The resulting mean absolute deviation across deciles is 1.1 percentage points and the standard deviation amounts to 1.5 percentage points. The third effect due to changes in the population is small and also less important than idiosyncratic inflation rates.

See Prammer and Reiss (2023) for an overview of government measures in Austria compared to the EU.

According to the European Commission’s Winter Forecast 2024, the annual inflation rate (HICP) for the euro area was 8.4% in 2022, 5.4% in 2023, and is forecast to be at 2.7% in 2024.

Inflation rates across the income distribution are calculated using the structure of equivalised consumption expenditure from the household budget survey. The resulting aggregate annual inflation rates for the years from 2022 to 2024 are, on average, about 0.3 percentage points lower than the official inflation rates shown in Figure 1, which use the CPI-basket.

The first adult is assigned a weight of 1.0, further persons aged 14 and over receive a weight of 0.5 and children under the age of 14 are given a weight of 0.3.

For COICOP categories with no exact match in the CPI we use the lowest available CPI category. As part of the survey period of the Household Budget Survey 2020 (March 2020 until mid-June 2020) falls into the period of COVID-19 restrictions, we apply a correction factor to consumption in this period. We calculate the correction factor as the percentage difference between the actual average per capita consumption during the COVID-19 period and the per capita consumption that would be observed if the seasonal pattern in 2019/20 was similar to the seasonal pattern in the last Household Budget Survey in 2014/15. We only apply this correction factor to goods for which consumption data is collected via a household book, while we don’t correct consumption data that is collected via a questionnaire covering the past six or twelve months. The quantitative effects of applying this correction are minor.

In the past, tax brackets were adjusted at irregular intervals and the last major reform (eco-social tax reform) was passed in early 2022. The resulting share of income tax revenues to taxable income would have been comparably low in the years 2022 to 2024. The counterfactual for our analysis is that there wouldn’t have been any further changes to the income tax code until 2024.

In Table 3 we differentiate between means-tested measures at the household and the personal level. From the latter also households in the upper part of the income distribution can benefit, if a person benefiting from such a measure lives in a high-income household (e.g. one-time payments for unemployed persons).

To estimate the distributional effect of the price cap on basic electricity consumption, it was assumed that electricity contracts do not differ systematically across the income distribution. Estimated differences between households result only from the amount of their consumption and the location of their main residence.

It should be noted that this measure undoes the burden from bracket creep that would similarly increase with income.

Differences primarily exist due to some means-testing and taxation for high-income earners and due to the unequal distribution of children.

Formally, price-reducing measures increase real disposable income of households. We calculate their relief as the reduction of equivalised household expenditures relative to (nominal) equivalised disposable income.

There is not enough information available on electricity price differences between households to estimate the dispersion due to the electricity price subsidy. Therefore, we cannot calculate the heterogeneity for the price reducing measures.

The main parts of the reform were passed in early 2022. Some measures targeting low incomes and the first tax bracket were already introduced from 2021. Our counterfactual income in 2021, which serves as the basis for the calculation is income in the absence of the eco-social tax reform.

See Appendix A.3 for a decomposition of this income component into earned income, pension income, unemployment income, and other income sources.

Looking at income and expenditure per person, the effect is slightly negative in two deciles. This is due to the equivalisation of income when calculating relative changes as opposed to displaying the (non-equivalised) per capita change in absolute terms. The same absolute increase in income raises the average equivalised income by more when it goes to multi-person households than when it goes to single-person households.

Here, the heterogeneity of average propensities to consume within deciles plays an important role. We do not show this dispersion across households due to concerns of overestimating the variance. Because parts of the consumption expenditures of an individual household are only recorded over two weeks and extrapolated for the whole year in the household budget survey, some households might spend exceptionally high or low amounts in those weeks. For income data, this is less of a concern as it is recorded for the whole previous year for each household in the dataset.

Deciles are classified in each year according to equivalised disposable income in the baseline scenario without fiscal reforms.

Fessler et al. (2023) find that the group of households that struggle predominantly from high inflation consists of those already facing a challenging financial situation during periods of low inflation (e.g. unemployed, working poor, single parents).

The climate bonus was introduced along with the introduction of a tax on carbon emissions in the non-EU-ETS sectors. The amount is differentiated depending on the rural typology and the access to public transport.

The voucher and the one-off payment to self-employed are partly means tested. High-income households (approximately 10% of households) are not entitled to redeem the voucher. Regarding the voucher, a non-take-up rate of 30% of eligible households was taken into account.

We take into account that in some months local energy providers also allowed existing customers to change to a cheaper contract.

Note that, for this definition, fiscal measures not only include those in response to inflation but also those that are part of the eco-social tax reform.

In 2022, the recovery from the COVID-19 crisis was still taking place and the number of unemployed fell by 20.7% compared to 2021 (see Table 2 in Section 3).

All components are net of social insurance contributions and taxes. If a person has different kinds of income (e.g. pension and employment income), the average income tax rate is applied to all parts.

Only in 2024, gross pensions are expected to grow faster than gross average wages.

The income support measures introduced by the government to cushion the effects of high inflation are the following:

• One-off payments to all households: In 2022, the climate bonus, with a regular amount of €100 to €200 per adult¹⁹ (halved for children), was increased to €500 per adult (halved for children). In addition, each household received a voucher of €150 to reduce the electricity bill and a subgroup of self-employed persons received an (additional) one-off payment of €410 to tackle high energy cost.²⁰

• One-off payments to low-income households: Vulnerable groups, like recipients of unemployment benefits, social assistance, or a minimum pension received additional one-off payments. To also support low-income earners and low-income pensioners excluded from these one-off payments a tax-credit for employees and self-employed and two separate one-off payments for pensioners (2022 and 2023) were introduced. In addition, in 2023 local state governments received a transfer from the federal government to fund transfers related to higher housing and heating costs.

• Measures for families: In August 2022, the regular child benefit was increased by €180 per child. From July 2023 to December 2024, vulnerable families (transfer recipients, lone parents, and lone earners with low income) receive an additional €60 per child and month. An already scheduled increase in a general tax credit for children was brought forward, and a tax credit for children targeted at low-income parents was increased permanently.

• Commuter support measures: Commuter tax allowances and tax credits are increased from. 2022. to June 2023.

• Indexation of income tax brackets: Starting in 2023, tax brackets are indexed with inflation each year.

• Indexation of further social transfers: Some social transfers, most importantly a non-means tested child benefit, were not indexed with inflation in the past. From 2023 onwards, the list of indexed social transfers was extended.

Most transfers are exempt from income tax, an exception being the top-up to the climate bonus, which is partly taxable for persons with a taxable income above €90,000 in 2022.

In addition to these income support measures, the government introduced the following price reducing measures which also lower the inflation rate:

• Energy tax measures: From May 2022 to December 2024, the tax rate on electricity and gas consumption is reduced by approximately 90% to the minimum tax rate according to EU legislation. In addition, the introduction of the carbon price on emissions not covered by the EU-Emissions Trading System was postponed by three months to October 2022.

• Electricity price subsidy: A price cap of 10 Cent per kWh for basic electricity consumption (2,900 kWh per year per household) was introduced from December 2022 until the end of 2024. The difference between 10 Cent and the household’s actual electricity price is subsidised by the government up to a maximum of 30 Cent per kWh until June 2024 and up to 15 Cent per kWh from July to December 2024 Households with more than three persons and low-income households receive additional transfers via the electricity bill. The relief volume of the price cap is subject to uncertainty as it depends on the individual household’s electricity contract. In our estimation, the fiscal cost of this subsidy is €2.7 billion, well below the provided funds (€3.8 billion) according to the law. A further measure to reduce electricity cost is a subsidy which partly covers power grid loss cost in 2023 with an additional fiscal cost of €0.3 billion.

This appendix describes how we model the electricity price subsidy, in particular the price cap for the first 2,900 kWh per year and household. The household budget survey contains data on expenditures for electricity, which we convert to kWh per household by using Eurostat’s electricity prices for household consumers from the survey period. In particular, we use a fixed cost of €135 per year and a variable cost of €0.171 per kWh, including all taxes and levies.

The relevant electricity price for the subsidy from December 2022 to December 2024 does not include taxes and levies. For the relevant period from 2021, we collected public data from the Austrian regulator E-Control, which are available on a monthly basis. For each region, we use the prices of the local electricity provider as well as the prices of the cheapest available provider. In line with E-Control’s statistics, 80% of our households in the sample are allocated the price of the local provider. For the remaining 20% we use the cheapest price. Minimum contract periods are applied such that only some fraction of households may change the contract each month.²¹

The resulting time series of energy prices are matched to our sample by region. In terms of the income distribution, the underlying assumption is that contracts do not differ systematically by income. However, due to the fixed cost in the contracts, households with lower consumption pay a higher price per kWh on average. This is in line with the statistics published by E-Control.

We can compare our estimated volume of the subsidy to estimates by the Austrian Ministry of Finance and to actual payments in 2023. In its Draft Budgetary Plan 2024 from October 2023, the Ministry of Finance specified a total volume of €2.19 billion for the original period from 2022 December 2022 to July 2024. Our corresponding estimate is €2.22 billion for this period out of €2.66 billion in total for the extended period until the end of 2024.

In our analysis, we allocate the relief on an accrual basis to the year of electricity consumption. Under the assumption of an equally distributed time lag of 1 to 13 months between the consumption of electricity and the payment of the subsidy to electricity providers (annual electricity bill), we can obtain an estimate of these payments. According to our model, €0.92 billion are paid to electricity providers by the end of 2023, which is close to the actual number (€0.90 billion).

In Section 5, it was shown that a substantial part of nominal income increases in 2023 and 2024 is due to factors that are not directly explained by fiscal measures (“other income changes” in Figure 5).²² These factors include wage and pension increases as well as nominal changes in transfers. In addition, they also reflect changes in the population, mainly due to lower unemployment and higher employment.²³

In this appendix, we show components of households’ disposable income across deciles in 2021 and how they contributed to the nominal changes in years 2022, 2023, and 2024 compared to 2021. We distinguish between the following four components.²⁴

• Earned income from employment and self-employment

• Pension income

• Unemployment income

• Other income, including further transfers (e.g. family benefits, social assistance), investment income, and alimony payments

The first part of Table A3 shows the importance of these components across deciles in 2021. In the total population, earned income amounted to 66% of equivalised disposable income. Its share varied from 32% in the bottom decile to 76% in the top decile. The share of pension income was more equal across deciles, varying from 14% in the top decile to 29% in the second and third deciles. Unemployment income was only 2% of total disposable income, but its share was 14% of income in the bottom decile. The remaining income was also more important in the lower parts of the distribution due to means-tested transfers and because the same absolute amount (e.g. child benefits) is more important in the lower deciles. In the upper deciles, investment income becomes a substantial share of this component.

The bottom parts of Table A3 show changes in equivalised income per capita without fiscal measures compared to 2021 and how the four components contributed to them. In all cases, a 2% trend growth is subtracted as in the calculations in Section 5. Therefore, the total nominal change in each year and decile corresponds to the shown effect of other income changes in the right panels Figure 5.

Lower unemployment and higher employment have a noticeable effect especially in the bottom deciles. Therefore, earned income per capita has one of the highest contribution to income changes in the first decile in 2022, even though its share of total income is the smallest in the first decile.

Pension income has a negative contribution in some upper deciles in 2022 and 2023, which means that the net equivalised pension per capita grew by less than 2% per year. To some extent this is due to a smaller adjustment of gross pensions in 2022 (at most +1.8% for pensions above €1,300 per month). In addition, also composition effects matter, because households with pension income tend to drop in the income distribution compared to households with earned income.²⁵

To provide a better statistical match of EU-SILC 2020 and the Household Budget Survey 2020, we obtained additional data for households in both datasets from Statistics Austria. They include the commuter allowance and the quality of public transport at their main residence. This is especially relevant for measures targeting commuters and the regional component of the climate bonus.

We match households using a random hot-deck method as in Lamarche et al. (2020) based on the following characteristics:

• number of adult women, adult men and children,

• age, education and employment status of the reference person,

• usage of heating oil, natural gas and cars,

• deciles of equivalised disposable income.

We recalibrate the weights of households and uprate income and transfers in the EU-SILC data to account for changes in the population and the macroeconomic environment since 2019. The underlying assumption is that for example households with employed persons in 2019 are representative for the additionally employed persons from 2022 onwards. As employment increases and unemployment falls, the weights of households with employed persons are increased while those of households with unemployed are decreased.

For each year, we recalibrate the following variables:

• number of households with 1, 2, 3, and 4 or more persons,

• number of women and men,

• number of persons in five age categories,

• number of women and men receiving pension payments and number of households receiving a minimum pension,

• employment and unemployment of women and men,

• number of households using natural gas, heating oil and cars and amount of consumption of natural gas, heating oil and fuels,

• number of commuters within seven distance categories and joint distribution of commuters in four income categories and two commuting categories.

Our calibration targets for the adjustment of weights are mostly taken from the demographic forecast from Statistics Austria from 2023 and the economic forecast by WIFO from December 2023. The forecast for the number of women and men receiving pension payments is based on WIFO’s long term forecast from 2022. The number of commuters is assumed to remain constant. As a result of the higher CO₂-price from 2022, the number of households using heating oil is assumed to decrease by 2% per year and the number of households using natural gas and cars with fossil fuels by 1% per year. To implement the calibration, we use the logit method of the calib-function in the sampling-package in R, which implements the methods in Deville and Särndal (1992).

We are thankful for comments from participants of the 2^nd Banca d’Italia Workshop on Microsimulation Modelling and the 9^th World Congress of the International Microsimulation Association. The comments from three anonymous referees helped to improve the paper substantially. The views expressed in this paper are the sole responsibility of the authors and do not necessarily reflect the views of the Austrian Parliamentary Budget Office.

1. Version of Record published: December 11, 2025 (version 1)

© 2025, Eckerstorfer et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.