ILO is a specialized agency of the United Nations

International Labour Conference
85th Session 1997

The ILO, standard setting
and globalization

Report of the Director-General


International Labour Office    Geneva



I. Laying the groundwork for social progress and maintaining its momentum:
    A more universal approach

II. More targeted standards for a greater impact





In the thirteen years that have followed the last Report of the ILO Director-General on our Organization's standard-setting action, the economic, political and social environment has undergone a change of a magnitude that is probably unprecedented in the ILO's history. The fall of the Berlin Wall was a decisive turning-point, as it marked the end of a bipolar world which had witnessed the birth of the International Labour Organization -- and which, as some cynics would have it, was its main raison d'Ítre; and the disappearance of the planned economy system hastened the advent of a global market, whose laws, promises and risks now extend throughout the entire world.

In the Report Defending values, promoting change, submitted to the Conference on the ILO's 75th anniversary, I already referred to several aspects of the impact of these changes on the ILO's standard setting. Now that a few years have passed, I feel the time has come to stand back and re-examine the matter in a more systematic way. Three major considerations have convinced me of this. First of all, in 1994, the Conference, in the resolution concerning the 75th anniversary of the ILO and its future orientation, expressed the wish to be kept informed of the outcome of the examination of the ideas and proposals of the Director-General with a view to revitalizing the ILO and adapting its means of action to the changing world environment. The ground we have covered since then now fully justifies complying with this request. Second, the incidents which occurred during the discussion on the Convention on home work last year confirmed that there was a breach in the wide consensus which has always prevailed within the ILO in this area -- and which is indispensable to its functioning. Finally, the many flattering references made recently to ILO standard setting -- whether at Copenhagen or Singapore, at the G7 Summit on employment in Lille or the OECD -- have clearly shown us the extent of the renewed expectations and hopes placed in this aspect of our work and, consequently, our historic responsibility to act without delay so as not to fall short of these expectations.

Although this Report carries on directly from the one submitted on the ILO's 75th anniversary, I do not intend going over all the developments which have occurred since then, except to mention the opening up of two vitally important areas of activity which resulted from the 1994 Report. The first area of activity concerns the social dimension of the liberalization of trade, to which I shall refer in the first part. From the standpoint of standard setting, this work has focused attention on the particular significance of workers' fundamental rights in the context of trade liberalization and the globalization of the economy. The Working Party set up by the Governing Body to look into this issue and the Committee on Legal Issues and International Labour Standards have both examined various practical ways to strengthen the universal application of these principles. The second area of activity concerns the revision and updating of the body of standards, of which the ILO may be justly proud; considerable progress has been made and one of the first tangible results of this work is already before the Conference this year in the form of a proposal for a constitutional amendment to authorize the Conference to abrogate Conventions it considers obsolete.

The thoughts I shall be sharing with you in the following pages are mainly intended to help us continue examining in a systematic way not only the challenges facing the ILO but also the unprecedented opportunities opening up for the future standard setting of the ILO -- especially for adopting new standards as a result of the changes in the international environment.

As I already pointed out in 1994, ILO standards are not an end in themselves; they are one of the means -- undeniably the most important -- that the Organization has at its disposal to attain its objectives and ensure that the values enshrined in its Constitution and the Declaration of Philadelphia are put into practice. These values are those of the dignity of the human person, in particular the dignity of human beings at work and through work, which is expressed in the solemn affirmation of the principle that labour is not a commodity. We cannot therefore hope to renew and strengthen the consensus on the ILO's standard setting without considering the impact that the new international environment has both on the actors of social progress, to whom new standards are directed, and on the values of which they are the instruments. I feel it relevant to spell out briefly the nature of this dual impact.

Let us first examine the actors. The standard-setting action is necessarily, if not exclusively, to be carried forward by the member States. But although the complex phenomenon of economic interdependency resulting from trade in goods and services and capital flows, conveniently known as globalization, would seem to call for a global and universal approach, this same phenomenon affects the ability of States, under the pressure of increased international competition, to take up their role under the ILO Constitution.

Replies to the questionnaire on the social dimension of the liberalization of trade, sent out at the request of the Governing Body Working Party, show that globalization has forced many States to carry out legislative reforms to be able to cope with international competition as best they can. It is likely -- although this is not entirely clear from the replies -- that the relative decline in the ratification rate of Conventions might, at least in some cases, be due to a reticence to make long-term international commitments in these circumstances.

Globalization might also, albeit in a more indirect way, make it more difficult for States to assume their roles in standard setting because it encourages the establishment, extension or strengthening of economic blocs which are fairly autonomous. Once they reach a certain degree of integration, they find themselves in a situation similar to that of federal States -- at least as far as ratifying new Conventions is concerned.

Finally, globalization ushers onto the scene a host of new social actors, other than the ILO's non-governmental partners, who put forward their own independent demands for standards alongside those of the States' -- which may to a certain extent be in competition with them.

In global terms, the convergence of economies seems at any rate to hasten the need to make difficult choices or decisions -- hence the temptation to take the easy way out by not intervening or even by deregulating. And the result of this is that the ILO might also be forced increasingly to make choices, even to question the priority of its objectives, as its basic texts do not assign any hierarchy of these objectives. To give but one example, Part III(a) of the Declaration of Philadelphia recognizes the obligation of the Organization to further programmes which will achieve "full employment and the raising of standards of living", while advocating the extension of social security measures. Given the debate which has arisen around the existing links between unemployment, job creation and guarantees of job stability given to workers, the ILO can scarcely continue its standard setting -- if it is to retain any credibility or relevance -- on the premise that all these objectives can be taken on at the same time. It must, with facts in hand, demonstrate that this is possible or, as the case may be, propose a list of priorities among its objectives.

Let us now turn to values. Of course, the way in which the ILO's values have been expressed is not inalterable. It is in the very nature of things that they might sometimes have to be reinterpreted in the light of growing technological constraints or changing mentalities. However, globalization and the permeability of state structures raise a problem of an entirely different magnitude which affects the very concept of social justice or, in any case, a number of its vital principles such as "equal remuneration for work of equal value" laid down in absolute terms in the Preamble of the Constitution. Implicit in the Constitution is the notion that the context in which social justice is to prevail is that of each State, which is to be democratic; and in this the Constitution is no different from most philosophers who try to give this concept new life and make it operational. Globalization destroys this framework by placing greater emphasis on comparisons between workers in the same trade or industrial group and not only within the same country. With the deepening inequalities in the developed countries, there is a danger that globalization might increasingly make this principle seem a threat rather than a promise. Of course, there is no proof that globalization is the direct cause of these inequalities -- and I shall refer you on this matter to the last World Employment report.(1) Leaving aside the various interpretations which divide the specialists, it is highly likely that public opinion will continue to believe widely that globalization inevitably implies a downward levelling of pay for jobs of equal (low) skills in a market in which goods and capital can freely circulate.

Although globalization obliges us to re-examine the basic concepts and values of standard setting and specify their meaning or relevance, it also instils, in an insidious and more radical way, doubts as to the pre-eminence of these values. The end of the ideological, social and political confrontation which divided the world, combined with globalization, is ushering in a new Weltanschauung with the result that globalization is becoming an end in itself. By a strange irony of history, the "dawning of a new age" is no longer expected to occur with the end of the class struggle, which finally reaches its fulfilment in the withering away of the State; this time will only come now once the State has been stripped of its social and economic prerogatives and a global civil society emerges which is answerable only to the laws of economic rationality, itself the sole guarantee of a future so full of prosperity and promises that people forget the harshness of their present circumstances. This new form of the ideology of progress persists in affirming, like all that have preceded it, that human progress is more important than actual human beings; and it might well, as the same causes engender the same effects, result in the same disillusions.

Will the international community be spared these new disillusions? Will it be able to comprehend that, although globalization is undoubtedly an unequalled factor of progress and peace, it cannot be left entirely to its own devices? This will depend very much on us. I firmly believe the challenges I have just described also provide ILO standard setting with a new and exceptional opportunity, by placing it at the heart of the concerns of the twenty-first century. There are already signs of a healthy reaction to the infatuation with globalization, the obsession of com-petitiveness and the casting aside of values -- even amongst those who are its strongest advocates. Taking one example from among many, the views of one of the "gurus" of financial markets(2) are more than of merely passing interest. Criticizing the widely held view that globalization is irreversible because of the power of financial markets, he recently warned of possible serious repercussions because of the way the movement is spreading too rapidly and too widely; indeed he felt that "to have a stable system, you need some fundamental values to sustain it". He did not believe that unregulated competition automatically led to the best allocation of resources and stressed that we should "strive for certain fundamental values, such as social justice, which cannot be attained by unrestrained competition". These words are very close to the ILO's basic tenet: labour is not a commodity. Even if it were proved that child labour brings economic advantages to those resorting to it, it must still be abhorred by anyone with a healthy conscience.

The same concerns are starting to be expressed by the heads of international economic and financial organizations. The new Secretary-General of the OECD insisted very recently on the need to pay heed to the social tensions linked to globalization, particularly to discrepancies in wages which were tending to increase far too much and risked undermining social stability. This warning seems to echo the words of an American observer,(3) who recently published an alarmist and controversial article -- but which on this point, at least, is difficult to dispute: "The best thing that can be bequeathed to the next generation is social peace".

And this is precisely the basic and unswerving cause of the ILO. Although its role may no longer be seen as guaranteeing "universal peace" (in the sense of lack of conflicts between States), because, as stated in the Preamble of its Constitution, there can be no universal peace without social justice, its raison d'Ítre is still to guarantee social peace, without which neither the multilateral trade system, nor the financial system -- and by extension the global economy -- would be able to develop or even survive. It is certainly no coincidence that, contrary to what is widely believed, the financial markets tend rather to turn away from the paradise of ultra-liberalization and prefer regulated systems guaranteeing lasting and adequate economic growth, to avoid major social unrest and acute political crises.

To take up this challenge and live up to these expectations, the ILO must show that it is capable of making a choice between its objectives and the means of achieving them and ensure the necessary qualitative advance in its standard setting; the object would be, first, to go beyond the limits inherent in its dependence upon the action of States, in a world in which the realities of globalization know no borders; and, second, to provide each State on an individual basis with useful and specific guidelines enabling it to progress steadily towards its objectives. This qualitative advance would entail two approaches which are complementary and interdependent:


The Working Party on the Social Dimensions of the Liberalization of International Trade was set up by the Governing Body after the Conference in 1994. Despite its inevitable differences of opinion and the fact that it sometimes seemed to be going over the same ground -- or at times even to have come to a deadlock -- it succeeded in identifying the most important issues of trade liberalization which relate to the ILO's work -- and consequently to pave the way for useful and realistic solutions.

Without wishing to go over this subject again, I feel it useful to mention the essential conclusion that seems to emerge from these discussions -- that the liberalization of trade and the realization of the ILO's objectives are interdependent. Indeed, lifting restrictions on international trade lays the foundations for social progress -- as the ILO has always implicitly acknowledged, even during the worst years of economic depression. But, at the same time, this liberalization carries the risk, as the Preamble to the Constitution of the ILO warns us, that international competition, by inhibiting the will of certain Members to introduce progress, might be "an obstacle in the way of other nations which desire to improve the conditions in their own countries".

The ILO is indispensable for the consolidation of the multilateral trade system; however, this is not, as some might believe, because it should compensate for the distortions in competition likely to result from differences in levels of protection but because it should avoid the hopes this system raises in the mind of the general public from being dashed to such an extent that it would lose credibility. Whatever differing views are held concerning the impact of globalization on social issues, one thing is clear. The increasingly global economy is accompanied by restrictions and inequalities giving rise to tensions and upheavals, which are reported almost everyday in the press of the industrialized countries -- but also more and more so in developing countries.(4) The social unrest which abruptly took hold of the Republic of Korea, a country in which economic progress during the past 20 years has been nothing short of spectacular, clearly bears witness to the fact that these difficulties are not restricted to the traditional industrialized countries. Moreover, many observers pointed to the similarities between this unrest -- and the support given to it by public opinion -- with that which had rocked France one year earlier.

In the light of these developments, it is not out of the question that globalization might be jeopardized and give rise to the temptation of retreating back to protectionism. Above all, it might result in a tendency to form regional economic groups which are sufficiently autonomous from an economic standpoint and socially homogeneous.

As we now know what is at stake, we can determine our actual objective in more realistic terms and more easily define the ways in which to attain it.

Indeed, the aim is not for the International Labour Organization to achieve uniformity in the level of social protection in order to ensure a proper international competition. Rather the idea is simply to place social progress into a relationship with the economic progress expected from the liberalization of trade and globalization. Two vital and inextricably linked conditions to bring about this development have emerged from the discussions:


The discussions on the social dimension of the liberalization of trade, which followed on after my report in 1994, are far from over and, as I have already said, I do not intend taking stock of the situation here. It is, however, important that the Conference, which quite naturally and justifiably wishes to be kept informed of the results, should be aware of the progress achieved -- at least in mutual understanding -- and of the first stirrings of a consensus on a principle of vital importance, which have been felt as far away as Singapore.

In 1994 the "debate" over whether there should be a link between the liberalization of trade and the protection of workers' rights took the form, as between proponents and opponents of such a link, of mutual accusations of social dumping and protectionism. In my Report to the Conference in 1994 and in the document submitted to the Governing Body in November of the same year,(5) I tried to point out that this was a false debate because it was based on the implicit premise that trade liberalization should be subject to a certain level of standardization with regard to social protection -- but that this was in no way realistic or in line with ILO principles. Differences in conditions and levels of protection are linked to a certain extent to differences in levels of development. Denying developing countries the advantages (relative and transitory) which ensue from these differences would be tantamount to denying them a share in the profits of globalization and, by extension, the possibility of subsequent social development. The Declaration of Singapore, to which I shall refer later in the text, shows that we have come a long way in universally accepting these principles.

But if this approach is to be formalized, it presupposes the respect of certain common rules of the game. All the partners in the multilateral trade system must guarantee certain fundamental rights, without which workers cannot be assured of receiving their fair share of the fruits of economic progress generated by the liberalization of trade. The list of these rights seems no longer open to dispute: freedom of association and collective bargaining; the prohibition of forced labour, including forced labour of children; and non-discrimination (particularly in the form of "equal remuneration for work of equal value", stated in the Constitution).

This is why, since 1994, I have been insisting how important it is, in the context of trade liberalization, to guarantee fundamental rights which should allow the social partners to claim freely their fair share of the economic progress generated by the liberalization of trade.

The first time this approach was acknowledged outside the ILO was at the World Summit for Social Development in Copenhagen. After undertaking to promote the goal of full employment as a basic priority of their economic and social policies and to enable all men and women to attain secure and sustainable livelihoods through freely chosen productive employment and work,(6) the participants at the Summit agreed on the need to freely promote respect for relevant ILO Conventions, including those on the prohibition of forced and child labour, freedom of association, the right to organize and bargain collectively, and the principle of non-discrimination. In application of this commitment, I requested all governments which had not yet done so to indicate the measures they intended taking to ratify -- and apply -- the relevant instruments.

This approach has recently received further support from an OECD study(7) which analyses the impact of core standards on the position of the countries concerned in the light of international trade. After demonstrating from a theoretical standpoint that the implementation of certain core standards is likely to boost development and guarantee a distribution of labour resources in line with the requirements of a free market, the report shows -- on the basis of a more empirical study on the impact of two of these standards -- that any fears that the application of these standards might influence the competitive positioning of these countries in the context of liberalization are unfounded. On the contrary, they might even in the long term tend to strengthen the economic performance of all countries.

The latest important recognition of the particular significance of fundamental rights in the context of trade liberalization was at the Ministerial Conference of the World Trade Organization in Singapore; in their declaration, the ministers of trade, while agreeing that the comparative advantage of countries, particularly low-wage developing countries, should in no way be called into question, renewed their "commitment to the observance of internationally recognized core labour standards"; they also recognized that the ILO was "the competent body to set and deal with these standards".(8) The question which must now be raised is whether (and how) the ILO can ensure the respect of these fundamental rights by all the partners of the multilateral trade system, in line with the commitments they made at Singapore.

My reply is quite definitely "Yes".

After all, in order to ensure that these rights are universally guaranteed, it would be enough for the few partners in the multilateral trade system who have not yet done so to ratify the relevant international labour standards. The fairly encouraging success obtained by the ratifications campaign I launched in 1995 shows that this goal is perhaps not beyond our reach.

A more systematic -- and perhaps more inventive -- use of the means already at the ILO's disposal, particularly article 19(5)(e) of the Constitution, might consolidate these individual efforts.

Indeed, this provision allows for a leverage which has no equivalent amongst other international organizations. Although it respects the prerogative of each State to decide whether or not to ratify an ILO Convention, it nevertheless makes it binding for them to report on their law and practice in the area covered by the Convention. It is entirely up to the Governing Body to decide on the frequency of the reports it might request in this respect. It goes without saying that in the event of Conventions pertaining to rights considered fundamental, the Governing Body has every reason to request that these reports be submitted at more frequent intervals: this is precisely what it has already decided to do, at the Office's suggestion, by making arrangements for reports to be requested every four years for each of these rights.

There would be nothing to prevent us from going even further. The reports in question (or their absence) might not only encourage the States who have not yet done so to ratify them, but would also allow for discussions to take place on a regular, in-depth and tripartite basis (taking into account the comments made by the employers' and workers' organizations) on the situation of the countries in question concerning these fundamental rights. To have the required impact, this new approach would undoubtedly call for some adjustments being made to the procedure at present applied to all reports under article 19.(9) It might perhaps in particular increase the role of the Governing Body but it would remain perfectly faithful to the letter and intention of this provision.

Although making greater use of article 19 should make it possible to cast light on the situation in countries which have not ratified the fundamental Conventions, and thereby help to convince them that it might be preferable to do so, it could not impose basic obligations on them. There is another possible approach -- moreover complementary -- which might succeed in obtaining universal application of these rights; this would be to raise the question of whether, even in the absence of ratification of the relevant Conventions, all member States, by virtue of their acceptance of the Constitution, and the objectives and principles of the ILO, are not bound to a minimum of obligations with respect to fundamental rights.

There is already a precedent in the area of freedom of association, as the ILO has succeeded in attaining recognition of the universal value of the principles of freedom of association and the need for all its member States to respect them.

By taking as a basis the affirmation of freedom of association in its Constitution, it established a special procedure supplementing the general supervisory machinery for the application of international labour standards; under this special procedure, governments or workers' and employers' organizations may submit complaints concerning violations of trade union rights by States, irrespective of whether or not they have ratified the Conventions on freedom of association. These complaints are examined by the Committee on Freedom of Association, a tripartite body of the Governing Body with an independent Chairperson. The Committee carries out a preliminary examination of the complaints taking into account the observations submitted by the governments. It may recommend to the Governing Body, as appropriate: that a case requires no further examination; that it should draw the attention of the government concerned to the problems that have been identified and invite it to resolve them; or attempt to obtain the agreement of the government concerned for the case to be referred to the Fact-Finding and Conciliation Commission -- a much more cumbersome procedure which is used sparingly.

Since it was set up in 1951, the Committee has examined more than 1,900 cases, which has enabled it to build up a very full body of principles on freedom of association and collective bargaining, based on the provisions of the ILO Constitution and the relevant Conventions, Recommendations and resolutions,(10) and to take action which, even in the eyes of the outside world, is considered "reasonably effective".(11)

This prompts us to wonder whether this type of solution could not also be applied to other rights considered as fundamental in this context. This has been the main issue of discussion within the Governing Body. The debate has mainly focused on whether the Constitution and the Declaration of Philadelphia are explicit enough to consider that the respect of these basic rights is inherent in the membership of the ILO, even if the particular Member has not been bound by more detailed obligations resulting from the ratification of the corresponding Conventions. There are some who feel that this would require supplementing and clarifying existing texts by a possible "declaration" which would in some way spell out the intentions of the Constitution and the Declaration of Philadelphia. Others do not see the need for this, although they do not necessarily oppose it.

Without anticipating the outcome of the discussions, I feel that it would be useful for the Conference itself -- which is primarily concerned by this discussion because, when the time comes, it will be called upon to give a final form to its outcome -- to identify a number of points relating to the legal and constitutional scope of any such solution, having regard to the precedent of freedom of association. If we are to be successful in our venture, I feel it important to stress that, although the precedent of the freedom of association procedure is indeed relevant for the reasons the Office explained in some detail in November 1996, the fact that we have established its constitutional relevance does not necessarily mean that it can be transferred more or less automatically to other fundamental rights under consideration.

First of all, demonstrating that a precedent is valid merely provides a constitutional basis upon which to act; there is still a need to take a political decision to turn this potential to good account and set up an actual procedure. This need is even more important because the competence of the Organization to take action -- which derives from the existence of a constitutional basis -- does not require the same action to be taken in respect of each of the basic principles under consideration (and we do not have to be reminded of the misgivings that arose in connection with intervening in the fundamental but vast and multifaceted area of discrimination). A decision must therefore be taken; and this decision, whatever it is called and whatever form it takes, must reflect as wide a consensus as possible amongst all our constituents. Instead of continuing the legal debate on the relevance of the precedent of freedom of association, I therefore feel the time has come to outline a strategy for the promotion of other fundamental rights which would receive wide support, as this alone would guarantee the credibility and forward movement needed for a venture so vital to the future of the Organization.

Secondly, a decision of this nature is indispensable to circumscribe the rights to be protected, taking into account each one's specific characteristics. The Governing Body and the Conference -- and I feel it important to stress this point -- based the protection of freedom of association procedure on Conventions Nos. 87 and 98, as well as on the Preamble to the ILO Constitution, because this was perfectly natural and logical in the historical context. Indeed, this choice was entirely justified in the light of the initial approach which led to the present procedure -- that of the Fact-Finding and Conciliation Commission, which in any case depends upon state consent. Furthermore, these Conventions had just been adopted and only included straightforward principles, which seemed to spell out in a fairly succinct way the principles contained in the Constitution.

The same considerations do not necessarily apply to other fundamental rights. Unlike the Convention on freedom of association which sets forth principles or applicable rules in terms which could be described as timeless, these Conventions might, as in the case of forced labour, be strongly imbued with the concerns of their day -- or, as in the case of child labour, contain a complex provision aimed at attaining their objectives (in the Convention in question, fixing in stages the age of admission to employment). In this context, a declaration or any other text enshrining principles might help to extract from Conventions the universally acknowledged essence of the fundamental right, without running the risk of seeming to go back on or undermine the corresponding Convention.

Third, the fact of not following word for word the precedent of freedom of association should in no way imply taking a step backwards. It goes without saying that any extension should not undo what has been achieved in the area of freedom of association. Our task is to supplement the existing mechanisms and not to replace them. This matter will not, of course, be brought before the Conference until the political will to take a decision with a fairly well-defined strategy has crystallized within the Governing Body.

Having thus defined the basis and the content of the obligations inherent in membership, we would finally have to specify the mechanism for follow-up and implementation. Here again, nothing obliges us to copy the procedures adopted for freedom of association -- even if they provide a useful and interesting reference and case history. It will, of course, be up to the Governing Body and, subsequently, the Conference to elaborate the most appropriate procedure for the protection of the fundamental rights under consideration -- once these have been defined -- by taking advantage as much as possible of existing mechanisms and considering all the relevant logistic and financial implications. The follow-up mechanism could, as the case may be, be incorporated into the decision, whatever name or form we decide to give it, which would aim at establishing the principle and general framework of the protection of these fundamental rights and obligations -- as was done, in a certain way, in the case of freedom of association.


The emphasis placed in the previous paragraphs on the particular significance of fundamental rights in the context of globalization must not be misconstrued. Although guaranteeing these fundamental rights is a sine qua non for ensuring that social progress is "self-perpetuating", this will not be enough. A State cannot merely sit back and agree on the rules of the game; it is supposed to participate actively by seeing to it that liberalization is accompanied by social measures, for instance through the implementation of other ILO standards. There are many other rights which, without being termed "fundamental" (meaning that their implementation is considered a priority), are nevertheless of fundamental -- one might even say vital -- importance for workers; for an example, one has to look no further than certain occupational safety and health standards, without which there could be a heavy loss of human lives. Technological progress and transfers that accompany the liberalization of trade, and relocation of production, must be complemented by national safety legislation to avoid what some authors have called a "transfer of hazards". Reference was made to this notion when drafting the Tripartite Declaration of Principles concerning Multinational Enterprises and Social Policy.

More generally speaking, I feel it important to stress once again that the liberalization of trade and social progress must go hand in hand. To ensure the stability and consolidation of the multilateral trade system, upon which social progress depends because of the wealth it produces, there must be evidence that its promises are not vain or illusory -- or that its fruits are not too unfairly distributed. The main issue is not -- and I have not stopped repeating this since the beginning of this discussion -- to impose standard regulations from the outside. What matters is that each Member should try to act in accordance with its possibilities, in consultation with its social partners, and that these efforts can be objectively demonstrated at the international level.

There are no doubt those who would reply that the constitutional system of the ILO is already supposed to take care of this objective. On the one hand, in fact, the Constitution obliges every Member to submit within a time-limit of 12 to 18 months after its adoption any new Convention and Recommendation to the competent authority, which is usually Parliament; on the other hand, the instruments must be drafted in such a way to take account of the possibilities of each country. Consequently, any Member intending to fulfil its obligations in good faith should be able to progressively implement ILO standards, according to the means at its disposal.

For many years, this system did seem, albeit indirectly, to ensure that social measures were taken alongside those of an economic and trade nature. During the pre-war period, disappointments were voiced about the low level of ratification of Conventions which were most sensitive as regards international competition -- and this sometimes cast doubts on this system. However, after the war, in the wake of the constitutional reform of 1946, these doubts were overshadowed by a sort of "virtuous circle" of social progress which lasted for more than 30 years. Economic growth allowed the industrialized countries, firmly in the lead with their technological advance, to be generous; the ideological rivalry upheld a sort of emulation between the countries of both camps to prove their social superiority, including through ratifications. Decolonization did nothing to stop this trend because the newly independent countries acknowledged the ratifications of the former colonial powers -- thus multiplying even more the number of these ratifications.

This exceptional combination of circumstances is, however, far behind us and changes in the political and economic climate have revealed the limits of this system; competition has become much fiercer and, especially in the case of relocations, has given the impression that a high level of social protection might become a competitive disadvantage. Other trends, to which I have already referred, have all contributed towards giving the impression that progress is becoming sluggish, at least as far as the fairly tangible drop-off in the ratification rate of new Conventions is concerned. There can be no denying that trade liberalization and globalization have brought about visible improvements in the material conditions of many workers throughout the world; but there is no guarantee that the fruits of this progress, often spontaneous, are distributed reasonably fairly to one and all. The ILO's role must therefore first and foremost be to make people more aware of the facts and, in so doing, increase the awareness beginning to take shape that progress and social justice can -- apart from their intrinsic merits -- be a sound investment for stability and the long-term competitiveness of the economy. Country studies and various research projects under way should contribute towards this aim.

When examining the issue from the angle which particularly concerns us here, i.e. the institutional framework and standard setting, we must clearly ask ourselves whether the ILO, without straying from its promotional approach in which Members are free to decide upon the obligations they will assume, has the necessary institutional means. These means would encourage and evaluate in a more direct and systematic way the efforts made by States to turn the benefits of globalization to good account in terms of social progress -- whether or not this takes the form of the ratification of Conventions -- and at the same time make public opinion objectively aware of the positive repercussions of globalization in social affairs. I am personally convinced that, subject to a few adjustments, the opportunities provided by the ILO constitutional system itself, as it stands now, have been far from exhausted.

There are two possible approaches we should examine in this respect: the first would seek to encourage countries directly by a system of accompanying measures and mutual assistance; the second would set out to encourage them indirectly by mobilizing non-governmental actors.

1. A system of emulation between States

The ILO's constitutional system might be considered "vertical", as it sets out to encourage each of its Members separately in their efforts to attain the objectives of the Organization. As I shall attempt to demonstrate in the second part, the extremely wide- ranging opportunities it provides in this respect have not yet all been exploited. If there were a better follow-up of Recommendations, we could evaluate more closely the progress achieved in each State in another form than by means of ratification of Conventions. These various procedures can, however, only give a relatively fragmentary idea of the situation. They do not provide an overall "horizontal" view of progress achieved, taking account of the increased opportunities (or obstacles) arising from globalization over a given period of time. The question now is whether we might not be able to guide, support and even evaluate in a more direct and overall way these individual efforts by means of a system of accompanying measures and mutual encouragement for these efforts, taking into account the possibilities open to the Members concerned. Three aspects could be examined in the search for a response. Which principles could govern such a system? To what extent does the ILO have the necessary mandate to guarantee their follow-up? What form or legal instrument could be adopted for such a system?

(a) What principles could govern this system?

We should first determine whether, over and above the irrefutable principle that ILO Members should promote social progress as far as their resources allow, the Organization might provide them with principles or useful methods to achieve the desired result -- without encroaching upon the choices and decisions of the Members which, beyond being bound to guarantee basic rights, must continue to be made entirely by them. From discussions held at the ILO and in other fora, a number of fairly straightforward common principles seem to have emerged.

The first, which was reaffirmed by the ministers of trade in Singapore, is that a comparative advantage linked to a certain level of wages or social protection is legitimate, particularly if it is a factor of economic growth -- and thus, by extension, of social progress. I believe, however, that this principle should have as its corollary that any such comparative advantage would, if it were artificially maintained to the detriment of social progress as a mere means of winning markets, lose all legitimacy.

The second principle is that although it is up to each member State to decide upon the areas and social priorities which should benefit from the fruits of growth and prosperity generated by globalization, there is nevertheless a "minimum programme" that each should try to achieve. During recent discussions of the Governing Body, reference was made on several occasions to the former article 41 of the Constitution (Article 427 of the Treaty of Versailles) concerning the so-called "workers' clauses". In addition to stating certain fundamental rights, these clauses list a number of basic objectives (in the area of wages, working hours, weekly hours of rest) which, without being "complete or final", would "if adopted by industrial communities ... and safeguarded in practice by an adequate system of ... inspection ... confer lasting benefits upon the wage-earners of the world".(12) Although the list of fundamental rights contained in this provision was overshadowed by the Declaration of Philadelphia and later developments, these basic objectives of progress remain to a great extent valid. Given that this provision was omitted from the Constitution without being formally abrogated,(13) it could therefore easily be "revived" by the Conference.

The third principle is that, over and beyond this "minimum programme", all workers in a country, and not only those working for the world market, should be able to have a fair share of the fruits of globalization. And this principle is neither unrealistic nor impossible to achieve. Indeed, the way to achieve this stems quite naturally and concretely from the ILO's tripartite approach. Although the final decision on the priorities and specific content of social progress should continue to be a matter for the member States themselves, it should logically take shape during discussions between the social partners on the use of the benefits reaped from globalization and the distribution of its costs, in an extension of the Consultation (Industrial and National Levels) Recommendation, 1960 (No. 113), and the Tripartite Consultation (International Labour Standards) Convention, 1976 (No. 144), and its accompanying Recommendation (No. 152).

There is another consideration which should help us achieve our goal. Experience has shown that the principles that are most widely recognized in one organization are often overlooked in others. In addition to the list of principles and means of action which have just been mentioned, it would not be out of place to remind Members of the ILO that they remain bound by the obligations to which they committed themselves by joining the ILO, whenever they participate in the work of other international bodies, in particular economic and financial institutions.

(b) Does the ILO have the necessary mandate?

To ensure that these principles and methods reactivate the emulation of social progress and to convince international pub-lic opinion objectively that the promises of liberalization and globalization of the economy are not a mere illusion, they must obviously be the subject of a certain follow-up. This inevitably raises the question of whether such a follow-up is within the mandate of the ILO, especially since its Members would have to accept the principle of a comprehensive and reciprocal examination of the uses to which they have put the benefits of globalization.

Once again, the answer to this question is clearly in the affirmative. The Declaration of Philadelphia not only contains a very wide mandate which authorizes the ILO to judge economic and financial measures at national and international level in the light of its own objectives, but obliges it to further, among all its member States, programmes likely to achieve "policies in regard to wages and earnings, hours and other conditions of work calculated to ensure a just share of the fruits of progress to all". Furthermore, Members of the ILO, by accepting the commitment to work towards achieving its objectives in good faith, have acknowledged the necessary interdependency of their efforts -- and consequently a reciprocal right to see what other Members are doing. The well-known passage in the Preamble to the Constitution, according to which "the failure of any nation to adopt humane conditions of labour is an obstacle in the way of other nations which desire to improve the conditions in their own countries", should be read in this light. Although it assumes the existence of an open world market, this, in turn, assumes that all the partners in this market should "play the game" of social progress properly. Although, from this standpoint, introducing a social conditionality would be nonsense because the opening up of markets constitutes to a great extent a kind of precondition to social progress, it would be just as inconsistent to claim that all partners, in the name of social progress, should have access to all markets without having to give account to anyone on their practices in this field.

(c) On the basis of which instrument?

The third question concerns the instrument that would implement these principles. In view of the extremely wide mandate of the Organization to which I have already referred, this instrument could take the form of a declaration or charter adopted by the Conference -- a suggestion that has already been raised in connection with fundamental rights. But it would also be entirely feasible to give it the form of a separate "traditional" standard-setting instrument. The Social Policy (Basic Aims and Standards) Convention, 1962 (No. 117), might serve as a model or at least a precedent.(14) Its goal is not dissimilar to that of the text envisaged here because it set out, at a time when many former colonies were gaining their independence, to establish guidelines and priorities for action in the specific area of social policy. The decision taken at the time to incorporate this approach into a Convention would not, however, be the most appropriate in the present context.

However, a Recommendation accompanied by a follow-up procedure, to which I shall refer in the second part of the Report, might well serve the promotional purpose we are seeking to achieve. There is no reason why the means of supervision to be applied to the instrument should not be spelled out in the instrument itself. This follow-up could, for example, take the form of a regular report by the Director-General on "social progress in the world", which could, as appropriate, be discussed by an ad hoc committee of the Conference.

2. The mobilization of non-governmental actors

In my Report of 1994, I pointed out that one of the two limits placed on the ILO's standard-setting action in the context of globalization was that its interlocutors were member States whose will and capacity to follow its guidelines were inevitably affected by international competition. This was because capital had become mobile, whereas this was not true for workers -- and even less so for workers in the member States. Although my preceding analyses and suggestions should normally encourage Members to step up their action in the spirit of emulation, social progress is no longer only a matter for States; it is increasingly becoming a matter for other actors, in particular manufacturing enterprises, wholesalers and retailers, and consumers. This may be attributed to two factors, the first of which is closely linked to the second. First, there is a growing awareness among large or medium-sized enterprises of the social or environmental repercussions of their activities. Second, there is a growing awareness among consumers and their organizations (especially in the developed countries) of the responsibilities they undertake when they make a certain choice in relation to the protection of the environment or human rights. This two-pronged, converging movement can lead to the adoption of charters or codes of practice for producers and of labels guaranteeing the conditions under which consumer articles are manufactured.

As long as these voluntary arrangements do not become a technical barrier to trade, they seem to be free from the criticisms or condemnation levelled at social clauses on account of their protectionist connotations in the context of multilateral trade, since these voluntary measures merely set out to inform consumers without encroaching upon their choice.(15) From the standpoint of principles, the fact that consumers are prepared to pay for their environmental or social preferences is completely different from a situation in which a social preference would be forced on them through a linkage resulting in an increase in the price of the products. As far as the multinationals are concerned, voluntary arrangements of this type might ward off the temptation national legislators might have one day to make their subsidiary enterprises subject to certain overriding rules in social matters, justifying them by reference to the consequences that the infringement of these rules could have on their territory.

The developing countries are somewhat perplexed by these labels; they are sometimes even afraid that this type of decentralized solution might be used in a discriminatory way for protectionist or political purposes. This is apparently why some of the countries have already been concerned by the fact that the Office has started to gather some information on the subject. These concerns call for a clarification.

Although, at first sight, the objectives of this movement and those of the ILO seem in principle to be the same, labelling may, depending on its origin or the methods used, risk being arbitrary, singling out a particular right or product or being put to improper use. Indeed, there is a danger of:

Should these dangers make the ILO turn away from the trend or ignore it? I am firmly opposed to this for three reasons.

First of all it is obvious that it is not because the ILO turns its back on this movement, that it will go away. On the contrary, there are strong grounds to believe that it will continue with all sorts of initiatives exposing practices, each more shocking than the last, which will be seized upon by the media to stir the conscience or sensitivity of the general public -- and thus the consumers. In addition to the dangers I have just listed, there might well be the one of growing confusion, as consumers will be expected to find their way amongst a mass of labels of all origins, natures and motivations.

Secondly, there is a danger that by singling out a specific product or problem, with all the possible ensuing negative repercussions, this might, whether one likes it or not, call into question the ILO's standard-setting role (including potentially the definition of fundamental rights) and the value of its instruments. Granting a social label is tantamount to creating a parallel authority which assumes responsibility for pronouncing on what is socially "desirable" and on the manner (limited but extreme) of achieving this by tackling the cause by getting rid of the products.

The third reason, which is on a more positive note, is that in spite of the risks inherent in this movement, it gives an impulse and force to the ILO's standard-setting action beyond its normal audience. The fact that it tries to achieve certain humanitarian objectives by appealing to the real interests of partners in international trade is, in this respect, completely in line with the ILO's philosophy and methods.

What must be determined, however, is whether, and how, it would be technically and politically possible to reconcile these two forms of action, in view of their extremely different characteristics and concerns.

There is no doubt that this question must be looked into in depth and the competent services in the Office have scarcely begun the task. Without wishing to influence the outcome of this work, I would put forward a few preliminary considerations on the matter to try to assess whether and how this process might naturally reinforce the standard-setting action of the ILO.

The major difference between the ILO's approach and that of these spontaneous movements is that the ILO refuses to be selective. The ILO cannot turn away from workers whose work is not related to the export product selected for labelling, or who work only for the domestic market. There is also no reason to prefer one fundamental right over another -- even if it does have more emotional impact. As far as the ILO is concerned, labelling should therefore aim rather at promoting law and practice which meets the demands of fundamental standards (thus also benefiting workers whose products are not identifiable or exported). In other words, the ideal scenario would be that the label should encourage and reward those countries whose legislation is in line with a predetermined set of fundamental principles or rights. But if these labels are to have any credibility at all, they must guarantee that legislation has been complied with in actual practice. However, neither spontaneous initiatives nor the present procedures of the ILO can provide such a guarantee because there is no way of carrying out an international inspection on the spot which is reliable and legally independent. But it would be perfectly feasible to provide for such a system of inspection under an international labour Convention which, because of its voluntary nature, would allow each State to decide freely whether to give an overall social label to all goods produced on its territory -- provided that it accepts the obligations inherent in the Convention and agrees to have monitoring on the spot. Such a system should not interfere in any way with multinational trade practices for at least three reasons. First of all, even if the Convention contained the obligation to promote the label, this label would remain voluntary in the sense that ratification of this Convention would be, as in the case of any other Convention, an entirely free choice. Second, granting a label does not imply any trade discrimination against "unlabelled" products. Third, from the more general standpoint of international law, the members of the WTO not wishing to adhere to such a Convention would find it difficult to criticize the principle, as it would aim to promote the ILO's objectives to which they are also committed.

This approach would still leave room for charters or codes of conduct for manufacturers or distributors not wishing to see their products bear the stigma of coming from countries which do not put themselves in a position to grant the social label or, even if they do, wish to go beyond the national legislation. In such cases, it would be advisable to harmonize the priorities as to which obligations which should preferably be singled out (questions of occupational safety and health, consultation and collective bargaining, etc.) and, above all, to define how an impartial and objective certification of the sincerity of the commitments and compliance with practices may be carried out. In this respect, the ILO might provide a useful framework for discussion and serve as a reference for future action, thus extending or supplementing the procedure it set up nearly 20 years ago with the Tripartite Declaration of Principles concerning Multinational Enterprises and Social Policy.

* * *

My preceding comments show that, although globalization might make it more difficult for member States to apply ILO standards -- or even make them hesitate to apply them -- it nevertheless provides a golden opportunity for the ILO to renew its standard-setting activities in a number of ways:

These prospects for new action should help the Conference to envisage with more equanimity the possibility of adopting a more qualitative approach in future standard setting, on which I shall now seek to convince the Conference.


In my Report for the 75th anniversary of the ILO, I mentioned that I placed as much credence in those who said that our standard-setting activities had just about exhausted their possibilities as I did in those who claimed that we had reached the end of history. And I have not changed my opinion. Contrary to what some advocates of deregulation undoubtedly believe, the objectives contained in the Preamble to the ILO Constitution cannot, by their very nature, ever be entirely fulfilled. This does not imply there can be no progress but rather that every time you approach a goal, another one is visible in the distance. As technology gathers speed, it creates new hazards -- and only experience tells us that these hazards call for a new set of regulations. Man's ability to think up new forms of exploitation is one step ahead of the legislator. Moreover, a glance at the agenda of recent sessions of the International Labour Conference is enough to realize that changes in the world of work linked to technology or other factors not only require normative action by the ILO but also provide the subject-matter for this standard setting. This is even true in the case of home work which gave rise to so much controversy; although some contested the form and content of the instrument, the subject certainly concerns a considerable number of workers throughout the world, whose living and working conditions are amongst the most precarious.

Although globalization might well inhibit state action because of keener competition, it does not necessarily, except in very rare areas, alter their capacity to act on matters which constitute the crux of standard setting. Indeed, in many cases, it calls for increased cooperation between States and requires them to bring their activities more in line with others. It is no paradox to suggest that globalization might provide new subjects for standard-setting activities, as I suggested in the first part of this Report.

But there can be no denying that the absorptive capacity of the States, as that of ILO management, has its limits. After all, ILO standards only account for a fraction of the international standard-setting production with which its Members have to cope. Beyond a certain threshold, the production of new standards is subject to the law of diminishing returns -- as may be witnessed, with very few exceptions (Conventions Nos. 159 and 160) by the drop-off in the rate of ratifications for recent Conventions.(16) From a more political standpoint, the consensus kept alive by the cold war is wearing thin; for the first time in the history of the Organization, a group refused at the last session of the Conference to take part in the discussion of a draft Convention on a question which all three groups in the Governing Body had agreed to include on the Conference agenda. It would seem, in this particular case, that this position was taken because of the form of the proposed instrument. Yet it appears that, generally speaking, the pace of standard setting, resulting from the compromise reached in 1992 when improvements were made to the functioning of the Conference,(17) does not allow for the flexibility desired to take account of the degree of interest in potential subjects.

The way to try and reconcile these conflicting tendencies is not merely to "take a break" from standard setting, as has sometimes been suggested; the solution is to be more selective in producing new standards to ensure they are more relevant and have a greater impact. The idea is certainly not very original and might even pass for wishful thinking. To avoid this, we must examine ways to ensure that we do not lose sight of this relevance at every stage of standard setting: when choosing the subject for a standard; when deciding upon the form of the instrument; and when finally defining the content of the instrument itself. But above all, we must undertake, in a more comprehensive way, a systematic evaluation, after the fact, of instruments adopted to bring them back on course at any stage.


The ILO Constitution wisely provided, under article 14(2), that the Governing Body should undertake the necessary steps "to ensure thorough technical preparation and adequate consultation of the Members primarily concerned, by means of a preparatory conference or otherwise, prior to the adoption of a Convention or Recommendation by the Conference".

The aim of this provision was that future standards should be commensurate with real needs so as to have real impact. Today, now that the ILO membership has grown so much and the Organization is involved in so many areas, this task is just as indispensable but has become increasingly difficult to fulfil. Recent experience has shown that the difficulties and vicissitudes inherent in the subject have not always been gauged before embarking irreversibly upon the drafting of standards. To try and overcome these difficulties and comply with the objective in the Constitution, we should examine three possible additional approaches: extending the range of choices; applying more strictly the selection criteria to assess better the "added value" that a new standard might bring; and adopting a more rational procedure.

1. Extending the range of choices

Under article 10 of the Constitution, it is the Office's task to examine the subjects which it is proposed to bring before the Conference with a view to the conclusion of international instruments.(18) The Office's Report, intended to allow the Governing Body to take a decision on the choice of subjects, must be based on the widest possible information concerning national law and practice. This information should also include the provisions adopted by regional institutions (NAFTA, MERCOSUR, European Union, etc.) in the areas under consideration, provided that they have an impact on the law or practice of their members.

At present, the studies of the Office are carried out on the basis of information available at headquarters. It may sometimes be necessary to call meetings of experts on the subject. For instance, the decision to include on the Conference agenda the issue of the protection of workers in the event of their employer's insolvency was preceded by a meeting of experts, which produced a report allowing the Governing Body to take a decision in full knowledge of the issues involved. In the case of child labour, the information contained in the document submitted to the Governing Body was gathered from studies and research carried out by the competent technical service or contained in reports sent by governments in application of United Nations instruments.

The fact remains, however, that the choice of new standards is usually centralized and carried out in a rather random way. A better solution would be to be able to mobilize more effectively all the constituents, so that we obtain wider information on what they feel to be their real needs. Although this institutional approach is supposed to exist, this is mostly in theory.(19) A more viable solution might be to make more active use of the Office's decentralized structures and direct contacts with the tripartite constituents.

The external offices and the multidisciplinary teams should therefore make every effort to consult the people they normally deal with on the issues that might be included in the "portfolio" of future standards, whether or not these have been proposed by the technical services (including matters concerning the revision of older instruments). The regional offices might also, as appropriate, raise the question in pertinent regional bodies (meetings of ministers of labour, regional meetings, employers' and workers' organizations' congresses, etc.) and communicate the reactions and proposals voiced. All this information would help the Governing Body have greater insight into the issues involved and a clearer idea on the impact an instrument might have in the area under consideration, or the ratification prospects of a possible Convention.

Many of the preceding ideas have already been examined by the Committee on Legal Issues and International Labour Standards of the Governing Body, as part of a proposal to establish a regularly updated "portfolio" of proposals. This would allow the Governing Body to have a wider overall view of possibilities of action when called upon to select agenda items for the Conference submitted with a view to the adoption of standards. It would enable the Governing Body to make strategic choices rather than opting for a subject which is neither ready nor acceptable to anyone -- a situation bound to lead to disagreement and frustrations during discussions at the Conference and disappointment at the ratification stage.

2. Applying selection criteria more strictly: Looking for standards
    with the highest "added value"

In the past, it was asked whether it might not be better to proceed in a more systematic way and to choose subjects in the light of a number of objective criteria. In the course of the first in-depth review on standards, it was suggested to select several criteria to guide the choice of items, such as the number of workers affected, the extent to which the subject would affect workers in the lower economic stratum, and the severity of the problem.(20) These criteria are obviously useful and the Office suggests using them to establish a "portfolio" as just mentioned; but they also have their limits and some useful instruments would perhaps never have seen the light of day if these criteria had been mechanically applied. They therefore need to be given different weights, but such a system might be to the detriment of their apparently straightforward application.

In fact, there is a more general criterion which could be universally applied. This would consist of evaluating, in the case of each subject envisaged for standard setting, the extent to which a new standard might add a lasting qualitative input to the instruments already existing both inside and outside the Organ-ization and gauging whether the subject really lends itself to standards worthy of this name. To illustrate the problems that need to be addressed in the elaboration of new standards, I shall refer to three fundamental sets of alternatives: the suitability of a particular area to become the subject of obligations, rather than of mere guidelines of a political or moral nature; the added value of having overlapping instruments or a whole series of them in the same area, as opposed to that of consolidating those instruments; and the added value obtained from an accumulation of protective provisions as compared with provisions that rely on general principles of responsibility.

(a) Subjects that lend themselves to standard setting

As its standard-setting activities are quite rightly considered to be the pride of the ILO, there is a great temptation to draft a standard on any subject which seems at a given moment to call for action from the Organization. This approach would imply that as soon as any labour issue gains any international significance whatsoever, it should be made the object of a standard so as to acknowledge its importance. There is a tendency to confuse law and principles of good conduct. This attitude is even more surprising in that the idea of supplementing the formal standard-setting process by use of a sort of "soft law" has met with strong resistance. However, it is not by including in traditional instruments provisions which merely state noble principles, which few would contest, that these instruments assume the mantle of true legal standards -- in other words precise stipulations whose application in practice lends itself to being monitored. Indeed, provisions of the first kind cannot be the object of specific legal texts or regulations; they can be merely reflected in policies or statements and, consequently, their impact is difficult to assess. Of course, this particularly holds true for Conventions concerning economic and social policy.(21) But it might also start applying to more traditional fields of action, such as living and working conditions. Confronted with the myriad existing situations and solutions, the temptation is great indeed to prescribe merely the adoption of "national policies" as a means to meet goals defined in such a general way that they leave scope for complete freedom of action -- or create confusion about how they may be accomplished. This obviously does not imply that ILO instruments must refrain from giving general policy guidelines. It only means that such guidelines, especially when they are intended for Conventions, should be drafted in sufficiently specific terms to be able to give rise to rights and obligations worthy of the name. Failing this, they should be included in Recommendations -- accompanied by a follow-up mechanism to which I shall refer later -- or in other non-binding instruments.

In a similar vein, the increasingly rapid changes in the world of work and the emergence of atypical trends raises the whole problem of deciding, for the purposes of standard setting, when these trends stop becoming atypical and require regulation that has a reasonable chance of being effective. Trying to strike a strategic balance must take into account the inevitable interaction between regulations and the ingenuity of the social actors to get around them. Standard setting should not exhaust itself uselessly by pursuing subjects of a more or less ephemeral nature. It is better to wait until there has been a reasonable opportunity to understand the many diverse forms such subjects may take so that the regulations may be sufficiently global, lasting and effective. Here too, it would be more prudent, when dealing with these trends, to opt first for the solution of Recommendations.

(b) Overlapping or consolidation of instruments?

Anyone casually reading through the index of the compendium of Conventions and Recommendations could not fail to have the impression that there is a proliferation of subjects and many variations on the same theme. The same subject may be treated within a particular sector or on a more general level. This situation may be attributed to the very nature of the standard-setting system and historic circumstances which have meant that sometimes it was necessary: to deal with a specific problem when it arose before drafting an instrument of a more general scope; at times to circumvent obstacles created by a general instrument by means of sectoral instruments; and at other times to supplement a general instrument by more specific standards. Taking the area of occupational safety and health as an example, it may be seen that since the beginning of the 1960s the ILO has adopted standards concerning specific hazards (radiation in 1960, benzene in 1971, occupational cancer in 1974), a general instrument on the working environment (air pollution, noise and vibration in 1977) and an even more general instrument on occupational safety and health in 1981, before turning once again to specific subjects (asbestos in 1986, construction in 1988, chemicals in 1990 and mines in 1995). All these instruments respond to specific needs and also require specific provisions. It is obvious, however, that they all have a common approach and contain a large number of provisions which are similar if not identical.

Apart from the fact that this fragmentation accounts for the low rate of ratification of each of these instruments (although, there is a relatively regular flow of ratifications for the occupational safety and health sector as a whole) an overlapping of this nature runs a double risk: that of more or less accidental differences or even contradictions, and a dilution of the impact of the provisions that are common to all the instruments. Similar problems arise in the case of instruments devoted to special categories of workers who are not covered by legal instruments -- or only inadequately. In a well-intentioned attempt to provide these categories of workers with overall protection when they are not already covered by general instruments, provisions are reformulated at the risk of curtailing or even denying the entitlements that they do have. Although the Office's task is, at various stages during the preparation of the text, to draw attention to links with other instruments, there is unfortunately no absolute guarantee that the final result will be consistent.

Putting this state of affairs to rights is obviously not easy. Whenever there is already an instrument of general scope, as in the case of occupational safety and health, it might be better, rather than drafting a completely new Convention, to adopt a protocol which would refer to the relevant provisions of the main Convention and be open to ratification even to those countries not having ratified the main Convention itself. A "standard formula" should also be drawn up to define the relationship between specific instruments and the general instruments covering the same sector or the same workers. I shall return to this later in the annex when referring to my proposal for establishing a "code of good drafting practices". These improvements would not, however, resolve the problem arising from the overlapping of already existing instruments. The consolidation of existing instruments is an extremely complex undertaking and has been raised during discussions of the Working Party on Policy regarding the Revision of Standards;(22) but it goes beyond the scope of this Report.

In view of the difficulties involved in making any official consolidation of standards and the time required to do so, we could perhaps adopt another solution which, although ambitious, could be very useful and much quicker to implement. This would be to embark once again -- but in a different and more concise way -- upon the unofficial "codificiation" started by the Office under its sole responsibility just before the Second World War, and which bore fruit in 1951 in the International Labour Code. Without affecting in any way existing instruments or obligations ensuing from these instruments, such a code could give a more coherent summary, by distinguishing between general principles applicable to all workers and standards truly specific to certain sectors or categories of workers. An exercise of this nature would no doubt require considerable resources; but it would allow the ILO, and its standard-setting machinery, to make a dazzling entrance into the third millennium. Depending upon the quality of the result, nothing would prevent the Conference from making it more official by "promulgating" it as a sort of recommendation of general scope.

(c) Accumulation of protective provisions or referral to regulations relying
principles of responsibility?

The decision of whether to accumulate standards or to consolidate them is further complicated by having to decide between the accumulation of protective provisions and reliance on principles of responsibility. In order to cope with the new hazards arising from technology in particular, there is a tendency, especially in the area of occupational safety and health, to amass detailed regulations on preventive provisions or the safety of equipment. But this means being caught up in a constant struggle to try and identify new hazards as they emerge and yet being often unable to do more than have provisions of a very general nature. Such provisions may be to have well-designed tools or equipment, replace hazardous substances by those which are less so, and ensure that ladders (where these are necessary if work cannot be safely done on or from the ground, as specified in the Safety and Health in Construction Convention, 1988 (No. 167)) are properly secured against inadvertent movement, etc. Instead of amassing details which are often of very little practical use, it might be more effective to establish general rules of responsibility, application of which could be assessed by reference to safety precaution standards, the state of the art or other practices in a particular trade or occupation. These could appear in codes of practice which might be referred to in legislation, collective agreements or court decisions. This very complex question certainly warrants deeper reflection than is possible within the content and scope of this Report.

Another more traditional approach would be to incorporate these more detailed regulations in an annex to the instrument which could be revised and updated in accordance with a different and more simplified procedure than that applying to the Convention itself. In cases in which regulations are by their very nature relatively transitory (because they might for instance be linked to a specific level of technology or knowledge), it might be preferable to restrict the instruments to objectives and leave the actual details of implementation to the annexes to ensure greater flexibility. As the Director-General already pointed out in his Report in 1964: "certain Conventions contain provisions of a technical nature, necessary for their proper application, which do not involve matters of general policy in the same sense as the basic obligations of the Convention and are more liable than the provisions of a more general nature to call for periodical amendment to adapt them to changing needs and circumstances. ... The provisions in question are, however, an integral part of the Convention and, unlike regulations giving effect to a statute in national law and practice, cannot, as matters now stand, be amended without revision of the Convention itself."(23) The Report went on to state that it had not been uncommon for international conventions to permit the amendment of annexes, schedules and other provisions of a technical nature by simplified procedures specified by the Conference (in other words not necessarily involving ratification) "without infringing the principle that the obligations of States should not be extended without their consent".(24)

3. Should the procedure for selecting standards be less definitive?

My preceding remarks in no way offer a ready-made solution to the problem of choosing the most appropriate subject for standard setting in every case. They rather suggest that the inclusion of an item on the agenda with a view to standard setting should be the final outcome of a process during which any subject which does not seem ripe for standard setting -- which therefore would only make an inconsequential or ephemeral contribution to the body of standards -- should be eliminated. The present procedure for selecting items rarely enters into a discussion in the Governing Body of the possible content of the instrument. The result often reflects an overall compromise which also deals with agenda items for a general discussion. Indeed, subjects for standards are often chosen on the basis of the lowest common denominator -- in other words, because there is nothing better on offer.

To encourage a more rational selection procedure, ways should be seriously examined to separate items included with a view to standard setting from those for a general discussion. In short, to permit items up for general discussion to be more in touch with topical concerns, it should be possible to include them in the agenda at a later date than items for standard setting.

Above all, it should be possible, when finally including an item on the agenda, to have a clearer view of the possible content of the instrument under consideration compared with that of existing instruments -- including those outside the Organization. According to the present procedure, however, the Governing Body makes its choice on the basis of a comparative analysis and very general indications as to objectives, without entering into any detail on each subject. Here again, a "portfolio" could improve the situation by helping, after a number of successive examinations, to define the outline of the instrument on the subject envisaged.

Whenever the subjects under consideration raise complex technical or political issues on which the Office does not have enough information and the Governing Body itself is unable to provide the necessary guidelines, it might be envisaged to go even further and resort more systematically to a solution which has a potential that has perhaps not been adequately used: that of having a discussion in the International Labour Conference to assess the viability of the subject intended for the new instrument and to give specific guidelines on the drafting of the questionnaire. Only once the preliminary discussion was over would the Conference confirm the final inclusion of the item on the agenda. In the present context, this solution would appear to be more economic and more universal in approach than that of meetings of experts and preparatory technical conferences. I shall return to this subject in more detail in the questionnaire in the annex.


Amongst all the issues discussed in the successive Reports of the Director-General, this is undoubtedly the one that seems to come up with the greatest insistence; indeed, it is of fundamental importance. On this matter, I shall merely refer to the very explicit comments made in the 1964 Report, to the effect that "there is no inherent virtue in a Convention as such as compared with the Recommendation ...". Although a Convention creates obligations, in certain fields "a standard which can be widely accepted as such may well be more effective in practice than obligations which are unlikely to be equally widely assumed". It concludes that the Recommendation should no longer be considered as "a poor relation of the Convention".(25)

This subject was taken up again in the 1984 Report (to which I also referred in 1994) in which the Director-General did not hesitate to write that "one of the basic questions for the future is therefore whether greater use should not again be made of non-mandatory instruments, reserving Conventions for important issues capable of precise definition and action".(26)

Despite these concurring remarks and successive and urgent recommendations, the situation still has hardly changed. Out of the 17 Recommendations adopted by the Conference between 1985 and 1996, not one of the Recommendations was an autonomous instrument unrelated to a Convention. This confirms the decline of autonomous Recommendations, which accounted for 55 per cent of the Recommendations adopted in the period from 1951 to 1970 and 7 per cent in the period from 1971 to 1983. This seems even more regrettable given the fact that the drop-off in the rate of ratifications (due to excessive workload in parliaments, the limitations of federal States and now, to a certain extent, of Member States of the European Union, which, if it expands further, might result in a gradual drying up of ratifications in Europe, which until now has been a champion in this area) should be encouraging the ILO to make better use of the range of instruments at its disposal. Rather than repeating the same reproaches, I feel it important to analyse the reasons for this situation and to try and find a solution. Two factors, which moreover are closely linked, seem to be decisive in this respect: the attitude of governments and that of workers.

The ease with which the Conference opts to adopt instruments in the form of a Convention may be primarily attributed to the fact that many governments are prepared to vote for a Convention without seriously envisaging pressing for its ratification by the competent authorities. This attitude is, it should be stressed, not in keeping with the initial intentions of the founders of our Organization. Although they had to bend to reality and give up the revolutionary idea of giving the International Labour Conference a legislative power which was directly binding on States (in particular to ensure that legislative progress did not become hostage to a reactionary minority), it remained clear in their minds that a government, by voting in favour of the adoption of a Convention, undertook the moral commitment to bring about its ratification.(27) Only this can explain the presence -- which today seems so incongruous -- of special provisions aimed at certain countries expressly mentioned in a number of the first Conventions.(28) It is in this context, that one can also understand better the conviction, so widely shared before the Second World War, that international labour legislation should contribute effectively towards equalizing conditions for competition (and a contrario the criticisms of the Office made by the British employers at the beginning of the 1930s that it had not succeeded in adequately promoting ratifications more than ten years after the creation of the Organization).(29) It might admittedly be conceded that a government could in all good faith and in full knowledge of its responsibilities vote for a text knowing well that it would be impossible to ratify it for the time being, provided that it felt that this instrument might further the cause of international labour legislation and that it envisaged ratifying it in the longer term. This reasoning cannot, however, be carried too far because it might all too well create a situation -- unfortunately far too common today -- in which the Convention would become obsolete before having received enough ratifications -- and even sometimes before entering into effect! Interestingly, there has been a certain change of attitude at the Conference; some member States have abstained or voted against a text, explaining that they took this stand because they did not want their intentions to be misconstrued.

One way of dealing with this inconsistency has, although in an entirely different context, been examined by successive working parties examining the revision of standards. They have proposed, for instance, increasing the number of ratifications required for Conventions to enter into force. But even supposing that this idea was accepted and that the number of ratifications also increased, it would only take care of the consequences rather than the causes. Yet it should not be too difficult to return to the causes themselves, merely by using more systematically the procedure provided for under the Constitution to confront certain governments with their inconsistencies. Indeed, under article 19(5)(b)(30) of the Constitution, Members are obliged to bring the Convention before the competent authority and, subsequently, under article 19(5)(c), give account of the measures they have taken. In so far as the intention of the constitutional provisions is perfectly clear, it would be logical that governments having voted in favour of the Convention should be called upon to give account either, in the first case, of the recommendation they made to the competent authority or, in the second case, of the reasons why ratification has not occurred. This very moderate requirement should not discourage member States from voting for a text that they consider as a sound basis for legislation even if they are unable to ratify it at once; it should merely encourage them to be more consistent in their attitude, both with regard to their recommendation to the competent authorities and with their subsequent action.

The Workers have been unremitting in their preference for the adoption of Conventions, despite the low level of ratifications; they have argued, without being contradicted by the Employers, that even unratified Conventions can have an influence on national law and practice. In itself, this statement is difficult to deny, as the Director-General already noted (with regret) in his Report to the 27th Session (1945!) of the Conference: "it may well be ... that at certain periods in the history of the Organization, the form of the Convention acquired an undue symbolical importance, as the result of which it came to be used in cases in which a Recommendation would have been more appropriate".(31) But this approach does not stand up to close scrutiny and brings damaging repercussions for the efficiency and credibility of standard-setting activities as a whole in its wake. As pointed out more than 50 years ago, "it tends to discredit the Convention technique by resulting in widespread failure to ratify, and it equally tends to discredit the Recommendation by gratuitously implying that because a Recommendation is not an instrument for the creation of obligations it is therefore ineffective as an instrument designed to influence policy and legislation by the definition of an international standard".(32) I believe it is important to explain briefly why.

The aim of Conventions, as indeed that of Recommendations, is to ensure that the law and practice of ILO member States comply with the provisions contained in the standards. The advantage that Conventions have over Recommendations is obviously that they express the desired progress in terms of legally binding obligations. This has two consequences:

The argument that a Convention can have an influence on the practice of States even if it has not been ratified is therefore irrelevant because, by definition, this is the role of a Recommendation. Furthermore, if it is not ratified, a Convention is neither subject to the "blocking mechanism" nor able to contribute to the levelling of conditions for competition. This distorted view with respect to the influence and prestige of Conventions can be in part accounted for and excused by aberration in a practice which, by overlooking the required follow-up, has obscured the fact that Recommendations are instruments in their own right. Like Conventions, they are supposed to be followed up to measure their impact; and to be updated to retain their relevance. And now we come to the crux of the problem: if Recommendations are to regain their rightful place, they should once again become autonomous instruments unrelated to a Convention; they should then, and this of paramount importance, be followed up on a regular basis as provided for by the Constitution, to verify both their application and their relevance. Only those which meet with present-day needs should be retained. I shall now turn to each of these points.

1. Restoring the autonomous nature of Recommendations

Even a cursory reading of the compendium of Conventions and Recommendations is enough to see that Recommendations do not enjoy much prestige. As already pointed out, most of them are not autonomous. What is more, they often merely reiterate certain provisions of the Convention, adding details which cannot be contained in a Convention or which have been rejected by the technical committee during discussions on the draft Convention. Some of their provisions may at times more resemble resolutions rather than texts to serve as a guide and model for the ILO constituents' future action.

The few autonomous Recommendations are often obsolete or have been long forgotten. However, they still involve the ILO's legal and moral credibility. It could even be argued that their obsolescence might affect the Organization's credibility even more than in the case of Conventions (which are indirectly shelved when they fail to enter into effect or, depending upon the case, are denounced). Fortunately, the work carried out and the proposals for reform put forward by the Working Party on Policy Regarding the Revision of Standards, as approved by the Governing Body, provide a very clear way out of this situation. During the discussions of this group, it was acknowledged that although the abrogation of international labour Conventions required an amendment to the Constitution -- which is now before this Conference for examination -- nothing prevented the Conference from withdrawing any Recommendation considered as obsolete by means of an acte contraire (i.e. an instrument to undo what it has done), adopted in accordance with the same procedures and majority requirements as those that applied to the Recommendation in question.(34) This opens up a whole new area of activity before us, as we do not have to wait for the entry into effect of this constitutional amendment.

2. Restoring a procedure of regular follow-up for Recommendations

The second condition for restoring the status of Recommendations concerns their impact and follow-up. This is easy to fulfil, at least in theory, since it merely requires more effective implementation of the provisions of article 19(6)(d) of the Constitution. This provision places on all Members the obligation to "report to the Director-General of the International Labour Office, at appropriate intervals as requested by the Governing Body, the position of the law and practice in their country in regard to the matters dealt with in the Recommendation, showing the extent to which effect has been given, or is proposed to be given, to the provisions of the Recommendation and such modifications of these provisions as it has been found or may be found necessary to make in adopting or applying them".

It is worth recalling in this respect that this provision, as also the parallel provision under article 19(5)(e) concerning unratified Conventions, was introduced after the Second World War following a proposal made by the Delegation on Constitutional Questions -- after yet another discussion on whether it was appropriate or feasible to empower the Organization to take decisions that would be binding on its Members.(35) In 1919, a proposal on these lines, although not accepted, led to the introduction, in article 19, of the obligation for all Members, whether or not they had ratified an instrument, to submit it to their legislative authorities within a period of one year, or no later than 18 months, from its adoption "for the enactment of legislation or other action".(36) In 1946, the same concern led to the extension of the follow-up obligation to all instruments.

One thing, however, is sure; after this reform there was no longer any reason whatsoever to consider Recommendations as the poor relations of Conventions. Even though Recommendations obviously did not themselves create obligations, member States were nevertheless under an obligation to submit a report on them at the request of the Governing Body. Consequently, Recommendations were intended to "exercise in the future an even more far-reaching influence on policy and legislation than they have done in the past", as pointed out in the report of the Delegation on Constitutional Questions.(37) The problem is that this intention is not -- or is no longer -- really being put into practice. A change in approach of reports under article 19 ("general surveys" of the Committee of Experts) has meant that Recommendations are not followed up on their own; this has in turn perpetuated the imbalance in favour of Conventions. Since 1975, the Governing Body has not selected any autonomous Recommendation for reports under article 19, paragraphs 6(d) and 7(b)(v), of the Constitution, although the examination of such instruments has been proposed. The general surveys carried out on Conventions and their accompanying Recommendations by the Committee of Experts rarely refer to the provisions of Recommendations (they do not create substantive legal obligations) or to their implementation by governments (which rarely refer to them in their law and practice reports). Whenever any references are made -- and these are few and far between -- they concern the interpretation of the provisions in the Convention covered by the general survey, in the light of those contained in the Recommendation; this only accentuates the latter's secondary role. If we are to return to the practice provided for under the Constitution for the follow-up of Recommendations, we shall also have to re-examine the implementation of article 19(5)(c) of the Constitution.


As already pointed out in the 1964 Report: "A Convention should deal with essentials; it should not contain rigid requirements in regard to matters in respect of which national practice may reasonably vary widely; it should not enter into too much administrative detail".(38)

The same Report also warned the Conference against constantly striving for perfection: "A Convention is not an opportunity to secure a victory for the winning team but a contribution to the common law of the world, the value of which depends on the measure of general assent which it commands".(39)

The 1984 Report pointed out that the situation was not improved by the amendments procedure. The Conference was obliged to discuss "large numbers of amendments in the limited time available" and this "inevitably reflected on the quality of the instruments adopted". It stressed that: "Any difficulties resulting from amendments adopted in the course of a first discussion can be eliminated but this is not so with amendments adopted in a second discussion ...". It went on to say that the texts adopted could have differences in meaning between the languages as a result of pressure on the translators and the drafting committee, which was weighted down by amendments and called upon to solve questions of substance and not merely of drafting. They performed a major task "sitting on average for six or seven hours, at a stage in the Conference when the strain of two weeks' meetings is beginning to be felt, but the Conference timetable does not permit of any delay in the completion of their task".(40)

None of these statements has unfortunately lost any of its validity over the years. The drastic cutting down of the Conference schedule has, on the contrary, only added to the difficulties, including the time required by the drafting committees, which should on the contrary have more time. The various suggestions made, of which some seemed at first to be fairly sound, such as for example that of trying to obtain amendments in advance, have not encountered much success -- although the reduction in the time available for the work of the committee would have made these suggestions more useful. The only way of really improving the drafting process would seem to be to tackle the problem directly, i.e. at the institutional level and in its rules. I therefore felt it would be useful to have an annex containing a fairly detailed description of some of the aspects of the standard-setting procedure which are of direct concern to the Conference and its functioning -- and which, I believe, might, if they were re-examined or reviewed, contribute in a limited but effective way to improving the quality of the content and wording of standards. The annex first deals with the need to review the way in which the questionnaire is drafted; this questionnaire serves as a basis for preparing instruments and often fixes their structure and even their content at far too early a stage, before the Office has enough information at its disposal. It then looks at the amendments procedure in the technical committees which does not lend itself to finding solutions likely to meet with the widest approval. Finally, the annex discusses the difficulty of maintaining uniformity and consistency in the drafting principles and techniques in the technical committees and in the instruments. It points out the need to clarify and facilitate the role of drafting committees in this respect.

In addition to the various approaches and reforms which have just been suggested to improve the selection of instruments, their form and content, I believe it indispensable that standard setting should have a more general self-correcting mechanism so that it can respond more effectively to the objectives it sets. In other words, there should be an efficient system whereby the bodies which produce the standards should be able to evaluate them in order to ascertain the impact of these standards and their relevance, thus drawing lessons for the future.


An objective and systematic evaluation of a product is an integral part of any modern system of production. And there is no reason why the product of standard setting should be an exception. It might be thought that the Organization has already embarked on such an evaluation by setting up successive working parties on the revision of standards, the most recent one being that set up by the Governing Body following discussions at the Conference in 1994; this group has carried out extremely valuable work, to which I have already referred. There is, however, a world of difference between an overall "clean up" operation carried out every 25 years or more which results, much later, in the realization that an instrument has become irreparably obsolete, and an evaluation which allows the body that gave birth to the instrument to re-examine it within a fairly short time so as to be able to draw conclusions, not only about the instrument in question, but in a more general way about the choice of subject-matter and content of future standard setting.

This need to carry out a systematic and objective evaluation of instruments seems to me even more vital today because it probably provides the only way of settling once and for all, and in a plausible way, the false debate on "flexibility". And by evaluation, I mean an overall evaluation of the impact of instruments in terms of legal, economic and social effects, which would attempt not only to measure the success achieved in fulfilling the specific objective set forth in the Convention or Recommendation, but also to identify any possible indirect or adverse repercussions there might be with respect to other ILO objectives -- for example that of employment. This is truly a multidisciplinary task requiring an analytical framework, a body and a procedure for evaluation that are appropriate.

The basis for such a system already exists in the Constitution and only requires -- once again -- the necessary practical adjustments. This is contained under article 19(5)(e), concerning Members' obligations as regards the follow-up of Conventions they have not ratified, and under article 19(6)(d) -- mentioned earlier in the text -- concerning the follow-up of Recommendations. As explicitly mentioned during the preparatory work of the Conference Delegation on Constitutional Questions, these provisions have a dual complementary objective: to assess the impact of these instruments on national law and practice; and to evaluate the flaws in the instruments, which might explain why they have fallen short of their goal (low number of ratifications in the case of Conventions, limited impact on law and practice or policies in the case of Recommendations). In the early days of the ILO, a critical evaluation of this kind, even if not carried out in accordance with these provisions, was not considered in any way to be out of place. We only need to see how quickly the Conference corrected the flaws in its first instruments: Conventions on night work of women in 1919 and revised in 1934; on workmen's compensation in the case of occupational diseases adopted in 1925 and revised in 1934; on protection against accidents of dockers adopted in 1929 and revised in 1932; on hours of work in coal mines adopted in 1931 and revised (without much more success) in 1935, etc.

For various reasons, and in particular for practical reasons (the growing volume of work resulting from the supervision of an increasing number of instruments), this article was not implemented in the way it was originally intended. It was entrusted to the Committee of Experts for the Application of Conventions and Recommendations, which did not accept this additional burden without misgivings. And these misgivings are not hard to understand. After all, the Committee of Experts fulfils a quasi-judicial role of evaluating the way in which member States fulfil their obligations; it is not really expected to pronounce on the shortcomings of the Organization's own legislative work. Be that as it may, it has carried out the task conferred upon it by conducting "general surveys", which already existed. As may be inferred from their title, these are mainly comparative surveys which, although highly interesting, are not intended to measure the legal, political and economic impact of standards or, as the case may be, to identify the shortcomings or flaws responsible for their limited impact.(41) Any evaluation on the impact, as has been proposed, would obviously require far greater resources than those at present earmarked for general surveys; it would also imply close involvement with the decentralized teams and a more active and substantive participation on the part of employers' and workers' organizations. The stakes now seem high enough to warrant such efforts, however.

Assuming that we agree to widen the scope of reports and general surveys, we would then have to envisage the appropriate body and procedure for carrying out a critical and multidisciplinary examination of the findings. At present, general surveys are examined by the Committee on the Application of Standards at the Conference; they are only submitted to the Governing Body as a formality. However, the Committee on the Application of Standards, although it manages to devote a few sittings to the general survey, is far too busy fulfilling its mandate of supervising application to be able to spend enough time critically examining the instruments and their impact. Furthermore, its mandate is specifically defined by the Standing Orders of the Conference and is therefore somewhat removed from the multidisciplinary approach to which I am referring. Once again, I am not saying anything new. This observation has been made -- from a similar but much narrower perspective -- since at least the 1960s. The Director-General's Report in 1964 stressed that a "defect [of the present system] is that neither the Conference nor the Governing Body has a standing revision committee which can undertake over a period of years a continuing task of systematically revising existing instruments on the basis of a widely agreed general policy. The Governing Body Committee on Standing Orders and the Application of Conventions and Recommendations and the Conference Committee on the Application of Conventions and Recommendations are well qualified to draw attention to cases in which revision may be desirable, but the negotiation of the precise terms of revision goes somewhat beyond their scope."(42)

At the time, these ideas were well received and even resulted in some groundwork being laid. However, they did not result in a lasting and thorough reform and this may perhaps be attributed to the fact that they failed to raise the problem from the wider perspective of effectively evaluating adopted standards. To ensure that a systematic evaluation of standards develops on an institutional basis, it must first return to the objectives of the surveys provided for under article 19 of the Constitution. It must then opt for the most appropriate approach, which might, depending on the case, group together instruments dealing with the same subject-matter so as to draw lessons from them of a much wider scope. Finally, it must be able to depend upon a body which ensures a certain continuity in its outlook and action to carry out this evaluation in a consistent and systematic way. This body should, in the first instance, be the Governing Body itself; as it is responsible for determining the agenda, it would be led to evaluate the outcome of its own choices. But this body could also be a standing committee of the Conference. This is undoubtedly not the place to develop in detail the possible ways of putting these principles into practice. Depending on the level of support received for these ideas, I shall submit more specific proposals to the Governing Body.


Three sets of conclusions seem to emerge clearly from the observations I have made.

The first is that the global economy is opening up a new vista for ILO standard-setting action. While the first part of this Report has revealed that globalization is giving rise to new, extremely topical subjects for standard setting, the second part confirms that the concern to be more selective does not reflect a disenchantment with creating new standards but aims at giving them a greater impact by developing new, basic activities to follow up the progress of Recommendations and by carrying out a thorough evaluation of standard setting in the way suggested in the last part.

The second set of conclusions concerns the ILO's means of action. From what we have seen in the Report, the ILO does not need to move away from its constitutional framework or even to change it in order to be able to take on any new standard-setting activities it might assume in this age of globalization. All it has to do is to make a more judicious use -- more in line with their original objectives -- of the unique means of action placed at its disposal. In particular, the conclusions in the second part of the Report concerning the strengthening -- or rather the reinstatement -- of the procedure to follow up Recommendations -- might fruitfully be applied to reactivate and guide efforts to have social measures develop alongside globalization. A Recommendation with an appropriate follow-up procedure at sufficiently close intervals could, at least initially, offer an adequate framework to allow the Organization to reaffirm its doctrine and responsibilities in this area in an all-englobing yet concise form.

The last set of conclusions stems from the first two: whereas the scope for standard setting is vast and no complex constitutional reforms are required for the Organization to embark upon this work, taking action is simply a matter of political will. My hope of seeing such a will materialize is justified by three considerations:


Most of this Report was written before the last session of the Governing Body. Upon reflection, I thought it was preferable, for two reasons, to publish it as it stands rather than attempt to supplement it to take account of the discussions held at this session in the Committee on Legal Issues and International Labour Standards (LILS) and the Working Party on the Social Dimensions of the Liberalization of International Trade.

First, I did not feel that the very encouraging results of these discussions affected the validity of the analyses in the Report; quite the contrary, in fact.

Second, it would be difficult at this stage to develop much further the ideas or suggestions it contains.

This is particularly true for the content of a possible declaration of fundamental principles which has been discussed in the LILS Committee. As yet, I have only had relatively limited indications as to how this should be drafted -- which do not always tally. And in order to attain its objective, such a declaration should reflect as wide a consensus as possible; moreover, its drafting requires a particularly methodical and meticulous preparation. I therefore felt that it would be better, before engaging further in this exercise, to sound out the views of all the constituents at the Conference.

As concerns the re-examination of the ILO's means of action in this age of globalization, raised by the Working Party on the Social Dimensions of the Liberalization of International Trade, the Report already mentions quite a few possible approaches which have hitherto being overlooked or inadequately exploited: indeed, rather than trying to think up new means of action, it is rather a matter of finding the necessary will to exploit those with which the Organization is already particularly well endowed.

By publishing this Report as it is, I therefore hope to stimulate ideas on the subject which may be discussed in an open a way as possible, so that they can take shape before the end of the Conference and provide us with the basis of a wide consensus so necessary for renewing the standard-setting action of the ILO.

As the Conference Delegation on Constitutional Questions already pointed out in 1946: "no constitution can work successfully unless there is general agreement on its fundamental principles".(43) This observation remains valid, even though today we are not setting out to draw up a new constitutional charter but simply to rediscover, to meet the challenge of globalization, the relevance of the objectives contained in the ILO Constitution and to give greater effect to the means of action it provides.



According to the Standing Orders and the practice in force, the Governing Body decides in March the agenda of the Conference session to be held in June of two years hence. The Governing Body makes its choice on the basis of a document containing a summary of law and practice which is supposed to give an idea of the issues that the possible instruments concerning the subjects under examination might address (in other words, the possible content of these instruments). This summary is, however, very general; in the overwhelming majority of cases, the Governing Body takes a decision without discussing or giving precise guidelines on the content of the instruments in question. Consequently, the Office, in accordance with article 39 of the Standing Orders of the Conference, is often left to its own devices to prepare a report and questionnaire which already give a fairly detailed outline of the structure and the content of the instrument. This responsibility is of course entirely within the constitutional functions of the Office. But it is regrettable that it does not have the chance to benefit from some sort of preliminary guidance on issues deemed essential. The system of a portfolio "of proposals", to which I have already referred, should allow the Governing Body at its second examination in March to provide more specific guidance on a more limited number of possible agenda items. When the subject-matter is basically of a technical nature, already well prepared by the Office, this situation should be manageable. When, however, it deals with sensitive issues, as in the case of home work, this procedure might prove to be inadequate.

It is relevant to recall that the method of discussion based on a questionnaire determines to a very great extent the structure of a text. Since those to whom the questionnaire is sent are supposed to reply separately to each of the questions, it is difficult for them to propose another structure.(44) And attempts to reach an acceptable compromise within this framework by a flood of amendments can result in inconsistency. This was also pointed out in the 1984 Report, which noted that: "amendments tend to be considered in the context of the provision to which they refer without there being time to examine their possible impact on the instrument as a whole"(45) (and even less, one might be tempted to add, on the body of standards as a whole).

Although these constraints are to a certain extent inherent in any legislative procedure in a parliamentary context, there is no denying that the system of the questionnaire only makes the situation worse. Several possibilities might be envisaged to overcome this problem, whilst also ensuring the coherence and viability of the product as a whole for which the Office is responsible.

The first option would be to ensure that the questionnaire is not finalized until after a preliminary discussion specifically designed to provide guidance. The matter could for instance first be dealt with in the context of a preparatory technical conference or preparatory general discussion, after which the Conference would confirm that the item in question would be included on the agenda with a view to the adoption of new instruments. This general discussion would help gather the information required to draft the questionnaire on the basis of an outline contained in the report (but without entering into a discussion item by item with amendments). The possibility could also be envisaged of involving the Governing Body in the preparation of the questionnaire. In any case, the aim would be to comprehend better the needs of the constituents by testing whether the subject is amenable to standard setting and to achieve, in the end, texts of much higher quality. The other side of the coin is that it would obviously prolong the whole standard-setting process; this could therefore be reserved for the most difficult or uncertain issues (such as home work or contract labour) for which discussions within the Governing Body would not have provided clear or consistent guidance from the three groups.

A variation on this approach would be to amend the Standing Orders in such a way that the outline or structure would not be fixed prior to the first discussion, but finalized immediately afterwards. This solution would be tantamount to considering the second discussion as a "single" discussion (the Conference would still have the possibility, as in the previous case, to drop the item from the agenda for standard setting if this preliminary discussion did not produce conclusive results as to the viability of the instrument or instruments envisaged). A drawback would be that this system would not allow enough time (at least if the issue were to be dealt with at two consecutive sessions as is normally the case): an accelerated procedure, such as that provided for under article 38(4) of the Standing Orders, would have to be applied and a final report containing a draft Convention or Recommendation would have to be directly established on the basis of the first discussion. The problem of the questionnaire would be solved once and for all because it would no longer exist -- a solution that might seem too radical to some.

Another solution would be to review the presentation of the questionnaire itself. This would mean overhauling it and adopting a procedure closer to that used for negotiating international agreements in other bodies. Indeed, other international instruments often come into being on the basis of an outline of their possible content, supplemented during discussions by draft texts. In the ILO, proposals are always submitted in the form of questions -- a procedure which does not give a very clear idea of the final product. It would be entirely possible to elicit reactions to a proposed text rather than asking questions which subsequently have to be transformed into a draft text. This would give the constituents an initial idea of the text which might emerge, whilst providing them with the opportunity to propose alternative wording before the discussions at the Conference -- while still leaving them free to submit amendments during the session. Amongst other things, this would avoid the confusion that frequently arises from the use of the conditional tense in the proposals for the first discussion. The questionnaire could also be drafted in a form that would speed up the analysis of replies; for example, it could contain two boxes for marking acceptance or non-acceptance of the proposals submitted, with enough space left to put forward other proposals. The conclusions drawn up by the Office could take up the suggestions that fit within the conceptual framework of the proposed instrument, whilst continuing to mention all the constituents' suggestions in the report. Such an approach could contribute towards signifi-cantly improving the quality of ILO instruments.


The amendments procedure is mainly based on article 63 of the Standing Orders of the Conference; it has been enriched by long-standing jurisprudence often stemming from national parliamentary practices and has been adjusted to take account of the specific constraints under which the Conference has to accomplish its task -- in particular the extremely limited time at the committee's disposal and the usually considerable number of amendments. This procedure lends further authority to the chairpersons of the committees, whose decisions may not be challenged. Questions are normally decided by being put to the vote. Again in the interest of rapidity, the system for deciding as between amendments on the same provision is the most expeditious one: voting begins with the amendment which differs the most from the proposed text; this tends to result in the selection of the amendment likely to receive a majority in the shortest period of time, even if this majority is the narrowest. Although it would take longer, a more widely acceptable result could be achieved by using a more empirical and incremental approach.

These somewhat technical considerations can help explain the context in which feelings of frustration might arise and develop -- and the decision of a group not to take part in the discussion in the Committee on Home Work in 1996 provided a particularly striking example. In this particular case, it was unfortunate that the protest focused on a particular text which had already been placed before the Conference, instead of trying to analyse the reasons for this choice. Indeed, it is difficult to find a solution to this situation and the ensuing frustrations without making the whole process less rigid. In my opinion, the most viable solution would be that of the preparatory general discussion to which I have already referred, as this would allow us to judge to what extent we might expect agreement on the basis of the outline for an instrument. In this respect, it may be recalled that under article 16(3) of the Constitution, the decision to include formally an item on the agenda of the following session or a later session of the Conference requires a two-thirds majority, which already provides a certain guarantee that the item will not be included if there is strong opposition. It would moreover be feasible to strengthen this guarantee even further in the Standing Orders of the Conference.


If the Governing Body were to adopt a code of good drafting practices (for example practices concerning the drafting of preambles, the way to refer to other instruments, the way to avoid needless repetitions between a Convention and its supplementary Recommendation or between various instruments, the status of final clauses, etc.), this might dissuade the Conference from adopting provisions which are difficult to apply in practice or lack legal precision, diverge in their form in a way that undermines consistency of drafting in the Organization's body of standards or give rise to long and fruitless discussions within the drafting committee.

Although the role and composition of the drafting committees are defined under articles 6 and 59 of the Standing Orders of the Conference, delegates are sometimes confused about, and even mistrustful of, their functions and terms of reference. Their legitimate concern is that the solutions of compromise reached during the technical committee's discussions remain unchanged. However, these solutions have to be examined by the drafting committees in the wider context of the body of standards which has grown up over the years, in accordance with a certain number of drafting rules designed to preserve the consistency of the instruments as a whole.

The drafting committee therefore has to retain the substance of the outcome of the Committee's work, while examining it from the standpoint of clarity and form. If the wording is not very clear, it should be possible for it to be sent back to the technical committee for further discussion. As regards the form of the instrument, the adoption and application of a code of good drafting practices would inform delegates of the rules for drafting instruments developed by the Office from the outset; the technical committees of the Conference and their members on the drafting committee would thus be aware of these basic rules before beginning their work. They would then have more time to devote to their mandate in the strict sense of the term, as defined by the Standing Orders.

Having all this information in advance in a code which would be authoritative because it had been approved by the Governing Body would thus allow Conference delegates more time to concentrate on the difficult substantive work they face.

1 ILO: World Employment 1996/97: National policies in a global context, p. 7: "In the case of the industrialized countries there has been a lively debate in the academic literature over how large an impact trade with low-wage countries has had on employment. On balance, the empirical evidence suggest that trade with developing countries and the relocation of industries has been only a minor explanatory factor behind the rise in unemployment and the declining wages of unskilled workers in these countries"; p. 71: "Most authors tend to agree that, although international trade has contributed to income inequality trends to some extent, it has not played a major role in pushing down the relevant wage of less-skilled workers".

2 G. Soros: Soros on Soros: Staying ahead of the curve (New York, John Wiley and Sons, 1995), p. 196.

3 Ethan B. Kapstein: "Workers and the world economy", in Foreign Affairs, May-June 1996, p. 32.

4 At Singapore, several ministers of trade stressed that the advantages expected from trade liberalization were sometimes slow to make themselves felt.

5 Document GB.261/WP/SDL/1.

6 United Nations: Report of the World Summit for Social Development (doc. A/CONF.166/9, 19 Apr. 1995), Commitment 3, p. 14.

7 Organisation for Economic Co-operation and Development: Trade, employment and labour standards: A study of core workers' rights in international trade (Paris, OECD, 1996).

8 World Trade Organization (WTO): Singapore Ministerial Declaration (doc. WT/MIN(96)/DEC, 18 Dec. 1996), para. 4.

9 See proposals on these lines under section D below. "The need for an overall evaluation after the fact" (pp. 60 et seq.), as well as the discussions held at the beginning of the 1930s concerning reports on unratified Conventions, although article 19 (Article 408 of the Treaty of Versailles) did not at the time stipulate any obligation to report on reasons preventing governments from ratifying certain Conventions (see minutes of the 48th Session of the Governing Body of the International Labour Office, Paris, April 1930, pp. 398-401).

10 ILO: Freedom of Association, Digest of decisions and principles of the Freedom of Association Committee of the Governing Body of the ILO, 4th edition (Geneva, 1996).

11 OECD, op. cit.

12 For the full text of article 41 see: ILO: Record of Proceedings, International Labour Conference, 29th Session, Montreal, 1946, pp. 604-607.

13 ILO: Constitutional Questions, Part I. Reports of the Conference Delegation on Constitutional Questions, 29th Session of the International Labour Conference, Montreal, 1946, para. 23.

14 It includes several examples of shortcomings to which I shall return in the second part -- for instance when it advocates "the prevention and elimination of congestion in urban areas".

15 See the interview of G. Bagwati: "Counsel from a trade guru", in Business Times, 8 Dec. 1996.

16 A drop-off which is sometimes masked by the effect of ratifications by countries which emerged after the breakup of the USSR, the Socialist Federal Republic of Yugoslavia and Czechoslovakia. For instance, among some 750 ratifications registered between 1992 and 1996, more than 500 were in fact confirmations by the new States of Conventions applicable on their territory before becoming Members of the Organization.

17 Document GB.254/16/19.

18 Article 10 refers only to "international Conventions".

19 The Conference, as a universal body, has absolute authority over the agenda drawn up with a view to adopting instruments and can express an opinion on the subject (or, in any case, express its disagreement by deleting an item from the agenda). But it is fairly obvious that the present schedule for determining the agenda, which extends from November to the following March, does not allow much scope for useful intervention on its part; furthermore, it is difficult to see how it would be possible to organize a coherent discussion on this matter, unless this were to take place as part of a preliminary discussion on a specific proposal.

20 ILO: Report of the Director-General, Part I: International Labour Standards, International Labour Conference, 70th Session, Geneva, 1984, p. 17.

21 By way of example, see Paragraph 26 of the Workers' Housing Recommendation, 1961 (No. 115), or Part V of the Utilisation of Spare Time Recommendation, 1924 (No. 21). This observation also holds true for a Convention rightly considered to be priority, such as Convention No. 122 (which calls upon States having ratified it to declare and pursue "an active policy designed to promote full, productive and freely chosen employment"). Given the wide variety of theories and schools of thought on the causes of unemployment, the content of this obligation can only but remain extremely vague from a legal standpoint. In these circumstances, how is it possible to measure the impact that the Convention might have as regards the attainment of its goal, namely the disappearance of unemployment? It is enough to consider the varying results obtained in their employment policies by countries having ratified this Convention to understand that its directives are clearly not a factor that can explain the differences. A Convention of this nature might be more practical if, in line with the spirit of the Declaration of Philadelphia, it called upon the States ratifying it to adopt a consistent attitude in international organizations which determined the conditions under which it would be possible to apply an active employment policy.

22 The Working Party on Policy regarding the Revision of Standards of the Committee on Legal Issues and International Labour Standards was set up by the Governing Body at its 262nd Session (March 1995) following discussions at the 1994 Session of the Conference on international labour standards. To date, it has held four meetings at the 264th Session (November 1995), 265th Session (March 1996), 267th Session (November 1996) and 268th Session (March 1997) of the Governing Body.

23 ILO: Report of the Director-General, Programme and structure of the ILO, International Labour Conference, 48th Session, Geneva, 1964, pp. 156-157.

24 ibid., p. 15.

25 ILO, ibid., p. 117.

26 Report of the Director-General, 1984, op. cit., p. 16.

27 This is the meaning a contrario of a speech given by Albert Thomas to the Third Session of the International Labour Conference, 1921, Record of Proceedings, Vol. 1, p. 219.

28 See, for example, Conventions Nos. 1, 4, 5 and 6.

29 ILO: Record of Proceedings, 15th Session of the International Labour Conference, Geneva, 1931, Vol. II, pp. 3-5.

30 This procedure creates an institutional link between the Conference, the body entrusted with standard setting, and national parliaments, upon which the binding effect of these instruments in the countries concerned is contingent. The aim of bringing the instrument before the competent authorities is obviously to allow parliaments to pronounce, in full knowledge of the facts, on the follow-up to be given to the instruments adopted by the ILO and to reach a decision, whatever the nature of this decision. Supervision of the obligation to submit instruments to the competent authorities has, over the years, become merely a formal exercise with very few exceptions. This constitutional obligation should assume once again the character that it had at the beginning: to encourage member States to ratify or implement the instruments they have adopted; to provide assistance to parliaments; to provide for a wider distribution of statements or proposals accompanying the submission, etc. This mechanism does not work for the early Conventions. This is one of the reasons why, following the Social Summit in Copenhagen, I asked those member States who had not done so to communicate their intentions to ratify the seven core Conventions. I do not, however, believe that this solution should extend to other instruments. It must remain exceptional and be reserved for core Conventions; if not it will cease to be effective.

31 ILO: Report of the Director-General, International Labour Conference, 27th Session, Paris, 1945, p. 155.

32 loc. cit.

33 It is for this reason that the Conference Delegation on Constitutional Questions tried in 1946 (without much success) to counter the undue competitive advantage enjoyed by federal countries.

34 See GB.267/LILS/WP/PRS/1 as well as draft article 45bis to be included in the Standing Orders of the Conference, which the Conference is called upon to examine this year in the context of the proposed constitutional amendment.

35 ILO: Constitutional Questions, Part I: Reports of the Delegation on Constitutional Questions, 29th Session of the International Labour Conference, Montreal, 1946, paras. 43 and 58.

36 This mechanism, which has no equivalent in other international organizations, informed decision-makers of the follow-up to be given to the instruments to be adopted by the Conference. It is possible that, over time or with the prospect of a change in the functions of parliaments which are the "natural" competent authorities, the obligation to bring the matter before the authorities was considered a formal obligation. However, paragraphs 5(c) and 6(c) of article 19 provide that the Office should be informed of measures taken. It would be timely to examine means of breathing new life into this obligation on the lines of the memorandum adopted by the Governing Body, so that it might once again be fully meaningful.

37 ILO: Constitutional Questions, op. cit., para. 74.

38 Report of the Director-General, 1964, op. cit., p. 169.

39 ibid., p. 170.

40 Report of the Director-General, 1984, op. cit., p. 20.

41 ILO: Report of the Committee of Experts on the Application of Conventions and Recommendations, International Labour Conference, 33rd Session, Geneva, 1950, Report III (Part IV), pp. 2 and 3; and idem, 53rd Session, 1969, Report III (Part IV), p. 186.

42 Report of the Director-General, 1964, op. cit., p. 157.

43 ILO: Constitutional Questions, op. cit., para. 12.

44 The case did, however, occur a number of years ago when, at the opening of the discussion, a delegation proposed that the discussion proceed on the basis of an alternative text. This occurred during discussions on the revision of the Migration for Employment Convention at the 32nd Session of the International Labour Conference in 1949 (Record of Proceedings, pp. 565 et seq.).

45 Report of the Director-General, 1984, op. cit. p. 20.

To order the Report

The print version of the Report of the Director-General is available in:
ISBN 92-2-10982-6
ISBN 92-2-20982-X
ISBN 92-2-30982-3
ISBN 92-2-60982-4
ISBN 92-2-50982-0
ISBN 92-2-70982-8
ISBN 92-2-40982-7
15 Swiss francs; US$13.50; £8.10

Updated by VC. Approved by RH. Last update: 26 January 2000.