22.pdfchild labor and multinational conduct: a comparison of international business and ...

ABSTRACT. Increasing attention to the issue of childlabor has been reflected in codes of conduct thatemerged in the past decade in particular. This paperexamines the way in which multinationals, businessassociations, governmental and non-governmentalorganizations deal with child labor in their codes.With a standardized framework, it analyzes 55 codesdrawn up by these different actors to influence firms’external, societal behavior. The exploratory studyhelps to identify the main issues related to child laborand the use of voluntary instruments such as codesof conduct. Apart from a specific indication of thetopics covered by the code, especially minimum-agerequirements, this also includes monitoring systemsand monitoring parties. Most important to companycodes are the sanctions imposed on business partnersin case of non-compliance. Severe measures may becounterproductive as they do not change the under-lying causes of child labor and can worsen thesituation of the child workers by driving them tomore hazardous work in the informal sector. Thisunderlines the importance of a broad rather than arestrictive approach to child labor in codes of conduct.The paper discusses the implications of this study,offering suggestions for future research.

KEY WORDS: business associations, child labor,codes of conduct, international organizations, multi-national enterprises, non-governmental organizations

ABBREVIATIONS: BSG – Business Support Group;ILO – International Labor Organization; IO –International (governmental) Organization; MNE –Multinational Enterprise; NGO – Non-governmentalOrganization; OECD – Organization for EconomicCooperation and Development; UNICEF – UnitedNations Children’s Fund

Introduction

Since the early 20th century, the issue of childlabor has been the subject of widespread regula-tory and societal attention. After several indus-trialized countries had adopted laws that limitedthe minimum working age of children and theirworking conditions, international organizationswere requested to advance similar measuresworldwide. The main vehicle of these interna-tional attempts has been the International LaborOrganization, created in 1919 with the abolish-ment of child labor as one of its fundamentalobjectives. This was reflected in the adoption ofthe Minimum Age Convention (No. 5) in thesame year.

In the course of the century, several otherinternational initiatives have been taken toprotect children, particularly the 1924Declaration of Geneva, the creation of UNICEF(1946), ILO’s 1973 minimum age convention(No. 146), the 1989 UN Declaration on theRights of the Child and, most recently, the 1999ILO Convention to combat the worst forms ofchild labor (No. 182). These different initiativesaimed to regulate children’s working conditionsin the traditional sense: governments, both

Child Labor and Multinational Conduct: A Comparison of International Business and

Ans KolkStakeholder Codes Rob van Tulder

Journal of Business Ethics 36: 291–301, 2002.© 2002 Kluwer Academic Publishers. Printed in the Netherlands.

Ans Kolk is associate professor in sustainable managementat the University of Amsterdam, Department ofAccountancy and Information Management.

Rob van Tulder is professor in business studies at theErasmus University, Department of Business-SocietyManagement.

nationally and internationally, that imposed lawsor tried to stimulate different company behavior.

In the 1970s, the role of multinational enter-prises, and the negative social and environmentalimplications of large-scale foreign investment,became controversial for the first time. Thisdebate, that re-emerged in the 1990s, hascentered around the relocation of production todeveloping countries with lower social and envi-ronmental standards, where an appropriateregulatory framework to protect (child) workersand the environment was either missing or notimplemented, and multinationals’ cooperationwith or implicit support for oppressive regimes(Kolk et al., 1999). It gave rise to academicreflection on multinationals’ ethical behavior,including their fundamental rights and duties,and universal moral norms (e.g. Bowie andVaaler, 1999; Buller et al., 1997; Donaldson,1997).

In response to these concerns, attempts weremade to regulate multinational behavior,especially through codes of conduct. In the1970s, this involved guidelines drawn up byinternational organizations. In spite of pressureby a number of governments in developed anddeveloping countries, and non-governmentalorganizations (NGOs), it proved impossible tomake these codes mandatory. The 1990s wit-nessed new efforts to formulate global standardsfor multinational conduct. Besides internationalorganizations, governments and NGOs, com-panies and their business associations also startedto draw up codes in which they voluntary com-mitted themselves to a particular set of norms andvalues.

In view of large societal attention for childlabor, an increasing number of codes hasaddressed this issue. This paper analyzes the wayin which multinationals and stakeholders dealwith child labor in their codes of conduct. Itgives an overview of 55 codes drawn up bymultinationals, business associations, internationalorganizations and NGOs – those stakeholdersthat have adopted child labor codes. For thispurpose, codes of conduct are defined as “guide-lines, recommendations and rules issued byentities within society (adopting body or actor)with the intent to affect the behavior of

(international) business entities (target) withinsociety in order to enhance corporate responsi-bility” (Kolk et al., 1999, p. 151). In thisdefinition, codes therefore always aim atinfluencing company conduct, focusing on theirexternal, societal, usually international, behavior.It excludes more internally oriented (ethical)codes, designed, for example, to regulateemployees’ ethical conduct when confrontedwith dilemmas such as conflict of interests, gifts,theft, insider trading and bribery. In view of itsfocus on the business-society interface, the paperwill neither enter into important debates on the(macro)economic and ethical dimensions of childlabor (Basu, 1999a; Hartman et al., 1999;Hindman and Smith, 1999).

The first part of the paper gives some back-ground information on the issue of child labor,particularly the status of international standards,and the characteristics of companies that employunderage workers. Subsequently, the results ofthe code analysis will be given, comparingdifferent actors’ positions, based on a standard-ized framework that examines specificity andcompliance mechanisms. It also deals with theimplications of sanctions, including the possibledrawbacks of strict compliance. The final sectiondiscusses the role of codes of conduct, and offerssuggestions for future research in this field.

Child labor: definition and characteristics

The international standards, or codes, drawn upby organizations such as the ILO reveal theexistence of a continuum that ranges fromacceptable to unacceptable forms of child labor.This has also been labeled as respectively “childwork” and “child labor” (George, 1990, pp.22–23). At one end of the continuum, tolerablework can be found, which is, as defined in ILOconvention No. 138 (1973), “light work whichis not likely to be harmful to [children’s] healthor development, and which is not such as toprejudice their attendance at school”. The con-vention stipulates a minimum age of 13 years forthis type of employment; in exceptional cases,this may be lowered to 12 years. Given the socialand economic situation, such work can be a

292 Ans Kolk and Rob van Tulder

reasonable means of earning a living for thechildren and their families; sometimes this is evenregarded as potentially beneficial for children’sdevelopment (UNICEF, 1997).

At the other end of the spectrum, untolerableforms of abusive, exploitative and dangerousworking conditions can be found. ILOConvention No. 182 (1999), for example, char-acterizes worst forms of child labor as “all formsof slavery or practices similar to slavery, such asthe sale and trafficking of children, debt bondageand serfdom and forced or compulsory labor,including forced or compulsory recruitment ofchildren for use in armed conflict”. This descrip-tion also covers the use of children for prostitu-tion, illicit activities, and other forms of workthat “is likely to harm the health, safety or moralsof children”. According to this convention (andto others, such as the UN convention on therights of the child), children are defined as allpersons younger than 18 years.

Although perceptions on child labor alsodepend on cultural traditions, and levels ofeconomic development and social conditions,there is a large consensus on the unacceptabilityof these worst forms of child labor. This is under-lined by the relatively rapid acceptance of C182,which, after one year, has been ratified by 56countries, with ratification processes under wayin many others (ILOLEX data, see Kolk et al.,2001). In between the two extremes, however,there is a large grey area in which much childlabor falls. According to the ILO (1998), a largenumber of the estimated 250 million childworkers between 5 and 14 years is confrontedwith some kind of hazards; in some countries,this even amounts to two-thirds (69%). To whichcountries this latter percentage refers, is notspecified by the ILO.

The problem with the implementation ofinternational norms, however, is that the majorityof child labor takes place in the informal sector,in agriculture, services and small-scale manufac-turing. These are usually not adequately coveredby national legislation. While attention focuseson child labor in export industries, they employonly a very small percentage, probably less than5%, of the child work force (UNICEF, 1997,p. 21). This means that unilateral international

sanctions to ban child labor seem to have aminimal impact, and might well be counter-productive as it drives children to the informalsector where control over labor conditions islacking. Especially for small, local companies,employing children is a necessary component ofa cost-reduction strategy and vital to increase thevery low profit margins. While the cost savingsachieved in this way are a relatively small per-centage of the final consumer price, the use ofchild labor can double the local entrepreneur’sown small income, as an ILO study in the Indiancarpet and bangles industries showed (Fyfe andJankanish, 1997).

Child labor in codes of conduct

Given these characteristics of child labor, how dodifferent stakeholders and companies themselvesdeal with the issue in their codes of conduct?To assess this, the next section examines thecontents of 55 codes adopted by multinationalfirms (MNEs), business associations (businesssupport groups, BSGs), international organiza-tions (IOs) and non-governmental organizations(NGOs). These codes explicitly address the issueof child labor.

Fourty company codes were drawn from a setof approximately 100 codes of the largest MNEsor of companies that have been pioneers in thefield of corporate social responsibility (Cf. Kolket al., 1999; see the Appendix). This selectionprocess has revealed that only a relatively smallnumber (13%) of the largest (Fortune 500) firmshas a company code that includes provisions onchild labor. The added selection of slightlysmaller firms, but pioneers in the adoption ofcodes, shows a substantially higher share of childlabor provisions. The sectors in which theseleading firms operate are the ones with thehighest likelihood of child labor: retail andapparel. Of the 24 important international codesof business associations, only 6 (25%) includechild labor provisions.

Most codes of international organizations dealwith child labor issues. The IO codes in thisarticle include the main standards and conven-tions with regard to child labor (ILO conventions

Child Labor and Multinational Conduct 293

Nos. 138 and 182, and the UN Convention onthe Rights of the Child) and to multinationalconduct (the recently revised OECD guidelinesfor multinational enterprises). Approximately40% of the most well-known international NGOcodes include provisions on child labor. The issueof child labor is, therefore, least addressed bycompanies and business associations, and mostby NGOs and international organizations.Nevertheless, the discussion on child labor seemsto focus primarily on the ability of companiesto effectively address the issue through a code ofconduct.

All codes were examined with a standardizedframework of analysis that was used in previousresearch, but adapted to suit the peculiarities of

child labor (see Table I). It focuses on thespecificity of the child labor provisions includedin the codes, particularly the minimum-agerequirement, and the compliance mechanisms,including monitoring and the type of sanctionsin case of violations. Non-compliance leads todilemmas with regard to the fate of the children:will they be dismissed, leading to the impressionof “clean hands”, but without addressing theproblem as such, or are alternative measuresoffered to help the child workers? Some of themain features of the code analysis will be brieflydiscussed below, divided into three sections:specificity, monitoring, and types and implica-tions of sanctions.


TABLE IA model to analyze and compare codes of conduct on child labor issues

Criteria Short elaboration Classification

Specificity

1.1. Minimum age Does the code include a minimum age to employment? yes (age); noto employment If so, what age?

1.2. Applicability Is this a universal minimum age or are country-specific n.a.; universal; exceptions indicated? country-specific

1.3. Organization To whom is the code addressed? General, governments; actor category targeted internal operations of specific firms; business partners (exact wording)

(suppliers, subcontractors, vendors, manufacturers)

1.4. Reference Is reference made to international standards (ILO, UN), either none; home; host;implicit or explicit, or to home-country or host-country laws? international

(implicit/explicit)

1.5. Nature of Are alternative measures included in the code (such as education broad; strictcode for children)? Or does the code only prohibit child labor?

Compliance

2.1. Monitoring Good insight into system and process (clear); reference to some clear; clear to systems and parts, but criteria or time frames are lacking (clear to vague); vague; vague; processes only general reference to monitoring without details (vague). none

2.2. Position of Firms themselves (1st party); BSGs (2nd party); external ranging from: monitoring professionals paid by firms (3rd party); combinations of different 1st to 6th partyactor actors (4th party); NGOs (5th party); legal authorities (6th party).

2.3 Sanctions Measures have no large implications, e.g. warnings and exclusion none; mild; of membership (mild); threat to business activities (severe). severe

2.4 Sanctions to Measures such as fines, or demands for corrective action (mild); n.a.; none; mild; third parties severance of relationship, cancellation of contract (severe) severe

Source: Adapted from Kolk et al. (1999), p. 155.

Specificity

As examined in the previous section on thedefinition of child labor, one of the main topicsconcerning the specificity of codes has been theminimum-age requirement. Especially ILOConvention No. 138 has been important in thediscussion on international standards. In view ofthe exceptions that it makes to the universallyapplicable minimum age of 15 years, this ILOcode is not included in Table IIa. The table givesan overview of the minimum ages stipulated inthe other three IO codes, the five NGO codes,the six BSG (business support groups) codes, andthe 40 MNE codes.

Almost half of the codes does not mention aminimum age. If they do, however, 14 is mostfrequently indicated, which applies particularlyto business codes (both MNE and BSGs). Thisage is incompatible with Convention 138. A fewcompany codes explicitly refer to this ILO code,which means that 15 years is mentioned, with 14as developing-country exception. Only oneNGO and one MNE code stipulate 18 as

minimum age; the company concerned, Nike,however, includes two minimum ages: 18 for theproduction of footwear, and 16 for apparel,accessories and equipment.

Of the codes that mention a minimum age toemployment, only a small minority has universalapplicability (Table IIb). Usually, codes arecountry-specific, which means that they formu-late a minimum age, but that a higher age willprevail if host-country laws stipulate this higherage for employment or for finalizing educationalrequirements. The Reebok code of conduct canbe cited as example:

The company will not work with business partnersthat use child labour. The term ‘child’ generallyrefers to a person who is less than 14 years of age,or younger than the age for completing compul-sory education if that age is higher than 14. Incountries where the law defines ‘child’ to includeindividuals who are older than 14, the companywill apply that definition.

More generally, if business codes refer to stan-dards, this usually involves the host country


TABLE IIaMinimum ages to employment mentioned in codes of conduct (in % of code type)

No age 14 15/14 15 16 18

International organizations (n = 3)a 100NGOs (n = 5) 020 20 20 20 20Business support groups (n = 6) 067 33MNEs (n = 40) 043 28 8 15 05 0b3b

a ILO convention 138 is not included as it defines exceptions to the minimum age of 15. b This code (Nike) gives two minimum ages: 18 for the production of footwear, and 16 for the production ofapparel, accessories and equipment. As this is the only MNE code that mentions 18 years, it is included in the18 category only.

TABLE IIbApplicability of minimum ages stipulated in codes of conduct (in % of code type)

Universal Country-specific Not defined

International organizations (n = 4) 25 75NGOs (n = 5) 20 80Business support groups (n = 6) 33 67MNEs (n = 40) 8 70 23

(Table IIc). Some MNEs explicitly refer to inter-national standards, as already mentioned, or dothis implicitly by including Convention 138’sminimum age of 15 in their codes of conduct.The majority of NGO codes explicitly refers tointernational standards, while all IO codesmention other international conventions.

The final aspect of specificity concerns theorganizations targeted by the code. By the natureof the definition used in this paper, all stake-holder codes focus on business, either generally(all employers), or specifically (one sector). Mostinteresting in this regard are the organizations atwhich MNEs aim. Codes use a variety of terms,ranging from business partners, suppliers,vendors, subcontractors to manfacturers. Themajority (63%), however, does not define theconcepts, failing to explain the intended scopeof their requirements. One quarter of the MNEsclearly define the business partners at which theircodes are targeted. An example is Stage Stores,which states that:

We have defined business partners as vendors, man-ufacturers, contractors, subcontractors and othersuppliers who provide labor and/or materialincluding fabric, sundries, chemicals and trimutilized in the manufacture and finishing ofproducts that are ordered by or through us.

The remaining 12% of the MNE codes fall inbetween by defining the targeted organization,but without providing insight into the extent towhich it is to be implemented in the subcon-tracting chain.

Monitoring

A majority of the codes indicates how the pro-visions of codes are monitored, and by whom.80% of the NGO codes clearly specifies themonitoring process and system, including thecriteria for assessment; this also applies to 28% ofthe MNE codes. Most BSG codes, and half ofthe IO codes envisage a monitoring process butdo not mention criteria or time-frames. Theremaining just vaguely refers to monitoringwithout giving any further details (Table IIIa).

With regard to the monitoring party, mostMNEs prefer to do this themselves, whereasinternational organizations rely on legal author-ities. Third-party monitoring, by external pro-fessionals, is mentioned in 10% of the MNEcodes (i.e. 17% of the codes with monitoringprovisions). Most NGOs prefer monitoring bycombinations of actors, such as NGOs and BSGs(4th party); some firms have included this in theircodes (Table IIIb).

Sanctions and implications

A vital issue concerns the sanctions that firmsimpose on their business partners in case of non-compliance with the provisions included in thecodes. Almost half of the MNE codes does notinclude sanctions, while 45% takes strongmeasures, such as the termination of businessrelationships or the cancellation of orders. Morethan 80% of the BSG codes also announces totake such severe sanctions.

This raises the question, however, what will


TABLE IIcReference to home-country, host-country or international standards in codes of conduct (in % of code type)

No Home- Host- International International standards country country standards standards

standards standards (implicit) (explicit)

International organizations (n = 4) 100NGOs (n = 5) 20 20 060Business support groups (n = 6) 67 33MNEs (n = 40) 18 55 18 010

happen to the children in such cases. If the MNEterminates the relationship or if the children arefired, the problem of child labor does notdisappear as economic conditions have notchanged. In fact, the situation can even deteri-orate because the children are forced to findwork elsewhere, usually in the informal sectorwhere any form of control on labor conditionsis lacking. Studies in Bangladesh, for example,showed that the threat of U.S. sanctions, and theconsequent expectation of future restrictive lawson child labor, led to the dismissal of 40 000children in the garment industry; between 5000and 7000 girls moved to prostitution (Basu,1999b, p. 86; Fyfe and Jankanish, 1997, p. 87).

Therefore, “responsible” business conductwould mean that firms take measures that helpimprove the situation through compensatoryprograms for the children, such as access toeducation, food and health care, and alternativesources of income for the children’s families (e.g.UNICEF, 1997; Save the Children, 2000). Of the40 MNE codes, only three take such a broadapproach. Hennes & Mauritz is an interestingexample:

“H & M does not accept child labour. We areconcerned about the situation of children in manyparts of the world. We acknowledge the fact thatchild labour does exist and can’t be eradicated withrules or inspections, as long as the children’s socialsituation is not improved. We want to activelywork with factories and with NGOs in third worldcountries, to try to improve the situation for thechildren affected by our ban on child labour.

If a child is found working in any of the fac-tories producing our garments, we will request thefactory to make sure that the measures taken arein the child’s best interest. We will, in co-opera-tion with the factory, seek to find a satisfactorysolution, taking into consideration the child’s age,social situation, education etc. We will not ask afactory to dismiss a child without a discussionabout the child’s future. Any measures taken shouldalways aim to improve, not worsen, each individualchild’s situation. Any costs for education, etc. haveto be paid by the factory.”

Such an approach, however, also raises questionsabout the boundaries of corporate responsibilityand what should be provided by governments, inboth developed and developing countries.


TABLE IIIaClarity of monitoring systems and processes in codes of conduct (in % of code type)

None Vague Vague/cleara Clear

International organizations (n = 4) 50 50NGOs (n = 5) 20 80Business support groups (n = 6) 33 67MNEs (n = 40) 40 15 18 28

a Vague/clear means that monitoring is envisaged, but that criteria for assessment or time-frames are lacking.

TABLE IIIbMonitoring actors mentioned in codes of conducta (in % of code type)

None 1st party 2nd party 3rd party 4th party 6th party

International organizations (n = 4) 100NGOs (n = 5) 20 20 60Business support groups (n = 6) 83 17MNEs (n = 40) 40 43 03 10 05

a 1st party: firms themselves; 2nd party: BSGs; 3rd party: external professionals paid by firms; 4th party: com-binations of different actors; 5th party: NGOs; 6th party: legal authorities.

Conclusions and research implications

This exploratory study of 55 internationalbusiness and stakeholder codes has providedinsight into the way in which child labor issuesare addressed in these formal documents aimedat regulating corporate conduct. Using a stan-dardized framework of analysis, specificity andcompliance were examined, paying particularattention to minimum-age requirements, moni-toring and sanctions. Codes drawn up by NGOsturned out to be most specific, and those devel-oped by business associations the least. Thisconfirms other research on codes of conduct(Kolk et al., 1999), and the role of businessassociations in providing so-called “club goods”(Prakash, 2000).

Companies are supposed to be leading inaddressing the issue of child labor. This articlehas shown that this claim needs considerablemodification. Not only is the number of largemultinational companies that has child laborprovisions in its codes relatively limited (less than20%), the codes with such provisions score mixedresults with regard to code specificity and com-pliance likelihood. Many company codes suggestuniversal applicability (of a minimum age, forexample), are country-specific and prefer vagueor internal monitoring procedures, whereasothers include clear monitoring mechanisms andsanctions for their business partners in case ofnon-compliance.

The imposition of severe sanctions, however,proves to be a complicated issue. At first sight,one is inclined to applaud strict measures to banchild labor as this seems the logical consequenceof a business policy, usually developed after strongsocietal pressure. Nevertheless, on further con-sideration, such “clean hands” may well becounterproductive because it does not tackle thechild labor problem and its underlying causes.Moreover, it can worsen the situation for thechildren who are forced to find work elsewhere,often in the informal sector. In this regard, it isimportant to note that only a small fraction ofthe estimated 250 million child workers isemployed in export-related industries. More than95% works in agriculture, small businesses orprovides services outside the formal economy.

In spite of this caveat, however, the positiontaken by multinationals can be important. Theyset a standard for corporate conduct and influ-ence other firms and governments, thus perhapshelping to further international attempts toaddress child labor. This is even more the case ifthey take responsibility for improving children’sworking conditions, education and health, orassist in finding alternative ways of income-generation for the families. Such a broadapproach is only taken by three multinationalsout of the set of 40 analyzed in this paper.Because the 40 codes only represent a smallportion of the multinationals with a companycode (let alone of the whole set of multinationalswithout one), it can be assessed that the broadapproach to the child labor issue is currentlyembraced by less than 5% percent of companies.Nevertheless, the potential role of large interna-tional companies remains interesting, but cannotbe considered in isolation.

In this regard, it must be noted that thespecificity of codes drawn up by other societalstakeholders remains higher, whereas the actionof international organizations such as FIFA canbetter trigger broader-defined codes than com-panies themselves (Cf. Van Tulder and Kolk,2001). At the same time, however, governmentalratifications of international agreements such asthe ILO child labor convention are still sur-rounded by many caveats. This implies that theissue of child labor cannot be resolved bybusiness-government interaction only, butrequires the involvement of other national andinternational stakeholders as well. Researchefforts in this direction might focus on theimportance of firms’ country of origin, particu-larly the regulatory and societal context,including the role of stakeholders (Langlois andSchlegelmilch, 1990; Van Tulder and Kolk,2001). Available knowledge on ethics manage-ment in different cultures can be helpful in thisregard (Buller et al., 1997; Jackson, 1997;Weaver, 2001). Apart from the analysis of largersets of multinationals with different nationalities,sector-specific case studies of codes of conductcan be useful to understand the dynamics.

This article has provided insight into thecontents and compliance likelihood of codes of


conduct that address the very important moralissue of child labour. The actual effectiveness ofcodes of conduct in tackling this child laborproblem is difficult to assess, however, because ofits multi-faceted nature. Some of the companiesthat we approached for this study suggested thatit might be more effective not to have a code ofconduct (Cf. Kolk et al., 2001). Firstly, theyargue that a strict code (without the broaderapproach as distinguished in this article) mighteven aggravate the problem. Secondly, a broadercode might damage a company’s reputation as aresult of negative media coverage in its countryof origin (this depends on the country inquestion, see Van Tulder and Kolk, 2001).Thirdly, the company view on child labor, asincluded in the code, can differ from that of thehost government, possibly leading to barriers toentry for companies that might otherwise havehad a positive effect on the emancipation ofchildren.

Additional research is required to obtaininsight into business and stakeholder ideas aboutthe effectiveness of codes, and what roles firmsand governments ought to play in attempts toaddress child labor. This could also include aconsideration of instruments and mechanismsother than codes of conduct, and an analysis ofthose companies that claim to be concernedabout child labor but do not have codes ofconduct (or codes with only modest monitoringprovisions or that take a strict approach).Moreover, the communication and implementa-tion process of codes of conduct, and the relatedorganizational aspects, such as company structure,culture and leadership, that influence staffbehavior deserve further attention. This couldbuild on insights from research on ethical codesand the organizational and individual factors thataffect ethical conduct (e.g. Cleek and Leonard,1998; Somers, 2001). Future studies on this topicmight also focus on those parts of the supplychain that are usually not covered by codes ofconduct, but that employ the overwhelmingmajority of the children.

This article has identified a “broader” and a“stricter” approach to child labor. Hence, theprevention of child labor seems to be much lessof a universal “hypernorm” than is sometimes

suggested. This offers ample opportunities formore research into the actual, “realistic” and“pragmatic” strategies that are adopted for thepurpose, and their effectiveness in addressingchild labor.

Acknowledgements

This article is one of the publications resultingfrom a joint, long-term project on multinationalenterprises and corporate social responsibility.Inge Sloekers is gratefully acknowledged for hercontribution to creating the dataset on which thispaper is based. The authors would also like tothank the anonymous reviewers for their helpfulcomments.

Appendix. List of codes of conduct (analyzed in thisarticle)

International governmental organizations (

n=4)ILO (Minimum Age Convention No. 138, 1973)ILO (Convention 182 on Elimination of Worst

Forms of Child Labor, 1999)OECD (OECD Guidelines for Multinational

Enterprises, revised version, 2000)UN (United Nations Convention on the Rights

of the Child, 1989)

Non-governmental organizations (n = 5)Catholic Institute for International RelationsClean Clothes CampaignChristian AidDanish Confederation of Labour UnionsFédération Internationale de Football Association

Business associations (n = 6)American Apparel Manufacturers AssocationAthletic Footwear AssociationAssociation Merchandising CooperationBritish Toy and Hobby AssociationInternational Council of Toy IndustriesToy Industries of Europe

Firms (n = 40)Body ShopBurton GroupC&A


ConverseDayton HudsonDress BarnElf AcquitaineFederated DepartmentFruit of the LoomGapHennes & MauritzJCPenneyJones Apparel GroupKellwoodKmartLands’ EndLevi StraussLimitedLiz ClaiborneMercantile StoresMerckNestléNikeNorthern TelecomOxford IndustriesPhillips-Van HeusenPricecostcoPumaReebokSalant CorporationSara LeeStage StoresTalbotsTultex CorporationVenture StoresVF CorporationWal-Mart StoresWalt DisneyWarnacoWE

References

Basu, K.: 1999a, ‘Child Labor: Cause, Consequence,and Cure, with Remarks on International LaborStandards’, Journal of Economic Literature XXXVII,1083–1119.

Basu, K.: 1999b, ‘International Labor Standards andChild Labor’, Challenge (September-October),80–93.

Bowie, N. and P. Vaaler: 1999, ‘Some Arguments forUniversal Moral Standards’, in G. Enderle (ed.),International Business Ethics. Challenges and

Approaches (University of Notre Dame Press, NotreDame, IN), pp. 160–173.

Buller, P. F., J. J. Kohls and K. S. Andersen: 1997, ‘AModel for Addressing Cross-cultural EthicalConflicts’, Business and Society 26(2), 169–193.

Cleek, M. A. and S. L. Leonard: 1998, ‘CanCorporate Codes of Ethics Influence Behavior?’,Journal of Business Ethics 17, 619–630.

Donaldson, T.: 1997, ‘Fundamental Rights andMultinational Duties’, in N. E. Bowie (ed.), EthicalTheory and Business, Fifth Edition (Prentice Hall,Upper Saddle River, NJ), pp. 535–545.

Fyfe, A. and M. Jankanish: 1997, Trade Unions andChild Labour (ILO, Geneva).

George, I.: 1990, Child Labour and Child Work (AshishPublishing House, New Delhi).

Hartman, L. P., B. Shaw and R. Stevenson: 1999,‘Balancing Sweatshop Ethics and Economics’Labor Standards Working Paper No. 9901<http://www.ssrn.com>.

Hindman, H. D. and C. G. Smith: 1999, ‘Cross-cultural Ethics and the Child Labor Problem’,Journal of Business Ethics 19, 21–33.

Jackson, K. T.: 1997, ‘Globalizing Corporate EthicsPrograms: Perils and Prospects’, Journal of BusinessEthics 16, 1227–1235.

ILO: 1998, ‘Statistics on Working Children andHazardous Child Labour in Brief ’, Bureau ofStatistics, Geneva.

Kolk, A., R. van Tulder and C. Welters: 1999,‘International Codes of Conduct and CorporateSocial Responsibility: Can TransnationalCorporations Regulate Themselves?’, TransnationalCorporations 8(1), 143–180.

Kolk, A., R. van Tulder and I. Sloekers: 2001,‘Werken gedragscodes kinderarbeid?’, ESB86(4295), 160–163.

Langlois, C. C. and B. B. Schlegelmilch: 1990, ‘DoCorporate Codes of Ethics Reflect NationalCharacter? Evidence from Europe and the UnitedStates’, Journal of International Business Studies 21(4),519–539.

Prakash, A.: 2000, ‘Responsible Care: AnAssessment’, Business and Society 39(2), 183–209.

Save the Children: 2000, Big Business, Small Hands –Responsible Approaches to Child Labour (The WolseyPress, London).

Somers, M. J.: 2001, ‘Ethical Codes of Conduct andOrganizational Context: A Study of theRelationship Between Codes of Conduct,Employee Behavior and Organizational Values’,Journal of Business Ethics 30, 185–195.


Tulder, R. van and A. Kolk: 2001, ‘Multinationalityand Corporate Ethics: Codes of Conduct in theSporting Goods Industry’, Journal of InternationalBusiness Studies 32(2), 267–283.

UNICEF: The State of the World’s Children (OxfordUniversity Press, New York).

Weaver, G. R.: 2001, ‘Ethics Programs in GlobalBusinesses: Culture’s Role in Managing Ethics’,Journal of Business Ethics 30, 3–15.


Dr. Ans Kolk, University of Amsterdam,

Faculty of Economics, Department of Accountancy & Information

Management, Roetersstraat II,

1018 WB Amsterdam, The Netherlands.

E-mail: [email protected]

22.pdfchild labor and multinational conduct: a comparison of international business and ...

Documents