Duration of the trial period in fixed-term employment contracts
RULING No. 632/2008
Case no. 977/2008
Rapporteur: Justice Maria Lúcia Amaral
Ruling of the Constitutional Court sitting in Plenary
1. Under the terms of Article 278(1) of the Constitution of the Portuguese Republic (CRP) and Articles 51(1) and 57(1) of the Law governing the Organisation, Functioning and Process of the Constitutional Court, the President of the Republic has asked the Constitutional Court to consider whether the rule set out in Article 112(1)a of the Revised Labour Code approved by Decree of the Assembly of the Republic no. 255/X, which the Presidency of the Republic received on 5 December 2008 for enactment as a law, is in conformity with the Constitution.
The grounds for this constitutionality review request can be summarised as follows:
– The rule provided for by the aforesaid Article 112(1)a concerns the duration of the trial period in indefinite labour contracts, and significantly extends that duration in the case of unspecialised workers.
– This rule is a rule which restricts constitutional Rights, Freedoms and Guarantees, inasmuch as the longer the trial period is, the more precarious the legal/labour relationship and the more fragile the guarantee of job security.
– The rule laid down by Article 53(1) of the Constitution, which has systematically been included in the domain of ‘Workers’ Rights, Freedoms and Guarantees’, grants workers the right to “job security”, and results in the guarantee that “dismissals without fair cause” are prohibited.
– Given the nature of the right to job security and the restrictive nature of the rule in question, it is necessary to verify whether, under the terms of Article 18(2) of the Constitution, the restriction thus operated complies with the principle of proportionality, within the context of the latter’s sub-principles of appropriateness, necessity and, complementarily, reasonability.
– In the light of the corollary requirement for appropriateness, the increase in the duration of the trial period for unspecialised workers (from 90 to 180 days) raises doubts as to whether that increased duration is fit for the purpose attributed to the trial period (the assessment by the parties of their interest in pursuing the binding labour relationship). The doubts arise, first of all, from the historical/legal context of the trial period (which provides both a framework for the way in which this format evolved here in Portugal, and elements for a comparison of different laws).
– On the one hand, a worker is unlikely to need double the existing trial period in order to gauge whether the conditions which an enterprise is offering him are suitable – all the more so in that he can always terminate his labour contract without any need for fair cause, as long as he gives the required notice; on the other hand, in the case of unspecialised professions – ones in which no significant specialisation is needed – it does not seem clear that an employer objectively needs half a year to assess its workers’ professional qualities in order to enter into an indefinite labour contract.
– The rule set out in Article 112(4) does not compensate for the extension operated by the rule under analysis.
– Nor does the extension of the trial period for unspecialised workers appear to be in harmony with the value-measurement criterion of necessity that is textually imposed by Article 18(2) of the Constitution, under the terms of which the law can only restrict a constitutional Right, Guarantee and Freedom if that restriction is a requirement for, or even indispensable to, the safeguarding of other rights or interests that are protected by the Constitution.
– The extension of the trial period for unspecialised workers does not serve the interests of workers (who can always terminate their contracts by giving prior notice) and of employers (who can only terminate a labour contract without fair cause and avoid binding themselves to undesirable contracts during that period) to the same extent.
– The increase in the trial period for unspecialised workers does not seem to be a resource which employers need in order to assess the quality of those of their workers who are employed under the general scheme; this means that inasmuch as it constitutes an unfit or unnecessary measure, it is very difficult to see it as a requirement for, or indispensable to, the safeguarding of their contractual autonomy.
– If we weigh up the legal value affected by the restriction (the right to job security and to the guarantee that dismissal without fair cause is prohibited – Article 53 of the Constitution) on the one hand, and the autonomy of which the parties are deprived and the right to private enterprise on the other (Articles 26 and 61 of the Constitution respectively), we do not find the extension of the trial period for unspecialised workers to be a required measure.
– Within the overall context of the system, we can see that the Revised Labour Code approved by Decree of the Assembly of the Republic no. 255/X implies that the requirements for entering into fixed-term contracts have become more rigorous and onerous, thereby limiting the use of this form of precarious labour, which has often been excessively resorted to as a way of attenuating the alleged rigidity of the system governing the termination of indefinite labour contracts.
– The extension of the trial period as discussed above can act as a factor which compensates for the greater difficulty involved in, or more onerous requirements for, entering into fixed-term contracts.
– However, the practical concord between two – somewhat opposing – rights which results from restrictions that the legislative authorities may impose on the way in which those rights are exercised, is limited by the need for those restrictions not to subvert the purpose of their underlying legal institutions – namely the teleology of fixed-term contracts and of the trial period. In the case in question it does not seem possible for there to be a practical concord between the safeguarding of the right to private enterprise and the sacrifice of the right to job security.
– However fixed-term contracts may have been used in the past, an increased restriction on their use (as a way of bringing the system back into compliance with the applicable legal and constitutional presuppositions) cannot be legitimately compensated for in an indirect way by an increase in another form of precarious labour in the shape of a substantial extension of the trial period, in breach of the principle of job security set out in Article 53(1) of the Constitution.
– The trial period in indefinite labour contracts would thus abruptly be changed into a new way of hiring people for a six-month fixed term. On top of this, this would be a labour system whose termination would not be subject to a specific form, a right to compensation, or the requirement for fair cause. It would also be a factor that worsened the precariousness of labour arrangements, to the extent that unlike fixed-term contracts, such an arrangement could not be renewed.
– The right to private enterprise must be exercised in conformity with the public interest, and an increase in the precariousness of labour arrangement is against that interest.
– Establishing a trial period without taking account of the distinction between unspecialised and specialised workers raises doubts as to a possible breach of the principle of equality.
The President of the Republic concludes as follows:
In conclusion, there are doubts as to whether, in the light of the Constitutional Court’s jurisprudence (Ruling no. 54/91 sic), the legislative authorities’ exercise of their freedom to shape legislation in such a way as to set the trial period for unspecialised work included in the general scheme as they have done in Article 112(1)a of the Decree, exceeds the constitutional limits that the principle of proportionality imposes when it establishes a “period which is so long that it results in the distortion of the principle of job security”, because its duration is extended for too long a period, which can become a “means (…) of permitting dismissal without fair cause”.
The President of the Republic requests a review of the measure’s constitutionality in the following terms:
Given the grounds for doubts as to its constitutionality that are set out in this request, I hereby ask the Constitutional Court to consider the conformity of the rule set out in Article 112(1)a of Decree of the Assembly of the Republic no. 255/X, when applied to workers who are subject to the general scheme and engage in unspecialised labour, with the rule set out in Article 53 of the CRP, taken in conjunction with Articles 18(2) and 13(1) and (2) of the CRP.
Under the terms of Article 278(1) of the CRP, I order that the period within which the Constitutional Court must pronounce itself be reduced to fifteen days.
2. The request reached this Court on 12 December 2008 and was admitted on the same day.
3. When notified for the purpose provided for by Article 54 of the Law governing the Constitutional Court, the President of the Assembly of the Republic replied saying that he had nothing to add.
4. Now that the memorandum referred to by Article 58(2) of the Law governing the Constitutional Court has been presented and discussed, it falls to the Court to take its decision in accordance with the guidelines it established during that process.
Grounds for Decision
The question of constitutionality
5.The request which has been made to the Court is limited – as we can see from the above report – to the consideration of a single rule.
In effect, the author of the request would like the Court to pronounce itself preventively on the provisions of Article 112(1)a of the Revised Labour Code approved by Decree of the Assembly of the Republic no. 255/X.
Under the heading “Duration of trial period”, the aforesaid Article 112(1) reads as follows:
1.In indefinite labour contracts the duration of the trial period shall be as follows:
a) 180 days for workers in general.
b) 240 days for workers who occupy positions at director level or senior managerial positions.
This precept was written with a view to changing the provisions of Article 107 of the Labour Code approved by Law no. 99/2003, which is currently in force and reads as follows:
In indefinite labour contracts the duration of the trial period shall be as follows:
a) 90 days for workers in general.
b) 180 for workers who do jobs that are technically complex, entail a high degree of responsibility or presuppose a special qualification, as well as those who perform functions that require particular trust.
c) 240 days for director-level staff and senior managers.
Comparison of the two texts clearly shows that Assembly Decree no. 255/X seeks to change the existing rules governing the duration of the trial period in indefinite labour contracts, in a certain sense. According to the new text approved by the Decree, the 180-day period, which the existing Code says is only valid for contracts with workers who are especially qualified (or who do jobs that are technically complex, entail a high degree of responsibility, or presuppose functions that require special trust), would also come to apply to contracts entered into with workers who are unspecialised, or who do nor possess special qualifications. Instead of the previous provision for trial periods (in indefinite labour contracts) with three different durations – 90 days for workers in general, 180 for especially qualified workers, and 240 for those who occupy positions at director level – there would now be only two periods, and the system applicable to qualified labour would be “consumed by” the general system. This would result – and here we have the essence of the legislative change which the Decree proposes to bring about – in a doubling of the trial period in indefinite labour contracts entered into with unqualified workers, or, as they are otherwise known, workers who belong to the general scheme.
It is precisely this extension – its sense and its measure – which arouses the doubts as to its constitutionality that the author of the request has raised before the Court.
In reality, it is possible to summarise the essence of the argument before us (which often supports itself with reference to constitutional jurisprudence) in the following way. It says that any extension of the “trial” period or “experimental phase” in labour contracts (above all indefinite ones) leads to a restriction of a constitutional Right, Freedom and Guarantee – in this case the right to job security enshrined by Article 53 of the Constitution. It goes on to say that this restriction will only be lawful if it fulfils the requirements set out in Article 18 of the Constitution – especially that of compliance with the principle of proportionality included in the final part of paragraph (2). Finally, inasmuch as this principle can be broken down into the requirements for appropriateness, necessity and reasonability – the terminology used in the request – the latter’s author does not see how a 180-day “trial” period can be considered appropriate, necessary or reasonable in the case of contracts under the general scheme.
Quite apart from anything else, the author of the request contests the appropriateness of the length of the trial: he believes that neither will an unspecialised worker need “double the existing trial period in order to gauge whether the conditions an enterprise is offering him are suitable”, nor will the employer with whom he is entering into a contract need “half a year to assess (his) professional qualities” (paragraphs 13 to 24 of the request). The author then contests the need for such a period of time, because he considers that its quantum is neither a requirement, nor indispensable, for safeguarding other rights or interests protected by the Constitution – particularly the right to private economic enterprise enshrined by Article 61 of the Constitution (paragraphs 25 to 29 of the request). Finally, he also seems to contest the reasonableness of the new “trial” period, in that he concludes that “within the overall context of the system” – and also in accordance with the dual goal of ensuring practical concord between the necessary protection of job security and fulfilment of the right to free private enterprise – the restrictions (on job security) must respect the limits of the non-subversion of the “purpose of their underlying legal institutions – namely the teleology of fixed-term contracts and of the trial period” (paragraph 32 of the request).
At one point (as the initial report has already said), the author also argues that the new rules approved by the Decree will also injure the principle of equality laid down by Article 13 of the Constitution, “when conjugated” with the requirement for proportionality (paragraph 24 of the request).
6.Inasmuch as this is the question of constitutionality which the request places before the Court, it is important to its resolution to clarify whether or not the author’s arguments in relation to the reasons that have just been summarised are correct.
Given that these reasons are, apart from anything else, based on the affirmation that by its very nature, an extension of the trial period represents a restriction on the right to job security, it is important to begin by determining whether this is so. This in turn means that we must investigate the content which, according to what is already a firm body of jurisprudence, can and should be attributed to that right. We shall thus begin with this point.
It is impossible to know whether the rule that has been brought into question effectively contains – as the author of the request argues it does – a restriction on the right enshrined by Article 53 of the Constitution, without first characterising the essential elements of the trial period system, as it has been established – above all in what are currently called “indefinite labour contracts” – by the infra-constitutional legal order. Given that this is the only way to determine whether there are any points which link this system to the content of our Constitutional Law (and if so, what nature they possess), we will then proceed with this topic. Lastly, if we reach the conclusion that such links do exist, and – as the author of the request argues they do – they represent a restriction on the right in question which is capable of leading to the necessary application of the limits on restrictive rules contained in Article 18 of the Constitution, then it will be important to investigate whether, in the case before us, those limits have been complied with or not. This investigation will especially look at compliance with the requirements for proportionality, which are of particular import here.
On the guarantee of job security
7. Since the 1982 revision of the Constitution, the list of constitutional Rights, Freedoms and Guarantees has included the rights, freedoms and guarantees applicable to workers. The enshrinement by Article 53 of workers’ right to job security, with the prohibition of dismissal without fair cause or for political or ideological reasons, “opens” the constitutional catalogue of this particular type of fundamental right, whose inclusion in Title II of Part I of the Constitution means that they are systemically seen as rights to a defence, or negative rights.
In its jurisprudence, the Court has already offered enough pronouncements on the scope of the significance which the inclusion of workers’ rights in the category ‘constitutional Rights, Freedoms and Guarantees’ has for an understanding of the unity of meaning of that category. On various occasions (in this respect see, for example, Rulings nos. 372/91, 581/95 and 683/99, all of which are available for consultation at www.tribunalconstitucional.pt) the Court has said that by laying down the existential condition of person, citizen and worker as a criterion for the modulation of the possession of constitutional Rights, Guarantees and Freedoms, the Constitution makes it clear that the rights to a defence which it enshrines cannot just be seen as rights to abstentions on the part of the state or rights that only concern the state, because they also include – particularly in the Chapter on workers’ rights – rights which, inasmuch as they refer to relations between citizens, possess a special ability to “bind private persons and bodies” in the words of the final part of Article 18(1).
This “fact” cannot but be related to the principle laid down by the final part of Article 2 and by Article 9d of the Constitution. If we begin with the assertion that the term ‘workers’ as used by the Constitution certainly includes everyone who does subordinate labour for someone else – and if we are also of the opinion that “subordinate labour relationships do not take the form of true relationships between equals, unlike those established in the civil-law system of contracts” (Ruling no. 581/95) – then the Constitution’s intention to provide special protection for the existential condition of worker as a possessor of Rights, Freedoms and Guarantees can only reveal another, deeper intention – that which is included in the principle of the effective implementation of economic, social and cultural democracy and in the special state tasks derived therefrom (Article 9d). As the abovementioned Ruling no. 681/95 also says, when it protects the legal value ‘subordinate labour’ in the way it does, at the end of the day the Constitution seeks to guarantee that “the validity of (labour) contracts (depends) not only on the consent of the parties in the particular case in question, but also on the fact that that consent “has been given within a legal/normative boundary which ensures that the autonomy of one of the individuals is not subordinated to that of the other.” (C. S. Nino, Ética y Derechos Humanos, Buenos Aires, 1984, p. 178).
It is within this context that we must determine the meaning that should be attributed to the right enshrined by Article 53.
As can be gathered from what is already a firm body of jurisprudence (in addition to the abovementioned Rulings nos. 372/91, 581/95 and 683/99, also see [and again, just as examples] Rulings nos. 148/87 [Series II of Diário da República no. 178, dated 5 August 1987, p. 9674] and. 64/91, 373/91 and 306/2003 [all of which are available for consultation at www.tribunalconstitucional.pt]), the guarantee of job security, seen as a Constitutional Right, Freedom and Guarantee, or as a right to a defence, possesses a content that cannot be dissociated from the right to work enshrined by Article 58 of the Constitution. As a social right, right to state benefits or services, or right to the design and adoption of public policies that tend to promote employment, the scope of the right to work undoubtedly includes various complex, multifaceted subjective structures which this is not the right place to analyse. However, it does seem certain that they include the right to look for work, as a meritorious means of pursuing personal life projects. To this extent the legal value protected by this specific aspect of the right enshrined by Article 58 must be contiguous with the other legal value that is protected by the freedom to choose a profession enshrined by Article 47.
Now, in the Portuguese Constitution the right to look for work has a negative side or aspect – that derived from the right not to be arbitrarily deprived of a job that one has looked for and secured; and this is precisely the first aspect of the “guarantee” enshrined by Article 53. When the text of the Constitution prohibits “dismissal without fair cause or for political or ideological reasons”, what it is really seeking to do is to prevent subordinate labour relationships from being terminated by any arbitrary, discriminatory or unjustified act on the part of the employer. The right not to be arbitrarily deprived of a job that one has obtained is thus certainly the first of the subjective structures included within the scope of the protection offered by the rule set out in Article 53 of the Constitution – the first, but not the last.
The fact is – as was said in Ruling no. 372/91 – that “nothing enables us to conclude that the normative content of Article 53 of the Constitution is limited to the prohibition of dismissal without fair cause or for political or ideological reasons”. The literal formulation of the precept includes the expression job security, which is that which is guaranteed. If this guarantee were to be limited to the constitutional requirement – addressed first of all to the ordinary legislative authorities – to prohibit the abovementioned types of dismissal, we would have to conclude that the Constitutive Assembly had expressed itself here in an ostensibly superfluous manner. As such, we can say that the protective scope of the rule set out in Article 53 also includes – and cannot fail to do otherwise – another “subjective structure”, which can generically be seen as a right to the possible degree of stability of the job that one has looked for and secured. This implies that the state in general, and the legislative authorities in particular, need to avoid unjustified precarious employment situations.
8.Just what is an “unjustified precarious employment situation”, or just what is the right to its possible degree of stability, cannot be precisely determined without taking into account the right to free private economic enterprise enshrined by Article 61 of the Constitution. The thing is – and there do not appear to be any doubts about the idea – that “the necessary counterweight to labour-related rights, without which there are neither employment nor workers, is the freedom of enterprise and private initiative” (in this sense, see Bernardo Xavier, “A Constituição Portuguesa como fonte do Direito do Trabalho”, in Estudos de Direito do Trabalho em Homenagem a Manuel Afonso Olea, Coimbra, 2004, p. 165). We should add that this idea actually works both ways, inasmuch as private enterprise cannot operate without workers and jobseekers.
As the Court has repeatedly said (in this respect, see the summary in Ruling no. 187/2001, which is available at www.tribunalconstitucional.pt), the systematic inclusion of Article 61 in the Title concerning economic, social and cultural rights and duties does not prevent the right enshrined by paragraph (1) from possessing a certain aspect of a Freedom – an aspect which, since the 1997 revision of the Constitution, has been emphasised by the current text of subparagraph c) of Article 80. In effect, the respect for the legal value which the freedom of enterprise seeks to protect means that, within the context of an open society and a market economy, the production and distribution of goods and services are not things that are excluded from the activities in which private persons and bodies may engage. This in turn means that the scope of the protection offered by the rule set out in Article 61(1) not only includes the freedom to take initiatives in a certain area of economic activity, but also – and coming after that freedom – the freedom to organise and order the institutional resources needed to undertake the activity thus begun. However – and the text of the Constitution states it outright – these two freedoms must be exercised “within the overall frameworks laid down by this Constitution and the law and with regard for the general interest”.
This means that when they shape the way in which freedom of enterprise – above all as regards the second of the aspects we identified above: that of the freedom to organise the institutional resources needed to undertake a certain economic activity – is exercised, the ordinary legislative authorities must bear in mind the bounds which are imposed on them by the guarantee of job security, as defined by Article 53. This is essentially to say that here in Portugal, “(the) ownership of an enterprise does not entail being its proprietor and master, with the absolute and exclusive nature of the latter conditions, but is limited from day one: an enterprise is subject to other legal positions, rights and expectations on the part of the workers, which the employer / owner of the enterprise in question must respect in legal terms” (Bernardo Xavier, op. cit., p. 177). Among the legal positions to which an “enterprise is subject”, we immediately find those which the law must define in accordance with two injunctions that the constitutional guarantee of job security imposes on it: first, the prohibition of arbitrary dismissals; and second, the necessary avoidance of unjustified precarious employment situations.
The need to avoid unjustified precarious employment situations leads to a clear conclusion, which this Court’s jurisprudence has constantly emphasised. As we can read in Ruling no. 372/91 (among many others), “the labour contract without a fixed term (is) the type of contract which best ensures both the worker’s interests and the social goals which labour-related activities seek to attain”.
So, here in Portugal, the establishment between workers and employers of labour relations constituted by indefinite labour contracts must be the rule, and fixed-term contracts must be the exception. This is why the current law limits the possibility of entering into “fixed-term contracts”, by resorting to constricting elements that are not present in the indefinite contract scheme. As we know, apart from anything else the requirements for being able to hire people under fixed-term contracts include that there be a reason founded on the assumption that what is at stake is the need to meet non-permanent labour needs (Article 129 of the present Labour Code; Article 140 of the text approved by the Decree); the relationship which arises from such a contract must last for a minimum period of time (Article 142 of the present Code; Article 148 of the text approved by the Decree); the contract must be drawn up in a certain form (Article 131 of the present Code; Article 141 of the new text); and in the event the labour bond ends “by means of a declaration made by the employer”, the latter must compensate the worker (Article 388 of the present Code; Article 344 of the new text). On top of all this we have the limitation derived from Article 132(1) of the Code (the essence of which is retained by Article 143 of the new text), under which, as a rule, “(the) cessation, for a reason that cannot be attributed to the worker, of a fixed-term labour contract shall preclude new recruitment to the same position for a time equal to one third of the duration of the said contract, including any renewals thereof”.
While this system is undoubtedly not the only one that is possible under the Constitution, the truth is that its existence – and the essential distinction which arises from it as regards the modelling of indefinite labour contracts and the modelling of fixed-term contracts – puts the injunction derived from Article 53 of the CRP into practice. In infra-constitutional law, resorting to the precarious bond of the so-called practice of “fixed-term contracting” can only be stamped “exceptional”, given the duty on the part of the ordinary legislative authorities to avoid unjustified precarious employment situations.
The trial period in indefinite labour contracts
9.According to Article 105(1) of the current Labour Code, “(d)uring the trial period, either party may terminate the contract without prior notice or any need to claim just cause, whereupon there shall be no right to compensation unless there is written agreement to the contrary”. The essence of this formulation is retained by Article 114(1) of the version of the Code approved by Assembly Decree no. 255/X.
The fact that the law provides for a “trial” or “experimental” period – with the consequences we identified earlier – which coincides with the initial phase of the execution of the labour contract, reflects a tradition that has to some extent already taken root here in Portugal.
The first time the term “trial period” was used was by Executive Law no. 47,032 of 27 May 1966, which stated (Article 44) that “(i)n indefinite contracts there shall always be a two-month trial period, save only if something else is agreed in writing”. This provision already had consequences in relation to the rules governing the cessation of contracts: there was a presumption that when either resignation or dismissal occurred during the trial period, the former “took place on the worker’s initiative” and the latter “on the employer’s initiative”, “because of unfavourable working conditions, or lack of aptitude for the work, respectively” (Article 109). This solution, which was evidently based on constitutional assumptions that differed greatly from our own, vaguely matched a system that had been tried for the first time in 1937 (with Law no. 1952), under which a worker could be dismissed during the first two months of a contract’s duration, on the grounds of the presumption that he was inept for the work.
However, (following changes in 1975 and 1976, which it is now of no use to recall) it was the Law governing the Cessation of Labour Contracts approved by Executive Law no. 64-A/89 that laid down the post-constitutional regulations for the “experimental” period. This Law stated that, in the absence of a written agreement to the contrary, either party could rescind the contract during its initial phase, without prior notice and without any need to claim fair cause, whereupon no compensation was due. The “trial” or “experimental” period was then sixty days for indefinite contracts, and thirty days for fixed-term ones. There was also a provision that in the former type of contract, the “experimental” period for jobs with a special degree of complexity or responsibility could be extended to six months.
Over time this “tradition”, which took root in the way described above, was always accompanied by a certain justification of the need for there to be a “trial” or “experimental” period (one that coincides with the initial phase of the execution of the contract), above all in indefinite contracts. To quote current legal theory, which the Court has already repeated in Ruling no. 64/91: “(the) labour contract (format) implies the constitution of a lasting legal relationship – one that will tend to accompany the worker throughout the whole of his active professional life. Once formed, various factors make it difficult to end. So various legal systems require, or accept, that once a labour contract has been entered into, there is first an experimental period which enables the parties to reflect on both the viability of the labour situation that has been created and their own will, which will now have been clarified by a real working experience.” (António Menezes Cordeiro, Manual de Direito do Trabalho, Coimbra, 1991, p. 577). So the trial period – above all as of 1989, when the rules governing it were determined in harmony with the constitutional framework – became identified as an institution which combines four essential elements: (i) it forms a natural part of the contract, given that it must correspond to the initial phase of the latter’s execution, the only exception being the existence of a written agreement to the contrary; (ii) it is limited in time; (iii) while it is in effect, either party can end the bond between them without giving a reason for doing so, without (as a general rule) prior notice, and without any compensation or indemnity; and (iv) its purpose is solely that of experimentation. The “trial” period exists so that the parties can determine – against the background of a legal/labour relationship that they have already experienced – whether their earlier idea as to the suitability of entering into the contract matches the actual conditions under which the labour is in fact being provided.
It is this institution, with these characteristics and purposes, which continues to be regulated by the Labour Code – both in the formulation which is currently in force, and in that which is the object of consideration herein. However, and despite the institution’s constant overall pattern, that which has substantially changed – and has done so in the sense of a “constantly expanding movement” (in this sense, see Júlio Gomes, “Do uso e abuso do período experimental”, in Revista de Direito e de Estudos Sociais, Year XXXXI, no. 1/2, 2000, p. 49) – is the element ‘temporal duration of the experiment’. In fact, from the initial period of 60 days for workers in general (with the possibility of extension to six months for jobs with a special degree of technical complexity or responsibility) provided for by the 1989 Law, we have gone (via an intermediate change made in 1991) to the solution that is in force at present. As we have already said, and according to Article 107 of the 2003 Labour Code, the trial period is currently: 90 days for workers in general; 180 days for workers with special qualifications (who do jobs that are technically complex or possess a high degree of responsibility, or whose functions presuppose special relationships of trust); and 240 days for director-level staff and senior managers. It is also this general movement towards extension that is now before us. As we know, the new text approved by the Decree proposes to increase the “trial” period for workers who are unqualified or who come under the so-called “general scheme”.
10. As the Court has already said (Ruling no. 64/91), the existence of an experimental period configured in this way and with these essential elements cannot be criticised in constitutional terms. The need for a “trial”, with all the ensuing consequences, above all for the worker, falls within the teleology of the indefinite labour contract – the contract which, as we already know, is the “paradigm” for legal/labour relationships that are shaped in full accordance with the constitutional value ‘job security’. It is legitimate to believe that such long and lasting relationships need – in principle for the benefit of both parties – a breathing space, which is furnished by the legal provision for a period of time during which both the worker and the employer can freely release themselves from a commitment which, to their minds, cannot be expected to be viable. However, there is in this respect a problem that we cannot fail to consider.
In reality, one of the most widely discussed issues concerning the trial period format is whether its purpose truly includes a symmetrical idea of “bilaterality”. We said earlier that both parties to the contract possess an interest in there being a “trial” or experiment: it is in the worker’s interest to get to know the environment in which he is going to work, and in being able to establish a prognosis as to the viability of continuing a relationship that will last for a long time; it is in the employer’s interest to get to know the worker’s personal and technical aptitudes and characteristics, so that it can also predict the potential of a binding relationship that will prolong itself in time. However, it is clear that inasmuch as in our law the two parties do not possess the same abilities to choose to end the bond once the trial period is over – the worker can do so at any time by giving prior notice, without any need for fair cause (Article 447 of existing Labour Code; Article 400 of the new version approved by the Decree), whereas the employer can only do so under the terms of Articles 396 to 439 of the Code (Articles 338 to 392 in the new text) – the existence of a trial period becomes of special interest to the latter. To this extent it is possible to argue that any increase in the duration of the trial period will lead to a benefit for the employer and to a matching “compression” of the worker’s interests.
It thus seems clear (and Ruling no. 64/91 says the same thing) that such a period must be regulated by law. For reasons which concern the defence of the worker’s interests – and for reasons derived from the constitutional principle that employment cannot be unjustifiably precarious – the legislative authorities must obviously impose a maximum on the duration of the “experiment”. In principle, those authorities possess the freedom to shape the quantum of the “trial”, but not the freedom not to shape it.
Having said this, this conclusion does not resolve another problem: that of whether this freedom to shape legislation – with reference solely to the choice of the concrete duration of the trial period – should not also be limited. The answer, which has already been given in Ruling no. 64/91, appears to be unquestionably “yes”. As the latter Ruling says, the duration of the trial period “cannot be made so long that it results in the distortion of the principle of job security – an undeniable consequence in cases in which this duration were to be so long (two or three years, for example) that we would have to consider ourselves in the presence of a fraudulent decision – a disguised means of permitting dismissal without fair cause”.
If we look at things in this way, there does not seem to be any further doubt as to the potentially restrictive (and what is more, restrictive of the Right, Freedom and Guarantee enshrined by Article 53 of the Constitution) nature of legal measures which extend the duration of the trial period. The fact is that if we see the restriction of a right as being everything involved in a “state action or omission which, by eliminating, reducing, compressing, or posing difficulties for, the possibilities of gaining access to the legal value protected by constitutional law […], or by weakening the duties and obligations in the broad sense of the terms which that legal value imposes on the state, disadvantageously affects the content of a fundamental right” (Jorge Reis Novais, As Restrições aos Direitos Fundamentais não expressamente autorizadas pela Constituição, Coimbra, 2003, p. 247), then, and in the light of everything we have said so far, it is easy to argue that the abovementioned measures make it difficult to gain access to the value protected by constitutional law (job security) and weaken the duties which it imposes on the state.
It remains for us to determine – because this is another question – whether, in the case before us, the extension brought about by the rule set out in Article 112(1)a (in the version approved by the Assembly Decree) constitutes a constitutionally unlawful restriction because it implies a breach of the limits on the limits on rights enshrined by Article 18 of the Constitution.
As we know, the problem particularly concerns that special limit on such restrictions which is derived from the necessary observance of the principle of proportionality.
The principle of proportionality
11.Just what the precise content of the proportionality which is textually referred to by the final part of Article 18(2) of the Constitution is, has been adequately covered by the Court’s jurisprudence.
As the Court said in Ruling no. 634/93 (also referred to in Ruling no. 187/2001) for example, the idea of proportion, or of prohibition of excess – which, in a democratic state subject to the rule of law governs the actions of every public authority – fundamentally concerns the need for there to be a balanced relationship between means and ends: the actions which the state undertakes in order to achieve its ends must not employ means whose weight entails excessive (and thus unbalanced) burdens for the persons and bodies at which they are targeted. However, to say this is not to say much. As we can read in Ruling no. 187/2001 (which develops Ruling no. 634/93):
the principle of proportionality can be broken down into three sub-principles: the principle of appropriateness (measures that restrict constitutional Rights, Freedoms and Guarantees must prove to be a means of pursuing the desired ends, while safeguarding other rights or values protected by the Constitution); the principle of requirement (these restrictive measures must be required in order to achieve the desired ends, because the legislative authorities do not dispose of other, less restrictive, means by which to achieve the same goal); and the principle of fair measure, or proportionality in the strict sense of the term (the legislative authorities cannot adopt measures that are excessive or disproportionate in order to achieve the desired ends).
For now, we only need to add three clarifications to this general definition of the three sub-principles (into which it is possible to break the principle of proportionality down for analysis). The first concerns the exact content that must be given to the third test listed above, which jurisprudence and legal theory commonly call proportionality in the strict sense of the term or the criterion of fair measure. That which is really measured here is the relationship that exists in concrete between the coactive burden derived from the measure which has been adopted, and the specific weight of the public-interest gain which the provision is designed to achieve. Or, as the Court also said in Ruling no. 187/2001, “(t)he issue here…is the requirement that the restrictive or damaging effects of the intervention achieve a “calibrated” (fair measure) relationship with the desired ends – something that demands a weighing up and graduation of, and a match between, the possible effects and measures”.
The second clarification we need to add involves the logical order in which the three sub-principles should be applied. They should relate to one another in accordance with a rule in which the most abstract one takes precedence over the most concrete one, or over the one that is closest (because of its content) to the necessary assessment of the specific circumstances of the real-life case under consideration. What this means exactly is that the test of proportionality logically begins with the use of the sub-principle of appropriateness. Under the latter we must only gauge whether a certain means is, in abstract terms and as a typical means, fit or apt to achieve a certain end. The formulation of a negative judgement as to the appropriateness of the means logically prejudices the need to apply the other tests. However, if we do not conclude that the means is typically inappropriate to the end, we must then turn to the examination of the extent to which it is required, which is also known as the need to choose the most benign means. This is a “finer” examination, or one that is closer to the specificities of the concrete case in question: it assesses whether or not, in this life situation, there are various (equally fit) possibilities for achieving the desired end, in such a way as to determine whether, in the case at hand, the possibility which is most benign or least onerous for private individuals has been chosen, as it should have been. If we reach the conclusion that this is not the case – always a possibility, given that there may be measures which, albeit deemed appropriate, nonetheless do not prove necessary or required – then the inevitability of moving on to the last test of proportionality is logically prejudiced.
The third clarification we need to add is related to the particular aspect that must be subjected to the judgement of proportionality (now in its broad sense), when applied to the decisions taken by the legislative authorities. We said earlier that in a democratic state subject to the rule of law, the principle in question applies to the actions of every public authority. This means that it will apply as much to the acts of the administrative function as to those of the legislative one, in that the state (acting via its different authorities) can never, in either case, employ means which prove unnecessary or not “proportionate” in the light of the ends it wishes to pursue. What is certain, however, is that the legislative authorities are distinguished from their administrative counterparts precisely by the freedom they possess to, within the framework of the Constitution, select the goals that will guide their choices – indeed, this is what we are referring to when we talk about the legislative authorities’ freedom to shape legislation. This in turn means that any judgement that a certain legislative measure is invalid on the grounds that it fails any of the tests which make up the test of proportionality must always – as the Court said in Ruling no. 187/2001 – be based on a manifest failure by the legislative authorities to fulfil the duties imposed on them by the constitutional principle of the prohibition of excess.
12.The author of the request argues that the extension of the trial period in indefinite labour contracts involving unspecialised workers from 90 to 180 days is, quite apart from anything else, a legislative measure which is inappropriate to achieving the end it seeks to achieve – an end which is consubstantiated in the purpose of the “trial” or “experiment” itself. The trial period exists so that the parties can find out, from experience, whether or not that which they have contracted for is appropriate to that which actually exists; as such – says the author of the request – a “trial” which lasts six months (exactly double the duration to date) is not a typically fit, or apt, means for an unspecialised worker to get to know the environment in which he works, or for an employer to get to know the person he employs.
However, we should note that proof is needed to demonstrate that, considered abstractly and in the light of its typical content, a certain legislative measure is an instrument that is unfit, or not apt, for achieving the desired end. At the end of the day what is needed is evidence that the means so used in itself proves to be innocuous, indifferent or even negative when it comes to achieving more or less the desired effects (in this sense, see Jorge Reis Novais, Os Princípios Constitucionais Estruturantes da República Portuguesa, Coimbra, 2004, p. 168). Now – and we say so straight away – none of the arguments advanced by the author of the request appears to support the existence of such proof.
In the first place, the argument based on comparative law does not support it.
When it analyses the “historical/legal context of the trial period in the light of Portuguese and European Law”, the request says that the duration of the trial period in Portugal is “on the same level as the longest ones in the legal orders that are traditionally close to our own”, and goes on to list the durations in Spain, France and Italy in this respect.
However, the comparative argument should be used with caution. In order to be fruitful, a comparison cannot be limited to an analysis of an institution taken in isolation, without taking into account the broader regulatory system of which it forms a part. One can only compare that which is comparable; and the truth is that, because they are incorporated into system contexts that differ from our own, the formats which the trial period adopts in other legal systems – namely those mentioned above – can acquire a meaning and an importance that do not legitimate automatic transposabilities to the “interpretation” of the Portuguese system.
This appears to be the case with France and Italy, for example. There – where in fact, contrary to what happens here in Portugal, there can only be a trial period if the parties agree to it in writing, and it is therefore not a natural element of the contract – the possibility of entering into a fixed-term contract (while nonetheless retaining its exceptional nature) is less restricted in terms of the causes that justify doing so than is the case in Portugal. According to Legislative Decree no. 368/2001 of 6 September 2001, in Italy fixed-term contracts can only be entered into “for reasons of a technical, production-related, organisational, or substitutive nature” (in this sense, see F. del Giudice/F. Mariani/F. Izzo, Diritto del Lavoro, 13th edition, Naples, 2006, p. 344); and in France, the new recruitment contract format (albeit only applicable to enterprises with less than 20 workers) governed by Ordinance no. 2005-893 of 2 August 2005 lies (according to the legal theorists) “halfway between the indefinite contract and the fixed-term contract (…) and consists in freeing oneself for two years from the rules governing the unilateral termination of a labour contract” (Laurence Boutitie, “Le contrat nouvelles embauches: un contrat à durée indéterminée précaire”, in Recueil Dalloz, no. 40, 2005, p. 2777).
Given this backcloth, it would seem logical to conclude that the trial period systems in these countries possess outlines that differ from our own. Here in Portugal, the more demanding requirements for the way in which fixed-term contracts are modulated (see point 8 above) mean that the provision for a “trial” period in indefinite contracts takes on a meaning which, from the systemic point of view, does not possess an equivalent in the legal systems mentioned above. What is more, in those systems the duration of the trial period – in indefinite contracts – does not seem to be that different from the current rules here in Portugal (In Italy the patto di prova [trial period] lasts a maximum of 6 months [Article 10 of Law no. 604/1966]. In France the law provides for périodes d’essai of differing lengths: 2 months for unspecialised workers, renewable up to 4 months; 3 months for specialised workers, renewable up to 6 months; and 4 months for “managers”, renewable up to 8 months; in all three cases renewal depends on the existence of a collective labour agreement [Article L 1221-19 of the Labour Code]). The system which appears to be closest to ours is thus that laid down by Spanish Law, under which in any case there can only be a trial period if there is a written agreement to that effect between the parties.
For all these reasons, it does not seem that resorting to comparative law furnishes arguments that prove the inappropriateness of the legislative measure before us in relation to the end it is designed to pursue; and it is also true that we cannot see any other reasons which would conclusively offer the same kind of proof. Nothing would appear to legitimate the conclusion that the solution set out in Article 112(1)a of the revised version of the Code is, in its own right, a typical unfit or inapt means of achieving the desired end. To say that, after a certain time, a “trial” period ceases to serve as a means for the “parties” to a contract to “get to know one another” – thereby turning (and merely by the passage of time) into an instrument which is abstractly inappropriate to the achievement of such an end – would be to sustain an argument which is based on premises that remain impossible to demonstrate. The Court is therefore unable to uphold the author of the request’s position in relation to this point.
13.It can happen that a legislative measure which is not, in itself, inappropriate to the pursuit of a certain end comes to prove itself something which is unnecessary or which cannot be required in this respect. As we said earlier, “passing” the first test of proportionality does not render the examination operated by the second test useless or superfluous.
It is precisely the use of this second test – which we shall call a “measure of the value ‘need’” – to which we shall now turn our attention. In this we are now accompanied by the idea of a practical concord: the request says that the measure of the value ‘need’ must be gauged in terms of that which is indispensable, or required, in order to safeguard other interests or values that are protected by the Constitution. In the case before us (the request goes on to say) the other value which the legislation wants to safeguard, and which must be in practical concord with the value ‘job security’, is that of free private enterprise which, as we have seen, is enshrined by Article 61 of the Constitution.
This is to say that the requirement that a measure which restricts a right must, in a balanced way, fulfil the imperative need for practical concord between conflicting (and equally important in constitutional terms) values or interests is not only valid when it comes to getting a clearer idea of the content of the second test of proportionality. On the contrary. Inasmuch as at the end of the day it contains the embodiment of the very idea of proportionality, the inevitability of its presence makes itself felt right across the board in the application of all the sub-principles included in the constitutional value in question. We can deduce exactly this from the final part of Article 18(2) of the Constitution, which states that restrictions must be limited to that which is necessary to safeguard other rights or interests protected by the Constitution.
Having said this, there can be no doubt that the test of necessity or requirement itself requires a specific form of weighing up, or assessing, the way in which the legislative restriction of a right operates the necessary undertaking of the task of achieving practical concord between conflicting values or interests. We have already seen what this specificity is reflected in. The issue here is to gauge, in the case before us: whether there were alternative means of achieving the same end; whether there are differences in the extent to which those means are onerous for the targets of the restrictive measures; and lastly whether, from among the different means, the legislative authorities chose the most benign or least onerous one.
To this end, at this point – contrary to the procedure used to examine appropriateness – we need to descend to an analysis of the specificities of the concrete case in question. It is now important to investigate the way in which, in a certain life situation, the legislative authorities ‘weighed up’ different values that are protected by the Constitution and conflict with one another, in such a way as to restrict the right which one of them protects, to the benefit of another, which is also under constitutional protection. So it is not without reason that in this step we must verify the way in which the rule that has been questioned operates the practical concord between the right to job security on the one hand and the right to private economic enterprise on the other. The thing is that, as we have already seen (point 10 above), any extension in the length of the trial period can, in principle, work to the benefit of employers (and thus to the benefit of the legal value protected by the right enshrined by Article 61 of the CRP), with the ensuing compression of workers’ “interests”, and thus of the value specifically governed by Article 53.
However – and this is the fundamental point – if this is, in principle, the case with any extension, the truth is that in the case before us the increase in the duration of the trial period is not just any increase. The effect of the end which the change in legislation seeks to achieve is the application of an increase in the trial period to unspecialised workers only – an increase which doubles that provided for by the existing system, and which means that the duration of the trial period imposed on them matches – exactly – the duration applicable to specialised workers. As we know, the period laid down by the contested rule is the same as that set out in Article 107(1)b of the existing Labour Code, which applies solely to workers “who do jobs that are technically complex, entail a high degree of responsibility or presuppose a special qualification, as well as those who perform functions that require particular trust”. Given that the special reference to an “experimental” time applicable only to contracts entered into with the latter category of worker has disappeared in the new text approved by the Decree – and given that the trial period which used to be targeted at them now applies to “workers in general” – there is apparently no other possible conclusion. For unspecialised workers, the duration of the trial period is not only doubled, but extended to a point at which it matches the “experimental” time applicable to qualified workers.
Consequently, the question that poses itself is whether such an increase (in the trial period) is necessary, or required, for the end which the legislative measure seeks to achieve – and which cannot be other than to ensure a duration of the “trial” or “experiment” that is suitable for both parties in labour contracts. To this end, is it necessary, or can it be required, that in indefinite labour contracts unspecialised workers come to have a trial period that is not only equal to double the existing time, but also equal to the duration of the “trial” to which specialised workers are subject? An immediate answer in terms of empirical reasons is “no”. By the very nature of things, a “trial” must be less demanding to the same extent as the content of the work that is to be provided is less demanding in terms of its “quality” or “specialisation”: less qualified work can be “tested” or “experimented with” (by both the employer and the worker) in less time than that needed for other, more qualified, work.
What is more, it tends to be precisely workers without special qualifications, or without a special technical autonomy, who are most in need of the protection of the Law, because it is also they – due to their ‘replaceability’ – who are potentially most exposed to unjustified precarious employment situations. So any legislative measure which is aimed at them and which serves to increase the precariousness of their labour situation cannot but lead to a significant “compression” of the legal value protected by Article 53 of the Constitution.
In the case before us there is nothing in the legislative precedents that enables us on the one hand to identify a manifest insufficiency of the time periods which are currently in force, or on the other to find a justification for the extension of the trial period applicable to the labour contracts of unspecialised workers from 90 to 180 days. Neither the Green Paper on Labour Relations (April 2006, in Portuguese), nor the White Paper on Labour Relations (December 2007, in Portuguese) say anything about this point, nor does the Justification of Reasons in the Draft Government Bill presented to the Assembly of the Republic. At the same time, during the parliamentary debates on the Bill this question was emphatically posed, but not conclusively answered (on this point, see the debate on the general principles published in Series I of Diário da Assembleia da República (DAR) no. 2, dated 19 September 2008, and the debate on the details published in DAR no. 18, dated 8 November 2008. On both occasions the Assembly discussed the possibility that this “legislative measure” would cloud the boundaries between the fixed-term labour format in its minimum normal duration, and the trial period in indefinite labour contracts.
So, in the absence of any clear insufficiency on the part of the existing periods, and given that in the precedents mentioned above we are unable to find any justification for the change which the Decree seeks to bring about, everything indicates that the measure which contains the rule that has been questioned – to the extent that it refers to unspecialised workers – is not compatible with the test of that which is necessary or required to which all infra-constitutional rules which restrict fundamental rights are subject.
The fact is that any marginal increase in efficacy which might be derived from an extension of the trial period does not possess, in its own right, the potential needed to justify that that extension be from 90 to 180 days for unspecialised workers, who would thus be placed on the same footing as their specialised counterparts. For this reason we must conclude that the legislative authorities did not protect unspecialised workers from unjustified precarious employment situations as they should have done in the light of the provisions of Articles 53 and 18(2) of the Constitution.
For the reasons set out above, the Court hereby holds unconstitutional the rule set out in Article 112(1)a of the Revised Labour Code approved by Decree of the Assembly of the Republic no. 255/X, when applied to workers who do unspecialised work, because it is in breach of the provisions of Articles 53 and 18(2) of the Constitution.
Lisbon, 23 December 2008
Maria Lúcia Amaral
Maria João Antunes
Carlos Pamplona de Oliveira
João Cura Mariano
José Borges Soeiro
Joaquim de Sousa Ribeiro
Mário José de Araújo Torres
Carlos Fernandes Cadilha
Rui Manuel Moura Ramos
Justice Ana Guerra Martins is in agreement with this Decision, although her absence has prevented her from signing it.
Maria Lúcia Amaral