Foreword
From Carneades’ shipwrecked sailor, who saves himself by shoving someone else from a plank,Footnote 1 to H.L.A. Hart’s park where no vehicles are allowed,Footnote 2 fictional cases have allowed generations of legal scholars to show the particulars and the practical implications of their theories. Looking back at this tradition, this paper discusses an imaginary case brought by a destitute person against the Italian Republic, alleging that her constitutional right to social assistance has been violated.
The recourse to fiction is in part inevitable, for only rarely have indigents had sufficient resources to bring a case to the attention of the Italian Constitutional Court. When they have, their pleas have usually been narrowly framed.Footnote 3 In addition, the fictional scenario which I imagine here pressures the law along lines which are more closely related to the main question which I intend to answer: is the lack of a universal means-tested scheme against absolute poverty unconstitutional? If this is the case, which branch of government should amend the flaw?
Instead of discussing the case from my own viewpoint, I adopt the perspective of one of the Justices on the bench – a choice reinforced by the use of the first person throughout the piece. While unorthodox in legal academia, this has also been the choice of Lon FullerFootnote 4 and Duncan KennedyFootnote 5 when they dealt with adjudication. The advantage of this narrative expedient is to bring the reader closer to the stage, and to let her look at judicial decision-making through the protagonists’ eyes.
Unlike Fuller and Kennedy, however, I combine the use of the judge’s point of view with a ‘bad-man’ characterisation.Footnote 6 My judge, in other words, is an anti-hero. To be sure, judges like to think of themselves as ‘good judge[s]’.Footnote 7 But, however implausible or subliminal, the bad-man characterisation allows me to achieve two results: first, it creates a narrative distance between my character and me – a distance which I use to denounce the judge’s manipulating approach to the law as both professionally censurable and ultimately detrimental to labour-market outsiders;Footnote 8 second – and here I am looking at Oliver Wendell Holmes – a bad-man perspective allows for a test of the degree of resilience and manipulability of legal institutions.
The bricks of this paper are Italian, but the building may be of interest to the foreign observer. Some of the dynamics of constitutional adjudication discussed in the article are common to other jurisdictions. In addition, occasional reference is made throughout the piece to foreign sources. Finally, the study of Italian exceptionalism in matters of social assistance, as compared with other European countries,Footnote 9 can help scholars better understand the role of constitutionalised social rights, if any, in the historical evolution of welfare regimes.
This article is organised into an Introduction, which briefly illustrates the case to be discussed, and three main parts. The first part discusses the choice of rejecting the challenge as procedurally flawed, while the second part assesses the grounds for rejecting the challenge on the merits. The final part discusses the possibility that the Court endorses the challenge and issues a mandate to the Government.
Introduction
Sometimes, on evenings like this, when most of the employees have left, and the Palazzo della Consulta remains dark and silent, I like to have a solitary walk across the empty corridors. In pensive and slow steps I wander in the dormant rooms,Footnote 10 hands folded behind my back. Occasionally, I indulge in the view through one of the tall windows, and I let my gaze fly above the roofs of Rome.
Today, however, I force myself to remain in my office. I strive to concentrate on the briefs, but the alluring embrace of my large leather armchair conspires against my efforts and stimulates my reveries. Tomorrow’s case will not be an easy one.
My name is B.J., and I am one of the 15 Justices sitting on the Italian Constitutional Court. We will meet tomorrow for a public hearing, and the President chose me as rapporteur in one of the cases to be discussed.Footnote 11 These are the facts. A certain Federica Carmini, an unemployed young woman with a two-year-old daughter, recently applied for the only maternity benefit available to her, the so-called Assegno di Maternità del Comune. Footnote 12 Given the limited length (five months) and amount of the benefit (€338 per month) – inadequate, if compared with the cost of basic necessities in the city where she dwellsFootnote 13 – Ms Carmini soon ran out of money. She fortuitously found shelter in a facility managed by a charitable organisation. There, Ms Carmini was able to secure pro bono legal assistance from Ms Leire Pancrazio, and sued the State.
Ms Pancrazio argued that the State had failed fully to implement the ‘right to maintenance and social assistance’, which Article 38 of the Italian Constitution grants to ‘every citizen unable to work and devoid of the means which are necessary to live’. According to Ms Pancrazio, this provision mandates the Republic to enact a universally accessible (albeit need-based and conditional) scheme against absolute poverty, that is, a scheme to the use of which any permanent resident should be entitled, provided that she finds herself in a condition of extreme need which she cannot overcome on her own.
The promises of the Constitution, however, remained on paper. In place of a universal measure against absolute poverty, an incoherent mass of sectorial social assistance benefits has been introduced since the late 1960s.Footnote 14 In a welfare system already marked by a high level of fragmentation and inequality – legacy of workers’ friendly societies,Footnote 15 fascist corporatismFootnote 16 and post-war clientelismFootnote 17 – these selective social assistance programs had the effect of perpetuating the disparities between categories of needy people.
More specifically, Ms Pancrazio, and the trial judge who endorsed her claims and raised the question before the Constitutional Court, put forward four distinct challenges, with the hope that at least one would make its way through the strict standing requirements.Footnote 18 The following norms were challenged: (i) the statutory rule that limits the duration of the maternity benefit to a maximum of five months, thus failing to guarantee a continuous flow of income for the entire duration of involuntary indigence;Footnote 19 (ii) the statutory rules that set the amount of the maternity benefit at a nationwide flat-rate level, failing to take account of the specific family situation, actual needs, and place of residence of the beneficiary;Footnote 20 (iii) the statutory rule which links unemployment insurance (Nuova prestazione di Assicurazione Sociale per l’Impiego, NASpI) to a certain status-related condition, namely previous employment, thus unreasonably discriminating within the group of indigent people;Footnote 21 and, finally, (iv) the entire corpus of Italian social assistance legislation,Footnote 22 for failing to establish a general, means-tested scheme against absolute poverty.
In all cases, the constitutional rules allegedly violated are either Article 38 (right to social assistanceFootnote 23 ), Article 3 (equal dignity, factual and formal equality, actual freedom, full development of the human personFootnote 24 ) or both. In addition, Ms Pancrazio infers from these and other clauses of the Constitution, as well as from the works of the Constitutional Assembly, a general commitment to the principle of universality in the provision of social services.
I am concerned about Ms Pancrazio’s performance at tomorrow’s hearing.Footnote 25 I’ve always known her as the type of person who does things in unconventional ways. Her eloquence is inspiring; her line of thought clear and logical. I know I will have to work hard to refute her points. It is true that the decisive forum will not be the hearing per se, but the discussion which my colleagues and I will have behind closed doors soon afterwards. But the more persuasive Ms Pancrazio is tomorrow, the harder it will be for me to repair the damage.
This evening I want to prepare a few strategies to deal with the case.Footnote 26 And since I know I can count on your silence, my friend, I will reflect out loud.
Rejecting the challenge as procedurally flawed
The Court can decide the case in several ways, which can be clustered into three broad groups. Specifically, the Court can: (i) declare the case inadmissible on procedural grounds; (ii) hear the case, but reject the constitutional challenge; or, finally, (iii) hear the case and accept part or all of the challenges raised by Ms Pancrazio.
The first option is to declare the case inadmissible on procedural grounds. This is my primary preference: by denying the appeal altogether, we leave the legal framework (almost) unaltered, while simultaneously sparing the Court the critique of public opinion.Footnote 27 Unlike our American peers, however, Italian Justices have no discretion over the acceptance of cases. As a general rule, the Italian Constitutional Court must accept and hear all cases, except in cases of procedural violations. But what exactly counts as a procedural flaw? The law says that the plaintiff’s failure to specify ‘the rule of law [...] flawed by unconstitutionality’ is a legitimate ground to declare a case inadmissible [my emphasis].Footnote 28 Perhaps induced by the definite article (‘the’), the Court has interpreted this rule as requiring that the challenge is addressed against a specific flaw in a specific rule of law.Footnote 29
The fourth of Ms Pancrazio’s challenges seems particularly vulnerable to this procedural requirement.Footnote 30 That challenge is not addressed to a specific rule of law but, rather, to the Italian social assistance legislation as a whole. To be sure, Ms Pancrazio was in part compelled to formulate this last claim of hers in generic terms, because the failure to introduce a universal means-tested scheme against poverty can be hardly attributed to any particular statute or article. In addition, if it is legitimate to challenge a single rule of law in front of the Court, a fortiori there must be a procedurally-permissible way to scrutinise a wider and more severe constitutional violation. But, however logical this argument, the Court is bound by its rules of procedure. If the fourth of Ms Pancrazio’s claims is to stand, then, one of these two options must be followed: either the Court overturns its restrictive reading of the rules of procedure, thus making general challenges admissible by way of interpretation; or, alternatively, the Court could call into question the reasonableness of its procedural rules by raising a constitutional challenge to itself.Footnote 31 Both solutions, however, would involve the Court recanting its previous approach to standing requirements. Some of the Justices could perceive such a shift as a potential source of embarrassment, and would therefore be hesitant to take either of these two paths.
The traditional restrictive approach to standing requirements, however, is surrounded by controversies. Support for the admissibility thesis, for example, can be found in the post-war writings of the Italian jurist (and soon-to-be Justice) Vezio Crisafulli. In his seminal 1952 book on constitutional rules and principles, in particular, Crisafulli argued that, as a rule, only government actions could be challenged in courts. But, just after making this general claim, Crisafulli conceded that, in ‘very rare’Footnote 32 cases, legislative inertia could be challenged, too. In order to bring omissions to the attention of the Court, Crisafulli continued, it would be sufficient to address the challenge to existing statutes, as long as they impinge on the same policy area as the alleged omission.Footnote 33 Clearly, Crisafulli considered this strategy to be procedurally permissible, or he would not have suggested it. Crisafulli’s remarks, however, remained unheeded, as the Court gradually embraced a restrictive stance that excludes the admissibility of challenges raised against omissions or bundles of statutes.Footnote 34 By exploiting this tradition, then, I can try to defeat on procedural grounds the fourth and most ambitious of Ms Pancrazio’s claims.
There remain three other challenges, which attack the duration, amount, and eligibility requirements of existing benefits. The precedents of the Court on standing requirements seem to offer some material to neutralise these challenges without even examining the merits. Arguing in terms of separation of powers, the Court has frequently felt legitimised to declare a case inadmissible whenever some suggestion of legislative discretion is at stake.Footnote 35 Yet, the power of drawing the line between legislative discretion and constitutional boundaries is a delicate one: as such, the Court should not engage in this sort of enquiry hastily, during the preliminary check of admissibility, but scrupulously, during the discussion of the merits. Despite this powerful objection, however, the Court has shown over time a certain propensity to remain silentFootnote 36 : this consolidated practice will help me achieve my conservative ends.
I can see from your ‘frowning brows’Footnote 37 that you are sceptical of my ability to convince my colleagues to refuse to hear the case. You argue that more than 60 per cent of Italian residents now favour the introduction of a general social assistance scheme against poverty.Footnote 38 My colleagues and I, you say, will eventually bend to social expectations and will declare the case procedurally admissible, because – and here you quote US Justice Harry Blackmun – ‘a Court’s consciousness is necessarily the product of prevailing public opinion’.Footnote 39 These observations of yours are generally correct, and I have no reason to see this case as an exception. Your prediction, however, is based on imprecise assumptions. First, while most Italians do care about the Government’s social policies, cash benefits for the poor form a subset of policies that receives far less attention than, say, those impinging on old-age pensions. Second, among major Italian institutions, the Court is the one that attracts less media coverage, hence less attention on the part of the general public.Footnote 40 Lastly, opinion polls show that one out of two Italians is sceptical about the feasibility of introducing a general social assistance schemeFootnote 41 : we like to dream of an utopian society without poverty, but our daily attitude towards political institutions resembles a sort of fatalist indolence.Footnote 42 For these reasons, I do not expect any particular reaction from the public if the right to subsistence remains unenforced.
Another solid ground for a declaration of inadmissibility would be a federalist argument. I can pretend to believe, in particular, that the plaintiff mistakenly sued the national, rather than the regional, Government. To be sure, the text of the Constitution, as amended in 2001,Footnote 43 is not clear about the allocation of legislative power on matters of social assistance.Footnote 44 On the one hand, social assistance is not mentioned in the enumerated prerogatives of the State,Footnote 45 suggesting that these policies fall within the residual jurisdiction of the Regioni.Footnote 46 On the other hand, a general clause entitles the State to ‘determine the essential levels’ of social security across the country.Footnote 47 If I want to make my point persuasive, then, I’ll have to argue against a broad reading of the essential-levels clause. In particular, I could claim that the national Government is allowed to legislate, but is not compelled to do so. In addition, the State is only empowered to ‘determine’ the essential levels of social security, a verb which seems to allude to a purely ‘setting-of-the-levels’ activity, one which must necessarily be integrated by regional legislation.
Ms Pancrazio could respond by invoking a 2010 decision, whereby the Court upheld a national statute granting a pre-paid credit card to qualified indigents.Footnote 48 In that case, not only did the Court recognise the national Government’s power to grant a social assistance benefit directly, but it also claimed that the economic crisis was so serious that the direct provision of the benefit was ‘unavoidable’ (sic). Thus, the Court hinted at the possibility that the State had a constitutional obligation to guarantee subsistence uniformly across the country.Footnote 49 In order to defuse the force of this precedent, I will have to distinguish it from the case before us. In particular, I can emphasise that, in 2010, the Court allowed the national statute to stand by insisting that the economic juncture was ‘exceptionally negative’. Nothing is said in that decision about the legitimation, not to say the obligation, of the State to act during less ‘exceptional’ times.
In order to build an even larger consensus around the choice of declaring the case inadmissible, I can attract other colleagues of mine with the promise of including an admonition in the decision. The Court, in other words, can dismiss the case on procedural grounds, but at the same time warn the legislature that its inaction will not be tolerated for long. Once I secure the support of a sufficient number of colleagues over this choice, I will then be able to draft the admonition in a sufficiently loose formulation, in order to leave an ample margin of discretion to the legislature over the amount, distribution and timing of the benefit. As American constitutional history teaches, even an innocent-sounding expression like ‘with all deliberate speed’ can be interpreted as leaving a wide leeway for legislative discretion.Footnote 50 Finally, and in addition to vagueness, I can also try to adopt a particularly cryptic style, which will come in handy in order to hide the lack of solid arguments and prevent an hostile reaction from the media and the public.Footnote 51
Rejecting the challenge on the merits
If a decision of inadmissibility is not popular among my colleagues, I can try to convince them to enter the merits and reject the challenge. If properly written, such a decision would put the question to rest for quite some time. The strongest argument in support of this solution is a textual one. The text of the Italian Constitution makes the right to social assistance subject to three prerequisites: (i) being a ‘citizen’; (ii) being ‘devoid of the means necessary to live’; and (iii) being ‘unable to work’ (inabile al lavoro).Footnote 52 It is this last condition, that of being ‘unable to work’, that sets the Italian Constitution apart from, say, the Spanish one, which makes the provision of social assistance subject to citizenship and need alone.Footnote 53
The mere presence of the unable-to-work clause does not, however, automatically lead to the conclusion that the Italian Constitution is indifferent to the needs of able-bodied people. For this too is a matter of interpretation. Everyone agrees that the expression ‘unable to work’ encompasses people who are disabled, aged or otherwise personally impaired.Footnote 54 But how about those persons who are incapable of finding a job because of an economic slowdown, or because of diffuse prejudice against them, or by reason of any other external obstacle?Footnote 55 Are they too entitled to social assistance? Answering this question is crucial to our case, because Ms Carmini is physically and mentally able to work: she ‘just’ can’t find a job, despite her efforts.
The scope of constitutionally-mandated social protection under Article 38, paragraph 1 turns around the interpretation of the words ‘inabile al lavoro’ (unable to work). Most Italian dictionaries qualify as ‘inabile’ the person who ‘lacks aptitudes or requisites needed to accomplish a certain task’.Footnote 56 In modern definitions, then, the term ‘inabile’ does not cover cases of incapacity originating outside the individual, for example, the inability to work that derives from widespread unemployment. Instead, the contemporary understanding of the word ‘inabile’ tends to be limited to instances of physical or psychological impairment. In keeping with this interpretation, most Italian scholars read the unable-to-work clause narrowly, circumscribing the sphere of constitutional protection to people with disabilities.Footnote 57
The contemporary understanding of the word ‘inabile’, however, is the result of a centuries-old process of semantic change. The Romans employed the adjective ‘inhabilis’, from which the Italian ‘inabile’ originates,Footnote 58 to refer not only to personal, but also to external sources of incapacitation. Livy, for example, describes the strategy of the Romans towards the conquered city of Capua as aimed at making the population ‘ad consensum inhabilem’.Footnote 59 Here, the quality of being ‘inhabilis’ does not refer to any personal condition or trait. In fact, Livy tells us that the Capuans’ incapacity to gather and revolt would derive from their being deprived of autonomous political institutions,Footnote 60 i.e. from institutional or environmental, as opposed to personal, factors. To be sure, the term’s susceptibility to refer to environmentally-induced forms of incapacitation was progressively lost, and by the fifteenth century the vulgar ‘inabile’ started to be used to refer to personal impairments only.Footnote 61 Yet, some residue of the original meaning must persist, if today most Italian dictionaries and thesauri stop short of equating ‘inabile’ with ‘disabile’, ‘inability’ with ‘disability’.Footnote 62 Interpreters, then, should be cautious in drawing hasty conclusions about the clause’s scope.
Since the analysis of the text seems to lead to no clear answer,Footnote 63 the interpreter is compelled to look elsewhere for additional guidance. The debates at the Constitutional Assembly constitute one of the sources where such guidance can be found. A new obstacle immediately arises, however, for the opinions of the framers cannot be treated as monolithic. Hundreds of delegates gathered in the Palazzo di Montecitorio in 1946–47 to draft the new Constitution, and their views sometimes changed over time. In fact, when the topic of assistance to the needy was debated within the Third Sub-Commission in September 1946, sevenFootnote 64 out of nine delegates spoke in favour of a universalist approach: any kind of ‘impossibility to work’ was deemed sufficient to trigger social protection, irrespective of whether it was due to ‘age, physical or mental condition, or any other contingency of a general nature’ [my emphasis].Footnote 65 A few weeks later, the First Sub-Commission confirmed this stance, recognising the right to social assistance of everyone who, ‘for whatever reason, and without fault of her own, is unable to work’ [my emphasis].Footnote 66
The explicit universalist solution endorsed by the two Sub-Commissions, however, was rejected by the Commission for the Constitution, whose ‘Project of Constitution’ recognised a right to social assistance only to ‘citizens unable to work’ (inabili al lavoro).Footnote 67 The omission of any reference to external obstacles in the new formulation reflected a new ideological consensus: the universalist spirit that prevailed in the Sub-Commissions had given way to corporatism, consistent with the political priorities of the Communist, Socialist and Christian-Democratic Parties.Footnote 68 In the Plenary Assembly, most representatives endorsed the restrictive language proposed by the Commission.Footnote 69 This shift from universalism to corporatism lends support to a narrow interpretation of the unable-to-work clause. Thus, a prudent invocation of originalism – i.e., one that passes over in silence the universalist approach of the Sub-Commissions – can advance my cause.
In addition to its original understanding, guidance on how to interpret the unable-to-work clause can come from a comparative analysis. The Preamble to the French 1946 Constitution of the Fourth Republic, incorporated in the 1958 Constitution of the Fifth Republic and so still in force, is particularly helpful in this respect. Like the Italian basic law, the French Preamble recognises the right to ‘adequate means of subsistence’, subject to the condition that the recipient is unable to work. But, in a sort of authentic interpretation, the French text explicitly acknowledges that such an inability can arise ‘by reason of [...] the economic situation’.Footnote 70 We find here a broad interpretation of the unable-to-work requirement: somebody’s inability to work, so the framers of the French Constitution tell us, can originate in objective conditions like an economic slowdown.
Even if we were to adopt a restrictive reading of the unable-to-work clause, however, the condition of a mother with a small child could still be subsumed within the Article’s sphere of applicability. Empirical data are unanimous in showing that Italian mothers struggle in their ability to maintain their jobs and keep pace with their male colleagues in terms of career and salary advancements.Footnote 71 Some colleagues of mine, then, may use this evidence to argue that Italian women are constrained in their ability to work fully and equally: as such, they should be entitled to the protection stemming from Article 38, paragraph 1.
The controversies surrounding the unable-to-work clause could be circumvented by resorting to the second paragraph of Article 38, which grants protection against various risks in workers’ lives, including ‘involuntary unemployment’,Footnote 72 an expression which seems broad enough to include Ms Carmini’s conditions.Footnote 73 A drastic demarcation, however, has historically been made between the first and the second paragraphs of Article 38.Footnote 74 The first paragraph, according to the traditional reading, protects indigents as such, and is meant to provide relief subject to certain specified conditions (need, inability to work, citizenship). The second paragraph, by contrast, refers to workers alone: as such, it functions only when certain risks (e.g., unemployment) materialise in the course of a worker’s life.Footnote 75 Since Ms Carmini is not grounding her claim on her ‘worker’ status, the argument goes, she falls outside the area of protection of that second paragraph.
Ms Pancrazio, however, may raise at least two objections to this line of thought. To begin with, no unemployed person can by definition be a worker. If you are unemployed, you might be a former worker, or a prospective worker, but not, strictly speaking, a ‘worker’. Hence, the word ‘worker’ must be taken as a matter of degree, for otherwise the entire second paragraph would protect nobody.Footnote 76 From this perspective, then, Ms Carmini can be entitled to social security, despite her present lack of ‘worker’ status, perhaps on the condition that she commits herself to actively looking for a job. This is a powerful objection, and my only defence to it will be the decades-old obstinacy that legal scholars and this very CourtFootnote 77 have shown in limiting the efficacy of Article 38, paragraph 2 in and around the workplace.
There is another objection that Ms Pancrazio could raise. A too-strict demarcation of the various paragraphs and articles of the Constitution would produce the effect of leaving a shadowy area, where no social protection exists. This result, Ms Pancrazio could note, conflicts with the universal altruistic design of the Constitution. ‘Starting from the relatively less disadvantaged, up to those in extreme deprivation, the Constitution was conceived as a universal shelter for people in need’, the counsel could say. ‘Wouldn’t it be paradoxical if, after ensuring paid vacations (Article 36), the framers created discrimination inside the group of persons at the apex of need, granting protection to some (workers), but not to others?’
This argument sounds persuasive. The only way for me to weaken it will be to argue that some of the framers cared at least as much about their voters as about needy people in general. ‘Reflecting the preferences of their respective constituencies’,Footnote 78 I could say, ‘the delegates at the Constitutional Assembly drafted a Constitution that is as much based on reciprocity as it is on altruism’.Footnote 79 In support of my thesis, I could point to Article 1, which solemnly declares that ‘Italy is a democratic Republic founded on labour’, and to Article 3, which guarantees ‘the effective participation of all workers in the political, economic and social organisation of the Country’ [my emphasis].Footnote 80 Taken alone, these expressions seem to allude not much to universalism but, rather, to the social-insurance logic typical of workers’ friendly societies.
I must admit, however, that this piecemeal approach, this picking-and-choosing among the clauses and words of the text, does violence to the Constitution as an integrated document. Those very articles that I have just mentioned commit the Republic to the ‘factual equality of citizens’, and bind people to the ‘imperative duties of political, economic and social solidarity’. The Court looks at the Constitution as an integrated whole,Footnote 81 and rightly so: from this perspective, it is hard to ignore an overall commitment to universal solidarity,Footnote 82 nuanced though it may be.
Apart from strictly legal arguments, I can count on other weapons to induce my colleagues to reject the challenge, including reasons of expediency. Courts do not merely apply the law. They also perform a signaling function. They are like lighthouses sending a message to the sailors of the legal community.Footnote 83 The case before us could be a valuable occasion to send the message that this Court will keep a low profile vis-à-vis the legislature, and will refrain from setting off a new expansionary phase in Italian constitutional law. Imagine for a moment what could happen if the Court introduced a nationwide social assistance benefit. Social-rights holders would feel emboldened to raise new constitutional challenges. Employers, in turn, would demand compensation for the higher costs deriving from the increase in workers’ bargaining power.Footnote 84 Firms might then ask political institutions – and courts – to enforce the Republic’s commitment to education (Articles 34 and 35) and technological development (Article 9), so that they can remain competitive with the help of better-skilled employees and new technologies. In short, the Court could be submerged by a waterfall of constitutional challenges. By portraying the consequences of the decision in these apocalyptic terms, then, I can hope to convince some of my colleagues to reject the challenge.
My strongest ally, however, is the doctrine of the separation of powers.Footnote 85 In particular, I can point to the democratic costs which we, as a community of people, would have to bear if political questions were decided by unelected judges. Judicial review, even in its mildest forms, always involves some assessment or ‘alteration’Footnote 86 of the will of a parliamentary majority, but the democratic cost varies from case to case. That cost is relatively small, for example, when a court marginally extends the sphere of protection of an existing social benefit; but it would be much higher if a court were to introduce an entirely new social benefit.
Some scholars, however, have persuasively argued that not all forms of judicial review must be at odds with democracy. Gustavo Zagrebelsky, for example, claims that a reconciliation is possible, as long as we abandon a purely electoral conception of democracy and we extend our perspective to encompass within the notion of ‘democracy’ the pactum societatis at the basis of society.Footnote 87 From this viewpoint, constitutional courts can be considered the supreme guarantors of that pact: the highest expression, rather than a hindrance, of the democratic principle.
In order to counter this argument, I can resort to the theory of the rime obbligate—literally, ‘pre-established rhymes’. According to Crisafulli, who formulated the theory, the task of a constitutional judge can be seen as filling in the lines of an incomplete poem, whose metre has been established in advance.Footnote 88 The poem metaphor is not dissimilar to the image of the collective novel conceived by Ronald Dworkin.Footnote 89 Dworkin, however, used his image to both limit and empower his judge. Most Italian scholars, by contrast, use the allegory of the poet as a reminder of the Justices’ limitations only. To an Italian audience, the introduction of a universal means-tested scheme against poverty will probably sound like writing a poem from scratch: an impermissible exercise of judicial discretion.Footnote 90 I can, therefore, exploit this popular metaphor to advance my cause.
My ‘save democracy’ argument, however, can be attacked on factual grounds.Footnote 91 At certain points in time and space, courts can be more in line with the prevailing sentiment of the public than legislators. This is especially true where the distance between the people and elected representatives is profound, as it is in contemporary Italy.Footnote 92 As already mentioned, some surveys show that most Italian residents favour the introduction of a social assistance scheme.Footnote 93 In this context, a judicial mandate to introduce a universal means-tested benefit against absolute poverty would bring the Court closer to the people.
To refute this point, however, I can powerfully object that the Court should not be the mere reflection of transient public moods. The relationship between the Court and social conscience (i.e. the foundational cultural principles of a given community, in Zagrebelsky’s language) must be kept separate from the relationship between the Court and public opinion. The former is desirable,Footnote 94 but the latter should be avoided, lest the Court lose its independence.
A doubt might now arise in your mind. The Court, you might observe, values its own precedents highly.Footnote 95 ‘How come you have not mentioned precedents much, so far?’, you could ask. It is a legitimate question. Luckily, no precedent is directly relevant to this case. There do exist a few cases that touch upon the right to social assistance, but in most of those the plaintiffs refrained from questioning the social assistance system in its entirety, seeking instead to be included in the recipient groups of existing, selective benefits.Footnote 96 The Court, for its part, condemned discrimination between similarly-situated categories of workers,Footnote 97 but said nothing about the discrimination between labour-market insiders and outsiders. As such, these precedents bear little relevance to the present case.
The only exception seems to be a 1986 decision in which the Court hinted at a constitutional obligation to set up a universalist welfare system.Footnote 98 On that occasion, the Court asserted that the Constitution requires a certain ‘un-differentiation, a uniformity, a single quantitative threshold, for all citizens’.Footnote 99 I can, however, defuse the force of these words by distinguishing the two cases on the facts.Footnote 100 The 1986 challenge was raised by a group of self-employed workers who claimed that the amount of their minimum pensions should be set at the same level as that of employed workers. The case we must decide tomorrow is different. Be it because Ms Carmini is ready to run the risks of a long and uncertain suit, or because she and her counsel highly value social justice, the challenge here is much more ambitious. Ms Pancrazio is asking the Court to mandate the Government to provide social assistance universally (albeit conditionally) to all involuntary indigents: something unheard of in the Palazzo della Consulta. Such novelty plays in my favour, for there is no directly relevant precedent to overrule.
Endorsing the challenge
Our third option is to argue that the lack of any universal means-tested scheme against poverty is unconstitutional. The best argument with which to oppose such an outcome is to invoke the difficulties which afflict the Italian public budget. In enforcing social rights, courts enjoy a space of manoeuvring that is, de facto, inversely proportionate to the amount of deficit and national debt. To my great joy, the Italian public debt is the fourth-largest in the world in GDP terms (2015).Footnote 101 As if this constraint was not enough, a 2012 amendment constitutionalised the ‘principle of the balanced budget’, placing an additional limitation on the Court’s capacity to make costly decisions.Footnote 102 This principle, which requires the Republic to maintain, as a general rule, an equilibrium between revenues and expenses,Footnote 103 has had an impact on the working of the Court in at least one important wayFootnote 104 : a new constitutional interest – the tentative preference for budgetary equilibrium – has been added to the list of interests to be balanced against one another. In practice, this means that the Court is now explicitly and legally required to take into account the soundness of the budget when it enforces costly constitutional rights or principles, which is almost always the case.Footnote 105
The annual cost of a universal means-tested benefit for people in absolute poverty cannot be determined with exact precision in advance, because the number of recipients fluctuates over time in response to many uncertain variables. Despite these difficulties, however, a rough estimate has put the yearly cost of such a scheme at seven billion euros.Footnote 106 This is a relatively small sum, corresponding to just 1.53% of Italian public social expenditure (2014).Footnote 107 In part, this limited cost is attributable to the savings that would come from the simultaneous suppression of existing benefits. In addition, the yearly cost of the program could drop in the long term, as the economy recovers and recipients enter the labour force. All this being said, the Court is unlikely to burden the public finances of the Republic with a decision that, however financially sustainable, would still cost billions of euros.Footnote 108
I know what you are thinking. You object that only a few months ago, in an old-age pension case, the majority of the Justices voted for a very costly decision.Footnote 109 The Court held that a two-year freeze on the automatic revaluation of retirement pensions (the amount of which exceeded the minimum by three times) was unconstitutional. This decision, you might argue, is indicative of the Justices’ readiness to make an expensive decision, even during an economic crisis and even in the face of a disapproving political and academic community.Footnote 110 It may be helpful, however, to see this decision in the context of the social and demographic conditions of its time. The elderly, unlike indigent people, constitute a politically and demographically strong group in Italian society, and one with whom the Justices, by reason of their age and prospective retiree status, can potentially identify. Compassion for indigent people is less obvious. People living in absolute poverty are not few – more than 4.5 million people, or 7.6% of residents – and the figure is growing.Footnote 111 But they have scarce opportunities to push through their preferences in the political arena.Footnote 112 Even when their interests have taken an organised form,Footnote 113 labour-market outsiders have often been unable to promote universal policies or to oppose welfare retrenchment, as the discontinuation of the Reddito Minimo di Inserimento in the early 2000s has shown.Footnote 114
Not only does the 2015 old-age pension case fail to provide any substantial help to Ms Pancrazio, but it may even obstruct her plans. Expensive decisions are unlikely to be followed by other expensive decisions in the short term. The 2015 decision was ‘expensive’ in a double sense: it imposed a high cost on the public finances; but it also extinguished the Court’s ‘stock of respectability’,Footnote 115 as severe criticism ensued from academics and commentators.Footnote 116 That stock being temporarily depleted, there is little chance that the Court will make another ‘expensive’ decision in the near future. Like trial judges, who draft their decisions in a way that minimises the risk of being reversed,Footnote 117 constitutional Justices are driven by a similar ‘reversal aversion’,Footnote 118 which encourages them to avoid an open fight with the country’s political élites. I can therefore exploit the memory of the recent backlash suffered by the Court in order to induce my colleagues to take a more cautious stance.
Despite the force of the budgetary argument, some of my colleagues may nonetheless solicit a decision that mandates the Government to provide social assistance to everyone in need. To support their position, as we have seen, they can resort to textual, comparative and historical considerations.Footnote 119 Alternatively, they can invoke certain non-written constitutional standards, such as rationality, reasonableness and proportionality.
Despite the semantic confusion arising out of the inconsistent use of these terms in the Court’s own decisions,Footnote 120 it is possible to identify certain recurring requirements which have historically been demanded of a statute to allow it to withstand a constitutional challenge. Among these requirements, we find the following two: (i) the consistency between the statute and the broader legal system;Footnote 121 and (ii) an adequate balance between the constitutional right or principle advanced by the statute, on the one hand, and conflicting constitutional principles or rights, on the other.Footnote 122 Ms Pancrazio could argue that the failure to implement a universal means-tested scheme against poverty violates both requirements. It violates the consistency requirement, because the omission conflicts with the principle of universality, set out in Arts. 2 and 3 of the Constitution and in the framework law on social assistance.Footnote 123 It also violates the requirement of sound balancing, because the legislator gave too much precedence to budgetary equilibrium over social assistance. How can I respond to these strong claims?
At first glance, the consistency requirement should be relatively innocuous for my plans. When the term of reference used to assess ‘consistency’ is a set of existing rules, the impact of the requirement can only be conservative.Footnote 124 Major legislative innovations, by definition, are ‘inconsistent’ with the pre-existing legislative framework.Footnote 125 Failure to enact major reforms, in contrast, is always in keeping with the existing legal system. We can see this conservative effect at work by looking at the specific case before us: the lack of a universal means-tested scheme seems to fit perfectly with a legislative context that is rife with incoordination and imbalances.Footnote 126 Ms Pancrazio, however, is making an unconventional use of the consistency requirement.Footnote 127 Her term of comparison is not a given set of existing rules, but a prospective principle, namely, the principle of universal social protection. This innovative interpretation of the consistency requirement could disrupt my plans. If the principle of universal solidarity is the yardstick, then it becomes difficult to defend the ‘consistency’ between the existing policies and that objective. Luckily (from my point of view), the Court has historically opted for the conservative reading of the consistency requirement.Footnote 128 If this line of precedents is confirmed, I can rest assured that this requirement will not only fail to obstruct my argument, but might even play in my favour.
Let us now discuss the requirement of sound balancing. Ms Pancrazio could argue that the balancing (implicitly) performed by the legislature (by failing to legislate) is not as effective and expansive as it could be.Footnote 129 How can I convince my colleagues of the contrary? To begin with, I can point to the fact that the right to social assistance has not been entirely neglected. For decades, specific benefits have been granted to a variety of recipient groups, including the elderly, people with disabilities, families with three or more children, and mothers with children.Footnote 130 Ms Carmini herself benefited from one of those benefits, the Assegno di Maternità del Comune. If we look at the evolution of social policies over time, we can see that the list and scope of these specific benefits is expanding.Footnote 131 And even if the next parliamentary elections are approaching, a draft law is currently being discussed in Parliament in order to authorise the Government to enact by decree what could potentially become, if properly funded, a universalist reform of the entire social assistance system.Footnote 132 All this is to say that the legislature is acting to implement the constitutional right to social assistance—even though it is doing so according to its own political priorities. This evidence may induce some of the Justices to dismiss a ‘strong’ mandate as premature and unnecessary.
The mere fact that welfare reforms are under way, however, does not excuse the Court from checking the soundness of existing legislation and from performing its own balancing test; a test to which I now turn. When Italian scholars describe the way in which two constitutional interests are to be balanced, they usually resort to the theory of the ‘essential core’.Footnote 133 According to this doctrine, the Court’s task is to make sure that neither of the two terms in question (in our case, the principle of a balanced budget and the right to social assistance) is completely sacrificed. The problem, however, lies in what exactly that ‘essential core’ consists of: a question that most scholars answer in very generic terms claiming that the answer should not be predetermined by abstract reasoning, but is instead contingent upon the existing socio-economic and cultural situation.Footnote 134
When it comes to balancing social rights, the ‘essential core’ doctrine assumes a more precise shape in the form of the ‘vital minimum’ doctrine.Footnote 135 The ‘core’ of a social right is deemed to be protected as long as, at a given time and place, the right-holder can satisfy her basic vital needs. But can Italian residents, today, satisfy their basic needs? Statistics tells us that 7.6% of them cannot.Footnote 136 By definition, people in ‘absolute poverty’ are those who lack the means to satisfy their basic subsistence needs. As such, the ‘essential core’ of these people’s right to social assistance seems to be violated.
This sort of social enquiry, however, offers me the opportunity of a rebuttal: if, as Ms Pancrazio suggests, we can legitimately look at empirical evidence, why should the Court limit itself to ascertaining the absolute needs of indigents? Should we not also consider additional data, like the foreseeable social and economic impact of a strong-enforcement decision?Footnote 137 If so, I could then try to convince my colleagues that the judicial creation of a universal benefit would be deleterious. To begin with, I can claim that cash benefits for mothers would be paternalistic, and reminiscent of a time when women were forced into a rigid social role, that of the non-working, breastfeeding wife.Footnote 138 The task of the State, I could say, is not to grant ‘social consumption’ benefits that would keep women at home, but to engage instead in ‘social investment’ policies, like childcare or care for the elderly: policies, in other words, that facilitate women’s participation in the labour market.Footnote 139 In addition, I can pretend to believe that cash assistance for the poor generates laziness or, at best, a decline in productivity.Footnote 140
These obstacles, I must admit, can be bypassed by making access to the benefit conditional upon participation in training programmes and the acceptance of suitable offers of employment. With respect to productivity, the existing incentives to work, such as the prospect of career advancement and higher income, will remain in place.Footnote 141 However, given the incertitude surrounding these probabilistic scenarios, I am confident that I will be able to persuade my colleagues of the contrary. A universal means-tested benefit, I will say, would be bad for the taxpayer (burden on public finances), bad for employers (decline in productivity) and bad for indigents themselves (poverty trap and paternalism).
Now suppose that my efforts do not pay off, and that the majority of the Justices opt for a declaration of unconstitutionality. In that case, I will work to limit the damage. The margins of doing so are not narrow, because an unwritten rule prescribes that the initial rapporteur (in this case, me) drafts the majority opinion even when she belongs to the minority.Footnote 142
As a start, I can try to re-open the discussion on the merits. This strategy is especially likely to succeed if the majority, unlikely as it may be, were to decide to enforce the right ‘strongly’, i.e. not only mandating the introduction of a universalist scheme, but also setting out the details of the new policy.Footnote 143 My first objective, in this case, would be to convince my colleagues to opt for a ‘soft’ decision instead. I have already shown how the invocation of legislative discretion could be useful.Footnote 144 Here, that argument will sound even stronger, because I would be able to point out one by one all the instances where a strong mandate risks usurping the authority of Parliament.Footnote 145 Judges, I could say, can hardly set the frequency with which the benefit is to be disbursed, the conditions of access, the dispensing authority, the financing mechanisms and the nature of the means tests,Footnote 146 without transcending their powers and expertise.
In reality, some of these details can be judicially determined without exercising much discretion. The amount of the benefit, for example, can be tied to the poverty line, as set by the National Institute of Statistics. That threshold is updated yearly, and is differentiated according to age, residence and family size. In short, it is a good approximation of the actual subsistence needs of an Italian resident.Footnote 147 If the Court were to mandate the State to disburse the difference between the family income and that threshold, it would be hard to mourn the death of our parliamentary democracy.
And yet, this seemingly discretion-free act would not suffice. An additional condition is needed to avoid the benefit being granted to people whose assets, in spite of their low income, permits self-sufficiency. Ownership of conspicuous real estate and financial assets, for example, constitutes a strong indication of one’s ability to maintain oneself. At the other extreme, ownership or use of, say, a modest vehicle should not per se constitute a ‘conclusive presumption’ of self-sufficiency, as recently recognised by the Supreme Court of Israel.Footnote 148 In Italy, the existence of an individualised measure of one’s assets (ISEE), made more accurate by a recent legislative reform,Footnote 149 would save the Court the effort of making a detailed list of what count as a sign of self-sufficiency. However, the act of setting the specific ISEE threshold under which one becomes eligible for income support does require the exercise of discretion.
An even higher degree of discretion is required to choose the financing mechanism.Footnote 150 In this policy area, courts can hardly substitute themselves completely for elected representatives, both as a practical matter (lack of adequate administrative resources for implementation) and as a matter of principle (the degree of discretion involved in budgetary decisions being particularly high). These are exactly the boundaries to judicial activism acknowledged in the famous Grootboom decision. In that case, the South African Justices mandated the State to ‘devise and implement [...] a comprehensive and coordinated’ housing programme, but not without acknowledging that ‘the precise allocation [of financial resources] is for national Government to decide in the first instance’.Footnote 151
Coming back to Italy: no doubt the money for a benefit of the kind demanded by Ms Pancrazio would flow from the National Fund for Social Policies. The question, however, is precisely how to supply the fund. One possibility for the Court would be to re-route some of the resources collected through general taxation. But even this choice would require some adjustment on the part of the Government after the case is decided, so that both the overall equilibrium of the budget and respect for European budgetary rules are preserved. As the unsatisfactory enforcement of Grootboom seems to show,Footnote 152 at least some degree of co-operation from the other two branches of government is indispensable, if the desiderata of the Court are actually to be enforced.
Ms Pancrazio could argue that yes, the Government’s involvement is inevitable, but such co-operation should be secured authoritatively, by means of a ‘strong’ decision.Footnote 153 But in a country where even a ‘soft’ Grootboom-style mandate to ‘devise and implement’ a new welfare scheme is unprecedented, the possibility of taking an even stronger stance vis-à-vis the Government and its parliamentary majority would seem to most of the Justices a step too far.
Similarly delicate is the choice of the granting institution, i.e. the entity or entities that should receive and review the applications, perform the means tests, and issue the checks. Should it be the national agency that administers social security or the municipalities, acting as agents of the national Government? Most of the literature is inclined towards the local solution, because it allows for a greater proximity to the needs and conditions of the beneficiaries.Footnote 154 The local solution, however, could also increase the risk of geographical fragmentation and parochialism. In any event, who are we, unelected judges, to make such a political evaluation of costs and benefits?
If the majority of the Justices hold to their preference for strong enforcement, my capacity of rapporteur gives me some margin of manoeuvre to smooth such a drastic solution. Among the various things I can do, one of the most effective is to attach strict citizenship or duration-of-residence requirements to the benefit. More specifically, I can attempt to exclude from the benefit four categories of people: (a) non-Italian EU citizens; (b) documented migrants; (c) refugees, asylum seekers, and stateless persons; and (d) undocumented migrants.
European and international law may limit my ability to restrict eligibility. In reality, however, EU law seems to have only a modest impact on national social assistance policies.Footnote 155 Unlike social security measures like sickness or old-age benefits – with respect to which no member state is allowed to discriminate between its own citizens and other EU nationalsFootnote 156 – social assistance policies tolerate such discrimination. Member States are allowed to deny social assistance benefits to EU citizens of another Member State during: (i) the first three months of residence; or (ii) the entire period in which she is actively seeking employment.Footnote 157 After that period (whichever is longer) has elapsed, the host state is allowed to expel EU citizens who cannot show that they have ‘sufficient resources [...] not to become a burden’Footnote 158 on the national welfare system. Apart from the inappropriate use of the word ‘burden’ to refer to human beings, people in absolute poverty – by definition – lack ‘sufficient resources’. As such, EU law gives them little help against expulsion measures.
The European Court of Justice has, on the one hand, mitigated this approach by allowing parents whose children attend school to remain in the host state and to access social assistance benefits, without their needing to show ‘sufficient resources’.Footnote 159 But, apart from this exception in favour of children and their parents, the Court has made it clear that even an EU citizen who has not been expelled can be legitimately denied access to social services, if she is found to lack ‘sufficient resources’.Footnote 160 In terms of evidence, the European Court of Justice demands that the host state makes an individualised examination of ‘the financial situation of each person’Footnote 161 before it can proceed to expulsion. However, no substantial proof concerning the nature or scope of the alleged ‘burden’ is required: the mere word of the state – i.e., the claim that providing the benefit in question would be ‘burdensome’ – seems to suffice. Overall, it seems, the idea of making the right to free movement within the Union conditional upon personal wealth (an idea that appalled the majority of US Justices in the famous Shapiro v Thompson case of 1969Footnote 162 ) does not much trouble European judges and lawmakers.
Turning to refugees, international law grants them ‘the same treatment with respect to public relief and assistance as is accorded to [...] nationals’.Footnote 163 This constraint, however, can be easily bypassed by levelling down social protection. Under standards of formal equality like this one, a state is allowed to provide relief to both citizens and refugees, or to neither of them. I could argue, ‘To the Italian Republic, with its strained budget, only the second option is viable’. Some colleague of mine could reply that the cost of a national means-tested benefit could be maintained at affordable levels if effective border controls and an EU-wide system of allocation of refugees were in place. A good point, but one to which I can rejoin that strict border control by sea is difficult, if not immoral and illegal, during a humanitarian crisis. As for the existing, limited, EU agreements for the allocation of migrants, they have not been enforced so far.Footnote 164 If some of my colleagues, confused by this unusual political discussion, were to object that all this must be left to politics, I would candidly reply that that is my point, too.
The European Court of Human Rights has occasionally treated some nationality requirements attached to social benefits as impermissible under Article 14 of the European Convention on Human Rights, which prohibits discrimination.Footnote 165 The principle laid down in the Court’s precedents is that restrictions on the basis of nationality must be justified by ‘very weighty reasons’.Footnote 166 I can, however, exploit the ‘wide margin’ of discretion that the Strasbourg Court leaves to Member States in the design of their social policies.Footnote 167 In particular, I could try to make use of that discretion by attaching to the benefit a long residence requirement, which would be almost as restrictive as a nationality requirement, but less vulnerable to the Court’s scrutiny.
The most stringent limitations upon my capacity to attach restrictive eligibility requirements to the benefit, however, come directly from the precedents of the Italian Constitutional Court. To be sure, the text of the Constitution seems to lend support to my restrictive plan. The basic protections of Article 38, paragraph 1 are textually granted to Italian ‘citizens’ alone.Footnote 168 And while paragraph 2 entitles all ‘workers’ to social security, thus encompassing foreign workers,Footnote 169 it says nothing with respect to people who, lacking a stable occupation, cannot be qualified as such. The Court, however, significantly expanded the textual guarantees for non-citizens and short-term residents. Limiting access to holders of an ‘EU long-term residence permit’ has been repeatedly declared unconstitutional.Footnote 170 Similarly suspicious to Italian Justices is discrimination on the basis of citizenship.Footnote 171 As for duration-of-residence requirements, their likelihood of withstanding the scrutiny of the Court varies, depending on the nature of the benefit in question. As a general rule, access to benefits aimed at satisfying the ‘primary needs of a human being’ cannot be made conditional upon prolonged residence.Footnote 172 Since social assistance benefits are meant exactly to meet those primary needs, they seem not to tolerate any requirement of this sort.Footnote 173
The stance of the Italian Constitutional Court on duration-of-residence requirements looks particularly valiant when seen in comparison with constitutional standards elsewhere in Europe. For example, the limitation of the Revenu de Solidarité Active (RSA) to five-year residents, upheld by the French Constitutional Council in 2011,Footnote 174 would probably fail to meet the requirements set by the Italian Court. Yet, this strict judicial stance on duration of residence is of little use until a nationwide benefit, comparable to the French benefit, is introduced in Italy.
The last of Ms Pancrazio’s assaults on my arguments could be framed as a defence of human dignity and factual freedom, both of them fundamental values of the Italian Constitution.Footnote 175 Drawing from the tradition of social democracy and progressive liberalism,Footnote 176 widely represented at the Constitutional Assembly,Footnote 177 Ms Pancrazio could argue that a conditional, means-tested, minimum income would allow workers to withhold their labour from exploitational contracts and give full, real content to their otherwise merely formal freedom.
To be sure, the centrality of liberty and dignity in our constitutional system does not make the right to social assistance any more unconditional or absolute.Footnote 178 Still, by arguing in terms of liberty and dignity, Ms Pancrazio would lift the veil which is distracting my attention away from the case’s moral foundations. True, whether her endeavour would be successful is uncertain. Arguing in these terms could even be counterproductive, as principles such as dignity and liberty have no space in what some of the Justices see as a purely technical enterprise. Other Justices, however, might be stirred by this line of thought and would set to work to find a legal solution that accommodates those moral values.
Conclusion
Outside my window, the night has bathed the roofs and walls of Rome in blue. I am now assaulted by doubt. Alone in my office, I look at myself, and I see a Doctor Azzecca-garbugli in front of a mountain of paper.Footnote 179 But there is life outside legalism. The window is slightly open, and long, oblique moonbeams stretch out on the Persian carpet beneath my desk. In the distance, I can hear the waiters clearing the dinner tables which populate the warren of alleys below the Quirinale Hill. Maybe Federica and her child are there too, somewhere in the dark.Footnote 180
So far I have played with legal arguments, but should I not devote at least as much attention to the underlying ethical question – the question that, in Dworkinian words, aims at the decision that ‘best fits the background moral rights of the parties’?Footnote 181 Farewell, my friend. A long night separates me from tomorrow’s hearing, and I must now prepare for an arduous and self-questioning ethical exercise.Footnote 182
