Nordic Economic Policy Review 2025

This very interesting article about the wage (and employment) effects of extended collective bargaining agreements (mostly) in the Nordic countries reviews the evidence and finds that extensions raise wages modestly, especially for low-paid workers, but may lead to employment trade-offs. The article also considers recent developments in Nordic labour markets, such as the Finnish shift from industry to company agreements and discusses the wider implications for labour standards considering falling trade union density and the rise of posted workers. The article is well structured, easy to follow and speaks to a broader audience. It also draws on relevant literature. The following feedback focuses on complementing the article, introducing contrasting perspectives and adding additional context.

While similar along several dimensions, the Nordic countries have chosen different minimum wage regimes. Denmark and Sweden rely solely on collective agreements to regulate minimum wage levels, whereas Finland, Iceland, and Norway have also introduced extensions of collective agreements, referred to as erga omnes instruments (Eldring & Alsos, 2014). None of these countries have introduced statutory minimum wage systems. The Nordic resistance to statutory wage regulation is illustrated by the ongoing debate surrounding the EU minimum wage directive, introduced in 2022. While Denmark and Sweden have strongly opposed the directive to preserve their autonomous wage-setting models, Finland has taken a more pragmatic approach, concluding that its existing extension mechanisms largely comply with the directive’s requirements. This divergence reflects broader differences in how the Nordic countries balance collective agreements with state intervention.

The article provides an overview of extension practices in the Nordic countries, highlighting key differences in their implementation. I think a short discussion should be included on why the practices vary between countries. Finland has chosen to extend collective agreements more extensively than Sweden and Denmark, despite all three countries having similar levels of union density and collective bargaining coverage (Norway lower, partly due to the Ghent system in the other three countries). Sweden and Denmark have strong traditions of maintaining labour market autonomy, which may explain their reluctance to use extensions, while Finland has been more willing to adopt them. Including such a discussion would not only clarify the institutional differences but also provide readers with a deeper understanding of how historical and political factors, such as the greater role of the state in labour market governance in Finland, have influenced these divergent approaches.

The article distinguishes between compulsory and voluntary extensions. Compulsory extensions are mandated by law or a government agency, applying collective agreements to all employers in a sector. Voluntary extensions are when employers adhere to collective agreements, even though they are not legally obliged to do so.

In Norway, voluntary extensions are often referred to as the norm-setting impact of collective agreements. In practice, this means that the (wage) terms of collective agreements are often followed in workplaces without an agreement because they are considered industry standards. This informal extension through norm-setting is particularly significant in sectors with low union density or collective agreement coverage, such as the retail trade and hotel, restaurant, and catering (Alsos et al. 2021).

The paper could benefit from a more thorough discussion of the voluntary mechanisms, which can be significant. Recognising the prevalence of voluntary extensions is important when assessing the impact of compulsory extensions on wages, as failing to account for them could lead to an overestimation or underestimation of their true effects.

There are also differences in the scope of extensions. In Norway, for example, extensions typically only apply to minimum wage rates, whereas in Finland, nearly the entire collective agreement can be extended. This distinction is significant because it reflects variations in policy design that influence both labour market outcomes and the role of collective agreements as tools for regulating labour standards. Svarstad & Oldervoll (2018) suggest differentiating between extensions of collective agreements and extensions of minimum terms regulated by those agreements, a framework that could help clarify the discussion. As noted in the article, Finland’s use of broader extensions contrasts with Norway’s more targeted approach, reflecting differing strategies to achieve labour market objectives such as combating social dumping or maintaining comprehensive labour standards. Expanding on these nuances would improve the article by offering a clearer understanding of the implications of different extension practices.

The article highlights various effects of extending collective agreements, but more attention could be paid to how these effects are shaped by the specific context in which extensions are applied. This is particularly important given the article’s aim to evaluate the impacts of extending collective agreements. The wage and employment effects of extensions are undoubtedly related to a variety of factors specific to each case. One such factor is the level of the wage rate being extended. This probably has implications for the wage effects of extensions. However, it is perhaps just as important when considering employment effects that there are considerable differences between industries in the extent to which firms are able to pass costs onto prices or substitute labour for capital, which could have longer-term implications for employment.

These contextual differences are illustrated by examining how extensions have shaped wages in industries with different types of collective agreement, such as construction and cleaning in Norway. This is shown in Figures 1 and 2.

Figure 1. (Hourly) wages among workers covered by extended collective agreements in construction in Norway 2022.

Source: Svarstad & Dapi, 2024

Figure 2. (Hourly) wages among workers covered by extended collective agreements in cleaning in Norway 2022.

Source: Svarstad & Dapi, 2024

The extended collective agreement in construction (‘Fellesoverenskomsten for byggfag’) is what is called a minimum wage agreement, while the cleaning agreement (‘Renholdsoverenskomsten’) is a normal wage agreement. Minimum wage agreements set lower wage floors, with local negotiations expected to supplement these rates, whereas normal agreements establish fixed, higher wage rates that do not allow for local adjustments. Consequently, normal agreements tend to result in higher overall wage levels.

Empirical data confirms these distinctions: in construction, 78% of workers earn more than 15% above the extended minimum wage, while only 11% of workers in cleaning surpass this threshold, with wages clustering around the extended normal wage rate.

In addition, the impact of extensions depends on the initial wage dispersion in an industry. If most workers earn close to the extended rate, the effect on wages will be smaller compared to industries with a greater gap between actual and extended wages. These observations underscore the significance of both the wage floor level and industry wage structures, illustrating how challenging it can be to make generalisations about the effects of extensions in different contexts. 

A few times, the article suggests a one-to-one relationship between membership of employers’ associations and coverage by collective agreements. This does not apply to Norway, where many workplaces are members of employers’ associations without being a part of a collective agreement. In Sweden and Denmark, membership more or less automatically implies coverage under a collective agreement.

Moreover, in recent years, there has been a noticeable increase in membership of employers’ associations in Norway. This rise has not been accompanied by a corresponding increase in collective agreement coverage, suggesting that many firms prefer the benefits of membership of employers’ associations without committing to the obligations that come with collective agreements (Alsos et al., 2021).

The paper analyses the case of a “reversed extension” in the Finnish forest industry, where sector-wide collective agreements were abolished and firms moved to company-level bargaining. This shift provides a unique opportunity to study the effects of removing generally binding agreements, raising important questions about wage developments and labour market dynamics. However, I have a few reservations.

Firstly, research shows that wages are much harder to lower than to raise, a phenomenon referred to as downward wage rigidity (Holden & Wulfsberg, 2008; 2009; Vainiomäk, 2020; Engebretsen, 2021). In practice, it means that removing an extension may not lead to wage reductions in the same way that introducing it leads to wage increases. Comparing the two processes directly without acknowledging this asymmetry could oversimplify the analysis, as wage-setting behaviour is influenced by institutional and behavioural factors that resist downward adjustments. I believe the paper would benefit from a more thorough discussion of the literature on downward wage rigidity, as it would provide important context for interpreting the observed wage patterns after the extension was removed. 

I am also curious about how the effects of a “reversed extension” can be disentangled from the broader impacts of decentralisation. Decentralisation may introduce new bargaining dynamics, such as company-specific wage adjustments and productivity-based pay schemes, which could blur the distinction between the effects of removing an extension and the structural shift toward more local negotiations. While this is undoubtedly a complex task, acknowledging this potential overlap and discussing possible strategies for separating these effects would strengthen the analysis and provide a clearer interpretation of the results. 

Finally, the forest industry in Finland has strong unions (as acknowledged in the paper), which could be an important reason why wages have not (yet) declined. One could question how much this specific case truly says about the importance of extensions in general. One hypothesis could be that strong unions can enforce hängavtaler or otherwise maintain conditions equivalent to the collective agreement even in non-unionised companies and that formal extensions play a more important role for wages in industries with lower union density.

Alsos, K., Nergaard, K., & Svarstad, E. (2021). Arbeidsgiverorganisering og tariffavtaler. Fafo-rapport, 7, Oslo.

Eldring, L., & Alsos, K. (2014). European Minimum Wage: A Nordic Outlook – an update. Fafo-paper, 21, Oslo.

Engebretsen, I. (2021). Nedadgående lønnsrigiditet i Norge. Master’s thesis. University of Oslo.

Holden, S., & Wulfsberg, F. (2008). Downward nominal wage rigidity in the OECD. The BE Journal of Macroeconomics, 8(1).

Holden, S. & Wulfsberg, F. (2009). How strong is the macroeconomic case for downward real wage rigidity? Journal of Monetary Economics, 56(4), 605-615.

Svarstad, E. & Dapi, B. (2024). Lønn i allmenngjorte bransjer 2022. Fafo-notat, 8, Oslo.

Svarstad, E. & Oldervoll, J. (2018). Mellom lov og avtale - Minstelønnsreguleringer i Nord-Europa. Fafo-notat, 12, Oslo.

Vainiomäki, J. (2020). The development of wage dispersion and wage rigidity in Finland. Fin