EUROPEAN COMMISSION

Brussels, 28.9.2022

COM(2022) 494 final

COMMUNICATION FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT, THE COUNCIL, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF THE REGIONS

Better assessing the distributional impact of Member States’ policies

{SWD(2022) 323 final}

Commission Communication

Better assessing the distributional impact of Member States’ policies

Evidence-based policymaking is crucial for ensuring successful policies and the credibility of policy actions. In particular, impact assessments allow to make informed policy choices and distributional impact assessments help understand how they are likely to affect different parts of the population. This is especially relevant as the European Union and its Member States strive for socially fair digital and green transitions and face challenges related to the rising cost of living due to the Russian war of aggression in Ukraine.

Europe’s unique social market economy is the bedrock of its prosperity. In her political guidelines, President von der Leyen stressed the importance of leaving no one behind in the digital and green transitions. She also highlighted the contribution of the European Pillar of Social Rights proclaimed by the European Parliament, the Council and the Commission in 2017, setting out 20 principles for a strong social Europe that is fair and inclusive and ensures equal opportunities.

At the Social Summit in Porto in May 2021, EU leaders welcomed the target of lifting at least 15 million people out of poverty and social exclusion by 2030, as presented by the Commission in the European Pillar of Social Rights Action Plan, in line with the United Nations 2030 Agenda for Sustainable Development 1 . Just one year on from the Social Summit, the EU Member States have committed themselves to reaching national targets to collectively achieve that goal. In parallel, within the framework of the Action Plan, new initiatives at EU level aim to contribute to reducing income inequalities, including the directive on minimum wage and the proposal for a Council recommendation on minimum income schemes. 

Poverty and income inequality are relatively low in the EU by international standards, and poverty and social exclusion have declined over the last decade, although income inequality did increase for people with lower incomes following the recession in 2008-2012. Action taken by the EU and national governments (including temporary job support schemes and automatic stabilisers) cushioned the socioeconomic blow of the COVID-19 crisis and prevented inequality from further increasing. Still, the pandemic has had a disproportionately negative impact on women and some groups like young people, low-skilled workers and persons with disabilities.

While Europe is recovering from the COVID-19 pandemic, social cohesion could be affected by current developments. Russia’s unjustified and unprovoked invasion of Ukraine has brought high economic uncertainty, especially in the supply of energy and food prices. High inflation risks exacerbating the situation of low-income households and other disadvantaged groups, while supply chain uncertainty could have an impact on employment in certain sectors.

The EU and its Member States are working to cushion the economic impact of the war. To wean Europe off Russian fossil fuels, the Commission’s REPowerEU Plan has put forward several options: saving energy, diversifying energy supplies, and accelerating the roll-out of renewable energy to replace fossil fuels in homes, industry and power generation. The Commission’s “Save gas for a safe winter” plan aims at accelerating the phasing out of Russian fossil fuels and decreasing EU gas demand while paying specific attention to vulnerable consumers. In addition, the Commission proposals for an emergency intervention in Europe’s energy markets aim at tackling recent dramatic price rises and ease the pressure they are putting on households and businesses across the EU. To counteract rapidly increasing food and energy prices, Member States have adopted measures aimed at mitigating the impact of inflation on households, in particular the most vulnerable ones.

Climate change and digitalisation are also expected to have an impact on income distribution and inequalities. The green transition has the potential to create up to 1 million additional jobs by 2030 2 and 2 million by 2050 3  as well as to improve the job quality, but its repercussions on the labour market may vary across sectors, regions, degree or urbanisation (urban/rural), skills needed or types of jobs. The European Green Deal and the Fit for 55 package present the EU’s mid-term response to addressing climate change and integrate the social dimension into its policymaking from the outset. To reach the EU’s energy and climate targets for 2030, Member States have drawn up integrated national energy and climate plans to be updated regularly. They are also committed to adopting a comprehensive policy for a fair transition towards climate neutrality. The Council Recommendation on ensuring a fair transition towards climate neutrality 4 sets concrete guidance to Member States on how to address the relevant employment and social aspects linked to the transition, including by analysing the distributional impact of the policy measures, while making full use of funding opportunities. The Commission has also presented its vision for Europe’s digital transformation by 2030 and proposed a Digital Compass for the EU’s Digital Decade 5 . Basic digital skills for everyone and the opportunity for the workforce to acquire specialised skills in information and communications technology feature prominently, as does the importance of reskilling and upskilling in order to prevent exacerbating the digital divide between low- and high-skilled workers.

Recent surveys show that social inequalities are Europeans’ biggest concern (followed by employment, environmental issues and climate change) 6  and the biggest increase in terms of concerns at EU level relates to rising prices/inflation and the cost of living 7 . Even before the pandemic, over 80% of EU citizens wanted their national governments to take further measures to reduce income inequalities 8 . High income inequality can have damaging effects on economic growth and endanger social cohesion 9 , putting the social market economy model at risk, thereby undermining European values. Income inequalities and lack of social fairness are a growing concern not only for people at risk of poverty or social exclusion, but also for most middle-income people.

To make sure that the long-term trends and short-term shocks described above do not exacerbate existing inequalities, it is imperative to improve the quality of policymaking by better assessing the distributional impact of existing and new policies and reforms. When designing them, it is vital to understand their impact on different socioeconomic groups and different geographical areas. This is necessary to better target policies and mitigate their possible adverse effects, in order to reach the 2030 poverty reduction goals and ensure sustainable, inclusive and fair transitions.

This Communication provides guidance for Member States on how best to conduct distributional impact assessments on the income of different socio-economic groups 10 and incorporate them into their policymaking processes. It sets out a process for further developing existing methodologies together with the Member States, and presents the support the Commission can make available to the Member States 11 .

The concept of distributional impact assessments

Impact assessment is a process that gathers evidence to support policymaking 12 . Distributional impact assessments (DIAs) entail an analysis, usually quantitative, to assess the distributional effects of policies (reforms, investments, etc.) on the income of various groups across the population. DIAs provide useful information and help policymakers to choose between different reform options. DIAs can also help identify the need to design flanking measures to protect vulnerable groups from the potential negative impacts of certain policies and provide valuable information to calibrate those measures.

DIA techniques can be used to obtain quantitative estimates of the impact of taxes and social protection and social inclusion reforms, and also provide indications of the impact of reforming publicly provided in-kind services. The importance of equal access to in-kind services such as healthcare and education was highlighted during the COVID-19 pandemic 13 . Monitoring the redistributive effects of in-kind benefits would contribute to tapping into their potential to mitigate and reduce poverty and can also contribute to making DIAs more comprehensive.

DIAs can quantify, in advance of their implementation, how specific policies and reforms will affect the income of various groups. They can also help to establish in detail the cost of reforms (including by considering different options) and design corrective measures. Assessments done after measures have started to be implemented contribute to the evaluation of measures once data are available, indicating whether or not further measures need to be taken or reforms corrected. DIAs are often done as part of budgetary preparations, but it is equally important to do them for the design of reforms that might not have a (direct) impact on public budgets, but are liable to have a greater impact on certain socioeconomic groups. DIAs should, where appropriate, be integrated in wider analyses covering economic, social, and environmental impacts of new policy initiatives.

DIAs can also help significantly improve the quality of public spending and fiscal policies. Public finances have been put under strain by the COVID-19 pandemic and the gradual normalisation of monetary policy is expected to increase government borrowing costs. Well-calibrated reallocations of expenditures or shifts in revenues can help to ensure that public budget deficits are kept under control, but this can only be sustainable if the distributional consequences are well identified and weighed up. The presentation of the expected distributional impact of measures in draft budgetary plans submitted to the Commission by euro area Member States is already recommended in Regulation (EU) No 473/2013, but rarely provided and the practice could be further improved. In the context of the European Semester, the 2022 Employment Guidelines recommend the use of DIAs to improve the effectiveness of social protection systems 14 .

Key components of a good quality DIA

Key components of quality DIAs can be identified on the basis of good practices in the Member States. These concern their timing, what policies to analyse, the tools (models and data) that should be used and how to disseminate the results.

a.Who and when should do a DIA and at what level of granularity?

DIA should be conducted both before and after the implementation of relevant policies. Doing a DIA during the design of reforms and investments is particularly important as it allows policymakers to gauge the impact of the foreseen policies on different types of households, even in cases where the reform is not initially intended to address social objectives (for instance in the case of reforms on energy consumption). Post-implementation analyses allow to evaluate the impact of reforms and investments on different types of households a certain period of time after the reform has been implemented. This can help develop potential further steps or corrections in reforms. They are based on hard data, but require the disentanglement of the effects of reforms from other changes. Credible post-implementation analyses help to ensure the ownership and quality of assessments before implementation. Ideally DIAs should be done by authorities at national, and where relevant at regional and local levels.

It is good practice to present the effect of a full package of policy changes as some interactions among them may be missed if their effect is presented separately. This is all the more so, given budgetary changes are often decided as a package, with some measures compensating for others. Measures that have a considerable impact also warrant individual analysis.

Entrusting independent bodies with doing DIAs can help make them more credible and accurate. Thanks to their expertise, academia or research institutes are well placed to develop advanced DIA techniques. Moreover, they are not affected by the same political motives as administrations when assessing decisions to which they have often not contributed. The replication of results by various institutions can also make for a more comprehensive analysis, and enhance its quality and credibility.

b.What policy areas to choose for doing DIAs and what length of time to consider?

DIAs should cover policy areas that have an impact on household income and their distribution, which relates typically to taxes and monetary benefits. Standard models can give an idea of the impact of reforms to traditional direct taxation and monetary benefits that are generally the focus of DIAs 15 . However, efforts should be made to assess the impact of reforms in other areas, such as labour market or pension reforms, as well as investment programmes.

It is worthwhile assessing consumption taxes given their impact is often regressive (as lower-income households spend more of their income on goods and services). This is getting increasingly important with the growing need to reduce carbon emissions and practise ‘green budgeting’. Ideally, Member States should also do DIAs on wealth tax reforms, which have a major impact on the long-term dynamics of inequality, and on the effect of changes to in-kind benefits, such as healthcare benefits.

DIAs can be used in a broad range of policy areas. They are crucial for assessing the impact of reforms designed to address megatrends, such as the green and digital transitions or the impact of recent crises. However, various reforms differ in the extent to which they lend themselves to being analysed. For example, DIAs are very useful for assessing the effects of reforms that have a direct impact on households’ incomes, such as carbon pricing and the use of carbon revenues or heating allowances for low-income households. However, it is often less straightforward to gauge the impact of new regulatory measures (e.g. to improve energy efficiency standards). Post-implementation assessment strategies can nonetheless provide good insights into the distributional consequences of such reforms.

DIAs before implementation should not only focus on the impact of new policies and measures on outcomes immediately after implementation, i.e. in the following year (comparative statistics), but should also adopt a multiannual perspective. An incremental policy – pension reform, for example – that induces changes slowly over a number of years may not have a very large impact in any 1 year, but cumulatively it might have considerable impacts. In these cases, DIAs can require long-term projections to take account of the full impact of reforms. Adopting a multiannual perspective might also be useful when taking demographic trends into account is particularly important. In this case the use of dynamic models might be relevant.

c.What tools to use?

Microsimulation models are essential as they typically allow users to estimate the net budgetary cost of changes to taxes and benefits, the pattern of gains and losses from a policy change and the impact of reform packages on poverty and inequality. Advanced models make it possible to estimate the impact by gender, or additional equality criteria such as disability or minority background, as well as to assess the change in work incentives and the labour market responses to policy changes.

Reforms of tax and benefit systems can impact people’s behaviours and change the macroeconomic environment. Incorporating behavioural responses, macroeconomic data 16 and demographic trends into DIAs can further increase their accuracy.

DIAs can be done using a country-specific model (currently the case of 10 Member States), or the EUROMOD model only (currently the case of 4), or both (currently the case of 12) 17 . EUROMOD is a tax-benefit microsimulation model maintained by the Commission and available for all the Member States to use. It covers the main aspects of direct taxation, social contributions and benefits 18 . It can be particularly useful for Member States with limited experience of assessing distributional impacts 19 . One distinctive advantage of EUROMOD is that it allows the calculation in a comparable manner across Member States, of the effects of direct taxes and benefits on household incomes, poverty and income inequality and work incentives. Country-specific models, involving more tailor-made approaches, tend to be used by Member States with a long-standing tradition of doing DIAs.

Quantitative analysis can be complemented with more qualitative considerations to identify the groups most likely to be affected by a set of policies.

d.What data to use?

A good DIA requires comprehensive and timely data. Survey data – such as European Union Statistics on Income and Living Conditions (EU-SILC) 20 – provide detailed sociodemographic and income information, crucial for the quality of a DIA. However, there is usually a time lag between when a survey is conducted and when the results are available 21 . Administrative data can fill in the gaps, with more accurate and, in most instances, relatively more timely information, a larger sample size and greater geographical detail. Combining a survey with administrative records on income, welfare and taxation can also significantly improve the quality of a DIA. This can also enable the broadening of the scope of modelling, while retaining the detailed socioeconomic information from surveys.

More must be done to make national administrative data more easily available 22 .  Anonymised (unsampled) administrative data are a valuable resource for precise DIA analysis 23 . Such data should be accessible to various public bodies and independent researchers in a transparent way. It is also important to stress the crucial role of national statistics institutes in providing timely, comprehensive and detailed surveys and other data that can be used for DIA analysis.

e.What indicators to choose?

Using common metrics enables policymakers to better compare the results, and assess the impacts of, various new measures. DIA outputs should ideally gauge the impact of reforms on: 1) the distribution of income 24 ; 2) the level of inequality, by relying on a selection of key indicators (such as the income quintile share ratio S80/S20) 25 ; 3) the level of poverty, by using key related indicators such as the at-risk-of-poverty rate and the relative median at-risk-of-poverty gap 26 .

There may also be other relevant output, depending on the national situation and the available data. This is why DIAs are often done to assess the impact of reforms on various groups 27  differentiated by age and gender (also for gender budgeting 28 ) and type of household (such as working age households, retirement age households, single parent households, single adults and couples, with and without children). Other analyses, such as those focusing on people with migrant backgrounds, with disabilities or any other disadvantaged group, as well as groups differentiated by degree of urbanisation or region, could be worth doing.

f.How to disseminate DIAs?

Publishing the analysis of the impacts on different income groups increases transparency of policy making. It allows for more scrutiny of the impacts of proposed measures and thereby can improve the quality of the public debate and enhance trust in decision-making processes. Higher transparency on the impact of reforms can be expected to lead to the adoption of measures that will reduce or eliminate the negative impact on poverty or the widening of inequalities. It can also help debunk false perceptions about the ultimate effect of policies by giving clarity on beneficiary groups. For instance, draft budgetary plans are a good way of presenting the impact of related policy measures on different parts of the population.

When done by administrative bodies, making DIAs public in an accessible way nurtures public debate and might help gain acceptance of measures and reforms. DIAs should be presented in publicly accessible budgetary documents (e.g. draft budgetary plans) and any other type of document accompanying the design, implementation and assessment of reforms and investments, such as (integrated) impact assessments. This way, apart from enabling more evidence-based decisions, DIAs can give a more credible picture to inform public debate on the effects of policy reforms, including social partners and civil society.

To garner the interest of the media and the public, DIA results need to be considered reliable and reasonably easy to understand. The publication of DIAs should therefore be accompanied by a presentation of the key modelling decisions underlying the analysis. A presentation of DIA results planned at given moments in time (e.g. every year) can be helpful as it signals that results are communicated systematically. DIA results may also be considered more reliable and credible if the analysis has been done, or can be verified by, independent researchers. By facilitating the replication of the analysis, the publication of all relevant modelling assumptions and decisions and data can also enhance credibility.

How the Commission relies on DIA and supports Member States

The Commission’s track record in evidence-based policymaking has continuously improved in recent years. It will continue to systematically assess the economic, social and environmental impacts of its policy actions to ensure the high quality of proposed legislation, as envisaged in its ‘better regulation’ system 29 / 30 . 

The Commission will continue doing DIAs in order to give more prominence to distributional considerations relevant for the design of reforms and investments. The Commission regularly uses EUROMOD to do DIAs of selected reforms in the Member States. The ensuing analysis is also presented in the country reports that constitute policy assessments as part of the European Semester, the framework for integrated surveillance and coordination of economic and employment policies across the European Union.

The Commission will further support Member States in setting up or developing their DIA practices. This support will be provided by maintaining and further developing the EUROMOD model. The Commission makes the model available to all Member States and updates it annually. The JRC provides a range of training courses on EUROMOD, including on how to do DIAs. Member States can also request further assistance in developing DIA practices under the Technical Support Instrument 31 .

The Commission will further support mutual learning by bringing together representatives of Member States who are actively involved in doing DIAs in national administrations or, in the case of Member States where DIAs are not yet so widespread a practice, those who could be involved in doing them in the future. Discussions in mutual learning events cover various aspects of doing DIAs, thereby enabling the exchange of best practices. Since Member States’ DIA practices differ significantly from each other, there clearly is added value in learning from each other’s experience, drawing on best practices that can be adapted to country-specific contexts.

As part of the mutual learning process with the Member States, the Commission will monitor the use of DIAs, in draft budgetary plans or in other contexts, and inform the Social Protection Committee and the Employment Committee, the advisory bodies for the Employment and Social Affairs Council, and the Economic Policy Committee, the advisory body of the Economic and Financial Affairs Council.

Conclusion

The confluence of various long-term trends and short-term shocks described in this Communication underscores the importance of paying greater attention to distributional considerations while designing reforms and investments. Therefore, Member States are invited to follow the guidance provided to expand the use of and improve their DIA practices. The resulting analyses should give a credible impact of policies on different socio-economic groups and be made available to broader public to enhance the quality of the policy debate. The Commission stands ready to support Member States with assistance on developing DIA methodologies, monitoring the existing practices and providing a forum for exchange of views on how to improve DIAs further.

(1)

 Sustainable Development Goals 1 and 10. 2030 Agenda for Sustainable Development.

(2)

According to the Communication from the Commission ‘2030 Climate Target Plan’, COM/2020/562 final, and SWD(2020)176 final: projections based on E-QUEST using a ‘lower taxation low-skilled labour’ scenario, i.e. assuming a ‘targeted reduction in labour taxation which stimulates low-skilled labour supply via higher net wages while simultaneously lowering low-skilled labour costs for firms, thereby leading to higher overall employment’.

(3)

Based on the 2018 Clean Planet for all impact assessment, the ESDE 2019 reports a potential gain of about 2 million jobs by 2050.

(4)

https://eur-lex.europa.eu/legal-content/EN/ALL/?uri=COM:2021:801:FIN

(5)

The Digital Decade is the European Commission forward-looking strategic vision for the development of the digital economy and the transformation of European businesses by 2030. The plan, presented by the European Commission on 9 March 2021, aims to support a prosperous digital future for all.

(6)

  Future of Europe EB from January 2022.

(7)

April 2022 standard Eurobarometer .

(8)

Special Eurobarometer 471  Fairness, inequality and inter-generational mobility - April 2018 - - Eurobarometer survey (europa.eu)

(9)

See for instance Organisation for Economic Cooperation and Development (OECD) (2015), Divided We Stand: Why Inequality Keeps Rising.

(10)

Further efforts should be made in the future to also capture distributional impact of reforms on equality of opportunities, but this Communication focuses on income inequality that is much easier to quantify than the other aspects.

(11)

The accompanying Staff Working Document provides more granular information, including on current practices in the Member States.

(12)

It helps developing a policy response (either a policy reform or an investment) to a certain policy problem by providing the evidence base for - and the impacts of - various options. Impact assessments address a number of key potential impacts such as the economic, environmental and social impacts as well as the impact on human rights.

(13)

The evidence shows that in-kind benefits have a significant redistributive impact. It is important to ensure that policy changes serve the needs of the most disadvantaged groups and, at the very least, do not have unintended negative consequences for them. For instance, World Health Organization (WHO) Europe’s work on the affordability of healthcare shows that some countries will need to redesign health coverage policy, improve the financial protection of particular groups and at the same time seek additional public investment in the health system (see WHO/Europe Publications – ‘Can people afford to pay for health care? New evidence on financial protection in Europe’ (2019)).

(14)

  Guideline 8 .

(15)

Direct personal taxes, as well as social security contributions and social benefits.

(16)

 Microsimulation models often do not account for behavioural responses (such as the estimated effects on labour supply of changes in the income tax) or macroeconomic feedback (for instance increased labour supply may increase GDP and government revenues). Incorporating behavioural responses and macroeconomic feedback can be done by building more advanced models taking into account the various effects (such as incorporating behavioural replies into a microsimulation model or linking it to a macroeconomic model).

(17)

See the accompanying Staff Working Document for more details.

(18)

It is kept up to date by the Joint Research Centre (JRC) in cooperation with Eurostat and national experts.

(19)

The software and models are readily available, but national authorities or researchers still need to apply to Eurostat for access to the underlying data.

(20)

  Overview - Income and living conditions - Eurostat (europa.eu)  

(21)

In in the case of EU-SILC, in the framework of the Regulation 2019/1700, Eurostat receives microdata from Member States for operation year N by the end of year N, and complete datasets with final income data is received by 28 February of year N+1. Country indicators are published as soon as the data is validated and approved.

(22)

In line with A European strategy for data , stressing the importance of data sharing between public authorities, as this can significantly improve policymaking and the delivery of public services.

(23)

The reason being that uniform samples are not adapted to measure inequality when concentration is high.

(24)

Equivalised disposable income, meaning income after direct taxes and benefits and adjusted for household composition. Household members are equalised or made equivalent by weighting each according to their age, using the so-called modified OECD equivalence scale. (https://www.oecd.org/economy/growth/OECD-Note-EquivalenceScales.pdf)

(25)

European Pillar of Social Rights Social Scoreboard headline indicator; S80/S20 is the share of income going to the richest 20 per cent compared to the poorest 20 per cent of the population. It is also useful to present its decomposition between the upper part S80/S50 and lower part S50/S20 of the income distribution. It can be complemented by the income share of the bottom 40% of the population along the income distribution (S40 which constitutes the basis for the SDG goal on inequality reduction) or the Gini index (which is the most widely used synthetic measure of income inequality, capturing what percentage of the domestic income of a country each cumulative percentile of the population owns and converting it into an index).

(26)

The at-risk-of-poverty rate (AROP), calculated as the proportion of people with an annual equivalised disposable income below 60 per cent of the median annual equivalised disposable income, is one of three components of the at-risk-of-poverty-and-social-exclusion (AROPE) indicator that underpins the EU target to reduce poverty and social exclusion by 2030. The relative median at-risk-of-poverty gap shows the difference between the median equivalised disposable income of people below the at-risk-of-poverty threshold and the at-risk-of-poverty threshold (cut-off point: 60% of median equivalised income).

(27)

DIAs are useful to present both the differences in the initial situation of various groups and the changes expected from policy interventions (relevant for both and environment and social dimensions where socially disadvantaged groups tend to experience poorer environments).

(28)

According to the Council of Europe’s widely used definition, gender budgeting is an application of gender mainstreaming in the budgetary process. In short, it is a strategy and a process with the long-term aim of achieving gender equality goals by focusing on how public resources are collected and spent. See for more details Bova, E., Jerosch Herold da Costa Reis, J. (2022) Gender Budgeting Practices: Concepts & Evidence. European Economy Discussion paper 165, European Commission. Gender Budgeting Practices: Concepts and Evidence

(29)

  Better Regulation – joining forces to make better laws (COM/2021/219 final).

(30)

  Better Regulation Toolbox (November 2021), in particular Tool#30. Employment, working conditions, income distribution, social protection and inclusion .

(31)

In this context, the JRC has been collaborating since 2017 on projects to deliver technical support to Member States on microsimulation and the use of administrative data for assessing tax-benefit reforms. This makes it possible to develop a model for DIAs by making greater use of EUROMOD, drawing on survey and administrative data. Such projects have been taking place with Greece, Slovakia, Lithuania and Romania. Capacity-building and ad-hoc projects are also being carried out with Spain.