The Italian Index of Equivalent Economic Situation

Transcription

1 The Italian instrument of means-testing in the light of the Portuguese experience Cristina Berliri Ministry of Labour and Social Affairs Roberta Zizza Bank of Italy Introduction 1 In Portugal, to face the adverse impact of the crisis, a law approved in 2010 changed the means testing criteria to be held eligible for minimum income provision, reforming the so called Social Integration Income Programme (RSI). While striving for improving efficiency and effectiveness of the public social security system, the envisaged measures were clearly aimed at restraining the growth of public expenditure. It is since the late Nineties that Italy is endowed with a means testing tool for assessing the entitlement to non-contributory social assistance benefits, the Index of Equivalent Economic Situation (ISEE). The ISEE regulates the access to a wide variety of social benefits, both as a tool for selection of beneficiaries and to calculate the shares of participation to the costs of services. Notably, a value of the ISEE below a certain threshold is one of the requisites for the so-called Social Card, a purchase credit card issued since 2009 by the Italian Government with the aim of supporting particular types of households in absolute poverty 2. In this note we compare the indicators for means testing in these two countries, identifying communalities and differences and hence potential margins of improvements of the two experiences. This could be of particular interest for Italy as last June the Government has been delegated to reform the ISEE (delega assistenziale), in particular in the direction of paying more attention to the family composition. The Italian Index of Equivalent Economic Situation The Italian welfare system provides non-contributory social assistance benefits only in reference to the protection of weaker population because of reached retirement age limits or disability conditions. In fact, there is not a national framework for minimum income benefits to cover the whole population from the risk of economic deprivation. 1 Opinions expressed are those of the authors and do not necessarily reflect those of the institutions of affiliation. for correspondence: roberta.zizza@bancaditalia.it; cberliri@lavoro.gov.it. 2 The Social card was introduced on the occasion of rising price of energy products and food, to support families in need including either children under three years of age or people older than sixty five years. It currently covers a limited target of beneficiaries, however, the proposal of its wider use as a tool for fighting absolute poverty is under discussion. Portugal 1

2 It should be clarified, however, that since the constitutional reform of year 2000, the legislative competence in the field of social assistance has become regional, while management and administration of social services continue to be entrusted to the Municipalities. The State has jurisdiction only in the determination of basic levels of social assistance, competence that is in fact not yet fully exercised. The result is a system of social interventions rather uneven and geographically differentiated. Before the constitutional reform, following the work of the Onofri Commission 3 (1997) which suggested to structure the delivery of welfare services and benefits according to the principle of selective universalism, it was established to identify a unique tool for determining the economic status of families. The ISEE represents a unified assessment criteria of the economic situation of people asking for economic support (maternity and large household allowances, Social card, local income support benefits, exemption from schoolbook and transport costs, scholarships, university attendance, housing beneficiary lists and lower rent, free of charge access to health care, nurseries, crèches, refectories and school buses, etc.). Currently about 30 per cent of Italian families apply for benefits by presenting the ISEE. In what follows we first describe the main features of the ISEE with special respect to the areas that were the subject of the Portuguese reform: the concept of family unit, the income and asset categories considered, the tapers and the equivalence scale. Then, we provide a more general assessment of the ISEE. The main features of the ISEE: a comparison with its Portuguese analogue The concept of family In the ISEE the family is made by the declaring person, her/his spouse and their children, as well as other people living with her/him and other people dependent on her/him in reference to the personal income tax. The Portuguese reform has properly widened the definition of family reference unit, in a way that better proxies the concept of household; it now includes every relative that is living and sharing resources with the applicant. The income declared for tax purposes In the ISEE income coincides with the one declared for tax purposes, to which it is added the income from financial assets with a deduction for the cost of hiring. In addition, the 20 per cent of financial assets and real estate, with the application of tapers (see below), is considered. The reference to the income declared for tax purposes must be held problematic, at least in the Italian case (it is not clear if the RSI refers to the same income definition). There are two critical issues in this case. First, tax-free transfers, as social pensions, pension for accidents, disability etc., are 3 Commission, chaired by Professor Paolo Onofri, appointed by the President of the Council of Ministers to analyze "the macroeconomic compatibility of social spending." Portugal 2

3 excluded from the computation of the ISEE. The exclusion of such transfers is often motivated by the fact that their being exempt from tax would signal a kind of privileged status granted to them by the law, qualifying them as being an essential benefit, which therefore should not be significantly affected by taxes and withdrawals, either explicit or implicit (in the form of lower probability of assignment of other benefits). Especially for certain types of benefits, particularly when they are designed to tackle extreme poverty, a lack of consideration of transfers exempt from tax, combined with the presence of high tapers, which do not take into account the presence of assets, has made it difficult to graduate the potential beneficiaries on the basis of differences in the actual living standards. This consideration is linked to the previous one concerning the uniqueness of the instrument. In Italy this passepartout feature of the ISEE can lie behind its low capacity as a screening device in the lowest tail of the distribution (10 per cent of households have an ISEE equal to zero; more than 650,000 households are indistinguishable in terms of ISEE; Berliri and Tangorra, 2011). A further weakness affecting the ISEE indicator is its reference to incomes received long before than the submission of the request for access to the benefit or service. In fact, the incomes declared for tax purposes are those produced one year or even two years before submitting the application. Therefore, a different tool is needed if the means test should be designed to act promptly to deal with emergency situations related to substantial changes in the economic status of families. Since the onset the ISEE has been conceived as a combination of an income component and an asset component (both financial and real items), where the inclusion of the asset component was meant to attenuate the distortionary effects of the income underreporting, which is held very high in the Italian economy (more than one sixth; see Istat, 2011). As also the Portuguese hidden economy accounts for a relevant part of GDP (almost one fifth; see Dell Anno, 2007, or Schneider, 2011), the recent inclusion of financial and property income among the income categories considered in the evaluation of the resources of the potential beneficiaries seems appropriate, as assets are more difficult to hide than income. In the same vein including the housing benefits in the determination of total income and benefits in kind should render the instrument more effective in selecting needy families. As income and asset underreporting can imply a benefit misallocation, the traditional means testing could be coupled with a so called proxy means testing, based on additional information correlated with the standard of living (e.g. owning a luxury car). Tapers Regarding tapers, the ISEE has tapers on property and capital assets and a deduction for rent expenditures. If on the one hand the rationale for these tapers is the special consideration of the house of residence as a basic need, on the other hand they severely undermine the selective capacity of the indicator and in particular might lead to an understatement of the role of the inclusion of the assets in the light of the high income underreporting. The use of tapers, as well as the exclusion from the computation of the ISEE of some exempt income sources, can reflect in a paradoxical situation where the ISEE favours households that already receive some kind of benefits, at the expenses of those completely excluded by our welfare system (that, inter alia, is one of the few in Europe lacking a minimum income scheme). Moreover, homeowners are held favoured vis-à-vis renters due to the values of the respective tapers. Portugal 3

4 Still on this, while the income categories considered for the computation of the ISEE are much less comprehensive than the Portuguese analogue, the ISEE does not envisage any taper on work income. Reducing potential disincentives to work, together with the activation elements introduced by the Portuguese reform for some kinds of benefits (training, education or integration into the labour market policies, all targeted to people aged between 18 and 55) seem consistent with the workfare paradigm. The unique work activation element envisaged in the current ISEE stands in the additional weight granted to families with working parents (see infra). Definition of the equivalence scale The Portuguese reform has changed the definition of the equivalence scale, a crucial determinant of the selective capacity of this kind of means-tests. The equivalence scale increases with the number of components, but weights assigned to the different members of the family can differ significantly (see table). Comparison between equivalence scales No. components Italian ISEE Portuguese RSI* Portuguese previous RSI * OECD MODIFIED OECD Italian ABSOLUTE POVERTY * The scale described here refers to the case where the second component was adult and subsequent components were minor. In Portugal the original - very generous indeed - scale was replaced by a scale that is equal to the OECD one in case of a household with only two persons aged more than 18. This despite the fact that most recent OECD releases use either its modified version or the square root scale (implying that, for instance, a household of four persons has needs twice as large as one composed of a single person), both less generous. The scale currently used for the calculation of the ISEE lies in between. More generally, the scope of the weights is to realize the equalization of the well-being among households with different compositions, and this should be done on the basis of sound and rigorous econometric estimates. The weights set at the national level have been derived by the estimates produced by Carbonaro (1985) on the basis of consumption data available at that time and using the approach pioneered by Engel in 1895, centered on the share of food consumption on total consumption/income. In this regard an update would be recommendable, especially considering that the share of food consumption is rapidly decreasing so that it is now closer to other consumption items, such as transports or renting. In addition the following ad hoc increases are envisaged, which have been arbitrarily set by policy makers taking inspiration from those used by the University of Trento for her provisions: 0.35 for each additional component; 0.2 in the presence of children and one parent; Portugal 4

5 0.5 for each person with disabilities; 0.2 in the presence of a minor and of both parents working (or lone parent working). Also these specific bonuses that acknowledge the state of particular need of given family-types could be derived in a transparent way, on the basis of econometric estimates. The ISEE equivalence scale has also been recently questioned as it would penalize larger families: for example a family with two adults and two children would correspond to a value of 2.46 in the ISEE scale. This criticism would in principle apply with more strength to the OECD modified scale, which would assign a value equal to 2.1 to the same type of family. In general the existence of economies of scale, particularly relevant in high-income countries where the share of necessary good 4 on total consumption is lower, should go in favour of a not-increasing structure of the weights within household. Another critique that could be raised deals with the fact that the weights are independent of the age of the child 5, whereas the maintenance costs and the availability and use of public services could vary dramatically by age. The last column of the Table shows the equivalence scale that can be implicitly inferred from the Italian National Statistical Institute (Istat) measure of absolute poverty, based on the cost of a basket of consumer goods of low income families. The comparison among scales suggests that the scale that is able to render equivalent families of different composition may be different for the families who occupy the lower part of the income distribution. Finally, while acknowledging that a unique harmonized scale is helpful in the context of international comparisons, e.g. at the OECD level, scales should be properly defined at the national level, as several costs are highly country-specific (e.g. health care and child care). The cost of children, for example, could be better judged at national level, depending on the extent by which the welfare system takes care of children of different ages and other features of the national context. A sound approach could involve the use of national scales based on a harmonized methodology. A more general assessment of the ISEE The uniqueness of the instrument A first general comment deals with the aim of providing a harmonized framework for noncontributory benefits of both the ISEE and the RSI, meaning that the same tool is used for assessing the conditions of admittance to an extremely wide range of social benefits in the respective countries. If on the one hand simplicity and uniqueness in the definition of well-being represent pros of this option, on the other hand a one-size-fits-all tool can have disadvantages in terms of rigidity and capacity of selecting the right pool of beneficiaries, against the background of target populations ranging from households at risk of extreme poverty to households demanding for childcare services. 4 Individual consumption goods. 5 This criticism is not applicable to the Portuguese system. Portugal 5

6 It would be then recommendable to tune in a finer way the ISEE or the RSI on the basis of the type of service demanded. There is an interesting case in Italy that works along these lines, the ICEF (Economic and familiar conditions index), which is used for welfare provisions in the autonomous province of Trento (Barbieri, 2011). Despite the fact that potential beneficiaries fill in a unique data declaration, income categories and members of the household considered change according to the type of benefit demanded. A prerequisite for this is a well-functioning informational system. Selectivity and vertical equity The use of the ISEE has been gradually extended under the principle of selective universalism, aimed at substituting the existing categorical welfare model with a model which is universal in beneficiaries and selective in the delivery of benefits. Currently the means-test by ISEE determines the entitlement or the shares of participation to the costs for a wide range of non contributory benefits or services, provided at national or local level. However, the wide spread of the instrument can lead to situations of unfair distribution. In fact, the principle of vertical equity would require that government intervention, other conditions being equal, such as the needs deserving of protection, do not change the pre-existent order in the income distribution. However, conditioning the access to social assistance benefits and services to having incomes below a certain threshold may lead to a reversal in the ordering of families on the basis of the disposable income, penalizing families with incomes slightly higher than the threshold values. These effects will be stronger the wider the range of facilities subject to means-testing. Another way of seeing the same problem is to consider the effective marginal tax rate of the system of taxes and public transfers. For many low-income families it could be well over 100 per cent. This means that an increase in labor income could result in a negative effect on disposable income due to increased taxes, decreased benefits and reduced facilities. In Italy, where benefits are set at different levels of government (national, regional and municipal level), the combined effect of the provision of means-tested benefits and services should be carefully assessed, in relation to both the principle of vertical equity and the incentives to work in the formal sector of the economy rather than the informal one. The efficiency issue One of the important issue is that of increasing the efficiency of non-contributory social assistance with respect to the aim of not introducing distortions in the choice of participating to the labour market. As Italy lacks a national minimum income scheme 6, the efficiency theme is held less crucial, when interventions at national level are concerned, as social assistance covers people out of the labour force (the elderly, people with disability). As already discussed in the previous paragraph, distortionary effects may descend from the combination of several means-tested benefits and services. Distortions may also arise when one considers the interaction between social assistance for the elderly and contributory pensions, as the latter could be not larger for low-wage workers or for 6 Attempts of introducing an insertion minimum income (reddito minimo d inserimento) have been made at local level (see Granaglia and Bolzoni, 2010, for a review). Portugal 6

7 workers with short or discontinuous working histories. This makes it not neutral the choice between working either in the formal or in the informal sector 7. Non-take up and fraud Non-take up of social assistance is not an issue in Italy with reference to social pensions and assistance to people with disability. A major concern is instead for fraud. In particular the Social Security Institute improved the audit activities in order to detect the false positives among those receiving disability benefits. In the same vein a recent law (decree-law 78/2010), not yet implemented, established that all institutions in charge of providing benefits or services based on means-test proofs should provide the list of beneficiaries to the INPS, that cross-checks the information with that available at the tax authority to verify if eligibility criteria hold. This exchange of information should be the base of a comprehensive dataset able to assess the extent of multiple benefits granted to the same individual/household to protect for the same risk. It is a prerequisite for a rationalisation of the welfare system and for a multidimensional treatment of those in need. References Barbieri, G. (2011) ICEF: Indicatore della condizione economica familiare, Prospettive sociali e sanitarie, n , p Berliri, C. and R. Tangorra (2011) L ISEE: 10 anni di prova dei mezzi, Prospettive sociali e sanitarie, n , p Carbonaro (1985) Nota sulle scale di equivalenza, in Rapporto sulla povertà in Italia, Roma, p Dell Anno, R. (2007) The shadow economy in Portugal: An analysis with the MIMIC approach, Journal of Applied Economics, vol. 10, p Granaglia, E. and Bolzoni, M. (2010) Il reddito minimo di inserimento. Analisi e valutazioni di alcune esperienze locali, Quaderni CIES no. 3. Istat (2011) La misura dell occupazione non regolare nelle stime di contabilità nazionale, press release. Schneider, F. (2011) The Shadow Economy and Shadow Economy Labor Force: What Do We (Not) Know?, IZA Discussion Papers Marano, A., Mazzaferro, C. and M. Morciano (2011) The strengths and failures of incentive mechanisms in notional defined contribution pension systems, manuscript. Portugal 7