By Prof. Pedro Martins

#collectivebargaining #extensions #microdata #policyevaluation #socialdialogue #EuropeanUnion #collectiveagreements #employment #wages #inequality

This note summarises the research presented in a policy workshop held last week in Brussels. The studies were conducted under the ‘Economic Analysis of Collective Bargaining Extensions’ (CoBExt) project, funded by the European Union, and focused on the cases of Greece, Italy, Portugal and Spain.

In my introduction, I presented a comparison of collective bargaining (CB) across the four countries. Despite generally low trade union density rates, particularly in the private sector, these countries exhibit very high CB coverage, precisely because of widespread and nearly automatic (explicit or implicit) extensions. The exceptions to these practices were Greece and Portugal but only during their adjustment programmes, when extensions were entirely suspended (Greece) or made conditional on representativeness criteria similar to other EU Member States (Portugal). In Greece, firm-level CB agreements were also boosted through the suspension of the favourability principle, which allowed for greater differentiation in working conditions across firms.

Jonathan Thomas discussed a number of theoretical aspects in CB, including the tension between the benefits of extensions – the promotion of a level-playing field, savings in transaction costs, lower wage inequality (and gender wage gaps), potential encouragement of investments in training and product quality – and their costs – anti-competitive practices, which can lead to restricted firm entry, accelerated firm exit and lower employment levels. He also presented ongoing research on the financial implications of extensions, in particular in cases in which large arrears arise due to time gaps between the original agreement and its extension.

Daphne Nicolitsas discussed the role of trust and cooperation in labour relations in assessing the relative merits of sectoral and firm-level CB. Using data from the European Company Survey and the Wage Dynamics Network, in work co-authored with Claudio Lucifora, she finds that sectoral CB appears more likely in contexts of lower levels of trust and higher incidence of industrial action. However, trust can compensate for the generally more limited adjustment following negative shocks in firms that operate in a sectoral CB environment.

Claudio Lucifora provided a detailed comparison of CB in the metalwork industry across the four countries. Despite the relative homogeneity of the industry and its strong international exposure, CB institutions and practices are very different across countries in terms of number of social partners, agreements and the role of firm-level CB. Moreover, the degree of ‘bite’ of the minimum wages (MW) of those agreements was generally very limited. Moreover, when focusing on the case of Italy, he finds evidence of strong responsiveness of employment to firm-level shocks, but which is entirely driven by temporary employment contracts.

Ernesto Villanueva presented the available evidence on the employment effects of CB extensions: during the last recession, a large fraction of the wage increases implied by collective contracts are linked with employment losses. He also presented research, with Efi Adamopoulou, on the metalwork industry in Spain, based on linking CB MW and anonymised individual social security records. They find that, in 2008, only 4% of workers are paid the MW according to their skill level, while 13% are paid at most 10% above the minimum. However, there was substantial mobility in total wages during the recession, and employment losses were relatively more prevalent among workers whose wages were close to the CB minima.

My research considered the case of Portugal from two perspectives. First, I contrasted the roles of the national and the CB MW on a number of firm-level outcomes, using different measures of their ‘bites’. I found generally negative effects of these two types of MW on employment, but much stronger in the case of the national MW. On the other hand, both types of MW appear to reduce firm-level wage inequality, although operating at different parts of the distribution: the national MW at the bottom and the CB MW at the top. The second perspective focused on an analysis of the levels of monopsony and its wage effects. While several local labour markets are found to display high levels of employer concentration, most workers are employed in labour markets where concentration is relatively limited. Moreover, while monopsony does have a negative impact on wages, its effects appears to be quantitatively small.

Finally, Daria Vigani analysed the effects of the new, non-representative (sometimes referred to as ‘pirate’) CB contracts in Italy. The number of these contracts has more than doubled since 2010. Using social security data, in a study co-authored with Claudio Lucifora, she finds that these contracts, while paying lower wage levels, are associated with significant employment gains.

Thank you to all participants, including Alfonso Arpaia, Matteo Duiella and Marco Cantalupi (DG-EMPL), David Rodrigues (AIMMAP) and François Rycx (ULB), for the discussion of these results.

Related research

• ‘Collective bargaining through the magnifying glass: A comparison between the Netherlands and Portugal’
• ‘No Extension without Representation? Evidence from a Natural Experiment in Collective Bargaining’
• ‘The Microeconomic Impacts of Employee Representatives: Evidence from Membership Thresholds’ 
• ‘The effect of self-employment on health: Instrumental variables analysis of longitudinal social security data’
• ‘Economic effects of overtime premium flexibility: Firm-and worker-level evidence from a law reform’