Freedom of Association and Collective Bargaining
The committee's template to evaluating a state's compliance with freedom of association and effective recognition of the right to collective bargaining requires examination of three categories of indicators:
- the legal framework (at all levels of government),
- the government's performance (at all levels of government) in implementation, and
- the overall outcomes.
Legal Framework
An assessment must begin with an examination of the content of labor laws on organizing, bargaining, and engaging in strikes. Some analysts first determine whether a given country has ratified Conventions No. 87 and No. 98. But the use of ILO ratifications as an indicator of respect for workers' freedom of association is fraught with problems. Many countries widely considered to violate labor rights in serious ways have duly ratified the two conventions, while other countries widely considered as having better records have not.
Turning to national laws, it may be possible to determine with relative ease whether the legal system affords freedom of association to workers. Laws that require workers to become members of government-run federations, that place extensive prohibitions on bargaining, or that ban strikes altogether or in "essential" industries so as to support a particular national economic policy are markers of a lack of commitment to freedom of association and the right to collective bargaining.
For many countries, the inspection of its laws and regulations would immediately reveal that the legal framework does not meet the principle embodied in Convention No. 87, that:
. . . all workers and employers, without any distinction whatsoever, shall have the right to establish and, subject only to the rules of the organization concerned, to join organizations of their own choosing.
Many laws exclude agricultural workers, for example, from freedom of association and collective bargaining rights, and some countries restrict or prohibit public sector workers from organizing or collective bargaining. Some countries exclude domestic workers or other workers who do not work under contracts of employment.
Most countries' laws are a mix; some features comply with freedom of association standards and some do not. For example, Compa notes (2002, p. 12):
U.S. law forbids discrimination against workers because of union activity, a sine qua non for protection of the right to organize. However, U.S. law also excludes from coverage large segments of the labor force -- agricultural workers, domestic workers, low-level supervisors, and 'independent contractors' who are really dependent on one employer for their livelihoods. The National Labor Relations Act excludes about twenty million private sector workers. These workers can be fired for union activity with impunity. In twenty-seven U.S. states, collective bargaining by public employees is prohibited. On its face, this is a clear violation of the right to bargain collectively. But the situation is multi-faceted. The First Amendment protects public employees' rights of association, prohibiting their discharge for union activity. It also guarantees their right 'to petition the government for a redress of grievances. In many states that prohibit collective bargaining, public employees form unions and "bargain" their terms and conditions of employment with the legislature through the political process.
Spain's constitution and labor laws permit the right to organize and to strike, but the country's immigration law takes away these rights for undocumented workers. South Korea's labor regulations include many restrictions on the ability to organize and the ability to strike.
Once the assessment of compliance moves beyond a threshold of "easy" analysis of labor law texts for obvious conflicts with guarantees of freedom of association and right to collective bargaining, increasingly complicated issues arise. These issues include possible legal restrictions on the ability of workers to organize (by category of workers, such as civil servants or workers without contracts of employment, or by sector such as in agriculture). Some countries' laws may make distinctions or discriminate in the right to organize (excluding, for example, legal migrant workers). National laws may also restrict the right to collective bargaining -- by category of worker, by sector or by the subject of bargaining (such as exclusion of some issues). Laws may permit government interference, for example, by allowing the government to dissolve unions without legal recourse, to impose burdensome union registration procedures, limit the formation of national unions, prohibit or limit multiple unions within a single plant (including minority unions), or restrict who may serve as a union member, official, or adviser.
National laws and constitutions may restrict unions' political activities, either by establishing close relationships between unions and political parties (at one extreme) or by prohibiting union contributions to parties and candidates (at the other).
Other legal provisions might forbid certain kinds of strikes (such as "sympathy strikes" or "protest strikes") or strikes under certain conditions (such as during an "economic crisis"). Some laws limit actions associated with strikes (such as picketing or occupation of the workplace), while other laws may allow employers to fire striking workers and to hire permanent replacements. Yet other laws may restrict union members' rights to ratify or reject agreements negotiated by their leaders. Finally, legal provisions may limit the formation of worker-advocate nongovernmental organizations (NGOs), or the participation of external union representatives, worker-oriented NGOs, or compliance monitors in investigating and reporting on labor practices. In assessing these complex aspects of a nation's legal framework, an assessor must decide whether simply to follow ILO jurisprudence as far as it goes or to move beyond ILO jurisprudence on issues that are more contentious and less settled.
The committee generally followed ILO interpretations in developing its indicators for assessing compliance (International Labor Organization, 1994; International Labor Organization, 1996; Tajgman and Curtis, 2000). These 21 indicators are a partial -- but by no means exhaustive -- checklist for evaluating a country's legal framework relating to freedom of association and effective recognition of the right to collective bargaining, including the related issue of right to strike:
A-1. whether the country has ratified ILO Convention No. 87;
A-2. whether there are legal provisions that entitle workers or employers to establish and join organizations of their own choosing without previous authorization;
A-3. the extent to which there are legal restrictions on the ability of certain categories of workers to organize (such as civil servants, teachers, or workers without contracts of employment), and numbers of workers in each such category;
A-4. whether there are certain sectors where there is no right to organize (such as in export processing zones or in other tradable sectors or in agricultural and informal sectors);
A-5. whether there are other forms of distinction or discrimination in right to organize, such as race, nationality, sex, opinion, political affiliation, or citizenship (for example, excluding legal immigrants);
A-6. whether there are legal provisions that permit the government to interfere in freedom of association by workers or employers;
A-7. whether there are legal restrictions on political activities of unions or employers' organizations, either by establishing a close relationship between trade union organizations and political parties, or by prohibiting all political activities for trade unions (including making financial contributions to a political party or candidate;
A-8. whether national laws protect workers from discrimination if they join a union or participate in union activities;
A-9. whether the country has ratified Convention No. 98;
A-10. whether collective bargaining is protected in law;
A-11. whether there are legal restrictions on the mechanism of collective bargaining, (such as the mandatory exclusion of some issues or the ability of employers to refuse to bargain with a recognized union);
A-12. the extent to which there are categories of workers not permitted to negotiate a collective bargaining agreement (such as civil servants, teachers, or workers in special activities), and numbers of workers in each such category;
A-13. whether there are certain sectors where there is no right to negotiate a collective bargaining agreement (such as in export processing zones or in other tradable sectors, or in agricultural and informal sectors);
A-14. whether the principle of a strike as a means of action of organizations is generally recognized;
A-15. the extent to which the government can forbid certain kinds of strikes (such as "protest strikes," "sympathy strikes," "go slow" strikes, or "work-to-rule" slowdowns), or strikes under certain conditions (such as economic or political "crisis"), or strikes requiring a majority of workers involved to authorize a strike;
A-16. the extent to which the government can limit picketing or occupation of the workplace.
A-17. whether legal regulations ban employer lockouts;
A-18. whether workers are legally able to ratify or reject agreements reached by union leaders;
A-19. whether legal regulations permit employers to dismiss striking workers, or permit hiring of permanent strike replacement workers;
A-20. assessment of extent to which "right to work" laws or other "free rider" provisions undermine the ability of workers to organize and the extent to which laws require workers to join a given trade union as a condition of employment or that new workers be hired through a given trade union; and
A-21. whether workers' and employers' organizations may legally affiliate with international bodies.
Government Performance
The task of assessing a government's performance in implementation of freedom of association and effective recognition of the right to collective bargaining requires investigation of both the level of effort a government devotes to this objective and the effectiveness of the process. Effort and effectiveness depend on the magnitude of resources available to the government and the urgency of competing claims on those resources. The assessment involves an appraisal both of inputs and of outcomes.
Assessing compliance begins with investigating a government's performance in enforcement. This assessment involves whether and how often, union organizers are jailed, exiled, fired, or murdered without prompt and effective prosecution. It also considers whether and how often there is retaliation against strikers, and, if so, whether the government promptly and effectively prosecutes such retaliation. Other issues to consider are possible defects in the complaint process (such as excessive delays or expenses or light punishments) and whether the government enforces freedom of association and collective bargaining laws uniformly across all sectors and areas (including the farm sector and informal employment and export processing zones). Assessing government performance also requires scrutiny of the extent of labor-related corruption (such as control of unions by criminal figures) and the likelihood of prosecution. The assessment also involves examination of the independence of the judiciary in handling labor cases. Finally, it is necessary to examine whether trade unions are independent of control by the government or political parties and the ability of trade unions to influence public policy, such as through contributions to parties or candidates.
Assessing a government's performance cannot be limited to enforcement, but should also include the positive agenda. As the ILO and other organizations frequently point out, ensuring observance of core labor standards has a pro-active, educational, and facilitative dimension (International Labour Organization, 2000a, 2000b). Many workers do not know what their rights are or how they might pursue them. They may not even understand the concepts of freedom of association and collective bargaining. They might not know where to seek assistance.
The positive agenda of promoting compliance includes governmental efforts and effectiveness in educating workers about their rights and remedies, in building capacity for government officials with responsibility for labor matters at the federal, state or provincial, and municipal levels, and in facilitating dissemination of best practices in workplace cooperation, through unions, employer groups, labor-management organizations, labor-oriented NGOs, and what the ILO calls tripartite social dialogue (International Labour Office, 2002). Such dissemination includes encouraging the use of domestic and international channels about problems, difficulties, or violations of freedom of association and right to collective bargaining (such as submission of "Observations" by workers' and employers' groups to the Review of Annual Reports under the Follow-up to the ILO Declaration on Fundamental Principles and Rights at Work).
Finally, the assessment of government performance must consider the sheer magnitude of resources devoted to compliance with freedom of association and effective recognition of the right to collective bargaining. This assessment involves the adequacy of budgets and personnel assigned to departments with responsibility for labor regulation in comparison with the number of workplaces, adjusted for per capita gross domestic product [GDP], the caseloads of labor administrative bodies and labor courts, and the frequency of labor inspections of workplaces. As indicated above, the level of development and the competing needs faced by a government will limit the amount of resources that might be able to be devoted to enforcing compliance with core labor standards.
The committee proposes 13 indicators -- an extensive, but, again, by no means exhaustive list -- for assessing a country's performance in implementing freedom of association and effective recognition of the right to collective bargaining, including the related issue of right to strike:
B-1. the extent to which union organizers are jailed or exiled by the government, or fired, injured, or murdered without prompt and effective prosecution on the part of the government;
B-2. the extent to which strikers suffer retaliation without prompt and effective prosecution on the part of the government;
B-3. whether there are defects in the government's complaint process, such as excessive delays or expenses, light penalties, or nonpunishment of offenders;
B-4. the extent to which the government interferes in freedom of association (e.g., by declaration of martial law, "state of crisis," or by suspension or dissolution of associations by administrative authority);
B-5. the independence of trade unions from political control;
B-6. the ability of trade unions to provide support for political parties and candidates;
B-7. government actions to combat labor-related corruption (such as control of unions by criminal figures for use as a protection racket or for financial skimming) without prompt and effective prosecution;
B-8. the adequacy of personnel and budgets of labor regulation departments compared to number of workplaces, the frequency and adequacy of labor inspections, the caseloads of labor administrative bodies and labor court, and whether bribes are paid to labor inspectors by employers without effective prosecution;
B-9. the efforts on by national, state or provincial, and municipal authorities to educate workers about their rights and remedies and of the effectiveness of those efforts;
B-10. the efforts on the part of national, state or provincial, and municipal authorities to engage in capacity-building for government officials with responsibility for labor matters and of the effectiveness of those efforts;
B-11. government actions in encouraging consultation and in facilitating dissemination of best practices in labor-management cooperation, through unions, employer groups, labor-management organizations, labor-oriented NGOs, and tripartite social dialogue;
B-12. government actions in hindering or facilitating the formation and functioning of labor-advocate NGOs and of independent labor inspection, monitoring, and certification organizations; and
B-13. government actions in encouraging and enabling utilization of domestic and international channels about problems, difficulties, or violations of freedom of association and effective recognition of the right to collective bargaining.
Overall Outcomes
General indicators that are not necessarily under the control of a government may show compliance with freedom of association and effective recognition of the right to collective bargaining. These indicators include union density, frequency and length of strikes, and the percentage of workers covered by collective bargaining agreements. Each of these indicators presents special difficulties in assessing compliance.
Union density might be considered a good objective indicator of compliance with freedom of association, but the information must be considered in context and cannot simply be used alone. First, union density numbers are often not reliable. Trade unions often try to create an impression of strength by inflating their membership numbers, and governments also manipulate trade union membership data. Second, high union density numbers may not be indicative of genuine freedom of association because national governments or political parties may exercise control over membership. China and Egypt, for example, impose explicit governmental control over unions. In India, each major political party has its own union. The presence of "yellow" unionism -- unions controlled by employers -- can also affect union density figures, as can control of unions by government officials or criminal elements.
Even where numbers are accurate, one still finds a lot of room for interpretation. Both France and Spain exhibit low union density when measured in terms of dues-paying membership, on the order of 8-10 percent. But they both have relatively high density when measured in terms of the percentage of workers covered by collective agreements, estimated at more than 70 percent. Countries with a large informal sector can have low union density relative to the entire labor force, but it may be high in the formal sector.
Low union density in countries where freedom of association is respected could occur if workers have simply not opted for union representation or where other forms of worker-management interaction are prevalent. The use of union density as a proxy for compliance with freedom of association is built on the implicit assumption that, in the presence of genuine freedom of association, all workers would join a trade union, but this assumption might be at variance with reality (Freeman and Rogers, 1999).
Finally, while several organizations (such as the ILO and U.S. State Department) may provide estimates for union density (and for other general indicators discussed below, the numbers may all derive from a single origin, such as official government reporting. Alternatively, the numbers may come from distinct sources using diverse methodologies (such as those from studies by The North American Free Trade Agreement (NAFTA) secretariat).
Similar critiques apply to other overall measures. For example, the extent of coverage of collective agreements may, or may not, be an accurate measure of the right to bargain collectively. Many countries have a master collective agreement bargained between union and employer federations. The terms of these agreements are extended to all firms in a sector, even though workers and employers in the sector were not directly involved in the bargaining process. Such master agreements can result in wage levels that workers' productivity (or market clout) would afford them with free bargaining, but governments and government-run unions can also impose master agreements that hold wages below (or boost wages artificially above) the outcome that would result from free bargaining. The assessment of master agreements requires investigating the extent to which members of the unions in the federation delegate their bargaining rights to the federation and are able to approve or disapprove of the bargains struck.
There are other difficulties with using the sheer numbers of workplace unions and collective agreements to measure respect for workers' freedom of association and bargaining rights. One difficulty springs from whether a country permits minority unionism and multiple unionism and multiple agreements in the same workplace. In France, for example, Confederation generale du travail (CGT), Confederation Francaise Democratique du Travail (CFDT), and Force Ouvrière bargaining groups represent workers in the same plant. In Spain, the Union General de Trabajadores (UGT) federation and the Comisiones Obreras bargain on behalf of employees in the same plant. Each of these multiple unions negotiates a separate contract. These two countries might be compared with other countries where labor regulations stipulate exclusive representation by a single union required to be supported by a majority of workers to obtain one contract. For two countries that would appear similar in terms of size of the labor force, firm size, and other characteristics, indicators for the country that allows multiple unionism in a single workplace might display three or four times as many unions and collective agreements as indicators for the country with exclusive representation.
A large volume of strikes -- measured in terms of frequency, length, or person-days -- can be taken to indicate respect for the right to strike with a healthy collective bargaining system, or it can indicate the reverse. A small number of strikes might mean that labor and management bargain freely and readily compromise to reach agreements. But a small number of strikes can also be symptomatic of a repressive system in which workers are afraid to strike for fear that they will suffer reprisals or be replaced.
Thus, each of the committee's four indicators for assessing overall outcomes needs interpretation to be used to measure compliance with freedom of association and effective recognition of the right to collective bargaining:
C-1. union density;
C-2. frequency, length, and person-days of legal strikes;
C-3. percentage of workers covered by collective bargaining agreements; and
C-4. incidents of discrimination against union organizers, unions, or employer associations.
