Trade unionism in Sweden (executive summary)

By Reinhold Fahlbeck, Labour and Society Programme, DP/109/1999, ISBN 92-9014-617-6, 1999, revised 25 January 2000

Changing patterns of employment

During the 1990s the unionization rate in Sweden rose from 80 per cent to 83 per cent. Unions do not rely much on formal organizing campaigns. There is no need for that. Employees join either of their own volition or as a result of group pressure. Being outside a union is a risky business, as unions control much of labour regulation. Though any form of discrimination against non-unionized employees is against the law, such discrimination can be hidden by employers acting together with unions.

Unionization rates in the service sector are significantly lower than in industry, at least among blue-collar workers. Still, they hover around 70 per cent, a figure that is high enough not to cause serious union worry.(Endnote 1)

Atypical work has mushroomed and now accounts for some 30 to 35 per cent of total employment. However, atypical employees are unionized to the same extent as others, perhaps slightly more so. The reason seems to be that they are more vulnerable to exploitation and so have more to gain from union membership.

Distance work has become much more common and so has union interest in protecting employees engaged in such work. No generally applicable collective agreement on distance work has been concluded so far (1999) but many local agreements exist at company level. Several unions have issued recommendations and advice concerning distance work and model agreements have been prepared for members to sign with employers.

Effective collective bargaining arrangements exist in all sectors of the economy. Even the temporary work business is unionized and covered by a comprehensive collective agreement.

Unions represent employees with widely diverging interests. This is not new so unions have experience of handling this situation. The 1990s have brought little change in this respect. The leading TCO (Central Organization of Salaried Employees) union (Union of Swedish Salaried Industry Workers, SIF) offers an extreme example. It represents employees ranging from low-paid office staff to highly paid executives. Tensions between employees are more related to inter-union than intra-union competition.

The greatest difficulty is to make salaries respond to changes in market demand. The pay structure is highly rigid in the sense that it is extremely difficult to change the relative level of different categories of employees. If one union manages to negotiate extra high pay increases for its members, perhaps due to market forces, all other unions will immediately demand—and fight hard for—the same increase, threatening to undo the initial effect. Unions have a de facto monopoly on collective pay regulation since there is no free labour market in the sense of a market not governed by collective agreements. This makes free market-type adjustments extremely difficult to achieve. The emergence of temporary work might change this to some extent since temporary work agencies create competition for staff. This particular problem has little to do with union weakness but much to do with human greed!

Unions have positioned themselves in the vanguard of gender equality. This is a fairly new phenomenon. In a historical perspective the record of Swedish unionism is far from flattering. Special—often discriminatory—treatment of women was commonplace in collective agreements until well into the 1960s. Even today the labour market is far from gender neutral. Many professions are primarily female and jobs are usually lower paid than in male-dominated professions. Attempts to lift female occupations to male pay levels do not meet with resistance in themselves, but if a change in the relative pay structure is achieved all other unions will ask for the same rise.

Changing labour management relations

Decentralized industrial structures have led to some decentralization in collective bargaining. Individual pay structures are now replacing collectively decided pay levels, even in the public sector. To some extent this development reflects increased market pressure but there is no significant decrease in union participation. Unions adamantly defend industry-wide talks as the nucleus of bargaining.

Regional and local union bodies have been able to cope with developments in the 1990s and new systems for workplace representation have not been created. A sharp decrease in the number of regional unions reflects a need to increase their professional standard.

Human resource management is much practised. Management has not tried to use this to circumvent unions, or at least has not succeeded in doing so. Unions have not needed broad community support to resist employer attacks as few attacks have occurred and they have not posed any serious threat to unions.

Public status of trade unions

Unions enjoy strong support from their members and events during the 1990s have not brought any noticeable change. There may possibly be a lack of respect, even a degree of contempt for higher union officials. Parachutes and golden handshakes have become quite common for top union officials as have various perks and fringe benefits. These favours have not caused real discontent or seriously tarnished the union image but they are certainly much debated. Union sincerity and dedication in representing members are not in doubt. The legal or actual status of unions has not been threatened in any way in the 1990s. They continue to build new alliances and strengthen old ones. In recent years they have become increasingly involved in environmental questions and fair trade issues.

Challenges in a hostile economic environment

The political influence of unions has not decreased during the last ten years and there is no reason to believe that it will do so in the foreseeable future. Unions have played a very active role during the period of record high unemployment in the 1990s. In the international arena Swedish unions have worked hard to broaden the aim of the European Union to embrace labour market and social issues. The employment platform of the 1997 Amsterdam Treaty is hailed as a victory for European workers and also for union efforts. In a 1992 position paper commenting on an ETUC proposal for Collective Bargaining in Europe LO strongly agrees with the idea that the principal aim of the European trade union movement must be full employment. LO goes on to say:

However, we have no understanding of the idea that this aim would be ‘utopian’.—It is the definite responsibility of the different governments to give priority to full employment in their economic policy. As this is not done at present, the European trade union movement must underline in a stronger way that the Community will have to take a greater responsibility with regard to employment matters.

Unions actively promote vocational training. This is a contribution to supply-side measures aimed at improving the situation of unemployed people and the young.

Alternative forms of work organizations based on quality circles, team work or the like rather than union representation do not exist to any great extent in Sweden and the 1990s have seen no significant increase in such approaches.

The second half of the 1990s has witnessed a reduction in social security benefits. Compensation levels used to be 100 per cent or slightly lower but at times they have dropped to 75 per cent. A social-democrat government made most of these cuts, in spite of strident protests from LO, which fiercely defended existing benefits.

The international economy: A threat to national unions?

Swedish unions maintain a very high profile in international union cooperation. This reflects a long-standing tradition that originated in Nordic cooperation. This has now been superseded by cooperation within the EU as well as on a global level.

Unions have financed information campaigns concerning international labour matters for decades. Sweden promotes free trade and imports many consumer products, such as clothing and textiles, from developing countries. For these and other reasons most Swedes are aware of the importance of international issues.

Swedish unions are at the forefront in promoting human rights, including union rights. They strongly advocate the inclusion of basic ILO Conventions in international instruments such as the WTO charter.

Role of unions

It is sometimes said that the role of unions is threefold: first an economic role to facilitate production and ensure equitable distribution; second a democratic and representative role; and third a social integration role. Swedish unions continued to fulfil these roles during the 1990s. Sweden has been a land of immigration for the last 50 years and the integration of immigrant workers plays an important role in union domestic programmes. The democratic voice is probably even more persuasive now than ten years ago because of the increasing role of the third largest union federation, SACO. This federation is rapidly attracting new members and becoming more vociferous.


The purpose of this text is to present Swedish unionism to a foreign readership in a succinct but analytical way. The emphasis is on attitudes, structures, trends and overall characteristics rather than minute factual information. Since the approach is analytical it is imperative to highlight the elements that set Swedish unionism apart from unionism in most other (non-Nordic) countries.

Sweden is a parliamentary democracy. Its present constitution is quite new—the 1974 Instrument of Government—but constitutional traditions stretch back several centuries. The Head of State is the King who, however, plays only a formal role. Power is divided between three independent bodies: government, headed by the Prime Minister, the legislature (the single-chamber Riksdag) and the judiciary. Political representation is proportional, majority representation never having been part of the Swedish system. Since the mid-1930s the Social Democratic Party, SAP, has completely dominated political life. With few interruptions it has been in power since that time. The SAP is close to the main union federation in Sweden, the LO.

Sweden is a unified country and legislation is under the exclusive authority of the Riksdag. However, the government enjoys a strong position vis-ā-vis the legislature. This is demonstrated by the fact that government, rather than the legislature, is the main initiator and architect of new legislation. Provincial or local (municipal) regulations of a statutory type do not exist in labour questions.

Sweden has no equivalent to the NLRB (National Labor Relations Board) in the United States nor is there a state labour inspectorate. The public authorities have a limited role in administering day-to-day labour relations.

This is in sharp contrast to the role of public authorities in administering labour market policy schemes, e.g. full employment policies, employment exchanges, vocational training and retraining. Such policies, which are very important in Sweden, are administered by the Labour Market Board together with its provincial and local branches. Despite its name the Board is not concerned with labour and employment law or labour relations generally.

The Industrial Safety Board and its local branches are responsible for health and safety at the workplace as well as the working environment.

Employers and workers play an important role in regulating the labour market. The collective agreement is the instrument primarily used and these exist in every sector of the Swedish economy. Employers accepted collective regulation early this century.

Strong elements of trust, cooperation and mutual understanding between employers and workers characterize the Swedish industrial relations system. Acceptance of trade unionism on the part of employers and appreciation of the trade union contribution to the daily running of the enterprise is matched by a pragmatic acceptance on the part of the union movement of employers' freedom to manage the business and make decisions on technological change. In most instances the relationship between an employer (or an employer organization) and the union is firm and of long standing. The parties live together in something like a marriage of convenience with no possibility of divorce, as it were. Despite this rather cosy relationship there is little collusion between the parties and featherbedding is unknown. By and large the parties deal with each other at arms' length, while preserving their marriage of convenience.

The population of Sweden was between 8.8 and 8.9 million in 1997.(Endnote 2) The total labour force (including the unemployed) in the same year was around 5.5 million, i.e. a participation rate of about 77 per cent of the population between 15 and 64 years of age. The participation rate for men was slightly higher than for women, about 79 per cent versus 74.5 per cent. The overall participation rate has gone down from 84.5 per cent in 1990, mainly due to ageing of the population. The employed population stood at 3.9 million, of whom 11 per cent were self-employed or family members. Women accounted for 48 per cent of the total.

Women are strongly over-represented in atypical work such as fixed-term contracts and part-time employment. For example, the total percentage of fixed-term contracts stood at 14.5 per cent in 1997, nearly 17 per cent for women and 12 per cent for men. The difference is far more dramatic in part-time work, i.e. less than 35 hours per week. About 38 per cent of women work part-time, compared with only 9 per cent of men.

The primary sector (agriculture) accounted for only 2.8 per cent of the total labour force. Manufacturing and construction (the secondary sector) accounted for 26 per cent of the working population, down from the 1991 figure of 28.3 per cent. Public-sector employment accounted for 37 per cent of total employment in 1997, primarily in local government, down from 40 per cent in 1990. Women dominate in the public sector, particularly in health and welfare.

In 1990 manual workers accounted for 44 per cent of all employees, 46 per cent of them being women. Salaried employees, white-collar workers as well as professionals, accounted for 41 per cent, of whom 53 per cent were women. The remaining 15 per cent were self-employed or unclassified.

1. Overall characteristics of Swedish unionism

Trade unionism in Sweden has a number of special features. To a great extent these are common to all five Nordic countries (Denmark, Finland, Iceland, Norway and Sweden). The most conspicuous are highlighted here: some of them distinguish Sweden from any other country.

The uniquely high rate of unionization is the single most outstanding point. Union density rates are well above 80 per cent of the employed population. In some sectors they are over 90 per cent and increasing, not declining. Between 1990 and 1996 the overall figure rose from 80 to 83 per cent.(Endnote 3)

Another rather unusual phenomenon is that employers do not resist unions. Since a compromise was reached in 1906 between the then infant organizations on both sides, private sector employers belonging to the dominant Swedish Employers Federation, SAF, (Svenska Arbetsgivareföreningen), have accepted unionism. A cooperative attitude on the part of employers has prevailed ever since, despite some bitter conflicts.

Unions have traditionally pursued a highly ideological agenda. The transfer of the means of production to society was long a stated goal of the dominant blue-collar federation of employees, the Swedish Federation of Trade Unions, LO (Landsorganisationen i Sverige). Although that goal was never pursued with much determination, the transformation of Sweden into a welfare state based on political and economic democracy and on equality has been relentlessly pursued. However, and this is the outstanding feature, LO and its member unions have always maintained good lines of communication with their employer counterparts. This has enabled them to sign traditional collective agreements on wages and other terms and conditions of employment. It has also enabled them to negotiate master agreements on employer/worker cooperation on a variety of issues, such as grievance procedures, limitation of industrial action, health and safety at work, and gender equality. In other words, one outstanding feature of Swedish unionism is its pragmatism in dealing with the employer community.

Yet another outstanding feature is that unions look upon themselves as organizations with a mission, a kind of secular religion. Unions also see themselves as the vanguard of a better society. The task of unions is to help create this society and to lead their constituents into it. Closely related to the last characteristic is that inter- and intra-union disputes have always been rare.

Continuity is yet another characteristic of trade unionism and industrial relations generally. Existing unions can look back on an unbroken history since their foundation: LO celebrated its centenary in 1998. It is much the same organization today as it was in 1898, only society has changed considerably. This continuity permeates unionism and imbues the movement with a feeling of tradition coupled with responsibility.

2. Background and structure

Labour market organizations are voluntary, non-profit associations. There is no general legislation governing such organizations in Sweden, and no specific regulations. The requirements for a legally recognized union are minimal, making it extremely easy to form one. No registration is necessary. Some general principles of law exist but unions enjoy a considerable degree of self-governance.

There have been few allegations of abuse of this freedom and these have generally not been concerned with serious misconduct. Corruption or dictatorial practices are very rare. Calls for statutory regulation have been frequent but unions have consistently and adamantly opposed them. Given their strong position in Swedish life and the fact that union conduct has never caused truly serious concern, moves for legislation have never gained any significant momentum.

Union recognition is not really an issue since they all enjoy basic union rights as bargaining agents for their members.

As already mentioned, the Swedish labour market is highly organized. A recent study put the overall rate of unionization at 77.7 per cent in 1980, 81.6 per cent in 1990 and 83.6 per cent in 1996.(Endnote 4) Unionization rates are somewhat higher among white-collar than blue-collar workers. Union membership is fairly evenly distributed among the three main sectors of the labour market: private, local government and central government (state), though it is higher in the public sector than in the private sector. The size of the company is not particularly relevant, nor is the branch of industry. Age and geographical location are reflected since unionization rates are higher among older workers and in small towns rather than big cities. Women are unionized to a slightly higher degree than men. The rate among part-time employees is slightly above average. About 70 per cent of employees on fixed-term contracts belong to a union. Temporary workers, i.e. people working for agencies that place their employees with third-party clients, are organized at about the average level. They are covered by a nation-wide collective agreement which addresses the concerns of temps (cf section 6).

The union movement is divided into three main federations: for blue-collar workers, white-collar employees, and professionals. This division is largely an anachronistic remnant of the more class-oriented society of the early twentieth century. Mergers between unions are quite common within federations, in particular LO. To a certain extent mergers are defensive, but the chief motivation is offensive. Small unions realize the need to join a bigger union or else to form a bigger union together with other small union(s) in order to represent their members better. Moreover, technological developments may have rendered an existing union structure obsolete. All these points were relevant when the three unions in the typography, printing and bookbinding industries merged some 25 years ago. Three proud craft unions with long-standing traditions decided to merge, but it was a painful process.

So far no major amalgamation has taken place between unions belonging to different federations. Historically transmitted traditions and attitudes still obstruct trans-federation mergers despite a growing awareness of the need for common platforms and programmes. To an increasing extent, the relevant distinctions between employees are more likely to be found within the industry-wide unions of the three separate federations than between them.

Employees tend to cooperate across traditional borderlines. In some sectors collective agreements on wages and other conditions of work now cover all employees, but this trend is still very much in its infancy. The 1995 industry-wide agreement in the pulp and paper industry is a pioneer, covering some 35,000 blue-collar workers, white-collar employees, professionals and supervisors belonging to unions affiliated with the three federations plus the independent union of supervisors.

Cooperation between unions belonging to different federations has resulted in several industry-wide bargaining bodies. The best known is the Cartel of Private Salaried Employees (Privattjänstemannakartellen, PTK), a federation of 27 industry-wide unions, 17 from Sveriges Akademikers Centralorganisation (SACO) and 10 from the Central Organization of Salaried Employees (Tjästemännens Centralorganisation, TCO). In the past PTK entered into binding collective agreements on wages and other conditions of work, but that mandate was removed some years ago. The central, industry-wide mandate for PTK today is limited to matters concerning retirement, insurance and employment security, retraining and adjustment. At enterprise level PTK usually represents all salaried employees and their local unions.

In the international field cooperation and common action between the three federations are becoming the norm. Some differences of opinion exist between LO and TCO on the one hand and SACO on the other hand about union aims and strategies in the international arena. In a broad perspective these differences are minimal.

The Swedish Federation of Trade Unions (LO) completely dominates blue-collar unionism. LO was founded in 1898, which makes it by far the oldest of the three employee federations. It is a federation of 20 (1998) industry-wide unions.(Endnote 5) LO organizes blue-collar workers (and some white-collar workers as well, e.g. insurance company employees) over the whole labour market, private sector as well as public. With 2.1 million members (1998) it represents slightly more than half of the working population in Sweden. Membership peaked in the mid-1980s and declined somewhat in the 1990s. The density rate has gone up slightly, both phenomena reflecting the diminishing role of blue-collar work in the economy generally. Women account for 45 per cent of overall membership but they dominate the single biggest union in the LO family, the Swedish Municipal Workers' Union. The overall unionization rate of the blue-collar sector of the economy hovers slightly above 80 per cent (1997).

LO has a close relationship with the main political party, the Swedish Social Democratic Workers' Party, SAP. This alliance gives it a strong position. LO also commands a very powerful position from another point of view in that it represents slightly more than half the workforce, which in turn accounts for half of the entire population. The membership figures alone mean that LO speaks for one quarter of the entire population. Counting those who are dependent on members, the percentage increases further. This means that LO can speak to the government and public or private bodies on more or less equal terms and with great confidence.

The combination of political affiliation and membership figures has given LO a unique position in national life for most of this century. It has played an important role in shaping Swedish society.

Without exception LO members are industry-wide unions organizing employees throughout the entire country. Most are industrial unions, organizing all blue-collar employees in a particular branch of the economy regardless of occupation, skill or training. Historically, member unions were primarily craft based, i.e. organizing employees according to their skills and training. However, the number of craft unions has diminished during the century as they have merged and formed industrial unions. Some craft unions still exist, mainly in the construction industry.

Private-sector unions have generally dominated the LO family. The Swedish Metal Workers' Union was traditionally both the biggest and the single most influential member. Strong growth in public-sector employment after 1945 meant a concomitant increase in public-sector unionism. The private sector still accounts for more than 50 per cent of total membership. However, the single biggest member now is a public-sector union, the Swedish Municipal Workers' Union. No LO members are general unions, organizing employees regardless of occupation, education or skill.

The number of LO member unions has declined steadily over several decades. Traditionally, many unions were quite small in terms of total membership. Mergers have reduced the number of unions while ensuring that total membership has increased. There were 25 member unions in 1988 and 20 in 1997. However, there are still some quite small unions, such as the tinplate workers' union with 5,200 members and the musicians' union with 6,400 members (1997).

White-collar unionism is much more recent than blue-collar.(Endnote 6) Starting in the 1930s white-collar and professional employees began forming unions or turning existing associations into union-type organizations and demanding collective bargaining. Employers resisted. The government intervened and the 1936 Act on Freedom of Association and Collective Bargaining was passed. Building primarily on the experience gained by SAF and LO and copying most of the mechanisms that they had built, the statute extended collective bargaining rights to all private-sector employees, guaranteeing them freedom of association in the process. Under the protection of the statute, white-collar and professional unionism expanded quickly, albeit along different organizational routes. By and large white-collar unionism is now federated into the Central Organization of Salaried Employees (Tjästemännens Centralorganisation, TCO), founded in 1944.(Endnote 7) TCO is a federation of 18 industry-wide unions (1999) with a total membership (in 1998) of 1.2 million employees, 60 per cent of whom were women. TCO organizes employees in all sectors of the labour market: private, local government and central government, divided fairly equally between private and public employment. The unionization rate is about 90 per cent; it is particularly high in the public sector. Most members are industrial unions, many of which began as friendly associations for the promotion of professional standards and mutual assistance. Under its statutes TCO has no political affiliation or ties.

Professional employees are organized by unions federated into SACO, founded in 1947.(Endnote 8) (SACO is an acronym for Sveriges Akademikers Centralorganisation but the acronym has been adopted as the official name of the federation.) The total membership of its 26 industry-wide unions (1998) is 460,000 employees. Women account for around 45 per cent of total membership. Some 30 per cent of working members are employed in the private sector, 65 per cent in the public sector and the remaining 5 per cent are self-employed.

The predominant role of public-sector employment among SACO members distinguishes it from LO and TCO. Another distinguishing feature is the predominance of craft unions within SACO. Membership in the various unions is usually based on education, a university degree being required to join most of the member unions. Another distinguishing feature is the large number of unions and the modest size of most of them. This is because the recruitment basis of quite a few member unions is limited, e.g. physiotherapists, pharmacists or merchant navy officers. The smallest member union is the Swedish Veterinary Association with only 2,200 members. Yet another distinguishing feature is that 14 per cent of members are students or self-employed (5 per cent on a full-time basis and 5 per cent part time). The self-employed can be found in a variety of professions: they are architects, dentists, lawyers, physicians, or other specialists.

One SACO member is a general union, organizing employees who do not belong elsewhere. It is the only such union in any of the three federations but it is small, with less than 10,000 members. Most SACO members have an important role as a professional association as well. More often than not they grew out of professional associations, some dating back to the nineteenth century. The biggest member is the Swedish Association of Civil Engineers with a membership accounting for nearly 17 per cent of all SACO members.

SACO has experienced rapid growth in the 1990s, increasing its membership by 40 per cent since 1990. This is in sharp contrast to the other two federations, with little (TCO) or negative (LO) membership growth during the same period. Obviously SACO is growing from a much lower level and it is benefiting from higher educational standards in the population. In 1998 SACO also saw a new union entering the federation, increasing the number of member unions. This is noteworthy in an era when the number of member unions is decreasing in LO and TCO, although this is due to mergers between members rather than to unions leaving the organization.

SACO has faced considerable difficulties in gaining recognition as a federation of equal standing with LO and TCO. It was only in 1997 that SAC0 was admitted as a member of the European Trade Union Confederation (ETUC) and the Council of Nordic Trade Unions (NSF). There are several reasons for this somewhat discriminatory treatment. One is that SACO often pursued policies in labour market matters that deviated from and angered not only LO and TCO but also the social-democratic party. Another factor is that SACO competed with LO and TCO in some activities, such as trying to set up a European-wide employee organization other than ETUC (the CESI). A third reason, perhaps, is that SACO is strictly neutral in political matters. In addition, most of its members probably vote for the non-socialist bloc in Swedish politics!

Much of the acrimony between LO and TCO on the one hand and SACO on the other hand seems to have been overcome in the very recent past. The entry of Sweden into the EU seems to have united them into a common front vis-ā-vis employers and the EU bureaucracy in labour and social matters. They have shared an office in Brussels since 1997. The strong membership growth that SACO has experienced is also a factor behind its growing acceptance as a player on equal terms.

Minority unionism should not occur within the three federations since jurisdictional rules aim at preventing two member unions from organizing the same employees.(Endnote 9) These rules have generally prevented minority unionism from emerging inside the same federation. There are a certain number of jurisdictional agreements between unions belonging to separate federations but disputes are not all that uncommon. However, they have not led to any minority union situations since those involved in jurisdictional disputes are majority unions in their core field.

Independent unions, not affiliated with the three federations, are uncommon in Sweden, playing a very marginal role in the labour market. Independent unions are mostly minority unions, and independent unionism has never been a serious issue in Sweden. LO and its member unions aimed at organizing all (blue-collar) employees from the very start. The LO policy was strengthened by a corresponding policy on the part of SAF. Already in the early stages of modern industrial relations SAF preferred to deal with LO and its member unions (although there were instances where both SAF and individual employers promoted splinter unionism and independent unions). No communist union movement was ever formed or even seriously considered; no truly important communist party ever emerged in national politics. At the same time no religious union movement was formed either. Sweden was a firmly Protestant country so there was no ground for a union movement inspired by Catholic social thinking, as was the case in many European countries.

By and large the structure established in the first decades of this century of a unified labour movement rather than pluralistic unionism became the tradition of the land. The attitude that emerged is that multiple unions are ultimately detrimental to the employees concerned and to the country as a whole. This attitude partly reflects the strength of the larger unions. They dominate the field both in terms of actually unionizing employees in their various parts of the labour market and also in terms of creating an atmosphere in which union multiplicity is frowned upon. In other words, a classic example of the successful monopolist!

One consequence is that Sweden does not have an official system to establish majority status for unions. Some kind of border would have to be established to separate the employees represented by two (or more) unions from other employees and no such rules exist in Sweden. They are considered unnecessary because minority unions are very rare and they are unwanted because minority unions are considered undesirable. However, it usually poses no problem to ascertain which is the majority union since minority unions tend to be very small. If a problem arises it is usual to regard all associations in the business concerned as the unit for establishing majority status. Obviously this makes it very difficult for new unions or splinter unions ever to achieve majority status. An example will illustrate this point. In the ports longshoremen are traditionally organized by the Swedish Transport Workers' Union, a member of LO. This union organizes all workers in any job related to transport, e.g. trucking or bus driving. A splinter union was formed in the ports among longshoremen. This union achieved majority status in many ports, perhaps even among longshoremen in the entire country. However, it clearly did not have majority status if all transport work was to be included in establishing this. The fact that the entire transport business was taken into account evidently reflects the strength of the Transport Workers' Union. On the employer side the transport business is divided into several sectors, road hauling being separate from port handling of goods for example. Given the stern opposition of the powerful Transport Workers' Union and the entire LO federation, prospects for this splinter union were never very bright. The same is true for all unions that try to break into a field where there is already a union belonging to one of the three federations.

There are some independent unions that enjoy unchallenged majority status. The prime example is the union organizing supervisors. Now called The Leaders (formerly the Supervisors' Union of Sweden), this was formed as a professional association in 1905. In recent decades it has had a somewhat stormy history in terms of affiliation. It belonged to TCO for long periods of time but stayed outside at other times, as now (1999). Jurisdictional disputes are at the root of the friction between the two. The union organizes most supervisors in public and private employment. Supervisors enjoy the same basic rights of freedom of association and labour rights as other employees. Swedish law knows of (virtually) no exceptions for managerial employees. The Leaders is a very strong union and its position is unchallenged. The same is true of the Swedish Airline Pilots' Union. There are no other truly unchallenged and powerful independent unions apart from these two at the present time.

The syndicalism movement is federated into the Central Organization of Swedish Workers (SAC), which is a general union although most of its members are blue-collar workers. It was founded in 1910 as a splinter from LO. Today's total membership (1998) does not exceed 10,000 employees, i.e. less than 0.25 per cent of the employed population. Although it has been bigger than this at certain times it has always been small because the movement has been fought consistently and fiercely by LO, and SAF has never wanted anything to do with it. Despite its tiny size its influence on industrial relations in Sweden has not been quite negligible. It has served to challenge its mighty rival, LO. It sometimes manages to sign collective agreements with small, non-organized private employers. Problems usually ensue since the local branch of the LO-affiliated union will follow suit. By applying superior pressure it will obtain a collective agreement covering the same work, and a confrontation is inevitable.

How active are union members? In other words, how much active support can unions count upon from members? A study conducted by the Central Bureau of Statistics, published in 1996 but based on interviews carried out in 1992/93, provides the following information. Fifteen per cent of members actively participated in union work at that time. On average 40 per cent of members had attended a union meeting in the past year, with 53 per cent for SACO and 40 per cent for LO. A 1999 LO report on Union Activity and Union Work shows an increase in interest among women, particularly young women under 30, in union work and a concomitant increase in the number of women holding a union position of some kind (12 per cent in 1998 against 8 per cent in 1993). Corresponding figures for men reveal declining interest and also a decline in union positions held (16 per cent in 1998 against 19 per cent in 1993). Women held 40 per cent of voluntary union positions in 1998 against 27 per cent in 1993. Some 60 per cent of members affirm that they take an active interest in union matters. Women are still under-represented at the top of unions. Only two women are union presidents within the LO family of 20 unions. There are seven women presidents among the 18 member unions of TCO and the same number in the 26-member SACO family.

Inter-federation disputes between member unions are rare.(Endnote 10) Agreements between unions belonging to the same federation are common and several also exist between unions belonging to different federations. Still, jurisdictional disputes are not unheard of. Since they have not caused any real concern, calls for legislation to curb them have gone unheard. In most instances a settlement is reached without open conflict between employers and employees. LO has authority to make a binding decision: TCO and SACO can do so only if authorized by the unions involved. No public agency has authority to intervene, much less to settle a bargaining issue. There is no exclusive representation for a majority union.

Disputes between unions and individual members are rare and lawsuits are extremely rare. The number of reported court cases in the entire country is less than ten since 1945. Sweden has no rules on fair representation of union members but discriminatory treatment is unlawful.

3. Attitudes

What is a union? What are union attitudes and opinions regarding their work and their relations with members? One way of characterizing a union movement is to look at typical attitudes among members and officials to various aspects of unionism. Swedish unions display some very characteristic features in these respects.(Endnote 11)

Box 1.1: What is a union, its function?

(a) A voice for those who have none.

(b) An organization for those who prefer collective representation to individual representation.

(c) An alter ego of the enterprise, i.e. the company's alternative personality.

Box 1.2: What is a union, its common bond?

(a) An organization for people with a common attribute, e.g. skill or ideology.

(b) An organization to take labour out of competition.

(c) An organization for people with a common background, e.g. an enterprise.

Box 1.3: What is the relationship between unions and their members?

(a) Unions take care of their members' best interests.

(b) Unions present their members' views and wishes.

(c) Unions harmonize members' views with management views.

Box 1.4: What attitudes do union officers have towards their union work?

(a) Union work is a vocation, a kind of secular priesthood.

(b) Union work is a job among others, offering a career in the union business.

(c) Union work is an exercise in cooperation with management, even training for management positions.

These boxes may be used to compare Japan, Sweden and the United States. Union members and officials in these countries—and in other countries as well—respond to all the alternatives in the four boxes. Several exceptions would have to be made in the case of Sweden. Swedish unions, their members and officers do not respond at all to alternative (c) in boxes 1.1 and 1.4 and alternative (b) in box 1.4.

A classification should focus on the features that characterize unions and their officers. Table 1 attempts to do this with regard to the three countries, by summarizing responses to the questions in the boxes above.

Table 1. Attitudes of Japanese, Swedish and US unions
Box 1.1 Box 1.2 Box 1.3 Box 1.4
Japan c c c c
Sweden a a a a
USA b b b b

The differences are considerable, revealing important characteristics of the national union movements. Box 1.3 is of particular interest.

Swedish union officials, particularly in the blue-collar sector, see themselves as people with a mission. Their mission is to serve the employee community. Their gospel is the welfare of their members. Their mode of operation is that of a pastor leading his flock.

Some historical background will help to clarify the situation in Sweden. The blue-collar union movement was formed late in the nineteenth century. At that time blue-collar workers had lost contact with the established religion (i.e. the state-controlled Protestant church). They lived in a spiritual vacuum or void. The nascent socialist movement and the labour unions offered values such as solidarity, brotherly love and concern for others, equality and fair shares for all according to their needs. These principles closely resemble Christian ethics. Union campaigners and officials proclaimed them at that time and they still do. Unions were and still are communities of women and men inspired by these ideals and values: they are striving to better their lot in life by working together. There is an unbroken line of thinking between the early unionists and those of today. It is certainly true that the semi-religious fervour is less pronounced today, as poverty and exploitation no longer exist in Sweden and everyone, comparatively speaking, is affluent, but the spirit is still there. Unions are leaders but the purpose of leadership is to serve the community.

Describing the attitudes and values of the Swedish union movement as semi-religious reveals and explains certain aspects of the movement.

Alternative (a) in box 1.1 and box 1.3 means that union officials speak for their members, in the sense of formulating what is to be said and in actually saying it. The voice function (box 1.1) is much less important today than in the infancy of unionism, owing to vastly higher educational standards and workers' ability to speak for themselves, as well as the existence of many channels to express an opinion. But the voice function is still there because unions are supposed to represent rank-and-file workers and speak for them.

There is much less of a religious undertone in white-collar unionism. But unionism is still considered a noble activity, the unselfish pursuit of a fuller and more dignified life for members. Career thinking might be somewhat more prevalent but only marginally so.

The doctrinal role of taking care of members' best interests is also less conspicuous today, but it is still very strong and permeates the actual functioning of unions.

Two examples to illustrate this point are votes on issues at hand, e.g. strike ballots or collective bargaining agendas, and votes on proposed agreements. If the role of unions is to present their members' views, votes and referenda are the order of the day: at each and every juncture the rank and file must be asked for its opinion. If, on the other hand, the task of the union is to formulate, obtain and take care of its members' best interests the opposite becomes true. Votes and referenda are not the order of the day. They might even seem to disrupt the orderly running of things. In Sweden membership votes are very rare. Most unions never organize votes on current issues. When a vote is taken it is virtually never mandatory and the results are practically never binding on union officials. Reflecting this, Swedish labour law has no provisions on membership votes or referenda within unions. This situation is in sharp contrast to detailed regulation in the United States where votes and referenda are common, usually mandatory and almost always binding.

This all means that alternative (a) in box 1.4 is very strong and elements of (b) certainly exist as well. Unions do provide a career ladder. Traditionally the most coveted position for a working class person is to become a senior official of LO, the federation of blue-collar employees. And it is a fact that all those who have achieved this rank have come from the anonymous masses of the rank and file. They have little formal training. They have not graduated from senior high school and certainly not attended college. They have risen because of their dedication and skill. Contacts or hard elbows count for little and so do intra-union infighting and intrigue. Once there, they belong to the innermost circles of power in Swedish society. However, at least until very recently, they have been very modestly paid and enjoyed few fringe benefits. Any tendency on their part to forget the people they represent is quickly and resolutely quelled. This is true of all senior officials in blue-collar unions as well (although the Transport Workers' Union traditionally presents a livelier picture). Rare, for example, is the union president of a blue-collar union who has attended college.

Obviously educational levels have increased considerably among top union office holders even in blue-collar unions. It is still true, however, that the vast majority have attended neither high school nor college. On the other hand they have always benefited from extensive in-union education and have often attended non-formal educational institutions for long periods of time. Once they reach the top they are certainly very knowledgeable about all matters of concern to the union.

This means that the idea of union work as a vehicle for one's personal career is very remote indeed at the bottom of the hierarchy and is generally absent even among the higher echelons. The union is not a business and union work is not a job like others.

The above also applies to a great extent to white-collar unionism and—to a lesser extent—to professional unionism, for example SACO and its member unions. Obviously, educational levels are higher among union officials here since the border between the three different union federations is drawn along educational lines. Apart from that the picture is rather similar.

One illustration of the effect of this frame of mind is the length of tenure. The period in office tends to be quite short in Sweden, rarely exceeding ten years in the same post. This is in sharp contrast to some other countries, notably in the United States (where the (b) alternative in box 1.4 is very strong). Consider, for example, Samuel Gompers, who was president of the AFL for some 40 years until his death at age 74. Nothing similar is even conceivable in Sweden.

Another illustration of the importance of attitude (a) in box 1.4 (in particular when coupled with attitude (a) in box 1.2) concerns members' satisfaction with union leadership. All studies strongly indicate that the rank and file are satisfied with union leaders and trust them.(Endnote 12) There is nothing surprising in that. If union officials consider themselves to be people with a mission, charged with a noble and honourable task, and if they behave in this spirit, it is probable that members will have great confidence in them. If, on the other hand, alternative (b) is strong, even prevalent, a built-in source of conflict exists between members and their representatives. The risk of a gap between members and their representatives is always possible and that seems to be the main reason why extensive legislation is needed in the United States to prevent such gaps from occurring. Sweden lives in blissful ignorance of such rules.

Another way to illustrate the effects of alternatives (a) and (b) in box 1.4 is to study intra-union strife. Obviously the more alternative (a) prevails among union members and officials the less intra-union strife should be expected. The more alternative (b) prevails the more likely is intra-union strife. Sweden certainly illustrates this point. Intra-union strife is rare and when it does occur it is generally quickly eliminated.(Endnote 13) The situation in the United States is the opposite.

Yet another way to illustrate the effects of the various alternatives in boxes 1.1 to1.4 is to study the amount of legislation considered necessary to strike a balance between the interests of members and the interests of union office holders. Obviously a strong (a) attitude will reinforce the community of interests between members and their representatives, making statutory rules redundant. The absence of regulation in Sweden presumably reflects the strength of the (a) alternative.

The attitudes listed as alternative c) in boxes 1.1 to1.4 are generally alien to Swedish unionism. Box 1.4 is perhaps the most important here. It is true that Swedish unions cooperate with management and that personal relations between union officials and managers are usually friendly. But this does not mean that union officials tend to be co-opted by management. Swedish employers have never seriously pursued a policy of trying to reduce or even eliminate the influence of union officials by co-opting them. In the second place, union representatives have never shown a tendency to forget their roots or their mission. Rare indeed is the union representative who has assumed a true management position (apart from becoming a supervisor). Those who do accept a management job receive little sympathy or understanding from the rank and file, to put it mildly.

4. Potential and impact

In a unpublished paper I have listed 23 factors which influence union density. The purpose of the exercise is to arrive at a workable formula for explaining variations in union density rates among countries. Each factor is assessed on a five-step scale: very positive (for high union density rate), positive, neutral, negative and very negative (for low density rates). The result is quantified. The five alternatives are allotted a score, namely +6, +3, +/-0, -3 and -6 respectively. This means that the highest score possible is +138 and the lowest is -138. A total score of 0 means that prevailing conditions are neutral towards unionism.

A summary of the study is presented here because it illustrates important factors in the Swedish situation. Japan and the United States are included simply in order to highlight Sweden.

1. Degree of central union authority

The thesis is that strong central union authority is a positive factor for union density rates. Japan -3, Sweden +6, the USA -3.

Centralization is a feature of Swedish unionism. LO in particular is a very centralized organization.

2. Amount of substantive labour law (legislation or case law)

The thesis is that the more substantive labour regulation there is the less unions have to offer. Japan +/-0, Sweden +6, USA +6. Substantive regulation abounds in Sweden and this greatly favours unionism (c.f. sections 6 and 9 below for examples).

3. Degree of union involvement in day-to-day administration of public labour market policies

The thesis is that the higher the degree of union involvement the more unions have to offer employees. Japan +3, Sweden +6, USA +/-0.

4. Union participation in the administration of state unemployment benefits

The thesis is that the greater the role of unions, the more likely employees are to join. Japan +/-0, Sweden +6, USA +/-0.

5. Degree of union involvement in national politics generally

The thesis is that the higher the degree of union involvement the more unions have to offer. Japan -3, Sweden +6, USA +3.

6. Degree of union cooperation with and support for influential political parties

The thesis is that the higher the degree of cooperation the more unions will be able to offer employees. Japan -3 towards -6, Sweden +6, USA +/-0.

7. Degree of benefit to workers from belonging to a union (e.g. in terms of wages)

The thesis is that the more benefits unions provide the more attractive they become to employees. Japan +/-0 to +3, Sweden +/-0, USA +6.

The situation in Sweden is inconclusive because there are no non-unionized sectors of the economy. We do not know what levels would prevail if market mechanisms existed.

8. Group pressure and social custom

The thesis is that strong positive feelings about union membership influence employees to join. Japan +/-0, Sweden +6, USA -3.

9. Comprehensive collective bargaining

The thesis is that the more comprehensive collective bargaining is the more influential unions become and the more attractive to employees. Japan +6, Sweden +6, USA +6.

10. Depth of collective bargaining

The thesis is that the more detailed bargaining is the more active and influential unions become; this enhances their position. Japan +6, Sweden +6, USA +6.

11. Degree of activity of local unions at the workplace

The thesis is that the more active local unions are the more likely employees are to join. Japan +6, Sweden +6, USA +6.

12. Extent to which unions can channel and promote resolution of individual employee grievances

The thesis is that authority to process grievances will make unions more attractive to employees. Japan +/-0, Sweden +6, USA +6.

13. Degree of inter-union strife (jurisdictional disputes)

The thesis is that the more strife there is between unions the less attractive they are to employees. Japan +3, Sweden +6, USA -3.

14. Degree of intra-union strife

The thesis is that the more strife there is inside unions the less attractive they are to employees. Japan +3, Sweden +6, USA -3.

15. Degree of corruption in unions and vested interests

The thesis is that the more corruption there is the less attractive unions are to employees. Japan + 6, Sweden + 6, USA—3.

16. Comprehensiveness of union alternative to employer policies

The thesis is that the more comprehensive union programmes and platforms are the more attractive unions become to employees. Japan +/-0 to -3, Sweden +6, USA +6.

17. Union participation in managing the enterprise

The thesis is that the more unions participate the more they can deliver to employees. Japan +3, Sweden +6, USA -3.

18. Extent to which the union movement tries to organize all employees

The thesis is that comprehensive organization efforts lend legitimacy and relevance to the labour movement and makes unions more attractive to employees. Japan -3, Sweden +6. USA +/-0.

19. Employer resistance to unions

The thesis is that the more employers oppose unions the harder it becomes for unions to organize employees. Japan +/-0, Sweden +6, USA -6.

20. Degree of independence vis-ā-vis the employer

The thesis is that the freer the union is the more attractive it becomes to employees. Japan +/-0 to -3, Sweden +6, USA +6.

21. Risks involved in union membership

The thesis is that the higher the risk of retaliation by employers against unionized employees the less attraction unions will have. Japan +6, Sweden +6, USA -6.

22. Thresholds for union recognition and bargaining rights

The thesis is that the more difficult it is for unions to gain recognition and bargaining rights the more difficult it becomes for them to organize employees. Japan +3, Sweden +6, USA -6.

23. Freedom to undertake industrial action

The thesis is that the broader the range of freedom for unions to undertake industrial action with impunity the more powerful they become vis-ā-vis employers and the more attractive to employees. Japan +3, Sweden +6, USA +/-0.

Table 2. Summary of potential scores
(score +6)
(score +3)
(score +/-0)
(score -3)
Very negative
(score -6)
Total score
Japan5 (+30)6,5 (+19.5)6,5 (+/-0)4,5 (-13.5)0,5 (-0.5)35.5
Sweden22 (+132)-1 (+/-0)--132
USA8 (+48)1 (+3)5 (+/ -0)6 (-18)3 (-18)15

Unionization rates in Japan, Sweden and the United States in the late 1990s were around 24, 90 and 12 per cent, declining in Japan and the United States but increasing (or at least not falling) in Sweden.

The 23 dimensions listed and the answers provided are approximate and highly subjective. However, the scores given for Sweden do describe and characterize the Swedish union movement.

The dimensions listed mainly reflect union potential for organizing workers. Read in that way the 23 dimensions together go a long way to explain the surprisingly high union density rate in Sweden. All the factors combine to produce precisely that result. Together they show Sweden as a union paradise, as it were.

Read in another way these dimensions demonstrate the impact of unions in society at large. Points number 1, 3, 4, 5, 6 and 23 are highly relevant in that respect. Take, for example, number 23, i.e. freedom to undertake industrial action. The extent to which unions can resort to industrial action will greatly influence their impact on society. Industrial actions affect society and the social partners in a variety of ways, mostly detrimental. The very fact that union action can affect the social partners gives them power over these groups. In Sweden there is almost no restriction on industrial action. Both primary and secondary (sympathy) action is allowed. No social justification is required and no principle of proportionality applies, so that aggressive unions could wield enormous power. One reason why industrial action is not used as a weapon is that unions know that irresponsible conduct would lead to legislative curbs on their rights in this respect.

5. Agenda

Swedish unions represent their members in all negotiations on employment issues. The bargaining agenda is the broadest possible. As far as working life is concerned unions offer a total package from the cradle to the grave. In addition, the union platform has both a public, society-oriented, and a private, member-oriented side.

Unions maintain a very conspicuous presence in virtually every aspect of public life, although they are not political bodies per se. Still, LO proudly states that: We even formed a political party in order to pursue our demands in Parliament, the Swedish Social Democratic Party, a party which proved to be the most successful of all Swedish political parties in the twentieth century.(Endnote 14)

However, one of the characteristics of Swedish unions is that they do not pursue a strictly political course of action. It is true that LO and TCO have a strong ideology but they and their member unions pursue a pragmatic policy of not alienating themselves from the employer community. Federations and their member unions work closely with employer organizations, and they cannot jeopardize that cooperation in the political arena.

It is also true that Swedish unions are social creatures. They take part in public debates and policy discussions on social and economic questions. They also maintain a high profile in all matters concerning education and international solidarity. No field of human endeavour totally escapes them, even matters of a more private nature (see below).

Unions routinely participate in the legislative process, which they often set in motion. They have an intimate knowledge of social realities which enables them to pinpoint social problems, and other matters that need attention. Once the legislative process has begun, unions are involved in virtually every step. Since the process is meticulous and lengthy this gives unions considerable influence.

The first step is to set up a government committee to investigate the issue at hand and submit proposals. Unions are routinely represented on such committees and union representatives often chair them. Once the committee has published its findings and recommendations in a White Paper, copies are routinely sent to the organizations, agencies and other bodies concerned for comment. Unions participate in this round of consultations. Internal union efforts during the consultations differ considerably according to the importance of the question. When crucial issues are at stake internal union procedures might involve extensive member discussion before an opinion is given to the government. Once the government takes over the process, the influence of unions diminishes but public debate continues. Unions have no further formal influence once a bill is submitted to Parliament, but several Members of Parliament are or were union officials.

The private part of the union agenda concerns members as individuals. By far the most important activities here are the extensive training and education programmes conducted by virtually all unions. LO, for example, has a wide range of educational programmes and runs several schools. Some 11 per cent of LO expenditure in fiscal year 1997 went on education.(Endnote 15) Most of the programmes run by unions are directed at training members for union office, but some offer education of a more general nature. One LO school (Brunnsvik) plays an important role in Swedish cultural life, and it has trained many well-known writers. Vocational training proper is not usually part of union educational programmes.

Unions also provide services that have to do with the private lives of their members. Since unions can offer a huge number of customers to prospective business partners they are in an excellent position to secure a good price for their members. In most instances these services are optional but sometimes they are mandatory: the courts have accepted that unions have a wide margin of discretion here. Examples of union action include contracting insurance policies for their members, such as home and accident insurance. In addition, unions can arrange private bank loans or provide collateral for a loan (If you need a new car, contact your union!). Recently LO has entered the field of utilities. The electricity supply is being deregulated in Sweden, largely at the initiative of LO, which urged that competition should be introduced for the benefit of consumers. LO has made a deal with a big supplier, on behalf on 1.4 million households, giving them access to special rates if they so wish. LO also rents out computers to its members. Unions routinely assist members in realizing meaningful activities during time off, and also help in arranging vacations. One of the leading travel and tour operators in Sweden, which also runs hotels, is a creature of the LO movement (Reso).

Critics sometimes argue that unions meddle in things that are none of their business. Still, no restrictive legislation exists and courts found in favour of unions in one highly publicized case involving home insurance.

Swedish unions are very active in the international arena. Working either directly through international organizations, such as the ILO, or indirectly through international union bodies, such as the ETUC or professional associations, Swedish unions forcefully pursue an overall international agenda, looking upon themselves not just as participants but often as initiators as well. For example, LO takes the position that it was a main actor, if not the main actor, behind the creation of the ETUC and TCO feels the same. LO has pushed hard to have its member unions enter European federations.

First on the international agenda are strenuous and unrelenting efforts to have basic human rights, including core workers' rights, accepted everywhere and to have these included in international instruments, e.g. the Treaty of Rome as amended by the 1997 Amsterdam Treaty. Swedish unions adamantly support the inclusion of social clauses in international instruments, such as the WTO charter. More will be said on this topic in section 8.

Another important aspect of international activities concerns education and training. Unions spend increasingly large amounts on these in developing countries. Countries in Eastern Europe also receive sizeable financial assistance.

6. The information society

In 1982, the labour market parties in the private sector (SAF, LO and PTK) signed the Development Agreement, designed to promote cooperation, mutual understanding and business efficiency. It vibrates with the dynamism of change and also with the optimism of change. It stresses the need for business flexibility and adaptation, both for companies and for employees. It underlines the need for continuous learning and skill formation but at the same time acknowledges the legitimacy of employee expectations of a rewarding and fulfilling life at work. The agreement proves that the union movement is prepared to look ahead and take an active part in a fast-changing work environment.

The years since the Development Agreement was signed have shown that the vision of the agreement was correct. There has been a period of stunning technological advance: information technology and telecommunication systems have revolutionized our way of perceiving human interaction. The era of standardized mass production in huge factories is being replaced by smaller and leaner facilities where the contribution of each individual is much more quantifiable and visible. The importance of each individual worker's knowledge and skill has grown considerably. Manufacturing is increasingly computerized, turning many blue-collar workers into highly specialized technicians. At the same time, the number of employees in manufacturing has gone down dramatically, and services are becoming the dominant economic sector. Work processes and employee qualifications are much more individualized in services than in traditional manufacturing.

The structure of the labour force has changed as well, with a core staff working on a full time basis. These are considered permanent employees in the sense that they are not likely to be dismissed. Surrounding them are people employed on different types of contract. Part-time and fixed-term employees form one group. Many work for long periods as part-timers or on successive fixed-term contracts and many turn into core employees, only to be replaced by others. Another group is composed of workers employed by independent contractors. Temporary workers are a third group. Temporary hiring has become much more common and has spread into new sectors, such as accountancy and research.

Three closely related tables illustrate the process.

Table 3a. Relationship between buyers and sellers of work
Agrarian society Industrial society Information society
Where? Live and work together. Live separately but work together. Live and work separately.
Work and private life intertwined Work and private life separate Work and private life partly intertwined
How? Status-relationship Contractual relationship Contractual relationship
Work: family duty Work: commodity Work: transfer of knowledge
Strongly hierarchical Strongly impersonal Strongly mutual


Table 3b. Wealth and value-adding factors
Agrarian society Industrial society Information society
Wealth factor Land (real estate) Capital (money) People
Value-adding factor Manual work: Muscle power Combination of means of production:
Machine handling power
Knowledge and creativity:
Brain power


Table 3c. Capital
Agrarian society Industrial society Information society
OwnersFamilyBuyers of workSellers of work
FormLandMonetary capital
(physical capital)
Human capital
(intellectual capital)
PurposeMaximizing family survivalMaximizing monetary capitalMaximizing quality of life

In the information society(Endnote 16) personal contacts and closeness (table 3a) are based on what the parties agree at any given time. So is their interdependence, but typically there is a close professional relationship: this relationship is mutual and the seller often holds the trump cards. Traditional capital (i.e. money and equipment) becomes less important as a wealth factor (table 3b). Knowledge and creativity are what count. Brainpower increasingly replaces machine power and brainpower is primarily individual like knowledge and creativity. The transition from an industrial society to an information society profoundly changes the role of capital. Since the relevant capital is knowledge, ownership moves from the buyers of work to the sellers, from capitalists/employers to employees/self-employed. Capital is disseminated to an increasingly wide section of the population so that everyone becomes a capitalist. Possession and control of capital are increasingly atomized.

Elements of both the agrarian and the industrial society remain in the information society but their relative importance gradually diminishes. A multiplicity of activities with radically divergent structures is the hallmark of the information society. The evolution towards an information society also means a trend towards decentralization and flexibility as it is primarily individuals who possess knowledge.

The move towards an information society is a challenge to unionism. Knowledge and creativity are individual and have little to do with standardization and collectivization. The core idea of unions—to monopolize the labour supply and remove labour from competition—does not seem compatible with labour supply patterns in an information society. Due to the importance of individual knowledge and creativity, sellers of work will become increasingly independent and self-sufficient. Increased individualism is likely to follow. Unions will not benefit from this. They will not benefit from the increase in unemployment either, since unions do not represent workers who are jobless.

Yet another factor that will not benefit unions is anticipated change in the market place. Product output will become much more varied and adjusted to the needs and wishes of customers. This will result in less rigid price structures and businesses will focus on maximizing income rather than minimizing cost. This, in turn, will make unions less necessary to the sellers of labour.

An increasingly individualized demand for goods and services will enhance the trend towards an individualization of the labour supply. People will be more discerning and specific when they sell their labour if this is how they act as buyers of goods and services. Raised buyer expectations go hand-in-hand with raised seller expectations.

The paragraphs above reiterate some standard explanations for the recent decline in union density rates in highly developed countries. These can be summarized in four points. (1) Changes in industrial structure resulting in fewer big factories. (2) Relative increase in the number of atypical (non-permanent) workers. (3) Higher education and better living standards resulting in individualism and less interest in unions, combined with an increased emphasis on employee mobility and lower levels of employee identification with the enterprise. (4) Traditionally low unionization rates in medium and small companies, because of employer resistance, little union interest and greater difficulty in organizing them.

However, there are at least some countervailing factors. First, the atypical workforce is more vulnerable than the core workforce. Many people on atypical contracts have working conditions that are far less favourable than those of core workers. It is a fact that the decline in unionization rates and the increase in the atypical workforce have happened simultaneously. The standard thesis is that the peripheral workforce is less inclined to join unions. In most countries unionization rates among atypical workers are lower than among traditional full-time employees. Unions in many countries show little interest in organising atypical workers. Second, all those in agriculture and industry face harder times. Price competition will intensify and management will increasingly concentrate on cost cutting. Labour costs will not escape their attention.

Swedish unions are obviously aware of the potential in agriculture and traditional industry. There is nothing surprising here and that aspect needs no further comment. The important question is what they do with the new situation.

As indicated by the 1982 Development Agreement, Swedish unions want to be partners in the ongoing process and they want to make a contribution. One way is to be instrumental in formulating rules for flexible work organizations, and unions take part in all approaches to increasing flexibility. This has been done while preserving the Nordic model, i.e. the model that relies on collective bargaining and collective agreements. As far as possible the general standards in collective agreements also apply to atypical employees. Where this is not possible collective agreements often establish special standards for atypical employees. Statutes provide for bilateral flexibility since they allow room for derogation by means of collective agreements.

Atypical workers in Sweden are not unionized to a lesser degree than core workers. Indeed, the opposite is true in that part-timers are more often union members than people on fixed-term contracts. One explanation is that the unionization rate among women is slightly higher than among men, and women account for the overwhelming majority of part-time employees. Women also dominate among temporary workers and they—as well as their male colleagues—benefit from what the leading union has done for them.

How have unions responded to the changing work environment and avoided a declining unionization rate? There is no one simple answer to that question. A slogan might be hard work. To elaborate, one might say that unions have shown tenacity and stubbornness in defending positions already won. Much more important, they have displayed considerable enthusiasm and inventiveness in opening up new vistas, greeting the changing work environment, adapting to new technology, and accepting that fear is always connected with change. Evidently their success has been greatly helped by a positive social context that supports them in a variety of ways. Perhaps even more important is that the employer community has not tried to exploit the new situation to get rid of unions.

One remarkable achievement is the fact that unions have managed to organize the temporary work business and negotiate collective agreements covering temporary workers. No business is more difficult to unionize than temporary work, but it has been done. The leading role was played by the Clerical Workers Union (Tjänstemannaförbundet HTF), which belongs to TCO. Starting from nothing only a few years ago it has negotiated a series of collective agreements covering the core sector of the temporary work business. As of September 1999 it is negotiating a new contract, having terminated an existing agreement early in 1999 despite that fact that this was spectacularly good from an international point of view. The focus of the HTF agreements is income security for temporary workers. Obviously, the employers—temporary work agencies—only want to pay their employees—the temps—for time actually worked and consequently billed by the agency. Just as obviously the temps want to be paid regardless of whether they work or not (as long as they are available to work, of course). Up to a point the HTF agreement provides for precisely that. The now terminated agreement guaranteed 75 per cent of full salary for the individually agreed working time of each empolyee/temp. In other words, the employer and the employee shared the risk for non-billable time but the employer carried the greater risk. HTF is now demanding that the employer assume full responsibility for non-billable time. This would offer full income security for employees/temps (provided, of course, that they are available to work as ordered and covered by their individual contract of hire).

How has this seemingly impossible feat been accomplished? No simple and clear-cut answer can be provided. Industrial action is conspicuous by its absence, so the answer does not lie there. Hard, imaginative and tenacious union work account for much, as does the employer response. It should be noted that temporary work was illegal in Sweden for some 50 years until deregulation in the early 1990s legalized it. However, legalization did not come easily and it was accompanied by strong expectations that temporary work agencies would see that fair standards and socially acceptable practices were observed; collective regulation of labour conditions is standard practice in Sweden. Another factor has probably helped the union despite the fact that it works against the temps and puts them in a weaker position. Established case law holds that temps are not entitled to unemployment benefit for time not worked (and not paid) by the employer/agency. The rationale is that temps are employed regardless of whether they are paid or not: in that sense they are not unemployed. This legal ruling means that temps are in fact deprived of pay altogether for the time not covered by the employer guarantee in the collective agreement. It has been made clear that no change in the law can be expected, so that the labour market partners have to find a solution. In a way society is helping the union in its quest for full compensation from employers. The need for full compensation becomes more obvious when unemployment benefits are unavailable. This means that the possibility of a full payment guarantee in a new collective agreement seems quite high.

This achievement would probably not have been possible if the union had not in the process rendered a service to the temporary work agencies. Incidentally, this is one of the fastest growing business sectors in the entire national economy. The union contribution is to make temporary work agencies an accepted feature of economic life. As indicated above temporary work was strictly prohibited in Sweden for decades. A general feeling of discomfort or even suspicion in many circles greeted the total lifting of the ban by a non-socialist government. The union movement campaigned against lifting the ban completely and advocated a return to the middle ground in force before total deregulation. A government commission proposed a partial retreat from deregulation. But the social democratic government, with the tacit support of the trade union movement, decided not to heed that proposal. Without the introduction of a collective regime by the HTF things would probably have developed quite differently for the temporary work agencies.

Another union success is the organization of atypical employees. There is no hesitation in recruiting atypical workers as members and unions vigorously campaign for them. To give just one example: unions have campaigned hard for part-timers to have the right to increase their working time if they so wish. Rules to that effect have become rather common in collective agreements and a statutory rule was enacted in 1996. A final example concerns flexible working time and business cycle variations, where unions have helped shape socially acceptable schemes. The 1982 Working Time Act does not deal with flexitime as such, but it does not permit employers to introduce flexitime or business cycle variations unilaterally. The Act provides for collective agreements to that effect and the social partners have ensured that virtually all collective agreements include elaborate rules on flexitime. Rules on business cycle variations in the total number of working hours are closely related to flexitime arrangements. These can also be agreed by means of a collective agreement. A breakthrough 1995 blue-collar agreement introduced rules to that effect in engineering and metalworking. Proposals for lifetime flexisystems have attracted wide attention, as have proposals for regular sabbaticals. So far these have not produced any concrete results which are generally applicable, but at least one company (the insurance giant Skandia) has introduced a system of recurrent sabbaticals so that its employees can spend time studying. The unions wholeheartedly endorse the scheme.

Union efforts concerning atypical employment have been protective and offensive at the same time. In both respects they have served atypical employee groups well and it is not difficult to understand why union density rates are no lower than among other groups.

Structure and finance

The organizational structure of Swedish unions is fairly uniform despite the fact that their origins differ considerably. LO, the oldest union federation, has served as a model for the other two federations, TCO and SACO, and their member unions. However, since many SACO unions have a long history as professional associations they have often developed characteristics all their own. The by-laws of the various organizations spell out the exact structure of each union and its internal working.

Generally speaking union governance is highly centralized. LO in particular has a very strong position vis-ā-vis its member organizations. The by-laws of LO parallell the corresponding by-laws of the member unions. LO has proposed model by-laws for its members, which are not binding but which establish certain standards as a recommendation. To gain membership in LO a union must meet certain mandatory standards.

The most striking feature is that LO controls industrial action by member unions to a considerable extent. Other provisions also confer power on LO: a) disputes between member unions are settled by a binding decision of LO; b) member unions have to await an opinion from LO before signing industry-wide collective agreements; c) LO must be consulted on all matters of major importance. Despite the power of LO, the member unions are truly independent organizations. They pursue diverging policies in many respects while at the same time striving for unity. Frequent meetings are held between the presidents of the member unions. Nevertheless, union officials are responsive to the rank-and-file of that particular union. For example, member unions do not necessarily adopt master agreements entered into by LO. The 1938 Saltsjöbaden Agreement on collective bargaining, grievance handling and prevention of certain kinds of industrial action provides an illustration. This agreement is considered as a cornerstone of Swedish industrial relations. The rules laid down have served as a model for the entire labour market and also for legislation. The norms have achieved the status of principles of law. However, some private-sector unions in construction and transport, which belong to LO, have never adopted the agreement.

The prime role of LO is to serve as the vanguard of blue-collar unionism, indeed unionism generally. According to its by-laws LO shall perform the central governance of the efforts of the trade union movement to look after and protect the interests of employees on the labour market and within the economy and in this respect as in other respects to promote social development on the basis of political, social and economic democracy (Article 1). The model by-laws adopted by LO for its member unions propose a similar wording.

Though using slightly less leftist language the by-laws of TCO and SACO state the same aim. However, the power of TCO and SACO vis-ā-vis their member unions is significantly less than that of LO. Neither has any control over industrial action. Unless specifically authorized, neither has authority to settle disputes between member unions.

Industry-wide unions have regional and local branches. The local branches are bargaining agents, usually at enterprise or workplace level. Regional branches are usually not bargaining agents: they perform a variety of services for the industry-wide union and the local unions. Support to local unions is at the heart of their functions. Though employees are members of the industry-wide organization, union dues are levied at regional level. Regional and local branches are legally independent entities but since union by-laws contain detailed rules on their operations there is little room for manoeuvres. The number of regional unions has declined dramatically among LO members in recent decades as a result of a determined policy to streamline the organization and raise the professional level of each regional union. In 1952 there were 8,915 regional unions, and the number had dropped to 651 in 1997. In 1952 total membership was 1.3 million but had risen to 2.1 million in 1997, so the declining number of regional unions is not related to a decline in total membership.(Endnote 17)

Local branches form the basis of unions and all but the smallest workplaces establish local branches. Since three federations and their member unions operate side by side and since the union of supervisors is also represented at most workplaces, it is common to find four local branches at any given place of work. Local branches negotiate with the employer. Given the recent trend towards a more decentralized collective bargaining structure, the importance of local collective agreements on wages and other conditions of work has increased. So, in its wake, has the role of local branches. Thanks to statutory rules on information and cooperation between employers and employees, local unions take part in virtually every aspect of workplace operations. In their dealings with employers concerning long-term planning and day-to-day operations, they have considerable room for manoeuvre. It is not the task of regional or industry-wide unions to quell local inventiveness and creativity in dealing with individual employers.

Local union officials are elected by direct membership vote. Office holders higher up in the hierarchy are nominated by the elected representatives of members rather than by direct vote. By and large unions are organizations of the one-party type in the sense that it is very uncommon for two or more factions to fight for control of a union. This is true at all levels of the hierarchy. When two or more candidates present themselves for office they virtually never represent diverging union platforms. They are distinguished by their personal history, character, age and professional background. Tenure is for specific periods, spelled out in union by-laws. For example, tenure as president of TCO is four years. Re-election is possible and no maximum period is specified, but in most unions elected office is held for a relatively short period.

In addition to their elected or nominated office holders, all union bodies except the local branches employ staff.(Endnote 18) At federation level LO had a staff of 220 in 1997. SACO employed about 40 people. Employees range from office workers to highly specialized professionals, such as economic analysts. All three federations (LO, TCO and SACO) maintain research departments, particularly for economic matters. These are staffed by university graduates, many with PhDs, so speakers for LO and TCO play an important role in socioeconomic debate and analysis in Sweden. The various industry-wide unions also employ specialists. At regional level the core employees do most of the grassroots work. They are often appointed on the basis of a membership referendum, even though they are employees. Most of them have a background as elected local union officers. Despite the importance of employed personnel at various levels in the hierarchy, union governance is firmly in the hands of elected office holders.

Unions have far-reaching authority to represent their members. This authority is partly statutory, partly contractual, based on union by-laws. Unions conclude legally binding collective agreements, interpret them and represent employees in the grievance process and before the Labour Court. Unions have authority under most labour statutes to conclude collective agreements with employers derogating from the statute. Such agreements are binding on members and non-members alike. On the other hand, unions also have obligations towards their members, notably to support and represent them. However, the exact union obligations are far from clear. No statutory rules exist regarding the relationship between unions and their members and case law is practically non-existent. The same is true in situations where member interests clash, for example in agreeing to priority lists of employees in mass lay-offs and terminations. Rules on union duty of fair representation are conspicuously absent.

The financial situation of Swedish unions is very good. They have three main sources of income: member dues, income from investments and contributions from employers. Member contributions are the foundation of union finances. Unions are free to decide the amount and, except in cases of discrimination, the courts have no jurisdiction. No reported case on discriminatory fee structures exists.

The dues that are levied differ considerably among unions,(Endnote 19) and no pattern can be detected among the three federations. Union dues are often higher in absolute terms in LO member unions than in SACO members despite the fact that average incomes are higher among SACO members. Some unions charge a fixed percentage. A member of LO, the Swedish Metal Workers Union, charges 1.9 per cent whereas a leading member of TCO, the Union of Swedish Salaried Industry Workers (SIF), charges 1 per cent. Many unions have a ceiling. A survey of 23 major unions within the LO, TCO and SACO families, conducted by SIF, reported the following findings. In 1998 on a 24,000 kronor monthly income (approximately 2,750 euro) a high of 549 kronor (approximately 62 euro) was reported by the LO Swedish Food Workers Union (Livsmedelsarbetareförbundet) and a low of 218 kronor (approximately 25 euro) by the SACO Union for Civil Engineers (CF). However, unions differ in terms of what they offer in return for dues. They all offer standard union representation, of course, but in addition many provide members with other services, such as accident insurance or home insurance, or both. Some provide discounts for members using union recreational facilities. Others offer medical insurance as part of the package. Many also offer collateral-free bank loans. But, again, it is difficult to discern a pattern.

Unions are free to decide how to use member dues. Nothing prevents them from using the money for purposes other than strictly union business, e.g. political contributions. Employees cannot join a union on condition that their dues are not spent in such a way. There is no equivalent to an American type agency shop.

Historically membership dues represented the main, if not sole, source of union income.(Endnote 20) The situation is radically different today. In many unions, dues account for less than 50 per cent of income. Unions have accumulated wealth through the years, primarily by building strike funds. Today these funds are considerable, allowing unions to engage in protracted industrial action if need be. Many unions are in a position to fight not just one war, but two or more at the same time. Union assets are invested primarily in real estate, stocks and bonds. The financial management of assets has become an important part of union management. For example, in fiscal year 1998 financial transactions accounted for some 65 per cent of total income in SIF, the biggest TCO-member union. Though this figure is probably higher than for most unions it still represents a common trend. Membership dues do not cover expenses. For example, in 1998 SIF recorded a 150 million kronor deficit (approximately 19 million euro). That equals 425 kronor (approximately 45 euro) per member in a union that charges an average of 2,400 kronor (approximately 265 euro) annual dues.

Union wealth is primarily owned and administered by the industry-wide unions. Local unions do not dispose of any investment capital nor do the three federations to any significant degree.

Contributions from employers are either direct or indirect, although cash contributions are unusual. These occur only in a few blue-collar unions, primarily in the construction industry. Such contributions are really payment for services rendered by the union in measuring piece work and calculating pay for that work. Employer payments here are supposed to cover union costs, no more. Under some construction industry agreements employers cover union expenditure for supervising employer observance of pay provisions in the collective agreement. There is serious doubt about whether the money is actually spent on this purpose. No one asserts that the money represents featherbedding. Indeed the parties deal with each other at arms' length.

The overwhelming majority of employer payments are indirect, taking the form of time off for union work at full pay. Originally based solely on collective agreements, such indirect payments are now mandated in several statutes. The union does not actually receive any money. It is relieved of the expense of compensating its voluntary officers for union work at workplaces. It is not known how much the employer community pays for union work of this kind but it can safely be assumed that the total amount is considerable. Unions take the position that indirect payment of this kind is no different from other business costs since work performed by union representatives benefits the employer.

A hotly debated issue in recent years has been employee-union-controlled investment funds, financed by employer contributions. Such wage earners' funds were introduced by statute in the 1980s and employer payments were collected for some years; the funds were dissolved in the mid-1990s, as they were considered to disrupt the social balance between capital and labour. Not even the social democrats, who guided them through the legislative process in Parliament, were happy with them. They had become prisoners of their own propaganda to have them introduced. On the other hand, the private wealth accumulated by unions has attracted little attention and virtually no criticism. Unions are seen as one investor among many.

Doubts about union fund capitalism are of a radically different kind, being mostly concerned with the ethical aspects of union investment. Unions are supposed to pursue investment policies that do not conflict with with socio-political agenda, e.g. they do not invest in companies using child labour. In 1998 TCO adopted ethical guidelines for investment. LO is actively promoting a common union front vis-ā-vis multinational enterprises to make them respect human rights, including the core ILO Conventions. Fund capitalism is one way to exert pressure.

So far union fund power has not been a factor of any particular importance in financial markets. The resources of the truly important actors in financial markets dwarf union wealth, however impressive.

Evidently, union expenditure is primarily aimed at maintaining the ability to represent members. Much money is also spent on education, information and public relations. Federations spend more on information and lobbying activities. In 1997, for example, LO devoted 38 per cent of total expenditure to policy and lobbying and 14 per cent to information. In the same year 11 per cent was spent on education and 9 per cent on supporting kindred organizations in Sweden or abroad.(Endnote 21)

Regional and global action (Endnote 22)

As mentioned in Section 5, Swedish unions are very active in the international arena. The three federations cooperate closely in this field, maintaining a common office in Brussels. There are some differences of opinion between them, mainly between LO and TCO on the one hand and SACO on the other hand, but in a wider perspective these are not significant.

Unions display apprehension but no real fear of Europeanization and globalization of the economy. Pointing at the risk of increasing opposition in a global arena, Swedish unions call for increased international union cooperation to meet globalized business. Strong union cooperation is necessary to tackle run-away capitalism and fast moving investment.

Swedish unions believe that the present context is rather like the situation in Sweden at the beginning of this century. At that time and for most of the century, attention was focused on the domestic scene. Now the perspective has widened and the EU has become the domestic scene. Beyond the EU is the rest of the world.

LO and TCO are among the most ardent proponents of international cooperation. They pride themselves on being among the chief initiators of the ETUC, in which the Swedish federations are very active. The Brussels office is the main centre for that work as well as EU activities. The prevailing attitude among LO, TCO and SACO is dynamism and openness to the changing work environment and the challenges of internationalization. The three are united in wanting the ETUC to play a pivotal role in shaping the ever-changing realities.

In comments on the proposed General trade union policy resolution submitted to the 1999 ETUC congress, SACO strongly advocates:

...a policy programme that is proactive, outreaching, and marked by a desire to meet problems and challenges head-on. The resolution should not in any way be marked by defensive attitudes; the European trade union movement should take care of the new possibilities in the Amsterdam treaty and establish a high profile.

It goes on to state that economic growth is necessary for new jobs and sustained welfare for all. Significantly, it adds that economic growth is nothing others create, it is a shared responsibility of governments, social partners, industry and other economic actors and concludes that ETUC should have a framework programme on this issue. In an ever-changing world, so the document tells us, Swedish trade union experiences have taught us that trade unions must tackle these changes in a forward-looking way, trying to anticipate and prepare, and formulate union strategies and tactics even before the changes hit with full force. In line with Swedish trade union traditions it advocates that ETUC should demonstrate that trade unions are active in favour of industrial change and development. The document also stresses the need for free trade and adds that free trade must go hand-in-hand with the promotion and defence of core labour rights. All these statements reflect long-standing union opinions in Sweden.

Swedish unions want a European model of industrial relations. They see this model from the Nordic perspective, i.e. strong unions, heavy reliance on collective bargaining and collective agreements, strong tripartite cooperation between employers, unions and the government. They advocate European collective bargaining and European collective agreements but are not pressing for these at the present time. They feel that mechanisms for uniform enforcement and adjudication must first be created. However, LO pushed hard for European bargaining procedures and considers the social clause of the Maastricht Treaty a big victory. A revision of ETUC's by-laws to handle European-wide negotiations and European collective agreements has also been carried through with strong LO input. Important elements here were consultation with national unions and a well-defined mandate for ETUC prior to negotiations.

On the global front three issues are currently at the fore. First, promotion of human rights, including core union rights. Second, the struggle against social dumping and support for social clauses in international instruments, such as the WTO charter. Third, safeguarding of union rights to engage in international, cross-border sympathy action. Unions are making a strenuous effort to have basic human rights, including core union rights, accepted everywhere and to have these included in international instruments.

Swedish unions strongly advocate free trade. At the same time they insist that free trade should be coupled with respect for basic human rights, including union rights, and also minimum conditions of work. They strongly advocate the inclusion of the core conventions in the WTO charter or at least a mention of them. They also advocate close cooperation between ILO and WTO. The same policies are pursued with regard to other international institutions, such as the IMF and the World Bank. Non-regulated export processing zones cannot count on support from Swedish unions!

The eight basic ILO Conventions—the eighth on child labour was added in June 1999—are seen as vital for a sound and fair economic world order. Unions do not deny that their motives are not solely idealistic. Solidarity and self-interest are two sides of the same coin, states a 1998 TCO discussion brochure (Europe is Part of the World). Swedish unions worked hard to achieve the 1998 ILO Declaration on fundamental principles and rights at work. It certainly did not harm those efforts that one member of the ILO Governing Body is the LO staff official responsible for international union work. Also, Swedish unions, at least LO, strongly advocate wider authority for ILO to monitor observance of its Conventions.

Swedish unions strongly support the adoption of company codes of conduct and are staunch supporters of the 1976 OECD Code of Conduct of Multinational Enterprises. The 1997 unilateral decision by Renault to close its factory in Vilvoorde, Belgium demonstrated both the importance of the code and the need to strengthen it.

Finally, Swedish unions strongly favour multinational cooperation and rule making over regional or bilateral arrangements.

9. Collective action and institutional support

A characteristic of Swedish labour regulation and practice is collectivization. This comes to the fore in all aspects of labour market functioning.

Swedish law and industrial relations practice have no rules like those in France or Germany, where the employee community is represented by an elected body separate from the union (Comité d'entreprise or Betriebsrat). Unions have monopolized employee representation in Sweden, as it were.

The predominant role of collective bargaining is the oldest and most conspicuous feature of the collectivist system. Historically, collective bargaining is the method for rule making on the Swedish labour market. Though statutory regulation has become quite common since the 1970s, collective agreements still retain their position as the prime regulatory instrument. In addition, statutes defer to collective regulation to a large extent. Anti-trust legislation does not apply to collective agreements proper.

The subject matter of collective bargaining covers all questions concerning the relationship between employers and employees (including unions). With very few exceptions there are no managerial exemptions. Collective agreements are comprehensive, covering the entire employment relationship and they are often very detailed, in particular when dealing with issues such as working time or vacations. Collective agreement regulation is more or less exclusive in some areas concerning the individual employment relationship, notably with regard to pay. Sweden has no legislation at all on pay, not even a minimum wage. Swedish unions (and the business community) adamantly oppose the introduction of anything like a mandatory income policy or minimum wage. Tentative proposals by the ETUC for some kind of incomes policy or minimum wage in Europe has met with downright rejection by Swedish unions.

Collective bargaining is very centralized, adding to the collectivist structure. Few nations with a market economy and privately owned industry have equally centralized bargaining. In the 1990s SAF has worked hard to decentralize the bargaining system and has met with some success, particularly in wage setting. The process will probably continue but it does not seem likely to bring about a profound change. Unions adamantly oppose a dismantling of the system with industry-wide bargaining as the nucleus of collective bargaining. The result of the power struggle is mixed. Industry-wide bargaining still commands the field but much of the fighting has been delegated to negotiations at local level.

The union federations are not bargaining agents per se but that does not rule out participation in negotiations or wage bargaining. Industrial relations are marked by a series of agreements between LO and SAF. Starting in 1906 with a compromise that is the foundation of industrial relations LO has taken upon itself to negotiate master agreements intended to cover the whole private blue-collar labour market. These master agreements are not legally binding on member unions but most members will subsequently adopt them as legally binding collective agreements in collaboration with their employer counterpart. The scope of LO involvement depends on the willingness of SAF to engage in negotiations. On the whole SAF agreed to negotiate with LO concerning matters of overriding interest. Agreements between SAF and LO have been common during most of this century on matters such as collective bargaining procedures and timetables, grievance procedures, prevention of industrial action, information and consultation, as well as safety and health at the workplace.

Another area where LO has been an active partner is in establishing basic norms for pay and pay increase. During the golden age of Swedish industrial relations in the decades after 1945 such negotiations were quite common. SAF and LO agreed on certain basic principles for pay and specified percentage increases, leaving the rest of pay bargaining to industry-wide unions and local branches.

Early in the 1990s SAF adopted a policy of decentralization and non-participation in direct negotiations. This brought an end to the periodic wage negotiations that had held Sweden in suspense for decades. SAF also took itself out of negotiating master agreements. It still negotiates on highly technical matters which require uniformity, e.g. private pension schemes, but otherwise SAF has delegated negotiation to its member organizations and even some master agreements previously entered into. The late 1990s saw some renewed interest on the part of SAF in negotiating directly with LO but so far no substantive result has been produced.

Legally binding collective agreements are concluded at all levels of bargaining, national, industry-wide and local (company or plant) level. Due to their by-laws the three national federations are not authorized to enter into collective agreements that are binding upon their member unions, but many agreements entered into by them are subsequently ratified by member organizations. The industry is the traditional focus of bargaining and there are comprehensive, industry-wide collective agreements for every sector of the labour market, private and public. Uniform standards apply to workplaces regardless of size or location. The overwhelming majority of employees are covered by a collective agreement, including management representatives from production line and supervisors to senior executives in both the private and the public sector. Some agreements cover vast sections of the economy. For instance one single agreement for blue-collar employees covers the core of the engineering industry. The agreement dates back to 1905 and it has been renegotiated from time to time. It is obvious that the structure of collective agreements also adds to the collectivist nature of Swedish industrial relations.

The exact number of industry-wide agreements is not known and does not much matter since every sector of the economy is covered. There is usually just one leading agreement in every sector and the others are mostly adaptations of this. Most industry-wide agreements affect a large number of employees. For example, 20 agreements in the blue-collar municipal sector cover about 635,000 employees, or an average of 32,000 employees per agreement. LO reports that some 7,000 industry-wide agreements were in force in 1997. Excluding the Metal Workers Union with some 6,900 agreements, 269 agreements covered 1.2 million LO members.(Endnote 23)

Extensive labour legislation in Sweden has not undermined unions or the pre-eminence of collective bargaining. In fact, in many respects it is the opposite.

Another very conspicuous example is in the legislation on employment protection. Such legislation can differ profoundly in the role assigned to unions: they are not necessarily given a role at all. This is a political decision. The interesting point is that Swedish legislation has not deprived unions of arguments to persuade employees to join them. Indeed the legislation is structured in such a way that it gives unions tremendous influence over employers and employees alike.

The 1982 Employment Protection Act (and its 1974 predecessor) is comprehensive and detailed. It can be implemented without additional regulation, e.g. collective agreements, work rules or individual employment contracts. In that sense the statute is self-sufficient. However, it is unsatisfactory from the employer standpoint for two main reasons: (1) It imposes significant restrictions on employers, limiting managerial flexibility in running the workplace and (2) the rules are based on individual employee needs rather than employer or collective employee needs. For example, seniority is an important factor in the statute whereas business efficiency is not. For these reasons employers want to derogate from the statute in many instances. Collective employee interests may also point in that direction.

One point concerns redundancy. The statute accepts bona fide business considerations as just cause for termination of employment contracts. No substantive union participation is needed here. However, the statute regulates the sequence in which employees are to be dismissed in redundancy situations where not all employees have to be laid off. The statute does not permit managers to decide which employees to dismiss, but provides detailed, mandatory rules on priority rating. The rules are exclusively based on seniority, defined as aggregated time of employment. Ability is a factor only to the extent that those retained must be able perform the work. Superior ability is not considered above that minimum level nor are other factors such as qualification, training, motivation or past record generally. Employer interests in retaining only the best-qualified employees or in composing a workforce to meet some specific criterion is not considered either.

These statutory rules call for adaptations to meet specific needs and the 1982 Act authorizes broad derogation, except in cases involving abuse or discrimination. Employers cannot derogate at their own discretion, nor can a labour inspector or the like give permission. Further, the statute rules out agreements between employers and employees. The one and only route is by means of collective agreements and only unions can be parties to an agreement on the employee side. Consequently, unions effectively control derogation from the Act.

The combined facts that employers strongly want, and often need, to derogate from the 1982 Act and that unions are in control of such derogation give unions a very strong position indeed in the administration of the Act. But it does not stop at that. A further point is that collective agreements under the Act apply not only to union members but to all employees within reach of the agreement as well. Unionized employees can influence the content of collective agreements through the democratic decision-making structures of the union. Non-unionized employees have no similar channel of influence. Nor do they have recourse to the courts (or any administrative agency) other than in exceptional instances of abuse of statutory authority to derogate.

Obviously the Act is a powerful tool for unions and demonstrates that protective employment legislation does not necessarily affect union power negatively. It is perfectly feasible to construct statutes that strengthen unions vis-ā-vis both employers and employees. The 1982 Employment Protection Act is just one of many statutes which have that effect.

The collectivist tradition, strongly supported by legislation, also permeates co-determination and procedural labour legislation, the basic law on co-determination being the 1976 Joint Regulation Act, MBL. The Act invites the employee side to participate at its discretion in the dynamic process of managing the company and handling day-to-day work operations. Again, unions exclusively represent the employee side. However, with just a few exceptions, only unions that have concluded a collective agreement on employment conditions with the employer are entitled to participate in management. No works councils or similar bodies exist. If, exceptionally, there is no collective agreement at the workplace the consequence is that the employee side has no right to information and co-determination as provided in the Act. Obviously the 1976 Act works in favour of unions. What is more it favours majority unions since they are generally the only ones that are strong enough to obtain a collective agreement.

With regard to administering labour market policies the position of unions is also strong, though not as strong as in the field of employment proper. State agencies carry the main responsibility for administering labour market policies but close cooperation with unions is a prerequisite for success in many instances. In one way or another unions are represented on the board of state labour market policy agencies.

One very important scheme is primarily administered by the unions, i.e. the unemployment benefit insurance scheme. Historically unemployment benefits were exclusively for union members and financed by member contributions. This was replaced by a public insurance system. Today the system is 95 per cent funded by state grants, financed primarily by employer contributions. Employee contributions are marginal, if not symbolic, averageing 100 Swedish kronor per month (approximately 12 euro). Leaving technicalities aside, unions are in fact the exclusive administrators of the state scheme.

Union membership is not required (and never has been) for employees to receive benefits under this scheme. The fact that unions administer it has nevertheless given them tremendous influence. It takes courage on the part of an employee to stay outside the union but ask for assistance when in need. Despite this, non-membership is not uncommon, especially in white-collar employment where 15 to 20 per cent of members in the unemployment benefit associations do not belong to a union.

A 1997 statute changed the unemployment insurance system to some extent, slightly reducing union influence. Some of the responsibility for administration was transferred to a less union- dominated agent.

Unions make no secret of the fact that power is at the heart of union administration of the state scheme. For example, in a 1998 interview a key LO official (Hans Larsson) bluntly stated: Unions administer the unemployment scheme for egoistic and rational reasons. Union density rates are lower in countries where the unions are not in charge of unemployment schemes.(Endnote 24) The connection between unions and the unemployment benefit system is often quoted as the best recruitment argument that unions have. If this is really the case it means that employees are either ignorant or timid since union membership is not a requirement for unemployment benefits. The employer community is highly critical of the system and advocates a complete transfer to the state social security administration. The non-socialist parties share that opinion but their terms of office in the late 1970s and early 1990s were not long enough to undertake the transfer. Such a move would have caused uproar among unions so perhaps the political cost was considered too high.

Collective action and social alliances

Unions maintain close contacts with a variety of organizations. This is especially true of LO, because the federation either started or initiated many of the organizations that surround it, including the Swedish Social Democratic Workers' Party, SAP (c.f. section 5).Other examples are the consumer cooperative movement, Folksam, the home and housing cooperatives, HSB, and the travel agent and hotel operator, Reso.

Nothing comes even close to the relationship between LO and SAP. These are the two (main) components of what is commonly referred as the workers' movement. This movement is a socio-political concept, a non-organizational phenomenon. However, it has a very strong emotional and attitudinal substance indeed. It constitutes a frame of mind and a way of thinking that permeates Swedish society and that continues to play an enormous role in the social fabric. The ideological platform of these two organizations is basically the same. They have divided the work between themselves, SAP dealing with the political field and LO being responsible for the professional field. Since there is no clear demarcation between these fields the two organizations often find themselves on the same turf. But SAP implements policies through political channels (legislation and local rule making) while LO is at the bargaining table with employers.

Strong economic ties have traditionally linked the two branches in the sense that contributions from LO members to SAP were the main source of income for the party. These ties still exist but they are much weaker than they used to be.

Membership in a union affiliated to LO used to mean automatic membership in SAP. Until 1987 it was common practice for LO-affiliated local unions to collectively enrol their members in the party. This is no longer done but financial contributions from unions to political parties have not stopped, in particular LO-union contributions to its party. This issue becomes more sensitive every year as many individual LO-union members vote for other political parties. This means that LO cannot promise to deliver the union vote to any party, not even its own party.

Close personal bonds also hold the two organizations together. For example, the president of LO is usually a member of the select group that makes up the powerful steering committee of SAP. SAP routinely recruits people from the professional branch at all levels of the party hierarchy, including the top level. Since the central government in Sweden has been headed by SAP for most of the time since the mid-1930s, many top representatives of LO have become cabinet ministers or have taken up other senior political positions. Recruitment in the opposite direction is less common, perhaps because there is no real need for it.

LO is represented on the boards of numerous organizations, bodies and institutions: a September 1998 list contains no less than 453 agencies. Some of these are LO bodies proper but the vast majority are not. They range from organizations close to the heart of the workers' movement, such as the insurance company Folksam, to bodies of little immediate concern to the core business of LO, for example the Criminal Detention Board, the Traffic Injuries Commission or the Central Bureau for Statistics. LO, of course, is also represented at the tripartite Labour Court.

TCO has no political affiliation or ties, although its leadership has leaned towards social democracy in the past decades. Significantly a former president of TCO is now (1999) a prominent member of the social-democrat government. The first female president of TCO has just been appointed to head a government industrial relations research organization close to the social-democrat establishment.

SACO maintains strict political neutrality.

Has the new situation in economic life produced new alliances? Yes and no. Since many alliances have existed for long periods of time there has not been a great need for new ones. As new needs arise the organizations change their agenda to cover these as well. Nevertheless some noteworthy additions have been made.

In recent years unions have become increasingly involved in environmental ('green') issues. The thrust of their involvement here is social justice in international trade. Union campaigns focus on solidarity with growers in developing countries who are competing with big multinational companies. A variety of just trade and just grown symbols and products have hit the market. Another area where union cooperation with other organizations has increased considerably is in consumer goods other than food, notably clothing. Unions are at the forefront of campaigns to force multinationals to increase transparency in their operations in developing countries and also to impose strict conditions on subcontractors in these countries to respect human rights, including union rights, and to refrain from exploiting workers.

Parting words

The picture presented here of Swedish unionism might strike the reader as overly positive. Are unions really so strong? Do they really face the challenges of a new work environment and an internationalized economy with so much aplomb?

The picture is indeed a positive one in the sense that it depicts a strong and bold movement, afraid of no one and willing to meet the challenges that arise. The movement is capable of innovation and it is open to new developments; it does not only look back on past achievements but also—and primarily—faces the challenges of today and tomorrow.

Critics of Swedish unionism tend to say that it has become too strong. It has woven itself into every corner of the social fabric. It has monopolized labour output and labour conditions. It has in fact put Swedish society into a union straitjacket.

Unions say that human rights, the welfare state, equality and everything else that unions stand for has to be defended, indeed created, every day over and over again. Nothing can be taken for granted so a continuous struggle is necessary.

It is a fact that unions have largely monopolized labour conditions. It is a fact that union strength is awesome in the labour market. Those who stand up against it do so at their own risk and pay a high price, often a ruinous price. It is a fact that SAF often prefers to have the political process rule on labour and employment matters rather than have such questions decided at the negotiating table with unions. Political decision makers, so the reasoning goes, take issues into consideration in their entirety. They cannot concentrate solely on what is immediately beneficial for union members, so the outcome of the political process should be more balanced between workers and employers.

Sweden was at the top of the OECD wealth league about 20 years ago but now it has slipped to a position near the bottom of European OECD-member countries. Critics blame much of that on the suffocating influence of unions. Unions tend to look the other way but, if pressed, would answer that the welfare state and equality have high costs in financial terms but that the overall quality of life must also be taken into account.

It is a fact that unilateral flexibility on the part of employers has diminished considerably in the past 20 years. Critics here see one reason for comparatively poor economic performance. Unions do not deny that unilateral employer discretion has diminished. However, they point at the fact that employers can obtain virtually any kind of flexible solution when cooperating with unions. Critics retort that unions control most of the flexibility arrangements and that this has been achieved by cunning, indeed spurious, lawmaking more or less dominated by unions. Unions strongly reject the very idea that they control the lawmaking process, pointing out—quite rightly—that SAF turns to lawmakers rather than to collective bargaining in many instances.

Critics say that unionism is smug, even arrogant. Unions behave as if they owned not just industry but the country at large. Unions respond that they are built on membership support and approval. They point at the dedication of their voluntary officers and they insist that their prime goal is to serve.

Critics say that the union movement is self-congratulatory. Unions respond that they are proud. Some might add gratefully so. Some might even say that they feel humbly proud.

And so the debate goes on. It is not for the present writer to pass judgement. It is certainly true that unionism in Sweden is strong. It is also true that employers, existing and prospective (entrepreneurs), are severely restricted in acting unilaterally. But does that justify the position taken by critics?

Suffice it to say that the union movement verifies an old saying: Nothing succeeds like success!

Data on some unions and union federations*

Federation and unionTotal membershipWomen
Municipal Workers' Union633,567511,151
Metal Workers' Union422,87488,053
Municipal Workers' Union180,180112,617
Civil Engineers' Union72,97211,948
Teachers' Union62,21739,664

* The data here are from Table 226, Statistik Årsbok '99 (Statistical Yearbook of Sweden 1999); cf. note 1.

Abbreviations and acronyms

AD Arbetsdomstolen (Labour Court)

ETUC European Trade Union Congress

ILO International Labour Organization

IMF International Monetary Fund

LAS Lagen 1982 om anställningsskydd (1982 Employment Protection Act)

LO Landsorganizationen i Sverige (Swedish Federation of Trade Unions)

MBL Lagen 1976 om medbestämmande i arbetslivet (1976 Joint Regulation Act)

OECD Organization for Economic Cooperation and Development

PTK Privattjänstemannakartellen (Cartel of Private Salaried Employees)

SAC Sveriges arbetares centralorganization (Central Organization of Swedish Workers)

SACO Official name of the federation of professional employees. Originally the name is an acronym for: Sveriges Akademikers Centralorganization (the Central Organization of Swedish Professionals)

SAF Svenska Arbetsgivareföreningen (Swedish Employers Federation)

SAP Sverige socialdemokratiska arbetareparti (Swedish Social Democratic Workers' Party)

SIF Svenska Industritjänstemannaförbundet (Swedish Union of Salaried Employees in Industry)

SOU Statens Offentliga Utredningar (Government White Papers)

TCO Tjänstemännens Centralorganization (Central Organization of Salaried Employees)

WTO World Trade Organization


Works in English

No major texts in English are exclusively concerned with trade unionism in Sweden. The books listed are of a general nature but some devote important sections to trade unionism, in particular the book by Bruun et al.

Adlercreutz, A.: Sweden, in International encyclopaedia of law and social security (Kluwer, Netherlands, 1999).

Bruun, N. et al.: The Nordic labour relations model: Labour law and trade unions in the Nordic countries—Today and tomorrow (Dartmouth, UK, 1992).

Fahlbeck, R.: Reflections on industrial relations, in International Journal of Comparative Labour Law and Industrial Relations, Vol. 12, p. 289, 1996.

——: Flexiblisation of working life: Potentials and challenges for labour law: An international analysis, in Acta Societatis Juridicae Lundensis, ISBN, 91-544-2391-0, 1998a.

——: Flexibility: Challenges and potentials for labour law, in Comparative Labour Law and Policy (formerly Comparative Labour Law Journal), Vol. 19, No. 4, Summer 1998b.

——: Towards a revolutionised working life: The information society and the transformation of the workplace, in International Journal of Comparative Labour Law and Industrial Relations, Vol. 14, No. 3, 1998c.

——Labour and employment law in Sweden, in Acta Societatis Juridicae Lundensis, ISBN 91-544-2301-5, 1998d.

——: Sigeman, T.: Sweden, in European employment and industrial relations glossary (Office for Official Publications of the European Communities, forthcoming).

Jonsson, B., in World Wide Workers, a book to celebrate the centenary of LO (Stockholm, ISBN, 91-574-5395-0, 1998).

Kjellberg, A.: Sweden: Restoring the model?, in Ferner, A.; Hyman, R. (eds.): Changing industrial relations in Europe (Blackwell, UK, 1998).

Neal, A.; Victorin, A.(eds.): Law and the weaker party: An Anglo-Swedish comparative study, Vols. I-III (Professional Books, UK, 1981-1983).

Nordic Council: The future of the Nordic model of labour relations: Three reports on internationalization and industrial relations (Nord 1993, p. 36, Nordic Council, 1994, ISBN 92-9120-368-8, 1994).

Schmidt, F.: Law and industrial relations in Sweden (New Jersey, Rothman, 1977).

World Wide Workers. A book to celebrate the centenary of LO (Stockholm, ISBN 91-574-5395-0, 1998).

Works in Swedish

Standard text books on labour and industrial relations in Sweden are not listed here. The books listed deal specifically with trade unionism.

Elvander, N.: Skandinavisk arbetarrörelse (Liber, 1980).

——: Den svenska modellen (Allmänna förlaget, 1988).

Hadenius, A.: Facklig organizationsutveckling: En studie av Landsorganizationen i Sverige (Rabén & Sjögren, 1976).

Kjellberg, A.: Fackliga organizationer och medlemmar i dagens Sverige (Arkiv, 1997, ISBN 91-7924-105-09).

Lewin, L.: Hur styrs facket? Om demokratin inom fackföreningsrörelsen (Rabén and Sjögren, Stockholm, 1977.

Nilsson, T.: Från kamratföreningar till facklig rörelse. De svenska tjänstemännens organisationsutveckling 1900-1980 (Arkiv, ISBN 91-85118-974, 1985).

Sandberg, P.: Tjänstemannarörelsen. Uppkomst och utveckling (Stockholm, Tidens förlag, 1969).

Westerståhl, J.: Svensk fackföreningsrörelse (Stockholm, Tidens förlag, 1945).

Endnote 1: For detailed statistical data see Kjellberg, 1998.

Endnote 2: For statistical data see the Statistisk Årsbok (Statistical Yearbook of Sweden). Data are presented in Swedish and English. When not otherwise stated, figures (and percentages based on them) in the text are from this publication and refer to 1997.

Endnote 3: See Kjellberg, 1997.

Endnote 4: For extensive statistical data see Kjellberg, 1997.

Endnote 5: For statistical data concerning LO and its member unions see Kjellberg, 1997. For more detailed information see the Annual Reports of LO.

Endnote 6: On the emergence of white collar unionism see generally Nilsson, 78

Endnote 7: For detailed statistical data see the Annual Reports issued by TCO.

Endnote 8: For detailed statistical data see the Annual Reports issued by SACO.

Endnote 9: For extensive explorations of jurisdictional matters, both inter- and intra-union, see Government White Papers SOU 1988:49—50, Arbetsmarknadsstriden. Gränstvister. Stridsåtgärder mot småföretag. III—IV.

Endnote 10: Cf. previous note.

Endnote 11: See Fahlbeck, 1996.

Endnote 12: A leading study is Lewin, 1997.

Endnote 13: See generally op. cit. note 9.

Endnote 14: Jonsson, 1998.

Endnote 15: LO Annual Report 1997.

Endnote 16: For a more extensive discussion see Fahlbeck, 1998a, 1998b, 1998c.

Endnote 17: LO Annual Report 1997.

Endnote 18: For details see the Annual Reports of the trade union federations and individual unions.

Endnote 19: No survey of an official or semi-official nature exists. The figures given are based on a survey of 23 major unions within the LO, TCO and SACO families undertaken by SIF and published in its membership magazine, SIF-tidningen 1998:17.

Endnote 20: Figures in the text are based on a survey published by the daily Svenska Dagbladet, 16 May 1998.

Endnote 21: Data from LO Annual Report 1997.

Endnote 22: The text in this section is based on unpublished union material and informal interviews with union officials.

Endnote 23: Data from LO Annual Report 1997.

Endnote 24: Swedish monthly Sunt Förnuft 1998, p. 5.