by Federico Danilo Filetti and Emanuele Ferragina
Over the last decades, labour market protection in high-income countries underwent severe processes of change. Labour market liberalization first unfolded in the United States and the United Kingdom during the 1980s and served as a blueprint for reforms successively implemented in Europe since the 1990s. These processes of reform accelerated as a consequence of European integration and the Great Recession in the 2000s. Labour market protection has been predominantly reformed through the deregulation of employment protection, the weakening of collective bargaining institutions and the recalibration of compensatory benefits (i.e., unemployment benefits and minimum income schemes). Contextually, the reception of compensatory benefits has been increasingly conditioned to the participation to active labour market programmes. Moving from this context, our study provides a map of labour market protection generosity and change since the 1990s in 21 high-income countries.
Literature review and contribution
Our work engages with longstanding debates in Comparative Political Economy and Comparative Social Policy – and notably with Varieties of Capitalism and Worlds of Welfare representations – analysing similarities and differences among countries’ labour market protection. These works identified clusters of countries according to the levels and characteristics of labour protection (and other social policy realms) they guaranteed. On the one hand, they identify a liberal cluster – composed by the United Kingdom and United States and other countries like Australia, Canada, Ireland, Japan and New Zealand – with residual levels of protection and a high dependence of workers from market mechanisms. Liberal countries have a highly deregulated employment protection legislation, decentralized wage bargaining and meagre cash benefit generosity. On the other hand, these works witness the presence of a second group composed by most western European countries – with higher levels of labour market protection compared to liberal countries, but a marked cross-country heterogeneity. Welfare state theories help to deal with this heterogeneity further distinguishing between social democratic, Christian democratic and Mediterranean countries. While social democratic countries dispose of a universal welfare state and provide workers with encompassing labour market protection, in the Christian democratic cluster labour market protection reinforces the existent class structure – as labour market protection is mostly related to the level of social contributions. Labour market protection in Mediterranean countries, instead, is distinguishable from Christian democratic countries because of lower levels of generosity and a residual income assistance. Varieties of Capitalism and Worlds of Welfare analytical representations have often been accused of not accounting for dynamics of institutional change. To put it simply, they took a ‘snapshot’ of countries’ labour market protection in a determined point in time (the 1980 for Esping-Andersen’s work and the 1990s for Estevez-Abe et al.’s work), thus not being in position to analyse processes of change in labour market protection (i.e.: liberalization). Our work updates existing typologies and provides an illustration of the processes of change in labour market protection, that 21 countries went through since these typologies have been originally developed.
Seminal studies have tried to capture the tension between time-invariant clusters and the dynamics of change in labour market protection taking place in high-income countries over the last decades. We specifically refer to Thelen’s Varieties of Liberalization and the New Politics of Social Solidarity, in which she argued that trajectories of change in labour market protection are consistent within each country-cluster. Accordingly, liberal countries pursued a deregulation trajectory, which consisted in a classic form of liberalization that reduced social protection for the whole workforce. Christian democratic countries, instead, followed dualization, a form of liberalization that maintains high protection for the core of the workforce (especially in manufacturing), while deregulating the periphery (and in particular the low-skilled service sector). Finally, social democratic countries pursued embedded flexibilization: a process aimed at making the labour market more flexible, while contextually investing in active and passive measures. Although presenting important insights, this work is based on a restricted sample of countries and employs a rather limited selection of indicators to define labour market protection.
In our article, we address these issues looking at similarities and differences in labour market protection across four country-clusters and three decades. In doing so, we refine Thelen’s work by systematically analysing processes of change at the country-level in relation to (and independently from) existing country-clusters in a large-N setting. We also measure labour market protection more holistically, distinguishing between programmes for core workers and those targeted at atypical workers. Our operationalization of labour market protection employs macro-level data from OECD and other sources and is based on four different dimensions: employment protection, unemployment protection, income maintenance and activation. We also account for changes in the workforce composition including indicators measuring the unemployment rate, the share of permanent and temporary workers and the share of involuntary part-time workers (an indicator used as proxy for poor quality jobs).
We find that, in 1990, country-clusters retraced those identified by Varieties of Capitalism and Worlds of Welfare. Our figure shows a clear distinction between liberal countries (on the left), with more residual labour market protection, and Western European countries (on the right). Moreover, moving along the vertical axis, the figure helps to distinguish between Christian democratic (in red, with levels of protection close to the average of the sample), social democratic (in blue, with high levels of protection) and Mediterranean (in black, with low levels of protection) clusters. Due to processes of change in labour market protection, in 2015 the composition and characteristics of country-clusters changed significantly, despite the difference between liberal and continental Western European countries’ labour market protection remains noticeable. We find that liberal countries reformed their labour market protection only marginally (because labour market reforms took place before the 1990s, and starting from very low levels) and became more similar over-time (as countries get closer one another). Mediterranean countries clustered together in the fourth quadrant with France progressively approaching them. These countries experienced a significant increase of unemployment rates and levels of labour market precariousness, while their labour market protection has been overall reduced. Finally, we observe that the differences between Christian democratic and social democratic countries became more blurred over time, and in 2015 most of the countries belonging to the two groups cluster together in the first quadrant. Denmark (at the top of the figure, close to the vertical axis) is an outlier and reformed its labour market protection according to a flexicurity model. On this basis, we suggest a revised typology of country-clusters as a result of 25 years of institutional change: (1) a liberal cluster, whose composition and characteristics are coherent with previous works; (2) a Central/Northern European cluster that combines Christian democratic and social democratic countries; and (3) a Southern European cluster composed by Mediterranean countries plus France.
To study processes of labour market protection change at the country-level, we created a new multi-dimensional score allowing us to detect depth and characteristics of institutional transformation between 1990 and 2015. We find that countries became more similar to the United States over 25 years; a result of the deregulation of employment protection and the deterioration of labour market outcomes. At the theoretical level, countries’ trajectories can be summarised in a fivefold taxonomy that modifies Thelen’s original work – i.e.: liberalization, dualization, flexicurity, de-dualization and higher protection. Finally, we show that these trajectories are not always path-dependent and consistent within country-clusters.
In sum, it is important to systematically scrutinize and conceptually refine how labour market protection varieties and countries’ trajectories of change evolve over time, and this consideration also applies to other policy domains. Substantial institutional change can be obscured from sight if we take for granted the immutability of typologies or consider that trajectories of change are always inscribed within classic varieties. As Galileo said during his trial: Eppur si muove!
This blog post is based on an article published in the Journal of European Social Policy.
About the authors
Federico Danilo Filetti is a PhD student in Sociology at LIEPP and the Observatoire Sociologique du Changement (OSC) at Sciences Po, Paris.
Emanuele Ferragina is Associate Professor of Sociology at LIEPP and the Observatoire Sociologique du Changement (OSC) at Sciences Po, Paris.