Review of International Political Economy

ISSN: 0969-2290 (Print) 1466-4526 (Online) Journal homepage: www.tandfonline.com/journals/rrip20

Squeezing workers’ rights in global supply chains:

purchasing practices in the Bangladesh garment
export sector in comparative perspective

Mark Anner

To cite this article: Mark Anner (2020) Squeezing workers’ rights in global supply chains:
purchasing practices in the Bangladesh garment export sector in comparative perspective,
Review of International Political Economy, 27:2, 320-347, DOI: 10.1080/09692290.2019.1625426

To link to this article: https://doi.org/10.1080/09692290.2019.1625426

Published online: 27 Jun 2019.

Submit your article to this journal

Article views: 11140

View related articles

View Crossmark data

Citing articles: 83 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=rrip20
REVIEW OF INTERNATIONAL POLITICAL ECONOMY
2020, VOL. 27, NO. 2, 320–347
https://doi.org/10.1080/09692290.2019.1625426

Squeezing workers’ rights in global supply chains:

purchasing practices in the Bangladesh
garment export sector in comparative perspective
Mark Anner
The Pennsylvania State University, University Park, PA, USA

ABSTRACT
Workers’ rights violations have been pervasive in many global supply chains. In the
apparel sector, production workers often face precarious working conditions, includ-
ing persistently low pay, excessive and often forced overtime, unsafe buildings, and
repression of their right to form unions and bargain collectively. This article
explores how purchasing practices of lead firms adversely affect working conditions
and workers’ rights in supplier factories. It attributes these trends to a price squeeze
and a sourcing squeeze in which lead firms pay increasing lower prices to suppliers
while also imposing short lead times and high order volatility. To test this argu-
ment, trade data of apparel imports to the United States and the European Union
are explored. The article then turns to original surveys of Bangladesh supplier facto-
ries and workers carried out in 2016 and 2017. The final section of this paper exam-
ines the impact of the squeeze on working conditions and workers’ rights using the
Labour Rights Indicators.

KEYWORDS
Workers’ rights; global supply chains; global value chains; purchasing practices; apparel; garments;
Bangladesh; living wages; building safety; Rana Plaza

Introduction
The global value chain/production network literature assumes some degree of
power asymmetry in supply chains (Dicken, 2015; Gereffi, Korzeniewicz, &
Korzeniewicz, 1994). This imbalance is perhaps most noticeable in buyer-driven
apparel global supply chains (GSCs), where retailers and brands (‘buyers’ or ‘lead
firms’) consolidated their power through mergers, acquisitions and market concen-
tration (Abernathy, Dunlop, Hammond, & Weil, 1999; Bonacich & Appelbaum,
2000). At the same time, there has been a dramatic dispersion of apparel suppliers
across developing countries (Gereffi & Frederick, 2010). The combination of con-
solidated buyers and dispersed suppliers gives buyers an advantage when setting
prices and other terms of production contracts. Yet, some scholars have argued
that the shift from an assembly model of production through export processing

CONTACT Mark Anner [email protected] The Pennsylvania State University, University Park, PA
16801, USA.
ß 2019 Informa UK Limited, trading as Taylor & Francis Group
 REVIEW OF INTERNATIONAL POLITICAL ECONOMY 321

zones to full-package production in which suppliers source fabric and other inputs
resulted in a decline in power imbalance (Gereffi, Humphrey, & Sturgeon, 2005).
In contrast, this article argues that in the apparel sector the power imbalance in
GSCs has increased since the early 2000s. This is because state policies (including
shifting trade regimes), technology and pressure created by growing capital markets
(stock market and private equity firms) exacerbated power imbalances. The squeeze
on supplier firms does not lessen because they have begun to source fabric rather
than rely on buyers to source the fabric for them. Under conditions of supply
chain oligopsony (lead firm power concentration), lead firms can also squeeze sup-
plier firms on how, where and when they source their fabric and at what cost. All
else equal, lead firms have been able to reduce the real dollar price paid for apparel
made by suppliers. This ‘price squeeze’ impacts workers in the form of low wages
and increased work intensity.
Brands compete for market share in order to maintain or grow their revenue.
This includes pursuing continuous sales growth with low retail prices and ever-
changing products and shorter fashion seasons. This ‘fast fashion’ marketing model
requires increasingly shorter production lead times (Anguelov, 2016; Taplin, 2014).
Even non-fast fashion firms strive to increase their ‘speed to market’ in order to
better manage product inventory; the shorter the period from product design to
product sale, the better firms are able to judge consumer trends, which reduces the
likelihood of unsold inventory and costly markdowns. Lead firms also can use their
supply chain power to modify order volume and increase styles, which creates fur-
ther stress on suppliers (Locke, 2013). This ‘sourcing squeeze’ on suppliers – which
interacts with the pricing squeeze since lower prices help to increase inventory
turnover – impacts workers in the form of chronic and forced overtime, and
unauthorized outsourcing to unsafe factories. Indeed, three of the four worst fac-
tory disasters in the history of industrial garment production have taken place since
2012 (Nova & Wegemer, 2016).
To explore these arguments, this article employs mixed methods. It begins by
analyzing prices paid for imported apparel by U.S. and European buyers. These
findings are complemented by field research and an original survey of 223 factory
owners in Bangladesh and 188 factory workers. The impact of sourcing practices
on workers’ rights also is explored using structured interviews with workers and
the Labour Rights Indicators (LRIs). The findings provide support for my argu-
ment of a price and a sourcing squeeze, and they provide evidence that the buyer
squeeze on suppliers contributes to a supplier squeeze on workers.

Sourcing squeeze and workers’ rights

One of the original premises of the GSC1 literature is that there are power imbal-
ances across different nodes of chains (Gereffi et al., 1994; Henderson, Dicken,
Hess, Coe, & Yeung, 2002). Bonacich and Appelbaum (2000) find that high levels
of competitiveness among supplier factories and mergers among retailers increased
the power of retailers over suppliers. Abernathy et al. (1999) come to similar con-
clusions about the trend toward retailer consolidation, but for different reasons.
They argue that technology enables the largest firms to leverage lean manufacturing
to grow their market share, and forecast a continued trend for apparel retail indus-
try concentration. Hamilton and Petrovic (2011) document enormous consolidation
322 M. ANNER

of power among major retailers, as does Lichtenstein (2009), who focuses on the
rise of Wal-Mart. He argues that the power of this one corporation increased to
the point that it was able, ‘to “legislate” key components of American social and
industrial power’. Similarly, Phillips (2017) argues lead firms have consolidated not
only economic power, but also social and political power.
International institutions and mainstream media are coming to similar conclu-
sions. The International Labour Organization observes, ‘the sheer volume of its
purchases grants [buyers] substantial bargaining power in an asymmetrical market
relationship where a buyer can negotiate prices and specify, what, how, where, and
by whom the goods it sells are purchased’ (ILO 2016, p. 11). The Wall Street
Journal (WSJ) reports, ‘Declining enforcement of antitrust rules has led to bigger
mergers, less competition and higher profits.’2 Roni Michaely observes in the WSJ:
‘If you want to compete with Google or Amazon [ … ] you’ll have to invest not
just billions, but tens of billions of dollars.’3 The New York Times, in reference to
Amazon and other ‘superstar firms,’ finds: ‘Two of the most important economic
facts of the last few decades are that more industries are being dominated by a
handful of extraordinarily successful companies and that wages, inflation and
growth have remained stubbornly low. Many of the world’s most powerful eco-
nomic policymakers are now taking seriously the possibility that the first of those
facts is a cause of the second.’4
There is a debate in the literature regarding the degree of this power imbalance
and the direction of change. As Locke, Amengual, and Mangla (2009, p. 328)
observe, ‘Power relations among the key actors in the supply chains are far from
unidirectional or unambiguous.’ Gereffi, Humphrey, and Sturgeon (2005) argue
power relations are linked to information complexity, codification of transactions
and supplier capabilities. They develop five topologies of global value chains. For
our purposes, two of these topologies are particularly relevant, ‘captive value
chains’ and ‘relational value chains.’5 Captive value chains are characterized by low
supplier competence in the face of complex products and specifications. As a result,
in these chains we see the dominance of lead firms (Gereffi et al., 2005, p. 86).
Captive value chains thus represent ‘buyer-driven global commodity chains’ in
Gereffi’s (1994) original formation, with their high degree of power asymmetry.
In relational value chains, production specifications cannot be codified, transac-
tions are complex, and supplier capabilities are high. This results in the need for
the continuous exchange of complex information, and hence mutual dependence
(Gereffi et al., 2005, p. 86). Due to this mutual dependence and high supplier capa-
bilities, these scholars argue that in relational value chains power asymmetry is
moderate. What is notable is that, in the global apparel industry, they specifically
find, ‘the expansion and growing capabilities of its global supply-base have permit-
ted it to move rapidly from captive to more complex relational value chains over
the span of just a few decades’ (Gereffi et al., 2005, p. 91). That is, they see a
decline in power asymmetry in apparel global value chains. This decline, they sug-
gest, is the result of supplier factories moving from assembling garments through
an export processing zone model to full package production, which involves greater
capabilities on the part of supplier firms and more complex forms of coordination
between suppliers and lead firms. Facilitating this process was the 1995–2004 phase
out of the Multi-Fiber Arrangement (MFA) in accordance with the World Trade
Organization’s Agreement on Textiles and Clothing. The MFA had relegated to
 REVIEW OF INTERNATIONAL POLITICAL ECONOMY 323

developing countries the apparel assembly function. Hence, no longer incumbered

by the MFA, developing countries were free to develop other capabilities, such as
interpreting designs, sourcing inputs (notably fabric) and coordinating logistics
(Gereffi et al., 2005).
Having indicated that power imbalances were lessening in apparel GSCs opens
the questions as to whether economic upgrading is conducive to social upgrading.
For Barrientos, Gereffi, and Rossi (2011), economic upgrading is understood as
moving to higher value-added activities. This may be done in one of four ways:
through process upgrading in which changes in the production process result in
greater efficiencies; product upgrading in which more advanced product types are
introduced; functional upgrading where firms incorporate higher value-added tasks;
and chain upgrading where firms move to new industries or product markets
(Barrientos, Gereffi, & Rossi, pp. 323–24). Social upgrading draws on the
International Labour Organization’s definition of ‘decent work,’ which includes
quality employment, fundamental freedoms and rights at work, social protection
and social dialogue (ILO, 2008).
Barrientos et al. are careful to indicate that economic upgrading is not a suffi-
cient condition for social upgrading. Indeed, even under the best of circumstances,
only a sub-group of permanent workers at the bottom of supply chains may benefit
from economic upgrading. To increase the likelihood of social upgrading, they sug-
gest interventions by trade unions, states, multi-stakeholder codes of labor practices
and multinational initiatives to promote both economic and social upgrading, a so-
called ‘win-win’ scenario (Barrientos et al., 2011, pp. 336–337). I will argue that the
dynamics of economic upgrading and social upgrading must be understood within
the broader dynamics of international political economy and local institutional con-
texts. As Henderson and his collaborators (2002, p. 450) observe, power is not only
exercised by lead firms, but also by state and inter-state institutions, which have
the capacity to influence the investment and other decisions of lead companies and
other firms integrated into GSCs.
The phase out of the MFA agreement is only one of several important develop-
ments over the last two decades in the global apparel industry. The entry of China
and Vietnam into the WTO significantly increased cost competition among suppli-
ers. And technological developments facilitated greater control by lead firms over
their supply chain logistics. This has given market advantage to the largest of mass
merchandizers and firms specializing in electronic commerce (Kotha & Basu,
2011). In addition, growing pressure from investors on lead firms to reduce costs
and increase margins has created increased pressure on lead firms to squeeze their
suppliers on costs. Given these dynamics of the international political economy of
the apparel industry – as well as adverse domestic contexts (e.g. weak labor laws
and/or poor enforcement) – I argue that there have been increases in power asym-
metry in apparel GSCs both between lead firms and suppliers and between supplier
factories and their workers over the past 20 years.
Following the establishment of permanent trade relations between the US and
China in 2000, China entered the WTO on December 11, 2001. Vietnam entered
the WTO on January 11, 2007. Thus, as the MFA phased out, two countries with
very large workforces were now competing with traditional apparel exporting coun-
tries.6 This provided enormous leverage to lead firms, which could, with greater
force than a decade earlier, leverage suppliers to come down in price or face the
324 M. ANNER

risk of production moving elsewhere. Suppliers in some countries were not able to
reduce production costs, and closed down, notably in Mexico and the Dominican
Republic. Other countries squeezed down on costs and managed to grow their
share of the global market. This includes India, Indonesia, Cambodia, and particu-
larly Bangladesh, which, since 2009, has been the second largest apparel exporter in
the world.
Technology also played a role. Mass merchandizers using bar codes and check-
out scanning in their retail stores to track stock keeping units were able to process
sales data and improve inventory management and supply chain logistics
(Abernathy et al., 1999). When combined with high-quality software, faster com-
puters and accurate computer-to-computer exchange of documents in a standard
electronic format – a process known as electronic data interchange – it quickly
became much more feasible for US retailers to manage supply chains in geograph-
ically dispersed locations. And companies based on electronic commerce, such as
Amazon, began using digital technology to manage relations with an extremely
broad, global range of suppliers and a system of ‘virtual inventory’ and two-day
delivery times to dramatically capture market share (Kotha & Basu, 2011).
Finally, the growing role of finance capital has further contributed to supply
chain power asymmetries (Weil, 2014). Since capital markets are so competitive,
CEOs and managers in lead firms are under constant pressure to grow share values
or risk being replaced. This pressure is present whether it is public or private cap-
ital. The rise of 401(k) accounts and individual retirement accounts inserted con-
siderable capital into the market, mostly via mutual funds. Weil argues that money
flowing into firms from these mutual funds is ‘‘impatient’ and moves frequently in
search of better returns for a given level of risk’ (Weil, 2014, p. 46). Milberg also
finds that ‘the financialization of non-financial corporations’ has encouraged
restructuring of production through the growth of GSCs (Milberg, 2008). Private
equity firms also play a role in squeezing not only firms but also workers
(Appelbaum & Batt, 2014) as they leverage lead firms who, in turn, are pushed to
demand increased margins from their GSCs (Weil, 2014). Those lead firms that do
not deliver may be forced out of business. Those that remain consolidate their
market share. In 2018, Amazon surpassed Walmart as the top apparel retailer in
the U.S.7 For all products, its US 1 trillion market value made it roughly the same
size as the next seven largest consumer companies put together.8
To summarize, trade rules, technology and financialization have contributed to
growing power asymmetries in apparel GSCs. These current power imbalances in
GSCs are engendering two mechanisms which have deleterious effects on workers:
a price squeeze and a sourcing squeeze. The price squeeze refers to how much lead
firms pay suppliers for the products they produce. If, as I suggest, the growth of
the supplier base increased with the phase out of the MFA and the entry of China
into the WTO, and if, as I also indicate, buyer consolidation grew over the past 20
years as a result of technological innovations and growing capital markets, then we
can expect to see declining prices starting at some point in the mid to late 1990s
and a continued decline afterwards. If, as I also suggest, power imbalance is
increasing in apparel GSCs, then we can expect lead firms to use their leverage not
only to reduce prices, the ‘price squeeze,’ but also to reduce lead times, shift order
size and modify product specifications on short notice according to their needs.
This is what I define as the ‘sourcing squeeze.’
 REVIEW OF INTERNATIONAL POLITICAL ECONOMY 325

The price squeeze reflects lead firms’ interests in keeping production wages low.
Historically, this was the case under the Export Processing Zone model of apparel
assembly production, because wages were the most significant cost of production
in this system. Thus, a squeeze down on prices by lead firms pushed supplier facto-
ries to keep down real wages, often by not raising wages during periods of infla-
tion, by lobbying governments not to raise the minimum wage or by lobbying
governments to have a lower minimum wage for apparel export workers relative to
workers in other manufacturing sectors (Anner, 2011).
This is not the only potential impact of the price squeeze on workers. When it
is not possible to keep real wages down, supplier firms can turn to a long-held
practice in the sector: increasing work intensity by increasing worker production
targets. For example, a worker might be told she needs to perform 90 operations
per hour as opposed to the previous requirement of 80. It is important to note that
this is not a traditional piece-rate system where wages are directly and solely tied
to production. Such a system is most often not permitted due to minimum wage
laws. However, employers can demand workers perform a number of operations
per hour in order to keep their jobs. Under such a system, when lead firms lower
prices and suppliers are not permitted to lower wages due to minimum wage laws,
suppliers can demand workers work faster. In this way, a price squeeze often
pushes suppliers to increase production quotas to remain economically viable,
resulting in increased work intensity.
The sourcing squeeze includes a squeeze on lead times, the time given to sup-
plier factories to design, source inputs, make and ship an item. Speed to market –
manifest through shorter production lead times – has become part of a broader
trend epitomized by ‘fast-fashion’ (Anguelov, 2016; Taplin, 2014). Shorter lead
times allows buyers to better control inventory. This reduces the need to sell items
at a markdown and thus can result in significant savings for firms. Another elem-
ent of the sourcing squeeze is order size and number of styles. As argued by Locke,
‘some persistent labor problems originate in various upstream business practices’
(Locke, 2013, p. 153). These practices include small order volume combined with a
greater variation of styles. Nike acknowledged as much, noting, ‘Every time a fac-
tory has to change a style, it reduces productivity and overall efficiency, adding to
the total number of hours of work requested’ (Cited in Locke, 2013, p. 128).
Extreme order volume fluctuation can be expected to have three important
impacts on working conditions. Since factory owners are unsure how large their
orders will be month to month, they opt to hire a lower number of workers rather
than risk having idle workers during downturns. As a result, when order volume
increases, suppliers have workers put in overtime hours, often extensive and forced
(Anner, Bair, & Blasi, 2013). Even when all workers are doing, for example, over
70 h a week, this still may not be enough hours of work to complete order spikes
with short lead times. In these cases, factories may turn to sub-contracting whereby
they outsource production to another factory, typically smaller, with more precar-
ious working conditions, not inspected by building-safety or labor authorities, and
perhaps not authorized by a lead-firm to produce its products (see Figure 1,
green arrows).
For all factories – authorized and unauthorized by lead firms – the incentive to
use sub-standard buildings is a result not only of low prices, but also of fluctua-
tions in order volume. As noted by Piore (1997), whereas working hours can vary
326 M. ANNER

Figure 1. Order volume fluctuation, overtime, subcontracting and building safety.

with order volume, building expenses are a fixed cost. Hence, ‘to minimize that
[building] cost, the employer will seek out cheap – that is, substandard – factory
housing. [ … ] The attempt to reduce rent paid per worker is the chief cause of
congestion in sweatshops, affecting the way in which material inventories, supplies,
equipment and work-in-progress block aisles and exits. It is also the source of
unhealthy and dangerous conditions (poor wiring and ventilation, unsanitary or
nonexistent bathrooms, fire hazards) [ … ] (Piore, 1997, p. 137). What this suggests
is that, to avoid losing money during down cycles, the supplier opts for sub-stand-
ard buildings to reduce fixed costs during low production periods, and this trend
can be expected to accelerate due to increased order volatility (see Figure 1, red
lines and arrow).
We can expect the price and the sourcing squeezes to adversely affect respect
for workers’ rights. Workers’ rights are understood in accordance with ILO
Committee of Expert interpretations of ILO Conventions 87 and 98 in terms of the
rights of workers to form unions, bargain collectively and strike (Bellace, 2014).
These rights are interconnected. The formation of a trade union would be of lim-
ited impact if unions were not able to bargain collectively to improve wages, bene-
fits, and the conditions of work. And without the ability to threaten to withhold
labor through the right to strike, employers would have little incentive to partici-
pate in meaningful negotiations (Hicks, 1932).
The price and sourcing squeezes create an incentive for employers to pursue
union avoidance strategies, which may include violations of the law. This is because
employers assumed that unions will create pressure to raise costs by bargaining to
increase wages and benefits. Employers also assume that unions will strike and dis-
rupt production, which could have profoundly negative consequences for suppliers
with short lead time pressure. State representatives may fear that, were they to vig-
orously enforce their labor laws in the context of highly cost-competitive GSCs,
investors will go elsewhere (Anner, 2011). In sum, GSCs create pressures on
employers and states in producing countries to keep wages down and unions out
in order to keep investors in.
My framework for understanding the price and sourcing squeezes and their
impact on working conditions and workers’ rights is presented in Figure 2. The fig-
ure begins at the top where developed states and international institutions’ trade
and investment policies facilitate the geographic dispersion of suppliers through
shifting trade rules. Trade rules, technology and financial capital facilitate the con-
solidation of a few mass merchandizers and electronic commerce companies at the
 REVIEW OF INTERNATIONAL POLITICAL ECONOMY 327

Figure 2. Price and sourcing squeeze and workers’ rights.

expense of others. The result is depicted in the second tier of Figure 2: buyer con-
solidation on the one hand, and the expansion and dispersion of suppliers on the
other hand. The combination of these two trends – depicted in the third tier of the
Figure – is growing power asymmetries in apparel GSCs. Buyers then use their
leverage over suppliers through two mechanisms: the price squeeze and the sourc-
ing squeeze. This, in turn, results in deleterious impacts on labor via a squeeze on
wages and working conditions. Workers can be expected to respond to these pres-
sures through some form of collective action, unionization or otherwise
(Henderson et al., 2002), which, in turn, provokes employer efforts to suppress
workers’ rights in order to keep costs down (see Figure 2, bottom tier, green).

Methods/data sources
To explore the arguments presented above, this article draws on mixed methods
and two original data sources. The first section of the article examines sourcing
and pricing dynamics using US government and European Union (EU) trade data.
The US government provides data on apparel imports by country in US dollars
and in square meters. This allows one to divide imports in dollars by imports
measured in squared meters, which indicates dollars per square meter of imported
apparel. It is also possible to measure price per unit. This is useful when tracing
the same units (e.g. trousers) over time. It is less useful when tracking the price of
328 M. ANNER

a bundle of goods because a shift in value may reflect a shift in products (for
example, from dresses to socks). EU trade data is available in euros and kilograms.
This allows one to measure euros per kilogram of exported clothing to the EU
by country.
The second set of data is drawn from field research in Bangladesh between 2015
and 2017. The first trips to Bangladesh were used to conduct initial factory visits
and interviews. Over 30 interviews were conducted with employers, government
representatives, non-governmental organizations, trade unions, workers and local
researchers. These interviews lasted from 1 h to over 3 h. These interviews served
two purposes. First, they added rich contextualization and nuances to the quantita-
tive data. Second, they helped asked to shape the questions asked in the surveys.
The first survey I conducted was of garment supplier factory owners and man-
agers in Bangladesh between March 2016 and March 2017.9 Factory lists to begin
the survey process were drawn from a random sample of factories taken from the
government website focusing on the districts of Gazipur and Mirpur.10 To conduct
the survey, I hired a local team of surveyors, who I trained and supervised. The
survey focused on Gazipur and Mirpur to ensure variation in the sample of facto-
ries; Gazipur is known for relatively larger factories outside the city center, whereas
Mirpur is known for smaller factories inside the city center. When the survey team
exhausted the number of factories we could enter in those districts, it expanded
into a few other areas. By the end of the process, the team collected 74 surveys in
Gazipur, 61 surveys in Mirpur, 30 surveys in Savar, 36 surveys in Ashulia and 10
surveys in Sripur.11 Each survey was based on factory visits that lasted anywhere
from one to 3 h. After 13 months, the team managed to survey 223 factories.12
The mean number of production workers per factory surveyed was 1,128. For
comparison, it is estimated that the average size of all apparel export factories in
Bangladesh is 1,000 workers (Anner & Bair, 2016). The average size of the factories
in the Accord program is 1,569.13 This places the survey sample within this range.
Some 55% of factories in the sample produced directly for European brands. Some
13% produced directly for U.S. brands. And 26% produced via buying houses
(third parties). The three most common items produced in these factories were T-
shirts (28%), bottoms/trousers (25%), tops/shirts and blouses (24%) and sweaters
(18%). The survey data allow for an exploration of the nuances of price dynamics,
including by product and export market. The survey also provides data on non-
price sourcing dynamics, such as lead times and order specifications.
The second survey I implemented was of factory workers.14 This survey was
conducted in May and June 2017 in the Dakha region. A team of surveyors who I
trained and supervised went to working class neighborhoods and interviewed 188
garment export workers in their homes. Most workers were between 22 and
30 years of age (64.89%) and 74% were female. The average years of schooling of
the workers surveyed was 5 years. The average worker has been employed in her or
his position for 3.26 years and worked in the apparel export sector of 4.87 years.
These findings reflect other recent garment worker surveys in Bangladesh. For
example, Kabeer, Haq, and Sulaiman (2019) found in their survey of 1,500 workers
a mean worker age of 26.5 with an average of 5.9 years of schooling who had
worked in their current factory for 3.6 years and worked in the industry
for 5.6 years.15
 REVIEW OF INTERNATIONAL POLITICAL ECONOMY 329

Figure 3. US Apparel Imports, USD/Sq. m.

The final section of the article turns to data from the LRIs.16 The LRIs are based
on coding the findings of nine sources for 108 workers’ rights violation types. The
nine sources include reports of the ILO Committee of Experts on the Application
of Conventions and Recommendations, reports of the ILO Conference Committee
on the Application of Standards, Country baselines under the ILO Declaration
Annual Review, Representations under Article 24 of the ILO Constitution,
Complaints under Article 26 of the ILO Constitution, Reports of the ILO
Committee on Freedom of Association, national legislation, the International Trade
Union Confederation Survey of Violations of Trade Union Rights and U.S.
Department of State’s Country Reports on Human Rights Practices17 These data
allow me to explore trends in pricing and workers’ rights in practice.

Trade data findings: Pricing

I have argued that lead firms in apparel GSCs increased their leverage over supplier
factories over the past 20 years as a result of shifting trade regimes, as well as
technological developments and capital market dynamics. I have further suggested
that the shift from the export processing zone model of assembly production to
full-package production has not mitigated these power imbalances. What do the
trade data indicate? Dividing the price of all apparel imports to the United States
by the square meters of all apparel imports to the United States, there was a slight
increase and otherwise relative stability in prices during the 1990s. Then there was
a decline in price per square meter starting in 1998, 4 years after the MFA phase-
out began.
This price decline became notably pronounced from 2000 to 2005, the 5 years
following permanent normal trade relations between China and the United States
and then China’s entry into the WTO. The decline became even more dramatic
following Vietnam’s entry into the WTO in 2007. There was then a noticeable rise
330 M. ANNER

in prices from 2010 to 2012. This can be attributed to a dramatic spike in world
cotton prices (cotton prices rose between February 2010 and February 2011 by
164.94%18). As cotton prices stabilized, the price decline continued through to June
2018 (see Figure 3). It is important to note that these figures are presented in nom-
inal US dollars. If we were to control for inflation and present the figures in real
US dollars, the price decline would be more dramatic. Trade data thus provide sup-
port for the price-squeeze argument that changing trade rules resulted in supplier
dispersion and increased power imbalances in GSCs that translated into downward
pressure on prices.

Bangladesh in comparative perspective

To further appreciate the pricing dynamic, it is helpful to compare country and
product dynamics. Bangladesh offers an ideal case study because it is the second
largest apparel exporter in the world, following China. Yet, unlike China where
apparel exports have been in decline since 2014, apparel exports from Bangladesh
continue to rise. China is also an exceptional case, because it has not competed pri-
marily based on wages. Rather, it is the scale of production, the size and complex-
ity of its locally-based supplier networks, and efficiency of its logistics that made
China the largest apparel exporter in the world. All these factors help to bring
down the costs per unit and, for a long time, compensated for its higher wages
relative to other major apparel exporters. Indeed, precisely because China has
excelled so markedly in terms of production scale, domestic supplier networks and
logistics, the pressure on its competitors, who lacked these advantages, to gain mar-
ket share to do so based on labor costs became more intense. During this time,
exports from Bangladesh –which, at US 0.39 per hour, pays the lowest minimum
wage among major apparel exporting countries– surpassed all competitors with the
exception of China (see Figure 4).
Bangladesh is also an important case study because, unlike China, garment
exports dominate the country’s exports and have a profound impact on the domes-
tic economy. In 2015, the sector accounted for 76.33% of its exports in goods and
services and 13.64% of its Gross Domestic Product (GDP).19 Over 4,000 suppliers
and 4 million direct workers make their living off the sector (Anner & Bair, 2016).
And many millions more are indirect beneficiaries of the sector. Thus, the sustain-
ability of the RMG sector has a profound impact on the overall economic condi-
tions in the country and the well-being of millions of people.
To understand Bangladesh’s growth in the global garment export sector, it is
helpful to look at its top export by product category to the United States, men’s
and boys’ cotton trousers. The data indicate that Bangladesh is the main exporter
of trousers to the United States, even exceeding the value of China’s exports in this
category. Mexico had been the top exporter of trousers in the 1990s, with a
remarkable rise following the implementation of the North American Free Trade
Agreement (NAFTA) in 1994. But after peaking in 2000, Mexico’s market share
dropped precipitously. Vietnam and Pakistan’s trouser exports do not even amount
to one third of Bangladesh’s exports to the United States (see Figure 5).
How did Bangladesh surpass both China and Mexico to be the top exporter of
trousers to the United States? The answer can be found by looking at price per
unit. If one looks at these top apparel-exporting countries to the United States and
 REVIEW OF INTERNATIONAL POLITICAL ECONOMY 331

Figure 4. Garment exports, developing countries (USD Billions).

Figure 5. Exports of trousers to the US (millions).

takes price paid per unit in US dollars, two trends can be seen. First, the price
point comes down for the largest exporters, China and Bangladesh. Second,
Bangladesh comes down to the lowest level, even lower than China. Mexico goes
from being the number one exporter to a distant third and most likely will soon be
displaced by Vietnam as it gets priced out of the market. As prices rise in Pakistan,
it also starts to lose market share. What is noticeable is (with the exception of
Mexico) how the discrepancy in prices is also reduced, suggesting that buyers are
much more aware of price points among competitors across the global economy
when they negotiate prices (see Figure 6). What is also noticeable is that these
332 M. ANNER

Figure 6. USD per trouser, top exporters to US.

figures are presented in nominal US dollars. If we were to incorporate a deflator

for inflation, the real dollar price decline per units would be far more significant.
A similar dynamic can be observed with respect to T-shirts and exports to the
European Union. T-shirts account for the largest percentage of garment exports
from Bangladesh to Europe (EU-28). In 2017, Bangladesh exported over 250 mil-
lion kilograms of T-shirts per year to the EU, far exceeding the combined amount
of T-shirts exports from the second, third and fourth top exporters (China, India
and Turkey). As with trousers to the United States, Bangladesh came to dominate
the T-shirt market by providing the lowest price point. While the price of T-shirts
was 14.93 euros per kilogram in China, 15.78 euros in India and over 20 euros in
Turkey, the price of a kilogram of imported T-shirts from Bangladesh was 10.92
euros per kilogram (see Figure 7).
When examining prices paid for products exported from Bangladesh to the U.S.
and Europe, it is important to keep in mind the period from 2013 going forward,
because this is the period following the Rana Plaza building collapse when inter-
national pressure was placed on Bangladesh to improve building safety, wages and
workers’ rights. On December 1, 2013, the Bangladeshi government increased the
minimum wage by 77%, from USD 39 per month to USD 68 per month. Also,
beginning in 2013, three building safety programs – the Accord, the Alliance and
the National Initiative – inspected factories and determined the expenditures facto-
ries needed to make to improve the safety of their facilities. However, while build-
ing safety costs and wages were rising, what the data show is that, in general,
prices continued to decline. One argument for this decline is that it coincided with
the decline in the price of cotton. While the dramatic rise in cotton prices in 2011
can certainly explain the rise in garment prices in the 2011 and 2012 period, the
relationship does not hold in the post-2013 period. Indeed, what is most striking is
that, from 2015 to 2017, cotton prices rose by 22.13% while the price paid to
Bangladesh exporters of cotton trousers declined by 10.24%.
 REVIEW OF INTERNATIONAL POLITICAL ECONOMY 333

Figure 7. Price and Volume of T-shirts Exports to EU-28 (2017).

Bangladesh supplier factory survey findings

While international trade data provide support for my argument that lead firms in
apparel GSCs are squeezing down on price, one limitation of relying exclusively on
trade data is that such numbers include a range of costs that are added after a
product leaves a factory. These include international shipping, freight insurance
and duties. This is the difference between Free on Board (FOB) costs and
Estimated Landed Costs (ELC) Cost, Insurance & Freight (CIF). International trade
data reflect CIF. Yet, today most lead firms negotiate FOB prices with supplier fac-
tories. FOB costs are made up of the Cut-Make-Trim component (the traditional
apparel assembly operations), as well as inputs, most notably fabric. Hence, to
more fully appreciate pricing dynamics faced by suppliers, it is necessary to look at
FOB prices.

Survey findings: price squeeze

As noted above, a core measure of price for a garment producer is the FOB price.
As noted, this price includes the fabric, trims, embellishments, labor and local
freight to port costs. What the survey results indicate is that the average FOB price
was USD 4.64 in 2016, which is a 7.75% decline from the FOB price point of USD
5.03 in 2011. If we look at exports to the United States, the price point declined by
10.71%. For European buyers, the price point came down by 8.96%. Indeed, in all
major product categories we find a decline in nominal prices paid per unit (see
Table 1). Hence, not only is there a decline in CIF value as illustrated by the trade
data, but there is also a decline in FOB prices as indicated by the survey data. As a
result, the decline in prices cannot be mainly attributed to declines in international
transportation costs, freight insurance or duties.
The survey allowed suppliers to answer an open-ended question on their concerns,
insights and experiences. Of the 132 responses to the open-ended question, 82
334 M. ANNER

Table 1. F.O.B. prices, 2011–2016.

Product 2011 2016 Percent change
All $5.03 $4.64
7.75%
U.S. Buyers $5.32 $4.75
10.71%
European Buyers $4.91 $4.47
8.96%
Trading Houses $4.90 $4.63
5.51%
Knit $3.80 $3.40
10.53%
Woven $6.48 $6.05
6.64%
Bottoms $6.43 $6.10
5.13%
Tops $5.01 $4.58
8.58%
T-Shirts $2.72 $2.43
10.66%
Source: Author’s survey.

mentioned the declining-price issue. One supplier noted, ‘Though production cost is
rising day by day, the price rate is not increasing. Rather the price rate is reducing.’
Another stated, ‘Factories have to take orders at a less price in fear of losing the order
to someone else. We are not getting much profit, but we have to run our factory.’ This
concern is reflected in an ILO report, which found that 46% of textile and clothing
suppliers, ‘reported having accepted orders whose price did not allow them to cover
their production costs’ (ILO, 2017, p. 8). In Bangladesh, the ILO survey indicated that
52% of the Bangladeshi respondent suppliers reported selling below cost, and that
they did so under pressure from their buyers (ILO, 2017, p. 8).
We asked in our survey whether buyers increased the price they paid to suppli-
ers to produce their products following the 2013 increase in the minimum wage.
Some 13% of suppliers stated that their buyers did increase their prices whereas the
majority, 87%, indicated that their buyers did not increase their prices following
the increase in the minimum wage. Some suppliers who did receive a price increase
following the increase in the minimum wage noted in interviews that this lasted
for only one production cycle. With the following cycle, prices were lowered again.
This provides support for the finding based on U.S. and E.U. trade data that most
buyers continued to bring down prices even as the minimum wage increased.
Suppliers have been squeezed on costs in other ways, too. For example, we
asked about payment times after shipment. As in any business operation, cash flow
is a major concern. If a supplier makes a product and is not paid immediately, the
supplier lacks resources to cover past expenses, which may limit its ability to start
a new production cycle effectively. In some cases, a delay in payments by buyers to
suppliers may result in a delay of salary payments. What the survey indicates is a
19.72% increase in payment time after shipment. In other words, lead firms were
able to use their leverage to indicate to suppliers when they will pay, and they were
increasingly deciding to pay later. Of course, when the lead firms pay late, it is
often able to earn a financial return on those funds during the time it delays in
paying its supplier. For the supplier firm, this is lost income. Indeed, the supplier
firm may need to take out a costly loan in order to cover new business expenses as
it waits to be paid by the lead firm for products already shipped.

Survey findings: sourcing squeeze

The push for shorter lead times – the time given to factories to make and ship a
product – has received increased attention with the growth of fast fashion (Taplin,
2014). Many buyers are seeking to develop this model because, if they can get
 REVIEW OF INTERNATIONAL POLITICAL ECONOMY 335

consumers shopping for new styles with greater frequency, then they can sell more
products each year and increase their revenue. However, even retailers that do not
engage in a full fast-fashion model are looking for greater speed to market. This
allows them to more effectively plan product development closer to the actual time
in which consumers will be buying their products. The survey findings indicate
from 2011 to 2016 there was a reduction in lead times of 8.14%, from 93.4 days to
85.83 days. Hence, we find some evidence for a sourcing squeeze in terms of lead
times. This may contribute to excessive overtime and increased work intensity.
The concern about purchasing practices related to order volume was expressed
by suppliers through the open-ended survey question. One supplier explained, ‘The
costs are increasing, while orders are decreasing and the price is less, too. There
are not enough orders to survive.’ For example, instead of getting an order for
50,000 units, they might get an order for 10,000 units. This is part of the fast-fash-
ion model where buyers are no longer looking for a large order that sits on their
retail shelves for a couple of months, but rather they want small orders that might
only sit on the shelves for one month or even a couple of weeks. The problem for
suppliers is that this makes it increasingly hard at times to find enough orders to
keep a factory running at full capacity. And, since turning down the orders when
they come in risks loss of business, it also leads to production spikes, when facto-
ries are trying to make more orders than they can handle. This may contribute to
forced overtime and unauthorized outsourcing. One supplier noted, ‘There is no
order consistency. We do subcontract orders. We need regular orders and need a
system of receiving orders.’ Constant style changes also dramatically hurt product-
ivity, because just as the factory is getting skilled at making a new style, it is time
to terminate the order and start something new (Locke, 2013).
An additional sourcing concern explored through this survey is whether buyers
changed order specifications once production starts. What is not possible to change
once production starts is order volume, because that is related to prior raw material
purchases. But changes such as lead times or even the style of buttons are an issue.
Indeed, some 26% of suppliers surveyed report that buyers ‘often’ change order
specifications after production starts, and 55% report that buyers ‘sometimes’
change order specifications after production starts. Only in 19% of the time do
suppliers report that buyers do not change order specifications once production
starts. This is a concern for suppliers because every change involves production
reorganization and thus costs.

Order fluctuations and building safety

Of all the poor labor conditions and workers’ rights violations, Bangladesh will
always be remembered for Rana Plaza and the issue of building safety. Prices were
an important cause of what happened, because suppliers who are squeezed on price
tend to turn to less expensive buildings in order to lower costs. That is, many tier
1 factories have been shown to be unsafe. Order fluctuations associated with fast
fashion and other factors associated with the drive for speed to market also con-
tribute to sub-optimal buildings, resulting in unsafe working conditions. That is,
because order volume is unstable, factory owners are inclined not to turn down
orders for fear that their factory might be left idle at a later stage. As a result, they
frequently take orders that are beyond their factory capacity. As a result, they may
336 M. ANNER

Figure 8. Factory deaths in Bangladesh.

meet production obligations by outsourcing part of a production order to

unauthorized and potentially unsafe suppliers (see Figure 1 above).
I have argued that the pricing and sourcing squeezed began to intensify in the
early 2000s. What we know regarding life-threatening building safety is that three
of the four worst factory disasters in the history of industrial garment production
have taken place since 201220 (Nova & Wegemer, 2016). This includes the
September 2012 Ali Enterprises factory in Pakistan in which over 250 workers
died, the November 2012 Tazreem Fashions fire in Bangladesh in took the lives of
112 workers, and the 2013 Rana Plaza building collapse with the loss of 1,138 lives.
Labor rights experts have found that, from 2005 to 2013, nearly 2,000 workers
were killed in more than a dozen fires and building collapses (Nova & Wegemer,
2016). In the case of Bangladesh, in the 15-year period from 1990 to 2005 some
451 workers lost their lives in garment factory disasters, whereas in the 7-year
period after 2005 and before 2013, some 620 workers lost their lives (Ross, 2016).
Then, in 2013, the Rana Plaza building collapsed (see Figure 8). The data on fac-
tory fatalities indicate an increase in unsafe buildings since the sourcing squeeze
intensified in the early 2000s.21

Price and sourcing squeezes impact on labor conditions

Trade data and our survey findings provide support for our argument of the price
and sourcing squeezes. The question is how, given these squeezes, firms stay in
business. One way is by lowering profit margins. Bangladesh garment employers
were known in the 1980s and 1990s for having very high profit margins. For a
long period of time, an acceptable profit margin for suppliers set by lead firms was
10% (O’Rourke Group Partners, 2011), and Bangladesh employers appeared to be
well above that rate. In the early 2000s, as the price squeeze intensified, it appears
some firms reduced their margins and remained economically viable (Staritz, 2011,
p. 141). Our survey data provide support for this finding: the period of high profits
for most companies is no longer the case. According to the survey results, profit
margins decreased by 13.3% from 2011 to 2016, with a mean 2016 profit margin of
7.69%. As noted above, a significant number of suppliers will even accept orders at
a loss in order to keep a factory running with the hope of getting more favorable
contracts in the future.
 REVIEW OF INTERNATIONAL POLITICAL ECONOMY 337

Table 2. Bangladesh garment workers’ wages and hours of work.

Women (n ¼ 139) Men (n ¼ 48) t Sig.
Monthly wages without overtime $77.34 $92.94 6.022 0.008
Monthly wages (with O.T. & bonuses) $99.17 $120.29 6.342 <.001
Average hours of work per week 63.24 64.2 0.848 0.466
Average hourly wage $0.37 $0.44 6.342 <.001
Overtime hours are voluntary (’yes’) 24.50% 12.20% 1.768 <.001
Source: Author’s survey.

The impact of the price and sourcing squeeze are far more deleterious for work-
ers. This is because, just as there is a power imbalance between lead firms and sup-
pliers, so too is there a power imbalance between suppliers and their workers, most
of whom are poor, young women from the countryside. As Raworth and Kidder
(2009, p. 165) observe, ‘suppliers transfer the pressure onto workers, who bear it in
the form of precarious employment; workforces that are composed primarily of
women and migrant workers endure insecure contracts, low wages, excessive hours
and few benefits’.
To explore wage and hours of work, during the year we were conducting the
supplier survey in Bangladesh we also conducted structured interviews with gar-
ment workers in the Dhaka region. What the survey findings indicate is that, for
an average monthly wage, female workers earned USD 77.34 and male workers
earned USD 92.94. For average monthly wages with overtime pay and bonuses,
female workers reported earning USD 99.17 and male workers reported earning
USD 120.29. This amounts to an hourly take-home wage of USD 0.37 for women
and USD 0.44 for men (see Table 2). Some 12.2% of men and 4.3% of women indi-
cated that their wages always covered their living expenses. What these data suggest
is that the price squeeze has contributed to sub-poverty wages and that women
workers face additional discrimination based on their gender. This provides sup-
port for the arguments of gender-based wage discrimination in GSCs (Barrientos,
Dolan, & Tallotire, 2003).
Workers also reported an average work week of 63.49 h, with some 31.4% of work-
ers indicating they worked 72 h per week. Notably, there was no significant difference
between the number of hours men and women worked per week. When asked
whether overtime hours were voluntary or obligatory, some 77.2% of the respondents
indicated that overtime hours were obligatory (see Table 2). This suggests support for
the argument that short lead times and frequent adjustment to order specifications on
short notice are conducive to long hours and forced overtime.22 Finally, we asked
about work intensity. Specifically, we inquired whether their rate of production (pro-
duction targets) had stayed the same, gotten faster or gotten a lot faster over the last 5
years. Some 47.8% stated that their rate had stayed the same, whereas 52.2% indicated
the pace had gotten faster to a lot faster. However, these findings on work intensity
are very tentative because only 23 of the 188 workers surveyed answered this question.
Further research in this area is needed.

Price and sourcing squeezes and workers’ rights

These findings on working conditions lead us to my second proposition: that the
price and sourcing squeezes adversely affect respect for workers’ rights, because
338 M. ANNER

unions are presumed to push for better (e.g. costlier) terms of employment and
because unions are expected to lessen firm flexibility in terms of imposing working
hours and production targets. In the case of Bangladesh, there has been much dis-
cussion and debate regarding the issue of workers’ rights. Bangladesh has been
repeatedly notified by the International Labour Organization that its laws regarding
freedom of association need to come into compliance with international standards,
notably by allowing for unionization in export processing zones and facilitating the
unionization process in the garment sector by removing bureaucratic obstacles,
government discretionary authority and an unreasonably high membership thresh-
old requirement.23
Following Rana Plaza, strong international pressure contributed to labor law
reforms that modestly facilitated union registration. Perhaps most importantly, in
the year following Rana Plaza, workers, unions, employers and the government
realized the country was being closely monitored by the EU, the United States
and Canada (among others) and the International Labor Organization. This gave
unions and the workers they represented the sense of protection they needed to
increase their organizing drives while signaling to employers and governments
the need to restrain from actions that could be perceived as anti-union. In this
context, the number of unions applying for government recognition increased
dramatically (from 12 applications in 2012 to 392 in 2014). And the number of
government approvals also escalated. As of January 2018, there were 440 RMG
factories with unionization, up from less than 100 prior to Rana Plaza.24
However, the labor law reforms did not modify the requirement for a 30%
threshold for unionization formation (share of workers in a given workplace sup-
porting a union), and nor did it modify the exemption of Export Processing
Zones from the labor law. As attention on Bangladesh waned, the government
rejected more union registrations, anti-union repression intensified and union
applications declined (see Figure 9).

Figure 9. Trade unions in Bangladesh garment sector.

 REVIEW OF INTERNATIONAL POLITICAL ECONOMY 339

Figure 10. Prices (trousers USD/Sq. m) and workers’ rights violations in Bangladesh.

What role might the price squeeze have on respect for workers’ rights? Using
the LRIs, we are able to observe trends over time. The LRI data are organized in
three categories: violations in law, violations in practice and overall violations.
Using these data for Bangladesh, the first observation we can make is the depth of
workers’ rights violations. Since 2012, Bangladesh has scored more than 7 on a 10-
point scale in which 10 indicates a complete lack of labor rights. Excluding coun-
tries that received a default score of 10 for their lack of civil liberties, Bangladesh
has the fourth worst labor rights score in the world. Moreover, since 2000, there
has been an increase in the overall violation of workers’ rights in Bangladesh and a
general increase in law and in practice violations. To explore the relationship
between workers’ rights and pricing, it is possible to superimpose the price per
square meter for men’s and boys’ cotton trousers exported from Bangladesh to the
US over the figure worker rights volitions. While there are some ups and downs in
prices and worker violations, the overall trends are a decline in price and a rise in
labor rights violations (see Figure 10).
As the price squeeze results in a squeeze on wages, the squeeze on wages can
motivate worker collective action. There has been a long history of worker mobil-
ization to increase wages in Bangladesh (Siddiqi, 2017). In response, employers
have used their leverage over the state to suppress worker efforts to improve condi-
tions, which is captured by the data presented in Figure 9. For example, in
December 2016, some 1,600 garment workers were dismissed and 34 activists were
arrested and detained after workers protested and demanded higher wages.
National and international labor rights advocates demanded buyers doing business
in Bangladesh take action.
On June 14, 2018, the European Parliament passed a resolution stating that
Bangladesh needed to address the persecuting of trade union leaders and poor
340 M. ANNER

working conditions, including long working hours, low wages and hazardous con-
ditions.25 The ILO Committee of Experts noted in 2016, ‘with concern the numer-
ous allegations of anti-union discrimination and harassment of workers, including
dismissals, blacklisting, transfers, arrests, detention, threats and false criminal
charges combined with insufficient labor inspection, lack of remedy and redress
and delays in judicial proceedings.’26 The Committee, ‘urged the Government to
investigate as a matter of urgency all acts of anti-union discrimination, ensure the
reinstatement of those illegally dismissed and impose fines or criminal sanctions
(particularly in cases of violence against trade unionists) according to the law.’27
Yet, worker rights violations continued to escalate. In December 2018, when work-
ers demanded a minimum wage of USD 188 per month, some 11,600 were dis-
missed (WRC, 2019).

Workers’ rights in comparative perspective

We have seen how workers’ rights violations tend to increase with a decline in pri-
ces in Bangladesh. And we have seen that, in general, there has been a significant
decline in prices since 2000. How has this decline affected workers’ rights dynamics
in the apparel export sector more broadly? In the case of Vietnam, as the real dol-
lar price paid per blouse (a major garment export) shipped from Vietnam to the
United States from 2005 to 2016 declined by 29.09 per cent, there is evidence for
chromic worker violations there (Anner, 2018b). In a prior research project with
Bair and Blasi, we showed that, this pricing squeeze has affected all of the twenty
top apparel exporters to the United States, that this price squeeze intensified in the
late 1990s and early 2000s, and that this price squeeze was associated with a decline
respect for workers’ rights in major apparel exporters (Anner et al., 2013). The
data used for that project was based on a single source workers’ rights database
with a three-point scale, the Cingranelli and Richards Human Rights
Dataset (CIRI).
A more developed workers rights database is the LRIs. Rather than one source,
it codes nine sources. And in place of a three-point scale, it has 108 evaluation cri-
teria representing different types of worker rights violations, including violations of
fundamental civil liberties, workers’ right to establish and join organizations, rights
concerning the internal functioning of these organizations, rights to collective bar-
gaining and the right to strike. The database also provides a more recent picture of
worker rights violations.
Using trade date to determine the ten top apparel exporters to the U.S. in 2000
and 2015 and their market shares, it can be seen that some of the worst workers’
rights violators became the largest exporters, and countries with slightly better
scores became less prominent exporters. Furthermore, some countries, such as
Bangladesh and Honduras, increased their level of violations as they increased their
export share to the US. Defining top workers’ rights violations as countries with an
overall score of 6.00 or higher, we see that 42.83% of exports to the US in 2005
came from top violators whereas in 2015, 70.84% of apparel came from top viola-
tors, a 65.39% increase (see Table 3). What this suggests is that the price and
sourcing squeeze have increased workers’ rights violations not only in Bangladesh
but also among top apparel exporters around the world.
 REVIEW OF INTERNATIONAL POLITICAL ECONOMY 341

Table 3. Top ten apparel exports to the United States and workers’ right violations (2000; 2015) (Pink
shaded areas: ‘top violators’).
Labor Rights Violations Scores (1–10)
Sq. Meters Share of US
Country (100 million) imports Total In law In practice
2005
China 58.83 26.74% 10.00 10.00 10.00
Mexico 17.03 7.74% 4.75 3.10 4.43
Honduras 12.47 5.67% 3.91 3.17 3.03
Bangladesh 11.25 5.11% 7.04 6.62 4.52
El Salvador 8.66 3.94% 5.13 3.68 4.44
Indonesia 8.23 3.74% 7.43 5.25 6.52
Vietnam 8.01 3.64% 10.00 10.00 10.00
India 7.90 3.59% 8.41 6.36 6.96
Dominican Republic 7.15 3.25% 3.10 2.03 2.89
Cambodia 7.10 3.23% 5.66 2.61 6.35
Average 6.54 5.28 5.91
2005: Top Violators, Share of Total Imports: 42.83%
2015
China 113.84 41.85% 10.00 10.00 10.00
Vietnam 31.35 11.53% 10.00 10.00 10.00
Bangladesh 18.71 6.88% 7.63 6.37 5.71
Indonesia 12.65 4.65% 6.49 5.81 4.47
Honduras 11.14 4.10% 5.37 3.68 4.83
Cambodia 10.51 3.86% 5.56 3.14 5.66
India 10.23 3.76% 6.27 6.36 3.56
Mexico 8.99 3.31% 4.95 3.35 4.49
El Salvador 8.13 2.99% 4.95 3.98 3.86
Pakistan 5.92 2.18% 7.97 8.79 3.82
Average 6.92 6.15 5.64
2015: Top Violators, Share of Total Imports: 70.84%
Source: Author’s calculations bases on the Labour Rights Indicators and OTEXA data.

Conclusions
This article has explored how buyer consolidation and supplier dispersion have
contributed to a growth in power asymmetries in apparel GSCs since the late
1990s. This power asymmetry is expressed through two mechanisms: a price
squeeze and a sourcing squeeze. Further, this article illustrated how these two
mechanisms have had deleterious impacts on working conditions and workers’
rights. What might be done to address these dynamics? Barrientos et al. (2011, pp.
336–337) argue that social upgrading can be facilitated with interventions by trade
unions, states and multi-stakeholder codes of labor practices.
No doubt, the right interventions can go a long way to improving working con-
ditions and respect for workers’ rights in GSCs. The role of the state in establishing
and enforcing labor standards and workers’ rights is crucial (Seidman, 2007), par-
ticularly in the area of freedom of association rights (Anner, 2012; Locke et al.,
2009). Private and state mechanisms, under the right circumstances, could comple-
ment each other (Amengual, 2010). And, as Bartley (2018) finds, there may be a
positive impact when there is a ‘re-centering’ of the state’ as part of ‘place-con-
scious transnational governance’ that used domestic law as a point of reference and
state-imposed penalties on violators.
This article also suggests that addressing poor working conditions and workers’
rights violations involves addressing the price and the sourcing squeezes. The prob-
lem, for example, of inadequate wages is not simply a matter of finding the right
combination of private and state mechanisms to ensure compliance with standards.
342 M. ANNER

Rather, part of the problem in many cases is that state standards are so low. In
Bangladesh, factories that pay garment workers USD 0.51 per hour (a rate which
covers only a fraction of living expenses) are in full compliance with national law.
The pressure to keep those rates low comes from the price and the sourcing
squeezes across apparel export countries. For example, in 2018, the Cambodian
Prime Minister Hun Sen told an assembly of workers: ‘Employers won’t stay if the
price of labor is expensive. There’s no way they could endure to stay if the min-
imum wage continues to increase. Be reasonable for your employers because if they
go bankrupt, they will move to another location.’28 At the same time that
Cambodian garment workers were paid USD 0.83 per hour, the largest apparel
retailer in the world, Amazon, saw its stock valuation pass USD 1 trillion and Jeff
Bezos, Amazon’s CEO, net worth reach USD 166 billion.29
It was the beyond the scope of this paper to examine other economic sectors.
However, extant literature suggests there are adverse impacts on labor of a sourcing
squeeze in other GSCs, be that in the agriculture sector, such as bananas and cut
flowers (Riisgaard & Hammer, 2011), or in electronics (Locke, 2013). Indeed, while
some has pointed to the mega-supplier Foxconn as evidence that suppliers in the
electronics can grow in size and scope and thus, in theory, gain power relative to
their buyers (Locke et al., 2009), data on profits margins suggests even this mega-
supplier has been no match for the power of the lead-firm, Apple. In 2011, Apple
enjoyed a gross profit margin of 40.44% while Foxconn’s margins were only 1.98%
(Locke, 2013), a power asymmetry that has increased over time (Chan, Pun, &
Selden, 2016).
What these cases suggest is that achieving decent work requires addressing sup-
ply chain power asymmetries and to fundamentally alter lead firm sourcing practi-
ces. More stable orders allow suppliers to more adequate allocate human resources
without resorting to unauthorized outsourcing or forced overtime. And stable
orders allow suppliers to increase their efficiency on a given order (Locke, 2013).
The challenge is to then ensure that workers benefit from such potential changes.
These cases also suggest a need for mechanisms to address pricing practices. One
such approach for doing that is reflected in the Bangladesh Accord, which required
buyers to ‘negotiate commercial terms with their suppliers which ensure that it is
financially feasible for the factories to maintain safe workplaces and comply with
upgrade and remediation requirements’.30 This holds retailers and brands respon-
sible for ensuring, ‘ensure factories have the financial capacity to comply with
remediation requirements’.31 This suggests that cost-sharing approaches should
require lead firms to calculate the ‘total cost’ of socially sustainable production,
from adequate wages to safe buildings (Hasan, 2019). In addition to the
Bangladesh Accord, models of cost-sharing and shared responsibility are present in
agriculture, transportation and construction, among other sectors (Blasi & Bair,
2019). And worker-driven initiatives have had an impact. For example, the
Bangladesh Accord has addressed approximately 100,000 building safety issues
(Anner, 2018a), and the Fair Food Program has resulted in adequate wages for
thousands of farmworkers (Asbed & Hitov, 2017)
More broadly, the findings in this article also suggests the need to re-think the
state policies and practices that have increased finance capital’s leverage over sup-
ply chains while also facilitating lead firm consolidation and supplier dispersion.
The development strategy promoted by international agencies based on the
assumption that every poor country can simultaneously export apparel and achieve
 REVIEW OF INTERNATIONAL POLITICAL ECONOMY 343

development needs to be revisited. Rather than promote development, the result

has been a crisis of overcapacity. The retailer and brand response has been to keep
production costs low through squeezing workers and while getting consumers to
buy more and more low-priced garments. This approach is not sustainable for the
environment or for workers.

Notes

1. The literature also uses the terms ‘global value chains,’ ‘global commodity chains,’
and ‘global production networks,’ and there is considerable debate over the meanings
of these terms (Bair, 2009). I use the term ‘global supply chains’ as a more neutral
term and one that has been adopted by the International Labour Organization. The
ILO defines GSCs as, “the cross-border organization of the activities required to
produce goods or services and bring them to consumers through inputs and various
phases of development, production and delivery” (ILO, 2016, p. 1).
2. Jason Zweig, “The Disturbing New Facts about American Capitalism.” The Wall
Street Journal, March 3, 2017.
3. Cited in Zweig, “The Disturbing New Facts about American Capitalism.” The Wall
Street Journal, March 3, 2017.
4. Neil Irwin, “Are Superstar Firms and Amazon Effects Reshaping the Economy?” The
New York Times, August 25, 2018.
5. See Lakhani, Kuruvilla, & Avgar (2013) for how this topology corresponds to
employment relations practices.
6. According to World Bank data, China’s 787 million workers account for 22.81% of the
world’s worker. See: https://data.worldbank.org/indicator/SL.TLF.TOTL.IN (accessed
September 12, 2018).
7. CNBC, “Amazon’s 100 million Prime members will help it become the No. 1 apparel
retailer in the US.” April 19, 2018. https://www.cnbc.com/2018/04/19/amazon-to-be-
the-no-1-apparel-retailer-in-the-us-morgan-stanley.html
8. Laura Stevens and Amrith Ramkumar, “Amazon Hits $1 Trillion Valuation.” The
Wall Street Journal, September 4, 2018. See: https://www.wsj.com/articles/amazon-
hits-1-trillion-valuation-1536075734?mod=searchresults&page=1&pos=2
9. A shorter, online version of this survey was developed with considerable input from
Jeremy Blasi and Jennifer Bair, for which the author is very grateful.
10. See: http://database.dife.gov.bd/.
11. A handful of suppliers completed an online version of the survey, and some of these
suppliers did not identify their sub-districts.
12. The sample size yields a confidence level of 90%, with a margin of error of 5.4%.
13. http://bangladeshaccord.org/factories/list-factories/
14. Due to financial and time restraints, this survey was more limited than the
supplier survey.
15. For gender, Kabeer, Haq, and Sulaiman (2019) purposively selected their sample to be
made up of 66% of women and 33% of men.
16. Center for Global Worker’s Rights. (2019). Labour Rights Indicators. Retrieved April
1, 2019, from http://labour-rights-indicators.la.psu.edu/
17. For more information and a full, detailed explanation of the methodology see Kucera
& Sari (forthcoming).
18. See: https://markets.businessinsider.com/commodities/historical-prices/cotton-price/
19. In 2015, the year the WTO reported USD 26.60 billion in clothing exports from
Bangladesh, the World Bank reported USD 34.85 billion in total exports of goods and
services from Bangladesh and USD 195.07 billion in GDP.
20. The fourth disaster was the Triangle Shirtwaist factory fire of 1911 in which 146
garment workers in New York City lost their lives.
21. Of course, factory deaths are the most extreme, but not the only indication of unsafe
buildings. Data gathered from factory inspections by the Bangladesh Accord indicate
344 M. ANNER

that, as of April 2019, the Accord had detected 21,799 structural violations, 46,932
fire safety violations, and 73,239 electrical safety violations in garment factories. The
Accord covers 1,688 factories, which indicates that, on average, these garment
factories had 84 violations per factory (see Bangladesh Accord, https://
bangladeshaccord.org/, accessed April 29, 2019). Unfortunately, these data are not
available for the 1990s, and thus do not allow for a comparison across time periods.
22. It may also be conducive to unauthorized sub-contracting, but we were not able to
explore this possibility through our structured interview.
23. See: http://www.ilo.org/wcmsp5/groups/public/—ed_norm/—relconf/documents/
meetingdocument/wcms_543646.pdf
24. Solidarity Center, Dhaka Office.
25. See: http://www.europarl.europa.eu/news/en/press-room/20170609IPR77025/
bangladesh-should-do-more-for-its-textile-workers-meps-warn.
26. Observation (CEACR) – adopted 2016, published 106th ILC session (2017) Right to
Organize and Collective Bargaining Convention, 1949 (No. 98) – Bangladesh
(Ratification: 1972).
27. Ibid.
28. Cited in Leonie Barrie, September 11, 2018. “Overtime Up in Line with Garment
Worker Pay in Cambodia.” Just Style.
29. Wall Street Journal, September 4, 2018: https://www.wsj.com/articles/amazon-hits-1-
trillion-valuation-1536075734?mod=searchresults&page=1&pos=2
30. Clause 17 of the Bangladesh Transition Accord. The full clause states: “In order to
induce factories to comply with upgrade and remediation requirements of the
program, participating brands and retailers will negotiate commercial terms with their
suppliers which ensure that it is financially feasible for the factories to maintain safe
workplaces and comply with upgrade and remediation requirements. Each signatory
company may, at its option, use alternative means to ensure factories have the
financial capacity to comply with remediation requirements, including but not limited
to joint investments, providing loans, accessing donor or government support,
through offering business incentives or through paying for renovations directly.”
http://bangladeshaccord.org/wp-content/uploads/2018-Accord-full-text.pdf.
31. Ibid.

Acknowledgments
The author thanks Teri Caraway, Matthew Fischer-Daly and Richard Locke for comments on ear-
lier versions of this paper. The paper also greatly benefited from collaboration and many long dis-
cussions on the Bangladesh apparel industry with Jennifer Bair and Jeremy Blasi. Funding for this
research was made possible, in part, by a grant from the Embassy of the Kingdom of the
Netherlands in Bangladesh, for which the author is extremely grateful.

Disclosure statement
No potential conflict of interest was reported by the author.

Notes on contributors
Mark Anner is an Associate Professor of Labor and Employment Relations and Political Science
at the Pennsylvania State University. He is the founding director of the Center for Global
Workers’ Rights and the founding director of the Penn State MPS Program in Labor and Global
Workers’ Rights, which is part of the Global Labour University network. He holds a Ph.D. in
Government from Cornell University and a Master’s Degree in Latin American Studies from
Stanford University. He has been researching workers’ right in the global apparel industry for the
past two decades.
 REVIEW OF INTERNATIONAL POLITICAL ECONOMY 345

References
Abernathy, F. H., Dunlop, J. T., Hammond, J. H., & Weil, D. (1999). A stitch in time: Lean retail-
ing and the transformation of manufacturing: Lessons from the apparel and textile industries.
New York: Oxford University Press.
Amengual, M. (2010). Complementary labor regulation: The uncoordinated combination of state
and private regulators in the Dominican Republic. World Development, 38, 405–414. doi:
10.1016/j.worlddev.2009.09.007
Anguelov, N. (2016). The dirty side of the garment industry: Fast fashion and its negative impact
on environment and society. Boca Raton: CRC Press.
Anner, M. (2011). The impact of international outsourcing on unionization and wages: Evidence
from the apparel export sector in Central America. Industrial and Labor Relations Review, 64,
305–322. doi:10.1177/001979391106400205
Anner, M. (2012). Corporate social responsibility and freedom of association rights: The precar-
ious quest for legitimacy and control in global supply chains. Politics & Society, 40, 604–639.
Anner, M. (2018a). Binding power: The sourcing squeeze, workers’ rights, and building safety in
Bangladesh since Rana Plaza. University Park: Center for Global Workers’ Rights.
Anner, M. (2018b). CSR participation committees, wildcat strikes and the sourcing squeeze in glo-
bal supply chains. British Journal of Industrial Relations, 56(1), 75–98. doi:10.1111/bjir.12275
Anner, M., & Bair, J. (2016). The bulk of the iceberg: A critique of the stern center’s report on
worker safety IN Bangladesh. State College, PA: Center for Global Workers’ Rights.
Anner, M., Bair, J., & Blasi, J. (2013). Toward joint liability in global supply chains: Addressing
the root causes of labor violations in international subcontracting networks. International
Labor Law & Policy Journal, 35(1), 1–43.
Appelbaum, E., & Batt, R. (2014). Private equity at work: When wall street manages main street.
New York: Russell Sage Foundation.
Asbed, G., & Hitov, S. (2017). Preventing forced labor in corporate supply chains: The fair food
program and worker-driven social responsibility. Wake Forest Law Review, 52, 497–532.
Bair, J. (2009). Global commodity chains: Genealogy and review. In J. Bair (Ed.), Frontiers of com-
modity chain research (pp. 1–34). Stanford: Stanford University Press.
Barrientos, S., Dolan, C., & Tallontire, A. (2003). A gendered value chain approach to codes of
conduct in African horticulture. World Development, 31, 1511–1526. doi:10.1016/S0305-
750X(03)00110-4
Barrientos, S., Gereffi, G., & Rossi, A. (2011). Economic and social upgrading in global production
networks: A new paradigm for a changing world. International Labour Review, 150, 319–340.
doi:10.1111/j.1564-913X.2011.00119.x
Bartley, T. (2018). Rules without rights: Land, labor, and private authority in the global economy.
Oxford: Oxford University Press.
Bellace, J. (2014). The ILO and the right to strike. International Labour Review, 153(1), 29–70.
doi:10.1111/j.1564-913X.2014.00196.x
Blasi, J., & Bair, J. (2019). An analysis of multiparty bargaining models for global supply chains.
Conditions of Work and Employment Series No. 105. Geneva: International Labour Office.
Bonacich, E., & Appelbaum, R. P. (2000). Behind the label: Inequality in the Los Angeles apparel
industry. Berkeley: University of California Press.
Chan, J., Pun, N., & Selden, M. (2016). Apple, Foxconn, and China’s new working class. In R. P.
Appelbaum & N. Lichtenstein (Eds.), Achieving workers’ rights in the global economy (pp.
173–189). Ithaca: ILR Press, An Imprint of Cornell University Press.
Dicken, P. (2015). Global shift: Mapping the changing contours of the world economy (7th ed.).
New York: The Guilford Press.
Gereffi, G. (1994). The organization of buyer-driven global commodity chains: How U.S. retailers
shape overseas production networks. In G. Gereffi & M. Korzeniewicz (Eds.), Commodity
chains and global capitalism (pp. 95–122). Westport: Praeger.
Gereffi, G., & Frederick, S. (2010). The global apparel value chain: Trade and the crisis. Policy
Research Working Paper 5281. The World Bank Development Research Group, Washington,
DC.
Gereffi, G., Humphrey, J., & Sturgeon, T. (2005). The governance of global value chains. Review
of International Political Economy, 12(1), 78–104. doi:10.1080/09692290500049805
346 M. ANNER

Gereffi, G., Korzeniewicz, M., & Korzeniewicz, R. (1994). Introduction: Global commodity chains.
In G. Gereffi & M. Korzeniewicz (Eds.), Commodity chains and global capitalism (pp. 1–14).
Westport: Praeger.
Hamilton, G. G., & Petrovic, M. (2011). Retailers as market makers. In G. G. Hamilton, M.
Petrovic & B. Senauer (Eds.), The market makers: How retailers are reshaping the global econ-
omy (pp. 31–49). Oxford: Oxford University Press.
Hasan, M. R. (2019). Total cost of ownership of responsible apparel production in low cost country
(Doctor of Philosophy). Textile Technology Management, North Carolina State University,
Raleigh.
Henderson, J., Dicken, P., Hess, M., Coe, N., & Yeung, H. W.-C. (2002). Global production net-
works and the analysis of economic development. Review of International Political Economy, 9,
436–464. doi:10.1080/09692290210150842
Hicks, J. R. (1932). The theory of wages. New York: Macmillan. doi:10.1086/ahr/38.4.756
ILO. (2008). ILO declaration on social justice for a fair globalization, adopted by the international
labour conference at its ninety-seventh session. Geneva: International Labour Organization.
ILO. (2016). Decent work in global supply chains: Report IV for international labour conference,
105th session. Geneva: International Labour Office.
ILO. (2017). Purchasing practices and working conditions in global supply chains. INWORK Issue
Brief No.10. Geneva: International Labour Office.
Kabeer, N., Haq, L., & Sulaiman, M. (2019). Multi-stakeholder initiatives in Bangladesh after Rana
Plaza: Global norms and workers’ perspectives. Working Paper Series 2019, No. 19-193.
London: International Development, London School of Economics.
Kotha, S., & Basu, S. (2011). Amazon and eBay: Online retailers as market makers. In G. G.
Hamilton, M. Petrovic & B. Senauer (Eds.), The market makers: How retailers are reshaping the
global economy (pp. 155–180). Oxford: Oxford University Press.
Kucera, D., & Sari, D. (forthcoming). New labour rights indicators: Method and trends for
2000–2015. International Labour Review.
Lakhani, T., Kuruvilla, S., & Avgar, A. (2013). From the firm to the network: Global value chains
and employment relations theory. British Journal of Industrial Relations, 51, 440–472. doi:
10.1111/bjir.12015
Lichtenstein, N. (2009). The retail revolution: How Wal-Mart created a brave new world of busi-
ness. New York: Metropolitan Books.
Locke, R. (2013). The promise and limits of private power: Promoting labor standards in a global
economy. New York: Cambridge University Press.
Locke, R., Amengual, M., & Mangla, A. (2009). Virtue out of necessity? Compliance, commitment,
and the improvement of labor conditions in global supply chains. Politics and Society, 37,
319–351. doi:10.1177/0032329209338922
Milberg, W. (2008). Shifting sources and uses of profits: Sustaining US financialization with global
value chains. Economy and Society, 37, 420–451. doi:10.1080/03085140802172706
Nova, S., & Wegemer, C. (2016). Outsourcing horror: Why apparel workers are still dying, one
hundred years after triangle shirtwaist. In R. P. Appelbaum & N. Lichtenstein (Eds.), Achieving
workers’ rights in the global economy (pp. 17–31). Ithaca: Cornell University Press.
O’Rourke Group Partners. (2011). Benchmarking the Competitiveness of Nicaragua’s Apparel
Industry. Prepared for CARANA Corporation Enterprise & Employment Project (USAID) on
behalf of PRONicaragua, Ventnor, NY.
Phillips, N. (2017). Power and inequality in the global political economy. International Affairs, 93,
429–444. doi:10.1093/ia/iix019
Piore, M. (1997). The economics of sweatshops. In A. Ross (Ed.), No sweat: Fashion, free trade,
and the rights of garment workers (pp. 135–142). New York: Verso.
Raworth, K., & Kidder, T. (2009). Mimicking ‘lean’ in global value chains: It’s the workers who
get leaned on. In J. Bair (Ed.), Frontiers of commodity chain research (pp. 165–189). Stanford:
Stanford University Press.
Riisgaard, L., & Hammer, N. (2011). Prospects for labour in global value chains: Labour standards
in the cut flower and banana industries. British Journal of Industrial Relations, 49(1), 168–190.
doi:10.1111/j.1467-8543.2009.00744.x
 REVIEW OF INTERNATIONAL POLITICAL ECONOMY 347

Ross, R. J. S. (2016). The twilight of CSR: Life and death illuminated by fire. In R. P. Appelbaum
& N. Lichtenstein (Eds.), Achieving workers’ rights in the global economy (pp. 70–92). Ithaca:
ILR Press, An Imprint of Cornell University Press.
Seidman, G. W. (2007). Beyond the boycott: Labor rights, human rights, and transnational activism.
New York: Russell Sage Foundation.
Siddiqi, D. (2017). Before Rana Plaza: Toward a history of labor organizing in Bangladesh’s gar-
ment industry. In V. Crinis & A. Vickers (Eds.), Labour in the clothing industry in the Asia
Pacific. London: Routledge.
Staritz, C. (2011). Making the cut? Low-income countries and the global clothing value chain in a
post-quota and post-crisis world. A World Bank study. Washington, DC: The World Bank
Group.
Taplin, I. M. (2014). Who is to blame? A re-examination of fast fashion after the 2013 factory dis-
aster in Bangladesh. Critical Perspectives on International Business, 10, 72–83.
Weil, D. (2014). The fissured workplace: Why work became so bad for so many and what can be
done to improve it. Cambridge, MA: Harvard University Press.
WRC. (2019). Banning hope: Bangladesh garment workers seeking a dollar an hour face mass fir-
ings, violence, and false arrest. Washington: Worker Rights Consortium.