Reforming the Personal Income Tax and the Family Allowance in Italy: A Proposal

Matteuzzi, Massimo; Toso, Stefano Working Paper Reforming the Personal Income Tax and the Family Allowance in Italy: A Proposal Quaderni - Working Paper DSE, No. 266 Provided in Cooperation with: University of Bologna, Department of Economics Suggested Citation: Matteuzzi, Massimo; Toso, Stefano (1996) : Reforming the Personal Income Tax and the Family Allowance in Italy: A Proposal, Quaderni - Working Paper DSE, No. 266, Alma Mater Studiorum - Università di Bologna, Dipartimento di Scienze Economiche (DSE), Bologna, https://doi.org/10.6092/unibo/amsacta/5044 This Version is available at: https://hdl.handle.net/10419/159109 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc/3.0/ 1 Reforming the Personal Income Tax and the Family Allowance in Italy: a Proposal Massimo Matteuzzi and Stefano Toso* Dipartimento di Scienze Economiche, Università di Bologna September 1996 Abstract We illustrate a proposal to reform the personal income tax and the family allowance in Italy which aims at attaining a fair tax treatment of the family, laying emphasis on the redistributive processes in favour of large and poor households. Our proposal is based on a simplification of the tax rate schedule (with a reduction of income brackets) and a new design of the tax credits for dependents which implies the fiscal exemption of a sort of basic income, which is decreasing as the income of the household increases. The new scheme of supplementary tax credits for dependents, related to family income thresholds, would match the need to keep into account the economic conditions of the household with the need to avoid the abandonment of the definition of ability to pay on individual basis. The proposal also aims at attaining a closer integration between the personal income taxation and the family allowance, with a sizable revaluation of the amount of family allowances and its extension to self-employee. The distributional analysis, using the tax-benefit model DIRIMOD, shows a slight revenue loss and a significant increase, with respect to the 1995 system, in both the progressivity and the redistributive impact of the income tax. The distributional effects are even stronger if we keep into consideration the reformed system of family allowances. Acknowledgements This paper would not have come out without the invaluable cooperation of Daniela Mantovani. We acknowledge our debt of gratitude for her help in using the tax-benefit microsimulation model DIRIMOD. We are also grateful to Paolo Bosi, Maria Cecilia Guerra, Adriano Di Pietro, Silvia Giannini and Nicola Sartor for their useful comments and suggestions. This paper is part of HCM Network on Comparative Social Policy and Taxation Modelling. The usual caveats apply. JEL classification: D31, H24, H53 * Correspondence to: Massimo Matteuzzi, University of Bologna, Dept. of Economic Sciences, Strada Maggiore 45, 40125 Bologna, Italy, fax: IT-51-6402664, e-mail: [email protected] 2 Keywords: income tax reform, family benefits, personal income distribution 3 1. Introduction This paper illustrates a reform proposal of the personal income tax (IRPEF) and the family allowance (assegno al nucleo familiare) which aims at attaining a fair tax treatment of the family, laying emphasis on the redistributive processes in favour of large and poor families. Among the major shortcomings of Italy’s redistributive policies there are definitely those that are a result of the failure to consider the economic conditions of the family in the assessment of personal tax, as well as those that are a result of the increasingly more limited role played by the family allowance institute1. Quite naturally, a proper solution of these problems may not leave aside different aspects, including such requisites as the equity and neutrality of taxation with respect to an efficient allocation of resources; the coordination of personal income tax with schedular tax regimes relating to special categories; the simplification and transparency of the tax system; the elimination of unjustifiable discriminations among various categories of taxpayers and/or beneficiaries. These objects shall be taken into account by incorporating some of the issues that emerged from the debate on the reform of personal income taxation2. However, the core aspect of our proposal relates to the tax treatment of the family, as we believe that it is the preliminary issue to be solved. This is also due to the fact that the current system features some extremely striking gaps in the scheme of tax reliefs for dependents. 2. Personal taxation and economic conditions of the family The economic conditions of the family may affect the personal income tax assessment to a considerable extent, whether directly or indirectly. They have a direct influence whenever it is the family, rather that the individual, to be selected as the tax unit and, therefore, the tax base coincides with the aggregated income of the family3. 1 See Franco, Sartor (1990) for a systematic analysis of the problems related to the tax/benefit treatment of the family, as well as for an evaluation of the Italian experience. 2 See Messere (1993) and Sandford (1993) with respect to the international reform process of the 1980s. See Cnossen, Messere (1990) and McLure, Zodrow (1993) for an accurate synthesis of the theoretical debate that steered that process; for a critical review of the reform prospects in the Italian system see Bernardi (1989) and Visco (1992). As for specific proposals for the reform of the personal income tax in Italy, see the scheme devised by V. Visco (1988) and the more recent proposal included in the Tremonti White Paper. The preparatory studies for the implementation of such a reform, coordinated by L. Bernardi, have now been published in a monographic dossier in Rivista di diritto finanziario e scienza delle finanze: see Bernardi (1995). A synthesis of the main comments on the IRPEF reform proposal that was put forward in the Tremonti White Paper is included in Toso (1995). 3 This is the system in force in some leading Western countries, such as the USA and France, which adopt the income splitting and the family quotient system, respectively. An authoritative support to the selection of a family-based tax unit is provided by the Carter Report, which marked a significant turn in the history of tax reforms giving practical implementation to the concept of comprehensive income [see the Royal Commission on Taxation (1966), vol. III, part A]. The joint taxation system was in force in Italy for quite a time, both at central government level, through the Imposta complementare introduced in 1923, and the personal income tax (IRPEF) in its initial application, before the ruling no. 176/76 sanctioned its unconstitutionality (see paragraph 5), and at local level, through the Imposta di 4 Within the context of an individual-based tax system, they have an indirect influence as one needs to take into account how the taxpayer’s ability to pay is affected by its family situation. Insofar as the choice between an individual and a family basis is concerned, the arguments that may make a solution preferable to the other crucially depend on the choice of objectives assigned to taxation, dealing with a series of aspects (equity, efficiency and administrative costs). It should be pointed out that, as a general rule, it is neither adopting extreme solutions (individual or family taxation) nor selecting likely combinations of these two criteria that one may come to a solution that univocally prevails over all the others. Therefore, the problem cannot be solved without a proper arrangement of the virtually pursuable objectives. For instance, should the requirements of taxation include its neutrality with respect to the choice of marrying, it is obvious that while the choice of an individual basis reaches this object, the family solution does not. We get to the same conclusion when the requirement is stated in terms of the absence of any disincentive for either spouse to enter the labor market. On the other hand, if we promote the object of avoiding any discrimination of the overall tax burden of families with different individual income compositions, being the family income equal, it is obvious that the family taxation is to be preferred. Hence, it is unavoidable that a choice be made among likely objectives or, at least, their respective degrees of implementation. In any event, income taxation must definitely take the economic conditions of the family into consideration, at least -as previously pointed outon account of the fact that the latter affect the individual situation. This leads quite often to the adoption of systems that somehow mediate between the solutions founded on the individual and family basis. The most commonly used systems entail tax reliefs for dependents, splitting and family quotient4 methods. The tax credit for dependents granted to a single recipient is a method that is consistent with a definition of ability to pay on a strictly individual basis. It is the method adopted by our personal taxation system and, further to the suppression of the principle of income aggregation of spouses which characterized the initial phase of the famiglia. See the lucid remarks by C. Cosciani (1991, pp. 258-268) on the justifications of a joint taxation system. As a matter of pure curiosity, as proof of how natural it was to take as the tax reference the aggregated income of those living under the same roof rather than the individual sources of income, we should recall that in the original Meda reform project of 1919, the tax unit of the personal tax was extended to household members as a whole, “provided that they shared work and the fruition of assets, income and profits of any type whatsoever” rather than limiting it to the legal family. Such a formulation was abandoned in the De Stefani decree-law, where the concept of family was limited to the householder and “those persons who, being bound to the taxpayer by ties of relationship or affinity, are entitled to maintenance in pursuance of the provisions of the Civil Code and there is evidence that they exercise such a right”. As regards this subject, see L. Einaudi (1932, p. 145 onwards). 4 With the income splitting the sum of family incomes is imputed for tax purposes half to one spouse and half to the other, regardless of the share contributed by each spouse to the overall family income. The family quotient is a sort of per capita splitting revised for family needs. In fact, the formal rate scale is applied to the family income after being divided by a number (the quotient) which is equal to the sum of weights attributed to each family member. Once the basic tax is calculated, it is multiplied by the quotient itself in order to determine the total tax amount owing by the family. 5 tax application, is also the only tool for keeping into account the taxpayer’s family situation. It should be borne in mind that the tax reforms of the 1980s led in most cases to an abandonment of family taxation in favor of individual-based systems5. Even accepting the concept of ability to pay on an individual basis, it would seem that one of the limitations of this method, from the point of view of equity, is that it neglects the fact that the extent to which the presence of dependents limits the economic means of the individual taxpayer depends in practice on the economic conditions of the family as a whole. This becomes particularly evident when the spouses have limited means and the entire income from all the recipients is assigned to expenditures that are to guarantee the survival of the household members. The family quotient is not hindered by such a limitation. However, in assessing the tax according to an equivalence scale that keeps into account the needs of families having a different composition, this method is generally founded on the legitimacy of effecting a transition from a structure of individual preferences to one of familiar preferences that, from a methodological viewpoint, may appear rather questionable6. Furthermore, within the context of a personal conception of welfare and ability to pay, it is unquestionable that the most macroscopic limitation of the family quotient system is the fact that it entails tax cuts that may increase as the income of family members increases and, therefore, it seems more consistent with policies of demographic growth rather than with policies aimed to fight poverty. The recourse to equivalence scales would seem to be a suitable means for defining poverty thresholds depending on household composition. If the family income is below these values, the taxation should be annulled, regardless of its apportionment among individual recipients, as it is clear that -being this the caseeach individual income would be destined for the survival of the family. In other words, this would entail that the family should be entitled to a sort of fully tax-exempted basic income destined to guarantee the mere survival of its members. As a general rule, this exemption threshold should be guaranteed -regardless of the income level of the recipientsthrough a system of abatements at the tax base, the amount of which should be measured against the needs of the family which may vary in relation to the various family typologies7. 5 This was the case of Finland, Iceland, Great Britain, Belgium and Spain, in the end of the 1970s. The individual taxation principle is currently in force in over half of the OECD and European countries, with only a minority choosing the family as the tax unit (France, Luxembourg, Switzerland, Portugal and the United States) or allowing the possibility of opting for one or the other system (Germany, Ireland, and Norway). As regards this subject, see Messere (1993), pp. 248-162, and Oecd (1993). 6 See Longobardi, Patrizii (1993, pp. 195-199) for a review of the conditions that are necessary to insure that family welfare can be used as an indicator of the welfare of its members. See Patrizii, Rossi (1991, chapter 2) for a conceptual analysis of the equivalence scales as indicators of family welfare. 7 Having defined the family income threshold below which the tax exemption should be guaranteed and assuming that the definition of fiscal family is to apply only to married couples with minor children, the system of taxation on an individual basis could be implemented in practice as follows. Considering a one-earner family, the lump-sum exempt amount equal to the total basic income is deducted from the taxable base of the single recipient. Considering a twoearners family, the lump-sum exempt amount is divided in equal parts between husband and wife and, when filing 6 A more selective system might be founded on the logic of concentrating tax expenditures to a greater extent on individuals living in poor households, granting only partial tax cuts to those who are better off, within a context of a system acknowledging family maintenance costs which are decreasing as income increases. The choice of a more selective type of action may be warranted by the need to comply with strict public budget constraints without having to resort to exceedingly high increases in marginal tax rates which could give rise to disincentive effects on labour supply. This is the rationale of the proposal being examined which, by pursuing objectives that may be concretely attained, aims at avoiding any exceedingly extensive breach with respect to the current system. However, the pursuit of a redistribution in favour of needy families may make it advisable to resort to means other than tax reliefs. The recourse to other means, namely family allowances, can be justified by a number of reasons. Firstly, at moderate income levels, the extent of any relief granted through tax credits is limited by the amount of the tax itself which should be levied. Secondly, the tax cut does not bridge the likely gap between the taxpayer’s income and the basic income that may be deemed advisable to guarantee to the family. Thirdly, the extent of the redistribution attained by the tax system in favor of large families may be considered insufficient even at higher income levels, given the partial nature of the acknowledgment of family maintenance costs which decreases as income increases, in relation to the impossibility of fully burdening the public budget with these costs. Hence, it might be advisable to integrate the distributive effects attained through the tax system with more powerful family allowances. 3. The reform proposal The income tax reform being proposed is based on the assumptions and the purposes outlined below: a) to maintain the central position of IRPEF in pursuing redistributive objectives without any relevant revenue loss; b) to do away with the disincentive effects originating from an exceedingly complex structure of marginal rates; c) to extend tax credits, directing the resources especially toward the poor families; d) to integrate the redistributive role of the personal income tax with that one of the family allowance. In order to attain these objects, we suggest the following changes: 1. reduction in the number of brackets as well as the range of tax rates; their joint income returns, either one of the recipient is allowed to deduct, in addition to his/her own exempt amount, any amount in excess of the deductible amount of the other recipient with respect to her/his taxable income. 7 2. strengthening the role of tax credits to a considerable extent; 3. sizable appreciation of family allowances and their extension to self-employee. 1) reduction in the number of brackets According to our proposal, the structure should move from the current seven brackets to four brackets, as shown in table 1. Table 1 - Rate schedule in the current system and in the reform proposal current income brackets (million Lire) new income brackets (million Lire) current tax rates new tax rates difference 07,2 0-15 10% 20% + 10% 7,214,4 22% - 2% 14,4-15 27% - 7% 15-30 15-30 27% 27% -- 30-60 30-60 34% 35% + 1% 60-150 >60 41% 42% + 1% 150-300 46% - 4% >300 51% - 9% The minimum rate increase is founded on a dual consideration. On the one hand, the sizeable appreciation of the role of tax credits to discriminate taxpayers on the basis of the economic conditions of their families allows a suitable graduation of the actual marginal rates notwithstanding the presence of a minimum rate that is considerably higher than the one currently applied. On the other hand, it may prove useful for a coordination with the schedular systems to be applied to capital income, even in view of a tax harmonization with the systems in force in other European countries. The maximum rate reduction may limit disincentive effects, if any, on the labor market without any relevant revenue loss given the scarce density of the taxpayers to which such a reduction applies8. The considerable reduction in the difference between minimum and maximum rate is conducive to a potential reduction of the disparity of treatment of one-earner families with respect to two-earner families. In fact, according to our proposal, the maximum to minimum rate ratio drops from 5.1 to 2.1. 2) strengthening the role of tax credits With reference to the tax reliefs for dependents that were already provided by the 1995 tax system, regardless of income limits, it is our opinion that they should be maintained in the reformed system, even though they should be suitably appraised. 8 A representative sample of the 1993 income tax returns seems to point out that the density of taxpayers with a total taxable income in excess of 150 million Lire is merely 0.8 per cent of the total (Herr, 1995). 8 We are suggesting a slight revaluation of the tax credit for a dependent spouse with respect to the 1995 system, and a much more sizable appraisal of the tax credit for dependent children. The former should be raised from the current 818,000 to 1,000,000 Lire; the latter from 94,000 to 250,000 Lire. Notwithstanding the reduction in the rate scale with respect to the current one, the limited extent of the increase in the tax credit for a dependent spouse does not do away with the differences in tax treatment between one-earner and two-earner couples, but for moderate income levels or very slight tax base differences among husband and wife. In our opinion, a less onerous tax treatment for two-earners couple is justified by a few objective disadvantages that the latter has with respect to one-earner couple, as it bears costs for the production of the second income, as well as the relative work effort, which are not suitably recognized from the tax system. Besides, if the selected approach aims at discriminating in a right way the tax treatment of different family units, we are convinced that it is quite meaningless to differentiate two-earner from one-earner couples, but indeed one should differentiate within the context of one-earner couples whether the lack of a second source of income is occasioned by involuntary unemployment or the free choice of the individual. These are the reasons why we deem that the situation of one-earner couples should be protected, especially when it is characterized by a state of poverty, in line with the rationale of the introduction of further tax credits as outlined below. The increase in the tax credits for dependent children pursues the object of keeping into account -and not merely in a symbolic mannerthe costs for their maintenance. As previously pointed out, the proposed system substitutes for an ideal system of tax allowances to an extent equal to a minimum income that is the same for all taxpayers, as the latter would end up being too costly in terms of revenue loss. An appreciation in excess of 150% in the current tax credits is designed to bridge, in an however partial but significant manner, the gap with respect to that system. The method of tax credits is founded on the logic of granting implicit deductions from income subject to tax, which decrease as income increases, in compliance with a principle of social solidarity that legitimizes a progressive reduction in the tax relief granted to families as their economic means increase. Our proposal provides for a second type of tax credits which, indeed, characterizes it: additional reliefs should be introduced in support of poor families. This measure aims at acknowledging the right of indigent persons to adequate reliefs in order to avoid that taxation might cause them to fall below a poverty threshold, in line with the previously illustrated criterion of the basic income. Such additional tax credits should amount to 550,000 Lire per capita for taxpayers belonging to families with 1 or 2 members whose family income amounts to less than 10.9 and 19.5 million Lire, respectively. An additional tax credit is also awarded for dependent children in the amount of Lit. 300,000 per child if the family income is below the following thresholds (in million Lire): 26.1, 31.9, 37.2, 42, and 47 for families with 1, 2, 3, 4, 5 or more 15 Table 4 - Reform proposal: support for dependents and poverty gap in relation to the number of household members (one-earner households with householder in dependent employment) (thousand Lire) Family type Couple C. with 1 child C. with 2 children C. with 3 children C. with 4 children C. with 5 children poverty linea17498 23360 28557 33334 37621 42013 taxable income net income b poverty gap (1) + (2) poverty gap (1) + (2) poverty gap (1) + (2) poverty gap (1) + (2) poverty gap (1) + (2) poverty gap (1) + (2) 10000 9350 8148 2378 (29,2)c14010 3722 (26,6)c19207 5066 (26,4)c23984 6410 (26,7)c28271 7754 (27,4)c32663 9098 (27,8)c 20000 17000 498 3079 6360 4788 (75,3)c11557 6840 (59,2)c16334 8376 (51,3)c20621 9912 (48,1)c25013 11064 (44,2)c 30000 24300 -6802 1384 -940 3036 4257 5888 9034 8524 (94,3)c13321 11736 (88,1)c17713 12996 (73,4)c 40000 30800 -13302 1000 -7440 1500 -2243 2960 2534 4804 6821 10584 11213 12452 50000 37300 -19802 1000 -13940 1500 -8743 2000 -3966 2500 321 4920 4713 8966 60000 43800 -26302 1000 -20440 1500 -15243 2000 -10466 2500 -6179 3000 -1787 5420 (a) The poverty line for the various family types is calculated on the basis of the Carbonaro equivalence scale (b) Income net of tax and tax credit for work-related expenses, but exclusive of tax credits for dependents and family benefits (c) Degree of coverage of poverty gap (1) Tax credit for dependents (2) Family allowance 16 The income fraction that is exempt owing to tax credits decreases to 29% for couples without children which have a taxable income of 30 million Lire. In this income bracket, a total exemption may only be found among families with seven members (see table 2). However, the joint system of tax credits for dependents and family allowances is effective in restoring economic conditions over the poverty threshold for families with less than five members. With respect to all the families with at least three children, the value of the poverty gap coverage coefficient remains quite high, in excess of 94 and 88% for families with five and six members, respectively, while it drops to 73.4% for families with seven members, in respect of which the tax credits have indeed the effect of cancelling the taxation in full. In income brackets in excess of 40 million Lire, the tax system before tax credits for dependents leaves the taxpayers above the poverty threshold in case of families with up to four members. For larger families we find once again a virtual poverty gap, but it is more than offset by the system of tax credits and benefits. Assuming a taxable income that goes up to 50 million Lire, the same phenomenon appears in a more marked way, but only in respect of the last two family types taken into consideration (couples with at least 4 children). Close to a taxable income of 60 million Lire, family allowances have no role at all (to the exclusion of couples with 5 children) and tax credits have an increasingly slighter role in determining exempt income fractions, owing to both income growth and a corresponding increase in the marginal tax rates. In fact, for an income close to 100 million Lire the rate of tax base implicit abatement drops to a mere 5.6% with respect to couples without children and 11.5% with respect to families with seven members. These figures drop to 1.1 and 2.3%, respectively, in case of taxable income levels of around 500 million Lire (see table 2). The data shown in these three tables clearly stress how the new tax/benefit system, as suggested by the reform proposal, gives a particularly effective support to larger families in the lower income brackets, while this support becomes increasingly weaker as income grows, in line with the objects that the reform is striving to reach. In fact, while the reform aims at recognizing a considerable revaluation of the current monetary support in favour of all families, it does so with a logic of special effectiveness in the measures supporting needier ones. As pointed out at the beginning of this paragraph, we are advocating a system of tax reliefs for dependents that relates to an individual-based tax system and, therefore, aims at keeping into account the economic conditions of poorer families through arrangements that differ from those peculiar to family-based tax systems. However, the horizontal and vertical equity effects of the reform may be appreciated even in terms of those systems and in particular in terms of family quotient. Table 5 deals with the equivalence scales that are implicit in the IRPEF of the 1995 law and points to the weight ascribed by the legislator to such non-monetary characteristics as the household size with a view to determining the tax treatment of a 17 family with one earner working in dependent employment17. Table 5 - Implicit equivalence scales* in IRPEF (1995 system) for a one-earner family with householder in dependent employment income (million Lire) effective average tax rate single couple without children couple with 1 child couple with 2 children couple with 3 children couple with 4 children couple with 5 children 25 14,2 0,74 11,06 1,12 1,18 1,22 1,26 30 16,4 0,74 11,04 1,07 1,11 1,14 1,18 35 18,9 0,83 11,03 1,07 1,11 1,14 1,18 50 23,4 0,84 11,03 1,07 1,11 1,14 1,18 60 25,2 0,84 11,02 1,04 1,06 1,08 1,10 100 31,5 0,91 11,02 1,04 1,06 1,08 1,10 (*) Criterion of effective average tax rate parity. The family income on which the incidence is calculated is the income of the married couple with no children. The little consideration paid to the presence of dependent children in the determination of the incidence on households with different size is immediately evident. In fact, the tax system treats in much the same way (in terms of effective average tax rate) a couple without children with an income of 25 million Lire and a family, let’s say, with 3 dependents whose income is merely 18% higher than that of the couple with no children (therefore, equal to 29.5 million Lire) underestimating by far children maintenance costs. The equivalence scale adopted by the Commission of inquiry on poverty and deprivation is a useful term of comparison to see how the system in force underestimates family needs, at different income levels. Being the number of household members equal, the scale values (that remain unchanged as income varies) show a particularly marked difference with respect to the implicit IRPEF values (see table 6). Table 6 - Equivalence scale adopted by the Commission of inquiry on poverty and deprivation (Carbonaro scale) number of household members 1234567 coefficient 0,6 11,33 1,63 1,90 2,15 2,40 17 The implicit equivalence scale translates, in terms of monetary differences, a set of characteristics other than income which are ascribed an implicit relevance by the tax system in the assessment of the tax burden weighing on the different types of families. The calculation of the implicit scales in Table 5 is consistent with the horizontal equity criterion of the effective average tax rate parity (see Longobardi, Patrizii, 1993, pp. 202-210). The calculation method is as follows: given two families, A and B, with a different size but with the same monetary income, we have to estimate what would be the hypothetical income level of family A which would be subjected to the same effective average tax rate to which family B is subjected. 18 The system of additional tax credits for dependents that we are proposing is instead conceived in such a way that, for an equivalent family income equal to 22-25 million Lire of a childless couple (values that are just above the poverty threshold for this family type), the tax system reproduces an implicit equivalence scale quite similar to the Carbonaro scale (see table 7). Keeping to the previous numerical example, it may be noted that, further to the reform proposal, a family with three dependent children and an income of 29.5 million Lire would be burdened with a lower tax than that levied on a couple with no children and an income of 25 million Lire. An equal tax treatment in respect of the latter family type would be re-established at an income level for a 5-member family of 38.4 million Lire (see table 8). Table 7 - Implicit equivalence scales* in IRPEF reform proposal for a family with oneearner from dependent employment income (million Lire) effective average tax rate single couple without children couple with 1 child couple with 2 children couple with 3 children couple with 4 children couple with 5 children 21,5 11,19 0,71 11,23 1,50 1,74 1,96 2,19 25 13,40 0,71 11,15 1,33 1,54 1,73 1,93 30 15,67 0,71 11,09 1,17 1,31 1,48 1,65 35 18,43 0,80 11,09 1,17 1,26 1,34 1,47 40 20,50 0,83 11,09 1,17 1,26 1,34 1,43 50 23,40 0,83 11,09 1,17 1,24 1,29 1,34 60 25,34 0,83 11,05 1,10 1,15 1,20 1,25 100 32,00 0,90 11,05 1,10 1,15 1,20 1,25 (*) Criterion of effective average tax rate parity. The family income on which the incidence is calculated is the income of the married couple with no children. The horizontal tax equity taken as the reference criterion (horizontal equity meant as effective average tax rate parity) features another interesting property: the implicit equivalence scale that results from the application of the aforementioned criterion coincides in full with the coefficients of the hypothetical family quotient which would lead to the same effects of the system in force in terms of tax treatment18. Therefore, it should be noted that, although our reform proposal does not give up a definition of ability to pay on an individual base, it simulates a family quotient system with variable coefficients whose weights seem to be substantially close -with respect to low income levelsto those of the reference equivalence scale (Carbonaro scale). Quite naturally, having taken the road of lump-sum tax credits rather than tax credits growing with respect to income, the coefficients of the equivalence scales 18 In fact, the equivalence scale which is implicit in a family quotient system is given directly by the coefficients of the quotient itself, as the application of this method implies the same effective average tax rate for families with different size whose incomes are bound by a relationship given by the values of the quotient itself. 19 implicit in the reform tend to move off more and more, as income increases, from those of the reference scale, causing the tax system to move progressively away from the scheme inherent in a family quotient system with fixed coefficients19. In concrete terms, this amounts to a sort of “calibration” of the reform proposal around family income levels that are just above the poverty line, and it is in their respect that the horizontal equity is satisfactorily ensured. In the light of the above consideration, this falls within the objects that we had in view for the reformed tax system which considers the reference to the family income justifiable only in respect of low and medium-low income levels, while a tax system based on more strictly individual criteria should be preferred in respect of higher incomes20. 5. Income tax reform and constitutional law There are a number of constraints that need to be considered in order to ascertain the feasibility of implementing a reform of the personal income taxation in relation to the tax treatment of the family. Some of the most pressing definitely result from a largely consolidated constitutional law which expressed itself through three rulings, namely no. 179 of 1976, no. 76 of 1983 and, more recently, no. 358 of 1995. Starting from the first ruling, that did away with the joint taxation that had been provided for by the tax reform in pursuance of regulations introduced since 1958, the Court proposed a markedly individualist interpretation of the principle of ability to pay sanctioned by the first paragraph of article 53 of the Constitution. Even though the Court recognized that cohabitation has an influence upon the ability to pay of both spouses owing to the reduction of overhead costs, cooperation and mutual assistance, it questioned “even on account of the large variety of possible hypotheses and concrete situations, that in all instances this influence leads to an increase in the ability to pay of the two subjects taken together”. In different words, in the Court’s interpretation, the addressee of article 53 which lays down that “all the people have to contribute to public spending in relation to their ability to pay” is no one else but the individual, rather than the family. In the Court’s reasoning, the aggregation of incomes of the spouses is justified only if it were one of the spouses to have the income of the other at his/her disposal, and not also when the incomes were at their joint disposal, thereby denying any systematic relevance to the family as a decision-taking unit on such matters as consumption, savings, and the spending for the maintenance and education of children. 19 As income increases, the weight of tax credits decreases. This leads to a lessening of the difference in the IRPEF incidence, being the monetary income equal and, consequently, also to a lessening of the difference among equivalent incomes and the income of the family which is taken as the numeraire. 20 Finally, it should be noted that, notwithstanding the fact that the coefficients of the implicit equivalence scales move progressively closer as income increases, the values that are obtained with the reform in respect of incomes in the 60100 million range point to a scale that is less flattened than the scale of the 1995 system in respect of incomes of 30 millions. 20 What follows from this premise is that the aggregation of incomes of the spouses is inevitably prejudicial to both the ability to pay principle, in relation to the fact that the income earned by the wife is not necessarily at the disposal of the husband, and the equity principle. There is no doubt that the joint taxation determined distortions to the detriment of legal families and discouraged women from entering the labour market. Stressing how such a regulation was in conflict with the articles 29 and 31 of the Constitutions, the Court pointed to the need for a “moral and legal equality of spouses”, as well as the need to “facilitate through economic measures and other benefits the creation of families and the fulfillment of the relative tasks, with special regard to large families”. At the same time, the Court pointed to the risk of the taxation of one-earner families where “it is not only the husband who has the income at his disposal, as it is at the disposal of both spouses”. Besides, the Court expressed its hope for more suitable regulations governing that matter and called upon the legislator to implement a tax system “facilitating the creation and development of the family and considering the position of the woman as a housewife and a worker”. However, the attempt to give concrete implementation to the Court’s indications gives rise to problems that defy any easy solution just on account of the strictly individualist interpretation given by the Court to the concept of ability to pay. As a matter of fact, it is clear that it is indeed this interpretation that seriously hinders all those joint taxation systems, such as the family quotient and the income splitting, that presuppose a reference to a family-based concept of ability to pay. The issue of the fiscal discrimination of one-earner and large families was brought up again in some recent judgments. The Court reaffirmed with even greater emphasis the need to find a solution to this problem, even though it could not allow the objections of unconstitutionality raised as regards this subject just on account of the fact that ruling no. 179 had done away with the legal presuppositions for their recognition. In any event, we have no intention to enter into the merits of constitutional law on the matter of the tax treatment of the family21. What interest us is more simply to verify whether our reform proposal, founded on the criteria described in paragraph 2, is susceptible of objections as to its constitutionality. Firstly, the concept of ability to pay is defined on a strictly individual basis, as it is at the level of individual incomes that both the rate scale and the tax credits apply. The family income is only relevant for defining that individual condition of economic distress to which the application of the system of additional tax credits is subordinated. As regards this subject, it may be noted that the conditions of economic distress of a household fall within those instances where the bonds of solidarity among spouses for the maintenance of their children make it absolutely reasonable to prospect an 21 On this subject, see the lucid remarks made by F. Gallo (1977) and E. De Mita (1976) on the ruling no. 179/76. As for the recent ruling no. 358 of July 1995, De Mita’s comments (1995a, 1995b) appear extremely convincing. 21 economic unity of the family. Besides, such conditions seem relevant for the sole purpose of allowing additional tax reliefs that are granted on an individual basis. Therefore, this measure may indeed be viewed as one of the possible concrete ways to safeguard the development of the family with reference to article 31. Secondly, the limited increase in the tax credit for the dependent spouse and the flattening out of the rate scale objectively appear to be re-equalizing measures in favour of one-earner families and this fact was stressed, even quite recently, by the constitutional Court notwithstanding the fact that its interpretation on a strictly individual basis of the concept of ability to pay makes a comparison between oneearner and two-earner families quite meaningless, being family income equal. In any event, the additional tax credits (for dependent children) anchored to a condition of family poverty operate in an indirect manner mostly in defense of one-earner families as the presence of a single income makes it more likely for a situation of poverty to occur and this is the condition for granting reliefs. Finally, the sizable appreciation of the tax credits for dependents, which was made even greater in respect of poor families, aims at reducing if not even at cancelling the disparities of treatment to the detriment of large families. From all these arguments, it follows that while the reform being proposed is not at all in conflict with the complex bonds of the constitutional law, it is indeed likely to afford the opportunity of putting into effect the suggestions voiced by the Court. 6. Redistributive effects of the reform The IRPEF reform proposal has been reviewed by means of the static microsimulation model DIRIMOD. The model uses the sample data of the Bank of Italy’s Survey on the Budgets of the Italian Families and allows to analyze the impact on revenue and on the personal income distribution of both the 1995 system and the hypothetical reforms concerning IRPEF, social contributions and family allowances. The distributive analysis was performed at the level of both the individual income recipient and the household, after having rearranged the latter by growing levels of equivalent income (Carbonaro equivalence scale). The outcome of these estimates are shown in the tables below, which we will comment later on. Table 9 - Proposed IRPEF reform vs. the 1995 system: effects on revenue, progressivity and personal income distribution (% values) - analysis by income recipient 1995 system reform proposal revenue lossa-- 3,3 redistributionb3,19 3,07 incidencec15,67 15,34 progressivityd17,39 17,42 22 redistributione3,23 3,16 a Thousands of billions Lire b Difference in the Gini index between gross and net incomes c IRPEF incidence on the gross income d Difference in the index of concentration between tax liabilities and gross incomes e Difference in the index of concentration between gross and net incomes. Table 10 - Proposed IRPEF reform vs. the 1995 system: effects on revenue, progressivity and personal income distribution (% values) - analysis by household 1995 system reform proposal revenue lossa-- 3,3 redistributionb2,76 3,12 incidencec15,67 15,34 progressivityd15,32 17,56 redistributione2,85 3,18 a Thousands of billions Lire b Difference in the Gini index between gross and net incomes c IRPEF incidence on the gross family income d Difference in the index of concentration between tax liabilities and gross incomes e Difference in the index of concentration between gross and net incomes. Tables 9 and 10 clearly show that the reform entails a slight reduction of the incidence that drops from 15.7% to 15.3% of the gross income, with a consequent moderate revenue loss estimated at 3,300 billion Lire, equal to nearly 2% of the overall revenue for 1995. Table 9 shows that the redistributive effects at the level of income recipient, measurable through the traditional Kakwani indexes, appear to be substantially unchanged with respect to the 1995 system. On the other hand, the analysis by households points to an increase in the Kakwani redistribution index, notwithstanding the reduction in the incidence, due to an upsurge of the progressivity that increased by over 2.2 percentage points. The accentuation of the reform redistributive effects is confirmed by the variation of the differences in the Gini indexes of gross and net income, which moves from 2.76 to 3.1222. Table 10 bis points to an increase in both the progressivity and the redistributive effects of the reformed system which is particularly marked if one takes into consideration also the changes in family allowance and social contributions that are to guarantee its coverage. 22 The redistributive effect measured by the difference in the Gini coefficients of income before and after tax includes, with respect to the better known Kakwani index, the reranking effect produced by IRPEF through the system of tax credits which, as it depends also on non-monetary characteristics, may cause a different ranking in the income scale in the passage from before-tax income to after-tax income. 23 The variation of the social contribution rates affecting the administrations in charge of dependent employment and self-employment is 0.55% and 0.67%, respectively23 (table 11). On the basis of some preliminary estimates, the revaluation of the family allowance and its extension to self-employed workers would entail a total greater spending of nearly 5,900 billion Lire, broken down as follows: 3,900 billions in favour of families that are already granted the allowance and 2,000 billions to the families of self-employed workers. These two spending increases should be financed through a separate management of contributions. Table 10 bis - Reform proposal (IRPEF and family allowance) vs. the 1995 system: effects on revenue, progressivity and personal income distribution (% values) - analysis by household 1995 system reform proposal revenue lossa-- 3,3 redistributionb4,24 5,16 incidencec22,00 21,66 progressivityd15,96 19,60 redistributione4,50 5,42 a Thousands of billions Lire b Difference in the Gini index between gross and disposable incomes c Incidence of IRPEF and social contributions (net of family allowance) on gross income d Difference in the index of concentration between total revenue (IRPEF and social contributions, net of family allowance) and gross incomes e Difference in the index of concentration between gross and disposable incomes. Table 11 - Increase in family allowance spending and contribution rates with respect to independent workers and workers in dependent employment (money value in thousand billions) - analysis by income recipient increase in family allowance spending and social contributions absolute variation of the contribution rate % spending variation with respect to the 1995 system Workers in dependent employment 3,900 +0,55 +49,1 Independent workers 2,000 +0,67 (*) Total 5,900 +0,58 +75,4 (*) As the initial spending amount is null, the variation is a value tending to infinity. With reference to the variations of the incidence by decile, the analysis by 23 The outcome of the estimate of these rates should be taken with caution, as DIRIMOD does not consider any imputation procedure to allow for social contributions evasion, even though it does re-proportion the figures estimated on the basis of the sample of the Bank of Italy with respect to the statistics of the Ministry of Finance to keep indirectly into account the tax evasion problem. 24 income recipients (table 12) points to an incidence increase which is approximately equal to 2 points in the first decile and 1 point in the second and third deciles, a substantial invariance in the fourth and last decile and a reduction in the other five deciles that allows to assume an initially growing trend up to the seventh decile and then a decreasing trend until the ninth decile. The increased incidence in the second and third deciles may be explained by the considerable increase in the marginal tax rates as well as by the suppression of the additional tax credits for both dependent employment and self-employment, which are replaced in the reformed system by additional tax credits referred to the economic conditions of the family. The sizable increase in the first decile should not be a surprise as the analysis of sample data allows to note that this decile is formed in excess of 90% by persons who are not employed and, therefore, may be referred for the most part to the recipients of rent income, who are not entitled to allowances for work-related expenses, as the latter only concern workers in dependent employment and pensioners and their amount is readjusted to the new rate scale. Table 12 - Reform proposal vs. 1995 system: IRPEF incidence* by income recipient** deciles and absolute variation with respect to the law in force deciles mean decile income (million Lire) 1995 system reform proposal absolute incidence variation 13,398 1,05 3,08 +2,03 28,817 1,21 2,28 +1,07 313,175 3,97 4,94 +0,97 418,520 9,72 9,65 -0,07 523,676 12,46 11,57 -0,89 627,801 13,81 12,88 -0,93 731,972 14,64 13,58 -1,06 837,131 15,85 15,02 -0,83 945,239 17,64 17,25 -0,39 10 77,297 22,21 22,29 +0,08 average 28,707 15,67 15,34 -0,33 (*) Incidence on gross income (**) Income recipients ranked by gross income level. On the contrary, the analysis by households points to a satisfactory redistribution of the incidence by decile of the reform (table 13). In fact, notwithstanding the increase in the marginal rate applied to incomes in the first bracket, the most relevant tax cuts are indeed concentrated in the lower income levels. These reliefs are in excess of 3 percentage points for households whose income is below 26,268 million Lire and in excess of 1 point for households whose income is included between that amount and 30,317 millions. Besides, the profile of the 31 Table 16 bis - Reform of IRPEF and family allowance: overall incidence* by household** deciles and number of household members, and absolute variation with respect to the 1995 system. deciles 1variation 2variation 3variation 4variation 5variation 6variation 7variation 11,80 -0,29 -6,72 -3,48 -4,80 -5,63 -3,89 -10,13 -8,95 -11,49 -15,13 -12,14 -7,13 -13,69 25,11 +0,55 -1,85 -3,04 8,18 -4,68 7,89 -7,05 6,16 -9,08 4,88 -6,54 3,96 -7,14 37,75 +1,45 4,20 -1,46 14,06 -2,31 15,44 -3,25 13,64 -5,21 13,63 -2,75 13,12 -6,37 411,43 +0,89 10,62 -0,36 16,12 -0,90 19,23 -1,26 19,08 -1,57 17,44 -1,12 17,33 -3,12 515,01 +0,75 14,14 +0,27 18,54 -0,09 21,89 -0,58 20,67 -0,78 16,42 -0,60 17,65 -0,52 618,78 +0,64 17,48 +0,70 21,77 +0,35 22,26 -0,21 20,49 -0,27 21,33 +0,04 20,52 +0,40 719,87 +0,68 20,35 +0,85 21,53 +0,28 23,99 +0,36 23,64 +0,19 21,93 +0,41 18,96 -0,50 820,36 +0,49 22,89 +0,82 23,09 +0,67 25,88 +0,44 24,89 +0,61 23,60 +0,41 23,83 -0,24 924,00 +0,72 23,81 +0,71 25,53 +0,64 26,55 +0,58 29,86 +0,43 27,96 +0,71 20,73 +0,41 10 31,72 +0,49 28,33 +0,96 30,13 +1,02 30,33 +0,84 32,03 +0,20 29,41 +0,52 22,42 +0,52 mean 21,96 +0,61 19,33 +0,30 22,90 +0,13 22,74 -0,74 21,39 -1,49 17,76 -1,57 12,58 -4,02 (*) Incidence of IRPEF and social contributions, net after the family allowance, on gross family income. (**) Households ranked by gross equivalent income - Carbonaro scale. 29 Furthermore, the increase in the net incidence is regressive as the number of household members increases with respect to families of up to three members. Larger families allow to note a decrease in the tax burden which is positively correlated with the number of members. Tax increases of 0.6, 0.3, and 0.1% may be observed for singles, and households with two and three members, respectively, while incidence reductions may be observed with respect to households from four members upwards, which get up to four points for households with seven members. The tax cuts in the first deciles are more marked and the distribution of the net incidence variations per decile is very regular with respect to the 1995 system, with a clear regressive trend as income grows for all the families with at least two members. Tables 17 and 17-bis show the incidence of both IRPEF and the tax/benefit system, respectively, by one-earner and two-earner family deciles, and the variations with respect to the current legislation. Both household types, whether characterized by one or two income recipients, show the usual trend of the incidence variation which is regressive as income grows until they turn into tax increases for higher income levels. It is particularly interesting to note that tax cuts are on the average much higher for one-earner families when compared with two-earner families, with a considerable increase of the incidence differential between the two types when the effects of the family allowance are taken into consideration. Table 17 - IRPEF reform proposal: mean* incidence by household** deciles, considering oneearner and two-earner households. deciles one-earner households variation with respect to the 1995 system two-earner households variation with respect to the 1995 system 12,12 -4,06 2,00 -2,03 26,86 -3,89 4,52 -3,00 311,21 -1,23 8,05 -1,35 413,56 -0,77 11,41 -0,49 515,12 -0,40 12,89 -0,32 616,26 -0,02 14,19 -0,24 716,16 +0,05 15,33 -0,21 818,01 +0,11 16,82 -0,10 919,50 +0,14 18,23 -0,04 10 23,82 -0,05 21,60 +0,17 mean 14,91 -1,04 15,61 -0,34 (*) IRPEF on gross income (**) Households ranked by gross equivalent income - Carbonaro scale. In fact, while the 1995 system provides for an identical mean IRPEF rate of 15.95% for both household types, and an aggregate tax/benefit rate of 22.83 and 22.54% for one-earner and two-earner families, respectively, the new system would 30 provide for absolute reductions in both the incidence of IRPEF (-1.04%) and the aggregate incidence (-1.50%) that are much more sizable for the one-earner families than for the two-earner ones. This is a further proof that the system of additional tax credits, related to the family income and supplemented by a revaluation of the family allowance, is an effective means to reduce the taxation of one-earner families as the presence of a single income recipient increases the probability of living in a poor family. The extent of tax reliefs is considerably bigger in the first deciles by the redistributive effect of the family allowance. Table 17 bis - Proposal for the reform of IRPEF and family allowance: overall mean incidence* by household** deciles, considering one-earner and two-earner households deciles one-earner families variation with respect to the 1995 system two-earner families variation with respect to the 1995 system 1-4,89 -8,28 -8,03 -8,13 27,93 -6,01 3,33 -5,66 314,86 -2,14 9,09 -3,03 418,51 -1,07 15,49 -0,86 521,29 -0,26 19,16 -0,24 623,58 +0,50 21,51 +0,18 722,80 +0,63 23,59 +0,33 825,53 +0,68 24,98 +0,46 927,39 +0,72 26,65 +0,53 10 32,59 +0,53 30,29 +0,74 mean 21,33 -1,50 22,26 -0,28 (*) Incidence of IRPEF and social contributions, net of family allowance, on gross income (**) Families ranked by gross equivalent income - Carbonaro scale. 7. Conclusions The proposal to reform the personal income taxation, as outlined in this article (paragraphs 1, 2, and 3), focuses on both a reduction of the higher marginal rates and a simplification of the rate scale, but also and especially on a thorough review of the system of tax credits for dependents. In fact, it aims at getting to a fair tax treatment of the family, through the exemption of a sort of basic family income which decreases as income increases. In addition to ensuring the full exemption of the less well-to-do taxpayers, this system succeeds in determining effects in respect of moderate income levels that are not dissimilar from those that may be attained with the family quotient system (paragraph 4). The system of additional tax credits, subordinated to family income thresholds, 31 is an attempt to match the need to keep adequately into account the economic conditions of the family with the need to avoid the abandonment of the definition of ability to pay on individual basis, even with a view to getting over any objection raised in the matter of the tax constitutionality (paragraph 5). Besides, it was deemed advisable to see to a closer integration between the tax treatment of the family and the institute of the family allowance, providing for both a sizable revaluation of its amounts and the extension of its benefits to self-employed workers in order to avoid that the application of what may turn into an effective instrument to reduce the poverty gap might be limited to a single category of taxpayers. The analysis of the effects of the reform at a family level (paragraph 6) points indeed to a slight revenue loss, but also to a significant increase in both the progressivity and the redistributive impact of the tax. These effects result to be even stronger if we keep into account the effectiveness of the reformed system of family allowances whose increases are financed at a balance. The re-equalizing effect of the reform takes concrete form in an apportionment by family deciles of the incidence variation of both IRPEF and the aggregate tax/benefit system that implies a mitigated fiscal burden on the lower deciles countered by slight increases in the higher ones. The same type of effects may also be found, notwithstanding a few irregularities, at the level of the individual categories of income recipients. Special significance should be ascribed to the redistributive effects of the reform in the analysis that keeps into account the disaggregation by deciles as well as by household size. As a matter of fact, it points to a considerable increase -with respect to the 1995 systemof the tax reliefs provided for by the reform as the family becomes larger, as well as to decreasing incidence variations as income increases, leading to increases in the tax pressure on the deciles that are further away from the first. Even the analysis by number of income recipients within the family clearly points to higher tax reliefs for one-earner families with respect to two-earner families, with a definite reversal of trend which appears particularly evident if the effects of the family allowance are taken into consideration. The analysis of the redistribution impact of the reform proposal allows to conclude that, with respect to the current system, the reform is moving towards a considerable equalization of incomes as it removes a series of limitations that characterize the current regime. Even though we are convinced that the reform being proposed is likely to mark a significant step forward in reducing a few significant lacks of the current system, there is no denying that the ability of income assessment by the tax administration is still affected by dramatic limitations. Not only is the reform unable to remove those limitations, but it may even make them more evident, notwithstanding the fact that it confines itself to suggesting solutions that may be realistically and immediately put into practice. 32 In particular, the redistribution of the fiscal burden among households on the basis of a combination of tax credits and family benefits, and the extension of the latter to self-employed workers within the context of an administration kept fully apart from that of workers in dependent employment, strengthens the need to find a remedy for the problem of fiscal evasion in order to avoid any effect that is likely to multiply tax injustice. Until a proper solution to this problem is found, no easy optimism may be voiced with respect to the re-equalizing capacities of any reform. The problem of fiscal evasion needs to be dealt with and solved rather than being eluded by simply ignoring it or legalizing it through arbitrary systems of relief. Indeed, we are fully convinced that the actual difficulties in finding a solution to this problem should not be taken as a pretext for giving up an effective redistributive role of the tax system. 33 Bibliography Bernardi, L. 1989 “Per un’introduzione al dibattito sul sistema tributario italiano e alle proposte di riforma”. In La questione tributaria. Edited by Pedone, A. Bologna: Il Mulino. 1995 “Studi per un progetto di riforma del sistema tributario italiano: rapporto Irpef”. Edited by Bernardi, L. In Rivista di diritto finanziario e scienza delle finanze, no. 3, pp. 436-590. Cnossen, S., Messere, K.C. 1990 “Personal Income Tax Reforms in Oecd Countries”. In The Personal Income Tax: Phoenix from the Ashes? Edited by Cnossen, S. and Bird, R. Amsterdam: North Holland. Commissione di indagine sulla povertà e l’emarginazione. 1995 Verso una politica di lotta alla povertà. L’assegno per i figli e il minimo vitale. Rome, Prime Minister’s Office, Department for information and publishing. Cosciani, C. 1991 Scienza delle Finanze. Turin: Utet. De Mita, E. 1976 “La illegittimità costituzionale del c.d. “cumulo”. In Diritto e pratica tributaria, no. 2, pp. 337-340. 1995a “Famiglie monoreddito: l’illegittimità impossibile”. In Il Sole-24 ore, September 2. 1995b “Il trattamento fiscale dei redditi familiari: la sentenza della Corte costituzionale. Il Commento”. In Scuola Centrale Tributaria. Supplement to no. 9 of the tax news report, pp. 81-86. Einaudi, L. 1932 Principii di Scienza della finanza. Turin: La riforma sociale. Franco, D., Sartor, N. 1990 Stato e famiglia. Obiettivi e strumenti del sostegno pubblico dei carichi familiari. Milan: Franco Angeli. 1994 “Il sostegno pubblico alla famiglia tra nuove povertà e declino demografico”. In Economia & lavoro, no. 1, pp. 15-32. Gallo, F. 1977 “Regime fiscale della famiglia e principio di capacità contributiva”. In Rivista di 34 diritto finanziario e scienza delle finanze, no. 1, pp. 92-104. Herr, U. 1995 “Evoluzione ed aspetti quantitativi attuali dell’Irpef. In Bernardi, L., editor, pp. 485-506. Longobardi, E., Patrizii, V. 1993 “La tassazione dei redditi familiari”. In La crescita ineguale. 1981 - 1991. Edited by Rossi, N. Bologna: Il Mulino. Mantovani, D., Toso, S. 1996 L’assegno per i figli e il minimo vitale: una valutazione delle proposte della Commissione di indagine sulla povertà e sull’emarginazione. Prometeia. Memorandum 9601. Bologna. McLure, C.E.Jr., Zodrow, G.R. 1993 “The Study and Practice of Income tax Policy”. In Modern Public Finance. Edited by Quigley, J. And Smolensky, E. Harvard: Harvard University Press. Messere, K. 1993 Tax Policy in Oecd Countries. Amsterdam: Ibdf. Oecd 1993 Taxation in Oecd Countries. Paris: Oecd Publications. Patrizii, V., Rossi, N. 1991 Preferenze, prezzi relativi e redistribuzione. Bologna: Il Mulino. Royal Commission on Taxation 1966 Ottawa: Queen’s Printer. Sandford, C. 1993 Successful Tax Reform. Lessons from an Analysis of Tax Reform in Six Countries. Bath: Fiscal Publications. Toso, S. 1995 “L’Irpef: dinamica, interventi e proposte di riforma”. In La Finanza pubblica italiana. Rapporto 1995. Edited by Bernardi, L. Bologna: Il Mulino. Visco, V. 1992 “Fiscal System and Fiscal Reform in Italy in the 90’s”. In Design and Reform of Taxation Policy. Edited by Galeotti, G. and Marelli, M. Dordrecht: Kluwer 35 Academic Publishers Visco, V. et al. 1988 Proposta di Legge n. 2991, X Legislatura, introduced in the Chamber of Deputies on June 26, 1988.