Disruption of the Ghent effect: Disentangling structural and institutional determinants of union membership decline in Sweden, 2005–2010

• Published date: 01 November 2023
• Author: Jesper Prytz,Tomas Berglund
• Date: 01 November 2023
• DOI: http://doi.org/10.1111/irj.12404

DOI: 10.1111/irj.12404

ORIGINAL ARTICLE

Disruption of the Ghent effect: Disentangling

structural and institutional determinants of

union membership decline in Sweden,

2005–2010

Jesper Prytz|Tomas Berglund

Department of Sociology and Work

Science, University of Gothenburg,

Goteborg, Sweden

Correspondence

Jesper Prytz, Department of Sociology

and Work Science, University of

Gothenburg, Gothenburg, Sweden.

Email: jesper.prytz@gu.se

Funding information

Forskningsrådet om Hälsa, Arbetsliv och

Välfärd

Abstract

In 2007–2008, Sweden implemented changes to

the Ghent system of unemployment insurance,

leading to union density decline. We study the

interaction of labour force composition and institu-

tional changes, finding that the structural tenden-

cies of the decline strongly interacts with these

changes, exposing previous precarious categories

even further.

1|INTRODUCTION

Sweden and its Nordic neighbours boast one of the world's most robust labour movements. The

country's high union density (UD) surpasses that of many European countries and can be

attributed to several factors: close ties between unions and the politically dominant Social

Democratic Workers' Party, collective agreements as the primary regulating mechanism of the

labour market, and active workplace presence of the unions. As a result, a widespread culture

of union membership prevails in workplaces.

In addition, institutional mechanisms reinforce UD with the so‐called Ghent system

of union‐administered unemployment uninsurance as the most important of these

(Edebalk, 1996). While other European nations have moved from voluntary to compulsory

systems, the Ghent system still remains in several Nordic countries (Rasmussen &

Pontusson, 2017). The Swedish Ghent system is regarded as a key institution determining

the power resources of labour and capital, and therefore, highly politically contested

Ind. Relat. 2023;54:471–494. wileyonlinelibrary.com/journal/irj

|

471

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and

reproduction in any medium, provided the original work is properly cited.

© 2023 The Authors. Industrial Relations Journal published by Brian Towers (BRITOW) and John Wiley & Sons Ltd.

(Lindellee, 2018; Rothstein, 1990). The state has direct legislative access concerning

entitlements criteria, compensation levels and relationships between the funds and unions,

and several different governments have also taken the opportunity to reform the insurance

since the 1990s (Lindellee & Berglund, 2022; Gordon, 2019). Conservative governments have

made the most pervasive changes by reducing the generosity of the unemployment benefit,

while social democratic governments have complied with most of these changes and only

marginally restored the benefit (Lindellee, 2018). Those changes have created haves/have‐nots

divides concerning access to income protection (Gordon, 2019).

The most significant changes of the Ghent system were implemented by a centre‐right

government, that took office in 2006, putting forth a distinctly neoliberal policy agenda and

workfare regime (Bengtsson & Berglund, 2012). To provide for an earned‐income tax credit, the

government decided to triple membership fees to the unemployment funds, decrease the

compensation rate and make the eligibility criteria much stricter, all implemented in 1 January

that year (Kjellberg, 2011). The government also abolished tax deductions for union

membership (at 25%) and unemployment funds fees (at 40%) (Kjellberg, 2019; Prop. 2006/

07:1). Further changes were made in July 2008, when fee amounts became based on the

unemployment level within the branch of the specific unemployment funds. This change—in

addition to a flat fund reduction fee on1 July 2009—led to some white‐collar unions ending up

with lower fees than before the 2007 reforms, while several blue‐collar unions saw further

increases in fees (Kjellberg, 2011; Kjellberg & Lyhne Ibsen, 2016).

These changes to the unemployment funds resulted in a decline in union membership as

well as coverage of unemployment insurance (Lindellee & Berglund, 2022; Kjellberg, 2011).

Between 2006 and 2008, UD dropped from 75% to 69%, a decline in previous research clearly

relates to the changes in the unemployment insurance system (Kjellberg, 2011). Particularly,

the reforms hit younger employees and industries with large numbers of low‐paid workers

(Kjellberg, 2011; Kjellberg & Lyhne Ibsen, 2016).

The impact of the dual reforms of the Ghent system in 2007 and 2008 on unionisation is an

important research question in several respects. First, the Ghent system is a cornerstone for

some of the strongest labour movements in the world and key for the so‐called Nordic labour

market model. However, in the last decades weaknesses in this model have been visible (e.g.,

Baccaro & Howell, 2017), and the paradox how one of its key features (the voluntarism of the

Ghent system), can turn into one of its weakest spots warrants further study. Second, the

reform added to general tendencies of dualization visible in Sweden (Lindellee & Berglund,

2022; Emmenegger, 2014; Gordon, 2019). As will be shown in this study, precarious workers

became even more exposed after the reform, both by losing UB and the protection of union

membership. Third, previous research indicates that the changes decreased unionisation

among more precarious categories of workers (e.g., Kjellberg, 2011; Palm, 2017). However, by

not multivariate testing the timing of union decline, these studies are not able to account for

alternative possible explanations of the decrease and test reasonable theoretical explanations

for the decline.

Thus, the purpose of the current study is to scrutinise the effect of the reforms of the

Swedish Ghent system on union membership. We ask how UD was impacted in different

phases of the reforms, how different categories of workers were affected, and how the reform

interacted with other institutional changes in the direction of dualization, mainly the

liberalisation of temporary employment. For this purpose, we analysed detailed individual‐level

data from the Swedish Labour Force Survey (LFS) to follow monthly changes in unionisation

both before, during and after the changes to the Ghent system.

472

|

PRYTZ and BERGLUND