by Conor Cradden*

Anyone who has a passing acquaintance with economics will be familiar with the free rider problem. Free riders are economic actors who avoid paying for public goods because these goods are ‘non-excludable’ — their benefits are available to anyone, regardless of whether they have contributed to the cost of providing them. For trade unions, the free rider problem takes the form of non-membership in unionized workplaces.

Even though collective agreements generally cover all the workers in a particular category, membership is almost never 100%. Hence, workers who are not union members get a ‘free ride’. They benefit from the union’s bargaining power without paying for it.

Logically, there are two ways in which unions can try to solve the free rider problem. They can choose a regulatory solution, which is to say that they can aim to make it obligatory for everyone who benefits from union services to contribute to the cost of providing them, just as all citizens are obliged to pay taxes. Alternatively, unions can adopt a market solution: they can aim to convince workers that certain important benefits of union membership are excludable; that there are social or financial goods and services which are only available to union members and which are worth paying for. In this way unions can be sure that a high proportion of workers will join of their own free will. Putting this distinction in less academic terms, what it comes down to is that unions can either try to make non-members join, or they can try to persuade them to join.

The regulatory solution

Regulatory solutions to free riding take the form of union security practices. These are arrangements arising from collective agreements, or very occasionally from labour legislation, that commit employers either to insist that workers be union members, or to give union members some kind of preferential treatment, or to insist that workers pay dues to the union even if they are not members (see Table 1).

It may seem to be the most obvious answer to the free rider problem, but there can be little doubt that the use of union security practices is in decline. While it was once very common to find security clauses in collective agreements, particularly in the English-speaking world and in Scandinavia, over the last twenty years they have increasingly been subject to challenge by right-wing governments and employers (see Table 2). Only in South Africa, where COSATU successfully argued to include the protection of union security practices in the post-apartheid constitution, does the pendulum seem to be swinging the other way.

Whether union security practices are thought to be legally justifiable depends on the interpretation of the principle of freedom of association, as enshrined in articles 20 and 23 of the United Nations Universal Declaration of Human Rights (UDHR) and ILO Convention No. 87. While there can be no doubt that this principle is the legal bedrock of the positive right to organize, what is less clear is whether it also protects the parallel negative right: the right to refuse to join a union. Article 20 of the UDHR takes what appears to be a clear stand, stating without qualification that “No one may be compelled to belong to an association.” However, the ILO position is that “systems which prohibit union security practices in order to guarantee the right not to join an organization, as well as systems which authorize such practices” are compatible with Convention No. 87.[1]

Despite the ILO’s studied neutrality on the issue, arguments that the principle of freedom of association includes both the positive and negative right to associate have increasingly been used to justify the banning of the closed and union shops. This type of argument has been less successful with respect to the agency shop. Of the six main types of union security practice (see Table 1), the agency shop is unique in that it does not seem to represent a violation of the negative right to freedom of association. A 2003 decision by one of the EU’s law-making bodies seems to confirm this view,[2] and both the Australian and New Zealand Councils of Trade Unions have argued in favour of bargaining fees on the same grounds. At least in principle, those who pay fees are merely paying for the collective bargaining service they receive from the recognized union. They are not obliged to contribute to the cost of ‘political’ campaigning which they may not support.

Table 1: Union Security Practices

  • The (pre-entry) closed shop, the hiring hall and the preferential shop
    These three types of practice are intended not just to ensure that union membership is as close as possible to 100%, but also that the union has some control over the supply of labour. In a closed shop, employers agree to hire only workers who are already card-carrying union members. In the case of the hiring hall, the union also acts as the direct supplier of labour, providing recruits in response to employer requests and maintaining control over selection criteria. The preferential shop is a weaker form of closed shop in which an employer agrees to give preference to union members in hiring decisions, but where under certain circumstances non-union labour is allowed.
  • The union shop (post-entry closed shop) and maintenance of membership agreements
    Rather more common in practice than the union card approaches, these practices have as their sole aim the maintenance of as high a level of union membership as possible in a unionised workplace. In the union shop, the employer deals with recruitment in the conventional way, but all employees are contractually bound to join the appropriate union once hired. A maintenance of membership agreement is an element in a collective agreement that stipulates that all those employees who are members of the recognised union at the point at which the agreement is struck will remain in membership — or will at least continue to have union dues deducted from their wages — for the duration of the agreement. The same applies to new employees hired during the term of the agreement.
  • The agency shop (compulsory bargaining fees)
    In an agency shop, workers are obliged to pay fees to the union for acting as their bargaining agent, regardless of whether they are members. These fees may be less, equal to or more than normal union dues.

How important are union security practices to the labour movement?

It could certainly be argued that since the right to refuse union membership has not been clearly established in international law, appealing to the principle of freedom of association in order to justify outlawing union security practices is simply a cover for anti-labour policies that aim to make union organization as difficult as possible.

The ambiguity of international law could simply be a reflection of the wide variation in how national legal systems treat union security practices (see Table 2), something which itself may well reflect the fact that there is a genuine case to be made on both sides. There has always been a range of attitudes to free riding and to union security practices within the labour movement that cannot be explained by variations in national legal systems.

The extent to which the presence of free riders is actually experienced as a problem by union members and activists, and therefore the importance which union security practices have in each country, depends very much on the history and traditions of trade unionism in each national context. It cannot be assumed, therefore, that union security practices would be seen as desirable by unions in those countries where they are unlawful.

In Italy, for example, as Marco Biagi has argued, the basic attitude of the labour movement is that trade unions should represent the interests of all workers, not merely those who are members. The tolerant attitude to non-members — who are not really thought of in the pejorative sense of being free riders — arises from the perceived need to “guarantee a maximum of pluralism and spontaneity within the labour movement, and to avoid union bureaucratisation.”[3] Perhaps even more importantly, there is clearly a point at which the goal of promoting stable union organization and collective bargaining systems has to give way to the basic workers’ right of self-determination. For this reason, the protection of workers from pressure to leave one union in order to join or form another, and the parallel guarantee that they should be allowed to do so if they wish, are highly valued within the Italian labour movement.

Finally, a note of caution about the apparently attractive strategy of the agency shop. Enacting as a legal principle the argument that union activity can be split into political and non-political elements may in certain circumstances cause problems for union movements. It is very difficult to define what union goals and policies count as ‘political’, and so it may be possible for anti-union employers and governments to define the ‘legitimate’ and hence legally-protected activities of trade unions very narrowly.

Since April 2004, for example, US Federal Government contractors have been obliged to inform employees who are union members of their ‘Beck rights’, that is their right to demand that their union refund the (unspecified) proportion of their membership dues that are used for political contributions or other activity not related to administration of the collective bargaining agreement. This move, based on a 1988 Supreme Court judgement, was arguably only possible because the agency shop is permitted under Federal law.

The market solution

Regardless of the arguments for and against compulsory union membership and bargaining fees, the open shop is a reality for most national trade union movements. In the absence of any legal possibility of resorting to union security practices, these movements have had to ask whether the free rider problem is not in fact more usefully conceived as a recruitment and organizing problem. Indeed, in the English-speaking world, interest in new organizing strategies like those promoted by the AFL-CIO in the USA, the TUC in the UK and the ACTU in Australia seems to have come about as a direct result of the increasing difficulty of implementing union security practices and the increasing influence of the so-called ‘right-to-work’ movement.

From the organizing perspective, free riders actually represent less a problem than a good opportunity for unions to increase membership. The fact that the workplaces in question are already unionized means that recruitment activities can be conducted in “a relatively benign environment that… does not necessarily require an intense union-building effort.”[4] Given the relative ease of access to potential recruits and the active or passive support of the employer, unions can concentrate on communicating the substantive arguments in favour of membership.

Research on what arguments for membership are the most persuasive shows that there is no magic solution. In the end, and regardless of the national context, it comes down to persuading potential members both that the union is an effective force for change in the workplace and that membership carries significant individual benefits. Perhaps most importantly, recruitment depends on convincing non-members that in campaigning for more equitable employment relationships, for fair trade and for sustainable economic development, the unions have right on their side.

Table 2: Union security practices around the world

In most European countries outside Scandinavia, union security practices (USPs) are either extremely rare (for example in Belgium and Holland) or have historically been prohibited altogether as in France, Italy, Portugal, Spain, Germany, and Greece. The situation in European Union law as it applies to those member states where USPs have traditionally been used, notably Norway, Denmark and Sweden, is not clear. The Norwegian Supreme Court, for example, ruled the closed shop illegal in 2001, subsequent to the inclusion of the European Convention for the Protection of Human Rights in Norwegian law. While the Council of Europe’s Committee of Ministers has argued that the closed and union shops (but not the agency shop) are contrary to the Convention, the European Court of Human Rights has yet to rule on the issue.

In the USA, union security practices are permitted in Federal law, but individual states are allowed to legislate against them. Currently, 23 states have so-called ‘right-to-work’ laws, which outlaw the closed, union and agency shops. Campaigns in favour of these laws are supported and organized by the anti-union National Right-to-Work Committee and the National Right-to-Work Legal Defense Fund. George W Bush is among those who support ‘right-to-work’ legislation at Federal level, although attempts to get such a law through Congress have so far been unsuccessful.

In Britain before 1980 the union shop or ‘union membership agreement’ was a common feature of collective agreements. It and all other forms of union security practice were outlawed by the Thatcher government.

In Canada union security practices remain legal, but are subject to sustained attack by anti-union groups, notably the National Citizens’ Committee.

In New Zealand, the union shop was outlawed in 1991.

In Australia, the right-wing Howard government outlawed the agency shop, the last remaining legal union security practice in Australia.

Although there are still many countries — for example the Republic of Ireland, Japan, Mexico, the Philippines, and South Africa — where union security practices have a more or less unchallenged place in the legal and industrial relations systems, these appear to be increasingly unusual. Switzerland is also a rare exception to the middle-and-southern European rule, the agency shop being a common feature of collective agreements there.

References and resources

The academic debate
Unfortunately, the academic debate on the union free rider problem is rather specialized both in its approach (drawn overwhelmingly from labour economics) and its subject matter (workers and employers in the USA). This is not to say that it is not of some interest, however, particularly to the extent that many writers have tried to identify the key factors in the decision to join or not join a union in the already-unionized workplace. In this respect, the articles by Booth & Bryan, Chaison & Dhavale and Wheeler & McLendon are the most interesting. Of the other works listed, Olson’s book is the classic statement of the free rider problem, Harbridge & Wilkinson’s article is of interest because it considers a non-American industrial relations situation, and Delaney’s paper provides a useful overview of the US debate along with with a provocative but defensible argument that union security is an idea which is simply inappropriate in the contemporary socio-economic and political context.

Booth, A. & M. Bryan (2004) “The Union Membership Wage-Premium Puzzle: Is there a Free Rider Problem?”, Industrial and Labor Relations Review Vol 57: 3.

Chaison, G. & D. Dhavale (1992) “The Choice between Union Membership and Free-Rider Status”, Journal of Labor Research Vol XIII: 4.

Delaney, J. (1998) “Redefining the Right-to-Work Debate: Unions and the Dilemma of Free Choice”, Journal of Labor Research Vol XIX: 3.

Harbridge, R. And D. Wilkinson (2001) “Free Riding: Trends in Collective Bargaining Coverage and Union Membership Levels in New Zealand”, Labor Studies Journal Vol 26: 3.

Olson, M (1965) The Logic of Collective Action; Public Goods and the Theory of Groups, Cambridge, Mass.: Harvard University Press.

Sobel, R. (1995) “Empirical Evidence on the Union Free-Rider Problem: Do Right-to-Work Laws Matter?”, Journal of Labor Research Vol XVI: 3.

Wheeler, H. & J. McLendon (1991) “The individual Decision to Unionize”, in The State of the Unions, G. Strauss, D Gallagher & J. Fiorito (eds), Madison, Wis.: Industrial Relations Research Association Series.

Freedom of association in national and international law

Labour law is another rather specialized field, but the ILO has tried to make it as easy as possible for non-specialists to get access to information and guidance. Most ILO materials are available in French and Spanish as well as English.

The ILO’s International Observatory of Labour Law has produced an extremely useful series of national labour law profiles aimed at non-lawyers. 23 countries are covered.

Another indispensable ILO resource is the Labour Legislation Guidelines. These aim to “equip those involved in the process of formulating and reviewing labour legislation… with tools to make social dialogue on labour legislation more effective”.

Of more use to those with some legal background is the ILO Library & information services index of sources of national law, arranged by member state.

The Universal Declaration of Human Rights is available at:

ILO Convention number 87 is available at:

Recruitment and organizing

There is quite a lot of published work available on union organizing and revitalization around the world. For those interested in what’s going on, a good place to start is the ILO’s International Institute for Labour Studies Discussion Paper series. All the papers are freely available for downloading.

The British Trades Union Congress has recently been taking the issue of organizing very seriously. This is reflected in the quality of the materials available on the organizing and recruitment section of their website.

The AFL-CIO in the USA has a useful page with links to a wealth of generally applicable arguments and ideas about union membership.

[1]   ILO (1994) General Survey 1994. Freedom of Association and Collective Bargaining: Right of Workers and Employers to Establish and Join Organizations, Geneva: ILO; paragraph 100.

[2]   European Industrial Relations Observatory (2003). Report available at

[3]   Biagi. M. (1998) Report on Six National Case Studies in the Field of Freedom of Association, Geneva:ILO (Bureau of Workers’ Activities).

[4]   Heery, E., J. Kelly & J. Waddington (2002) Union Revitalization in the United Kingdom (Discussion paper 133/2002), Geneva: International Institute for Labour Studies.

* Conor Cradden is a research fellow in the Department of Sociology at the University of Geneva. He also runs the Twitter feed for this network. See Conor’s particular interests are in economic sociology, the sociology of work and employment, industrial relations, the labour movement, corporate social responsibility and business ethics. An abbreviated earlier version of this article first appeared in the Public Service International Journal Focus in 2004, and later in International Union Rights, the journal of the International Centre for Trade Union Rights. ( It is reproduced here with permission.

Copyright © Conor Cradden 2004-2012