Content uploaded by Jennifer Bair
Author content
All content in this area was uploaded by Jennifer Bair on Aug 04, 2014
Content may be subject to copyright.
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
1
TOWARD JOINT LIABILITY IN GLOBAL SUPPLY
CHAINS: ADDRESSING THE ROOT CAUSES OF
LABOR VIOLATIONS IN INTERNATIONAL
SUBCONTRACTING NETWORKS
Mark Anner,† Jennifer Bair,†† and Jeremy Blasi†††
I. INTRODUCTION
On the morning of April 24, 2013, Rana Plaza, an eight-story building
in Bangladesh that housed five garment factories, collapsed, killing at least
1,129 workers.
1
The collapse was the most deadly disaster in the history of
the global garment industry and, indeed, one of the worst industrial
workplace disasters in history.
2
Media coverage of the factory collapse and its aftermath fueled a long-
standing campaign by a coalition of unions and NGOs to address what had
become, well before this latest tragedy, a crisis in building and fire safety in
Bangladesh.
3
Under intense pressure, within a month of the Rana Plaza
collapse, more than threedozen of the world’s largest apparel brands and
retailers, including H&M, Inditex (owner of the Zara brand), Tesco, and
† Associate Professor of Labor Studies and Political Science, School of Labor and Employment
Relations, Pennsylvania State University.
†† Associate Professor of Sociology, Department of Sociology, University of Colorado at
Boulder.
††† Law Clerk to the Hon. Judge Dean D. Pregerson, United States District Court for the Central
District of California; Former Director of Investigations, Worker Rights Consortium. The authors
would like to thank the Kheel Center for Labor-Management Documentation and Archives at Cornell
University for granting us access to the I.L.G.W.U archives; Ethan Erickson for his outstanding research
assistanceand ProfessorsMarty Wellsand KateBronfenbrennerofCornell’sSchoolofIndustrial and
Labor Relations for facilitating our research. The authors also wish to thank Tim Bartley, Gary Blasi,
Peter Evans, Jeff Hermanson, Nina Pillard, Andrew Schrank, Jodi Short, David Weil, and the
participants of the IBS Institutions working group at the University of Colorado at Boulder for their
helpful comments on earlier drafts of this Article. Author names are listed in alphabetical order.
1
. See, e.g., Julfikar Ali Manik & Jim Yardley, Building Collapse in Bangladesh Leaves Scores
Dead, N.Y. TIMES, Apr. 25, 2013, at A1; Jim Yardley, Grim Task of Identifying Factories’ Dead
Overwhelms Bangladeshi Lab, N.Y. TIMES, May 31, 2013, at A1.
2
. Ali & Yarkley, supra note 1; Yardley, supra note 1.
3
. For example, just several months before, on November 24, 2012, in a fire eerily reminiscent of
the one at NewYork City’s TriangleShirtwaist factory in March 1911, 112 workers lost their lives at
the Tazreen garment factory, which produced for Wal-Mart among other brands. See, e.g., Vikas Bajaj,
Fatal Fire in Bangladesh Highlights the Dangers Facing Garment Workers, N.Y. TIMES, Nov. 26, 2012,
at A6. Even prior to the Rana Plaza collapse, the death toll of garment workers killed in the workplace
since 2006 exceeded 500. Id.
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
2 COMP.LABORLAW&POL’YJOURNAL [Vol. 35:1
PVH (owner of Tommy Hilfiger and Calvin Klein brands), had signed a
sweeping agreement with two global union federations, UNI and
IndustriALL.
4
As we argue below, this agreement—the Accord on Building and Fire
Safety in Bangladesh—represents a new paradigm in the enforcement of
global labor and human rights. Under the currently dominant corporate
social responsibility (CSR) model, brands and retailers voluntarily adopt
codes of conduct and then monitor their suppliers for compliance. This
approach assumes that labor violations are a factory-level problem and that
the only entity that needs to be regulated is the contractor factory.
By contrast, the Bangladesh Accord—and the new model of corporate
accountability it represents—sees apparel brands and retailers at the top of
the supply chain as jointly responsible, along with contractors, for the
conditions below. The Accord imposes on these buyers substantive
obligations—most notably, financial obligations to help suppliers pay for
safety upgrades, such as the installation of fire exits, or, in the case of
structurally unsound buildings, major repairs while guaranteeing the
payment of workers’ salaries for time lost at work. Unlike the existing
model, in which brands and retailers unilaterally adopt standards or join
programs from which they can withdraw at any time, the brand and retailer
commitments in the Accord are not merely general statements of intent, but
binding, contractually enforceable duties.
Although the Bangladesh Accord represents a significant departure
from the extant model of labor compliance that has developed over the past
two decades, it is not unprecedented. Indeed, one of the reasons we believe
the approach holds so much promise is that it reflects core elements of what
is, to our knowledge, the most successful effort to systematically eradicate
sweatshop conditionsin any nation’s appareli ndustry. As we describe in
thesepages,thenegotiationofcollectivelybargainedcontracts,or“jobbers
agreements,”between workers, contractors, and lead firms, which ensured
fair prices and stable orders in the domestic apparel supply chain, was the
key force behind a dramatic decline in sweatshop conditions in the U.S.
apparel industry during the middle part of the twentieth century.
This Article revisits the history of jobbers agreements in order to
situate the Bangladesh Accord and other campaigns for “buyer
responsibility agreements” with global brands and retailers in a much
longer struggle against sweatshops. The downward pressure that pervasive
subcontracting networks put on garment workers’ wages and working
conditions is not a new problem unique to the era of economic
4
. See Steven Greenhouse, Major Retailers Join Bangladesh Safety Plan, N.Y. TIMES, May 13,
2013. A full list of signatories is available at INDUSTRIALL, Bangladesh, http://www.industriall-
union.org/tags/bangladesh (last visited Sept. 10, 2013).
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
2013] JOINT LIABILITY IN GLOBAL SUPPLY CHAINS 3
globalization; indeed, it as old as industrial apparel production itself. Yet
the scholarly debate about how to combat contemporary sweatshops has
been peculiarly ahistorical. We remedy the amnestic tone of this debate by
reflecting on how the sweatshop problem was effectively solved, albeit only
for a time, in the United States. Our case for the contemporary relevance of
jobbers agreements unfolds as a three-part answer to a straightforward
question: Why are working conditions and labor practices in the apparel
industry essentially unchanged despite the widespread implementation of
codes of conduct and compliance auditing regimes at the factory level, and
what alternative approaches might prove more effective in securing garment
worker rights in global supply chains?
First, as anticipated above, the principal root cause of sweatshop
conditions in international subcontracting networks is to be found in the
sourcing practices of the brands and retailers that coordinate these supply
chains. This emphasis on lead firms makes our approach to the sweatshop
problem distinct. Although other analysts gesture toward the role of buyer
practices as a cause of labor violations, their proposed remedies target
different problems: state failure (poorly enforced or inadequate public labor
regulations), managerial failure (lack of capacity, competence or motivation
at the factory level), or market failure (not exploiting the niche of ethical
consumers who care about labor standards). While all of these arguments
have some merit, they are ultimately unsatisfying because they do not
address the primary source of labor violations in apparel supply chains.
Labor violations are not simply a factory-level problem that can be
corrected by improved compliance monitoring; they are a pervasive and
predictable outcome in an industry dominated by lead firms whose business
model is predicated on outsourcing apparel production via highly flexible,
volatile, and cost-sensitive subcontracting networks. To substantiate this
claim, we explore the relationship between the price that global suppliers
are paid to produce apparel and the status of workers’ rights in these
apparel-exporting countries.
Our second argument follows from the first: If the sourcing strategies
of powerful buyers are the root cause of poor labor performance, then an
effective solution to this problem requires the regulation of buyer practices.
We develop this element by drawing out the parallels between
contemporary supply chains and the emergence of domestic subcontracting
networksinNewYork’sgarmentdistrictduringtheearlytwentiethcentury.
The trend among clothing companies to outsource much of their production
to independent contracting shops precipitated a dramatic deterioration in the
wagesandworkingconditionsofNewYork’sgarmentworkers. However,
this decline was arrested and then reversed by the development of a novel
collective bargaining model. The centerpiece of this model was the
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
4 COMP.LABORLAW&POL’YJOURNAL [Vol. 35:1
“jobbers agreement”—a contract negotiated by the International Ladies
Garment Workers Union (ILGWU) and companies called jobbers that—
much like today’s brands—designed, but did not manufacture, apparel.
Under the jobbers agreements, jobbers were held contractually “jointly
liable” forwages and workingconditions in theircontractors’ shops. The
upshot of this discussion is that there are lessons to be learned from the
history of the jobbers agreements about how to regulate the power of lead
firms vis-à-vistheirsuppliersintoday’sglobalsupplychains.
Of course, we recognize that the contemporary apparel industry is
vastly different from the one that ILGWU organizers confronted during the
first third of the twentieth century; their efforts to secure jobbers
agreements were facilitated by relatively high union density, at least in the
northeast, and a government willing to enforce labor laws—conditions
lacking in many of today’s major garment-exporting countries.
Consequently, the question is what elements of the jobbers agreement
model are replicable today and what would induce today’s jobbers—the
brands and retailers coordinating global supply chains—to enter into
similarly binding agreements?
The final argument we make is that alliances between workers,
national and international labor unions, and other actors (consumers,
student activists, etc.) can help bring about such agreements. We first
identify specific ways in which jobbers agreements regulated domestic
subcontracting relationships between buyers and suppliers, and then
consider the contemporary relevance of these principles for current global
supply chains. Specifically, we review a number of recent developments to
show that garment workers and their labor unions can secure some degree
of joint liability by enlisting worker allies, such as students and nonstudent
consumers, into campaigns to secure commitments from lead firms. The
enormous investments that global buyers have made in their brands, and the
potential sensitivity of these brands to negative publicity, represents an
Achilles heel for lead firms that is much more pronounced today than in the
era of jobbers agreements.
This Article proceeds as follows: Part II situates our analysis in the
current scholarly debate on the causes of and solutions for the global
sweatshop problem. Part III elaborates our argument on root causes,
showing that worsening labor rights compliance globally has coincided with
a substantial reduction in the prices apparel brands have paid their overseas
suppliers. Part IV recounts the history of the emergence and success of
jobbers agreements in addressing similar dynamics in the United States.
Part V reviews how the key principles and policies of the jobbers
agreements created joint liability between contractors and their clients. Part
VI returns to the present day to review a series of recent developments and
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
2013] JOINT LIABILITY IN GLOBAL SUPPLY CHAINS 5
campaigns which incorporate elements of jobbers agreements. We also
take a brief detour from our focus on contractual instruments to consider
statutory and common law approaches to joint liability in international
supply chains. We conclude with thoughts on the practicalities of the
approach we describe here and initiatives that are complementary to it.
Our analysis relies on a mix of primary and secondary data. To
explore the relationships between trends in apparel prices, import
penetration,andworkers’rightsandwages,we turn to primary, time-series
data from government sources and to the Cingranelli and Richards Human
Rights Dataset (CIRI). In reconstructing the history of jobbers agreements
in the United States we draw from the considerable secondary literature on
the U.S. apparel industry and the ILGWU, as well as materials in the
union’sextensivearchivehousedat Cornell University’s Kheel Centerfor
Labor-Management Documentation, including thirty-two original jobbers
agreements. Finally, our discussion of contemporary supply chains is
informed by more than a decade of academic research conducted by two of
the authors on the global apparel industry, particularly in Latin America.
5
Additionally, the third author has served as Director of Investigations for
the Worker Rights Consortium, a university-backed labor rights monitoring
organization that works to protect the rights of workers in global supply
chains. Collectively, this work has taken the three of us into well over one
hundred garment factories in more than ten countries, including Mexico,
Guatemala, El Salvador, Nicaragua, Honduras, Dominican Republic,
Kenya, Swaziland, Turkey, Thailand, Vietnam, and Bangladesh. It has
enabled us to speak with and learn from many stakeholders in the labor
rights arena, including industry actors, government officials, and workers,
labor unionists, and other civil society activists.
II. THE SWEATSHOP RESURGENCE: DEBATING CAUSES AND
PROPOSED SOLUTIONS
There is a growing consensus, at least among social scientists, that
codes of conduct and auditing programs have failed to eliminate, or perhaps
even substantially reduce, incidents of labor violations in global supply
chains.
6
Yet, there is considerable debate about why this is so and what can
5
. See, e.g., MARK ANNER, SOLIDARITY TRANSFORMED: LABOR’S RESPONSES TO
GLOBALIZATION AND CR ISIS IN LATIN AMERICA (2011); Jennifer Bair & Enrique Dussel Peters, Global
Commodity Chains and Endogenous Growth: Export Dynamism and Development in Mexico and
Honduras, 34 WORLD DEV. 203 (2006).
6
. See INSTITUTE FOR DEVELOPMENT STUDIES, THE ETI CODE OF LABOUR PRACTICE: DO
WORKERS REALLY BENEFIT? Part 2(a) (2006) [hereinafter ETI REPORT]; Mark Anner, Corporate
Social Responsibility and Freedom of Association Rights: The Precarious Quest for Legitimacy and
Control in Global Supply Chains, 40 POL. & SOC’Y 604 (2012); Richard M. Locke, Fei Qin & Alberto
Brause, Does Monitoring Improve Labor Standards: Lessons from Nike, 61 IND. & LAB. REL. REV. 3,
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
6 COMP.LABORLAW&POL’YJOURNAL [Vol. 35:1
be done about it. Richard Locke and his collaborators, for example, have
focused their attention on the failure of factory managers to utilize best
practices in managing workflow and production challenges, leading to
avoidable labor violations. These authors argue that when improved
compliance does occur, it often comes about because suppliers have been
“assistedand/oreducatedtocomplywithregulationsandstandardsbyhigh-
performingcomplianceofficersandauditors.”
7
Andrew Schrank and Michael Piore agree that a more collaborative
approach to compliance is needed, but in contrast to Locke et al., who call
for a more cooperative ethos between firms, they focus on the role of state
actors.
8
Specifically, they argue that labor inspectors, instead of acting like
policemen when violations are detected, are more effective when they work
with noncompliantfactoriesintherole of “advisor or consultant,” helping
actors reconcile compliance with competitive pressures.
9
InLocke’srecent
work, he argues that improved working conditions and better respect for
workers’ rights can best be achieved by a mix of private voluntary
regulation and public (national state) regulation.
10
Still others claim that the potential buying power of ethical consumers
has not been leveraged in the fight against sweatshops. Robinson, Meyer,
and Kimeldorf observe that “[m]any markets appear to be responding
poorly if at all to the ethical concerns of consumers” and that companies
could do a better job of targeting shoppers who value “sweat-free”apparel
alongside other product characteristics, such as quality style and price.
11
Studies of ethical consumption underscore the importance of educating
shoppers about their options and creating market contexts that encourage
them to make purchasing decisions that are consistent with their
commitments.
12
5–6, 20–21 (2007). Lockewrites,“Aftermorethan adecadeofconcertedefforts byglobalbrandsand
labor rights NGOs alike, private compliance programs appear largely unable to deliver on their promise
of sustained improvements in labor standards in the new centers of global production.” RICHARD
LOCKE, THE PROMISE AND LIMITS OF PRIVATE POWER: PROMOTING LABOR STANDARDS IN A GLOBAL
ECONOMY 20 (2013).
7
. Richard Locke, Matthew Amengual & Ashley Mangla, Virtue Out of Necessity? Compliance,
Commitment and the Improvement of Labor Conditions in Global Supply Chains, 37 POL. & SOC’Y 319
(2009).
8
. Michael Piore & Andrew Schrank, Toward Managed Flexibility: The Revival of Labour
Inspection in the Latin World, 147 INT’L LAB. REV. 1 (2008); Michael Piore & Andrew Schrank,
Trading Up: An Embryonic Model for Easing the Human Costs of Free Markets, 31 BOS. REV. 1 (2006)
[hereinafter Piore & Schrank, Trading Up].
9
. Piore & Schrank, Trading Up, supra note 8, at 3.
10
. See LOCKE, supra note 6.
11
. Ian Robinson, Rachel Meyer & Howard Kimeldorf, The Strength of Weak Commitments:
Market Contexts and Ethical Consumption, in WORKERS’ RIGHTS AND LABOR COMPLIANCE IN GLOBAL
SUPPLY CHAINS: IS A SOCIAL LABEL THE ANSWER? (Jennifer Bair, Marsha Dickson & Doug Miller
eds., 2013).
12
. Jans Hainmueller, Michael J. Hiscox & Sandra Sequeira, Consumer Demand for the Fair
Trade Label: Evidence from a Field Experiment (MIT Pol. Sci. Dep’t Research Paper No.2011-9B,
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
2013] JOINT LIABILITY IN GLOBAL SUPPLY CHAINS 7
A stronger state with better educated workplace inspectors, joint
problem solving between buyers and suppliers, and more informed
consumers can each play a role in eliminating labor violations in garment
factories worldwide. However, no approach can be truly effective unless it
targets the structural characteristics of the global apparel industry, with
particular attention to the buying practices of the brands that coordinate
overseas production. Indeed, what we find peculiar about most
contributions to the sweatshop debate is that although observers
acknowledge the central role of buyers in creating conditions that encourage
labor violations, they simultaneously propose solutions that leave this root
cause unaddressed. The implication is that while we can and should try to
change the behavior of factory managers, government officials, and
individual consumers, the business practices of a relatively small set of lead
firms are somehow beyond reach.
13
For example, Locke et al. recognize that supplier performance is
undermined by instability and volatility in orders from buyers. In fact, they
go so far as to argue that “upstream” (buyer) sourcing practices are often
therootcauseofproblems“downstream”(factorylevel),suchasexcessive
overtime, low wages, and the exploitation of migrant and contingent
workers.
14
Consequently, crafting a more collaborative approach to
compliance requires moving toward a “commitment model” in which
brands make longer-term commitments to suppliers, thus giving buyers and
contractors both an incentive to engage in joint problem solving and the
time to do so. The problem with this approach is that it does not explain
why brands would willingly forego their current model of footloose
sourcing in favor of longer-term commitments. Volatile supply chains may
inhibit factory compliance, but they benefit brands that are looking to
maximize flexibility and minimize cost.
Similarly, although Schrank and Piore frankly acknowledge that the
sourcingpracticesofleadfirms“areheavilyimplicatedinthedeterioration
of working conditions that has taken place, especially those associated with
the pace of work and with working time,”
15
their proposed treatment is
orthogonal to this diagnosis: They recommend giving labor inspectors the
2011), available at http://ssrn.com/abstract=1801942; Neeru Paharia & Rohit Deshpandé, Sweatshop
Labor Is Wrong Unless the Jeans Are Cute: Motivated Moral Disengagement (Harv. Bus. Sch.,
Working Paper 09-079, 2009), available at http://www.hbs.edu/faculty/Pages/item.aspx ?num=35032;
Dara O’Rourke, Citizen Consumer, BOS. REV. (Nov.–Dec. 2011), available at
http://www.bostonreview.net/BR36.6/ndf_dara_orourke_ethical_consumption.php.
13
. Cf. JILL ESBENSHADE, MONITORING SWEATSHOPS: WORKERS, CONSUMERS, AND THE GLOBAL
APPAREL INDUSTRY (1999); DAVID WEIL, IMPROVING WORKPLACE CONDITIONS THROUGH STRATEGIC
ENFORCEMENT: A REPORT TO THE WAGE AND HOUR DIVISION (2010), available at http://www.dol.gov/
whd/resources/strategicEnforcement.pdf.
14
. See LOCKE, supra note 10; Locke et al., supra note 7.
15
. See Piore & Schrank, Trading Up, supra note 8.
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
8 COMP.LABORLAW&POL’YJOURNAL [Vol. 35:1
power to work with suppliers in jointly identifying the pressures driving
noncompliance at the factory level, and to collaborate with them on finding
solutions. Yet if it is true, as the authors themselves recognize, that buyers
are part of the problem, why are they not also part of the solution?
In short, an effective strategy for fighting sweatshops must grapple
with the structural dynamics of apparel supply chains. To further
substantiate this claim, in the next Part we offer a closer examination of the
relationship between garment sourcing and labor performance in exporting
countries.
III. FROM THE SWEATSHOP TO THE SUPPLY CHAIN:
THE APPAREL PRICING MECHANISM AND LABOR
(NON)COMPLIANCE
Over the last two decades, competition in the global apparel industry
has intensified as a result of a World Trade Organization mandated ten-year
phase out of the Multi-Fiber Arrangement, which ended three decades of
controlled trade in textile products. Combined with the entry of China into
the WTO, trade liberalization enables buyers to play suppliers in more
countries off against each other without concern for quotas or other barriers
that had earlier restricted their sourcing options. One reflection of changing
sourcing patterns is China’s growing share of the U.S. import market,
which increased from 13.55% in 1989 to 41.97% in 2010; Vietnam is now
the second largest apparel exporter to the United States and Bangladesh is
the third.
16
Changes in the global geography of apparel production have
occurred alongside changes in the organization of the apparel market in
importing countries. In the United States, consolidation and growth at the
retail end of the apparel supply chain over the past several decades has led
to“theoverwhelmingdomination of the marketby a handful of enormous
retailers.”
17
Technological advancements and lean retailing further facilitate the
expression of buyer power among major retailers.
18
Today’smanufacturers
must manage much shorter lead times because retailers, rather than stocking
extra items in store inventory, prefer to use point of sale technology to track
purchases and reorder replacements on a continual basis. Some buyers even
expect their large suppliers to track retail sales themselves so that they
16
. U.S. DEP’T COM., Office of Textiles and Apparel (OTEXA), http://otexa.ita.doc.gov (last visted
Sept. 9, 2013).
17
. EDNA BON ACICH & RICHARD P. APPELBAUM, BEHIND THE LABEL: INEQUALITY IN THE LOS
ANGELES APPAREL INDU STRY (2000); Will Milberg, Shifting Sources and Uses of Profits: Sustaining
US Financialization with Global Value Chains, 37 ECON. & SOC’Y 420 (2008).
18
. FREDERICK H. ABERNATHY, JOHN T. DUNLOP, JANICE H. HAMMOND & DAVID WEIL, A
STITCH IN TIME: LEAN RETAILING AND THE TRANSFORMATION OF MANUFACTURING—LESSONS FROM
THE APPAREL AND TEXTILE INDUSTRIES (1999).
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
2013] JOINT LIABILITY IN GLOBAL SUPPLY CHAINS 9
know when and how many of each item to replenish. In addition to the
shorter lead times created by new technology, fashion seasons have grown
ever shorter, as retailers and brands aim to introduce new products on a
regular basis. Zara’s“fastfashion”system,whichfeaturesa near-constant
stream of new garments into stores, exemplifies an extreme version of this
model.
19
One of the most direct impacts of the shift to shorter lead times, more
styles, and more volatile orders is in the area of working hours; forced,
excessive, and inadequately compensated overtime is an endemic problem
in the global apparel industry.
20
Because each new worker hired incurs
training costs and fixed-benefit costs on employers, many firms prefer to
maintain a smaller workforce and then demand that these employees work
excessive hours during periods of peak demand. In effect, the retail
revolution has meant that lead firms are increasingly able to shift the risks
associated with volatile product demand onto their suppliers, and the
suppliers in turn shift the burden onto their workers.
Figure 1 models the impact that variations in sourcing contracts have
on working hours in a given apparel factory over the course of a twelve-
month cycle. The dotted line indicates a standard work week with a fixed
number of workers. Anything above the dotted line is overtime. Anything
below the dotted line represents underutilized working hours. The first
panel models relatively stable sourcing contracts. Although there is some
overtime (represented by the gray area above the dotted line), it is limited
and, we can assume, consistent with maximum working hours as
established by local labor law. The second panel models highly volatile
sourcing contracts. Here, the problem for the employer is that hiring more
workers—which would shift the dotted line upwards and reduce the need
for excessive overtime—increases worker downtime during the ebbs in
production orders, while also increasing fixed payroll and training costs.
Consequently, employers are likely to maintain the same level of
employment, and require excessive overtime in order to meet demand
during the peak production cycles that result from volatile sourcing
contracts with short lead times.
21
In addition to physical fatigue and
disruption in family life, chronic overtime has been linked to significantly
19
. Barbara Mihm, Fast Fashion in a Flat World: Global Sourcing Strategies, 9 INT’L BUS. &
ECON. RES. 55 (2010).
20
. Michael Piore, The Economics of Sweatshops, in NO SWEAT: FASHION, FREE TRADE, AND THE
RIGHTS OF GARMENT WORKERS 135 (Andrew Ross ed., 1997).
21
. The relationship between short lead times and labor violations is not unique to supply chains
for apparel and footwear products, as underscored by a spate of recent articles on labor problems in
Apple’ssupplychain. See Charles Duhigg & David Barboza, In China, Human Costs Are Built Into an
iPad, N.Y. TIMES, Jan. 26, 2012 at A1. Richard Locke, How Waiting Longer for the iPhone Could Help
Workers, SOUTH CHINA MORNING POST (Oct. 18, 2012).
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
10 COMP.LABORLAW&POL’YJOURNAL [Vol. 35:1
greater health and safety problems at work.
22
Volatile orders also push
firms to turn to contingent workers and unauthorized subcontractors. And
the need to minimize fixed costs also creates pressure to maintain low
building rent, which is why production often ends up in unsafe structures.
Figure 1. Sourcing and Forced Overtime
These changes, and the increased pressures they generate for suppliers
and their workers, are occurring in a context of generally stagnant to
declining prices for manufacturers. The growing power of buyers vis-à-vis
suppliers, facilitated by trade liberalization and retailer consolidation, has
enabled them to resist pressures to pay their contractors more. Negotiations
withpotential suppliersfocusnoton whataspecificproducer’scostmight
be to make an item, but rather the price that the brand or retailer is willing
22
. Anne Spurgeon, J Malcolm Harrington & Cary L. Cooper, Health and Safety Problems
Associated with Long Working Hours: A Review of the Current Position, 54 OCCUPATIONAL & ENVT L.
MED. 367 (1997); Allard E. Dembe, J. Bianca Erickson, Rachel G. Delbos & Seven M. Banks, The
Impact of Overtime and Long Work Hours on Occupational Injuries and Illnesses: New Evidence from
the United States, 62 OCCUPATIONAL & ENVTL. MED. 588 (2005).
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
2013] JOINT LIABILITY IN GLOBAL SUPPLY CHAINS 11
to pay. This systematic shunning of responsibility for the costs of decent
working conditions—a break with the social contract enshrined in the
jobbers agreements of the mid-twentieth century U.S. apparel industry—has
hindered efforts to significantly improve conditions for garment workers in
the global economy.
A simple exploration of the price paid by U.S. importers for apparel
from 1989 to 2010, and the status of workers’ rights in the top twenty
apparel exporting countries to the United States during that time period,
helps to illustrate our point. The Office of Textiles and Apparel at the U.S.
Department of Commerce records apparel imports by volume and by dollar
values. Dividing the value of imports by import volume (measured in
square meters) gives us the price per square meter of imported apparel to
the United States from 1989 through 2010. In nominal terms, we see that
the unit price increased from $3.48 in 1989 to $3.77 in 1997 and then
declined to $2.89 in 2010. This is a drop of 23%. More important, the real
dollar price per square meter of apparel entering the U.S. declined by 48%
from 1989 to 2010.
23
How has this pricing trend affected workers? David Cingranelli of
Binghamton University and David Richards of the University of
Connecticut have developed the Cingranelli and Richards Human Rights
Dataset (CIRI), which contains standards-based quantitative information on
government respect for internationally recognized freedom of association
rights in approximately 195 countries for the years 1981 to 2010.
24
To
explore the relationship between pricing and labor performance, we first
identified the twenty top apparel exporters (by volume) to the United States
for each year from 1989 to 2010. This group of countries represents
between 83% and 95% of U.S. imports per year during this period. The
CIRI Workers’ Rights score for each country was recordedf oreach year
and averaged,givingusanindicatorofthestatusofworkers’rightsduring
the same period covered by our unit price of imports measure, 1989–2010.
We expect that the status of workers’ rights in apparel exporting
countries would be influenced by the declining unit price paid by buyers.
That is, as suppliers are increasingly pressured to produce goods more
quickly and at a lower price, they would also experience greater pressure to
23
. We use the GDP deflator provided by the U.S. DEP’T OF COM., Bureau of Economic Analysis,
http://www.bea.gov/index.htm (last visited Sept. 9, 2013).
24
. The CIRI dataset provides some of the best, up-to-date data available for the relevant time
period on the internationally respected rights of workers to associate freely, bargain collectively, and
strike. The focus of the variable appears to be on these freedom of association rights, but it also includes
a prohibition on the use of any form of forced and child labor, minimum wages, hours of work, and
safety and health. Ontheirthreepointscale,a scoreof0indicates that workers’rights wereseverely
restrictedascoreof1indicatesthatworkers’rightsweresomewhatrestrictedandascoreof2indicates
thatworkers’rightswerefullyprotected.
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
12 COMP.LABORLAW&POL’YJOURNAL [Vol. 35:1
violate the internationally recognized rights to freedom of association and
collective bargaining because labor is the most significant production cost
in apparel assembly, and violating these rights reduces union strength and
thus weakens an important mechanism for increasing wages and benefits.
Whatwefindisthatrespectforworkers’associationalrightsheldrelatively
steady in the 1990s, but has declined significantly since 2001.
Figure 2. Unit Price of U.S. Apparel Imports and Labor Rights
Performance Among Leading Exporters
25
It is important to note that the data on which Figure 2 is based are not
weighted. That is, China, which represented 41.97% of apparel imports to
the United States in 2010 and scores a 0 in the CIRI dataset, is weighted the
same as a country like the Dominican Republic, which now represents less
than 1% of U.S. apparel imports and scores a 2. If we were to use a
weightedscale,theindicatorofdecliningrespectforworkers’rightswould
be much more dramatic. By not using a weighted scale, our data emphasize
thatthe trend indecliningrespectfor workers’ rightsinapparel exporting
countries is widespread, and not solely accounted for by trends in one
country.
To further explore pricing dynamics in global supply chains, we
examinedunit price dataforoneapparelcategory,men’sand boy’scotton
trousers (HTS 347), from four top exporting countries: China, Mexico,
Honduras, and Bangladesh. Controlling for inflation, we see that in all four
countries the price per square meter has declined since 2000. Notably, in
25
. Authors’calculationsarebasedonOTEXAandCIRIdata.
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
2013] JOINT LIABILITY IN GLOBAL SUPPLY CHAINS 13
2010, the price per square meter remained the highest in Mexico, the
country that has faced the greatest decline in market share among the top
exportersofmen’spants.
Figure 3. Unit PriceofMen’sandBoy’s Cotton Trousers
26
USD/Sq. Meter (2005=100)
We are not making a general argument about an unambiguous, across
the board decline in working conditions and wages in the global apparel
industry. Certain state initiatives, private governance mechanisms, and
labor union strategies may have arrested some of the more notorious abuses
in the sector, including child labor.
27
Yet, overall it appears that working
conditions and respect for workers’ rights and labor standards have not
improved considerably over the last fifteen to twenty years. And, as we
note below, the fact that hundreds have died in factory fires and collapses in
recent years suggests that even egregious health and safety violations
continue to threaten the well-being of global garment workers.
The impact of downward prices on suppliers also emerges from the
primary data we have gathered via fieldwork. One contractor in Central
America,reflectingproudlyonhiscompany’sinvestmentinitsworkforce,
nevertheless acknowledged frankly that
No customer has ever come in here and paid a premium price because of
that, or pushed to get in the door [because of it], or said “we know you
cost a little more because of what you do, and that’s okay.” It’s all
26
. OTEXA data “Men’sand Boy’sTrousers”refertoU.S.TextileandApparelCategorySystem
“HTS 347.” See U.S. DEP’T COM., U.S. Textile and Apparel Category System (2013)
http://otexa.ita.doc.gov/correlat/cor347.htm (last visited Sept. 9, 2013).
27
. ANNER, supra note 5; ETI REPORT, supra note 6.
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
14 COMP.LABORLAW&POL’YJOURNAL [Vol. 35:1
about Vietnam. Seven cents [the differential between his price and what
the client would pay to have the same product made in Vietnam].
“Whatcanyoudo about that?” It’sno added value to them. They say
they care, but the incentive system is set up to care more about price.
28
These observations were echoed by a Honduran labor rights activist
who has been involved in numerous campaigns to improve conditions in
local factories:
In all of the negotiations we have had with maquila [apparel export]
factories, the first argument the managers put forward is that they cannot
improve working conditions, that they cannot provide a decent
production bonus, that they cannot provide a decent wage increase,
because the brands pay them such a low price.
29
Whether we examine statistics of the sort presented above on apparel
import values and labor performance, or the primary data we gathered
during fieldwork in garment-exporting countries, the empirical evidence
leads us to the same conclusion: any effort to target sweatshops must
address the sourcing and pricing dynamics between suppliers and buyers,
and ensure that buyers take greater responsibility for the terms and
conditions of work at suppliers’ factories. The question that remains is:
How do we arrive at a fair pricing and stable sourcing mechanism that
benefits workers? We believe the answer lies in binding agreements
negotiated between workers, suppliers, and buyers. There is a historical
precedent for achieving such a mechanism. During much of the twentieth
century, jobbers agreements brought stability to subcontracting relations
within the U.S. apparel industry and went a long way toward eradicating
sweatshop conditions in those regions where workers were organized.
IV. REGULATING BUYERS: JOBBERS AGREEMENTS
AND THE DOMESTIC STRUGGLE AGAINST SWEATSHOPS
The relationship between subcontracting and sweatshops dates back to
the origins of industrial apparel production in the United States. The
epicenter of the industry, New York City, was the birthplace of the so-
called “outside model” of production, in which manufacturers contracted
with small-scale, independent sewing shops to produce some of the
garments they designed and/or marketed. The outside model eventually
surpassedthetraditional“inside”systemwhereby manufacturersemployed
their own garment workers. Over time, it became so pervasive that a new
term—jobber—was coined to refer to individuals and firms that designed or
sold apparel, but did little to no manufacturing of their own. Already in
28
. Author interview with apparel manufacturer, Managua, Nicaragua (Sept. 30, 2010).
29
. Author interview with union organizer, San Pedro Sula, Honduras (May 25, 2012).
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
2013] JOINT LIABILITY IN GLOBAL SUPPLY CHAINS 15
1893, a New York State Bureau of Labor inspector reported that while
“therewereprobablyonehundredwholesalecloakhouses”inthe garment
district, “not over half a dozen provide their own factories and
workshops,”
30
meaning the vast majority were jobbers contracting with
independent sewing shops. The outside system was preferred because it
enabled jobbers to better manage volatility and fluctuating demand, while
also avoiding the union contracts that were becoming common in larger
factories, thanks to a wave of historic strikes orchestrated by the rapidly
growing International Ladies Garment Workers Union (ILGWU).
The ranks of the ILGWU increased in the early 1900s during a period
of intense labor conflict, including a massive strike at the Triangle
Shirtwaist Company in 1909. Two years later, in 1911, a fire at the
Triangle killed 146 workers, most of them young, immigrant women. In
the aftermath of the tragedy and the massive public response it provoked,
the governor of New York created a Commission to investigate factory
safety and to make recommendations for improving worker health and
welfare. One of the most active members of the Factory Investigating
Commission was its vice-chair, then-New York Assemblyman Alfred
Smith. By 1913, sixty of the sixty-four laws recommended by the
Commission had been enacted.
In 1924, now-governor Alfred Smith created another Commission to
investigate the apparel industry, this time in response to a period of
protracted labor conflict and the looming threat of an industry-wide strike.
In its final report, the members of this Commission acknowledged that
many of the problems in the industry were rooted in the outside system of
production and the competition that pervasive subcontracting networks
generated among sewing shops. They observed that the formal
independence of the contractor andthe jobber was belied by the former’s
reliance on the latter, emphasizing that the relationship should be viewed
more in terms of substance than form: “By whatever name he may call
himself, the jobber controls working conditions; he controls employment,
and that element of control imposes upon him the responsibility that he
shall so conduct his business that proper working standards may be upheld
insteadof undermined.”
31
The Report went on to recommend changes to
the structureofcontractingrelationsaswould“tendtoregularizetheflow
of work into sub-manufacturing shops . . . cause closer relations between
jobbers and manufacturers, and stabilize working conditions in the shop.”
32
Yet these advisory recommendations were never implemented.
30
. LOUIS LEVINE, THE WOMEN’S GARMENT WORKERS 17 (1924).
31
. LEON STEIN, OUT OF THE SWEATSHOP: THE STRUGGLE FOR INDUSTRIAL DEMOCRACY 280
(1977).
32
. Id. at 281.
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
16 COMP.LABORLAW&POL’YJOURNAL [Vol. 35:1
Changing the structural dynamics of the industry was imperative for
the ILGWU. Cutthroat price competition among contractors not only
created downward pressure on wages and working conditions; it also made
it difficult for the union to organize workers because jobbers could easily
move orders from an organized to a nonunion shop. Aware of the threat
that subcontracting posed to organized labor, the ILGWU argued as early as
1923 that jobbers and contractors were part of an “integrated process of
production,”
33
and as such, were jointly liable for wages and working
conditions in contracting shops. The challenge confronting the union was
how to institutionalize this liability. The solution it devised was the jobbers
agreement: a collective bargaining agreement between the union local
representing the garment workers in a factory, and the buyer (i.e., jobber)
whose apparel was being sewn in that shop.
Although the first jobbers agreements were signed in the 1920s, they
proved difficult to enforce, both because the process of negotiating
contracts with individual buyers was time consuming, and because jobbers
were shifting orders out of the city to nonorganizedfactories,or“run-away”
shops in lower-cost regions, such as northeastern Pennsylvania. This status
quo persisted until 1933, when the passage of the National Industrial
Recovery Act (NIRA) resulted in “Codes of Fair Competition” that
essentially extended the logic of the jobbers agreement to the entire
industry. Both of the Codes regulating the garment business (covering the
men’sandladies’segments,respectively)recognizedthe principleof joint
liability,andspecifiedthatthepricespaidbyjobbersshouldbe“sufficient
to enable the contractor . . . to pay the employees the wage and earnings
provided for in this Code,together with an allowance for the contractor’s
overhead.”
34
By the time the U.S. Supreme Court ruled the NIRA unconstitutional
in 1935, the ILGWU was strong enough to secure collective bargaining
agreements that included some of the key provisions of the now-invalidated
Codes. Organizedlabor’sposition was buttressed by otherNew Deal-era
legislation, including the National Labor Relations (Wagner) Act of 1935
and the Fair Labor Standards Act of 1938. Clearly, the political context
createdbytheNewDealfacilitatedtheILGWU’sgrowthduringthemiddle
third of the twentieth century and increased its ability to bring employers to
the bargaining table. In this sense, we would grant that the U.S. state
played an important role in the fight against domestic sweatshops. But
what happened once labor got capital to the table was also decisive: the
33
. Katie Quan, EvolvingRelationsintheWomen’sApparel Industry (2007) (unpublished article)
(on file with U.C. Berkeley Center for Labor Relations and Education).
34
. EMIL SCHLESINGER, THE OUTSIDE SYSTEM OF PRODUCT ION IN THE WOMEN’S GARMENT
INDUSTRY IN THE NEW YORK MARKET 37 (1951).
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
2013] JOINT LIABILITY IN GLOBAL SUPPLY CHAINS 17
union decided to leverage its strength in pursuit of a specific objective—
namely, negotiating collective bargaining agreements with both the apparel
contractorsthatdirectlyemployedtheunion’smembersas well as with the
jobbers that, as clients of the contractors, indirectly employed garment
workers. As the ILGWU grew more powerful, both the contractors and the
jobbers responded by forming associations of their own that could bargain
collectively with the union.
35
Agreements negotiated with these groups of
employers provided more coverage than the early jobbers agreements
because they applied to all the companies that were members of the
signatory association instead of just one firm.
Thus, the New Deal marked a turning point in the ILGWU’s battle
against sweatshops. Within a decade, jobbers agreements had become the
lynchpin of an industrial relations model described as triangular collective
bargaining—so named because the goal was to regulate, via a set of paired
contactors’andjobbersagreements,relationsbetweenthethreesidesofthe
production triangle: the workers as represented by the union, and the
jobbers and contactors, each represented by their own employers
association.
It is difficult to overstate the impact that the implementation of jobbers
agreements had on conditions of employment for unionized garment
workers. In August 1938, Life Magazine devoted a lengthy cover story to
thetransformation,noting that “[t]hirtyyearsagotheindustrystankofthe
sweatshop and the cruelest kind of exploitation. Workers toiled 16 hours a
day for $2 to $8 a week. Today, they get $15 to $35 for working 35 hours a
week,onlyafewhoursmorethantheyonceworkedintwodays.”
36
Time
series data indicate that gains on the wage front were indeed significant. As
Figure 4 demonstrates, wages varied within the apparel industry, but the
overall trend is clear: During the middle decades of the twentieth century,
the earnings of garment workers—a heavily feminized and, at least in New
York, largely immigrant workforce—kept pace with general manufacturing
wages.
37
35
. Employers’ associations for jobbers and contractors were typically organized around a
particular product and market niche. For example, in the dress segment of the industry, the National
Dress Manufacturers represented jobbers of higher-priced dresses, the Affiliated Dress Manufacturers
included mid-market jobbers, and the Popular Price Manufacturers was comprised of jobbers whose
dresses were sold at lower prices. This structure was more or less reflected in the organization of the
contractorstheemployers’associationsfordresscontractorsincludedUnitedBetterDress(up - to mid-
market) and United Popular Dress (low- to mid-market). See David Melman, The Cause and Effect of
the ILGWU Dress Industry General Strike of 1985 (Feb. 1994) (MA Thesis, Oberlin College) (on file
with authors).
36
. AGreatandGoodUnionPointstheWayforAmerica’sLaborUnions, LIFE, Aug. 31, 1938, at
43, 44.
37
. Because this is national wage data, it likely underestimates the gains for workers in New
York’s garment district, which boasted the highest rate ofu nion density and the largest number of
jobbers agreements. For example, the annual earnings of unionized dressmakers in New York City in
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
18 COMP.LABORLAW&POL’YJOURNAL [Vol. 35:1
Figure 4. Hourly Earning in Current U.S. Dollars, 1940-1975
38
While it is true that the capacity of the federal and state departments of
labor to enforce minimum wage and hour laws was much more robust
during the heyday of the jobbers agreements than it is today, such
enforcement cannot explain the increase in garment worker earnings. This
is evident from the fact that apparel wages during this period exceeded
minimum wages by a large margin. Between 1947 and 1990, for example,
annual average garment worker earnings never fell below 146% of the
federal minimum wage, and in many years they were more than 200% of
the minimum. Indeed, somewhat remarkably for an industry that has
become equated with low pay, by the late 1940s and early 1950s, apparel
workers received the highest hourly rate in the nondurable goods sector,
according to data compiled by the U.S. National Compensation Survey.
Thus, while national and state minimum wage laws provided a wage floor,
the jobbers agreements secured earnings for workers far beyond those
guaranteed by governmental regulation.
The wage premium is one important indication of the impact of
jobbers agreements, but their achievements were even more striking in the
area of benefits. Forexample,JoelSeidman finds thattheunion“usedits
bargaining power to achieve a comprehensive system of welfare programs,
including retirement pay, severance pay, weekly supplementary benefits,
and hospital, disability, and medical benefits-all of them entirely financed
1946 was 14.5% above the average for all manufacturing industries whereas their counterparts in New
Jersey and Pennsylvania, where many shops were also covered by jobbers agreements, earned 8.9%
above average. See Melman, supra note 35, at 9.
38
. U.S. BUREAU LAB. STAT., http://www.bls.gov/ (last visited Sept. 9, 2013).
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
2013] JOINT LIABILITY IN GLOBAL SUPPLY CHAINS 19
by employer contributions.”
39
Working hours was another area where the
jobbers agreements left their mark: In an industry notorious for chronic
excessive overtime, in 1960 apparel employees were working an average of
36.3 hours a week according to the National Compensation Survey. Indeed,
apparel workers often recorded the shortest work week of any sector in all
the nondurable and durable goods sectors.
40
In sum, these data indicate that
apparel workers, largely as a result of jobbers agreements, were able to
increase wages, extend benefits, and reduce hours of work well beyond
what was stipulated by law.
By 1959, the success of the jobbers agreements in eradicating
sweatshops was so universally recognized that that no less a conservative
than Senator Barry Goldwater would speak out in their defense on the
Senate floor:
We conferees are in the very peculiar position of every one of us
agreeing that we do not intend to upset the status quo of the garment or
apparel industries . . . . I have been engaged in the retail end of this
business all my life. I have watched what has happened in the garment
section of New York and the garment section of Philadelphia and St.
Louis and Chicago and on the west coast, and have seen sweatshops
disappear. I have seen order come out of chaos. I have seen unions
create profits for businesses which were unable to produce profits, and
Mr. President, none of us wants to disturb for one second the status that
the garment trade now occupies under the present law.
41
Goldwater’s statements were made in the context of debate over the
then-proposed Landrum-Griffin revisions to the National Labor Relations
Act (NLRA). It was feared that the jobbers agreements would be rendered
illegal by a proposed amendment prohibiting“any labor organization and
any employer to enter into any contract or agreement . . . to cease doing
businesswithanyotherperson.”
42
The jobbers agreements did just that: it
required jobbers to contract only with unionized contractors.
Supporters of the jobbers agreements insisted that the provision should
not apply where the companies involved have a unity of interest through an
integrated process of production. Senator Jacob Javits expressed his view
that the “elimination of sweatshops” was “heavily attributable” to the
system of joint liability,
43
while Senator John F. Kennedy argued that doing
awaywiththerighttonegotiatejobbersagreementswould“invitechaosin
the industry.”
44
Ultimately, the amendment was passed, but a Garment
Industry Proviso was added to the NLRA, specifically exempting
39
. Joel Seidman, The I.L.G.W.U. in the Dubinsky Period, 9 LAB. HIST. 55, 62 (1968).
40
. Id. at 55.
41
. See 95 CONG. REC. 8709 (1949).
42
. Pub. L. No. 86-257, 73 Stat. 543-544, 29 U.S.C. § 158(e).
43
. 105 CONG. REC. 17381 (1959).
44
. See 105 CONG. REC. 17327 (1959).
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
20 COMP.LABORLAW&POL’YJOURNAL [Vol. 35:1
companies in the apparel industry from the hot cargo and secondary boycott
prohibitions of the Act. In preserving the jobbers agreements as an
effective tool against sweatshops, Congress explicitly adopted the theory of
the garment supply chain as an “integrated process of production” into
federal law.
Yet while the jobbers agreements survived many legal challenges, they
were unable to withstand the globalization of textile and apparel
manufacturing that occurred during the last third of the twentieth century.
At one level, geographical shifts in production were nothing new. Jobbers
had long tried to circumvent union contracts by placing orders with
nonorganized contractors outside New York City and eventually outside the
northeast. The union responded by launching organizing drives in these
locations, and in this way, expanded its reach into new areas, including
states in the mid-Atlantic and mid-West and even Puerto Rico. However,
when new trade regimes encouraged U.S. manufacturers to open factories in
Mexico and the Caribbean Basin, the union confronted an even greater
challenge in organizing workers offshore.
Even more than the decision of traditional manufacturers to relocate or
expand operations in lower-wage countries, what transformed the
geography of global apparel production was the development of “private
label” (store-brand) lines by retailers and brands. Unlike traditional
jobbers, most of whom started as apparel manufacturers and continued to
maintain at least a minimal amount of in-house production, this new breed
of buyers has always relied on global subcontracting networks.
Consequently, they have no history of garment production and have never
negotiated a collective bargaining agreement with an apparel industry
union. According to one of our informants, a long-time ILGWU organizer,
the brands and retailers that dominated today’s apparel industry “got into
business by end-running or circumventing the jobber-contractor formula we
had developed in the U.S.”
45
In other words, by sourcing from overseas
contractors, global buyers avoid having to pay for the kinds of labor
standards that the ILGWU had, to a large degree, succeeded in imposing on
the domestic industry through jobbers agreements.
By the turn of the century, it was clear that what began as an emergent
trend toward offshore production by apparel manufacturers had morphed
into the fundamentally new paradigm of global sourcing by retailers and
brands. On the domestic front, this transformation is manifest in the twin
trends of increasing import penetration and declining garment worker
wages. Import penetration of the U.S. apparel market rose modestly
45
. See Jennifer Bair, The Limits to Embeddedness: Triangular Bargaining and the Institutional
Foundations of Organizational Networks (Inst. Behav. Sci. Insts. Gr., Working Paper No. IN ST2 012-10,
2012), available at http://www.colorado.edu/ibs/pec/pubs/wp.html.
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
2013] JOINT LIABILITY IN GLOBAL SUPPLY CHAINS 21
through the 1970s, and then increased dramatically over the course of the
next decade. Although unaffected by the initial rise in imports, real wages
began a significant and protracted decline around 1980 such that by 1990,
they were well below 1965 levels. Apparel employment tracked the decline
in wages, particularly in New York City, where the number of garment
workers plummeted from 140,000 in 1980 to about 90,000 in the mid-
1990s.
46
By the end of that decade, the union was no longer signing new
jobbers agreements because so many jobbers had either closed or moved
their business entirely offshore, essentially bringing to an end the system of
joint liability that had regulated much of the domestic apparel industry for
more than five decades.
V. KEY POLICIES OF JOBBERS AGREEMENTS
This Part explores how the jobbers agreements managed to stabilize
contracting relations and improve working conditions by analyzing the
agreements’ key elements.
47
Jobbers agreements institutionalized the
principle of buyer liability in three specific ways. First, they prevented
contractors from competing on labor costs by negotiating wages directly
with the jobber. Second, they stabilized subcontracting relationships by
requiring jobbers to register their designated contractors with the union.
Third, they made jobbers directly liable for certain labor costs beyond
wages. Our brief description of these policies enables an analysis in the
following Part of how the logic of the jobbers agreement model might be
replicableintoday’sglobalizedapparelindustry.
A. Paying for Decent Labor Conditions
Decent working conditions in the U.S. garment industry were not
established until competition among contractors on the basis of labor price
was arrested. The ILGWU accomplished this by reaching up to the top of
the supply chain and negotiating the cost of labor directly with jobbers.
The jobbers agreements included numerous provisions detailing the
jobber’sobligationsforwagesandworkingconditionsincontractingshops.
These included:
46
. Robert Ross, The New Sweatshops in the United States: How New, How Real, How Many, and
Why?, in FREE TRADE AND UNEVEN DEVELOPMENT: THE NORTH AMERICAN APPAREL INDUSTRY
AFTER NAFTA (Gary Gereffi, David Spener & Jennifer Bair eds., 2002).
47
. This Part draws from our analysis of a set of thirty-two collective bargaining agreements that
were negotiated by the ILGWU over more than sixty years, from 1930 to 1994. This data from the
ILGWU archive at Cornell was supplemented interviews carried out by the authors in spring 2012 with
former and current officials of the ILGWU (or its successor unions) with intimate knowledge of the
jobbers agreements.
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
22 COMP.LABORLAW&POL’YJOURNAL [Vol. 35:1
Minimum wages by occupational category (sewer, cutter, presser etc.),
both for “week workers” (cutters, sample makers, etc.) and “piece rate
workers”(primarily,sewingmachineoperators)
A process for setting piece rates sufficient to yield the minimum wage for
a sewer of average efficiency, as well as a process for resolving any
disputes that might arise in the negotiation of these rates;
Provisions triggering wage increases in the case of inflation or increases
to the statutory minimum wage;
Hours of work, including provisions regarding maximum permissible
overtime and the compensation rules for overtime work.
To be sure, such provisions are hardly atypical of collective bargaining
agreements, but it is important to emphasize that these provisions were
included in the union’s contracts with the jobber, even though the jobber
was not the direct employer of the workers.
The union also had a matching agreement with the employers’
association representing apparel contractors, but this contract, which was
signed after the jobbers agreement was finalized, simply repeated the
language regarding wages and working conditions that had already been
negotiatedbetweentheunionandthejobbers’association. The sequencing
of these negotiations reflected the fact that what the contractors were able to
pay to their workers depended largely on what the contractors were, in turn,
being paid by the jobber. Astheunion’sformerresearchdirectorexplained
to one of the authors: “The ILGWU recognized that the jobber was the
lynchpin of the industry. Contactors couldn’t pay anything unless the
jobber paid it to him or for him.”
48
The jobbers agreement also contained language regarding the price
negotiation between the jobber and the contractor:
A member of the Affiliated [the employers’ association for dress
jobbers] whose garments are made in contracting shops shall pay to such
contractors at least an amount sufficient to enable the contractor to pay
the workers the wages and earnings provided for in this agreement, and
in addition a reasonable amount to the contractor to cover his overhead
and profit.
49
Like other provisions, this requirement was subject to binding arbitration by
an independent chairperson designated to resolve disputes in the industry.
Importantly, jobbers agreements had broad coverage because they
were negotiated and signed not with individual firms but rather with
employers’ associations that included the majority, if not the entirety, of
jobbers and contractors making a particular type of apparel (e.g., dresses).
Because all workers were required to be paid the wage levels spelled out in
the contract regardless of where they worked, and because jobbers were
48
. Cited in Bair, supra note 45 at 8.
49
. E.g., Affiliated Dress and ILGWU 1936, at 14.
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
2013] JOINT LIABILITY IN GLOBAL SUPPLY CHAINS 23
required to pay all their contractors prices sufficient for these wages to be
paid,contractorscouldnolongervieforajobber’sbusinessbyundercutting
competitors’ labor costs. Consequently, competition among contractors
was based less on price than on quality and efficiency.
B. Regulating and Stabilizing Contracting
Relations in the Supply Chain
The second central component of the system created through triangular
bargaining was a regime for regulating and stabilizing relationships
between jobbers and contractors. Specific provisions designed to achieve
this objective included:
• Requiring jobbers to designate, at the outset of an agreement, all
contractors that it intended to use during the course of the (typically three
year) agreement, and to register designated contractors with an
Administrative Board comprised of representatives from the union and the
jobbers’employers’association
• Restricting the supply of contractors that could be so designated to
unionized shops;
• Permitting a jobber to register only the number of contractors actually
required to manufacture its garments;
• Requiring jobbers to distribute work evenly across all designated
contractors and prohibiting the use of new contractors except when
additional capacity was needed;
• Prohibiting the discharge of a designated contractor for any reasons other
than poor quality and/or late delivery.
The intent of these provisions was to promote long-term, stable
relationships between jobbers and contractors. Just as the inclusion in
jobbers agreements of language regarding wages and working conditions in
contracting shops was intended to arrest competition on the basis of labor
costs, so too were the detailed clauses regulating the jobber-contractor
relationship meant to preclude jobbers from constantly switching between
contractors to secure a short-term advantage on price.
This system had a number of important consequences. First, it caused
the buyer to consolidate its orders among a smaller number of contractors
whosefactorieswerethenfilledupwiththatjobber’swork,orperhapswith
orders from several buyers that were party to the same master jobbers
agreement. The goal was to minimize dramatic fluctuations in the amount
of work given to a contractor over the course of the year so that the
contractor could provide steady employment and pay workers the wage
levels established by the agreement. Second, it created a powerful incentive
for individual contractors to accept unionization and to agree to the higher
labor standards provided for in collectively bargained agreements because,
by doing so, contractors gained the opportunity to become a designated
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
24 COMP.LABORLAW&POL’YJOURNAL [Vol. 35:1
supplier of one or more jobbers and consequently to gain access to steady
orders at fair prices. Moreover, by stabilizing production contracts, it
eliminated the need of contractors to rely on long hours, including the type
of forced overtime to meet short-term production goals depicted in Figure
1. Indeed, stable contracts are what allowed the union to negotiate a 35
hour work week for apparel workers.
Additionally, the close regulation of the contracting relationship (e.g.,
the requirement that orders be distributed evenly among contractors), the
prohibition of discharge without cause, and the availability of a dispute
resolution mechanism in the event of a disagreement between the buyer and
the supplier gave all of the parties confidence in the bargain they had struck.
Cheating on the part of jobbers, through the placement of orders in
nonregistered or nonunionized shops, tended to be detected and sanctioned,
including by the weapon of the so-calledjobbers’strike.
Finally, because the registration period lasted for the life of the
agreement (typically three years), and because designated contractors in
good standing were assured steady business as long as the jobber had orders
to give, the system enabled economic security for contractors and job
security for workers. In short, the contractor registration system ensured
that decent wages and conditions were feasible, incentivized contractors to
agree to such conditions, and brought stability to inter-industry relations
that had been chaotic and chronically insecure.
50
C. Guaranteeing Payment to Workers
A third important aspect of the jobbers agreements was that they held
jobbers liable for direct payment of certain production costs. These
provisions helped to ensure that workers received the compensation and
benefits due to them in the occasion that their direct employer—the
contractor—went belly up or absconded without paying them, a critical
problem in the notoriously footloose garment industry.
First, the jobbers agreements included provisions requiring jobbers to
make direct contributions into benefit accounts for workers’ pensions,
healthcare, and accrued vacation. Under the jobbers agreement system,
employee benefits were paid not by the contractor, but rather by the jobber
as the indirect employer of the contractor’s workers. These funds were
administered by the union, not the contractor, which made them portable—
50
. The provisions of the jobbers agreements that regulated and stabilized contracting between
jobbers and contractors produced buyer-supplier relationships similar to the ones that Locke et al.
suggestarea preconditionoftheirproposed“commitment”approachtolaborcompliance. The authors
acknowledgethat“acommitmentapproachispossibleonlyif buyersworkwith factoriesformore than
justafewmonths”andifbothbuyerandsuppliershareastrongdesireto“cultivate long-term business
relationships.” Locke et a l., supra note 7, at 345.
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
2013] JOINT LIABILITY IN GLOBAL SUPPLY CHAINS 25
a desirable trait in an industry with high turnover among contractors and
workers. Crucially, this design also ensured that significant liabilities did
not accrue upon poorly capitalized contractors. As a result, workers did not
lose their benefits if their direct employer went under.
Second, the jobbers agreements held jobbers liable for a portion of
unpaid wages in the case of abrupt contractor closure. Typically, jobbers
were responsible for up to two weeks of compensation left unpaid by a
bankrupting contractor. Although the portable benefit accounts, overall
stability provided by the system, and the existence (since 1935) of
unemployment insurance in the United States made this less crucial in
ensuring payment, the guarantor provisions played an important role in
ensuring that workers were paid the compensation due to them when their
direct employers shut down operations.
VI. CONTEMPORARY RELEVANCE
A. Resurrecting Contractual Joint Liability in Global Supply Chains
The policies described in the previous Part were the centerpiece of the
most successful historical precedent for eliminating sweatshop conditions in
the garment industry. We believe that similar policies institutionalizing the
principle of buyer liability are necessary for contemporary efforts to
vindicate labor rights in today’s apparel supply chains. While such policies
will not be easily or quickly won, recent developments are promising. This
Section reviews a number of contemporary initiatives that move in the
direction of joint liability in global supply chains.
1. Global Framework Agreements
The most frequently cited contractual instrument to enforce labor
rights are International or Global Framework Agreements (GFAs).
Negotiated between international union federations and multinational
companies or global brands, these agreements outline labor rights
commitments and provide a grievance structure for contesting violations of
these commitments. Michael Fichter and his collaborators,
51
among others,
observe that such agreements are most likely to contribute to labor rights
compliance when local unions and management are fully involved in the
development and implementation of a GFA. Some supporters of GFAs
further hope that they might help unorganized workers form unions and
51
. See Michael Fichter et al., Globalising Labour Relations: On Track with Framework
Agreements (Friedrich-Ebert-Stiftung Working Paper, 2012), available at http://www.fes.de/gewerk
schaften/index_en.php.
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
26 COMP.LABORLAW&POL’YJOURNAL [Vol. 35:1
eventuallyinstitutionalizeorganizedlabor’sroleinimprovingconditionsin
the workplace.
52
More than fifty GFAs have been signed since 1988 in
numerous sectors, including manufacturing, telecommunications,
agriculture, and retail, but to date only two of these cover apparel
production workers.
53
One of these was a GFA negotiated in 2007 by the International
Textile, Garment, Leather Workers Federation (ITGLWF) and Inditex, the
Spanish company that designs and markets the apparel brand Zara.
54
In
their analysis of the ITGLWF-Inditex agreement, Gregoratti and Miller
underscore both the promise and the limitations of GFAs as tools in the
fight against sweatshops.
55
On the one hand, they find evidence that this
agreement has been used effectively to remediate violations of workers’
right to freedom of association by contractors in individual cases.
However,they also concludedthat Inditex’s failure to maintain consistent
production levels or adjust its prices to accommodate increased production
costs impeded factory-level bargaining and undermined the sustainability of
improvements.
GregorattiandMiller’sanalysisdiagnosesamoregeneralweaknessof
the GFAs: The scope of such agreements is generally limited to general
recitations of the core ILO labor standards. In this sense, GFAs do not
differ significantly in content from the codes of conduct that have been
adopted by virtually all major brands, albeit with the additional requirement
that the brand at least consult with the global union body regarding
violations by contractors.
56
GFAs do not contain any substantive provisions
regulating the sourcing or pricing practices of the lead firm itself, such as
the requirements included in jobbers agreements that buyers make long-
term commitments to suppliers and pay prices consistent with decent
wages. Admittedly, provisions regulating buyer conduct may be less
crucial in the heavy manufacturing and construction industries in which
GFAs were first developed, where global corporations directly employ
workers around the world or have strategic partnerships with highly
capitalized suppliers. Nevertheless, their absence is a significant
shortcoming in the footloose global garment industry, and one which
52
. Michael Fichter & Dmitiris Stevis, International Framework Agreements in the United States:
Escaping, Projecting, or Globalizing Social Dialogues?, 33 COMP. LAB. L. & POL’Y J. 667 (2012).
53
. Mark Thomas, Global Industrial Relations: Framework Agreements and the Regulation of
International Labor Standards, 36 LAB. STUD. J. 269 (2011).
54
. In November 2011, the Japanese sportswear company Mizuno Corporation signed a GFA with
UI Zensen, a Japanese union affiliated with the ITGLWF. In June 2012, the ITGLWF merged with two
other international federations to form a new global union, IndustriALL.
55
. Catia Gregoratti & Doug Miller, International Framework Agreements for Workers’ Rights?
Insights from River Rich Cambodia, 2 GLOBAL LAB. J. 84 (2011).
56
. Owen Hernstadt, Are International Framework Agreements a Path to Corporate Social
Responsibility?, 10 U. PA. J. BUS. & EMP. 187 (2007).
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
2013] JOINT LIABILITY IN GLOBAL SUPPLY CHAINS 27
underscores the limitations of GFAs as a tool in the fight against
sweatshops.
Furthermore, as explored in a recent issue of this Journal, it is an open
question whether and how the commitments companies have made through
GFAs are legally enforceable.
57
As is typical of GFAs, neither of the GFAs
in the apparel industry include binding arbitration clauses or other formal
processes for resolving disputes between the signatories in the event that
they cannot be resolved through dialogue. In sum, GFAs are like jobbers
agreements in that they involve direct negotiations between multinational
companies and labor over labor rights issues, but they differ from jobbers
agreements in three important ways: they do not address issues of pricing
and contract stability; they do not involve contractors in the negotiating
process; and they lack robust dispute resolution and enforcement
mechanisms.
2. Accord on Building and Fire Safety in Bangladesh
There has been movement in recent years toward the establishment of
agreements that do impose substantive, binding obligations on apparel
buyers. The most significant recent development is the Accord on Building
and Fire Safety in Bangladesh, discussed at the outset of this Article. The
agreement was signed in May 2013 in the wake of the Rana Plaza disaster,
which claimed the lives of more than 1,100 workers, a disaster that itself
followed a series of horrific mass-casualty fires and building collapses in
Bangladesh’sapparelindustry.Theagreement’ssignatoriesaretwoglobal
union federations, IndustriALL and UNI, and, at present, more than seventy
apparel brands and retailers. Two labor rights NGOs, the U.S.-based
Worker Rights Consortium and Europe-based Clean Clothes Campaign,
played central roles in developing the agreement and are participating in its
implementation.
Under the Accord, each participating apparel brands and retailer has
committed to take the following steps:
• Disclose its supplier factories in Bangladesh to recognized, independent
fire safety experts, require these factories to submit to rigorous fire safety
inspections led by these experts, and accept public disclosure of all
inspection reports;
• Make commitments of at least two years, at current production volume
levels, to its supplier factories;
57
. See, e.g., Alvin L. Goldman, Enforcement of International Framework Agreements Under U.S.
Law, 33 COMP. LAB. L. & POL’Y J. 605 (2012); Rüdiger Krause, International Framework Agreements
As Instrument for the Legal Enforcement of Freedom of Association and Collective Bargaining? The
German Case, 33 COMP. LAB. L. & POL’Y J. 749 (2012); Christopher D. Pigott, Freedom of Association
in Private Transnational Law: How Enforceable Are the Commitments of European Companies in
North America?, 33 COMP. LAB. L. & POL’Y J. 775 (2012).
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
28 COMP.LABORLAW&POL’YJOURNAL [Vol. 35:1
• Require its suppliers to implement all repairs and renovations necessary to
make their factories safe, as determined through the inspection process;
• Pay suppliers prices sufficient to make it possible for them to afford the
necessary repairs and renovations and to operate in a safe manner;
• Require suppliers to allow worker representatives into their factories to
educate workers about workplace safety and worker rights;
• Cease doing business with any supplier that fails to comply with any of
the above requirements.
As of the time this Article went to press in October 2013, the
signatories were in the initial stage of implementing the Accord.
There are at least four key ways in which the Bangladesh Accord
reflects core principles of the historical jobbers agreements. First, the
Accord regulates the buying practices of apparel brands and retailers.
These lead firms are specifically required to help pay for factory safety
upgrades, which brands may do by adjusting the prices they pay contractors
for products or financing upgrades in other ways (e.g., joint investments,
offering business incentives, or paying for renovations directly). Buyers are
required to make at least an initial multi-year commitment to supplier
factories, a major deviation from the current footloose norm of the industry.
And lead firms are required to drop contractor factories that do not adhere
to the program’s standards. Together, these obligations may result in
greater contract stability and, thus, enhanced job security for workers.
Second,theAccordcallsforworkers’representatives to be fully equal
and empowered participants. The steering committee comprises three union
representatives (one from each of the Global Union Federations that signed
the agreement and a representative of the Bangladeshi labor movement) and
three representatives of participating companies. In addition to workers’
representatives having co-equal power with participating brands in
governing the program, the Accord also builds worker involvement into the
program in a number of other ways. Half of the members of the Health and
Safety Committees that are to be established in the supplier factories of
participatingbrands are toinclude workers’ representativeselectedby the
union. Factory inspection reports must be shared with worker
representatives. Union representatives are also to participate in the training
sessions on safety issues that are to be regularly provided to workers and
management. These provisions ensure that union representatives have
consistent access to workers and worksites, which is a major departure from
thestatusquoinBangladesh’sapparelsector.
Third, as in jobbers agreements and collective bargaining agreements
generally, these commitments are not merely general statements of intent,
but binding, contractually enforceable obligations. Under the accord’s
dispute resolution process, disputes concerning implementation are first
submitted to the seven-member oversight steering committee, comprised of
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
2013] JOINT LIABILITY IN GLOBAL SUPPLY CHAINS 29
three representatives chosen by the trade union signatories and three
representatives chosen by the company signatories, with a representative of
the International Labor Organization serving as a neutral chair. Any
decision of the steering committee may then, at the request of either party,
be appealed to a process of binding arbitration. Thearbitrator’sawardmay
be enforced in a court of law of the home country of the signatory party
against whom enforcement is sought. Under the New York Convention,
signed by the home country of every signatory company, domestic courts
have a broad obligation to enforce foreign arbitration awards.
58
Fourth, like the jobbers agreements and most other successful
collective bargaining efforts, the agreement covers a broad portion of the
industry. Its more than sixty signatories include many of the world’s
leading global brands and retailers, including PVH (world’s largest shirt-
seller), H&M (largest global buyer from Bangladesh), Carrefour and Tesco
(the second and third-largest retailers in the world), and Inditex (world’s
largest fashion retailer). A high level of density is important because it
means that increased costs resulting from the agreement will not put any
participating company at a competitive disadvantage relative to any other
signatory, thereby easing implementation.
That said, a number of key companies have not joined the Accord.
Although several U.S.-based companies are signatories (including PVH,
Abercrombie and Fitch, and Sean John), most participating companies are
based in Europe. Beyond refusing to sign the Accord, a group of seventeen
North American companies announced an alternative factory inspection
program, the Alliance for Bangladesh Worker Safety, in July 2013.
59
The
announcement of this rival effort, which was spearheaded by Wal-Mart and
Gap, provoked sharp criticism from worker safety advocates who argue that
the program has been designed to superficially resemble the Accord while
58
. The New York Convention (formally the Convention on the Recognition and Enforcement of
Foreign Arbitral Awards) places the burden of establishing the invalidity of an arbitral award on the
party challenging enforcement. Underth eNew York Convention, an arbitrator’s awardmay only be
vacated for one of several narrow grounds, such as the incapacity of the parties at the time of the
underlying agreement to submit disputes to arbitration, the violation of a party’s fundamental right to
participate in a proceeding, or the irregular composition of the arbitration tribunal. See Article V. For
more information on the New York Convention, including a list of signatory states, visit N.Y. ARB.
CONVENTION, http://www.newyorkconvention.org (last visited Sept. 9, 2013). For further discussion of
the Accord’s enforcement provisions, see BENJAMIN HENSLER & JEREMY BLASI, MAKING GLOBAL
CORPORATIONS’ LABOR RIGHTS COMMITMENTS LEGAL ENFORCEABLE: THE BANGLADESH
BREAKTHROUGH, WORKER RIGHTS CONSORTIUM, June 18, 2013, available at http://www.cleanclothes.
org/resources/recommended-reading/making-global-corporations2019-labor-rights-commitments-le
gally-enforceable-the-bangladesh-breakthrough.
59
. See Steven Greenhouse, U.S. Retailers Announce New Factory Safety Program, N.Y. TIMES,
May 30, 2013, at B6. The website of the program can be found at ALLIANCE FOR BANGLADESH
WORKER SAFETY, http://www.bangladeshworkersafety.org (last visited Sept. 9, 2013).
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
30 COMP.LABORLAW&POL’YJOURNAL [Vol. 35:1
omitting its most important substantive elements.
60
For example, these
critics note that—unlike the Accord—the Alliance for Bangladesh Worker
Safety does not mandate that participating companies contribute any funds
for the renovation and repairs needed for factories to operate safely.
Indeed, the only reference in the Alliancefor Bangladesh WorkerSafety’s
program document to participatingcompany’scontributionoffunds—apart
from administrative fees—is a loan program that is “not a condition of
membership” and is subject to whatever terms the companies impose.
61
Additionally, safety advocates note that—unlike the Accord—the program
lacks enforcement provisions; the only penalty for abandoning the effort is
the payment of administrative fees.
62
Finally, unlike the Accord, the
program was developed without worker representatives and lacks any role
for worker representatives in its governance. In short, critics view the
program as an effort to undercut the Accord by providing a less onerous and
less rigorous alternative.
While the Accord corresponds to the jobbers agreement model much
more clearly than the nonbinding, retailer-dominated alternative proposed
by Wal-Mart and Gap, there are also ways in which it differs from, and is
more limited than, the jobbers agreements. Most notably, it covers only one
area of labor standards—worker safety—which was not the focus of the
jobbers agreements. Some of the original jobbers agreements did include
worker safety provisions—for example, requiring regular fire drills and
providing that an employee be designated a volunteer fire warden charged
with inspecting the factory and working with the local fire department and
management to eliminate hazards. But all sides essentially agreed to
recognize the state as having responsibility for the primary enforcement of
building safety standards and regulations, leaving them to focus their
bargaining over economic issues such as wages, working hours, and
benefits.
63
Of course, it is precisely the demonstrated inability or
unwillingness of the Bangladeshi government to enforce building codes and
other workplace safety regulations that makes the Accord so critically
necessary. Manufacturers in Bangladesh have also been complicit in poor
enforcement, frequently citing their inability to afford meaningful
60
. See, e.g., Statement of Worker Rights Consortium, Clean Clothes Campaign, International
Labor Rights Forum, Maquila Solidarity Network and United Students Against Sweatshops, Safety
Scheme of GAP and Walmart Only “Empty Promises,” Jul. 10, 2013, http://www.cleanclothes.org/
news/press-releases/2013/07/10/safety-scheme-gap-and-walmart; Steven Greenhouse & Stephanie
Clifford, U.S. Retailers Offer Plan for Safety at Factories, N.Y. TIMES, Jul. 10, 2013, at B1.
61
. See BANGLADESH WORKER SAFETY INITIATIVE, MEMBERSHIP AGREEMENT, § 2.3.1,
http://www.bangladeshworkersafety.org/wp-content/uploads/Alliance-Member-Agreement-FINAL.pdf
(last visited Sept. 9, 2013).
62
. See id. §§ 9.1–9.2
63
. At the same time, the improved economic standing of contractor shops under the jobbers
agreements system enabled the implementation of building safety improvements.
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
2013] JOINT LIABILITY IN GLOBAL SUPPLY CHAINS 31
improvements, given the pressure that foreign buyers exert on prices. The
Accord on Fire and Building Safety in Bangladesh (though not the industry-
proposed alternative) reflects this reality, and acknowledges the relationship
between buyer pricing and workplace safety.
3. University-Licensed Sportswear
A number of relevant developments have occurred in the area of North
American university-licensed sportswear, a $4 billion a year business
involving many of the world’s largest brands, which has come under
considerable scrutiny from anti-sweatshop activists on college campuses.
64
Since 2000, most major universities in the United States and Canada have
incorporated supply chain codes of conduct in their trademark licensing
contracts with major athletic apparel firms such as Adidas and Nike.
65
Unlike corporate codes of conduct, which do not subject the brands
adopting them to any legal obligations, university codes of conduct impose
binding legal duties on the brands to ensure labor standards compliance by
their suppliers.
The contemporary initiative that most fully resembles the jobbers
agreement model is the Designated Suppliers Program (DSP). The Worker
Rights Consortium (WRC) and United Students Against Sweatshops
(USAS) proposed the DSP after concluding that the existing code of
conduct regime, while achieving some gains in certain areas in individual
factories, was not producing sustainable improvements, in part because the
ability and willingness of buyers to jump from factory to factory put heavy
price pressure on suppliers.
66
Under the proposed program, universities
would impose on all of their licensees, including companies like Nike,
Adidas, and Russell, an obligation to designate supplier facilities they
intend to use to fulfill the licensing contract; to sign long-term production
agreements of no less than three years with these designated suppliers; to
ensure demonstrable respect for freedom of association at each facility; and
to pay each supplier a price sufficient for the factory to pay workers a living
wage. Just as the jobbers agreements specified a dispute resolution
mechanism for adjudicating conflicts about what constitutes a price
sufficient to meet the contractor’s wage obligations, the DSP would
establish an arbitration process for settling price disagreements between the
contractor and its client about what constitutes a price sufficient to meet the
program’s living wage requirement. Where the DSP would differ from
64
. For background concerning the campus anti-sweatshop movement, see LIZA FEATHERSTONE,
UNITED STUDENTS AGAINST SWEATSHOPS: THE MAKING OF A MOVEMENT (2002).
65
. See, e.g., COLLEGIATE LICENSING COMPANY, Labor Code, 207.238.221.31/clcweb/publish
ing.nsf/content/labor+code.html (last visited Oct. 3, 2013).
66
. Scott Jaschik, CodesDon’tWork, INSIDE HIGHER ED. Sept. 28, 2006.
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
32 COMP.LABORLAW&POL’YJOURNAL [Vol. 35:1
jobbers agreements is that the latter included local labor directly in the
negotiation process with retailers, as well as with suppliers, whereas the
DSP calls for collective bargaining between the union and the factory, but
would not require a direct agreement between the union and the buyer—in
this case, the licensee.
Although more than forty universities publicly pledged to implement
the DSP by 2008, progress was stymied after brands objected that the
initiative would violate antitrust laws, prompting universities to pull back
on implementation. AfavorablereviewbytheU.S.DepartmentofJustice’s
antitrust division in late 2011 reignited some interest in the program.
However, its future remains very much uncertain.
67
While the DSP has yet to be implemented, some of its key objectives
are being realized on a pilot scale through the Alta Gracia project. Alta
Gracia is an initiative involving the WRC, the Dominican labor federation
FEDOTRAZONAS,andKnightsApparel,the country’slargestsupplierof
university-licensed clothing, such as t-shirts and sweatshirts displaying
collegiate logos. In 2010, Knights Apparel launched a new line of
university licensed apparel, Alta Gracia, manufactured at a factory in the
Dominican Republic with labor standards that far exceed the industry norm,
including payment of a living wage and full respect for the right to organize
and bargain collectively.
68
To enable these conditions, Knights committed,
in a legally enforceable agreement with the other parties, to (1) pay prices to
the factory that fully reflect the cost of complying with the living wage
requirement, and (2) maintain consistent production volumes so that the
living wage can be paid year-round.
69
Knights Apparel is paying workers
3.5timestheprevailingwageintheDominicanRepublic’sexport-oriented
apparelsectorandhasrespectedworkers’righttounionize.
70
Knights is absorbing the higher production costs that the Alta Gracia
project entails, having decided not to pass these on to its clients or to the
ultimate consumer. The company has succeeded in recovering some of
67
. Connor Jones, University to Implement Anti-Sweatshop Regulations, GEO. VOICE , Mar. 22,
2012 News Release, U.S. Dep’t Just., Justice Department Will Not Challenge Worker Rights
Consortium’s Designated Suppliers Program for Collegiate Apparel Dec. 16 , 2011, available at
http://www.justice.gov/atr/public/press_releases/2011/278340.htm.
68
. JOHN KLINE & EDWARD SOULE, GEORGETOWN UNIVERSITY REFLECTIVE ENGAGEMENT
INITIATIVE, ALTA GRACIA: WORK WITH A SALARIO DIGNO (2011); Steven Greenhouse, Factory Defies
Sweatshop Label, but Can It Thrive?, N.Y. TIMES, July 17, 2010.
69
. The obligations that the factory making Alta Gracia apparel—the Alta gracia Project factory—
and the brand buying that apparel and bringing it to consumers—Knights Apparel/Alta Gracia—have
agreed to meet are outlined on the following site WORKERS RIGHTS, http://www.workersrights.org/
verification/Obligations.asp (last visited Sept. 9, 2013). Note that Knights is currently effectively
operating the factory (though it does not own it), something that distinguishes the arrangement from a
typical buyer-contractor relationship in the apparel industry.
70
. WORKER RIGHTS CONSORTIUM, VERIFICATION REPORT RE LABOR RIGHTS COMPLIANCE AT
ALTAGRACIA PROJECT FACTORY (DOMINICAN REPUBLIC) 7, 14–15 (Jan. 31, 2013), available at
http://www.workersrights.org/verification/index.asp.
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
2013] JOINT LIABILITY IN GLOBAL SUPPLY CHAINS 33
these costs through lowered turnover and operating efficiencies, but
Knights Apparel has thus far struggled to make the Alta Gracia line
profitable because orders from university bookstores have not met
expectations, leaving the factory underutilized.
71
These challenges reveal
one important way in which the Alta Gracia experiment differs from
jobbers agreements: whereas the jobbers agreement imposed higher
standards on a group of employers, thus leveling the playing field and
putting no individual jobber at a competitive disadvantage for increasing
worker wages, Knights Apparel must compete with other university
licensees whose products are being manufactured by workers earning far
less.
In another major development, the Central General de Trabajadores
(CGT) of Honduras, recently concluded a series of agreements with the
apparel giant Fruit of the Loom and its subsidiary Russell Corporation.
Fruit of the Loom is the largest private employer in Honduras, with seven
major apparel and textile facilities in the country. In February 2008,
Russell shut down one of its facilities, Jerzees de Honduras, terminating
more than 1,200 workers; an investigation by the WRC completed in the
wake of the incident concluded that the decision to close the plant was
made, at least in part, to defeat an effort by workers to unionize.
72
The
closure of Jerzees de Honduras sparked a sophisticated, year-long campaign
led by workers, students activists, and labor advocates. Honduran workers
and student organizers spent months touring college campuses in the United
States and United Kingdom, garnering strong media attention and
ultimately persuading more than 100 universities to sever their licensing
agreements with Russell. Off campus, retailers such as Sports Authority
and Dick’s Sporting Goods who sell Russell products to noncollegiate
consumers came under considerable activist pressure and, in turn, joined
universities in pressuring Russell to resolve its labor dispute.
Such tactics succeeded in bringing Russell to the table, and in October
2009, following several months of negotiations with Honduran union
leaders,Russell’s parentcompany,Fruitof the Loom, ended the campaign
by making a number of unprecedented commitments: to open a new
factorytoreplacetheoneitshutteredtorehiretheclosedplant’s1,200plus
employeestopayeachoftheworkersaffectedbytheclosureninemonths’
back pay ($2.5 million in total); to not oppose future organizing drives (a
union“neutrality clause”) and, to provide the Honduran union CGT with
access to workers at all of its Honduran facilities. The agreement with CGT
also outlines a process of third party mediation and arbitration that makes
71
. See KLINE & SOULE, supra note 68, at 22.
72
. See WORKER RIGHTS CONSORTIUM, FINDINGS AND RECOMMENDATIONS REPORT RE JERZE ES
DE HONDURAS (Nov. 7, 2008), available at http://www.workersrights.org/Freports/Jerzees Choloma.asp.
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
34 COMP.LABORLAW&POL’YJOURNAL [Vol. 35:1
fulfillment of these commitments enforceable in U.S. court.
73
Fruit of the
Loom has since negotiated a collective bargaining agreement with workers
at the reopened Jerzees plant. This agreement, which provides for a 27%
raise,isexpectedtoserveasatemplateforcontractsatthecompany’sother
Honduran plants where workers have already begun to organize.
74
Unlike the Alta Gracia Project, in which Knights Apparel volunteered
to adopt high labor standards in order to capture the emerging market of
ethically sourced collegiate apparel, Fruit of the Loom was forced to
negotiate substantive labor standards with the workers of its Honduran
subsidiary by an international campaign. The form of leverage that brought
about this outcome differs markedly from the ones used to secure the
jobbers agreements of yesteryear. Whereas garment workers in the
twentieth century used general strikes to bring jobbers to the table, and were
aided at crucial moments by key government allies, in the Russell case it
was a cross-border alliance of Honduran workers and activists both on and
off college campuses that compelled Russell to transform its labor practices
by pressing universities and retailers to cut ties with the company. This
victory was concluded without any involvement of the Honduran state,
whichwasinthe grip of a coup d’état during much of this period, though
the post-coup Honduran state did play a belated role by recognizing the new
union and registering the subsequent collective bargaining agreement
negotiated between the union and the firm. The CGT-Fruit of the Loom
agreement therefore suggests that workers in international supply chains
can secure binding, substantive agreements on labor practices and wages
with major global companies, even absent massive media attention, such as
that which followed the Rana Plaza disaster in Bangladesh.
The CGT-Russell/Fruit of the Loom agreement is not strictly
analogous to a jobbers agreement because the factories in question are
subsidiaries of Fruit of the Loom, not subcontractors.
75
However, insofar as
Russell designs and markets its own brands and sits atop a supply chain that
includes independent suppliers as well as owned and operated factories, it is
much more like a jobber than a traditional apparel contractor. Moreover,
there is no obvious reason why the same set of factors that brought Russell
to the table to negotiate the price and conditions of labor in its own factories
could not be used to leverage a similar agreement from Russell and other
73
. Steven Greenhouse, Labor Fight Ends in Win for Students, N.T. TIMES, Nov. 17, 2009; Worker
Rights Consortium, Statement on the Agreement Between Russell Athletic/Fruit of the Loom, the
General Workers Confederation of Honduras, and the Sitrajerzeesh Union (Nov. 17, 2009).
74
. WORKER RIGHTS CONSORTIUM, SECOND PROGRESS REPORT RE IMPLEMENTATION OF RUSSELL
ATHLETIC/FRUIT OF THE LOOM REMEDIATION AGREEMENTS FOR OPERATIONS IN HONDURAS (Aug. 16,
2011), available at http://www.workersrights.org/Freports/JerzeesCholoma.asp.
75
. Fruit of the Loom did, however, commit to working with the CGT and WRC to address labor
rights violations throughout its Honduras supply chain, including subcontractors.
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
2013] JOINT LIABILITY IN GLOBAL SUPPLY CHAINS 35
global brands that coordinate subcontracting networks in countries like
Honduras.
Progress has also been made in recent years toward holding apparel
brands responsible as guarantors for the wages owed to the workers
employed by their suppliers. The problem of finding redress when
bankrupt employers abscond without paying their employees is a key
challenge in the globalized apparel sector. It is not uncommon for
contractors to shut down owing workers six months to a year in terminal
compensation.
76
In a series of recent cases, however, garment workers and
their advocates have succeeded in compelling major brands to use their own
resources to fulfill financial obligations to a supplier’s workers. In2010,
under university pressure and pursuant to an agreement reached with the
CGT in Honduras, Nike became the first major buyer to make significant
direct financial payments to workers in such a case. Nike contributed a
total of $1.54 million to workers denied legally mandated pay upon the
closure of two Nike contractors in Honduras, Hugger and Vision Tex.
77
The agreement also includes a commitment to ensure priority hiring of
Hugger and Vision Tex workers in other Nike supplier plants in Honduras,
enforceable through binding arbitration in U.S. court. Nike appears to have
since adopted guarantor responsibility as a matter of company policy. In
2011, the company made another $1 million in payments to workers at two
facilities in Indonesia, PT Kizone and PT Dong One, which closed without
paying workers due compensation.
78
In 2012, under heavy pressure from
universities, Adidas followed suit, agreeing to pay approximately $1.8
million to make workers at the PT Kizone facility whole.
79
Adidas resolved
the case after seventeen universities cut ties with the company.
80
The
University of Wisconsin-Madison had also sued Adidas for breach of its
licensing agreement, a case in which the Indonesian workers successfully
moved for intervenor status.
81
To cite another recent example, in early
76
. For a detailed discussion of one such case and a summary of other cases, see MAQUILA
SOLIDARITY NETWORK, EMERGENCY ASSISTANCE, REDRESSAND PREVENTION IN THE HERMOSA
MANUFACTURING CASE (June 2007), available at http://en.maquilasolidarity.org/sites/maquilasoli
darity.org/files/HermosaReportFinal_1.pdf.
77
. WORKER RIGHTS CONSORTIUM, VERIFICATION REPORT RE LABOR RIGHTS COMPLIANCE AT
ALTAGRACIA PROJECT FACTORY (Dec. 8 2011), available at http://www.workersrights.org/
verification/Obligations.asp [hereinafter CONSORTIUM, VERIFICATION REPORT]; Steven Greenhouse,
Nike Agrees to Help Laid-Off Workers in Honduras, N.Y. TIMES, July 26, 2010.
78
. CONSORTIUM, VERIFICATION REPORT, supra note 77.
79
. WORKER RIGHTS CONSORTIUM, ASSESSMENT RE PT KIZONE (INDONESIA) (Jan. 18, 2012),
available at http://www.workersrights.org/Freports/PT%20Kizone.asp; Aaron Guggenheim, Adidas
Settles with Workers at PT Kizone, MICH. DAILY, Apr. 25, 2013, http://www.michigandaily.com/
blog/wire/adidas-settles-workers-pt-kizone.
80
. See supra note 79.
81
. Karen Herzog, UW-Madison Drops Suit Against Adidas Over Treatment of Workers,
MILWAUKEE-WIS. J. SENTINEL (June 3, 2013), http://www.jsonline.com/blogs/news/209952971.html;
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
36 COMP.LABORLAW&POL’YJOURNAL [Vol. 35:1
2013, H&M and Wal-Mart agreed to contribute $205,000 to workers of the
Kingsland factory in Cambodia, which shut down without paying them in
late 2012.
82
Finally, Adidas is now the target of the first transnational campaign
with the explicit goal of securing what would essentially be a jobbers
agreement—described by its proponents as a “buyer responsibility
agreement”—on a global scale. This effort was launched by a recently
formed coalition of garment worker unions in Latin America and Asia
called the International Union League for Brand Responsibility.
83
The
body’s coordinating committee includes unions from Bangladesh,
Cambodia, El Salvador, Honduras, India, Indonesia, Dominican Republic,
and Nicaragua, including many of the unions—such as CGT in Honduras—
that have waged among the most sophisticated and successful campaigns
calling upon brands to intervene to remedy labor violations.
84
The Adidas
campaign is being supported by United Students Against Sweatshops,
among other labor rights groups in the United States and Europe.
85
In
Spring 2013, activists launched a “Badidas” campaign, arranging for
workers from Adidas contract factories to speak at college campuses about
thelaborviolationstheyhaveexperiencedandprotestingatAdidas’annual
shareholder meeting in Germany.
86
Just as in the jobbers agreements of
years past, the unions and their supporters are calling upon Adidas to
participatein triangularnegotiationswithrepresentativesofthecompany’s
supplier factories and the unions representing workers at those factories.
Summing up, while none of the cases surveyed here approach the
comprehensive scope of jobbers agreements, they represent clear movement
in the direction of legally enforceable agreements with major brands. These
initiatives impose real obligations on buyers, and in this sense far surpass
the voluntary nature of the standard CSR code of conduct model. Most
crucially, like the jobbers agreements, each of the concluded agreements—
all reached in a four years period between 2009 and 2013—require lead
firms to contribute financially toward the elimination of sweatshop
conditions in the global supply chains they sit atop. There will be
Noah Goetzel, Court Says Former Adidas Workers Can Testify in Suit, BADGER HERALD (Jan. 24, 213),
http://badgerherald.com/news/2013/01/24/court_says_former_ad.php#.Ucznfz5tXyc.
82
. CLEAN CLOTHE S CAMPAIGN, Historic Win for Cambodian Workers (Mar. 4, 2013),
http://www.cleanclothes.org/news/kingsland-victory.
83
. Information on the campaign can be found at the International Union League for Brand
Responsibility’s website, ADIDAS WORKERS UNITE: INTERNATIONAL UNION LEAGUE FOR BRAND
RESPONSIBILITY, http://adidasworkers.org (last visited Sept. 9, 2013).
84
. Id. For example, one of the leaders of the body’s coordinating committee is Evangelina
Argueta, an intrepid organizer with the CGT in Honduras who led the campaigns discussed in this
Section involving Fruit of the Loom and Nike.
85
. United Students Against Sweatshops have launched a campaignstyled“Badidas.” Information
is available at BADIDAS: ALL IN AGAINST SWEATSHOPS, http://badidas.com (last visited Sept. 9, 2013).
86
. See id.
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
2013] JOINT LIABILITY IN GLOBAL SUPPLY CHAINS 37
considerable opportunities to build upon these precedents in the years to
come.
B. Statutory and Common Law Approaches to Joint Liability
We have focused in this Article on contractual approaches for
establishing joint liability in global supply chains. Although it has received
relatively little academic attention, this approach has an impressive track
record of success, which, we anticipate—given the developments reviewed
in the preceding Section—will be an area of important experimentation
going forward. However, lead firm liability for labor rights violations
committed by subcontractors may, of course, also be imposed without the
lead firm’s assent. This might involve private enforcement of corporate
codes of conduct through creative litigation on behalf of employee,
consumer, or investor plaintiffs, as Professor Jim Brudney has recently
explored in this Journal.
87
It may also involve the application of vicarious
liability and joint employment doctrines, as Professor Alan Hyde has
proposed.
88
While it is beyond the scope of this Article to thoroughly
examine these possibilities—which have been the subject of thoughtful
analysis and debate in the domestic U.S. context
89
—we take a brief detour
in this Section from our primary argument to cite several noteworthy
efforts.
One strategy to place a degree of responsibility on lead firms for
ensuring compliance in international supply chains is borrowed from the
FLSA’s “hot goods” provision. The provision empowers the federal
Department of Labor (DOL)—though not private individuals—to object to
theshipmentofgoodsmadeinviolationoftheFLSA’sminimumwageand
overtime requirements.
90
In such cases, federal district courts have
authority to enjoin shipment of such hot goods until violations are
remediated.
91
Since buyers and retailers have an interest in the release of
their embargoed goods (and avoiding the embargo of their goods in the first
87
. James J. Brudney, Envisioning Enforcement of Freedom of Association Standards in Corporate
Codes: A Journey for Sinbad or Sisyphus?, 33 COMP. LAB. L. & POL’Y J. 555 (2012).
88
. Alan Hyde, Legal Responsibility for Labour Conditions Down the Production Chain, in
CHALLENGING THE LEGAL BOUNDARIES OF WORK REGULATIONS 83 (Judy Fudge, Shae McCrystal &
Kamala Sankaran eds., 2012).
89
. See, e.g., Bruce Goldstein, Marc Linder, Laurence E. Norton & Catherine K. Ruckelshaus,
Enforcing Fair Labor Standards in the Modern American Sweatshop: Rediscovering the Statutory
Definition of Employment, 46 UCLA L. REV. 983 (1999); Brishen Rogers, Toward Third-Party Liability
for Wage Theft, 31 BERKELEY J. EMP. & LAB. L. 1 (2010).
90
. 29 U.S.C. § 215(a)(1) (2012) (“[I]t shall be unlawful for any person to transport, offer for
transport, ship, deliver, or sell in commerce, or to ship, deliver, or sell with knowledge that shipment or
delivery or sale thereof in commerce is intended any goods in the production of which any employee
was employed in violation of section 206 [statutory minimum wage] or 207 [maximum hours] of this
title[.]”).
91
. 29 U.S.C. § 217 (2012).
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
38 COMP.LABORLAW&POL’YJOURNAL [Vol. 35:1
place), this approach can be a potent means for compelling apparel buyers
to take responsibility for labor conditions in their supply networks.
During the late 1990s and early 2000s, the DOL creatively leveraged
hot goods injunctions to compel apparel buyers to participate in a DOL
program in which the buyers agreed to monitor (typically through contract
monitors) and assure compliance at their contract facilities. Although the
achievements were modest compared to the transformation achieved during
the era of the jobbers agreements, the program did prove effective in
reducing minimum wage violation rates in California and New York.
92
Of
note, buyers participating in the program were required to ensure that the
prices they paid to suppliers were sufficient to enable compliance with
FLSA’sstandards.
93
While it is not clear how or to what extent the DOL
enforced the pricing provision, an analysis of the program found that that
contractor’s ability to negotiate price terms was the single best predictor of
compliance, more important even than the most effective monitoring of
suppliers.
94
While the FLSA applies only to goods manufactured domestically,
95
there has been an effort to create a hot goods provision for goods produced
overseas and imported into the United States. Introduced in 2006, the
Decent Working Conditions and Fair Compensation Act would “prohibit
the import, export, and sale of goods made with sweatshop labor.”
96
At its
maximum support, the 2007 version of the bill was cosponsored by 167
Democrats and 7 Republicans in the House of Representatives.
97
While
political winds would have to change significantly for a law along these
lines to be enacted, it is not inconceivable.
Additionally, as Professor Hyde has noted, recent years have seen
significant experimentation at the state level with regard to supply chain
liability, with California leading the way. Following a decade-long
92
. See ESBENSHADE, supra note 13, at 85; WEIL, supra note 13, at 32.
93
. The initial agreement underlying the program required signatory companies to ensure the
economic feasibility of the price terms that are involved [in each purchase of apparel
goods from the contractor], in light of the compliance with the [Fair Labor Standards Act]
and the [Employer Compliance Program] required of the Contractor and in light of the
calculations and expectations of the parties to the purchase.
DEP’T LAB., Augmented Compliance Program Agreement (ACPA), § 3(a)(iii) (Dec. 29, 1994); see also
ESBENSHADE, supra note 13, at 105.
94
. ESBENSHADE, supra note 13, at 107.
95
. 29 U.S.C. § 213 (2012) (noting that the minimum wage, overtime, and homework provisions of
the FLSA “shall not apply with respect to any employee whose services during the workweek are
performed in a workplace within a foreign country”).
96
. See, e.g., H.R. 5635, 109th Cong. (1996). Sweatshop goods are defined to include products
“mined,produced,ormanufacturedwhollyorinpart”in violation of workers’ rights of association and
collective bargaining, using forced or compulsory labor or child labor, or in violation of domestic laws
with respect to minimum wages, hours of work, and occupational safety and health. H.R.1992, 110th
Cong. § 3 (2007).
97
. H.R. 1992 110th Cong. (2007). The Senate version had twenty-five cosponsors, all Democrats.
S. 367, 110th Cong. (2007).
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
2013] JOINT LIABILITY IN GLOBAL SUPPLY CHAINS 39
campaign by worker advocates, California now holds apparel brands strictly
liable as guarantors for wages owed by their domestic contractors.
98
New
York also holds garment manufactures liable for wages owed by domestic
contractors, though only where the manufacturer knew or should have
knownofthecontractor’sviolations.
99
These laws correspond generally to
the guarantor provisions of the jobbers agreements discussed above, but do
not involve the pricing or stable order provisions. As might be expected,
the general view of worker advocates is that they have made a difference in
getting workers paid in the case of fly-by-night contractors, but they have
not led to a significant reduction in overall non-compliance rates.
100
California has also made it unlawful to enter into a contract with a
contractor in a range of industries (including garment, construction, farm
labor, janitorial, security, and warehouse) with actual or constructive
knowledge that the price paid is insufficient to enable compliance with
local, state, or federal laws.
101
This law appears to have been little used.
102
Finally, under California’s Transparency in Supply Chains law, passedi n
2010, companies with annual worldwide gross receipts of more than $100
million are required to publicly disclose steps they are taking to eradicate
slavery and trafficking from their global supply chains, including auditing
suppliers for compliance with company policies.
103
Thus far, the law
appears to be encouraging the development of codes of conduct and
monitoring along the lines of what has developed in the apparel sector.
104
Although the developments reviewed above outline a number of
potentially promising avenues for pursuing buyer liability via statute as
opposed to contract, it should be noted that, with the exception of
California’s transparency law, none of the existing measures apply to
subcontractors operating overseas. Nor does the NLRA or, as noted, the
98
. CAL. LAB. CODE § 2673.1 (2013); Katie Quan, Evolving Relations in the Women’sApparel
Industry, UC BERKELEY CENTER FOR LABOR RELATIONS AND EDUCATION (2007) (unpublished article).
99
. N.Y. LAB. LAW § 345-a(1) (Consol. 2012).
100
. A study conducted six years after the California law was passed revealed that although it led to
someprogress in workers obtaining owedwages, the law’s full promisewas undermined by the state
labor department’s failure to effectively identify guarantors, conduct investigations, and pursue
sanctions and damages. The law does not contain a private right of action. MARCI SEVILLE,
REINFORCING THE SEAMS: GUARANTEEING THE PROMISE OF CALIFORNIA’S LANDMARK ANTI-
SEATSHOP LAW—AN EV ALUATION OF ASSEMBLY BILL 633 SIX YEARS LATER (Women’sEmployment
Rights Clinic, Sept. 1, 2005), available at http://digitalcommons.law.ggu.edu/cgi/viewcontent.
cgi?article=1003&context=werc.
101
. CAL. LAB. CODE § 2810(a) (2013).
102
. The only published or unpublished case involving the law is Castillo v. Toll Bros., Inc., 197
Cal. App. 4th 1172 (Cal. Ct. App. 2011).
103
. CAL. CIV. CODE § 1714.43 (2013). There is no private right of action and the only remedy is
that the company may be enjoined to comply with the disclosure obligation. § 1714.43(d).
104
. See, e.g., HEWLETT-PACKARD DEVELOPMENT COMPANY, CA Transparency in Supply Chains
Act 2010, http://www8.hp.com/us/en/hp-information/global-citizenship/society/california-transparen cy-
in-supply-chains-act-of-2010.html (last visited Sept. 9, 2013).
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
40 COMP.LABORLAW&POL’YJOURNAL [Vol. 35:1
FLSA.
105
The extent to which joint liability laws might be written to apply
extraterritorially is a fruitful topic for further research. One of the
advantages of the contractual mechanism we focus on here is that it would
not require the writing of new legislation or creative litigation under
existing statutes. As with any contract, a global jobbers, or “buyer
responsibility,” agreement could be enforced in any forum the parties
choose where jurisdictioncanbe exercised and withtheparties’choiceof
law. For example, such an agreement could be subject to binding
arbitration,enforceablein the domesticcourt of thesignatoryfirm’s home
country, as is the case with the Bangladesh worker safety accord and CGT
agreements with Fruit of the Loom and Nike discussed above.
VI. CONCLUSION
In recent years, the sweatshop debate has become characterized by a
new consensus about the failure of extant approaches to significantly
improve labor compliance in global supply chains for apparel products.
Although numerous approaches have been proposed, most of these fail to
address what the bulk of available data suggests is the principle root cause
of pervasive labor violations: the buying practices of global brands and
retailers. More effective states, more empowered labor inspectors, and
more enlightened consumers would almost certainly benefit garment
workers, but our view is that none of these would prevent the pressure on
laborstandardsand workers’rights that is created when a large number of
suppliers (mostly in developing countries) are constantly trying to secure
orders from a much smaller set of powerful buyers (primarily in developed
countries) by manufacturing goods more quickly and more cheaply than
anyone else.
Our emphasis on jobbers agreements is not meant to discount the role
of domestic and international labor law enforcement in the pursuit of
worker rights. Indeed, we expect that jobbers/buyer agreements will be
much more effective where they are buttressed by well-functioning
domestic labor law and institutions.
106
Because the contractual approach to
joint liability that we outline here depends on worker self-organization, any
program that enhances the will and capacity of governments to protect
workers’righttoorganizeandbargaincollectivelywouldcomplementthis
model. International legal mechanisms, particularly the inclusion of labor
standards in free trade agreements, also have a key role to play.
107
But
105
. See McCulloch v. Sociedad Nacional de Marineros de Honduras, 372 U.S. 10 (1963) (holding
that the NLRA does not apply to employees stationed overseas); 29 U.S.C.§ 213 (2012).
106
. See, e.g., Locke, Qin & Brause, supra note 6.
107
. See Sandra Polaski, Protecting Labor Rights Through Trade Agreements: An Analytical
Guide, 10 J. INT’L LAB. & POL’Y 13 (2003). However, research shows that such mechanisms are most
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
2013] JOINT LIABILITY IN GLOBAL SUPPLY CHAINS 41
domestic laws and international labor standards in and of themselves do not
ensure fair pricing mechanisms and stable contracts, elements of the jobbers
agreement model that were critical in the battle against sweatshops in the
United States. Pursuing jobbers/buyer responsibility agreements is a
strategic decision, one based on targeting the main sources of power and
control in buyer-driven supply chains—namely, brands and retailers. It also
reflects a labor strategy that opts for international solidarity over economic
nationalism.
In pursuing this strategy, workers and their advocates today confront a
very different economic and political landscape than the one that spawned
jobbers agreements. The ILGWU’s efforts to secure joint liability for
domestic garment workers were critically enabled by a federal government
that supported the rights of workers to organize and bargain collectively.
Garment workers in New York City also benefitted from a geographically
concentrated industry susceptible to labor pressure via strikes. In contrast,
today apparel production is dispersed—and easily transferred—among
dozens of countries. Furthermore, the governments of some of these
apparel-exporting nations are actively hostile toward unions and other
forms of labor advocacy, in part because they fear that higher wages and
better working conditions may negatively impact their comparative
advantage in low production costs. Given these dramatic differences,
jobbers agreements might be seen as a quaint but irrelevant relic of a
bygone era when capital had less mobility and labor had more power.
But there are aspects of the constellation of forces in the current era
that make something like jobbers agreements, at least in a limited form,
plausible. It is true that global garment workers are weak in terms of
traditional sources of structural and associational power,
108
but they and the
organizations that represent them are central participants in transnational
activist networks that include students, NGOs, and international unions.
109
effective when combined with strong cross-border organizing. See HENRY J. FRUNDT, TRADE
CONDITIONS AND LABOR RIGHTS: U.S. INITIATIVES, DOMINICAN AND CENTRAL AMERICAN RE SPONSES
(1998); Lance Compa, NAFTA’s Labor Side Agreement and International Labor Solidarity, 33
ANTIPODE 451 (2002). The Obama Administration’s July 2013 decision to suspend certain of
Bangladesh’s trade privileges under the General System of Preferences, as a means of pressing the
Bangladeshi government to accept reforms such as eliminating its ban on unionization in export
processing zones, is a welcome recent example of this approach (albeit a largely symbolic one, since the
country’sgarment exports, which account for 80% of total exports, do not receive GSP preferences).
Steven Greenhouse, Obama to Suspend Trade Privileges with Bangladesh, N.Y. TIMES, June 27, 2013,
at B1.
108
. BEVERLY SILVER, FORCES OF LABOR: WORKERS’ MOVEMENTS AND GLOBALIZATION SINCE
1870 (2003).
109
. See Mark Anner, Workers’Power inGlobal ValueChains: Fighting Sweatshop Practices at
Russell, Nike and Knights Apparel, in TRANSNATIONAL TRADE UNIONISM: NEW CAPABILITIES AND
PROSPECTS (Peter Fairbrother, Marc-Antonin Hennebert & Christian Lévesque eds., forthcoming 2013);
Jennifer Bair & Florence Palpacuer, From Varieties of Capitalism to Varieties of Activism: The Anti-
Sweatshop Movement in Comparative Perspective, 59 SOC. PROBS. 522 (2012).
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
42 COMP.LABORLAW&POL’YJOURNAL [Vol. 35:1
These labor-activist alliances have used consolidation at the top of the
apparel industry to their advantage by targeting brands, institutional buyers
(universities) and retail chains to bring about major agreements with
powerful corporations including Nike, Fruit of the Loom/Russell, and
Adidas.
110
Largely in response to these activities, lead firms have attempted to
protect their reputations by putting millions of dollars into CSR initiatives.
Yet, as the limitations of these initiatives are increasingly exposed through
disasters such as the one at Rana Plaza, and with the growing capacity of
the transnational anti-sweatshop movement, lead firms will be pressured to
accept bigger and bolder initiatives. This might just create the perfect storm
for a new approach, one that brings lead firms to the bargaining table and
forces them to engage in negotiations that reflect the spirit of the jobbers
agreements. The segments with the best prospects are those which have
been the focus of attention in the debates on global labor standards to date:
university licensed apparel, which is at the center of activist efforts to
reform the contracting system, government-procured apparel, which is
increasingly subject to anti-sweatshop ordinances,
111
and perhaps most
important, reputation-sensitive sportswear and fashion companies (the
Nikes,Adidas’s, H&M’s,and Inditex’softhe world), whichhavealready
committed themselves in recent agreements to various degrees of joint
liability.
Achievingaversionofjoint liability in today’sglobalsubcontracting
networks is a daunting task, but the odds may be less long than skeptics
contend. As the breakthrough accord on worker safety in Bangladesh and
the Adidas campaign suggest, the global labor movement appears to be
coalescing around this approach. Ultimately, the prospects for success
depend on the extent of power wielded by apparel workers and their allies
in the developed world—students, consumers, nonprofit organizations, and
unions—as well as the sustained attention of journalists and the beneficence
of forward thinking industry leaders.
112
But based on existing evidence,
including the contemporary case studies described in this Article, there is
110
. See ANNER, supra note 5; Ralph Armbruster-Sandoval, Workers of the World Unite? The
Contemporary Anti-Sweatshop Movement and the Struggle for Social Justice in the Americas, 32 WORK
& OCCUPATIONS 464 (2005).
111
. Bjorn Claeson, Are Social Labels Symbols of Resistance? A Case for Sweatshop-Free
Procurement in the U.S. Public Sector, in WORKERS’ RIGHTS AND LABOR COMPLIANCE IN GLOBAL
SUPPLY CHAINS: IS A SOCIAL LABEL THE ANSWER? (Jennifer Bair, Marsha Dickson & Doug Miller,
eds., forthcoming 2013).
112
. Reporting on the Bangladesh Rana Plaza factory collapse and sweatshop conditions were
considered so outstanding in shaping public debates by two prominent social scientists that they
suggested the New York Times receive a Pulitzer Prize. See Peter Dreier & Richard Appelbaum, New
York Times Deserves a Pulitzer Prize for Reporting on Bangladesh Sweatshop, HUFFINGTON POST, May
13, 2013, http://www.huffingtonpost.com/peter-dreier/new-york-times-bangladesh-sweatshops_b_331
7561.html.
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
2013] JOINT LIABILITY IN GLOBAL SUPPLY CHAINS 43
no reason to rule out such an outcome. Indeed, we believe there is reason
for guarded optimism, as long as our efforts to eradicate the century-old
scourge of sweatshops begin with a correct diagnosis of its cause.
ANNER ET AL 35-1 FINAL (DO NOT DELETE) 10/4/2013 6:09 PM
44 COMP.LABORLAW&POL’YJOURNAL [Vol. 35:1
