A primeira Revista de Direito Público Online em Portugal

Justice Unrobed: Judicial Review of Austerity Measures in Portugal

A Justiça sem Toga: Controlo Jurisdicional das Medidas de Austeridade em Portugal 


Abstract: The severe financial crisis which several European states have been facing during the last few years has generated a voluminous and highly interesting wave of constitutional adjudication, since the austerity measures taken by the States’ legislatures as a response to the economic recession were challenged before domestic courts under claims of human rights violations. The present paper will critically evaluate the relevant case law of the Portuguese Constitutional Court, with a view to examining the role of the judiciary in questions of resource allocation in a community, like the ones which lie behind such challenges. It will be argued that decisions on the allocation of resources must, first, respect the equal status of all citizens and, secondly, serve the common good in a way which benefits the community as a whole. The first requirement relates to the constitutional limits of the legislature’s competence, since the state is not empowered to act in violation of the equal status of its citizens; therefore, its observance can and must be ensured by the judiciary. On the contrary, the second requirement calls for an assessment of the various plausible answers, rooted on various conceptions of the “common good”; as such, it can and must be monitored through political debate and, ultimately, through representative elections. This entails that the argument that the judiciary do not address the second question is not one of deference to the political branches, but one of allocation of power between the branches of a democratic government: while courts are mandated to address the first (logically prior) question regarding the constitutional limits of the legislature’s competence, they lack the power to review the plausibility of measures of resource allocation, for example by employing proportionality analysis. In this respect, the relevant practice of many European supreme courts is regrettable.

Summary: Introduction; I. The nature of rights; II. Resource allocation and the authority of judges; (a) The equal concern requirement - the American approach; (b) The common good requirement; (c) Synthesis; III. The authority of judges as seen by the Portuguese Constitutional Court.

Keywords: Jurisprudence of crisis; constitutional courts; authority of judges; nature of rights


Resumo: A grave crise financeira que vários Estados europeus enfrentaram nos últimos anos gerou uma volumosa e interessantíssima vaga de decisões de Tribunais Constitucionais, uma vez que as medidas de austeridade tomadas pelos respectivos legisladores como resposta à recessão económica foram contestadas perante os tribunais nacionais com fundamento na violação de direitos humanos. O presente artigo avaliará criticamente a jurisprudência relevante do Tribunal Constitucional português, com o objectivo de examinar o papel do poder judicial em questões de alocação de recursos numa comunidade, como as que estão por detrás desses desafios. Argumentar-se-á que as decisões sobre a alocação de recursos devem, em primeiro lugar, respeitar o estatuto de igualdade de todos os cidadãos e, em segundo lugar, servir o bem comum de uma forma que beneficie a comunidade como um todo. O primeiro pressuposto respeita aos limites constitucionais da competência do legislador, uma vez que o Estado não está habilitado a violar o estatuto de igualdade dos seus cidadãos; portanto, o seu respeito pode e deve ser assegurado pelo poder judicial. Inversamente, o segundo pressuposto impõe uma avaliação das várias respostas plausíveis, ligadas às várias concepções de “bem comum”; como tal, pode e deve ser monitorizado através de debates políticos e, em última instância, através de eleições. Isto implica que o argumento de que o poder judicial não atalha a segunda questão não é de deferência para com o poder político, mas de distribuição de poderes pelos diferentes ramos de um governo democrático: se por um lado os tribunais estão obrigados a atalhar a primeira (logicamente anterior) questão sobre os limites constitucionais da competência do legislador, por outro não têm o poder de rever a plausibilidade das medidas de alocação de recursos, por exemplo, com recurso à formula da proporcionalidade. A este respeito, é lamentável a prática levada a cabo por muitos dos tribunais superiores europeus.

Sumário: Introdução; I. A natureza dos direitos; II. Alocação de recursos e a autoridade dos juízes; (a) O pressuposto do igual tratamento - uma abordagem Americana; (b) O pressuposto do bem comum; (c) Síntese; III. A autoridade dos juízes aos olhos do Tribunal Constitucional Português. 

Palavras-Chave: Jurisprudência da crise; tribunais constitucionais; autoridade dos juízes; natureza dos direitos



The phenomenon of the European debt crisis needs no introduction. By now, we are all familiar with the devastating consequences of the economic recession with which certain member States of the Eurozone were faced, and with the bailout agreements between the EU institutions and the governments of such States1. These agreements provided the necessary spending cuts including, inter alia, cuts in salaries and pensions of public servants, higher taxation and a series of other austerity measures. These agreed measures were challenged before national courts under claims of rightsviolation. Most prominent and leading judgments amongst them were those issued by the Portuguese Constitutional Court2. After a period of reluctance, in which the Court upheld the impugned measures, a second wave of case law sprangruling that these measures were incompatible with fundamental principles of the Portuguese Constitution.

What I would like to argue today relates, firstly, to the nature of the rights which have allegedly been violated and, secondly, to the institutional authority of judges to adjudicate on issues like the ones brought before the Court. These questions are closely connected to each other, and I will treat them in turn.


I. The nature of rights

In many cases,the applicants put forward three claims in support of theirconstitutional challenges to the specific policy measures enacted by the Portuguese legislature to achieve the broad objectives of the financial support and assistance programmes. They argued that the impugned measures ran counter to:

  • a general principle of legitimate expectations, which amounts to a claim to some form of economic security (Article 2 Const.)
  • the principle of proportionality,which requires that equivalent but less intrusive measures should have been adopted (Articles 2 and 18(2) Const.)
  • the principle of equality in the sharing of public burdens (Articles13 and 104 Const.)

Regarding the issue of legitimate expectations (princípio da proteção da confiança), it suffices to mention that to expect that one’s salary will remain uncut or that one’s job will remain unaffected by any changes in economy, society, or technology sounds as legitimate as to expect that a cataclysmic earthquake will not tear down one’s house or take one’s life. Leaving the financial crisis aside, it is worth wondering if any sensible judge would find that all in-house postmen in governmental services (the people who used to deliver internal post from one office to another) had a legitimate expectation to retain their jobs after emails were invented and widely used.

Now, the claim that one may have a right to legitimately expect that his or her salary will not be reduced arguably entails a claim that one has a right, under certain circumstances, not to have his or her salary reduced3. This claim, then, accords great weight to the question of proportionality as a test to determine the circumstances under which this alleged right is lawfully “infringed” (to borrow the terminology used by the Strasbourg Court). This, in turn, brings me to my main point in this part, because it directly relates to the issue of the nature of the rights compared. 

Taking a closer look at the claims concerned, we see that they are grounded on what in human rights law are termed “social welfare rights” – that is, rights which relate to the protection of the physical and economic well-being of the members of a society (either individually or collectively). These rights are different from basic liberties (or fundamental civil and political rights, or first-generation rights, or whatever you wish to term that category of rights comprising liberties like the freedom of expression, of religion, or of association). The latter constitute the framework within which the status and relationships between members of the community develop. As such, they are not only constitutionally entrenched in theory but they are also immune from social, financial or other conjunctures in practice – indeed no financial or social calamity would justify any infringement on the freedom of speech or of religion. At the same time, protection of these rights in a community can be asserted in black-or-white terms: it may sometimes be hard to determine whether what one sees is black or white (and indeed, the ones called to decide it often get it wrong), but it can only be one of the two – it can never be grey. In other words, one cannot claim that a right – for example, the freedom of religion – is protected “to some extent”, “a lot”, “a little” or anything like this in one state: the right is either protected or not. And each act of the state in question either violates the right or it doesn’t: it cannot violate a right “excessively” or “moderately” or “to some degree4

I doubt, however, that one could say the same with regard to the category of social welfare rights, like the ones invoked in the judicial decisions in question. For example (and knocking on wood just to be on the safe side), if Portugal went bankrupt, or if its GDP plummeted by 90%, would the Court still have ruled that salary cuts were contrary to the Constitution? I think that the Court has already suggested that it would respond in the negative because, firstly, it upheld measures of similar nature (but not of similar force) taken shortly after the sudden outburst of the crisis (see judgments no. 399/20105 and 396/20116) and, secondly, in its subsequent judgments (judgments 474/20137 and 862/20138) it held that the impugned measures were only in concreto (and not in principle) unconstitutional, due to the fact that no adequate proof of their necessity was furnished by the government. What is more, even when deciding that certain measures had actually violated fundamental constitutional principles, the Court weighed the restoration of those principles as less important in light of certain competing considerations, which compelled the effect9 of those judgments to be suspended (judgment 353/201210) or to be produced only ex nunc (judgment 413/201411).

If one, then, admits that radical socioeconomic fluctuations potentially alter the judicial outcome reached on a question of constitutional entitlements12, then one accepts that these claims are not decided on principle, but by relying on varying circumstances. Indeed, these claims, as well as the underlying social rights on which they are grounded, pertain to the process of resource allocation in a society: they seek to inform our decisions as to how scarce resources of a community are to be distributed among its members. As such, these rights are highly dependent on the relevant demand and supply of resources in a society at each given moment; if the supply of a resource shrinks, then the demand will inescapably have to adjust accordingly. Therefore, a severe financial crisis like the one which struck Portugal may reasonably be addressed by taking measures which modify the level or the model or the rationale of resource allocation in the society. This, in turn, means that the extent to which social welfare rights are realized in a society depends on the size of the available state resources and, most importantly, on the political choices of the elected government regarding resource allocation. As a result, the social right to employment or health care or housing is realized to varying degrees under different governments, in the sense that, depending on the current state of the economy and the socio-economic program of each governing party, different governments subscribe to different (richer or less protective) notions of social welfare rights.

The question, then, becomes who and how should make the choices regarding resource allocation in a constitutional democratic state. In other words, what part should the different branches of government (to put it simply: the legislature and the judiciary) play in the decision-making process when it comes to resource allocation?


II. Resource allocation and the authority of judges

I will submit here that our fundamental conceptions of democracy should lead us into assuming that the political branches of a democratic government should have the final say on questions of what the common good is, and therefore how resources should be allocated among the members of the society13. Indeed, these questions touch upon public policy issues for the tackling of which political leaders are selected by the electorate: politicians are selected by their constituents to choose and arrange, in the way they deem best, how the economic and social life of a community is to be run, and they are held politically accountable for their choices. Likewise, the electorate bears the economic and social consequences for its political selection. Under this scheme, the political branches of government are entitled and required to make choices as to whether, for example, to give precedence to health over education or over employment. In fact, this process of decision-making holds equally when it comes to allocating funds (e.g. subsidizing one sector of the economy instead of another) or reducing benefits and pecuniary resources (e.g. making budget cuts). Consequently, it falls upon the electorate to assess and ‘judge’ the politicians for their choices in the subsequent elections. Put simply, questions of policy are to be answered by policy-makers (elected representatives of the people) and reviewed by the polis (the community) in its entirety.

Clearly, the interesting question is how this choice should be made by the political authorities; in other words, whether there are constitutional constraints upon the political branches of government when making their policy decisions. And the answer is in the affirmative: allocation of resources – which, it must be noted, holds equally for distribution and for deprivation of wealth (presumably in times of economic prosperity or recession, respectively) –, must:

  • firstly, respect all citizens’ fundamental rights, that is not compromise the equal status of all citizens, and
  • secondly, serve the common good.

Let us examine these requirements of resource allocation in turn.

(a) The equal concern requirement– the American approach14

The first requirement essentially ensures that the policy-makers abide by the framework outside of which no political decision may be taken, for it would deny the equal status of citizens within the community. In other words, no decision may be taken lawfully, if it is based on arbitrary or discriminatory classifications among the members of society who receive the benefits (or bear the detriments) of the decision. This relates to the rule of equality of all citizens, in the sense that every person has a right to expect that his or her welfare will be considered with equal concern and respect in the decision-making process regarding resource allocation (or any other matter in society), and that (s)he will not be discriminated against on the basis of unacceptable criteria. These criteria are especially those touching upon a person’s fundamental freedoms – for example a criterion of in what (if any) god one believes, or what (if any) thoughts one expresses – but are not necessarily limited to them: for example, a cut in unemployment benefits only of the citizens whose last name begins with an “A” is also discriminatory, for it is based on a criterion which is “irrational and wholly arbitrary.15

However, in most European jurisdictions the equal protection guarantee of the Constitution is increasingly becoming a major doctrinal tool for analyzing, and sometimes striking down, economic and social measures which are unrelated to traditionally suspect classifications (such as race). A closer look at the prevailing judicial practice reveals a general methodology which virtually all courts use to resolve equal protection disputes. Broadly speaking, equal protection claims involve a challenge to laws that allocate benefits or impose burdens on a defined class of individuals. The plaintiff in these cases claims that the government has drawn the line between the favored and disfavored groups in an impermissible place. Of course, the fact of treating individuals differently cannot invariably give rise to an equal protection violation. Thus, the central question in equal protection cases involves deciding whether, under particular circumstances, a challenged classification is impermissible.  

In addressing that issue, the approach of the Supreme Court of the United States has focused on three basic questions: First, how has the government defined the group being benefited or burdened? Second, what is the goal the government is pursuing? Third, is there a sufficient connection between the means the government is using and the ends it is pursuing? When reviewing classifications in the context of state social and economic regulations, the U.S. Supreme Court has consistently applied a form of scrutiny called “rationality review”. Under this method, the reviewing court asks whether the line the government has drawn is rationally related to the achievement of a permissible government purpose16. In other words, to survive equal protection review, a classification must bear some connection to a permissible government end. But when is a classification sufficiently related to justify conferring benefits or imposing burdens on the basis of the difference it tracks?

In almost all cases, the classification will not be perfectly efficient but will be either overinclusive (it will disadvantage a larger class than is needed to achieve the state’s purpose) or underinclusive (some people will not be disadvantaged even though the failure to include them undermines the achievement of the state’s interest) or both. According to the U.S. Supreme Court, the equal protection clause does not require the state to demonstrate that every member of the disadvantaged class possesses the trait relevant to the state’s objective. Indeed, such a test would make all legislation virtually impossible, for almost all laws group people together based on generalizations that do not universally hold. It is, therefore, sufficient to uphold the classification to show that it advances the state’s purpose to some extent. In the same vein, the Court has sometimes said that the equal protection clause permits the legislature to deal with one problem at a time or to proceed step by step18.

In practice, the application of the rational basis review has usually19 led to validation of the legislative scheme, particularly in cases involving what the Court sees as straightforward economic regulation20. Differences in treatment can be justified by relevant differences between individuals and a difference is relevant so long as it bears an empirical relationship to the purpose of the rule21. Notice that the requirement of a relevant distinction provides no guidance as to how the social costs of achieving the state’s objective are to be distributed when different distributions are reasonably efficacious in achieving the state’s goal. Nor does it provide any protection against the concentration of extreme costs on a small group even when a different distribution of the costs over a larger group might be less burdensome for each targeted individual22.

Admittedly, this version of the rational basis requirement becomes meaningless unless some restriction is placed on the kinds of purposes the legislature may pursue23.What makes a legislative purpose invalid under the equal protection clause? There are some purposes that are forbidden by other constitutional provisions: presumably a classification designed to accomplish one of those independently forbidden goals would be unconstitutional as well. For example, Article 41 of the Portuguese Constitution provides for the freedom of religion. Thus, if the state denied a generally available government benefit – for example, drivers’ licenses – to a class of people because of those individuals’ religious beliefs, that law would be unconstitutional. But in such a case, reference to the general rule of equality before the law (under Article 13 Const.) would seem superfluous: A reviewing court could simply rely on the substantive constitutional provision to invalidate the law. Does the equal protection clause of its own prohibit the government from pursuing certain ends? If so, what ends does it prohibit?

My thesis is that the government is barred by the equal protection requirement to deny a benefit or impose a burden on a class of people because it disapproves of their beliefs or status24. Nothing could be a plainer violation of the principle of equal concern than acts of government that exhibit blatant prejudice25 (e.g. assumptions of supposed superiority of one caste over another). Similarly, the equal protection clause does not tolerate classifications on the basis of an individual’s chosen lifestyle26. Individuals have a personal responsibility to define success in their own lives and no state that diminishes a person’s capacity to take charge of his own life can claim that it embraces an acceptable conception of human dignity27.Thus, it is unconstitutional for the legislature to determine that one way of life is preferable to another and therefore more worthy of support. It would be unconstitutional, for example, for the legislature to subsidize small family farms, but not large-scale industry, on the theory that farm life is “wholesome”. Similarly, it would be a violation of the equal concern requirement for a state agency to provide health insurance to the children of its married employees but not to the children of its unmarried workers, on the theory that marriage is an essential component of the ethical culture that society deems best. In short, government attempts to discourage certain lifestyles by means of imposing special burdens or withholding benefits violate the equal concern requirement.28

This aspect of the method by which resource allocation in a society is decided suggests that the political branches of the government have no authority, within the framework of operation of a constitutional state, to decide a resource allocation in violation of the rule of non-discrimination. Therefore, there should be no doubt that courts can and must review the observance, on the part of the government, of the rule of equality, for it relates to the respect of the constitutional limits imposed on the government’s power. The requirement that the government does not exceed its legitimate authority is indeed a matter of constitutional justice, and as such it can (and must) be resolved by a court29. Indeed, this is what the role of (constitutional) courts boils down to: ensuring that all exercise of government on behalf of the state (all state acts, if you wish) falls within its constitutional boundaries; this means that, in making any choice of policy, the government respects the equal status of all citizens. Under the scheme which I describe, the essential function of the court is to verify that the choice made by the political branches – the one of zillion alternative choices a political branch could make for any given problem – does not compromise the equality of citizens. In this process, the court must argue, persuade, and prove why an act of government (a political choice), irrespective of its merits, fails to abide by the obligation to accord equal concern and respect to all citizens. After all, the constitutional requirement for reasoned judgments reflects precisely the idea that good reasons must be given for a democratically legitimated choice not to be implemented30.

(b) The common good requirement

Does this mean that these questions of policy may be answered in a way which does not purport to reflect the common good in the community? Of course not. This is what the second requirement calls for: that resource allocation should serve a notion of the common good. However, in any given society there may be various – if not infinite – conceptions of what the “common good” is or how it will be best served, and many conceptions among them are reasonable, which means that they can be rationally defended. For example, one opinion (advocated by, say, one political party31) could be that resources are best allocated if funds were taken from the public sector and used for the attraction of private investments, or if funds were taken from industries and used to subsidize the agrarian sector, or anything under the sun. Indeed, all policies advocated by all voices in a political debate (whether in the government, in the opposition, or among the citizens) claim to further the public good. But precisely this variety of conceptions of what the public good requires proves that these are not questions that can have only one reasonable answer32 and any democratic constitutional state should enable and encourage debates on these issues, both in and outside the legislature33. The requirement that a decision serve the public good is not of the same nature as the requirement that a decision serve constitutional justice, and therefore its respect cannot be reviewed in the same way. It is a requirement to be monitored and fulfilled by political debate and, ultimately, by the (s)election of the decision-makers by the whole citizenry. This is, after all, why citizens’ representatives are (s)elected: to govern the community in a way which they deem good. 

(c) Synthesis

My point – which, I hope, should be clear by now – is that, essentially, we are faced with two wholly separate questions, which should accordingly be addressed by two different branches: one question is whether a choice on a matter of resource allocation complies with the constitutional framework of the community, the other is whether the choice advances the common good. The former question can only be answered in one way, and indeed in a way rooted on a theory of constitutional justice – it is a question of principle and, as such, it should be answered by the judiciary. In contrast, the latter is a question of policy, and therefore it invites many reasonable answers, depending on our conception of the “common good” and on the subject which gives the answer – as such, this question should be addressed to the political branches of government and, ultimately, to the electorate34. In other words, when it comes to the assessment of the requirement that a policy choice serve the public good, it is not that courts must defer to the political branches (and the electorate) – it is that courts have no authority to make such assessments in the first place35. Conversely, it lies outside the constitutional power of the political branches to make decisions which violate a fundamental right or fails to accord equal concern and respect to all citizens.

In fact, this order reflects a logical sequence between the two questions: the question whether a legislative enactment serves the public good presupposes an affirmative answer to the question whether that enactment is lawful. To give an example (and leaving international law aside for a moment36), a response to the severe financial crisis might have been to extradite all population living below the poverty line – in fact, some people, I fear, based on their economic or social perceptions, might embrace this measure as an effective way to advance the “common good.” Irrespective of whether this would indeed be a good solution, it would certainly be (fundamentally) unlawful, and as such it should be struck down by a court of justice. Conversely, raising taxes on the middle class may arguably not be an advisable way to react to the financial crisis – and many economists or other citizens may have solid grounds to believe it will fail to promote the “public good”– but it should be upheld in court, for it can hardly be seen as an “unlawful” legislative decision. 


III. The authority of judges as seen by the Portuguese Constitutional Court

Now, let us apply this scheme to the question of austerity measures introduced by the Portuguese government, as reviewed by the Portuguese Constitutional Court. What emanates from secondary sources and translations of the pertinent case law37 is that the Court – in its landmark judgments 353/201238 and 187/201339, as well as in subsequent judgments such as 862/201340 and 413/201441–upheld the argument that the cuts were contrary to some concept of (proportional) equality between, on the one hand, the civil servants and the pensioners, who bore the consequences of the salary cuts, and, on the other hand, the other citizens42.In ruling this way, the Court erred in analogizing public with private sector servants, not because their position is different in terms of public security, trust or benefits but rather because their employer is different43. In principle, the same occurs with respect to private sector businesses when the latter are in economic trouble. If we were to follow the equation between the two categories for reasons of equality, the salaries of public servants should be cut in such circumstances to break even with those of their private sector colleagues. Hopefully, it is clear by now that a judgment would trespass into questions of resource allocation if it stated that the cuts on salaries of public servants amount, for example, to indirect taxation and that they should be mitigated, therefore, by taxation on the rest of the population44.

What is more, the Portuguese Constitutional Court went on to rule that the austerity measures taken were essentially affecting “disproportionately” the right of the civil servants not to be dismissed or not to have their salaries and pensions cut45. Given that – alas! – the state funds would not change (that is the financial figures would not improve by a wave of a magic wand), this inescapably meant that resources from other beneficiaries (other citizens, broadly conceived) had to be transferred to the benefit of civil servants: in this sense, the Court indeed advocates for a certain method of resource allocation46.

More importantly, the Court ruled that the measures ought to be struck down because the government had failed to argue plausibly that the measures taken were indeed proportionate to the harm suffered by the civil servants or the pensioners. There are two conclusions drawn from this argument. First, that the same measure can fall within or outside the government’s constitutional authority (and therefore be compatible with the Constitution or not), depending on (a) the substantiality of the “public interest” which the government seeks to advance and (b) the intensity of the setbacks suffered by the groups targeted by the measure. Indeed, if the financial crisis had been harsher or the salary cuts had been less severe, the measures would be potentially upheld by the Court. And secondly, that it falls upon the government to prove that the impugned measures merit precedence over the individual rights infringed.

Let us go back for a second to the example of the government deciding to extradite all poor citizens in order to preserve resources: this measure would always be considered as falling without the government’s legitimate authority (and therefore unconstitutional), no matter how compelling reasons – economic analyses, sociological reports or anything under the sun – the government put forward to support it. It would be a measure instrumentalizing people to serve other ends, and it would deny the targeted persons’ equal standing with all other members of the community. At the same time, it would fall upon the Court to argue that the extradition measures exceed the constitutional powers of a democratic government. It would obviously be fairly easy in this fictional example, but the point is that it is the Court which should argue in a reasoned manner that a measure fails to observe individual rights or the equality between citizens; it is not the government which bears the onus of proving that the measure does not violate the principle of equality or the rights of individuals47.

Coming back to the actual case of the austerity measures in Portugal, the difference is striking. Firstly, whether or not the same measures are a violation of constitutional rights is not a single-answer question decided by reference to unconditional propositions but rather depends on variable circumstances such as economic contingencies emerging amid the crisis. Secondly, the government bears the onus of proving why these measures must be upheld by the Court. The question, then, becomes: is this judicial operation more akin to an inquiry of principle or of policy? Does it investigate whether the governmental action is lawful or does it explore, instead, whether the legislature came up with an advisable course of action in furtherance of the public good? Doesn’t the reasoning advanced by the Court constitute an assessment of the quality of the decisions taken by our elected representatives, an evaluation which is alarmingly resemblant to a second-instance decision with regard to the wisdom or desirability of the resource allocation selected by the government? Arguably, it does constitute such a second-instance assessment. In fact, given that it falls upon the government to prove the plausibility of its actions, the government is called effectively to persuade the Court that the action taken was the best among many which could alternatively have been taken – and this task of persuading arguably resembles all too much the task of politicians before their constituencies48.

Indeed, if it is the Constitutional Court which decides whether a measure strikes a fair balance among the numerable conflicting or diverging interests within a community, then arguably we would need no elections at all: we could draw by lot an executive committee and let it issue decision after decision, feeding the Court with cases and hoping that some of them would eventually please the bench. If the judicial review undertaken by courts with respect to constitutional rights is perceived as broadly as the Portuguese (and other) Constitutional Court(s) suggest, then every piece of legislation potentially gives rise to a human rights issue, and therefore the judiciary will be called to decide on virtually any question of public policy, from fines for parking violations to the fluctuation of interest rates49. Furthermore, in doing so, it will be bound to employ a standard that is much more intrusive than mere rational connection, just as the Portuguese Constitutional Court did in the austerity cases50. Such a scheme, however, would tilt the institutional balance between the legislature and the judiciary impermissibly in favor of the latter: the courts would effectively not act as fora of review of the acts of public authorities, but as bodies of appeal of policy choices made by the representative and accountable officials – a task for which courts lack the resources, the know-how and (most importantly) the legitimacy.

What this paper has attempted to argue is not that courts must perform no judicial review of political decisions, nor that they should defer, as a matter of comity or who-knows-what, to the political branches when it comes to “political questions” or other sensitive subjects; on the contrary, a Constitutional Court can and must review all decisions which may compromise the equal concern and respect owed by the state to the citizens. This is in fact, a second “sin” committed by the Portuguese Constitutional Court: when it was constitutionally obligated to uphold the equality of citizens by ruling that a ban on same-sex marriage was discriminatory against same-sex couples (in 2009)51, it preferred to abstain from exercising its mandate, based on an ill-conceived notion of “deference” when it comes to “political questions”52. But ruling that the question of equality of citizens must be decided in the courtroom, while the question of implementing spending cuts must be decided by the legislature, is not a matter of some fluctuating and undefined notion of “deference”; rather it is a matter of institutional balance that is of proper allocation of power between the branches of a democratic government. And while it will always be unlawful to deny same-sex couples the benefit of founding a family enjoyed by opposite-sex couples, no matter what the legislature regards as “good” or “wise” public policy, it makes all the difference in the world whether the legislature believes that it is public spending which must be reduced or not.

The Portuguese Constitutional Court – like most of its European counterparts, I fear – misplaces its constitutional boundaries, depriving citizens both of their constitutional rights’ protection, when this is most needed, and of their political voice, when this is most essential. We must resist this.


* Professor of Constitutional Law, University of Athens. An early version of this paper was presented in the Conference on the Portuguese Constitutional Court’s Jurisprudence of Crisis held on 14 March 2016 at the University of Lisbon and I benefited a lot from comments of my colleagues that participated in the Conference. I am grateful to Nikiforos Panagis and Panagiotis Tsialas, without whose assistance this article might not have been the same. I would also like to thank Dimitrios Kyritsis and Spiros Skliris for comments in earlier drafts. Responsibility, however, lies solely with me.
Economic and political stability in Portugal has been deeply affected by severe financial trouble. Faced with the imminent threat of bankruptcy, Portugal became a bailout country in 2011, and thus subject to strict conditionality, and exited the financial and assistance program in May 2014.

1 This paper has taken into consideration judgments 399/2010, 396/2011, 353/2012, 187/2013, 474/2013, 574/2013, 575/2013, 794/2013, 602/2013, 862/2013 and 413/2014.

2 The Court held that there is no right to continue to be paid a salary or a pension of a particular amount: Judgment 396/2011. This is in full alignment with the case law of the ECtHR: see Vilho Eskelinen and others v. Finland App No 63235/00 (ECtHR [GC], 19 April2007) at para.94 (for salary), Andrejeva v. Latvia App No 55707/00 (ECtHR [GC], 18 February2009) at para.77 (for pension).

3 The Court held that there is no right to continue to be paid a salary or a pension of a particular amount: Judgment 396/2011. This is in full alignment with the case law of the ECtHR: see Vilho Eskelinen and others v. Finland App No 63235/00 (ECtHR [GC], 19 April2007) at para.94 (for salary), Andrejeva v. Latvia App No 55707/00 (ECtHR [GC], 18 February2009) at para.77 (for pension).

4 For a more elaborate discussion of this argument, see STAVROS TSAKYRAKIS , Proportionality: An assault on human rights?, I • CON, Volume 7, Number 3, pp. 468 – 493.

5 In 2010, while Portugal was already under an excessive deficit procedure, the Court was asked to decide on the compliance of Laws 11/2010 and 12-A/2010 with Article 103 (3) (prohibition of retroactive legislation), and Article 2 (principle of protection of legitimate expectations) of the Constitution. A majority of the Justices upheld the validity of the challenged measures. According to the Court, there was no authentic retroactivity, the only kind forbidden by the Constitution, and the expectations of tax-payers were not violated insofar as the two laws pursued a legitimate aim, were adopted as urgent measures to counteract the financial crisis, and were announced well in advance as measures to reduce the public deficit and debt. In this case, the Court did not investigate whether the proportionality principle had been observed - it did not, that is, examine whether less restrictive means could have been used, and thus aligned its position with that of the legislature.

6 In this case, the Justices tried to set a standard to be applied in future cases to assess whether the cuts were consistent with the Constitution. The constitutional challenge was brought against the Budget Act for 2011 whereby cutbacks in public salaries were increased from 5 to 10 per cent. Although it was not the first time that public salaries were reduced, the particular economic circumstances entailed the threat that public wage cuts might become a persistent feature of Portuguese fiscal policy, as the annual Budget Acts could confirm the salary reduction year after year, or even increase it. The Budget Act for 2011 was challenged before the Court on the grounds of a violation of the principle of equality (Article 13 Const.), of the principle of protection of legitimate expectations (Article 2 Const.) and of the principle of proportionality (Article 2 Const.). The Court dismissed the challenges holding that there was no legitimate expectation at stake, given the lack of a constitutional right not to have a wage reduction. Moreover, according to the Court, public salary cuts were justified on the bases of the transitional nature of the measure, and of the existence of a compelling interest (to enforce the Growth and Stability Pact and to ensure fiscal sustainability by means of the most effective tools to achieve the target as soon as possible). Finally, the limitation of public salaries was also justified in the name of the public interest as, unlike private workers, public employees are paid with public money. As previously mentioned, in this ruling, the Court fashioned its two prong test (temporary nature of the cuts plus suitability/necessity to pursue the targeted objectives of fiscal stability and reduction of the public deficit) which was used as a constitutional standard in almost all its rulings since then (above, n 3).

7 In this case, a decree of the Portuguese Parliament, which made the dismissal of public employees for objective reasons easier, was declared unconstitutional as it violated the principle of the protection of legitimate expectations and the principle of legal certainty, both derived from Article 2 of the Constitution. The Court stated that the national measures – the Budget Acts – implementing the Financial and Economic Assistance Program had justified the salary cuts of public workers in exchange for the greater job stability that they enjoyed compared to workers in the private sector. Such a justification created an expectation among public employees that, according to the Court, had to be protected and that the decree contradicted. The relevant provision of the decree ultimately did not survive the Court’s usual test for temporary effects and proportionality of the measures. According to the Court, the challenged provision was certainly not transitional and the government failed to demonstrate that the changes in labor relationships in the public sector were really necessary and adequate to match the need for a more efficient public administration.

8 In Judgment 862/2013 of 19 December 2013 several legislative provisions aiming to amend the statute governing the retirement of public sector staff were declared unconstitutional on the ground of a violation of the protection of legitimate expectations. These provisions (retrospectively) cut the pensions paid by the public pension fund (CGA) by 10 per cent and introduced a less favorable formula for calculating pensions. As in its previous judgments on public salary cuts (judgments 353/2012 and 187/2013), the Constitutional Court highlighted that only some of the pensioners, namely public servant pensioners, were subject to cuts. According to the Court, in this case too the measure had an indefinite duration and it could only be justified if and insofar as it was part of an overall structural reform of the public system in which several factors affecting financial sustainability were addressed at the same time to improve intra- and inter-generational fairness.

9 The Portuguese Constitution provides the Constitutional Court with the authority to ‘rule that the scope of the effects of the unconstitutionality or illegality shall be more restricted’ than what is prescribed as the general rule of retroactivity and its exception (Article 282(4) Const.). The limitation of the effects of the declared unconstitutionality must be used ‘for the purposes of legal certainty, reasons of fairness or an exceptionally important public interest, the grounds for which shall be given’. The Court has used Article 282(4) Const. twice during the years of the financial crisis (see nn 12-13).

10 In judgment 353/2012, while the Court declared the suspension of certain allowances to public workers unconstitutional, it did not go so far as to irremediably impair the government’s duties and commitments vis-à-vis the other Eurozone countries and the Troika. When the decision was taken, the execution of the budget for 2012 was already underway. Thus, the Court considered that the consequence of a declaration of unconstitutionality, namely the annulment of the law ex tunc, could have put the state’s solvency in danger, as the State would have to give the suspended allowances back to their legitimate holders. Therefore, the Constitutional Court used the prerogative it has under Article 282.4 Const. and restricted the effects of the declaration of unconstitutionality and decided not to apply them to the suspension of the 13th and 14th monthly allowances in 2012.

11 In this case the Portuguese Court decided to restrict the effects of its declaration of unconstitutionality again but in a rather different manner to that in the 2012 ruling. Indeed, the Court held that a further reduction of public salaries was in contrast with the principle of equality, but this time it did not suspend the effects of its ruling. It only prevented its judgment from establishing retroactive effects and the wage cuts were annulled only ex nunc starting from the date of the ruling – 30 May 2014, i.e. the wage cuts introduced before were not affected. This time, the standpoint of the Court appears to be the protection of the level of public salaries for the remaining period of execution of the 2014 budget, whereas in 2012 the point of view taken by the Court was that of preserving the level of public revenues estimated for the whole fiscal year.

12 Starting from 2010 through 2014, the economic situation became worse and Portugal obtained financial assistance in exchange for a more austere fiscal policy. This latter element led to a redistribution of the already limited public resources, to the detriment of public workers and pensioners in particular. According to A M GUERRA MARTINS, Constitutional Judge, Social Rights and Public Debt Crisis: The Portuguese Constitutional Case Law, Maastricht Journal, 22, 2015, p. 682: ‘“it was not the case law of [the Portuguese] Constitutional Court that changed from 2011 to 2012; it was the austerity measures that became harsher.’”.

13 In cases brought before it, the ECtHR emphasizes that the choice of the domestic policymaker should be given special weight: Hatton and others v. UK App No 36022/97 ([GC], 8 July2003) at para. 97; Valkov and others v. Bulgaria App Nos 2033/04 and others ([GC], 25 October2011) at para. 92; Vistiņš and Perepjolkins v. Latvia App No 71243/01 ([GC], 25 October2012) at para. 98. This reasoning, however, has not always been applied correctly, i.e. it has been applied in cases where no margin should have been afforded: see for example S.A.S. v. France App No 43835/11 ([GC], 1 July2014) at para. 129.

14 For a neat overview of the American caselaw on the equal protection clause and the rational basis review see GEOFFREY STONE et al., Constitutional Law, 7th edition, New York, 2013: p. 497-520, 707-719, 750-776, 824-841. This subsection of the paper largely follows the relevant expositions of this casebook.

15 Similarly, a state cannot, without any further justification, provide more money for the education of students who live in the northern part of the state. For a discussion see GERALD L.NEUMAN, Territorial Discrimination, Equal Protection and Self-Determination, University of Pennsylvania Law Review, 135, 261,1987. In practice, wholly arbitrary classifications are quite rare. easement. The Olechs alleged that the village required only a fifteen-foot easement from other property owners, that the difference in treatment was ‘irrational and wholly arbitrary.’ In a per curiam opinion, the U.S. Supreme Court upheld the equal protection challenge. According to the Court, a plaintiff can bring an equal protection claim even if (s)he belongs to a ‘class of one’ when (s)he ‘alleges that she has been intentionally treated differently from others similarly situated and that there is no rational basis for the difference in treatment.’ In addition, the Court came close to detecting sheer arbitrariness in at least two more challenges to legislative classifications. See: Enquist v. Oregon Department of Agriculture, 553 U.S. 591 (2008)(dismissal of public employee for allegedly arbitrary, vindictive and malicious reasons); Armour v. City of Indianapolis, 132 S. Ct. 2073 (2012) (Indianapolis decided to forgive the assessment for sewer improvement projects levied on property owners who had chosen to pay their assessment in periodic installments but refused – in order to reduce the administrative costs – to refund the payments made by property owners who had chosen to pay in a lump sum).

16 ‘[L]egislative classifications are valid unless they bear no rational relationship to the State's objectives.’Massachusetts Bd. of Retirement v. Murgia, 427 U.S. 307 (1976); ‘[The] constitutional standard under the Equal Protection Clause is whether the challenged state action rationally furthers a legitimate state purpose or interest.’ San Antonio Independent School District v. Rodriguez, 411 U.S. 1 (1973).

17 A good illustration of the Court’s approach can be found in New York City Transit Authority v. Beazer, 440 U.S. 568 (1979). In this case the Court upheld a New York City Transit Authority (TA) rule which prohibited employment of persons who used narcotic drugs (including methadone, a drug widely used in the treatment of heroin addiction). Two justices dissented on the grounds that: “[Even] were successfully maintained persons marginally less employable than the average applicant, the blanket exclusion of only these people, when but a few are actually unemployable and when many other groups have varying numbers of unemployable members, is arbitrary and unconstitutional. Many persons now suffer from or may again suffer from some handicap related to employability. But petitioners have singled out respondents—unlike ex-offenders, former alcoholics and mental patients, diabetics, epileptics, and those currently using tranquilizers, for example—for sacrifice to this at best ethereal and likely nonexistent risk of increased unemployability. Such an arbitrary assignment of burdens among classes that are similarly situated with respect to the proffered objectives is the type of invidious choice forbidden by the Equal Protection Clause.” (White, J., dissenting). To this argument, the Court responded that “because [the classification] does not circumscribe a class of persons characterized by some unpopular trait or affiliation, it does not create or reflect any special likelihood of bias on the part of the ruling majority. Under these circumstances, it is of no constitutional significance that the degree of rationality is not as great with respect to certain ill-defined subparts of the classification as it is with respect to the classification as a whole. No matter how unwise it may be for TA to refuse employment to individual car cleaners, track repairmen, or bus drivers simply because they are receiving methadone treatment, the Constitution does not authorize a federal court to interfere in that policy decision.” Similar statements can be found in Vance v. Bradley, 440 U.S. 93 (1979), which ruled that “[even] if the classification involved here is to some extent both underinclusive and overinclusive, and hence the line drawn by Congress imperfect, it is nevertheless the rule that in a case like this 'perfection is by no means required' "as well as in San Antonio Independent School District v. Rodriguez, 411 U.S. 1 (1973): “[The] existence of ‘some inequality’ in the manner in which the State's rationale is achieved is not alone a sufficient basis for striking down the entire system. […] Nor must the financing system fail because [other] methods of satisfying the State's interest, which occasion ‘less drastic’ disparities in expenditures, might be conceived.” The classical statement of the required nexus between means and ends under rational basis review, however, can be found in Dandridge v. Williams, 397 U.S. 471 (1970), which held that: “In the area of economics and social welfare, a State does not violate the Equal Protection Clause merely because the classifications made by its laws are imperfect. If the classification has some ‘reasonable basis,’ it does not offend the Constitution simply because the classification ‘is not made with mathematical nicety or because in practice it results in some inequality’. [Although the regulation may be both over and underinclusive,] the Equal Protection Clause does not require that a State must choose between attacking every aspect of a problem or not attacking the problem at all. It is enough that the State's action be rationally based and free from invidious discrimination.”

18 Williamson v. Lee Optical, 348 U.S. 483 (1955). “The problem of legislative classification is a perennial one, admitting of no doctrinaire definition. Evils in the same field may be of different dimensions and proportions, requiring different remedies. Or so the legislature may think. Or the reform may take one step at a time, addressing itself to the phase of the problem which seems most acute to the legislative mind. The legislature may select one phase of one field and apply a remedy there, neglecting the others. The prohibition of the Equal Protection Clause goes no further than the invidious discrimination.” The Court took a similar approach in Railway Express Agency v. New York, 336 U.S. 106 (1949). In this case, a New York traffic regulation prohibited the operation of “advertising vehicles” in order to avoid distractions but permitted placing “business notices upon business delivery vehicles”, so long as such vehicles were engaged in the usual business or regular work of the owner and were not used merely or mainly for advertising. According to Court “The local authorities may well have concluded that those who advertised their own wares on their trucks do not present the same traffic problem in view of the nature or extent of the advertising which they use. [. . .] And the fact that New York City sees fit to eliminate from traffic this kind of distraction but does not touch what may be even greater ones in a different category, such as the vivid displays on Times Square, is immaterial. It is no requirement of equal protection that all evils of the same genus be eradicated or none at all.”

19 Since Dandridge, the Court has generally adhered to the view that rational basis review is the appropriate standard for the evaluation of welfare classifications. See, e.g. Califano v. Boles, 443 U.S. 282 (1979) (upholding a provision of the Social Security Act restricting “mothers’ insurance benefits” to widows and divorced wives of wage earners); Jefferson v. Hackney, 406 U.S. 535 (1972)(upholding a provision of a state welfare program authorizing payment of a lower percentage of need to recipients of AFDC than to recipients of other forms of categorical welfare assistance); Richardson v. Belcher, 404 U.S. 78 (1971) (upholding a provision of the Social Security Act reducing disability benefits for amounts received from workers’ compensation but not for amounts received from private insurance); Lindsey v. Normet, 405 U.S. 56 (1972) (upholding a state’s summary forcible entry and wrongful detainer procedures for the eviction of tenants after alleged nonpayment of rent against a constitutional challenge on the grounds that eviction actions were more summary than “other litigation” – according to the Court the “unique factual and legal characteristics of the landlord-tenant relationship [justify] special statutory treatment.”)

20 A rare exception was the Court’s opinion in Metropolitan Life Ins. Co. v. Ward, 470 U.S. 869 (1985). In this case, the Court struck down an Alabama statute that imposed higher gross premiums taxes on out-of-state insurance companies than on domestic ones. The Court reasoned that “Alabama’s aim to promote domestic industry is purely and completely discriminatory”. The Court grounded its reasoning on two propositions: (a) Under the circumstances of this case, promotion of domestic business by discriminating against nonresidents is not a legitimate state purpose. (b) Nor is the encouragement of the investment in Alabama assets and securities a legitimate state purpose.

21 This is presumably what the Court means in Beazer when it says that “legislative classifications are valid unless they bear no rational relationship to the State’s objectives”. Even though the Transit Authority’s no-methadone rule treated methadone users and nonusers differently, the Court thought that the rule did not violate the equality principle because there was a difference between the two classes relevant to the state’s objective of a safe and efficient transit system.

22 An interesting question is whether a state can defend an otherwise irrational classification on the ground that it is the product of compromise. In Bowen, v. Owens, 476 U.S. 1137 (1986), for instance, the Court defended certain classifications contained in the Social Security law as follows: “Congress' adjustments of this complex system of entitlements necessarily create distinctions among categories of beneficiaries, a result that could be avoided only by making sweeping changes in the Act, instead of incremental ones. A constitutional rule that would invalidate Congress' attempts to proceed cautiously in awarding increased benefits might deter Congress from making any increases at all.” Few people would disagree that raw interest-group deals, justified by nothing other than the political strength of the beneficiaries, are prohibited by the equal protection clause. This prohibition, however, may not subject to principled judicial enforcement because inquiries into the legislative process might prove unmanageable and strain judicial competence and authority. But although this prohibition is “underenforced,” it nonetheless remains binding on legislators and administrators who have an obligation to obey the Constitution. See generally LAWRENCE GENE SAGER, Fair Measure: The Status of Underenforced Constitutional Norms,Harvard Law Review, 92, 1978, p. 1212.

23 For example, imagine a Transit Authority whose director has decided that the public welfare would be best enhanced by promoting traditional family structures and that providing high paying jobs to men while deterring women from entering the workforce will serve this goal. In this scenario, sex is a relevant difference among job candidates.

24 In Romer v. Evans, 517 U.S. 620 (1996) the Court invalidated a Colorado constitutional amendment prohibiting local governments from enacting anti-discrimination measures protecting “homosexual, lesbian, or bisexual orientation, conduct, practices or relationships.” According to 6 Justices, “[Amendment] 2 fails, indeed defies, even [conventional rational basis] inquiry. First, the amendment has the peculiar property of imposing a broad and undifferentiated disability on a single named group, an exceptional and, as we shall explain, invalid form of legislation. Second, its sheer breadth is so discontinuous with the reasons offered for it that the amendment seems inexplicable by anything but animus toward the class that it affects; it lacks a rational relationship to legitimate state interests. [. . .] Laws of the kind now before us raise the inevitable inference that the disadvantage imposed is born of animosity toward the class of persons affected. . . Even laws enacted for broad and ambitious purposes often can be explained by reference to legitimate public policies which justify the incidental disadvantages they impose on certain persons. Amendment 2, however, in making a general announcement that gays and lesbians shall not have any particular protections from the law, inflicts on them immediate, continuing, and real injuries that outrun and belie any legitimate justifications that may be claimed for it.”
Much more complicated is the Court’s opinion in a case called Plyler v. Doe, 457 U.S. 202 (1982). In this case the Court held unconstitutional a Texas statute that authorized local school districts to deny free public education to children who had not been “legally admitted” into the United States (children of undocumented immigrants). Emphasizing the fundamental role of basic education as well as the fact that the law imposed its discriminatory burden on the basis of a legal characteristic over which the targeted children could have little control, five Justices ruled that a heightened form of scrutiny was constitutionally required by the equal protection clause (“the discrimination contained in [the statute] can hardly be considered rational unless it furthers some substantial goal of the State”) (emphasis added). Later cases, and especially Martinezv. Bynum, 461 U.S. 321 (1983) (upholding a Texas statute that authorized local school districts to deny tuition-free admission to public schools to minors who lived apart from their parents or guardians and whose presence in the district was “for the primary purpose of attending the public free schools”) and Kadrmas v. Dickinson Public Schools, 487 U.S. 450 (1988)(upholding a user fee for transporting students to and from public schools even though it was equally high for the rich and the poor kids) dramatically limited the holding of Plyer to its unique circumstances.

25 For instance, the City of Cleburne, Texas, had a municipal zoning ordinance that permitted a wide variety of structures to be built on a particular site, including “[hospitals], sanitariums, nursing homes or homes for convalescents or aged.” However, the ordinance specifically excepted “homes for [the] insane or feeble minded . . .” Pursuant to the ordinance, the city denied a special use permit for the operation of a group home for the mentally retarded. In a case called Cleburne v. Cleburne Living Ctr., 473 U.S. 432 (1985) the Court dismissed as illegitimate the claim that “the Council was concerned with the negative attitude of the majority of property owners located within 200 feet of the Featherston facility, as well as with the fears of elderly residents of the neighborhood.” According to the Court “mere negative attitudes, or fear, unsubstantiated by factors which are properly cognizable in a zoning proceeding, are not permissible bases for treating a home for the mentally retarded differently from apartment houses, multiple dwellings, and the like . . . The short of it is that requiring the permit in this case appears to us to rest on an irrational prejudice against the mentally retarded . . .” In his concurring opinion Justice Stevens added the following: “[In] my own approach to these cases, I have always asked myself whether I could find a ‘rational basis’ for the classification at issue. The term ‘rational,’ of course, includes a requirement that an impartial lawmaker could logically believe that the classification would serve a legitimate public purpose that transcends the harm to the members of the disadvantaged class. Thus, the word ‘rational’ -- for me at least -- includes elements of legitimacy and neutrality that must always characterize the performance of the sovereign's duty to govern impartially […] The Court of Appeals correctly observed that, through ignorance and prejudice, the mentally retarded ‘have been subjected to a history of unfair and often grotesque mistreatment.’ [ …] The record convinces me that [the special] permit was required because of the irrational fears of neighboring property owners, rather than for the protection of the mentally retarded persons who would reside in [the] home.” (Stevens, J., concurring). 

26 For example, Section 3(e) of the Food Stamp Act of 1964, as amended in 1971 (the so called “unrelated persons” provision), excluded from participation in the food stamp program any household containing an individual who was unrelated to any other member of the household. In United States Dept. of Agriculture v. Moreno, 413 U.S. 528 (1973), the Court held that “. . . that amendment was intended to prevent so-called ‘hippies’ and ‘hippie communes’ from participating in the food stamp program. The challenged classification clearly cannot be sustained by reference to this congressional purpose. For if the constitutional conception of ‘equal protection of the laws’ means anything, it must, at the very least, mean that a bare congressional desire to harm a politically unpopular group cannot constitute a legitimate governmental interest. As a result, ‘[a] purpose to discriminate against hippies cannot, in and of itself and without reference to [some independent] considerations in the public interest, justify the 1971 amendment.’” In fact, given the subsequent restrictions on the holding of Plyler, it is quite safe to say that in the realm of economic regulation, the only precedent that sets a real generic limit to the broad holding of Dandridge is the Court’s opinion in Moreno, which introduced a rational basis review with bite.

27 For a more elaborate argument on this issue see RONALD DWORKIN, Justice for Hedgehogs, USA, 2011, pp. 191-214, 327-399.

28 The opposite conclusion would hold if the government decides to give precedence to one sector of the economy over the other, articulating its reasoning, not in terms of preferred lifestyles, but rather in terms of economic efficiency (e.g. the subsidization of small family farms on the ground that the development of this economic activity will promote market efficiency and provide overall benefits for the economy would be constitutional).

29 Another important issue which makes the Judges feel uneasy and has produced a certain amount of judicial ambivalence turns on the question of “who decides what the purpose of a law is”. A more careful review of the established judicial practice in the U.S. reveals that the Court will usually hypothesize a legitimate legislative purpose, rather than inquiring into the legislature’s “actual motivation”. For example, in McGowan v. Maryland, 366 U.S. 420 (1961) the Court held that the equal protection clause is violated “[only] if the classification rests on grounds wholly irrelevant to the achievement of the State's objective. State legislatures are presumed to have acted within their constitutional power despite the fact that, in practice, their laws result in some inequality. A statutory discrimination will not be set aside if any state of facts reasonably may be conceived to justify it.” [emphasis added].  This view is stated more emphatically in U.S. Railroad Retirement Board v. Fritz, 449 U.S. 166 (1980), holding that “Where, as here, there are plausible reasons for Congress' action, our inquiry is at an end. It is, of course, ‘constitutionally irrelevant whether this reasoning in fact underlay the legislative decision’ because this Court has never insisted that a legislative body articulate its reasons for enacting a statute.” On this view, not only is the Court barred from reviewing the “actual purpose” of the statute, not only is the legislature not required to publicly reveal the purpose underlying its enactments but, most importantly, even if when the legislatures articulate an illegitimate purpose, the Courts should ignore the real purpose behind the challenge statute and try instead to invent normatively attractive legislative goals to uphold the law. For more recent judgments refusing to review the actual purpose of the law, see Nordlinger v. Hahn, 505 U.S. 1 (1992), Federal Communications Commission v. Beach Communications, Inc., 408 U.S. 307 (1993), Fitzgerald v. Racing Association of Central Iowa, 539 U.S. 103 (2003). For a contra holding see Allegheny Pittsburgh Coal Co. v. Webster County, 488 U.S. 226 (1989) as well as cases dealing with facially neutral statutes that have the effect of disadvantaging a racial minority. For a more moderate approach on whether a statement of legislative purpose is required see United States Railorad Retirement Board v. Fritz 449 U.S. 166 (1980) (Brennan, J., dissenting) and Schweiker v. Wilson 450 U.S. 221 (1981) (Powell, J. dissenting).

30 On that point see Dimitrios Kyritsis, Constitutional Review in Representative Democracy, Oxford Journal of Legal Studies, 32(2), 2012, p. 322; LON FULLER/KENNETH WINSTON, The Forms and Limits of Adjudication, Harvard Law Review, 92, 1978, pp. 366-367.

31 Interestingly enough, a significant number of constitutional complaints in Portugal, were filed by parliamentary minorities. This kind of action has provided the socialist opposition with an effective mechanism to overturn the austerity reforms put forward by the coalition government which was formed in 2011.

32 Even if we assume that there is one optimal answer, which can be discovered ex post by the judiciary, even then, I doubt that we must concede the power of the judiciary to correct the legislature by substituting its own economic arguments and financial estimates for that of our elected representatives. The idea of participatory democracy and collective self-government presupposes that the electorate can make mistakes and it rests with the electorate to correct itself. In other words, the participatory model of democracy turns not on making wise decisions; it instead turns precisely on whether the citizens identify with the governmental decisions as their own, even if those choices turn out (in retrospect) to be misguided.

33 It follows that the Court got the case right in Minnesota v. Clover Leaf Creamery Co., 449 U.S. 456 (1981). In this case, a Minnesota law banned the retail sail of milk in plastic nonreturnable, non-refillable containers but permitted such sale in nonreturnable paperboard milk cartons. A unanimous Court per Justice Brennan held that “[Although] parties challenging legislation under the Equal Protection Clause may introduce evidence supporting their claim that it is irrational, they cannot prevail so long as itis evident from all the considerations presented to [the legislature], and those of which we may take judicial notice, that the question is at least debatable. Where there was evidence before the legislature reasonably supporting the classification, litigants may not procure invalidation of the legislation merely by tendering evidence in court that the legislature was mistaken.” [emphasis added].

34 See generally RONALD DWORKIN, A Matter of Principle, USA 1985, pp. 33-71.

35 In Justice Stewart’s words: “[Here] we deal with state regulation in the social and economic field, not affecting freedoms guaranteed by the Bill of Rights, and claimed to violate the Fourteenth Amendment only because the regulation results in some disparity in grants of welfare payments to the largest AFDC families. For this Court to approve the invalidation of state economic or social regulation as ‘overreaching’ would be far too reminiscent of an era when the Court thought the Fourteenth Amendment gave it power to strike down state laws ‘because they may be unwise, improvident, or out of harmony with a particular school of thought.’”, Dandridge v. Williams, 397 U.S. 471 (1970).

36 Expulsion of nationals is prohibited under international law (see, for example, Article 3 of Protocol no. 4 ECHR), not least because in principle there is no obligation of other States to accept foreign nationals in their territory (and practically there are no terrae nullius in which expelled nationals could end up). Note, however, that in any case this is a fairly recent development of international law (see, for example, Vaughan Lowe, International Law, Oxford, 2007) on the gradual process of development of state boundaries).

37 Reliance on such sources was necessary due to my regrettable ignorance of the Portuguese language. The principal source of my knowledge is the recent comparative analysis published by Cristina Fasone, Constitutional Courts Facing the Euro Crisis. Italy, Portugal and Spain in a Comparative Perspective, European University Institute Max Weber Programme 2014/25. The opinion summaries included in the footnotes of the present paper have been drawn from this source and especially from pages 24-30 of Fasone’s paper.

38 In judgment 353/2012 of 5 July 2012 the Court declared unconstitutional the provisions in the Budget Act for 2012 that suspended the 13th and 14th monthly allowances for public workers (and pensioners) from 2012 to 2014 and saw it as an integral part of the salary. By applying the test outlined in its ruling of 2011, the Court found that this suspension amounted to a targeted discrimination against public workers. While in 2011 the Court linked the justification of the public salary cuts to the particular position of public employees, as their salaries are funded by taxpayers, one year later the Court argued that a non-legitimate violation of the principle of equality embedded in Article 13 Const. had occurred. The combined effect of the Budget Acts for 2011 and for 2012 led to a persisting impairment of many civil servants’ living conditions. The suspension of the allowances was indeed neither temporary, as public workers would have experienced salary cuts for the third year in a row, nor proportionate to the aims to be met. The lack of compliance with the principle of proportional equality derived, on the one hand, from the fact that some categories of public employees were exempted from the salary cuts without any clear reason, and on the other, the suspension of the 13th and the 14th monthly allowances in itself did not substantially contribute to achieving the medium-term objective.

39 In this case the Court applied what it had threatened to do in Judgment 353/2012. The Budget Act for 2013, Law no. 66-B/2012, was partially struck down on 5 April 2013 by judgment 187/2013, which jointly decided four constitutional actions brought before the Court by a variety of actors: the President of the Republic, parliamentary minorities and the ombudsman (nos. 2, 5, 8, 11/2013). The Court recognized that the following provisions were in breach of the Constitution: suspension of the holiday allowance for public workers (Article 29), for teachers and researchers (Article 31), and of the holiday allowance for pensioners (Article 77); and the duty imposed upon the beneficiaries of unemployment subsidies to pay social security contributions of 6 instead of 5 per cent. The first three articles of the Budget Act for 2013 were found in breach of the principle of equality (Article 13), since the Act again targeted the same categories of people as the previous Budget Acts and thus The forced payment of a contribution from unemployment subsidies, instead, was considered to violate the principle of proportionality: the same objectives could be achieved by less restrictive means and without damaging an already disadvantaged group of people. (The Court was also asked to review the constitutionality of other provisions of the Budget Act for 2013, such as article 27, which confirmed the wage cut for public workers for the third year, and article 45, about overtime payment, which were not eventually declared unconstitutional).

40 See above (n 10).

41 In judgment 413/2014 of 30 May 2014, Article 115 of the Budget Act for 2014 was declared unconstitutional as it required an additional sacrifice of unemployed people by again asking them – as in case 187/2013 – to pay a contribution from their unemployment subsidies. The measure was considered disproportionate as it affected a group who were already in a situation of particular vulnerability without achieving a substantive benefit in terms of public revenues. Furthermore, the Court declared the reduction of survivors’ pensions (Article 117 of the Budget Act) and public wage cuts (Article 33) to be in violation of the principle of equality. In particular, Article 33 increased the number of public workers who were subject to this increase in their salary cut – from 3.5 to 10 % of their wage – from those who earned more than 1500 euro (as had been upheld in case no. 353/2012) to those who only had an income above 675 euro. This minimum threshold was deemed too low. Given the financial impact of the judgment, however, the Court decided to exercise its power under Article 282.4 Pt. Const. and to declare the effects of the unconstitutionality of Article 33 to be ex nunc rather than ex tunc, as discussed above (n 13).

42 MIGUEL NOGUEIRA DE BRITO, Putting social rights in brackets? The Portuguese experience with welfare rights challenges in times of crisis, European Journal of Social Law, 1-2, 2014, p. 93.

43 See the case law of the ECtHR, which seems to support the reasoning that the state is different from an employer in the private sector: « la différenciation entre fonctionnaires des personnes morales et employés des entreprises privées se justifie non seulement par la nécessité de protéger le patrimoine des personnes morales, mais aussi du fait du statut spécial des fonctionnaires et du régime juridique différent qui s’appliquait aux rapports des fonctionnaires publics et des employés privés avec leurs employeurs respectifs » (Giavi v. Greece App No 25816/09 (03.10.2013) at [50]); see also “it is within the the State’s discretion to determine what benefits are to be paid to its employees out of the State budget. The State can introduce, suspend or terminate the payment of such benefits by making the appropriate legislative changes” (Kecho v Ukraine App No 63134/00 (08.11.2005) at [23]).

44 Sadly, this was roughly the reasoning of the Greek Supreme Court in similar cases regarding austerity measures: see, among others, judgment no 4741/2014 (para. 12) of the Council of State.

45 Above (nn 40, 41, 43).

46 Whenever the Portuguese Constitutional Court reviewed the austerity measures by reference to the principle of proportional equality, the net effect of its rulings was to pass judgment on what state policies would constitute a wise or balanced differentiation between the private and the public sectors, among different categories of workers and pensioners, and between different standards of social assistance across the national territory. According to the Court, whenever differential treatment between groups of people occurs, the proportional equality principle requires this treatment to be, not just rational, with regard to the state objective that is deemed to justify it; rather, the proportional equality test dictates that the sacrifices exacted from one group cannot be excessive (i.e. cannot supersede the benefits granted to another group) from the viewpoint of the targeted objective and of the treatment reserved. In this way, the finding of unequal treatment is based on a triangular relationship: a comparison between the group affected by the differentiation and others, and therefore between the two groups regarding the state objective. What is more, it is not unusual for the Court, precisely because it operates within this method of adjudication, to shift the object of review to the effectiveness of the public salary cuts to check whether they were proportionate; this effectiveness has been evaluated by the Court on the ground of the ability of the cuts to reduce the public deficit. Such an assessment, however, implies a judgment on the part of the Court about what was needed to overcome the crisis and in turn results in merging different levels of analysis: equality, proportionality and economic effectiveness. See CRISTINA FASONE, Constitutional Courts, pp. 45-48.

47 “States are not required to convince the courts of the correctness of their legislative judgments. Rather, those challenging the legislative judgment must convince the court that the legislative facts on which the classification is apparently based could not reasonably be conceived to be true by the governmental decision maker.” Minnesota v. Clover Leaf Creamery Co., 449 U.S. 456 (1981).

48 Contrast the approach of the ECtHR in situations relating to the proportionality of measures interfering with the right to property under Article 1 of Protocol 1 ECHR: “provided that the legislature chose a method that could be regarded as reasonable and suited to achieving the legitimate aim being pursued, it is not for the Court to say whether the legislation represented the best solution for dealing with the problem or whether the legislative discretion should have been exercised in another way” (Markovics and others (Decision) App No 77575/11 and others (24 June2014) at para. 39; similarly in James and others v. UK App No 8793/79 ([PC], 21 February1986) at para. 51; J.A. Pye (Oxford) Ltd v. UK App No 44302/02 (15 November2005) at para. 45; Koufaki and ADEDY v Greece (Decision) App Nos 57665/12 and 57657/12 (7 May 2013) at para. 48; Da Conceição Mateus and JanuárioApp Nos 62235/12 and 57725/12 (8 October 2013) at para. 28; Da Silva Carvalho Rico v Portugal (Decision) App No 13341/14 (1 September 2015) at para. 45.

49 STAVROS TSAKYRAKIS , Disproportionate Individualism, DIMITRY KOCHENOV/ GRAINNE DE BÚRCIA/ ANDREWWILLIAMS (eds), Europe’s Justice Deficit, Oxford, 2015, p. 245; this risk has also been identified by proponents of the proportionality analysis: see Stephen Gardbaum, A Democratic Defense of Constitutional Balancing, Law & Ethics of Human Rights, 4, 2010, p. 104.

50 For an illustration of the point that the principle of proportional equality entails a much stricter scrutiny than just checking for violations by reference to the rational basis test see above (n48).

51 Enshrined in the provision of Article 1577 of the Civil Code, as it then was, which defined marriage as a contract which could only be concluded between two persons of different sex: ‘Casamento é o contrato celebrado entre duas pessoas de sexo diferente que pretendem constituir família mediante uma plena comunhão de vida, nos termos das disposições deste Código’ (emphasis added).

52 In short, the Court held that the Constitution permitted the establishment of same-sex marriage, but it did not require so, as this was something for the legislator to decide; see Judgment No 359/2009 at [10]-[11] http://www.tribunalconstitucional.pt/tc/en/acordaos/20090359.html