Decline in union density threatens collective bargaining

Union density, as well as membership of unemployment funds in Denmark, has decreased steadily since 1994, according to a trade union study published in October 2010. In particular, the number of double members, those belonging to both a union and the unemployment fund connected to it, diminished. Unions fear this undermines the very foundation of the Danish model of labour market regulation as their representativeness is challenged and their legitimacy deteriorates.

Number of ‘free riders’ increasing

The study (in Danish, 1.4Mb PDF) published by the Danish Confederation of Trade Unions (LO), focused on membership of trade unions as well as of unemployment funds, the reasons behind these trends and the consequences for the collective bargaining system.

According to the study, the loss of membership is most pronounced among the trade unions and unemployment funds affiliated to LO, the largest of the three employee confederations. LO’s membership covers manufacturing, the metal industry, transport, agriculture and also considerable parts of the public sector, including large groups of commercial, clerical and other salaried employees. One reason for the decline lies in the changes to education, and in the way work is structured. But it is also a reflection of growing difficulties in recruiting and retaining employees, many of whom are more likely to stay completely outside trade unions and unemployment funds, or to join the alternative so-called ‘yellow’ organisations, which are independent in relation to the traditional ‘red’ trade unions and charge a much smaller membership fee. These employees are practically ‘free riders’ in relation to the collective bargaining system because the ‘yellow’ members and non-members are still covered by the sectoral collective agreements signed by the traditional social partners and thus receive a share of the collective achievements. In Denmark, the sector agreement signed by an employer covers all employees, regardless of union membership.

The same trend is also observed in the case of unemployment funds. These have lost many members, which means that the number of double members, those belonging to both unions and unemployment funds, has diminished as well. This is important because such members are the foundation of union strength. According to the study, however, a development seems to be emerging where ‘double choice’ is replaced by ‘double deselection’.

Membership of unemployment funds fell from 79% to 74% in the period 1994–2008. In the same period, union density declined from 77% to 71%. Double membership declined from 70% to 62% (see table).

Employees aged 18–66 years, according to organisational relationship (thousands and %)*
  1994 1997 2000 2004 2008 Change

1. Labour force (minus self-employed)

2,515

2,492

2,524

2,531

2,587

+72

+2.9%

2. Insured in an unemployment fund

1,992

79.2%

1,992

79.9%

1,984

78.6%

1,969

77.8%

1,921

74.3%

-71

-3.6%

3. Not insured in an unemployment fund

523

20.8%

500

20.1%

540

21.4%

562

22.2%

666

25.7%

+143

+27.3%

4. Double members

1,750

69.6%

1,739

69.8%

1,722

64.7%

1,662

65.7%

1,602

61.7%

-148

-8.5%

5. Single members (unemployment fund)

242

9.6%

253

10.2%

262

10.4%

307

12.1%

319

12.3%

+77

+31.8%

6. Single members (union)

187

7.4%

182

7.3%

200

7.9%

205

8.1%

233

9.0%

+46

+24.6%

7. Union members (4–6)

1,937

77.0%

1,921

77.1%

1,922

76.1%

1,867

73.8%

1,835

70.9%

-102

-5.3%

8. Non-union members

578

23.0%

571

22.9%

602

23.9%

664

26.2%

752

29.1%

+174

+30.1%

9. Double deselection (3–6)

336

13.4%

318

12.8%

340

13.5%

357

14.1%

433

16.7%

+97

+28.9%

Note: * Figures exclude self-employed workers.

Source: LO-dokumentation, No. 1/2010, Udviklingen I den faglige organisering: årsager og konsekvenser for den danske model, p. 17

Main findings

The study relies on the registration data, which offers a much more systematic overview of the characteristics of union and unemployment fund members compared with other wage earners than any of the earlier surveys. Different membership groups were analysed with regard to gender, age, ethnicity, education, income and professional affiliations. According to the study, this makes it possible to get closer to an explanation for the declining membership.

The main findings of the study are outlined below.

  • Total union density declined between 1995 and 2010 by six percentage points, from 73% to 67%. If the members of the ‘yellow’ unions are excluded, the density decreases by 10 percentage points, from 71% to 61%. LO shows the steepest decline from 65% in 1995 to 53% in 2010, during which period the yellow unions registered a slow but steady membership increase, from 3% to 10%.
  • Those not in unions tend to be younger workers (aged 18–39 years), immigrants and descendants of immigrants, low-paid employees without vocational education, and employees in the private sector, agriculture and fishing.
  • The density of membership in the unemployment funds declined in the period 1994–2008 from 79% to 74%.
  • In 2008, less than three in every four employees were insured against unemployment.
  • Those not in an unemployment fund tend, again, to be younger workers, immigrants and descendants of immigrants, and low-paid employees without education.
  • In 2008, the share of double members comprised a little less than two thirds of the labour force (not including self-employed workers). This is equal to an 8% decline since 1994.
  • Those most likely not to be double members are younger people, immigrants and their descendants, and persons without vocational training.
  • Without including the ‘yellow’ members in the calculation of double members, union density is 56%. This means that the core group of employees in the collective agreement system, that is members of both the union that signs a collective agreement and its connected unemployment fund, only covers a little more than half of the labour force.
  • It is particularly the LO-affiliated unions that have lost ground in union membership, membership of an unemployment fund and double membership. Partly because of the structural changes in the labour force, it could be said that people educate themselves into other jobs outside the coverage of LO. Also important, however, is the rising membership of yellow unions and a growing share of employees who choose not to join any trade union.
  • After being unchallenged for almost a century, the established trade union structure seems to have been weakening over the past 15 years. Strong, agreement-signing unions and employer associations have formed the basis of a labour market model marked by ‘undisputed’ self-regulation by the social partners. Growing membership of unions that do not take part in collective bargaining, combined with a growing share of employees who are not members of any union weakens the representativeness of the established trade unions, undermining an important prerequisite for the self-regulation of the social partners. Self-regulation requires strong, influential partners on both sides of the bargaining table.

Consequences of membership decline

The future of the Danish bargaining system seems a little dark. It should, however, be added that Danish union density, seen from an international perspective, is still among the highest. However, the analysis does raise this question: How much further does the continual weakening of the traditional unions have to go before we can speak of the beginning of the end of the traditional Danish labour market regulation model?

To answer this question, the authors have turned towards Norway which resembles Denmark in many ways, although it does not have the same close connection between the unions and unemployment funds found in Denmark and Sweden as a legacy of the Ghent-system. In Denmark, as well as in Sweden, the decline in membership is closely connected to the decline in membership of the unemployment funds. Thus, it appears that the unemployment funds are no longer an effective way of maintaining union membership.

In Norway, union density is generally lower. However, in the same period when Danish union density saw a decline from three-fourths to two-thirds of employees, union density in Norway has remained at between 50% and 55%. Therefore, it is not unlikely that the Danish system is in a transition period which might seem dramatic but nevertheless will stabilise at Norwegian levels.

According to the study’s authors, however, Norway suggests another possible scenario. In Norway, the collective agreement coverage in 2008 was 59% compared with 71% in Denmark. Moreover, there is a large discrepancy in coverage between different sectors. The consequence has been a need for a more centralised bargaining system and more political regulation. Thus, Norway introduced a mechanism extending sectoral agreements to cover all employees in that sector. This is not the case in Denmark, which relies on high voluntary membership. The authors therefore suggest the possibility that the continual decline in union density, combined with the resulting loss of legitimacy, could weaken the flexibility and decentralisation of the Danish system, requiring extension mechanisms as in Norway. According to the study, the companies and the social partners are unlikely to welcome such developments as they would lose the ability to influence the agreements.

Commentary

In the background to the study, the authors conclude that actual deregulation and disorganisation are not very likely, even if accelerating loss of membership is pointing in that direction. The more obvious possibility is that the government and the parliament would start regulating the labour market, bidding farewell to self-regulation by the social partners. This would mean a form of political ruling monitored through the supervising authorities, instead of local representatives of the social partners using the industrial conflict resolution system. Regulation by the authorities would probably mean less flexible regulation, less effective enforcement of the rules and more red tape for the companies. Seen from this perspective, there appears to be a common interest between the employee and employer organisations to strengthen the Danish model of self-regulation. It is the trade unions that are losing the members, but the employers risk losing a flexible system of labour market regulation that, on the whole, has been in the companies’ interest.

Carsten Jørgensen, FAOS, University of Copenhagen

Useful? Interesting? Tell us what you think. Hide comments

Eurofound welcomes feedback and updates on this regulation

Add new comment