Accueil » 20-2 ( 1965) » La convention collective de travail et la paix industrielle en Europe

La convention collective de travail et la paix industrielle en Europe

Jean-Réal Cardin


Dans un article précédent, l'auteur s'est efforcé de dégager le climat nouveau dans lequel s'inscrivent les relations de travail en Europe.

Dans la présente étude il tente de dégager, à même les systèmes européens de relations industrielles, les caractéristiques les plus significatives ainsi que les institutions de droit et de fait informant la négociation et la convention collective de travail et concourant à la paix industrielle dans les pays étudiés.


Collective Bargaining and Industrial Peace in Western Europe

This article is not intended to give a descriptive and detailed account of the institutions and legal structures which, in each phase of labour relations, organize relations between the partners in every country under review. Such an analysis would be too long and too exhaustive in view of the scope of this study and, in any event, would only repeat what has already been done in many excellent publications readily accessible to and well known by those who may be interested in such matters. Instead, an attempt will be made to determine from the European industrial relations systems the most significant characteristics, the most interesting formula for Canadians, the institutions in law and fact which can best be compared to our own institutions in conformity with the purpose of our study. We will attempt to do this in respect of each of the main phases of labour-management relations as they operate in the countries under review.


It should be noted at the outset that, in matters of negotiations and labour relations more generally, one distinction should first be made. Whereas some countries like Great Britain, Sweden and Denmark maintained almost intact their pre-war legislative and institutional frameworks, the collective negotiation systems of the countries of the European Economic Community, i.e., France, Belgium, Holland, West Germany and Italy, all underwent transformations, at times drastic since 1945, through new or renewed constitutional provisions or amended or entirely different legislation.

Most of the changes were based on an increasing acceptance of the existence of trade unions and on a better defined and more liberal status for collective agreements. During the period of intensive reconstruction, from 1945 to about 1950, the liberty of the parties in determining salaries was obviously kept at a minimum by restrictive regulation.1 But, with the exception of this brief severely « controlled » period, the trend was generally in the direction mentioned above. Even in the Netherlands where controls are still relatively severe, because of the prevailing situation, some relaxation has become apparent in recent years. One indication is the substitution of the Labour Foundation to the State Mediators College for the approval of collective agreements. Also, and this may appear to be paradoxical at first sight but quite logical after proper examination, it was precisely during this period, from 1945 to 1950, that new formulas now accepted were tried out and underwent their first practical tests.

The Status of Collective Agreements

It must be pointed out that, for our purposes, collective agreements generally in European countries, in Great Britain and Scandinavia as well as in the Common Market countries, consisted primarily of extensive regulations covering individual labour contracts. It is not comparable to our local agreements which cover and control every detail of industrial life at the plant level and in day to day working relations. It is relatively free of substantial regulations governing working conditions in the various industrial units to which it applies.

It is necessary to make these comments in view of the situation on this side of the Atlantic. It must be pointed out, however, that for the last fifteen years or so in Common Market countries and even longer in the Scandinavian countries, considerable progress was made in improving agreements either at the level where they are negotiated or through the development of enterprise agreements themselves, which are now much more prevalent than they were before the war.

Agreements consist primarily of « minimum compulsory standards of individual contracts between each employer and his employees ». 2 This situation derives from the conditions in connection with industrial relations systems in Europe, but it also demonstrates the still powerful influence of common law in Europe generally on labour relations, in contrast with what one might surmise at first sight. If it is true that « sociologically » several countries such as Great Britain and Scandinavian countries as well as Belgium have,in practice, gone far beyond pure legal formalism, the fact remains that legally individual contracts regulate in the final analysis day to day working relations between individual employees and management. This is apparent, for example, in the settlement of disputes concerning general working conditions. In any event, collective agreements have no legal status of their own in several countries, notably in England, Belgium 3 and Italy. Evidently, however, they are often in practice more effective there than in countries where they have a real legal status. The fact remains however that, since they are designed as minimum standards often without substance and leaving a great deal to the sole discretion of management in determining most actual working conditions, they do not contribute significantly to sound co-operation in industrial relations at the local level. 4

The Parties to Agreements

France and the Netherlands require that a union be a legal entity to sign collective agreements as defined in labour legislation. Other countries, such as Sweden and Germany, recognize the obligation for parties to an agreement to « keep the peace » while the agreement is in force and penalties are provided for in this matter.

This point is interesting for industrial peace. In England, where trade unions are not legal entities, and are protected by statute from legal actions in their endeavours in pursuing their economic interests in labour relations, certain parties recently expressed the opinion that, in order to check non-official strikes which have been so numerous in recent years, trade unions should be given legal entities, respect agreements as contracts and therefore be legally responsible for the actions of their members, particularly the shop stewards, as is the case in Sweden and West Germany, for example, in the event of strikes or any other action taken in contravention of duly negotiated agreements.5

But what is more important for the purposes of a study such as this one, in connection with parties to an agreement, is that, in Western Europe and in every country under review, collective agreements still come under the jurisdiction in the vast majority of cases not of labour units at the local level or individual employers, but of vast employer and labour groups, generally at the level of a complete industrial branch and, in some contries such as Sweden and Denmark, at the level of the economy as a whole. 6 On the whole, the statutes reviewed are very clearly directed in this manner. So much so in fact, that one could assume that legislators intended in those countries to center labour relations negotiations on the greater units (federations and confederations grouping wage-earners or management) so that the resulting agreements would be more readily integrated in general economic and social policies in these countries, and that agreements entered into would have a « community » feature and negotiating parties would consequently be more conscious of the implications of their decisions on working conditions for the economy as a whole in such countries.

All legislative texts to define collective agreements as capable of being entered into by individual employers and trade unions without giving further details, but a wide range of provisions actually favour clearly negotiations beyond the local level, notably the industry branch level.

The Extension of Agreements.

There is the possibility provided for almost universally of the local extension of labour collective agreements in Europe, under certain conditions. It is in a sense the logical consequence of the negotiations system in these countries, which consider much more readily that the determination of working conditions results from vast regulations, undoubtedly negotiated for the greater part, but always designed to be used as standards for large sectors of the economy and therefore capable of being more readily integrated in the framework of general salary and working conditions policies, and one can only agree that this is somewhat opposed to the spirit of our North American industrial relations systems. With the extension procedure, public powers in almost all of the countries under review were able to obtain full participation of labour and management associations in the determination of working conditions, even in those sectors where they are not fully represented, and laying down at the same time certain conditions so that standards set by private parties are truly representative of the areas to which they apply and do not disrupt industrial sectors. The precautions taken in such countries before granting extension should be remembered if ever we attempted to follow this course more extensively.

Briefly, what are these precautions ? In France, under the provisions of the statutes of February, 1950, only those associations representing management and labour can enter into an agreement capable of extension. Management representation must be an « association » under the terms of the act, i.e., have legal status. The same applies obviously to labour.De facto groups can enter into « ordinary » agreements but not an agreement capable of extension. This is all the more so for an individual employer who could never negotiate towards extension as can be done under the terms of the Collective Agreements Act of Québec. Furthermore, groups on both sides must be « representative » of the trade group or sector covered, be it local, regional or national. Nothing in the texts defines « representative », but the courts and administrative practice have provided for this. The detailed enumeration of the criteria would fall outside the scope of this study Furthermore, they must include a certain number of clauses whose provisions themselves obviously remain free. Generally, extension covers the whole of a pre-determined industry branch and even, in some cases, to the whole of the branches at the national level, for example, the national intertrade agreement on supplementary pensions.

In Germany, at least 50% of the workers involved must already be covered by an agreement before it can be extended, and there are certain other conditions. In Italy, nothing in the texts positively provides for legal extension but in practice, under the constitution and the powers given in the 1959 statutes, the government has practised extension in certain cases, at least in matters relating to salaries. In Belgium, an agreement can only be imposed on non-contracting parties if it has been entered into within a joint parity commission where the parties involved must also be « representative ». In Holland, extension is not considered an exceptional procedure, because it applies automatically when an agreement is approved by the Labour Foundation, a consultative parity organization at the national level with which we will deal further on. As for Great Britain, we are familiar with their wages councils system, which can extend salary standards in trade union sectors to the whole of a branch or region. In Sweden and Denmark, this extension procedure is non-existent as such in law, but trade unionism is so powerful and the authority of negotiation parties is so effective that, in practice, the few employees not covered by agreements obtain the same advantages as the others. Furthermore, in Sweden, the provisions of the 1928 Labour Collective Agreements Act permit terms of an agreement to be in a manner of speaking « imposed » through pressures during negotiations disputes between an employer who is not covered and his employees, by those who are already covered (strike or sympathy action).

It must be noted that extension procedure is not a new development achieved during recent years. The fact remains that it is always practised and that it is still in most European labour laws, and this is why we take note of it here without evaluating its chances of disappearance or survival during the next few years.

Negociation Through Joint Organizations

One of the most common formulas used in Western Europe for collective negotiations, and for that matter for several other labour relations aspects (for example dispute settlements ) is what is called « paritism ». Paritism consists of any formula or any institution within which the social partners meet, discuss, negotiate and determine « jointly », generally without the help of a third party and without resorting to open dispute, the relations systems between them in connection with labour relations, and every other aspect of their relations on problems of any type. Paritism implies necessarily a certain equilibrium of forces, a mutual recognition of the diverging points of view of the partners, a determination to achieve compromise where possible, a certain consensus on the long-term objectives to be reached by each party, and, finally, a certain maturity which will incite the partners to make all possible decisions on their own, while limiting as far as possible resorting to arbitration and compulsory decisions by the public powers.

In our opinion this constitutes the most spectacular post-war development in certain continental European countries in relation to institutionalized co-operation in labour relations.

What differentiates the situation in negotiations co-operation as between Great Britain and the Scandinavian countries on the one hand and the countries of the European Economic Community on the other hand, is that in the former countries government abstention is almost total (joint decision procedures were developed in those countries as safeguards against state intervention) while in those countries of the latter group where they are most prevalent, measures were initiated following increased intervention of the public powers in the field of labour-management relations, after the Second World War. In such countries, paritism is not the same as in the former countries and is not the result of agreements between parties (free « contractual » system), but rather of co-operation with the public powers who included these matters in a « policy », by instituting them and by maintaining them with the support of the interested parties (accepted « statutory » system ).


The dispute settlement system in labour relations vary considerably from one country to another. It is therefore extremely difficult to sift out common features which could, by themselves, give a good idea of the overall situation in this connection. As for individual labour dispute settlements certain West European countries, at least from a legal point of view, have no other procedure than common law. In Holland, for example, there are labour magistrates, but they are operating within ordinary jurisdictions. In France, Germany, Sweden and Denmark, as well as in Belgium labour tribunals deal with these disputes. In Germany and Sweden for example, the decisions of these tribunals are final and without appeal. On the other hand, in France, beyond a certain amount, appeal can be made to appeal courts and to the social chamber of the Appeal Court. In England, nothing in the Act imposes recourse to labour tribunals which do not exist. An individual dispute is legally based on the individual labour contract (as is the case for the other countries mentioned above) and there can be, in theory, recourse to ordinary tribunals. In practice, a whole series of conciliation stages is provided between the parties from the local level up to the national level and, eventually, the dispute can be brought before the Industrial Court, whose decision in fact is not mandatory, unless the parties agree to it beforehand.

As to the distinction between legal disputes and disputes in interests, i.e., the interpretation or the violation of an existing agreement, during negotiations or the renewal of an agreement, certain countries make a clear distinction between these two types of disputes while other countries make no distinction at all. In the former group, mention could be made of West Germany, Sweden and Denmark where any dispute arising on the interpretation of a collective agreement must be brought before a labour tribunal, whether it is collective or individual ; this springs from a principle recognized in these countries that the signatories must respect industrial peace when they are bound by an agreement. The second group includes Belgium, Italy, England and France, where nothing in the statutes, as far as we know, prevents the eventuality of a strike during the term of an agreement in connection with a collective dispute, unless the parties had excluded this action in the agreement.

None of the countries have compulsory arbitration procedures for disputes of interests, at least as far as private services are concerned. Employer associations as well as employee associations were always firmly opposed to this, since, according to them, this would be tantamount to state salary controls, as is more or less the case in New Zealand and Australia. Management and labour representatives on the Belgian National Labour Council, during the study recently made by it of a proposed bill to organize collective social relations between workers and employers, were once again opposed to having a clause deal with this because « in connection with arbitration, the execution force results from arbitration compromise », achieved between the parties themselves if they deemed it advisable to submit to it.


In most European countries, conciliation procedures are a purely voluntary matter subordinated to the will of the parties. Obviously most of the countries have placed at the disposal of the parties, for a long time in many cases, conciliation and mediation services and the social partners themselves have often developed such procedures either through basic agreements to this end or through collective agreement themselves. The fact remains, however, that it is interesting to note how, in Canada, our labour laws are by far more stringent than those of these countries in Western Europe. In England, Germany, Belgium, Italy, the Netherlands, Sweden and Denmark, there is no legal obligation to resort to conciliation procedures.

Only France, in spite of the fact that it liberalized significantly its legislation in this field since 1936, still maintains compulsory conciliation (without procedures) in case of industrial disputes. Section 5 of the statute of February 1950 states that all collective labour disputes must be submitted to conciliation. But this conciliation does not rule out strikes and can take place during strikes. Furthermore, no penalty is provided for in case this obligation was not fulfilled, except that a fine can be levied for failing to appear before the Conciliation Commission.

If, as we have just seen, legislation is most summary and not compulsory in relation to conciliation or mediation procedures in European countries, this does not mean, and far from it, that conciliation and mediation are non-existent in their labour relations systems. Quite to the contrary, in most countries, conciliation is common practice on a voluntary basis. The social partners preferred to rely on their own initiative in this field rather than see the government intervene.

Joint Organizations

It must also be said that where joint organizations exist at the level of the branch activity, as is the case for example in England, Belgium and the Netherlands, these organizations are expected not only to negotiate agreements but also to settle disputes arising between the parties and they constitute, because of this, the best channel through which these disputes are taken up and settled. Paritism plays an important role in this sector. Since collective agreements are generally centralized at their level, these joint organizations constitute the logical terminal for the settlement of disputes. In our opinion, this is an important example of institutionalized co-operation allowing the parties themselves, often without the help of third parties, to settle their own problems without resorting in most cases to tests of economic strength or violence. This is also due to the fact that, in Europe, management and labour confederations effectively control their affiliated bodies and members.

England's experience in this field is quite revealing. This may be where the control of the central authority over the basic elements at the local level is at its weakest and least efficient both for employers and workers. The presence of the shop steward, who represents the union but holds his authority from the members he represents, often excludes any « real » presence of the confederation in the establishment. In Great Britain, the number of days lost through strikes or lockouts has increased in recent years in comparison to the other countries of Western Europe, in spite of joint industrial councils and long tested procedures for private conciliation at the industry branch level.

All this tends to show that everywhere where control from top to bottom is even slightly effective, co-operation is made much easier when it is achieved at a higher level, where the dispute is out of the hands of the interested parties, to be studied, often more objectively and without emotional factors, within parity organizations designed to find objective solutions.

The Swedish Experience

In our opinion, the Scandinavian countries, Sweden and Denmark, offer the best examples of industrial peace. Sweden is probably the country which has the

least legislative texts on labour-management relations and it could be said that « the parties themselves constitute the main centre of regulations and government of industrial matters ». 7

In Sweden, the existing legislation presently confîrms what the parties have been practising by themselves, according to their experience and the positions of power and strategy they occupied in the economic and political contexts. By itself the fear of state intervention was enough in Sweden to bring the parties to discipline themselves in disputes.

It should be noted that since the turn of the century, the Swedish Management Confederation and the Swedish General Labour Confederation agreed to keep at a minimum all possible regulations by the public powers, to recognize that during the term of an agreement a dispute in law should never be solved by resorting to force, to determine amongst themselves the nature of the disputes and to attempt to solve them by themselves. In fact, this last point had already been the subject of a summit agreement in 1908.

The 1938 basic agreement was primarily featured by industrial peace. It covers almost all phases of relations between the parties, negotiation procedures for the protection of neutral third parties in case of disputes and of economic sanctions, through voluntary and final settlement of disputes arising out of collective labour agreements, as well as the establishment of regulations designed to solve problems of working conditions : employment procedures, separations, lay-offs, firings, and all questions related to personnel management. Finally, procedures were set up to solve disputes which could become a menace to the proper operation of essential public services. The highlight of the 1938 agreement was the creation of the Labour Market Commission, a parity commission with multiple functions, such as negotiations and, on occasion, conciliation and arbitration. It should be added that there are other basic agreements, too numerous to mention, covering every wage-earner sector, which were agreed to in Sweden. They constitute a « private » network of incomparable labour-management co-operation.

The Swedish and Danish systems are original8in comparison with other countries in Western Europe. Other countries such as Germany and the Netherlands enjoy widespread industrial peace but it results perhaps more from legislation, institutional structures responsible to the public powers, or simply from strong public opinion which checks trade union power in periods of full employment and economic prosperity. Yet, in the case of Sweden, this peace is primarily due to the quasi-public status of the social partners and to their substitution, for all intents and purposes, to the State in the determination and application of procedures designed to safeguard industrial peace. In France, the relative absence of economic disputes is due to the weakness of unions and to the fact that, at the enterprise level, it is still the employer in many cases who makes unilateral decisions for a great number of problems. In Great Britain, where labour movements are strong and militant at the base, the weakness of the confederations in relation to their local units and to the T.U.C. in relation to both its affiliated bodies and the public bodies, together with the deficient nature of the structures, make tensions much stronger there than elsewhere and inhibit co-operation as it exists in Sweden. This is in opposition to what is happening in other European countries for, under pressure of full employment, the power of shop stewards was increased proportionately so that confederations are increasingly isolated from the very life of labour relations at the base. Consequently, this gap in the processes of authority within trade unions prevents actual control at the local level and allows disputes at the local level.

(1) Some examples are in France the 1946 Act limiting liberty in determining salaries and, in the Netherlands, the requirement for the approval of the State Conciliators Office, as of 1945, for a collective agreement to come into force.

(2) GÉRARD DEHOVE, « Le droit et la pratique des conventions collectives dans les pays de la C.E.E. », political social séries No. 6,Etudes Collection, European Economics Community, Brussels, 1963, p. 14.

(3) In Belgium, for example, as well as in England for that matter, collective agreements cannot apply legally to individual labour contracts even as minimum standards, except under royal decree in certain circumstances.

(4) It must be noted, however, that in England tradition often supplements agreements in due form and sociologically « determines » numerous items in working conditions.

The same applies to Scandinavian countries, where trade unionism is prevalent at the local level and where there are establishment contracts provided for by Agreements.

Also, in Germany, « enterprise councils » or « works councils » actually negotiate many things at the local level with interested employees.

Finally, in France, following the 1950 Act, general collective agreements were improved and establishment contracts have spread, especially since the 1955 Reynaud contract.

Nevertheless, we feel that our statement remains true generally.

(5) B.C. ROBERTS, Edit,Industrial Relations, Contemporary Problems and Perspectives Metheiea & Co. Ltd., London 1962, p. 15.

(6) The trend, both in fact and in law, during recent years in Europe towards negotiations on a much narrower basis will be reviewed later.

(7) T.L. JOHNSTON,Collective Bargaining in Sweden, George Allen and Unwin Ltd., London, 1962, p. 115.

(8) The Danish system is quite similar to that of Sweden and could have been cited as an example. Limitations of this thesis compelled the author to use one illustration only.