F. CoBExt

Economic Analysis of Collective Bargaining Extensions (CoBExt)

Research project funded by the European Union (VS/2016/0340), 2017-2019


Executive summary - Key results

Collective bargaining (CB) and its extensions are key aspects of employment relations in many European Union Member States (MS). Extensions, issued explicitly by governments or representing legal benchmarks within sectors, widen the provisions of CB (including minimum wages by occupation) to a potentially large number of firms and workers that are not members of the employer associations and unions that conducted the bargaining. Extensions may also influence the membership rates of the latter organisations, as well as their incentives to conduct CB.

In a theoretical contribution [Paper 8], we extended a leading model that explains extensions and their effects to make it more realistic and consistent with existing data. The model finds that, in some cases, extensions can have an important effect in increasing wages and profits but this will come at the expense of consumers, which will be faced with higher prices.

We conducted a counterfactual evaluation of the impact of extensions [Paper 1], by exploiting the sudden suspension of extensions in Portugal during its adjustment programme. We compared the employment outcomes of industries which had their collective agreements extended and of those which concluded their bargaining a little later and therefore did not have their agreements extended. We found that the latter experience higher levels of employment growth, particularly amongst non-affiliated firms. These reforms, similar to those implemented in Greece, shaped collective bargaining along the model which has been adopted in other EU countries, including the Netherlands [paper 3], for some time.

Extensions are almost always based on sectoral collective bargaining. Their influence can therefore vary when firms make greater use of firm-level bargaining. We found that firms’ preferences regarding these two forms of bargaining is influenced by their implicit cost, including the existing levels of trust between employers and employees [Paper 11]. If trust is low, firms are more likely to implement sectoral agreements. These agreements also involve a lower probability of downward wage adjustments following negative economic shocks. Similarly, we also found [Paper 13] that wage rigidity to demand shocks is driven by the rigidity of the base wage set by collective agreements at the industry level.

The rigidity above may have led to a number of responses, including the emergence of ‘pirate’ agreements in countries like Italy [Paper 15]. These agreements involve significant negative wage gaps, below the collectively agreed minimum wage (defined at the industry-wide level).

In our detailed examination of collective bargaining practices across different MS, we found important institutional differences. For instance, there are important differences in the range of bargaining partners, frequency of bargaining, and wage setting and compliance in the case of metalwork industry [Papers 22, 10]. Moreover, when comparing Portugal and Spain [Paper 20], the agreements in the latter country are bargained jointly by the two main unions and involve multi-annual, price-indexed wage increases, which are awarded to virtually all workers. These practices are not present in Portugal, which can explain at least in part its much better labour market performance since the financial crisis. While collective agreements in Portugal are very long (typically over 100 clauses), their added value is limited [Paper 21] as only 25% of its norms are effectively more generous than general employment law (and 62% are exactly equal to the latter).

Additional results

We recommend that more attention is paid, at both the national and the EU levels, to increasing the clarity of the coverage of collective agreements. These agreements are multi-dimensional, involving different industries, regions and occupations, which may also change over time. Their domain can also depend on the affiliation of firms and workers to employer associations and trade unions, respectively, which is not always necessarily straightforward and raises important questions about the representativeness of social partners [Papers 1, 3].

Tracking collective agreements over time and matching them to their correct labour market domains can be a challenging task. This creates important hurdles in legal compliance [Paper 4] and economic evaluation. It may also create barriers against investment, particularly across countries, given the existing informational asymmetries. Greater convergence across the EU in labelling and characterising collective agreements, perhaps following a process of ‘benchlearning’ (dissemination of good practice) could be an important way to support and modernise collective bargaining.

Extensions may also have an important role in the relationship between productivity and earnings across firms. For instance, when focusing on the impact of union representatives in firms [Paper 2], they appear to have positive effects on the productivity in the firm where they are based but not on the wages of those workers. This may be because under sectoral bargaining with extensions, wage increases will be spread out across firms even if productivity increases may be differentiated across firms. This imbalance may have negative effects on resource allocation within sectors.

A number of papers consider the specific case of Greece [Papers 9, 10 and 12]. For instance, the evidence presented indicates that the bargained basic wage appears to be binding for small firms in the case of important hospitality industry.

We also consider ‘non-economic’ outcomes of collective bargaining, focusing specifically on the employee/self-employed dimensions [Papers 6, 7]. We found that self-employment is not necessarily associated to lower outcomes in either health or fertility, provided that social protection does not discriminate between work formats. From an international perspective, we consider the role of China. First, we find that rent sharing (the division of profits between employers and workers) is much smaller in its labour market compared to those of EU MS [Paper 18], which highlights the importance of collective bargaining in Europe. Second, we find important negative effects of the shock of China’s emergence in the international economy in the labour market of Portugal [paper 17], both in terms of wages and employment. The latter result may underline the importance of some degree of flexibility in collective bargaining so that firms can better accommodate these types of shocks.

Finally, we also studied the role of employer concentration and its labour market effects [Paper 5]. Exploiting comprehensive matched employer-employee data for Portugal, we find relatively low levels of concentration (less than 9% of workers are exposed to levels thought to raise market power concerns), compared to the few existing studies covering the case of the US. However, we also find that wages can be negatively affected by employer concentration. Future research will seek to examine the role of collective bargaining (and extensions) in the levels of employer concentration and its wage effects.