5. The case of Italy
This case study provides an overview of recent trends in income inequality in Italy, and discusses how considerations for inequality and distributional implications of public expenditure are brought to bear as part of the budget process. It discusses the practices currently in place in the country, how they are set up in the country’s public expenditure frameworks, and how they are supported at the technical level, through the range of models, and data tools that are utilised in policy practice.
5.1.1. Trends in income inequality
Italy’s tax and transfer system reduces income inequality at a rate slightly below the OECD average. In 2018, before taxes and transfers, Italy had a Gini coefficient of 0.433, as shown in Figure 5.1. However, taxes and transfers reduced this coefficient to 0.333. While this coefficient means Italy’s disposable income Gini coefficient is still higher than the OECD average of 0.312, it nevertheless represents a significant reduction in inequality.
Notes: Countries are ranked from the highest to the lowest difference before and after taxes. Before taxes and transfers data for Mexico are post taxes but before transfers. The latest data refer to 2019 for all countries except Costa Rica and the United States (2021); Australia, Canada, Latvia, Korea, Mexico, the Netherlands, New Zealand, Norway, Sweden and the United Kingdom (2020); Ireland, Italy, Japan and Poland (2018); Chile, Iceland and South Africa (2017). No data available before 2018 for Belgium and Japan or before 2015 for Luxembourg and South Africa. Earlier data for Brazil, Chile, Estonia, Sweden and the United States are from 2013.
Source: OECD Income Distribution Database
The impact of taxes and transfers on reducing income inequality has increased significantly in the past several years. As Figure 5.2 demonstrates, in 2006 the tax and transfer system reduced the Gini coefficient by just 0.172, a figure which had increased 11% at its peak in 2015. While this value has since decreased somewhat, the 2018 impact is still 5.5% higher than the 2006 base.
Source: OECD.Stat
These redistributive impacts should be considered against the fact that Italy has among the highest net social expenditures as a percentage of GDP, as evidenced in Figure 5.3, although still well behind France, Denmark, Finland and Belgium. However, Austria, Czechia, Germany, Greece, Hungary, Ireland, Portugal, and Slovenia all spend less on social benefits than Italy, but achieve higher levels of redistribution in terms of the Gini coefficient.
In recent years, Italy has introduced several significant redistributive measures, mainly through transfers, in order to combat poverty. Among these are the “Citizenship Income” (Reddito di cittadinanza), a guaranteed minimum income introduced in 2019, and the “Single and Universal Allowance” (Assegno unico e universale), a family allowance introduced in 2022. These measures, together with other temporary interventions to limit the economic effects of COVID-19 and of the war in Ukraine, contributed to reducing inequality.1
The distributional impacts of these measures can be seen in reports from Italy’s Official Statistics Office (ISTAT, 2023[1]). The lowest income quintile is the most significant recipient of Citizenship Income, receiving nearly half their average family income’s worth (Figure 5.4). This trend is less pronounced for the emergency COVID measures – while the lowest income quintile did benefit the most from these measures as a percentage of income, the top three quintiles benefitted the most in raw terms, likely due to the fact that many COVID-related policies were intended to benefit the population as a whole, and thus were often not income-specific.
Source: (ISTAT, 2023[1])
5.1.2. The Pension System
As Italy’s high social expenditure is predominantly due to pensions – pension spending in Italy is over double the OECD average (see Figure 5.4) – it is worth taking a special look at its pension system. Pension expenditure of course goes predominantly towards older people, and as wealthier people tend to live longer and thus receive higher pay-outs, such expenditure can actually increase inequality (Sanchez-Romero and Prskawetz, 2020[2]). There is also a risk that high pensions reduce public resource availability for other age groups, increasing income precarity and poverty risks for the working classes (Fornero, 2021[3]).
As it currently stands, Italy has seen a relative stagnation in income inequality measured by weekly income since the 1990s, while it has seen a consistent increase in income inequality measured by annual income. This pattern can be predominantly explained by an increase in the regularity of part-time work among the population (INPS, 2019[4]). As the pension system is heavily geared towards employer contributions,2 this change in working patterns serves as further evidence that pensions can actually work to increase net distributions of income, by favouring those who already have relatively high incomes. Indeed, this pattern has been evident in recent years – in 2022, 47.8% of pensions went to the more affluent north of Italy, while only 31% went to the south (Il Sole 24 Ore, 2022[5]).
5.1.3. Income inequality at regional level
Italy has significant variation in income across its regions, which is not majorly altered after considering pensions and transfers – indeed, aggregate income including transfers is often higher among wealthier regions, likely due to the impact of pensions identified above (Figure 5.5). It is also evident that aggregate income including transfers is often higher among wealthier regions, likely due to the impact of pensions identified above.
Income variation across Italy is consistent with variation in public transfers. As shown in Table 5.1, receipt of Citizenship Income is highest in the South and Islands, where average income is lowest, and risk of poverty is highest. In Sicily and Campania about 38% of the population has a net equivalised income below or equal to 60% of the median equivalised income. However, this trend is not evident when it comes to salary supplement receipts, where the Northeast sees the greatest proportion of people claiming some sort of salary supplement, despite having among the highest incomes in Italy. Again, this may well be due to the pensions effect previously discussed.
There is also significant variation in in-region inequality – ranging from Marche, where the wealthiest 20% earn 3.7 times as much income as the least wealthy 20% after transfers, to Campania, where they earn 7.5 times as much (see Figure 5.7). This variation does not necessarily correspond to regional variation in income (as seen in Figure 5.5) – Campania, Calabria and Sicily have the highest levels of inequality as measured by the ratio of the highest quintile of income over the lowest quintile, despite being in the bottom half of the income distribution after transfers and pensions. In a similar manner, Tuscany, Piemonte and Emilia-Romagna are the highest income-earners after transfers and pensions, but below-average income inequality levels.
Source: (ISTAT, 2022[7])
5.1.4. Income inequality and the informal economy
Understanding income inequality in countries with a significant informal economy, such as in the case of Italy, presents many analytical pitfalls. There is substantial evidence that a larger informal economy leads to greater levels of inequality. A large hidden economy can negatively impact the government’s tax revenues and therefore reduce the provision of public goods (Enste, 2003[8]). Furthermore, it can disincentivise innovation and instead encourage rent-seeking behaviours, with the aim of increasing monopolistic power. This can in turn reduce worker wages and consumer surplus, thereby increasing inequality (Eilat and Zinnes, 2002[9]). On top of this, as wealthier portions of the population are more likely to operate in the informal economy for tax evasion purposes, official statistics can actually underestimate the extent of inequality in an economy, which can subsequently impact policy decisions. This issue is notable in Italy, where it is estimated that the propensity for personal income tax evasion is under 3% for employees while it is nearly 70% for business owners (Italian Ministry of Economy and Finance, 2022[10]). As the latter group make up the majority of high earners in Italy, there is a real risk of statistical underreporting of inequality (Valentini, 2022[11]).
While the Italian informal economy has been steadily decreasing in size since 2014, it still makes up a significant component of the economy – in 2019, the Italian institute of Statistics (ISTAT) estimated it to be worth 203 billion euros, 11.3% of total GDP. Of this, 183 billion is estimated to be the hidden economy (i.e., legal activity concealed through misreporting of turnover and costs, or use of irregular labour), with the remainder being illegal activity (ISTAT, 2021[12]). A 2018 econometric study underlines the negative impact this can have on inequality, finding notable and highly significant correlation between the extent of irregular labour in a region and the inequality of income distribution (Clementi and Valentini, 2018[13]).
5.2.1. The Ministry of Economy and Finance
The Ministry of Economy and Finance (MEF) is divided between the Department of Finance, which focuses on taxation and government revenue, and the Department of the Treasury and the State General Accounting Department (RGS), which concerns itself with government expenditure and budgeting. While the Department of the Treasury traditionally analyses the macroeconomic impacts of policies and reforms, the Department of Finance uses microsimulation models to estimate the revenue effects of fiscal policies – although it can also provide distributional impact assessments on expenditure measures that have fiscal implications. Therefore, this section will discuss first the work in the Department of Finance that has more developed models for studying inequality, before moving on to the role of the Department of Accounting (RGS), in charge of the budget and the Department of the Treasury.
SOGEI (Società Generale d’Informatica), is an information technology company fully owned by the MEF. It is charged with the modernization and digitalization of the Italian public system, as well as with supporting the Government in its policy decisions, through the development of a number of tools.
SOGEI supports MEF in the development and management of a variety of models intended for short-, medium- and long-term forecasts of the Italian economy as well as for policy impact evaluation. Such support is predominantly intended for the preparation of institutional documents, such as the National Reform Plan (PNR), the Stability Program and the Draft Budgetary Plan. SOGEI maintains and operates models fit for macroeconomic and budgetary forecasts and projections, as well as those that look at distributional analysis (see Box 5.3).
5.2.2. The Department of Finance –Directorate for Economic and Fiscal Studies and Research (DSREF)
The Department of Finance’s main focus is on tax matters, which has traditionally involved using data from tax returns to provide estimates of the impacts of fiscal policies on tax revenue. This continues to be a key task of the Department, with their estimated impacts appearing in the Technical Notes to legislative documents. However, in the last ten years, there has been a significant increase in the attention given towards distributional impacts by policymakers. In order to support this, the Finance Department’s Directorate for Economic and Fiscal Studies and Research (DSREF) has developed various microsimulation models based on the different type of taxes3 in order to analyse the ex ante distributional effects of proposed or adopted fiscal policies and reforms.4 These models use representative samples of the Italian population, based on a large set of various administrative and survey data. While the models simulate all the relevant tax-benefits policies that it is possible to estimate with the available data, the Directorate is called upon by the government to provide support mainly on policies on the fiscal issues, implying that, in most cases, it only looks at the distributional implications of the relevant benefits and transfers to households and firms having fiscal implications, as is the case of the 2022 family allowance (see Box 5.2) (MEF, 2022[14]).
Source: (MEF, 2023[15])
In most cases, the DSREF conducts distributional analysis in order to assist government officials in choosing the best policy options. As such, the DSREF often runs many simulations, in order to provide the Minister of Economy and Finance and his advisers with a range of policy projects to choose from, and these simulations are internally discussed within the cabinet, and are not made publicly available. The definitive and publicly available simulations are only those on the definitive policies approved and published in the Technical Notes to legislative documents for the tax revenue impacts. Aside from that, Distributional Impact Analysis is also officially published on the Economy Sustainability and Wellbeing report.
In addition, from early 2020 onwards, the DSREF also started to produce policy notes5 in order to inform the public on the Covid-19 effects and key reforms, such as the last PIT and family allowance reforms in 2022. When conducting more extensive research on specific topics (e.g. the effects of COVID-19) or when developing new economic models used for the analysis, the DSREF also produces working papers in the Department Working Paper Series6 (written in English), as well as occasional scientific publications in specialized economic journals. For example, in the recent case of the pandemic, analysis on the effects of and policy response to COVID-19 was initially published in the form of policy notes, before being further developed in subsequent working papers and published in a scientific journal (Monteduro, De Rosa and Subrizi, 2023[16]).
Ex ante analyses are essential, considering that there is usually a lag of over a year between when income statements are received and when tax return and survey data are. However, the DSREF also carries out ex post analyses once they receive the relevant data, in order to assess the impact of past reforms, and thus support their improvement or justify their continued funding. Examples include the impact assessment of business incentives (‘Industria 4.0’ plan) (Bratta, Romano and Mazzolari, 2020[17]), the analysis on the effects on tax gap of the flat tax on real estate (‘cedolare secca’) (Di Caro et al., 2022[18]) and a recent analysis on the expansion of the flat tax for self-employed workers. Sometimes ex post analyses are expressly requested for institutional purposes, as in the case of the analysis provided to the Commission on Tax Expenditures.
The DSREF publishes the official statistics on tax returns annually, as well as those on real estate wealth in Italy7 and the above-mentioned Report on the tax gap and informal economy. Some publications receive wide media attention, with major national newspapers covering the main findings as was the case of the recent Well Being and Sustainability Report (Il Sole 24 Ore, 2023[19]). This kind of media attention has also occasionally been given to the policy notes which are more informal and less technical, and therefore easier for the general public to understand. The statistics on tax returns are of great importance as they form the basis for the DSREF microsimulation models, but also are used by existing external models in Italy.
On top of the internal data at its disposal, the DSREF also makes use of external data sources provided by the ISTAT and the Italian National Social Security Institute (INPS), including information on the socio-demographic characteristics of households, non-taxable incomes, pensions and social security contributions not available in tax returns (for a full description of data sources see the section “Data and information infrastructure”). ”). Model development and impact assessments are generally carried out by internal DSREF officials, as in most cases data are not available for external institutions. However, for some specific projects, the DSREF has worked in collaboration with universities or other research centres (e.g. the Italian EUROMOD team to develop the microsimulation module on labour supply, ECOMOD and the World Bank to develop the ITAXCGE-DF model).
PIT reform and Single and Universal Allowance
In 2022, Italy implemented two important reforms into the tax-benefit system, designing a new Personal Income Tax (PIT) scheme and a new family allowance system intended to make the previous system simpler and fairer. This previous system combined tax credits based on individual income (therefore benefitting only workers with adequate fiscal capacity) and only provided an allowance for employees, based on gross household income.
The AUU increases the number of family allowance recipients by introducing a single and universal allowance for all children under 21. The AUU has a progressive design based on the Indicator of the Equivalent Economic Situation (ISEE),1 which implies a maximum amount for households with ISEE lower than 15 thousand euros (with an adjustment to a minimum amount for ISEE greater than 40 thousand euros). The modification of the PIT aims to correct some structural flaws linked to the previous design of marginal tax rates, which produced unequal and inefficient treatment of taxpayers. The new scheme reduces the number of tax brackets from 5 to 4 and the two central tax rates from 27% to 25% 38% to 35%. It also innovates the design of the tax credits by type of income (employees, self-employed, and retired people) and the in 2016,design of the so- called “bonus PIT”, an additional benefit initially introduced in 2014 but modified by the most recent PIT reform, and intended for employees with an income of less than 15 thousand euros.
Although the AUU is a transfer to households and thus a spending measure, it fell under the analytical remit of the Department of Finance (DSREF) as it is strictly related to the PIT reform replacing the previous policy of tax credits for dependent children (which now only remains in force for children over the age of 20).
Using the internally developed TAXBEN-DF microsimulation model, the DSREF analysed the impacts of both the PIT reform and the AUU introduction. It found that the two reforms impacted 85% of Italian households, and although all households benefited, the lowest decile benefitted the most, with the two interventions have an 11.6% impact on household gross income (Table 5.2). When isolating the impact of these two reforms, the DSREF noted that these redistributive effects were predominantly driven by the AUU (Table 5.2, last columns), with an impact of 11.3% on the gross income of the poorest 10% of households, with the income tax reform contributing only the remaining 0.3 percentage points. Overall, the two reforms led to a 1.65% reduction in disposable income inequality (estimated using the post-tax and transfer Gini index), entirely explained by the introduction of the AUU (Italian Department of Finance, 2022[20]).
5.2.3. The Department of Accounting and the budget process
The Department of Accounting (Ragioneria Generale Dello Stato) is in charge of the budget process, which includes an annual budget cycle starting with the preparation of the Economic and Financial Document (DEF), a document that sets out the macroeconomic projections for the following three years, as well as an update for the current year, and provides fiscal goals for the same period. This initial document must be submitted to Parliament for debate and approval no later than 10 April, before it is then submitted to the European Commission by the end of April, in line with Italy’s Euro membership criteria (Blöndal, von Trapp and Hammer, 2016[21]).
In the second phase of the cycle, line ministries will update their baseline expenditure requirements from the previous year, and attempt to justify these requirements. These line ministries will then defend their claims in front of the MEF, who will review, and adjust the DEF document accordingly. This updated DEF document goes in front of Parliament again, usually around mid-September. This updated version is usually approved by Parliament very quickly, after which budget documentation is presented in mid-October. This documentation consists of the measures necessary to achieve the objectives set out in the updates DEF, and the planning expenditure by policy area. This document goes through several readings, and receives its final approval before 31 December (Blöndal, von Trapp and Hammer, 2016[21]).
Since 1989, the Italian Parliament has had two in-house non-partisan analytical units in order to help it go through the approval process, known as the State Budget Department and the Senate Budget Service. They both produce several difference types of reports, including analysis of the budget proposals and broader economic analysis. All reports it produces are in the public domain.
In 2016, Italy introduced gender budgeting, with the intention of creating a better understanding of how budgetary allocations impacted men and women differently. RGS prepares the budget, providing a set of indicators to monitor actions taken to address gender inequalities. The initial pilot examined the impact of budgetary policies on women and men in terms of expenditure, services, time and unpaid work, with a 2018 decree strengthening the initiative by explicitly stating that it should define and allocate resources from a gender perspective, and take into account the ESW indicators (Blazey, Lelong and Giannini, 2022[22]). Analysis of income inequalities by gender have increased steadily since 2016, and are now available in the Annual Gender Report, although as it currently stands this report does not impact the budget process.
5.2.4. The Department of the Treasury
The Department Treasury also plays a key role, given its strength as the powerhouse of economic analysis in the Ministry of Economy and Finance. Its role in terms of inequality and well-being covers two key areas: analysis of distributional implications (particularly for the labour market) using its microsimulation model T-DYMM, and preparation of some general indicators of well-being, which include inequality considerations. Usually analysis from the Treasury is brought to bear for significant structural reforms and long-term projections.
“Modernising Social Protection Systems in Italy” (MOSPI) was a project undertaken by INAPP and the Treasury, in tandem with the European Commission and the Giacomo Brodolini Foundation, and with technical support from SOGEI. It aimed at adapting the social protection system to the modern challenges of digitalisation, an ageing population, and the fragilization of the labour market. As part of this project, the T-DYMM simulation model was further developed (see subsequent section), so that it could then be used for impact and distributional analysis.
The project’s final report uses T-DYMM to analyse the impact of two policy options: an extension of the standard unemployment insurance (NASPI) to cover those under nonstandard contracts paired with an extension of an allowance for certain self-employed workers (ISCRO) so that it covers all self-employed workers (Policy A); and the introduction of a guaranteed pension for notional defined contribution (NDC) workers (Policy B). Amongst other things, it finds that Policy A extends the number of unemployment benefits recipients by 3%, and disproportionately benefits those with less education, women, and non-Italians. Furthermore, it has a positive impact on income inequality amongst the unemployed (Figure 5.9). The point of the exercise was to show how T-DYMM can help in assessing the impact of possible policy proposals.
Source: (De Minicis et al., 2021[23])
Inequality and well-being analysis
In 2010, ISTAT and the National Council for Economics and Labour (CNEL) set up the first phase of the measurement of Equitable and Sustainable Well-being in Italy, and in 2013, the first Equitable and Sustainable Wellbeing (ESW) report was published.8 Since then, the report has been published every year by ISTAT and in April 2023 the 10th edition was issued.
In 2016, the Italian Government passed a law which reformed the state budget, it states that government policies must be assessed not only in terms of traditional measures such as GDP, but also in terms of multidimensional well-being. In doing so it was formally recognised that these indicators should be included in the government’s planning documents for the economic cycle. The reform further stated that the MEF must include an annex in its Economic and Financial Document (DEF) report, which must show a) the trend over the last three years of the selected indicators, and b) forecasts on their evolution over the DEF’s reference period. Furthermore, by 15 February each year, a report (“The Report on the ESW Indicators”, prepared by MEF) must be submitted to Parliament highlighting the evolution of these indicators (Italian Ministry of Economics and Finance, 2018[24]).
In order to achieve this, a committee9 was set up in order to choose the key ESW indicators to be analysed in the annex. The chosen indicators, 12 in total including net income inequality (including direct transfers but not including benefits in kind), absolute poverty, and non-participation in employment by gender,10 were formally adopted in November 2017, making Italy the first G7 and EU country to include such welfare objectives in economic planning. The reform provided legal basis to the two new reports. Italian governments must now systematically assess, both ex ante and ex post, the impact of policies on the chosen indicators (Italian Ministry of Economics and Finance, 2018[24]). Much of the data for the indicators is broken down by gender at the national level, including unemployment rate (for both adults and youth), economic inequality and risk of poverty, although most data do not break down by gender at the regional level. (Istituto Nazionale di Statistica, 2021[25]).
Since 2017, Italy has incorporated into its budgetary planning a subset of 12 Equitable and Sustainable Wellbeing indicators among those published by ISTAT in its Annual Report, the first OECD country to do so (Blazey, Lelong and Giannini, 2022[22]).
The baseline level of these indicators is provided by ISTAT using data from the Italian EU-SILC Survey, the LFS Survey and several other sources. Since EU-SILC micro-data are available with a two year delay, in year “x” the models of the Ministry of Economy and Finance provide the estimate of the indicators for year “x-1” and the forecast for years “x+1, x+2, x+3”. These estimates are contained in two documents: the “Report on the Equitable and Sustainable Wellbeing Indicators” released to Parliament every February, and the Annex to the DEF, released in April.
The Report and the Annex are co-ordinated and produced mainly by the Department of the Treasury. The statistics within the report and annex concerning income inequality are estimated by the DSREF of the Department of Finance using the interquintile ratio (S80/S20).1 Following the methodology explained in the previous point, in the year “x”, the DSREF provides the impact on the inequality indicator induced by the fiscal policies and transfers of the previous year (“x-1”) and the forecast for the following three years accounting for according to currently planned the future policies already planned by the government (European Commission, 2020[26]).
← 1. The S80/S20 reports the ratio between the disposable income of the 20% richest and that of the 20% poorest households of the income distribution.
Since then, each year the Treasury Department of the MEF assesses the impact of any new reforms, and publishes its assessment in the two aforementioned key reports. The Report on the ESW Indicators highlights the measures introduced by that year’s Budget Law11 that are most relevant to the ESW indicators and provides forecasts of these indicators for the next three years, based on the notes of update to the DEF.12 The annex to the DEF, which is presented to the Minister of the MEF, updates the forecast of these ESW indicators by considering the new macroeconomic and financial framework defined in the new DEF. These forecasts require specific models that sometimes also consider macroeconomic variables. The resultant analyses often form part of the public debate and are sometimes considered during parliamentary debates.
While the production of the two reports is mandated by law, ministries are not required to use ESW in their documents, including in their integrative notes13 – the guidelines to preparing the integrative notes do suggest the use of ESW indicators, but in practice very few ministries do so. Indicators are also not a part of the Italian government’s performance budgeting framework, and thus do not need to be part of the performance plans that ministries submit to Parliament. This means that the budget often will not demonstrate the relevance of ESW to new initiatives that are to receive funding (Blazey, Lelong and Giannini, 2022[22]).
5.2.1. Other ministries and agencies
In Italy as in other countries, the analysis of the distributional implications of public expenditure is also conducted at the level of some of the social ministries or agencies. The section below discusses the role of two important institutions, the National Social Security Institute and the National Institute for Public Policy Analysis.
5.2.2. The National Social Security Institute
The National Social Security Institute14 (INPS) is the main Italian entity that concerns itself with managing social monetary transfers, both for pensions and for other forms of welfare. The Institute is run by several bodies, which together draw up, debate and approve both the strategic objectives and the budget of the INPS (INPS, 2023[27]).
There are several teams within the INPS responsible for the analysis of welfare issues. The first of these is the Central Credit Welfare and Social Facilities Directorate15 (CCWSS). The CCWSS is responsible for managing all the activities related to the provision of social benefits, as well as the use of various statistical tools in order to monitor and forecast the impact of these benefits. It also analyses regulatory changes in the areas within its jurisdiction, in order to assess their impacts on the INPS’ activities. To this end, the team will prepare reports on the trends in the services it provides, and subsequently provide these reports to the Bodies. Another related team is the Central Pensions Directorate16 (DCP), which is responsible for the management of pensions and redundancy benefits, and provides its own similar reports to the Bodies (INPS, 2023[27]).
Both the CCWSS and the DCP receive assistance for their analytical endeavours via the General Statistical Actuarial Coordination17 (CGSA). The CGSA creates medium to long term forecast models, and provides consultancy for the evaluation of new legislative measures. In collaboration with the CGSA, the CCWSS provides analyses on the trends of income support benefits and evaluates the short, medium and long-term effects of these benefits. The DCP’s collaboration with the CGSA allows it to manage the Pension’s Register, and prepare reports on the trends in pension benefits for submitting to the Bodies, as well as carry out its own analyses on the short, medium, and long-term trends in demand for pensions, and their impacts (INPS, 2023[27]).
Another key analytical team that the CCWSS, DCP and CGSA all work with is known as the Central Studies and Research Directorate18 (DCSR), whose overarching role is to provide technical support for the decisions that the INPS makes. This is the team responsible for co-ordinating the writing the INPS’ Annual Report, which, among many other topics, includes analysis in income inequality and gender income inequality, as well as the impact of policies aimed at reducing inequality and poverty. The team develops and manages the INPS’s microsimulation models in order to assess the impact of welfare policies on the conditions of families and businesses. In liaison with the CGSA, CCWSS and DCP, it promotes the development of new databases, aimed at improving the data frameworks of the INPS. Furthermore, it encourages the interaction between the world of research and the Institute’s staff (INPS, 2023[27]).
5.2.3. The National Institute for Public Policy Analysis
The National Institute for Public Policy Analysis (INAPP) is the research body of the Ministry of Labour. As such, its primary focus is labour market policies (for example, looking at the impacts of temporary vs permanent contracts on medium- and long-term economic growth) but it utilises distributional analysis insofar as it concerns impacts on the labour market. Some of this work has been conducted using LABSIM, an additional microsimulation model developed by the Centre for Microsimulation and Policy (CEMPA) in the Department of Finance, financed by INAPP. The model uses EUROMOD outputs to estimate how a new policy is likely to impact the labour market, as well as any other behavioural impacts it may have (Biagetti, Ferri and Marsiglia, 2023[28]).
The Baby Bonus was introduced in 2015 to help incentivise Italians to have children. In 2018, the policy was intended for families with children under 3 years old and with an income below EUR 25 000. In 2019, it was reformed to only include children under 1, and in 2020 reformed again to consider those with incomes up to EUR 40 000.
As this system was to be replaced by the AUU (see Box 5.2) INAPP deemed it valuable to look at how it had helped the population while it was in place. While government provisions such as the baby bonus do not fall directly under INAPP’s remit, it was deemed relevant due to the impact of such provisions on the labour supply of women. The work used the EUROMOD Italy micro-simulation model, and EU-SILC data between 2016 and 2018.
The analysis looked at four scenarios: one without the baby bonus, and then one with each of the three systems in place between 2018 and 2020. It found that the income distribution was most unequal without the bonus, and, according to the Gini coefficient, most equal in 2018. It also found that average family income was highest for the 2018 system. It thus concluded that the AUU would benefit from having a component dedicated to new-borns.
Source: (Ferri, Ricci and Scicchitano, 2021[29])
5.3.1. Microsimulation Modelling: TAXBEN-DF (Department of Finance)
The key model used by DSREF for analysing the redistribution effects of the tax-benefit system on households is called TAXBEN-DF (I). It is based on the Italian version of EU-SILC, a survey provided by the ISTAT that builds a representative sample of the Italian population made up of 21 325 households and 45 761 individuals. TAXBEN-DF uses both survey data from the Italian version of the EU-SILC survey and administrative data from administrative tax returns, exactly matched through individual tax codes, and managed by the Department of Finance (for more details on data sources see the section “Data and information infrastructure”). The model estimates PIT revenue and other wealth and income taxes (i.e. taxation of housing and financial assets, flat tax on certain types of incomes, etc.), as well as social contributions and the impact of household transfers (i.e. family allowances, Citizenship Income, etc.). All this information allows it to simulate the impact of new policies on tax revenue and income distribution (Miola and Manzo, 2021[30]) .
The model contains several different modules that allow for the transition from the individual gross income coming from tax return data to the household disposable income used for distributional impact analysis (see Figure 5.11). Firstly, using the tax identification number, six sources of administrative and survey data are matched, providing the model’s complete database (as the EU-SILC survey sample-weight do not allow to reproduce exactly the major fiscal aggregates, a re-weighting process is applied to replicate both demographics and income distribution resulting from statistics on tax return). Secondly, a first stream of procedures defines all individual incomes, which are normally updated to the current year by applying the growth rate of the latest macro-economic aggregates to wages and prices. All these incomes contribute to the identification of the taxable income. A second stream of procedures then simulates the tax-benefit system. Finally, the net PIT and the net individual income are derived by simulating the main benefits and transfers to households under current tax legislation. The last procedure is applied to harmonize the net tax income and the household disposable income, on which the redistributive analyses are carried out.
Source: (Subrizi and Miola, 2022[31])
The DSREF has recently developed a behavioural module of the microsimulation model, called TAXBEN-DF (II), which is used to evaluate the behavioural effects on labour supply of new policies, including labour participation rates, as well as examine second-order impacts of policies on income distribution, by analysing the impacts of behavioural changes. The model takes a discrete approach, grouping working hours into five distinct intervals, and uses a utility function to estimate income and hours of work. It assumes that women vary their labour supply while men work fixed hours, and excludes those who are retired, in education and self-employed (however, further developments of the models will make it more flexible). TAXBEN-DF (II) is fully integrated into TAXBEN-DF (I), meaning that any changes in income tax levels are exactly represented within it (MEF, 2022[14]).
Furthermore, the distributional effects that TAXBEN-DF (I) and (II) estimate are used as input into the Department of Finance’s computational general equilibrium model, ITAXCGE-DF.
The DSREF provides distributional assessments on a large amount of different outcomes, including the variation of household disposable income and how it is impacted by fiscal policies, benefits and transfers, in order to assess where the tax and benefits incidence is concentrated. It also looks at variation in equivalised19 disposable income and uses indicators to evaluate the distribution of benefits stemming from new policies, their impact on household income inequality and relative poverty, as well as the progressivity of the tax-benefit system. Disaggregated analysis by household composition, region, source of prevalent income, age, education, and other factors is also possible.
The DSREF has developed different microsimulation models to study revenue and distributional effects in a partial and general equilibrium setting of different types of taxes and benefits for firms and households (Table 5.4).
5.3.2. Microsimulation modelling: T-DYMM (The Treasury)
The Treasury Dynamic Microsimulation Model (T-DYMM) has been developed through three research projects sponsored by the European Commission, with the first in 2009 and the most recent in 2021. The model’s main aim is to provide long-term distributional analyses of the Italian social security system, hence in most cases it does not concern itself with the annual budget (although it is worth noting that the model has the technical capacity to do short term analysis, it just tends not to be used for this purpose). Its main aim is to look at the impact of pensions and social protection on income distribution, and to this end is able to use other models, including macrosimulation models within the Treasury, to study secondary impacts – for example, it is able to look at the impact of a given reform on employment rates, and then introduce those new employment rates into the model in order to examine their impact on income distribution. The model is able to analyse at both the individual and the household level, although it is not able to look at the regional level (Conti et al., 2023[32]). It receives the majority of its data from INPS and ISTAT, and some from the Department of Finance.
The model is organised into five modules (see Figure 5.12). For the demographic module, various assumptions are made in order to simulate individuals in the sample being born, aging, dying, migrating, getting education, forming couples and marrying, and becoming disabled. For the labour market module, logit regressions are used to determine whether individuals are employed, and if so, in what capacity. For the pension module, pensions (including disability and survivor pensions) are allocated based on legal requirements as well as the likelihood of the individual taking up the pension when eligible (Conti et al., 2023[32]).
The most recent version of T-DYMM has incorporated both a wealth module and a tax-benefit module. The wealth module simulates intergenerational wealth transfers, both as a result of donations and inheritances, and financial accrues from housing and other assets. The tax-benefit module simulates taxes and benefits at the national level.20 This tax-benefit module starts by calculating social insurance contributions, which are derived from EUROMOD, before moving on to calculating proportional taxes (i.e. capital income tax, some self-employment taxes) and income tax. It then calculates in-cash benefits, assuming a full-take up rate for all benefits except disability allowances, minimum income schemes and unemployment benefits, for which various probabilistic allocation mechanisms are used (Conti et al., 2023[32]).
Source: (Conti et al., 2023[32])
5.3.3. Microsimulation Modelling at INPS
The INPS use a microsimulation model in order to measure the redistributive impact of policy interventions, among other uses. The model breaks down results into various categories, including income, wealth, type of work and geographical area. INPS defines the model as somewhat in between a static and a dynamic model – meaning, in essence, that it is a static model with the ability to adjust incomes, demographic factors, revenues and costs, and other structural conditions. The model uses sample surveys and administrative data, matching the two through tax codes, where generally sample surveys are preferred for analyses of socio-economic issues such as inequality and poverty. The primary source of data for the model is ISTAT SILC data (Di Nicola, 2020[33]).
5.3.4. The ISTAT Microsimulation Model: FaMiMod
ISTAT’s main microsimulation model is known as FaMiMod, and is based on administrative data from the Ministry of Finance, matched to ISTAT survey data from EU-SILC. While the model is static, ISTAT incorporates macroeconomic data into the model using either National Accounts or forecasts from ISTAT’s macroeconomic mode, MeMo-It.
The updating of the model consists of three steps: 1) forward projection of monetary variables using either National Accounts or MeMo-It, 2) reweighting based on the most recent known populations breakdown by age, sex, and professional condition (i.e. employed, dependent, self-employed and unemployed), and 3) updating of the legislation within the model in order to have an up-to-date baseline model. At this point, the model is able to simulate the effects of new policies, comparing incomes and the distribution of incomes in the baseline scenario and a reform scenario. The definition of household income includes labour and capital gains, imputed rent from home ownership, the AUU, and extraordinary measures adopted to mitigate the impact of the recent cost of living increase. The Personal Income Tax Reform and the value of ISEE are estimated within the model (ISTAT, 2023[34]).
5.4.1. Income inequality data: combining administrative and taxation data
The administrative data for TAXBEN-DF (I) comes from personal income tax returns, the National Real Estate Cadaster, the Italian National Social Security Institute (INPS), the Archive of Reports with Financial operators (for financial assets) and the Single Substitute Declaration (the declaration of the income and wealth owned by each family member used to produce the ISEE). To study the effects of COVID-19 on incomes, the DSREF has also recently developed a nowcasting procedure that integrates most frequently available firm-level data, matched using a combination of 6-digit NACE sectors and Italian Regions. The data are managed by the Department of Finance (DSREF), which, as a member of the National Statistical System (SISTAN), is able to utilise data from other SISTAN members provided it is for analysis purposes within previously established and well-defined projects, while they are not available to external academia or research institutes (European Commission, 2020[26]).
The survey data for TAXBEN-DF (I) comes from IT-SILC (EU-SILC) elaborated by ISTAT. It includes information on incomes, personal characteristics, skills, education level, socio-economic status, and employment conditions. This data source, integrated with non-taxable income data from the INPS, is very important for providing analysis on household income distribution, helping fill in the gaps on the information not included in tax return data. Indeed, the use of survey data in conjunction with administrative data has several advantages. While administrative tax data tends to have greater coverage than survey data, it often provides little to no socioeconomic information on taxpayers or on non-taxable income, and limits coverage to the taxpaying population, meaning that the lower end of the income distribution can be underrepresented. Access to survey data also allows analysts to estimate the size of the shadow economy, through a comparison between survey incomes and tax returns incomes (Miola and Manzo, 2021[30]).
The Treasury’s T-DYMM model also matches survey data from ISTAT with administrative data from INPS, resulting in a final dataset that it called AD-SILC (although it is worth noting that T-DYMM, as a dynamic model, uses longitudinal data, while TAXBEN-DF, as a static model, uses cross-sectional data). More specifically, the model takes longitudinal data (up to four years) on various socio-economic characteristics for 254 212 individuals, while it takes longitudinal data on pensions and working history from INPS. Ad-SILC also contains information from tax returns and the Cadaster, as collected by the Department of Finance, and information from the Bank of Italy’s Survey on Household Income and Wealth ( (Conti et al., 2023[32]).
References
[28] Biagetti, M., V. Ferri and S. Marsiglia (2023), “Reddito di cittadinanza: aspetti economici ed effetti comportamentali [Citizen Income: Economic Aspects and Behavioural Effects]”, https://economiaitaliana.org/wp-content/uploads/2023/04/EI_2023_1_02_S_Biagetti_et_al_D.pdf (accessed on 23 October 2023).
[22] Blazey, A., M. Lelong and F. Giannini (2022), “The Equitable and Sustainable Well-being Framework in Italy: An Action Plan for its use in policy and budget decision making”, OECD Working Papers on Public Governance, No. 56, OECD Publishing, Paris, https://doi.org/10.1787/4f48c504-en.
[21] Blöndal, J., L. von Trapp and E. Hammer (2016), “Budgeting in Italy”, OECD Journal on Budgeting, https://doi.org/10.1787/budget-15-5jm0qg8kq1d2.
[17] Bratta, B., L. Romano and F. Mazzolari (2020), “The Impact of Digitalization Policies. Evidence from Italy’s Hyper-depreciation of Industry 4.0 Investments”, https://www.finanze.it/export/sites/finanze/.galleries/Documenti/Varie/dfwp6-1_ultimo.pdf (accessed on 28 July 2023).
[13] Clementi, F. and E. Valentini (2018), “Economia irregolare, criminalità e disuguaglianza dei redditi: un circolo vizioso? [Informal economy, crime and income inequality: a vicious cycle?]”, Moneta e Credito, https://www.proquest.com/docview/2181699206/fulltextPDF/2A1519D8E2774717PQ/1?accountid=163263&parentSessionId=iS9pnSd66aznqeRa7nIqNO2%2FsV4oKYl4ClEaGBD1whw%3D (accessed on 23 October 2023).
[32] Conti, R. et al. (2023), The Italian Treasury Dynamic Microsimulation Model (T-DYMM): data, structure and baseline results, https://www.dt.mef.gov.it/export/sites/sitodt/modules/documenti_it/analisi_progammazione/working_papers/WP-1-2023.pdf (accessed on 22 March 2023).
[23] De Minicis, M. et al. (2021), “MOSPI: Policy Recommendations Report”, https://www.inapp.gov.it/wp-content/uploads/2023/02/Inapp_MOSPI4_EN.pdf (accessed on 22 March 2023).
[18] Di Caro, P. et al. (2022), “One step forward and three steps back: pros and cons of a flat tax reform”, https://mpra.ub.uni-muenchen.de/113684/ (accessed on 28 July 2023).
[33] Di Nicola, F. (2020), “Modello di microsimulazione Tax Benefit e valutazione Reddito di Cittadinanza [Tax Benefit microsimulation model and Citizenship Income evaluation]”, https://www.inps.it/docallegatiNP/Mig/Dati_analisi_bilanci/Attivita_ricerca/Studi_e_analisi/modello_microsimulaz_stima_RdC.pdf (accessed on 31 January 2023).
[9] Eilat, Y. and C. Zinnes (2002), “The Shadow Economy in Transition Countries: Friend or Foe? A Policy Perspective”, World Development, Vol. 30/7, pp. 1233-1254, https://doi.org/10.1016/S0305-750X(02)00036-0.
[8] Enste, D. (2003), “The Shadow Economy and Institutional Change in Transition Countries”, https://www.brainguide.de/upload/publication/52/jtvj/e78473d06cd3faf157061dc7e5b12e32_1311535210.pdf (accessed on 17 January 2023).
[26] European Commission (2020), Questionnaire on the use of distributional impact assessment (DIA) in budgetary processes and planning of reforms: Italy.
[29] Ferri, V., A. Ricci and S. Scicchitano (2021), “Sostegno economic ai genitori: Il Bonus Bebè attraverso i più recenti cambiamenti [Economic support for parents: The Baby Bonus through the latest changes]”, https://oa.inapp.org/xmlui/handle/20.500.12916/940 (accessed on 19 April 2023).
[3] Fornero, E. (2021), “La réforme italienne des retraites”, Revue de l’OFCE, Vol. 170/6, pp. 59-84, https://doi.org/10.3917/reof.170.0059.
[19] Il Sole 24 Ore (2023), “Come il MEF misura l’impatto dell’economia sul benessere [How the MEF measures the impact of the economy on wellbeing]”, https://www.infodata.ilsole24ore.com/2023/03/17/come-il-mef-misura-limpatto-delleconomia-sul-benessere/ (accessed on 27 July 2023).
[5] Il Sole 24 Ore (2022), “Le pensioni Inps, il reddito e la disuguaglianza. Cosa è successo nel 2021? [The INPS pensions, income and inequality. What happened in 2021?]”, Il Sole 24 Ore, https://www.infodata.ilsole24ore.com/2022/03/24/le-pensioni-inps-reddito-la-disuguaglianza-cosa-successo-nel-2021/?refresh_ce=1 (accessed on 31 January 2023).
[27] INPS (2023), Istituto Nazionale della Previdenza Sociale, https://www.inps.it/ (accessed on 31 January 2023).
[4] INPS (2019), “Disugaglianze e iniquita’ sociali: sfide future [Social inequalities and inequities: future challenges]”, https://www.inps.it/docallegatiNP/Mig/Dati_analisi_bilanci/relazione_presidente_versione_ON_line_XVIII_RA.PDF (accessed on 31 January 2023).
[1] ISTAT (2023), “Condizioni di Vita e Reddito della Famiglia [Living Conditions and Family Income]”, https://www.istat.it/it/files//2023/06/REPORT-REDDITO-CONDIZIONI-DI-VITA2022.pdf (accessed on 13 July 2023).
[34] ISTAT (2023), “Private Interviews”.
[6] ISTAT (2022), “Condizioni di Vita e Reddito delle Famiglie [Living Conditions and Family Income]”, https://www.istat.it/it/files/2022/10/Condizioni-di-vita-e-reddito-delle-famiglie-2020-2021.pdf.
[7] ISTAT (2022), “Rapporto BES 2022: IL Benessere Equo e Sostenibile in Italia [2022 BES Report: Equal and Sustainable Wellbeing in Italy]”, https://www.istat.it/it/archivio/282920 (accessed on 9 June 2023).
[12] ISTAT (2021), Economia non osservata in calo prima dell’emergenza Covid-19 [Informal economy in decline before Covid-19 emergency].
[25] Istituto Nazionale di Statistica (2021), “BES 2021: Il Benessere Equo e Sostenibile in Italia [BES 2021: Equal and Sustainable Wellbeing in Italy]”, https://www.istat.it/it/files//2022/04/BES_2021.pdf (accessed on 15 December 2022).
[20] Italian Department of Finance (2022), Assegno Unico Universale e revisione dell’Irpef: effetti distributivi sulle famiglie italiane [Universal Single Allowance and Irpef revision: distributional effects on Italian families], Italian Department of Finance, Rome, https://www.finanze.it/it/inevidenza/Nota-tematica-N.6-Assegno-Unico-Universale-e-revisione-dellIrpef-effetti-distributivi-sulle-famiglie-italiane/ (accessed on 13 December 2022).
[35] Italian Government (2012), Descrizione dei domini e degli indicatori del Bes selezionati dalla Commissione scientifica e varati [Description of the domains and indicators of the Bes selected and launched by the Scientific Commission], https://www.istat.it/it/files//2018/04/12-domini-commissione-scientifica.pdf (accessed on 14 December 2022).
[24] Italian Ministry of Economics and Finance (2018), “Comitato per gli indicatori di benessere equo e sostenibile [Committee on Fair and Sustainable Wellbeing Indicators]”, https://www.mef.gov.it/ministero/comitati/CBES/index.html (accessed on 25 January 2023).
[10] Italian Ministry of Economy and Finance (2022), “Relazione sull’economia non osservata e sull’evasione fiscale e contributiva – anno 2022 [Report on the unobserved economy and tax and tax evasion - year 2022]”, https://www.finanze.gov.it/export/sites/finanze/.galleries/Documenti/Varie/Relazione-sulleconomia-non-osservata-e-sullevasione-fiscale-e-contributiva-anno-2022.pdf (accessed on 28 July 2023).
[15] MEF (2023), “MEF Organigramma (MEF Organigram]”, https://www.mef.gov.it/ministero/struttura/organigramma.html (accessed on 13 July 2023).
[14] MEF (2022), Private Interviews.
[30] Miola, E. and M. Manzo (2021), A Tax-Benefit Microsimulation Model for Personal Income Taxation in Italy, Italian Department of Finance, Rome, https://www.finanze.it/export/sites/finanze/.galleries/Documenti/Varie/WP_A-tax-benefit-microsimulation-model-for-PIT-in-Italy_final1.pdf (accessed on 13 December 2022).
[16] Monteduro, M., D. De Rosa and C. Subrizi (2023), “How to Nowcast Uncertain Income Shocks in Microsimulation Models? Evidence from COVID-19 Effects on Italian Households”, Italian Economic Journal, https://doi.org/10.1007/s40797-023-00232-8.
[2] Sanchez-Romero, M. and A. Prskawetz (2020), “The Impact of Reducing the Pension Generosity on Inequality and Schooling”, De Economist, Vol. 168/2, pp. 279-304, https://doi.org/10.1007/s10645-020-09359-w.
[31] Subrizi, C. and E. Miola (2022), “Module Flow for the TAXBEN-DF Model”.
[11] Valentini, E. (2022), “Le disuguaglianze che economia sommersa e criminalità portano con sé [The inequalities that the shadow economy and crime bring with them]”, https://ristretti.org/le-disuguaglianze-che-economia-sommersa-e-criminalita-portano-con-se (accessed on 17 January 2023).
Notes
← 1. On the impact of tax and transfers on inequality, see the dynamics of the interquintile ratio estimated in the “Equitable and Sustainable Wellbeing Report” in the years 2022 (https://www.mef.gov.it/inevidenza/2022/article_00071/Relazione-BES-2022_03_03.pdf) and 2023 (https://www.dt.mef.gov.it/export/sites/sitodt/modules/documenti_it/analisi_progammazione/documenti_programmatici/Relazione-BES-2023.pdf). On the Gini index up to 2020, see http://dati.istat.it/Index.aspx?DataSetCode=DCCV_GINIREDD; for 2022 see https://www.istat.it/it/archivio/277878.
← 2. Of the EUR 218 billion spent on pensions in 2021, EUR 195.4 billion (89.3%) came from social security schemes, with the remainder funded by welfare schemes (Il Sole 24 Ore, 2022[5]).
← 3. i.e. personal income taxes, corporate income taxes, VAT and excise duties.
← 4. Data on households’ incomes are available with a two-year delay, implying the necessity of microsimulation models to analyse the impacts also of recently adopted policies.
← 5. The policy notes are available here: https://www.finanze.it/it/il-dipartimento/Analisi-economiche-e-fiscali-note-tematiche/notetematiche.
← 6. The WPs are available at this link: https://www.finanze.it/it/il-dipartimento/collana-di-lavori-e-di-ricerca/working-papers.
← 7. These are published less regularly, usually once every two years.
← 8. For a full description of the nature and data source of each indicator, see Descrizione dei domini e degli indicatori del Bes selezionati dalla Commissione scientifica e varati (Italian Government, 2012[35]).
← 9. The “Committee for Fair and Sustainable Welfare Indicators” (Comitato per gli indicatori di benessere equo e sostenibile) was chaired by the Minister of Economy and Finance, and composed of the President of ISTAT, the Governor of the Bank of Italy, and two field experts.
← 10. The other indicators were as follows: adjusted gross disposable income per capita, healthy life expectancy at birth, proportion of population that is overweight, early exit from education rate, employment rate of women with young children and without children, predatory crime rate, civil justice efficiency, CO2 emissions, and unauthorised building rate.
← 11. Legge di Bilancio in Italian.
← 12. These notes, known as NADEF (Nota di Aggiornamento al Documento di Economia e Finanza) are presented to the Chamber before the end of September each year in order to update the predictions and policy objectives of the DEF due to the greater availability of information.
← 13. Integrative notes (note integrative) are documents written by the line ministries in order to demonstrate to the General State Accounting Office (Ragioneria Generale dello Stato, RGS) their objectives and the financial resources needed to realise them.
← 14. Istituto Nazionale della Previdenza Sociale in Italian.
← 15. Direzione Centrale Credito Welfare e Strutture Sociali in Italian.
← 16. Direzione Central Pensioni in Italian.
← 17. Coordinamento Generale Statistico Attuariale in Italian.
← 18. Direzione Centrale Studi e Ricerche in Italian.
← 19. Disposable household incomes are made equivalent by the application of the modified-OECD-equivalence scale: disposable household incomes are divided by a quotient that is the sum of individual coefficients, i.e. 1 for the first adult, 0.5 for every other adults and 0.3 for every component younger than 14.
← 20. T-DYMM does not include an internal migration module and no regional-level taxes and transfers are simulated. Furthermore, no COVID-19-related transfers are simulated in the model, for three main reasons: 1) much of the COVID-19-specific aid took the form of salary integration for those suspended from work, which the model is not able to consider separately from other labour income, 2) there was insufficient data on lump-sum transfers to self-employed workers , and 3) the medium- and long-term focus of the model means that emergency measures such as those related to COVID-19 are not of immediate relevance.