Collective bargaining in Portugal

Collective agreements are published in the MTSSS’s official bulletin and are legally binding. There are no collective agreements on wages in public administration. The peak of the economic and social crisis in Portugal, in 2011–2013, combined with the introduction of stricter criteria to extend the agreements (2012 and 2014) that were implemented under the MoU (2011–2014) resulted in a collapse of collective bargaining at all levels – although the greatest impact was felt in sector-level multi-employer bargaining – while the intended decentralisation did not take place. Although the number of company agreements as a percentage of total agreements increased, this was more the result of the fall in sector-level agreements than of an increase in company bargaining. In 2016, company agreements represented 39.6% of all collective agreements signed, while in 2018 they represented 44.5% and in 2021 they represented 52%. However, between 2016 and 2021, the proportion of workers covered by the company agreements in force represented only 3.5% to 5.0% of collective bargaining coverage (DGERT/MTSSS, 2022; GEP/MTSSS, 2022d; 2023).

Wage bargaining coverage

Overall, the coverage of the collective agreements in force has slightly declined over the years. The most recent edition (2021) of the government survey Quadros de Pessoal (Personal Records) found that the total coverage of all legally existing agreements for the whole economy (with the exception of public administration) was 86.2% compared with 91.4% in 2011. This ‘accumulated coverage rate’ includes a number of agreements that have not been reviewed for many years (DGERT/MTSSS, 2022; GEP/MTSSS, 2022d).

Following the legal changes to the collective bargaining legal framework as required by the European Commission, the European Central Bank and the International Monetary Fund (which together are known as the Troika), there was a dramatic decline in the level of renewals of collective agreements. Coverage dropped from around 1,242,000 workers in 2011, that is, around 53% of all employees covered by agreements in the market sector, to unprecedented levels of 19% in 2012 and around 11% in 2013 and 2014. Since 2015, a slow process of recovery has been underway, with an increase seen in the proportion of workers covered by collective agreements that were reviewed or newly published: from 568,000 (25%) in 2015 to 994,000 (40%) in 2018 and 883,000 (35%) in 2019. Nevertheless, this coverage did not reach the levels observed before the global economic crisis, when 1,895,000 employees were covered by renewals of collective agreements, representing 65% of all workers covered by agreements (Campos Lima, 2017, 2019). The number of employees covered by renewals of collective agreements dropped sharply in 2020, the first year of the pandemic, to 488,000. In 2021, there was a slight recovery: 636,000 workers were covered by renewals of agreements, that is, 26% of the collective bargaining coverage. With this being the situation on the eve of the escalation of inflation, it put additional pressure on collective bargaining, leading to an increase in the number of workers covered by collective agreement updates, reaching around 711,000 in 2022 (DGERT/MTSSS, 2022, 2023; GEP/MTSSS, 2022).

Collective wage bargaining coverage of employees from different sources

Notes: * Percentage of employees working in local units where more than 50% of the employees are covered under a collective pay agreement against the total number of employees who participated in the survey.

Sources: Eurofound, European Company Survey 2013 and 2019 (including private sector companies with establishments with >10 employees (Nomenclature of Economic Activities (NACE) codes B–S), with multiple answers possible); Eurostat [earn_ses10_01], [earn_ses14_01], [earn_ses18_01], Structure of Earnings Survey 2010, 2014 and 2018 (including companies with >10 employees (NACE codes B–S, excluding O), with a single answer for each local unit). For more information on the methodology, see DGERT/MTSSS (2012–2023) and OECD (2021).

Collective bargaining coverage – national data, 2011–2021

Since the creation of the Portuguese collective bargaining system in the 1970s–1980s, the most important level of bargaining by far has been sector or branch level. The agreements at this level cover more than 90% of the total workforce that can potentially be covered by all levels of collective bargaining. There are no collective agreements at cross-sector level. Bargaining at company level is important in some sectors (such as in public utilities). In practice, there is no decentralisation in collective bargaining in Portugal.

Levels of collective bargaining, 2022

Articulation

Until 2009, company agreements could be signed only by trade unions. Since 2009, company bargaining has also been able to be conducted by non-union bodies in companies with more than 500 employees and, since 2012, in companies with more than 150 employees, although still under the delegation of trade unions (Labour Code, Article 491(3), with the changes introduced by Law 23/2012). However, agreements of this particular type have not been signed since the possibility was established, as documented in the green paper on labour relations of 2016 (Dray, 2017). Moreover, there is no significant articulation between bargaining levels, as far as company agreements are not subordinated to the framework of sector agreements. For instance, it is possible that the trade union that signs a given company agreement is not the same trade union that signed the sector agreement in force. Competition between unions (affiliated to the CGTP-IN, the UGT or an independent union) and expiry of sector collective agreements contribute to that possibility. On the other hand, Law 23/2012 made it possible for collective agreements to include clauses of articulation between levels, but very few agreements signed since then have included such a type of clause (CRL/MTSSS, 2016, 2017, 2018, 2019, 2020, 2021).

The period between the signature of an agreement and its publication in the official bulletin of the MTSSS (Boletim do Trabalho e Emprego, BTE) may vary from a few weeks to a few months.

There are large variations in the duration of bargaining rounds between sectors and years. This is another source of difficulty in determining the time of year when bargaining rounds usually take place.

The present crisis in bargaining and the resulting low numbers of agreements do not allow significant patterns of timing to be identified, as was possible to do before the crisis.

In general, bargaining rounds take place on a yearly basis, in connection with wage bargaining.

Trade union confederations follow up on sector bargaining and provide some guidance, but it is at the sector or federation level that coordination with lower-level units takes place. High-level employers implicitly coordinate changes among their lower-level affiliates. During the first decade of tripartite concertation at macro level (1987–1997), wage bargaining coordination took the form of tripartite macro agreements on income policies (Campos Lima and Naumann, 2011). Non-binding tripartite agreements on the increase of the national minimum wage may have an influence on collective bargaining outcomes because they put pressure on the lower wages of the existing wage tables. This occurred during 2008–2010 and has also been occurring more recently, since 2015, because the national minimum wage after the increases exceeded the level of the lowest wage groups of many collective agreements. Recent studies highlighted this link between minimum wage policies and collective bargaining developments (GEP/MTSSS, 2019b; Martins, 2019; Campos Lima et al, 2021).

Collective agreements can be extended by a decree issued by the MTSSS. Until the economic crisis, this was a pervasive practice in many sectors. The 2011 MoU required that the extension of collective agreements should be based on representativeness, both of trade unions and of employer associations. The legal changes in 2012 and in 2014 referred only to employer representativeness/representation. In the 2012 version, they had to represent 50% of employment in the sector, which in many sectors is an impossible target. In the 2014 version, 30% of their membership had to be made up of micro, small and medium-sized enterprises for them to be allowed to extend the collective agreements.

These rules were withdrawn in 2017 for a number of reasons: the negative impact on collective bargaining as a result of reducing the number of extensions and the number of updated collective agreements, as well as their coverage; the weakness of employer associations as highlighted by the green paper on labour relations of 2016 (Dray, 2017), with only 19% of the companies in Portugal in 2014 claiming to be affiliated to employer associations; and the fact that both employer confederations and trade union confederations were opposed to or reticent about the criteria for extension based on representativeness/representation. In May 2017, Resolution 82/2017 replaced the criteria of representativeness/representation of employer associations with new criteria for the extension of collective agreements: the effect on the wage bill and economic impacts, the level of the wage increase, the impact on the wage scale and on the reduction of inequality, the percentage of workers to be covered (in total and by gender) and the proportion of women that will benefit.

Law 23/2012 established the possibility of ‘open clauses’ by allowing collective agreements to specify that rules on geographical mobility, working time and wages can be set by agreements at another level (as required by the Troika MoU). However, no cases have been reported of agreements including this type of clause (CRL/MTSSS, 2016; GEP/MTSSS, 2017).

Derogations in a strict sense were not possible until recently, but this changed in August 2014. The seventh revision of the Labour Code introduced the possibility of temporarily suspending collective agreements in the case of a severe crisis that ‘gravely effects the normal activity of the company’. The suspension is possible only if the employer organisation(s) and trade union(s) sign a written agreement for that purpose.

The 2003 Labour Code introduced mechanisms to speed up the termination of collective agreements and reduce their period of validity after expiring. In addition, the 2009 Labour Code facilitated unilateral ‘caducity’ of collective agreements and reduced their survival period. In line with Troika MoU requirements, legislation in 2014 further reduced the period of validity and survival period. Expiration becomes effective if one of the signing parties officially ‘denounces’ the agreement, thus triggering the process of caducity. This process takes at least 14 months (starting on the date of denunciation) for the agreement to be effectively cancelled. The employees who were covered by the agreement before its caducity individually keep a set of rights stipulated in the agreement, such as their remuneration, category/function, working time and social protection. The employer associations have been the main proponents of the unilateral requirements for the caducity of agreements. The legal measures have undermined union power and the quality and balance of collective bargaining in a number of sectors.

The tripartite commitment for a mid-term concertation agreement signed in January 2017 comprised a bipartite agreement between employer confederations and trade union confederations to suspend any requests for caducity for 18 months. While this initiative might have helped the recovery of collective bargaining, it was not a long-term solution, as the measure was temporary. The CGTP-IN has been calling for revision of the legal framework in order to re-establish the principle that a collective agreement can expire only when both signatory parties agree to do so, a principle that has been entrenched in the collective bargaining system since 2003.

Law 93/2019 amending the Labour Code, following the 2018 tripartite agreement, did not reverse the possibility of unilateral caducity of collective agreements. However, it introduced various mitigation measures, such as an obligation for the written request for the expiry of a collective agreement to state the economic and structural reasons or maladjustment of the collective agreement (Article 500), a strengthening of the role of the CES in relation to arbitration and mediation to prevent the expiry of agreements (Article 501(A)), and an extension of the range of rights that workers keep when collective agreements expire including occupational health and safety and parental rights (Article 501(8)). On the other hand, this law added the possibility of the expiry of collective agreements in the case of the extinction a signatory organisation, trade union or employer association (Article 502(1)(b)), a measure which generated controversy and is under examination by the Constitutional Court – arguing that this provision undermines the constitutional rights of collective bargaining, the left-wing parties BE, PCP and PEV, on 23 September 2019, requested that the Constitutional Court examine this provision. This examination has not yet been concluded.

In March 2021, in response to the collective bargaining crisis in 2020, in the context of the pandemic, the government suspended the deadlines concerning the expiry of collective agreements for a period of 24 months (Law 11/2021) to prevent the erosion of collective bargaining coverage (Eurofound, 2022).

According to the Labour Code (Article 542), collective agreements can include rules on peace clauses that determine industrial peace during the validity of a collective agreement, but peace clauses in collective agreements are extremely rare.

Collective agreements encompass a large number of issues but, in the last few years, collective bargaining has prioritised wage bargaining and working time flexibility. Regulations on working time accounts (banco de horas) and other forms of working time flexibility have been addressed in a number of collective agreements (CRL/MTSSS, 2016, 2017, 2018, 2019). Recent reports also highlighted provisions concerning the challenges of the digital economy (Ramalho, 2019).