Outsourcing Regulation: Analyzing Nongovernmental Systems of Labor Standards and Monitoring

Abstract

A range of new nongovernmental systems for advancing labor standards and enforcement have emerged over the last 5 years. This article comparatively assesses these multistakeholder systems of codes of conduct and monitoring, discusses their underlying models of regulation, and proposes a set of criteria for evaluating their effectiveness, including their legitimacy, rigor, accountability, and complementarity. Critical issues are raised about the transparency of existing initiatives, independence of monitors, convergence of standards, and dynamics among nongovernmental regulation, unions, and state enforcement. The article concludes by arguing that with increased transparency, improved technical capacities, and new mechanisms of accountability to workers and consumers, nongovernmental monitoring could complement existing state regulatory systems.

Full text

The Policy Studies Journal, Vol. 31, No. 1, 2003
0162-895X © 2003 The Policy Studies Journal
Published by Blackwell Publishing. Inc., 350 Main Street, Malden, MA 02148, USA, and 108 Cowley Road, Oxford, OX4 IJF, UK.
111
Outsourcing Regulation: Analyzing Nongovernmental
Systems of Labor Standards and Monitoring
Dara O’Rourke
Arange of new nongovernmental systems for advancing labor standards and enforcement
have emerged over the last 5 years. This article comparatively assesses these multistake-
holder systems of codes of conduct and monitoring, discusses their underlying models of
regulation, and proposes a set of criteria for evaluating their effectiveness, including their
legitimacy, rigor, accountability, and complementarity. Critical issues are raised about the
transparency of existing initiatives, independence of monitors, convergence of standards,
and dynamics among nongovernmental regulation, unions, and state enforcement. The
article concludes by arguing that with increased transparency, improved technical capac-
ities, and new mechanisms of accountability to workers and consumers, nongovernmental
monitoring could complement existing state regulatory systems.
Introduction
Recent accounts of the proliferation of “sweatshops” and deplorable condi-
tions of work in both developing and industrialized countries call into question
the effectiveness of existing systems of governance over labor practices (Chan,
2001; Block, Roberts, Ozeki, & Roomkin, 2001; Compa, 2001; Jeffcott & Yanz,
1999). Nongovernmental organizations (NGOs), unions, consumers, and even
some firms—for different reasons—criticize current systems of labor standards
and enforcement for their opaque processes, limited scope and coverage, weak
capacities of regulators, limited leverage over recalcitrant firms, and general inef-
fectiveness (Nadvi & Wältring, 2001). These critiques have grown louder in the
face of new challenges of regulating global firms and their mobile supply chains.
Traditional labor regulations and the government-implemented monitoring and
enforcement systems on which they depend appear simply to be outpaced by
changes in the global economy.
Anumber of strategies have emerged in response to these perceived limita-
tions of labor regulation. These range from efforts to strengthen the capacity of

local labor inspectorates to developing international systems of labor regulation
that operate above the level of the nation-state, such as attempts to institutional-
ize a “social clause” in the World Trade Organization (WTO) (Hughes &
Wilkinson, 1998; Maskus, 1997) and efforts to advance a set of “core standards”
for labor protections (Lee, 1997) through declarations of the International Labour
Organization (ILO). The ILO core standards are outlined in Table 1.
This article focuses on yet another strategy gaining momentum over the last
several years, involving the development and implementation of nongovernmen-
tal systems of labor regulation. These nongovernmental regulatory strategies seek
to function along the lines of outsourced production: regulating firms across their
supply chains through “voluntary” standards (sometimes developed in coopera-
tion with NGOs and unions), internal and external monitoring systems, and new
sanctions and incentives.
In order to advance these strategies, nongovernmental organizations are
taking on activities that were previously the sole purview of state and interna-
tional regulatory bodies. Participants are working to fill holes in traditional gov-
ernment regulation and to transform the nature of existing regulation. These
initiatives involve new forms of collaboration, new roles for nongovernmental
organizations, new responsibilities of firms, and new responses from local and
national government authorities. In many ways, these systems are in flux, as is
global labor regulation more broadly (Arthurs, 2001).
Nongovernmental systems of labor regulation are expanding extremely
rapidly across industries and regulatory arenas—from garments to shoes, toys,
2Policy Studies Journal, 31:1
Table 1. ILO Core Conventions
Core Principle Convention Countries U.S.
ratifying ratified
Freedom of association Freedom of Association and Protection of the 141 No
and collective Right to Organize, Convention No. 87, 1948
bargaining Right to Organize and Collective Bargaining, 152 No
Convention No. 98, 1949
Elimination of forced Forced Labour, Convention No. 29, 1930 161 No
and compulsory labor Abolition of Forced Labour, Convention No. 156 Yes
105, 1957
Elimination of Equal Remuneration, Convention No. 100, 1951 159 No
discrimination in the Discrimination (Employment and Occupation), 157 No
workplace Convention No. 111, 1958
Elimination of child Minimum Age, Convention No. 138, 1973 119 No
labor Worst Forms of Child Labour, Convention No. 132 Yes
182, 1999
Source: International Labour Organisation. Retrieved from
http://www.ilo.org/public/english/standards/norm/whatare/fundam/

forest products, oil and gas, mining, chemicals, coffee, electronics, and even
tourism (Herrnstadt, 2001; Gereffi, Garcia-Johnson, & Sasser, 2001; Wick, 2001;
Cashore, 2002; Utting, 2002). However, to date very little rigorous analysis has
been conducted on the impacts and implications of these potentially transforma-
tive institutions. And the analysis that has been conducted has been highly con-
tentious, either advocating programs or dismissing them out of hand.
Advocates tout these initiatives as more flexible, efficient, democratic, and
effective than traditional labor regulation (see, for instance, Bernstein, 2001),
whereas critics conversely assert that nongovernmental regulation is a corrupt
attempt to free industry from the last vestiges of state regulation and union organ-
izing (see, for instance, Justice, 2001). Proponents argue that these systems can
supplement and even support government regulation, but opponents assert that
nongovernmental regulation implicitly challenges the legitimacy and efficacy of
state regulation. Some fear nongovernmental systems of regulation will preempt
or “crowd out” worker organizing efforts and the current role of unions, whereas
others believe these systems can support worker empowerment and participation
in shop-floor negotiations. Some believe monitoring and certification will provide
consumers with a false sense that problems have been solved and will demobi-
lize international labor and environmental campaigns, but others see the infor-
mation generated by nongovernmental regulation as key to transforming how we
produce, consume, and regulate around the world.
This article seeks to critically and constructively engage this heated debate
and assess new systems of nongovernmental labor monitoring and regulation.
Based on interviews with staff of each of the leading initiatives in the United
States and Europe; interviews with multinational managers and advocacy organ-
izations; a review of the existing literature and program documents; and direct
evaluation of monitoring activities in China, Korea, Indonesia, and Mexico, this
article details key efforts at nongovernmental labor regulation, explains how these
systems function, describes the challenges they face, and evaluates their effec-
tiveness in improving labor practices.
The article begins by describing and evaluating the key nongovernmental
labor regulatory programs in the United States and Europe: the Fair Labor
Association, the Worldwide Responsible Apparel Production certification
program, Social Accountability International, the Ethical Trading Initiative, the
Fair Wear Foundation, the Workers Rights Consortium, private firm internal mon-
itoring initiatives, and independent investigations and organizing by NGOs. The
article comparatively assesses these systems, discusses their different models of
regulation, and proposes a set of criteria for evaluating their effectiveness. The
article concludes by arguing that nongovernmental regulation can be effective
under certain conditions—when monitoring and remediation are technically rig-
orous, publicly transparent, politically legitimate, and accountable to key stake-
holders (in particular to workers and consumers).
Before discussing the details of these programs, it is worth exploring briefly
O’Rourke: Outsourcing Regulation 3

why nongovernmental regulatory systems are emerging and what they seek to
achieve.
The Emergence of Nongovernmental Regulation
Much of the existing literature on “privatized” regulation explains non-
governmental regulation as a response to two connected trends: the weakening of
national governments (due to globalization, neoliberal movements to shrink the
state, or simply the failure of state bureaucracies) and the strengthening of multi-
national corporations (Strange, 1996; Evans, 1997; Schmidt, 1995; Cutler,
Haufler, & Porter, 1999). States are turning to “market-based” and private volun-
tary strategies as an alternative or a supplement to traditional regulation (Haufler,
2001). Firms support this trend as a step towards enlightened “self-regulation.”
Some governments clearly view nongovernmental regulatory strategies as
more “flexible” and “responsive” regulatory mechanisms (Ayres & Braithwaite,
1992) that can supplement overworked and underresourced state labor agencies
(Nadvi & Wältring, 2001; Reinicke, 1998). Market-based mechanisms and public
disclosure systems are also increasingly popular as methods of regulation
(Braithwaite & Drahos, 2000). Nonstate, market-driven systems (Cashore, 2002)
are attractive to governments, as they can take up some of the demands placed on
state agencies, an especially pressing concern in developing nations. By enlisting
the energies of multinational firms, private sector auditors, and NGOs in moni-
toring labor performance, governments can potentially increase compliance
without increasing state budgets or staff (Nadvi & Wältring, 2001).
Firms are “voluntarily” participating in or leading these programs because of
significant new pressures to improve their labor, environmental, and social per-
formance. Brand-sensitive firms are joining nongovernmental initiatives or creat-
ing their own codes of conduct and monitoring schemes in response to direct
pressures and demands from activists. Codes and monitoring systems are viewed
as a strategy to reduce reputational risks in the marketplace (Conroy, 2001). One
bad supplier can significantly damage a firm’s reputation and, in turn, its sales and
stock value. Firms with suppliers in countries with weak enforcement systems or
a poor track record on child labor are viewed with suspicion by informed stake-
holders. These firms thus need independent means of establishing their “good”
performance (Nadvi & Wältring, 2001, p. 28). Firms may also be advancing these
programs as a strategy to preempt stricter state regulation and to undermine the
legitimacy of the state and unions in regulating labor issues.
Codes of conduct and monitoring systems also offer several advantages
over traditional regulatory regimes in the eyes of firms. These systems build on
some of the central organizational principles of contemporary globalization—
outsourcing production, monitoring, and continuous improvement—and so can
advance a form of regulation that multinational firms find compatible with
business strategies (Wach & Nadvi, 2000; Sabel, 1994). Supplier firms in devel-
4Policy Studies Journal, 31:1

oping countries increasingly see compliance with new labor standards as a pre-
requisite to entry into global supply chains. Today contractors have to perform
not only to world-class standards on quality and price but also on labor and envi-
ronmental standards. Meeting these new standards can mean greater market
access, closer ties to global buyers, and in some cases price premiums (Conroy,
2001; Nadvi & Kazmi, 2001).
The development of nongovernmental regulatory systems, however, has also
been motivated by public perceptions of the failures of state and intergovern-
mental regulation and by growing demands from civil society actors for new
mechanisms of corporate accountability. One surprising trend in the emergence
of nongovernmental regulation has been the role and support of NGOs that have
historically been extremely suspicious of market mechanisms, weakening state
roles, and privatized regulation. However, for groups interested in strengthening
the enforcement of labor standards, nongovernmental regulation is attractive as a
supplemental system of monitoring and enforcement. Increasingly influential
NGOs are thus advancing market-oriented, nongovernmental standards and mon-
itoring systems as a supplement to state regulation in countries where it is inef-
fective and as a new point of leverage over firms operating globally (Cashore,
2002; Conroy, 2001).
Consumers are in many ways at the root of these processes. Increasingly
aware consumers are demanding more information on the products they buy and
on systems to help them avoid “sweatshop” products. Elliott and Freeman (2001)
report that over 80% of consumers polled in the United States are willing to pay
more for products that are made under “good” conditions, over 75% report feeling
a moral obligation to help improve workers’ conditions, and over 90% agree that
countries should be required to maintain minimum standards for working condi-
tions. Consumers in particular seek to avoid “bad” companies, such as those impli-
cated in “sweatshop” production. A recent poll by Environics International (1999)
found that 51% of Americans reported “punishing” a firm for bad social per-
formance in the previous year. Codes of conduct, monitoring systems, and label-
ing schemes have emerged as means to provide consumers with the information
needed to buy their desired level of workplace conditions and to send market
signals to firms to improve their performance (Liubicic, 1998; Freeman, 1994).
Outsourcing Labor Regulation
Nongovernmental systems of labor monitoring and regulation are both more
diverse and “messier” than traditional regulatory approaches. These initiatives go
beyond the past procedures of government stipulation of fixed rules and standards,
government monitoring and enforcement, and judicial review (Arthurs, 2001;
Lipschutz, 2000; Reinicke, 1998). Nongovernmental initiatives involve multiple
actors in new roles and relationships, experimenting with new processes of stan-
dard setting, monitoring, benchmarking, and enforcement. In a number of regards,
O’Rourke: Outsourcing Regulation 5

these new nongovernmental regulatory strategies are following in the footloose
steps of global production processes. As networks of production extend along
increasingly complex supply chains, interested stakeholders are exploring systems
of dispersed but interconnected regulation over production. These emerging reg-
ulatory systems are almost as complex as the supply chains they seek to monitor.
They include chains of standard setters, layers of monitoring and enforcement,
and competing systems of incentives and action.
Acritical shift in this process is the move from factory-centered, state regu-
lation focusing on individual sites of production to supply-chain and “brand” reg-
ulation focusing on multiple actors in a production chain. This shift involves
establishing systems of accountability and management of performance across
factories and nations. Outsourced regulation attempts to create a network of reg-
ulators, involving multiple stakeholders along global supply chains. Whereas
traditional regulation involves a national government establishing standards and
policing performance, outsourced regulation involves NGOs and firms in stan-
dard setting and monitoring, with the ILO core standards and local laws as
minimum standards. Whereas traditional regulation uses state sanctions to enforce
standards, outsourced regulation relies largely on “market” sanctions—either
through interfirm purchasing decisions or NGO consumer campaigns. Whereas
government regulation is hierarchical and arms-length, outsourced regulation is
networked at multiple levels and engaged with multiple actors in the supply chain.
Nongovernmental regulation is actually a diverse family of regulatory strate-
gies, many of which are currently in competition. Even the terminology used to
describe these systems is contested. In this article, I use internal monitoring to
refer to monitoring conducted by brands and retailers, external monitoring to refer
to monitoring conducted by third-party organizations, and verification to refer to
independent evaluations (not paid for by those being monitored) of the results of
codes and monitoring systems. Detailed descriptions of existing programs can
help elucidate these different models.
Codes of Conduct
Nongovernmental regulatory systems are based on voluntary standards,
usually embodied in codes of conduct, which specify norms and rules by which
to evaluate factory performance. These standards are sometimes quite specific,
detailing precise rules of action, while in other cases presenting only general prin-
ciples of good practice (Douglas, 2001; Braithwaite & Drahos, 2000). In both the
United States and Europe, NGOs are now at the forefront of efforts to develop
entirely new institutions (some nongovernmental, some public-private partner-
ships) to advance codes and to define institutional procedures to monitor compli-
ance. Increased pressure from labor and human rights groups has motivated a
growing number of multinational corporations to adopt codes of conduct and to
submit to some form of external monitoring.1
6Policy Studies Journal, 31:1

Codes related to labor standards were originally quite diverse (OECD, 1999;
Varley, 1998; Diller, 1999; Compa & Hinchliffe-Darricarrere, 1995) but now
appear to be converging around the ILO core standards and basic principles
regarding the protection of health and safety, wages and hours, and treatment of
women.2Although the general range of issues addressed in these systems is fairly
similar (van Tulder & Kolk, 2001), the details of codes can vary considerably.
Table 2 presents a summary of the codes of conduct advanced by the four primary
U.S. monitoring systems. Key debates continue around issues such as freedom of
association, wages (minimum vs. prevailing vs. “living”), and the scope of
“nondiscrimination” clauses.
Systems for implementing and evaluating code compliance are obviously
critical to the credibility of these codes. To these ends, a number of initiatives
have emerged over the last several years to foster the implementation, monitoring,
and verification of codes.
Firm Internal Compliance Monitoring
Many large brands and retailers have developed procedures for monitoring
supplier compliance with their newly created codes of conduct. The Gap, for
instance, has a vendor compliance department with over 100 staff responsible for
monitoring the implementation of the company’s code of conduct throughout its
global supply chain. Levi’s, Disney, Walmart, H&M, and other companies have
established similar programs. These systems can either be extensions of existing
supply chain management programs—simply adding labor, human rights, and
environmental concerns to current systems for evaluating quality, timeliness,
price, and so on—or they can involve entirely new systems for internal monitor-
ing and evaluation. Some companies are asking their quality control and pur-
chasing staff to take on code compliance as an additional task, but others are hiring
dedicated staff to conduct precertification audits of contractors and ongoing
assessments of code compliance.
Nike was one of the first companies in the apparel and footwear industries to
develop an internal compliance division. In 1992 Nike established a code of
conduct on labor and environmental practices for its network of suppliers (now
over 900 factories around the world employing over 650,000 workers). Supplier
compliance with the code is monitored through a program of internal evaluation
conducted first by Nike staff and then reviewed by external accounting, health
and safety, and environmental consulting firms. Nike has developed internal mon-
itoring tools, such as its SHAPE (Safety, Health, Attitude of Management, People
Investment, and Environment) audit and MESH (Management, Environment,
Safety, and Health) program, that allow the company to integrate the evaluation
of labor and environmental issues into broader management practices and train-
ing (Nike, 2002). MESH resembles the ISO 14000 management auditing program,
though it seeks to go further by evaluating actual factory performance. Nike now
O’Rourke: Outsourcing Regulation 7

8Policy Studies Journal, 31:1
Table 2. Variations in Codes of Conduct
Fair Labor Association Social Accountability Worldwide Responsible Workers Rights
(FLA) International (SA8000) Apparel Production Consortium (WRC)
www.fairlabor.org www.sa8000.org www.wrapapparel.org www.workersrights.org
Child labor Minimum Age: 15; or 14 if Minimum Age: 15; or 14 if Minimum Age: 14; or age Minimum Age: 15; or 14
country of manufacture meets developing country for completing schooling; if consistent with ILO
allows; or age for exemption; or local or minimum age established practices for developing
completing compulsory minimum age if higher. by law; whichever is greater. countries.
education.
Harassment and abuse No employees shall be No corporal punishment, No harassment, abuse or No employees shall be
subject to any physical, mental or physical coercion corporal punishment in any subject to any physical,
sexual, psychological or or verbal abuse. No sexually form. sexual, psychological, or
verbal harassment or abuse. coercive or exploitative verbal harassment or abuse.
behavior. No corporal punishment.
Nondiscrimination No discrimination in hiring, No discrimination in hiring, No discrimination on basis No discrimination in
salary, benefits, advancement, compensation, access to of personal characteristics employment, including
discipline, termination or training, promotion, or beliefs. Question about hiring, salary, benefits,
retirement, on basis of gender, termination or retirement discrimination based on advancement, discipline,
race religion, age, disability, based on race, caste, seniority. termination or retirement,
sexual orientation, nationality, national origin, religion, on the basis of gender,
political opinion, or social disability, gender, sexual race, religion, age,
or ethnic origin. orientation, union orientation, political opinion,
membership, or political or social or ethnic origin.
affiliation.

O’Rourke: Outsourcing Regulation 9
Freedom of association Where right restricted by law, Where right restricted by Lawful right of free No employee shall be
and collective bargaining employer shall not seek state law, employer facilitates association, including right subject to harassment,
assistance to prevent workers parallel means for free to join or not join association. intimidation or retaliation
exercising right to FoA. association and bargaining. in their efforts to freely
associate.
Health and safety Safe and healthy working Safe and healthy working Safe and healthy working Safe and healthy working
environment required. environment required. If environment required. If environment required.
Standard also applies to provided, housing should provided, housing should be
employer operated facilities be clean and safe. Steps safe and healthy.
part from production taken to prevent accidents
facilities (e.g. housing). and injury. Regular health
and safety training.
Wages Local minimum wage or Legal or prevailing industry Legal minimum wage. Legal minimum wage and
prevailing industry wage, wage, and meet basic needs/ benefits. WRC code
whichever is higher, and provide discretionary income. requires paying a “living
legally mandated benefits. wage.”
Hours of work 48 hours per week and 12 48 hours per week and 12 Shall not exceed the legal Not be required to work
hours overtime or the limits hours overtime maximum. limitations of the countries more than the lesser of (a)
on regular and overtime Personnel shall be provided in which apparel is produced. 48 hours per week or (b)
hours allowed by the law of with at least one day off in One day off in every seven- the limits on regular hours
the country. One day off in every seven-day period. All day period, except as required allowed by the law of the
every seven. overtime work shall be to meet urgent business needs. country of manufacture,
reimbursed at a premium and be entitled to at least
rate. one day off in every
seven day period, as well
as holidays and vacations.
Sources: Organizational websites and Maquila Solidarity Network (2001b).

has over 80 staff members who monitor labor and environmental conditions in
the company’s contractor factories. Reebok and adidas, Nike’s main competitors,
along with many other prominent footwear and apparel firms, have established
similar programs that combine in-house assessment with audits by consulting
firms. Reebok, for instance, has instituted a worldwide “Human Rights Produc-
tion Standards Factory Performance Assessment” system, and adidas has created
a “Standards of Engagement” for labor practices and health, safety, and the envi-
ronment for all its subcontractors (interviews conducted with staff of Nike,
Reebok, and adidas, 2000, 2001).
Through these auditing tools, companies such as Nike, Reebok, and adidas
now regularly rate their subcontractors for environmental and labor performance.
In the case of Nike, points are assigned for performance in a wide range of cate-
gories, with double weight given to labor and environmental performance rank-
ings. Subcontractors are then told how they rate against other subcontractors in
the same country. High scorers often garner more lucrative orders, whereas low
scorers risk losing contracts. Nike bases these labor and environmental programs
on long-standing quality control management systems for evaluating and ranking
subcontractors. Requirements to improve labor conditions simply extend the
scope of commitments agreed to in the code of conduct and subcontractor mem-
orandum of understanding. Providing some evidence that this effort is earnest,
Nike has cancelled a handful of contracts because of poor performance and an
unwillingness of factories to change.
It is hard to determine how much improvement firm-led codes of conduct and
monitoring programs have achieved. Little research exists on the impacts of codes
and self-monitoring on actual workplace conditions. Firms naturally assert that
these systems respond effectively and sufficiently to labor concerns. Many com-
panies continue to argue that they alone (perhaps with the assistance of a con-
sulting firm) have the knowledge and ability to solve labor problems. However,
judging by press reports, neither activists nor the general public put much cre-
dence in corporate self-evaluation and monitoring (Connor, 2001a). Based on
recent cases in which codes and monitoring have been used for public relations
rather than for improving labor conditions, many stakeholders criticize voluntary
codes and internal monitoring for their vulnerability to corporate manipulation
(O’Rourke, 2002).
External Monitoring and Certification
Growing public awareness and further activist pressure have led to a recent
profusion of programs in the United States and Europe to establish standardized
codes of conduct and systems of monitoring that are conducted by accredited
third-party auditors. Three major initiatives of this type have emerged in the
United States: the Fair Labor Association (FLA), Social Accountability Interna-
tional (SAI), and the Worldwide Responsible Apparel Production (WRAP) certi-
10 Policy Studies Journal, 31:1

fication program. Each of these programs has a code of conduct informed largely
by ILO core standards and a system in place for accrediting external organiza-
tions to monitor compliance with their codes. A small army of monitors, includ-
ing large accounting firms, professional service firms, and small nonprofit
organizations are emerging to provide these third-party monitoring services
(Bartley, 2001).3These external monitoring systems differ in key procedures for
auditing (who conducts the monitoring and how), certification (whether a factory
or a brand is certified), and reporting (what is publicly disclosed). Table 3 high-
lights differences in these systems.
Fair Labor Association (FLA)
The Fair Labor Association (FLA), convened originally by the Clinton admin-
istration in 1996 as the Apparel Industry Partnership (AIP), is both the oldest and
most controversial of current initiatives to establish monitoring and verification.
The FLA originally focused only on the apparel and footwear industries but has
recently expanded to cover other industries producing university-logo goods. As
of November 2002, the FLA consisted of 13 footwear and apparel firms, several
nongovernmental organizations, and about 170 university affiliates.4
The FLA has developed a “Workplace Code of Conduct and Principles for
Monitoring,” accredits monitors, reviews audits, and reports on audit results. The
FLA advances a monitoring system that requires companies seeking “certifica-
tion” to first inspect (internally) at least half of their factories during the first year
and all of their factories during the second year. Companies are then required to
hire external monitors to evaluate at least 30% of their supplier factories during
the first 2–3 years of the certification process, and 5% annually thereafter.5Over
2000 internal audits have been conducted, and the FLA expects to complete 300
external audits by the end of 2002.
The FLA model has come under fire from a number of unions, NGOs, and
student activists for being overly controlled by industry (Benjamin, 1998; Maquila
Solidarity Network [MSN], 2001a). Critics pointed out that firms could select and
directly pay their own monitors, have a say in which factories were audited, and
only disclose summaries of auditing results. Student activists have also criticized
the FLA for failing to advance a living wage and for not supporting union and
women’s rights sufficiently.
The board of the FLA responded to these criticisms in April 2002 by announc-
ing sweeping changes in the program’s external monitoring and disclosure pro-
cedures (Fair Labor Association [FLA], 2002). The FLAis now taking much more
control over external monitoring, with the FLA staff selecting factories for eval-
uation (from a sample of “high risk” firms), choosing the monitoring organiza-
tion, requiring that inspections be unannounced, and receiving all audit reports
directly. The FLA will then work with the brand and factory in a remediation
process and will publicly disclose summaries of the results of the audits and
O’Rourke: Outsourcing Regulation 11

12 Policy Studies Journal, 31:1
Table 3. U.S. Monitoring and Certification Systems
Fair Labor Association Social Accountability Worldwide Responsible Apparel
(FLA) International (SA8000) Production (WRAP)
www.fairlabor.org www.sa8000.org www.wrapapparel.org
Scope Apparel and footwear companies. Factories producing a wide range Apparel industry.
Licensees of affiliated universities. of products.
Governance 12-member board with 6 industry Governing board has 5 members, Board of 3 officers and 8 directors
reps, 5 NGOs, 1 university rep. composed of 1 rep from CEP form the Independent Certification
(Council on Economic Priorities), Board. Primarily industry reps.
3 lawyers, and 1 businessperson.
SAI also has an advisory board
with more diverse membership.
Monitoring process Companies must conduct internal Manufacturers or suppliers are Factories must undergo a three-step
monitoring of at least one-half of granted the status of “applicants” process: Self Assessment,
their applicable facilities during for one year until they are verified Independent Monitoring, and Final
the first year, and all of their by an accredited Certification Review and Follow-up.
facilities during the second year. Auditor. The SA 8000 Certificate
must be renewed every 2 years. Factories contract and schedule
selected Independent Monitors to
Companies commit to use independent Specially trained local audit teams perform onsite evaluations. Based
external monitors accredited will be briefed by local NGOs and on this evaluation, the Independent
and selected by the FLA to conduct unions, speak to managers and Monitor will either recommend that
periodic inspections of at least 30% workers and check the records of the facility be certified or identify
of their facilities during their initial 3 the factories. The SA 8000 areas where corrective action must be
year participation period. “guidance document” is the SAI taken before such a recommendation
manual which assists the accredited can be made.
Factories are selected by FLA staff, auditors in fulfilling this task. NGOs
with a focus on the largest and those are also encouraged to undergo the Factories self-select for
with greatest risk of non-compliance. process of becoming an accredited certification.
All monitoring must involve consulting SAI auditor.
local NGOs. Monitors will use a
combination of announced and Factories self-select for certification.
unannounced visits.

O’Rourke: Outsourcing Regulation 13
Certification FLA certifies an entire brand. A Certification means that a facility The WRAP Certification Board will
service mark will be developed so has been examined in accordance review the documentation of
consumers know which companies with SAI auditing procedures and compliance and decide upon
are participating and which have found to be in conformance. certification. The term of certification
met the standards for certification. Auditors look for evidence of will be specified by the Board, based
Timely remediation, assessed by effective management systems on a criteria of risk factors. Over the
monitors and FLA staff, is required and performance that prove term of the certification, the facility
for certification. compliance. Certified facilities are may or may not receive an
subject to semi-annual unannounced inspection to verify
surveillance audits. continued compliance.
Company requirements Companies must implement the FLA Manufacturers/suppliers adopt a Factories must meet WRAP principles
Code; internally monitor every factory program to pursue SA8000 and bear all costs of certification.
every year according to FLA certification. Retailers become Factories must apply, be registered in
monitoring principles; and participate “SA8000 Members” and publicly the WRAP Certification Program, and
in independent external monitoring announce their commitment to seek perform self-assessment of its facility
every year. All internal and external out socially responsible suppliers with the WRAP Handbook to
monitoring must include local NGOs. and assist suppliers in meeting the determine if their facility complies
SA8000 social standards. with the WRAP Principles.
Reporting All internal and external monitoring Audit reports go to the companies Audit reports are provided to factories
reports will be provided in full to the and to SAI. Other parties can only and the WRAP board.
FLA staff. The FLA will evaluate receive them after having signed a
audits, jointly develop remediation confidentiality agreement with the
plans, and then publish summary company management and the audit
reports of audit remediation results. company.
Public disclosure Annual reports on each company based The public is informed No public reporting. No mention of
on internal and external monitoring. only of factories granted certification. sites that receive, fail, or lose
Participating companies are publicly certification.
listed on website. No disclosure of
locations of certified factories.
Sources: Organizational web sites and Maquila Solidarity Network (2001b).

remediation efforts. FLA staff conduct their own follow-up inspections to verify
that remediation has occurred. The FLA is also expanding its external complaint
procedures and internal management systems reviews of brands.
Social Accountability International (SA8000)
SA8000, a voluntary workplace standard patterned on the International Orga-
nization of Standards system (e.g., ISO 9000 and ISO 14000), was created in 1997
by the Council on Economic Priorities (a U.S. NGO). The SA8000 standard is
administered by Social Accountability International (SAI), with an advisory board
made up of representatives from multinational firms, international unions, and
NGOs. SAI seeks to motivate factories as well as member brands in a wide range
of industries to implement the SA8000 code of conduct and to be audited by
accredited auditors. SAI is responsible for accrediting these auditing firms; for
conducting trainings for auditors, factory managers, and workers on the standards;
and for publishing a list of factories meeting the SA8000 standard.
The SA8000 system differs from the FLA in a number of regards, most
notably on the issues of wages, worker representation, and certification. SA8000
may be interpreted to include the requirement that factories pay workers a “living
wage,” or what SAI refers to as a “basic needs” wage, as opposed to the FLA’s
weaker requirements to pay the prevailing wage (see Table 2). SA8000 also
requires firms to “facilitate parallel means of independent and free association and
bargaining” in countries where it is not possible to form free trade unions (such
as China). Both of these provisions remain highly controversial, as it is not clear,
for instance, exactly what would qualify as effective parallel means of represen-
tation in countries such as China. SA8000 also includes a section on management
systems that requires policies and procedures to ensure ongoing compliance with
the standard.
SA8000 also differs in that it certifies manufacturing facilities, not brands or
retailers. The idea behind this system is that brands and merchandisers will seek
out factories that have received SA8000 certification, as they look to ISO 9000
certification to verify quality standards. SAI is also developing a Signatory
Member program,6which requires members to move their supplier factories
toward SA8000 compliance and to periodically report progress in meeting these
plans.
SAI discloses lists of certified facilities and their locations but does not pub-
licly disclose which facilities have lost their certification or were rejected in their
applications. One hundred fifty factories in 27 countries had been certified under
SA8000 as of September 2002. Of these, 28 were apparel and textile firms. It is
not clear whether any of the Signatory Members’ affiliated facilities have received
SA8000 certification.
Anumber of concerns have been raised about the SAI strategy. Critics have
argued that it is impossible to “certify” that any factory is in compliance with the
14 Policy Studies Journal, 31:1

SA8000 standard based on a 1-day audit, once per year. Others point out the limi-
tations of a voluntary, factory-centered initiative that has to date certified only 150
factories out of at least 100,000 factories producing for the U.S. market. The
SA8000 auditing procedures have also been criticized by NGOs for a perceived
corporate bias and weak controls on the quality of monitors (Labor Rights in China
[LARIC], 1999). No NGOs have been accredited as auditors within the SAI system.
Auditing is conducted by professional service firms and quality testing firms.
Worldwide Responsible Apparel Production (WRAP)
The Worldwide Responsible Apparel Production (WRAP) program might be
viewed as industry’s version of external monitoring and certification. WRAP was
developed in 1998 by the American Apparel Manufacturers Association (which
recently became the American Apparel and Footwear Association [AAFA]) and
began certifying factories in June 2000. WRAP’s board members include major
apparel brands, such as Vanity Fair Corporation, Sara Lee, Kellwood, and Gerber
Childrenwear. The WRAP Certification Board consists of individuals primarily
from the private sector, with a nominal claim that the majority of its members are
not directly affiliated with the apparel industry (MSN, 2001a).
Like the FLA and SA8000, WRAP has created its own code of conduct that
it calls the “WRAP Principles.” The twelve WRAP Principles include common
standards for child and forced labor and workplace and environmental protections.
However, WRAP also contains unique requirements for customs compliance and
drug interdiction efforts that support tighter security controls over suppliers and
shipments. The WRAP Principles are widely viewed as the weakest standards
of any of these systems and the least transparent monitoring and certification
program (MSN, 2002).
WRAP certifies factories not brands, similar to SA8000. The WRAP Certifi-
cation Board accredits firms to be external monitors of manufacturing facilities.
WRAP has accredited 10 firms as monitors to date (MSN, 2002), again primarily
professional service firms, such as ITS, Global Social Compliance (formerly
PwC), BVQI, and Cal-Safety. Certification of facilities is valid for 1 year, and
factories are required to undergo self-assessment and to submit to external mon-
itoring. External monitors submit Facility Monitoring Reports to the WRAP
Certification Board, which then reviews each report and makes the decision for
or against certification. Upon certification, facilities may or may not be subject to
unannounced inspections. As of April 2002, approximately 200 factories had been
certified by WRAP to meet its code (personal communication with L. Doherty,
WRAP).
WRAP has been criticized by a range of stakeholders for its perceived indus-
try bias and low level of public transparency. WRAP does not disclose the names
or locations of certified or audited factories, and it has not disclosed any audits
to date. WRAP also lacks any NGO or civil society participation in monitoring
O’Rourke: Outsourcing Regulation 15O’Rourke: Outsourcing Regulation 15

or verification, or a system for workers or NGOs to register complaints regarding
certified factories. All audits are preannounced and conducted by firms paid
directly by the factories being audited.
European Monitoring Initiatives
Anumber of monitoring initiatives are also emerging across Europe. The
Ethical Trading Initiative (ETI) was initiated in England in 1998. ETI is an
alliance of companies, NGOs, and trade unions working to “identify and promote
good practice in the implementation of codes of conduct of labour practice, includ-
ing the monitoring and independent verification of the observance of code provi-
sions.” (ETI, 2001) ETI was established explicitly as an experimental, learning
initiative, designed to help identify and disseminate information on how best to
implement labor codes that benefit workers in global supply chains. ETI has con-
ducted pilot projects in a number of countries, including apparel factories in Sri
Lanka. These pilots are key to the ETI strategy, as they generate information on
issues such as how to monitor for child labor, how to evaluate the quality of 1-
day audits, how different actors can contribute to auditing, how best to establish
worker complaint systems, and so on. ETI reports the findings of its pilots and
company internal auditing to member organizations but not to the general public.
The Dutch Clean Clothes Campaign established the Fair Wear Foundation
(FWF) in 1999 (after 5 years of discussions and negotiations on code issues) to
work with associations of small- and medium-sized enterprises to oversee the
implementation of a standard code of conduct. FWF requires company monitor-
ing of their supply chains, with independent verification and effective worker
complaint procedures. The foundation plans to certify companies that have a
system in place to gather evidence on factory conditions in their supply chains.
FWF is responsible for verifying that the code is being implemented in a per-
centage of each firm’s factories. FWF has conducted pilot studies in India, Poland,
Romania, and Indonesia to test its monitoring and verification procedures.
Independent Investigations and Verification
Amajor critique of many of these external monitoring systems has been that
auditors in these programs are paid directly by the brands or factories being mon-
itored. This potential conflict of interest raises concerns that monitoring is not
“truly independent” and thus may miss or cover up problems in factories. As men-
tioned, the FLA board has responded to these critiques by approving a change in
the process for hiring monitors so that firms will not be allowed to directly pay
monitors. Nonetheless, an additional layer of nongovernmental regulation has
emerged to respond to this credibility concern. This independent monitoring and
verification involves NGOs, unions, and private individuals monitoring firm per-
formance without direct payment from those with a stake in auditing results. There
16 Policy Studies Journal, 31:1

are several institutionalized versions of independent monitoring in the United
States and Europe.
Worker Rights Consortium (WRC)
The Worker Rights Consortium (WRC) was developed in 1999 by the United
Students Against Sweatshops (USAS) in cooperation with the Union of Needle-
trades, Industrial and Textile Employees (UNITE); the American Federation of
Labor-Congress of Industrial Organizations (AFL-CIO); and a number of human
rights, labor, and religious NGOs. The WRC had 108 college and university
members as of November 2002 and focuses primarily on factories producing
apparel with university logos.
The WRC employs three broad strategies: (1) inspections of factories from
which the WRC has received worker complaints; (2) proactive inspections in
countries with patterns of poor practice and worker organizing efforts; and (3)
information disclosure requirements. The WRC does not certify company com-
pliance with a code of conduct, conduct systematic monitoring, or accredit
monitors. Instead, the WRC encourages (but does not require) participating
universities to adopt codes of conduct that closely resemble the WRC’s model
code, which includes strong provisions for a living wage, women’s rights, and
recognition of worker’s rights to freedom of association. The WRC requires
member universities to commit to broad public disclosure and to develop mech-
anisms to verify information reported by companies and their suppliers.
The WRC’s goal is to ensure that factories that produce university-branded
apparel comply with a base code of conduct and, in particular, with rights to
freedom of association and collective bargaining. The WRC also seeks to educate
workers themselves about university codes so that workers may report code
violations to local NGOs or the WRC. The WRC aims to work collectively with
its university affiliates, the licensee corporations (the manufacturers), and local
NGOs to correct problems that have been identified. The WRC’s investigative
efforts rely on collaboration with local NGOs and activists; personnel from either
the WRC, its board, or affiliated university members; and labor and human rights
experts.
To date, the WRC has investigated conditions at four factories.7It makes all
of its factory investigation reports public. The WRC is also developing a data-
base of manufacturing facilities (www.workersrights.org/fdd.asp), which allows
anyone with Internet access to search by factory name, location, or university affil-
iate. The WRC is increasingly focusing on remediation processes, working with
universities and buyers (usually the brands) and workers’ organizations to nego-
tiate solutions to problems raised by workers, with the hope that there will be
some “ripple effect” to other factories in these regions.
The WRC is of course not without its detractors. It has been criticized (and
publicly opposed) by a number of firms and university administrators.8Opponents
O’Rourke: Outsourcing Regulation 17O’Rourke: Outsourcing Regulation 17

have accused the WRC of representing a “gotcha” model of monitoring, more
focused on identifying problems and embarrassing firms than on resolving prob-
lems (Nike refuses to comply, 2000). And the WRC’s in-depth inspection system
has been criticized for having a limited scope and coverage.
Related “Global” Initiatives
There are several other initiatives that while not explicitly focused on codes
and monitoring, are potentially supportive or complementary to nongovernmen-
tal regulation. The Global Reporting Initiative (GRI), for instance, is a project to
advance globally applicable guidelines for voluntary self-reporting of economic,
environmental, and social performance of firms. The GRI is working to set a
global standard for corporate reporting, creating a system analogous to financial
reporting procedures for environmental and social issues. This standard, and
broader efforts for corporate disclosure, could strengthen and help to standardize
existing codes and monitoring systems.
The United Nations (UN) is also advancing voluntary codes and reporting
systems through the Global Compact initiative, created by the UN General
Secretary in 2000. The Global Compact is in the words of the UN “not a regula-
tory instrument or code of conduct, but a value-based platform designed to
promote institutional learning. It utilizes the power of transparency and dialogue
to identify and disseminate good practices based on universal principles” (Global
Compact, 2001). Towards this end, the Global Compact asks companies to commit
to respecting nine principles, including respect for human rights, labor rights
(basically the ILO core standards), and the environment, and to report annually
on their progress on advancing these principles. The Global Compact and GRI
now work together, with GRI reports qualifying for Global Compact annual
reporting. It should be noted, however, that neither initiative requires external or
independent monitoring or verification of any kind. These initiatives remain
essentially self-reporting and disclosure systems.
Models of Outsourced Regulation
The diversity of current codes and monitoring systems has led to both con-
fusion and debate about the benefits and costs of nongovernmental regulatory
strategies. Versions of nongovernmental regulation range from individual facto-
ries paying to be certified, to multinational brands internally monitoring their
contractor factories, to multistakeholder initiatives accrediting third-party organ-
izations to inspect factories, to independent NGOs inspecting factories individu-
ally or in coordination with worker campaigns. In these different forms of
outsourced regulation, the substance, scope, implementation, participation, and
reporting of inspection results can vary considerably. And more importantly, these
18 Policy Studies Journal, 31:1

O’Rourke: Outsourcing Regulation 19
systems also vary in their underlying models for changing labor practices in global
supply chains.
Modern governance systems, of course, involve complex arrangements and
interactions between public and private actors. However, traditional labor regula-
tion, now often disparaged as “command-and-control” policy, might be thought
of as largely a state-centric, local “policing” model of governance. Clear, fixed
rules are established, and government inspectors police compliance with these
standards to advance a kind of “interventionist” regulation (Knill & Lehmkuhl,
2002).
Nongovernmental labor regulation represents a significantly different model
of regulation, involving multiple stakeholders participating in standard setting,
monitoring, and sanctions (or incentives). And as we have seen, among non-
governmental regulatory initiatives there are different underlying models. The
FLA, SAI, and WRAP are all firmly centered around enlisting market drivers for
improved supplier performance. WRAP and SAI advance a system based largely
on factory certification. These initiatives certify that management systems are in
place to guarantee acceptable performance in individual factories. Certification
provides a stamp of approval that is designed to attract customers to self-selecting
factories. WRAP and SAI tap into the motivations of individual factories seeking
to connect into socially concerned (and presumably high-value) buyers, as facto-
ries are audited only if they ask (and pay) for the evaluation. These systems
involve an advanced a form of “privatized regulation” (Knill & Lehmkuhl, 2002;
Utting, 2002).
The FLA advances market pressures by creating a supply chain policing
system involving multiple stakeholders. This advances a kind of “collaborative
regulation” or “regulated self-regulation” (Ayres & Braithwaite, 1992; Teubner,
1983) that depends on top-level commitment to the code from a brand or retailer,
both internal and external monitoring of suppliers, and participation of NGOs in
providing legitimacy to the system. The FLA also provides information to buyers
that can be used to influence supply chains.
The WRC is building essentially a “fire alarm” model of regulation
(McCubbins & Schwartz, 1984) that focuses on creating new mechanisms of
accountability for both firms and government agencies by gathering information
from workers and local organizations and then helping them to organize to win
demands. Alarms are triggered through complaints from workers and local NGOs,
which motivate WRC inspections. Factories and the brands purchasing from them
are targeted through this bottom-up process. The WRC then puts pressure on
brands to improve conditions and, at the same time, works to facilitate worker
empowerment and organizing to negotiate improvements. Supporting freedom of
association and collective bargaining are primary goals of the WRC.
So these different emerging systems create a spectrum of new regulatory
processes: from purely “privatized” regulation (firm internal monitoring and

WRAP), to more “collaborative” regulation (SAI, FLA, and ETI), to more “social-
ized” regulation (FWF and WRC).
Evaluating Nongovernmental Regulation
There is, unfortunately, virtually no public data available to analyze how well
these systems of regulation are currently performing. Developing measures to
evaluate nongovernmental regulation remains a critical area for future research.
However, there is some evidence from programs in the United States and from
sporadic reports from monitoring initiatives around the world that can help us
begin to evaluate nongovernmental regulation.
For instance, more than 60 companies have signed agreements with the U.S.
Department of Labor to conduct factory monitoring, and hundreds of other com-
panies are conducting monitoring on their own. Private monitoring firms conducted
over 10,000 audits of garment shops in Los Angeles alone in 1998, which is about
10 times the number carried out by state and federal authorities (Esbenshade, 2000,
p. 5). This monitoring is generally unsystematic, with little oversight, no trans-
parency, no sanctions for poor monitoring, and potential conflicts of interest of
monitors. This would appear at first blush to be a failure of nongovernmental reg-
ulation. However, somewhat surprisingly, this third-party monitoring has actually
led to increased levels of compliance. As Esbenshade reports:
Monitoring has significantly raised the rate of compliance in the indus-
try. The rate of compliance rose 20% between 1994 and 1996, in part
due to the proliferation of monitoring. In 1998, DoL statistics indicate
that the rate of violations in non-monitored shops is twice as high as in
monitored ones. However, the data also demonstrate that while moni-
toring helps, it has far from solved industry’s problems. Fifty six percent
of monitored shops are still violating labor laws. (2000, p. 5)
Many of the initiatives described above are still too new to evaluate fully. The
FLA has only recently begun its external monitoring program. The FLA has,
however, established a complaint response system and is developing remediation
strategies when problems are identified. For instance, in response to a third-party
complaint submitted by Nike, the FLA sent a team to the Dominican Republic to
assess a labor dispute at the BJ&B cap factory in January 2002. Workers alleged
freedom of association violations when 20 workers were fired from the facility
after having signed a petition to form a union. Within 24 hours of receiving the
complaint, FLA-participating company compliance staffs were on location and
conducted unannounced factory visits; off-site interviews with workers; and a
review of facility payroll, personnel, and timecard records. Staff of the WRC also
participated in this inspection. The dispute was resolved within several weeks.
As mentioned, SAI has certified 150 factories to meet the SA8000 standard.
SAI staff assert that member companies have conducted over 2,000 audits of
20 Policy Studies Journal, 31:1

supplier factories in preparation for these certifications. SAI has also conducted
a number of auditor, supplier, and worker trainings around the world. To date,
over 1,100 people have taken SAI’s supplier training course in 19 different coun-
tries. SAI is also working with the International Textile, Garment, and Leather
Workers Federation to support workers in 12 countries across Latin America,
Africa, and Asia on increasing worker involvement in workplace standards and
monitoring systems.
One of the problems of evaluating these systems is their limited transparency.
The FLA and SAI currently do not publish data on the factories audited, findings
of audits, or changes over time in performance. WRAP has yet to publicly dis-
close the number of factories it has certified, improvements it has made in factory
operations, or even how the organization is working to influence suppliers and
buyers.
The WRC, on the other hand, makes all of its reports public. The first—of
the Kukdong garment factory (now known as Mexmode) in Puebla, Mexico—is
touted as a major success, as the process has supported improvements in factory
conditions and pay (workers have won a series of pay raises since the WRC inves-
tigation), the formation of an independent union, and the signing of a collective
bargaining agreement (one of the first in a Mexican Maquiladora factory). The
WRC has also recently completed an investigation in Indonesia that has led to the
remediation of existing problems and an agreement to recognize multiple unions
in the factory.
Limits to Nongovernmental Regulation
There are obviously a number of weaknesses and challenges to making these
different regulatory systems effective. Nongovernmental monitoring faces many
of the same mundane challenges as traditional government monitoring and
enforcement, including coverage, corruption, incentives of monitors, training and
capacity of inspectors, and so on. The long and mobile nature of apparel supply
chains, which has strained traditional regulation, makes nongovernmental moni-
toring extremely challenging. The Gap alone sources from 4,000 factories in 55
countries; Disney is estimated to source from over 30,000 factories and Walmart
from even more (Wach & Nadvi, 2000).
The ability of firms to move production quickly among factories and to hide
behind multiple layers of ownership makes systematic inspections extremely
difficult.9Anumber of critics have raised concerns that nongovernmental mon-
itoring involves visits to factories that are too infrequent to evaluate normal day-
to-day operations. “Parachuting” monitors are able to identify the most obvious
problems but may miss many of the largest issues and are not around long enough
to actually solve problems (O’Rourke, 2002). Critics surmise, quite reasonably,
that NGOs will not be able to duplicate national labor inspectorates, as they cannot
cover the full extent of factories (Justice, 2001).
O’Rourke: Outsourcing Regulation 21

There are also concerns about how many workers these systems can actually
reach. Many markets—such as informal sector production, business-to-business
commodities, and production for domestic consumption in developing coun-
tries—lack ethically inclined consumers. As Pearson and Seyfang (2002) have
argued, voluntary codes and monitoring primarily influence “enclaves” in the
global economy rather than applying universally. Nongovernmental regulatory
systems focus on workers in first-tier suppliers and often large-scale factories.
They rarely reach down to informal-sector or home-based workers (Lee, 1997).
Codes and monitoring also entail fairly complex technical and social issues.
Consulting and accounting firms and NGOs will need extensive training to ade-
quately play the role of industrial hygienists or experts on wages and hours.
Counting on private actors or NGOs to provide these skills assumes that they, or
the firms being monitored, will assume the cost of training and conducting
rigorous inspections.
Even if nongovernmental regulation can overcome these critical implemen-
tation challenges, a number of deeper concerns remain. Critics fear that non-
governmental regulation will provide public relations cover to brands, may
confuse consumers with a proliferation of labels and certifications, and may have
unintended negative impacts on workers in developing countries (Liubicic, 1998).
Another critical concern is that nongovernmental regulation may be helping to
privatize governmental regulation. Some critics warn that companies are control-
ling these processes, co-opting NGOs by changing them from watchdogs to “part-
ners” and undermining strong local laws and unions (Justice, 2001, p. 6). Having
NGOs play the role of regulators may also ultimately undermine traditional reg-
ulatory processes (Nadvi & Wältring, 2001; International Labour Organization
[ILO], 1998). Others fear that elected governments are actually ceding some of
their sovereignty to consumers through these systems. Clearly the shift to non-
governmental regulation focuses more attention on consumers (rather than on the
state or unions) as the key constituent of monitoring and enforcement.
Some critics also argue that monitoring, when it is conducted by local NGOs,
can impede unionization or “crowd out” the efforts of local workers organiza-
tions. Compa (2001) discusses several cases in Central America in which NGOs
appear to be “supplanting the unions’ role as worker representatives by discussing
wages and working conditions with factory managers,” a process that will actu-
ally help “powerful companies to avoid union organizing, enforceable collective
agreements, and government regulation.” Others on the ground in Central America
disagree with this assessment, arguing conversely that NGO monitoring has sup-
ported several union campaigns in El Salvador and Guatemala (Quinteros, 2001).
The Kukdong case in Mexico highlights the potential of a coordinated union-
NGO, north-south strategy. Cooperation among local workers, the AFL-CIO,
NGOs, student activists, and the WRC and FLA supported the formation of an
independent union, the signing of a new contract with Korean management, and
important gains in pay and health and safety conditions (Brown, 2001). The spot-
22 Policy Studies Journal, 31:1

light of monitoring was useful in this case in protecting fledgling union organiz-
ing. Nonetheless, the debate in Central America underscores the reality that NGOs
and unions continue to be “wary allies” and need to develop better means to work
together to advance nongovernmental regulatory mechanisms (Compa, 2001).
Anumber of critics have also noted that codes and monitoring can hurt
workers (Esbenshade, 2001; Liubicic, 1998). Monitoring reports can lead firms
to cut contracts with poor performing factories, leading to job losses. Firms may
reduce overtime at a factory working beyond a code of conduct’s limit, despite
workers needing these wages to survive. Workers may also be punished after com-
plaining to auditors, as these systems often have limited protections for workers
who complain. Even when monitoring is effective, some of the most hazardous
jobs may be shifted further down the supply chain or into the informal sector to
avoid the selective gaze of nongovernmental regulation.
There are also many problematic versions of nongovernmental regulation.
For example, Global Social Compliance (formerly PricewaterhouseCoopers), a
monitor for many large multinationals, depends largely on data provided by man-
agement and conducts very cursory inspections of factories. Worker interviews
are conducted inside the factories. Factory managers know who is being inter-
viewed, for how long, and on what issues (O’Rourke, 2002). This kind of moni-
toring can divert attention from the real issues in a factory, provide a false
impression of performance, certify that a company is “sweat-free” based on very
limited evidence, and actually disempower the workers it is meant to help.
Although there is no single perfect way to monitor a factory, there are clearly
better and worse monitoring practices.
Conclusions
New nongovernmental regulatory systems hold out both potential and peril.
They offer the potential of opening up and strengthening regulatory systems and
bringing in new voices and mechanisms for motivating improvements in global
supply chains. They also harbor the peril of privatizing regulation, effectively
closing off democratic forms of regulation and bypassing local governance.
Questions thus remain on how to move these systems toward more credible
and complete global regulation. Can these systems be implemented beyond the
first tier of suppliers? Can improved practices “spill over” into firms not directly
tied to high-end global supply chains? Can southern stakeholders be brought into
discussions and have a real say in the structure and implementation of these pro-
grams? Can mechanisms of representation and democracy be formalized in these
nongovernmental systems? Can these systems provide workers and their advo-
cates real tools that will increase their space for organizing? Is it possible to move
towards interoperable systems of standards and monitoring? Can the ILO be
brought more fully into nongovernmental regulation? And ultimately, can these
new forms of regulation be designed to complement and support existing regula-
O’Rourke: Outsourcing Regulation 23

tory processes and to directly benefit workers and poor communities around the
world?
In some regards, the distinctions among these systems are beginning to break
down. There is some convergence underway in codes and monitoring regimes that
is blurring the categories presented in this article. Factory monitoring sometimes
includes union officials. Supply chain monitoring is employing NGOs to monitor
factories. And NGO investigations are sometimes coordinated with powerful
brands. Nonetheless, there are still critical distinctions among these initiatives on
issues such as the roles of workers and advocacy organizations, transparency of
results, and strategies for remediation of problems.
And there is certainly no guarantee that voluntary codes of conduct and mon-
itoring schemes will naturally converge on more complete or democratic systems
of regulation. They are just as likely to diverge into a plethora of initiatives com-
peting for the hearts and minds of consumers, serving to only confuse the public
and undermine the credibility of nongovernmental initiatives. However, with
strategic policies and coordinated efforts, nongovernmental regulation could
instead move towards more credible, transparent, and accountable systems. Orga-
nizations implementing monitoring systems should commit, at a minimum, to
making public their factory audits and auditing methodologies.
Another potentially promising avenue forward would involve efforts to build
complementarity and interoperability among these systems. The different models
of nongovernmental regulation are effective at different aspects of regulation.
Factory monitoring identifies willing factories and gives managers information to
support change. Supply-chain monitoring helps move standards down outsourced
chains of production and provides brands with information to better manage their
suppliers. Independent investigations help to expose the worst actors, provide
information to workers, and create incentives for brands to prevent problems in
their contractors. Connecting these initiatives in some interoperable way might
help to overcome the challenges of access, scope, and credibility.
Already, several of these initiatives have worked together to resolve specific
factory complaints. However, much more could be done to expand the comple-
mentarity and interoperability of these initiatives. As Fung, O’Rourke, and Sabel
(2001) have argued, it is possible to build on the core dynamics of nongovern-
mental regulatory systems by first expanding the transparency of monitoring
methods and results and then creating mechanisms to compare and benchmark
the performance of manufacturing facilities and monitors. Increased transparency
and comparison would allow stakeholders to evaluate individual factories—
comparing an FLA factory to a WRAP factory to an SAI factory—and then to
compare monitors and monitoring methods to evaluate the effectiveness and cred-
ibility of monitoring protocols. This would provide important information on best
practices in both monitoring and factory management; support a broad public dia-
logue on labor standards, monitoring, and enforcement; and build the credibility
of these nongovernmental initiatives.
24 Policy Studies Journal, 31:1

O’Rourke: Outsourcing Regulation 25
Each of the emerging nongovernmental regulatory systems has clear weak-
nesses and challenges. Nonetheless, under certain conditions, nongovernmental
regulation can influence factory labor practices. With increased transparency,
improved technical capacities, and new mechanisms of accountability to workers
and consumers, nongovernmental monitoring could complement existing state
regulatory systems. As they develop, new nongovernmental regulatory systems
should be evaluated along a number of criteria: (1) legitimacy—are key stake-
holders involved in all stages of standard setting, monitoring, and enforcement?;
(2) rigor—do codes of conduct meet or exceed ILO conventions and local laws,
are standards measurable, and is monitoring technically competent?; (3) account-
ability—is monitoring independent and transparent?; (4) complementarity—do
nongovernmental regulatory systems support state regulation and processes to
learn and improve standards and monitoring methods?
Regulation in the global economy remains a daunting challenge. If these
experiments in nongovernmental regulation can be made more transparent,
accountable, and democratic, it may be possible to build outsourced regulation
into an important response to the adverse impacts of globalization. At a minimum,
nongovernmental regulation offers a glimpse of emerging strategies to regulate
global supply chains and to begin the process of building new systems of gover-
nance over a fast-changing world.
Dara O’Rourke is an assistant professor in the Division of Society and Envi-
ronment in the Department of Environmental Science, Policy, and Management
at the University of California, Berkeley. Recent publications include Can We Put
an End to Sweatshops? (Beacon Press 2001, with Archon Fung and Chuck Sabel)
and Community-Driven Regulation: Balancing Development and the Environment
in Vietnam (MIT Press, forthcoming).
Acknowledgments
I want to acknowledge Sharon Chan for research assistance and Garrett Brown, Archon Fung, Bob
Jeffcott, Richard Locke, Katie Quan, Judith Tendler, and Lynda Yanz for comments on an earlier draft
of this article.
Notes
1. For comparisons of company codes, see van Tulder and Kolk (2001) or company web pages such
as www.nikebiz.com and www.gapinc.com.
2. As Nadvi and Wältring (2001, p. 34) note, “Despite the toothless nature of core labour standards,
they have become a model for private social standards.”
3. Private, for-profit monitors include PricewaterhouseCoopers (recently spun off as Global Social
Compliance), Cal-Safety Compliance Corporation (CSCC), International Certification Services
(SGS-ICS), Det Norske Veritas (DNV), Bureau Veritas Quality International (BVQI), Intertek
Testing Service (ITS), Merchandise Testing Labs (MTL), MFQ, Sandler & Travis (STR), Centro
per l’Innovzione e lo Sviluppo Economico (CISE), RWTUV (Rheinisch-Westfalischer Technischer

Uberwachungverein) Far East Thailand, Global Standards-Toan Tin Vietnam, and KPMG. Non-
profit groups include the U.S. NGO Verité, the Guatemalan Commission for the Monitoring of
Code of Conduct (COVERCO), the Independent Monitoring Group of El Salvador (GMIES),
Phulki (a Bangladeshi NGO), and the Honduran Independent Monitoring Team (EMI).
4. FLA members include Nike, Reebok, Liz Claiborne, Patagonia, Nordstrom’s, adidas, Eddie Bauer,
Polo Ralph Lauren, L. L. Bean, Nicole Miller, Phillips Van-Heusen, the Lawyers Committee for
Human Rights, the International Labor Rights Fund, the RFK Memorial Center for Human Rights,
and the National Consumers League. Notably, several union and NGO members of the original
AIP left the organization when it evolved into the FLA in protest of what they believed were insur-
mountable flaws in the organization and its monitoring procedures.
5. The FLA has to date accredited 14 organizations to carry out this “external” monitoring. Each of
these monitors is accredited to inspect factories in specific countries. These include A&L Group,
Cal Safety Compliance Corporation, Cotecna Inspections, COVERCO, Global Social Compliance
(formerly PwC), Global Standards/Toan Tin, Intertek Testing Services, Kenan Institute Asia, LIFT-
Standards, Merchandise Testing Labs Brand Integrity, Phulki, SGS, T-Group Solutions, and Verite.
As of June 2002, 982 companies had applied for certification, the majority of which were small
university licensees.
6. Member organizations include Amana, Avon Products, Cutter and Buck, Dole Food, Eileen Fisher,
Otto Versand, Toys R Us, and the UN Office of Project Services.
7. These include the Kukdong garment factory in Puebla, Mexico; the New Era baseball cap factory
in Buffalo, New York; the BJ&B cap factory in the Dominican Republic; and, the PT Dada apparel
and stuffed toy factory in Indonesia. These investigations have involved six to eight people for 5
to 6 days each.
8. Phil Knight, the CEO of Nike Inc., withdrew a planned $30 million donation to the University of
Oregon after the university joined the WRC.
9. As one retailer in the ETI commented, “I can know my supply chain at 9 a.m., then by 10 a.m. it’s
all different” (Ethical Trading Initiative, 2001).
References
Arthurs, H. (2001). Reinventing labor law for the global economy: The Benjamin Aaron lecture. Berke-
ley Journal of Employment and Labor Law, 22(2), 271–294.
Ayres, I., & Braithwaite, J. (1992). Responsive regulation: Transcending the deregulation debate.
Oxford, United Kingdom: Oxford University Press.
Bartley, T. (2001, August). The professionalization of scrutiny: The rise of labor-standards monitor-
ing organizations. Paper presented at conference of the American Sociological Association.
Anaheim, CA.
Benjamin, M. (1998). What’s fair about the Fair Labor Association (FLA)? Sweatshop Watch.
Retrieved from http://www.sweatshopwatch.org/swatch/headlines/1998/gex_fla.html
Bernstein, A. (2001, November 19). Do-it-yourself labor standards: While the WTO dickers, compa-
nies are writing the rules. Business Week, 74.
Block, R., Roberts, K., Ozeki, C., & Roomkin, M. (2001, April). Models of international labor stan-
dards. Industrial Relations, 40(2), 258–292.
Braithwaite, J., & Drahos, P. (2000). Global business regulation. Cambridge, United Kingdom:
Cambridge University Press.
Brown, G. (2001, Fall). Maquiladora Health and Safety Support Network Newsletter. Berkeley, CA.
Cashore, B. (2002). Legitimacy and the privatization of environmental governance: How non state
market-driven (NSMD) governance systems gain rule making authority. Governance, 15(4),
503–529.
Chan, A. (2001). China’s workers under assault—The exploitation of labor in a globalizing economy.
Armonk, NY: M.E. Sharpe.
26 Policy Studies Journal, 31:1

O’Rourke: Outsourcing Regulation 27
Compa, L. (2001, July 2). Wary allies. American Prospect, 12(12), pp. 8–9.
Compa, L., & Hinchliffe-Darricarrere, T. (1995). Enforcing international labor rights through corpo-
rate codes of conduct. Columbia Journal of Transnational Law, 33, 663.
Connor, T. (2001a). Still waiting for Nike to do it: Nike’s labor practices in the three years since CEO
Phil Knight’s speech to the National Press Club. San Francisco: Global Exchange. Retrieved
from http://www.globalexchange.org/economy/corporations/nike/
Conroy, M. (2001). Can advocacy-led certification systems transform global corporate practices?
Evidence and some theory (Working Paper Series No. 21). Amherst, MA: Political Economy
Research Institute, University of Massachusetts.
Cutler, C., Haufler, V., & Porter, T. (1999). Private authority in international politics. Albany: State
University of New York Press.
Diller, J. (1999). A social conscience in the global marketplace? Labour dimensions of codes of
conduct, social labeling, and investor initiatives. International Labour Review, 138(2), 99–130.
Douglas, W. (2001). Who’s who in codes of conduct. New Economy Information Service. Retrieved
from http://www.newecon.org/global/trade/CodesofConductDouglas-01-02-01.html#enforcing
Elliott, K., & Freeman, R. (2001, January). White hats or Don Quixotes? Human rights vigilantes in
the global economy (NBER (National Bureau of Economic Research) Working Paper No.
W8102). Cambridge, MA.
Environics International. (1999, May). The millenium poll on corporate social responsibility.
Retrieved from www.environicsinternational.com
Esbenshade, J. (2000, August). Globalization and resistance in the apparel industry: The struggle
over monitoring. Paper presented at the meeting of the American Sociological Association,
Washington, DC.
Esbenshade, J. (2001, Spring). The social accountability contract: Private monitoring from Los
Angeles to the global apparel industry. Labor Studies Journal, 26(1), 98–120.
Ethical Trading Initiative (ETI). (2001). Learning our trade [Annual Report 2000–01]. London:
Author.
Evans, P. (1997). The eclipse of the state? Reflections on stateness in an era of globalization. World
Politics, 50, 1.
Fair Labor Association (FLA). (2002, April 28). New changes to increase the transparency, inde-
pendence, and scope of the FLA [Press release].
Freeman, R. (1994). A hard-headed look at labour standards. In W. Sengenberger & D. Campbell
(Eds.), International Labour Standards and Economic Interdependence. Geneva: International
Labour Organization, 79–92.
Fung, A., O’Rourke, D., & Sabel, C. (2001). Can we put an end to sweatshops? Boston: Beacon Press.
Gereffi, G., Garcia-Johnson, R., & Sasser, E. (2001, July-August). The NGO-industrial complex.
Foreign Policy, 125, 56–65.
Global Compact (2001). Overview Retrieved from www.unglobal compact.org.
Haufler, V. (2001). Apublic role for the private sector: Industry self-regulation in a global economy.
Washington, DC: Carnegie Endowment for International Peace.
Herrnstadt, O. (2001). Voluntary corporate codes of conduct: What’s missing. Labor Lawyer, 16,
349–370.
Hughes, S., & Wilkinson, R. (1998). International labour standards and world trade: No role for the
World Trade Organization? New Political Economy, 3, 375–389.
International Labour Organization (ILO). (1998). Overview of global developments and office activ-
ities concerning codes of conduct, social labelling, and other private sector initiatives address-
ing labour issues. Geneva, Switzerland: Author.
Jeffcott, B., & Yanz, L. (1999, October 18). Voluntary codes of conduct: Do they strengthen or under-
mine government regulation and worker organizing. Maquila Solidarity Network. Retrieved from
www.maquilasolidarity.org/resources/codes/volcodes99.htm

Justice, D. (2001). The new codes of conduct and the social partners. International Confederation of
Free Trade Unions. Retrieved from http://www.icftu.org/
Knill, C., & Lehmkuhl, D. (2002). Private actors and the state: Internationalization and changing pat-
terns of governance. Governance, 15(1), 44–63.
Labor Rights in China (LARIC). (1999). No illusions: Against the global cosmetic SA8000. Hong
Kong: Author.
Lee, E. (1997). Globalization and labour standards: A review of issues. International Labour Review,
136(2), 173–189.
Lipschutz, R. (2000, December). Regulation for the rest of us? Global civil society, social regulation,
and national impacts. Paper prepared for workshop on Human Rights and Globalization,
University of California-Santa Cruz.
Liubicic, R. (1998). Corporate codes of conduct and product labeling schemes: The limits and possi-
bilities of promoting international labor rights through private initiatives. Law and Policy in
International Business, 30(1), 111–158.
Maquila Solidarity Network (MSN). (2001a, September). The changing terrain in the codes of conduct
debate. Retrieved from http://www.maquilasolidarity.org/
Maquila Solidarity Network (MSN). (2001b). Codes update (Issues no. 3–9). Retrieved from
http://www.maquilasolidarity.org/resources/codes/
Maquila Solidarity Network (MSN). (2002). Codes update (Issue no. 12)., Retrieved from
http://www.maquilasolidarity.org/resources/codes/memo12.htm
Maskus, K. (1997). Should core labor standards be imposed through international trade policy (World
Bank Development Research Group Policy Research Working Paper no. 1817), World Bank,
Washington, D.C.
McCubbins, M., & Schwartz, T. (1984). Congressional oversight overlooked: Police patrols versus
fire alarms. American Journal of Political Science, 28, 165–79.
Nadvi, K., & Kazmi, S. (2001). Global standards and local responses (Working Paper). Sussex, United
Kingdom: Institute for Development Studies.
Nadvi, K., & Wältring, F. (2001). Making sense of global standards (Draft IDS-INEP working paper).
Sussex, United Kingdom: Institute for Development Studies.
Nike (2002). Management of Environment, Safety, and Health (MESH). Retrieved from
www.nikebiz.com/labor/mesh.shtml.
Nike refuses to comply with WRC, cancels U. contract. (2000, April 3). Brown Daily Herald,
Retrieved from www.browndailyherald.com.
Organisation for Economic Cooperation and Development (OECD). (1999). Codes of corporate
conduct: An inventory. Paris: Trade Directorate.
O’Rourke, D. (2002). Monitoring the monitors: A critique of corporate third-party labor monitoring.
In R. Jenkins, R. Pearson, & G. Seyfang (Eds.), Corporate responsibility and labour eights:
Codes of conduct in the global economy. London: Earthscan, 196–208.
Pearson, R., & Seyfang, G. (2002). Codes of conduct as enclave social policy. In R. Jenkins, R.
Pearson, & G. Seyfang (Eds.), Corporate responsibility and labour eights: codes of conduct in
the global economy. London: Earthscan.
Quinteros, C. (2001, October). Independent monitoring seen from the Central American region. Paper
prepared for the SOMO Conference: From Codes to Compliance, Brussels, Belgium.
Reinicke, W. (1998). Global public policy: Governing without government? Washington, DC:
Brookings Institution Press.
Sabel, C. (1994). Learning by monitoring: The institutions of economic development. In N. Smelser
& R. Swedberg (Eds.), Handbook of Economic Sociology (pp. 137–165). Princeton, NJ:
Princeton-Sage.
Schmidt, V. (1995). The new world order incorporated: The rise of business and the decline of the
nation state. Daedalus,124(2), 75–106.
28 Policy Studies Journal, 31:1

Strange, S. (1996). The Retreat of the State. Cambridge, United Kingdom: Cambridge University
Press.
Teubner, G. (1983). Substantive and reflexive elements in modern law. Law and Society Review, 17,
239–285.
Utting, P. (2002). Regulating business via multistakeholder initiatives: A preliminary assessment, in
P. Utting (ed), Voluntary Approaches to Corporate Responsibility. Geneva, Switzerland: United
Nations Non-Governmental Liaison Service, 61–130.
van Tulder, R., & Kolk, A. (2001). Multinationality and Corporate Ethics: Codes of Conduct in the
Sporting Goods Industry. Journal of International Business Studies, 32(2), 267–283.
Varley, P. (1998). The Sweatshop Quandry: Corporate Responsibility on the Global Frontier,
Washington, D.C. & Investor Responsibility Research Center.
Wach, H., & Nadvi, K. (2000). Global Labour and Social Standards and their Implications for Devel-
oping Country Producers: A Bibliographic Overview (Working paper). Sussex: United Kingdom:
Institute for Development Studies.
Wick, I. (2001). Workers’Tool or PR Ploy? A Guide to Codes of International Labour Practice. Bonn,
Germany: Friedrich-Ebert-Stiftung.
O’Rourke: Outsourcing Regulation 29O’Rourke: Outsourcing Regulation 29