Content uploaded by Luisa Lupo
Author content
All content in this area was uploaded by Luisa Lupo on Jan 24, 2025
Content may be subject to copyright.
Do promises of support
from distant buyers bolster
or undermine local
demands for reform?
Evidence from the
Indonesian apparel industry
during the pandemic
Luisa Lupo
Geneva Graduate Institute, Switzerland
Tim Bartley
Georgetown University, USA
Abstract
Calls to improve labor conditions in global supply chains frequently urge more robust
responsibility-taking by brands/buyers, various forms of state intervention, and respect
for worker collective action. In this article, we examine perceptions of these paths to
reform among workers and managers in Indonesia’s export-oriented apparel sector dur-
ing the COVID-19 pandemic. Extending research on labor regimes and global supply
chains, we consider how both workplace cleavages and signals from buyers may shape
local demands for reform. We use survey data from 96 factories to gauge perceptions
among workers and managers, including a survey experiment on the effects of a distant
brand/buyer pledge of support. Amidst conflicting interests and pandemic-era precarity,
we find broad agreement among workers and managers on some measures, including
local government intervention, general references to collective action, and even the
contested issue of labor market flexibilization. A hypothetical promise of support
from distant buyers had limited effects overall but did dampen some local demands
Corresponding author:
Luisa Lupo, Department of International Relations and Political Science, Geneva Graduate Institute, Chem.
Eugène-Rigot 2, 1202 Geneva, Switzerland.
Email: luisa.lupo@graduateinstitute.ch
Article
Journal of Industrial Relations
1–26
© Australian Labour and Employment
Relations Association (ALERA) 2024
SAGE Publications Ltd, Los Angeles,
London, New Delhi, Singapore and
Washington DC
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/00221856241300714
journals.sagepub.com/home/jir
among production workers. These findings imply that while combining responsible sour-
cing from above, mobilization from below, and state intervention is often desirable,
buyer initiatives could hinder grassroots mobilization in some circumstances.
Keywords
Indonesia, industrial relations, corporate social responsibility (CSR), supply chains, labor
regulation, local reform
The COVID-19 pandemic added insecurity to already-precarious global supply chains.
As markets stalled and large buyers shifted or canceled orders, there were cascading
negative consequences for employment and incomes, from factory closures to layoffs,
unpaid leave, and delayed wage payments (ILO, 2020a; Foley and Piper, 2021; Judd
et al., 2022). These disruptions not only increased power asymmetries between
workers and employers but also catalyzed regulatory reforms and local solidarity-
building efforts (Anner, 2022). While several studies have documented how the pan-
demic shaped labor markets, trade unions, and labor law (Anner, 2022, 2020; Ford
and Ward, 2021; Tejani and Fukuda-Parr, 2021), much less is known about the percep-
tions of workers and managers at the point of production during this moment.
More broadly, despite decades of research on labor in global supply chains, scholars
know relatively little about how workers and managers view various paths to reform.
Scholars have studied high-profile transnational labor rights campaigns (e.g., Connor
and Haines, 2013), historic corporate accountability projects (such as the Bangladesh
Accord on Fire and Building Safety), and contentious reforms of industrial relations insti-
tutions (e.g., Alford and Phillips, 2018; Caraway, 2023), but many questions remain
about how concrete strategies for improving immediate circumstances are viewed—
and how various interventions might intersect.
The objective of this paper is to fill this gap by posing two main research questions:
First, under what conditions do workers and managers in global supply chains see various
kinds of local interventions as helpful? Answering this question involves examining per-
ceptions of several types of potential interventions, broadly conceived, including special
accommodations from factory owners during moments of crisis, varied forms of local or
national government support, and collective action by workers. Second, are workers’and
managers’local demands for reform shaped by promises of support from distant brands/
buyers? A growing body of research indicates that promises from purportedly responsible
brands and retailers—apart from any direct influence on production conditions they might
have—can influence other actors’perceptions of what is desirable and feasible. To date,
this idea has been assessed by looking at the effects of hypothetical promises of voluntary
reform on the perceptions of citizen-consumers in affluent countries (Malhotra et al., 2019;
Kolcava et al., 2019; Amengual and Bartley, 2022) and, to a lesser extent, factory managers
in global industries (Malesky and Mosley, 2018). But workers’perceptions could similarly
be influenced by distant corporate promises of support—that is, promises to help workers
and factories through a difficult period—whether by bolstering their confidence or under-
mining demands for more proximate local reforms.
2Journal of Industrial Relations 0(0)
Two bodies of scholarship provide foundations for this research. First, research on
labor regimes highlights durable configurations of power and inequality within the work-
place and at larger scales (Baglioni et al., 2022). This literature draws attention to a
number of cleavages—from divisions between management and organized labor to gen-
dered power structures—that are expected to structure perceptions of reform. Second, a
growing literature on the political economy of signals and incentives in global supply
chains draws attention to actions and statements from distant buyers that may foster or
undermine local improvements (Distelhorst and Locke, 2018). This includes both hope
that high expectations in hypothetical destination markets can promote social upgrading
(Malesky and Mosley, 2018) and concerns that corporate social responsibility (CSR)
initiatives serve to undermine other pathways to reform (LeBaron and Lister, 2021).
Our analysis is also situated within a specific context of calls for greater responsibility-
taking by brands/buyers during the COVID-19 pandemic. Amidst controversy over can-
celed orders, some apparel brands and retailers pledged to pay in full or otherwise support
suppliers through the COVID-19 pandemic, whether by endorsing the ILO’s Call to
Action in the Global Garment Industry or issuing their own commitments (ILO,
2020b; Worker Rights Consortium, 2021). We seek to understand if commitments of
this sort affect perceptions of workers and managers.
We address these questions through a study of the apparel industry in Indonesia, where
many export-oriented factories experienced severe disruptions from lockdowns, canceled
orders, and delayed shipments (Ford and Ward, 2021). Alongside a variety of local and
factory-level efforts to cope with this crisis, the Indonesian national government intro-
duced a highly controversial Omnibus Law on Job Creation, which sought to make
labor markets more flexible and reduce labor and environmental regulations.
Labor-intensive factories in Indonesia have thus become especially important sites for
understanding the intertwining of the contentious domestic politics of labor relations
and the challenge of precarity in global supply chains (see Ford et al., 2023). In addition,
since many apparel factories in Indonesia produce for reputation-sensitive brands and
retailers, this is an important location for assessing whether promises of support from
distant buyers can shape local demands for reform. Researchers have previously exam-
ined Indonesian trade unions’attempts to gain leverage from corporate reputations and
responsibility norms (Connor and Haines, 2013; Bartley and Egels-Zandén, 2016;
Siegmann et al., 2017), but this work has been focused on particular transnational brand-
boomerang campaigns, often without comparable attention to domestic reform efforts.
Amengual and Chirot (2016), in contrast, examined the combination of transnational
and domestic labor regulation, with a particular focus on the interplay between the
ILO-IFC Better Work Indonesia program and local government actors. We seek to
follow in their footsteps, with a topically similar but methodologically different approach.
Our analysis is based on surveys of workers and managers in 96 export-oriented
apparel factories, conducted in 2020 and 2021. This individual-level survey data
allows us to examine how different types of employees evaluated various local interven-
tions, from collective action by workers to labor market flexibilization. The survey also
included an experimental component, in which we randomly assigned some respondents
to see a pledge of support from international brands and retailers. This allows us to see
Lupo and Bartley 3
whether, how, and for whom promises of support from foreign brands shape local
demands for reform.
We proceed by contextualizing COVID-19 era disruptions in the Indonesian apparel
industry and the variety of interventions that emerged in response. Drawing on labor
regimes scholarship and studies of signals and incentives in global supply chains, we
then consider the factors that are most likely to structure workers’and managers’percep-
tions of reform in this context. This leads us to consider how responsibility-taking from
distant buyers might shape local demands for reform, extending research on other settings
and audiences. After describing our survey data, we present the results of our observa-
tional and experimental analyses. We conclude by discussing the implications of our find-
ings for interventions combining responsible sourcing from above, mobilization from
below, and state intervention.
Supply chain disruptions and contentious reforms in Indonesia
Our study examines local demands for reform in an industry characterized by instability
at an especially precarious moment amid the global disruptions caused by the COVID-19
pandemic (Frenkel and Schuessler, 2021). The global economy was thrown into profound
disarray, with an estimated 88–93 million people worldwide being pushed into extreme
poverty by pandemic-related economic shocks (Mahler et al., 2020). With children out of
school and overwhelmed health services, unpaid care workloads at home intensified,
especially for women (UN Women, 2020; Kabeer et al., 2021). Disruptions to global
supply chains—for apparel among many other products—were unprecedented in both
upstream and downstream locations.
The suspension of transportation of raw materials from China during the first months
of the COVID-19 outbreak was soon compounded by drops in consumer demand for
apparel in Europe and North America as shutdowns expanded. A wave of buyer cancel-
lations, postponements, and refusals to pay for orders, either completed or in progress,
triggered adverse effects in supplier factories, especially in Asia (Anner, 2022; Judd
et al., 2022). An estimated 60% of garment manufacturers in Asia and the Pacific experi-
enced disruptions in the supply of intermediary products and raw materials, while apparel
imports dropped by approximately 70% in the first half of 2020 (ILO, 2020a).
In Indonesia, social restrictions were first put in place in April 2020, when the govern-
ment declared a health emergency and introduced specific measures on a provincial basis.
The apparel industry in the country was severely hit, with layoffs and dismissals affecting
over 350,000 workers (FW, 2021). In April 2020, according to the Indonesian Textile
Association, COVID-19 restrictions halted operations at least temporarily in 80% of
apparel factories across the country, while exports reduced by 60% between May and
June 2020 relative to the previous year (ILO, 2020a). Among factories in Indonesia par-
ticipating in the ILO-IFC Better Work program, nearly 20,000 workers lost their jobs and
more than 165,000 workers had reduced working days, hours or wages, with short-term
contract workers and women bearing the brunt of the crisis (BW, 2021).
Research by Ford et al. (2023) highlights various strategies that trade unions and gov-
ernment agencies in Indonesia took in response to these disruptions. While government
4Journal of Industrial Relations 0(0)
agencies used wage subsidies, credit provision, and social insurance programs to ease the
pain among factories and workers, trade unions fought to preserve job security and to
improve workplace safety, as well as providing direct assistance to displaced workers.
In the midst of the pandemic, though, the Indonesian government also pushed through
the Omnibus Bill on Job Creation, which sought to flexibilize labor markets and
reduce regulation. Among many other things, the Omnibus Bill proposed to change
minimum wage-setting processes, reduce the scope of industrial courts, restrict unions’
capacity to intervene against dismissals, and curtail severance payments, which have
been a central safety net for Indonesian workers. As Ford and Ward (2021) detail,
trade unions had long contested this sort of reform, and they mobilized during the pan-
demic to threaten large-scale protests against the Omnibus Bill. Unions agreed to nego-
tiate instead, but these proved fruitless, and the Omnibus Bill was quickly passed into law
in October 2020, precipitating a wave of strikes. The Indonesian constitutional court
rejected some parts of the law in 2021, but the Jokowi administration responded with
a wave of emergency regulations that accomplished similar goals in late 2022. In
2023, these regulations became permanent, and the Omnibus Law on Job Creation was
enacted.
In essence, reforms amidst the COVID-19 pandemic in Indonesia reflected a mix of
efforts “from above”and “from below”to buffer firms and workers from supply chain
disruptions and to push through labor market flexibilization reforms that had long been
demanded by the international business community. Brands and retailers sourcing
from Indonesia took few meaningful steps to promote workplace safety (Ford et al.,
2023), but some did eventually pledge to pay for their orders and otherwise support sup-
plier factories through this tumultuous period, rather than completely disavowing respon-
sibility (ILO, 2020b; Worker Rights Consortium, 2021). More broadly, and beyond
Indonesia, this period was marked by a variety of labor mobilizations, from collective
action among healthcare workers around the world (Trappmann et al., 2022) to strikes
against workers’displacement, some of which were repressed (Ford et al., 2023).
Against this background, we seek to understand the perceptions of workers and managers
themselves, to which we turn in the next section.
Theoretical foundations
Despite large bodies of research on global value chains, scholars have only rarely sought
to measure perceptions of workers and managers (but see Kabeer et al., 2020;
Wickramasingha, 2023; Holzberg, 2023). In this section, we develop a theoretical frame-
work for our inquiry, with an emphasis on the factors that are expected to pattern percep-
tions of reform within precarious and unequal supply chain structures. We draw guidance
from two main bodies of scholarship: First, theories of labor regimes, which emerged
from neo-Marxist theories of the labor process, highlight a range of durable configura-
tions of power and inequality that are reflected in and reproduced in workplaces embed-
ded in global industries (Baglioni et al., 2022; Frenkel et al., 2022). While this is
primarily a qualitative and case-based body of research, it highlights several categorical
distinctions that can be examined in our survey data. Second, research on the political
Lupo and Bartley 5
economy of global value chains (Barrientos, 2019) has increasingly drawn attention to the
incentives and signals that flow from lead firms to various audiences—from factory man-
agers looking to upgrade their firms’positions (Malesky and Mosley, 2018) to consumers
trying to assess fairness (Steinberg, 2023). This research suggests a different set of factors
that are likely to influence perceptions of reform at the point of production, as well as
varying accounts of the direction of influence. We take both bodies of research as foun-
dations for our study.
Labor regimes and the structuring of perceptions
A growing body of scholarship on labor regimes provides tools for understanding
inequality and power in global industries. Originating with neo-Marxist accounts of con-
flict on the shop floor (Burawoy, 1985), the labor regimes approach emphasizes that what
appear to be mere “labor markets”reflect a profound influence of institutions and social
relations that “bind capital and labour in a form of antagonistic relative stability in par-
ticular times and places”(Baglioni et al., 2022: 1). This perspective requires attention
to collective action by workers and the socio-political factors that unite, divide,
repress, or amplify their agency (Wickramasingha, 2023). It also requires attention to
structures of production that recruit and shape different types of workers, whether oper-
ating at factory, national, or global levels (Frenkel et al., 2022). In Indonesia, labor
regimes in manufacturing sectors (especially labor-intensive sectors such as apparel)
over the past two decades have combined a seemingly protective set of labor regulations,
growing precarity in practice, and a trade union movement that has been fragmented at
the workplace level but increasingly politically active (Caraway and Ford, 2022).
While our study is concerned with perceptions of reform rather than variations among
labor regimes, we draw guidance from this literature’s emphasis on durable antagonisms
among different groups of employees. Perhaps most fundamentally, labor regimes schol-
arship foregrounds trade unions as central and distinctive actors in the politics of produc-
tion, while also recognizing varieties of unionism and the role of disorganized labor
unrest. While some trade unions are far more independent, insurgent, and activist-minded
than others (Caraway, 2008), even moderate and management-friendly unions have been
at the forefront of some efforts to promote labor rights in global supply chains (Bartley
and Egels-Zandén, 2016; Caraway and Ford, 2022). At the individual level, then,
union membership should generally be associated with more demanding stances on
local reform, regardless of whether this is a result of union membership or simply reflects
differential selection into union membership.
The distinction between production workers and managers/supervisors is also
expected to be stark, since these groups most clearly represent antagonistic interests on
the shop floor. Although many production workers in global supply chains are margin-
alized and precarious, they have nevertheless been at the forefront of many labor strug-
gles (Anner, 2022; Lee, 2007). In contrast, employees in managerial or supervisory roles,
even at low levels of the hierarchy, may be more likely to identify with top factory man-
agers and/or worry about production workers becoming disruptive in their demands.
Workers in nonmanagerial and nonproduction occupations (such as security guards,
6Journal of Industrial Relations 0(0)
clerical workers) are in ambiguous positions; sometimes they have joined with produc-
tion workers in making collective demands, yet they may not have strong ties to produc-
tion workers or dramatic experiences of mistreatment on the factory floor.
Research on gendered labor regimes examines how patriarchal social relations have
been capitalized on and reproduced in global production networks (Bair, 2022).
Feminist scholars have long drawn attention to the “cheapening”of women’s work as
an underpinning of global production (Enloe, 2000; Elson and Pearson, 1981). The
sorting of workers into occupations within factories is also deeply gendered, as managers
prioritize gender in constructing the norms for “ideal workers”for different tasks (Pun,
2005). In the apparel industry, this has often meant that women have been sorted into
low-wage and precarious positions, often involving sewing or finishing, while men
have been channeled into operating cutting machines or into managerial/supervisory
positions (Barrientos, 2022). Beyond occupational differences, women are more likely
to be expected to bear the costs of social reproduction, even when they are also
working long hours in factories.
For these reasons, one would expect women, as multiply marginalized workers, to be
more aggrieved and demanding than men on average, perhaps especially in the wake of
the COVID-19 pandemic, which disrupted school and childcare as well as manufacturing
in Indonesia (Kusumaningrum et al., 2022). On the other hand, even when women are
overrepresented in the workforce, they are underrepresented in union leadership roles
(Ford, 2009) and often excluded from arenas of collective empowerment. Moreover,
since women are more likely to be in precarious positions and have shorter employment
histories (Mezzadri and Majumder, 2020), they may feel more vulnerable, less strident,
and perhaps more obliged to feel like events are beyond their control. Taken together,
while prior research suggests that gender nearly always a structuring force in global
production networks, there are several possibilities for precisely how it may be related
to perceptions of local reform in our study.
The political economy of supply chain signals and incentives
A different set of insights comes from a growing body of research on how perceptions of
reform are shaped by signals and incentives in supply chains. Research on the political
economy of global value chains has long emphasized the power of lead firms that coord-
inate dispersed sites of production (Gereffiet al., 2005), and several recent strands of
work draw attention to the messages, signals, and incentives for reform that these
firms may provide. Here, we describe three different expectations about how these mes-
sages may matter, with the goal of applying these to our study of hypothetical promises of
support from lead firms.
First, proponents of social upgrading argue that responsible corporations can—and
sometimes do—provide incentives for improving working conditions in supplier firms
(Greenhill et al., 2009; Distelhorst and Locke, 2018; Gereffiet al., 2005). Some evidence
suggests that buyers’voluntary commitments can thereby bolster local reforms. Based on
a survey experiment in export-oriented factories in Vietnam, Malesky and Mosely (2018)
find that factory managers were more willing to invest on labor improvements when told
Lupo and Bartley 7
they would be selling to markets with more responsible, reputation-sensitive brands.
Studying the apparel industry in Cambodia, Oka (2010) shows how producing for
reputation-sensitive buyers can improve labor conditions, especially when combined
with transnational advocacy or on-site interventions. Studying campaigns by unions
and labor NGOs in Indonesia, Bartley and Egels-Zandén (2016) find that CSR commit-
ments of buyers and brands could sometimes be leveraged into concrete gains for acti-
vists, even if these were less than desired.
More broadly, scholars of social upgrading in global supply chains often find that con-
ditions are generally better (though not across the board) and workers are sometimes
more empowered (though not always so) when producing for reputation-sensitive
firms than when producing for undifferentiated markets (Barrientos, 2019; Alford and
Phillips, 2018; Oka, 2012). By these accounts, when workers and managers receive infor-
mation about buyers/brands being responsible, it should bolster their demands for local
reforms, in essence by giving them confidence that reforms will be supported and
rewarded.
A second strand of research suggests that messages about corporate responsibility can
have the opposite effect, by reducing pressure for change, providing an illusory sense that
problems are addressed, and crowding out other types of interventions. Corporate pro-
mises of reform have often served to guard corporate power and existing systemic imbal-
ances, deflecting attention from systemic causes of violations and bringing hidden costs
that weigh more heavily on vulnerable workers and local communities (LeBaron and
Lister, 2021; Keahey and Murray, 2017).
We know of no prior research that has systematically examined this possibility among
workers and managers directly. But a growing body of research shows that consumer-
citizens in affluent countries can indeed have their concerns ameliorated by rather thin
hypothetical assurances of corporate responsibility. Specifically, survey experiments con-
ducted in the USA (Malhotra et al., 2019; Amengual and Bartley, 2022), Switzerland
(Kolcava et al., 2019), and Germany (Bartley et al., 2023) have found that public
support for state intervention decreases when individuals are told that companies are
taking voluntary measures. Though not all corporate statements resonate equally
(Amengual and Bartley, 2022), researchers typically find that even vague corporate pro-
mises with minimal credibility can appease observers’concerns. If a similar dynamic
operates among workers and managers at the point of production, this would suggest a
dampening effect, whereby promises of support from distant buyers/brands decrease
the perceived utility of local interventions. To be clear, we cannot test this possibility
across times and places, but we can at least see if there is evidence of it among one set
of employees navigating a moment of instability and regulatory change.
A third idea posits that corporate promises of responsibility may be viewed quite dif-
ferently by managers and workers, consistent with the labor regimes research described
above. Managers may see buyer responsibility as an opportunity for upgrading, greater
material rewards, and collaborative problem-solving, consistent with Malesky and
Mosley (2018) and Distelhorst and Locke (2018). But for workers, the involvement of
distant buyers may look like a rescue from above, rather than boosting their confidence
to press for change from below. Workers may even see buyer involvement as making
8Journal of Industrial Relations 0(0)
local reforms harder, especially during precarious times, since there is a constant threat
that factories may lose orders if workers are too demanding (Anner, 2022). In line with
this research, one would expect a promise of support from distant buyers/brands to have a
heterogeneous effect; it might increase the perceived utility of local interventions among
managers/supervisors but decrease the perceived utility of local interventions among pro-
duction workers.
Amidst the devastating effects of the pandemic on working conditions and livelihoods
and mounting demands for action among workers and their representatives, it remains to
be seen if voluntary corporate responsibility has any effects at all among employees at the
point of production. It may be that distant promises are largely irrelevant, especially when
domestic/local reforms are highly contentious, as they have been with regard to the flex-
ibilization of labor markets.
Methodology and data
Our analysis is based on surveys conducted with workers and managers employed in
export-oriented apparel factories across five provinces/regions of Java, the main
apparel manufacturing hub in the country, during the COVID-19 pandemic. This pro-
vides a unique and valuable source of data for both observing patterns in perceptions
of reforms within Indonesia and examining the influence of a promise of support from
distant brands/buyers. While case study and interview-based research haves revealed
key contours of contention in this setting (e.g., Ford and Ward, 2021) a quantitative ana-
lysis of survey data has the potential to highlight structured similarities and differences in
perceptions across large numbers of employees (e.g., Frenkel et al., 2022; Li and
Kuruvilla, 2023).
The surveys were conducted by the ILO-IFC Better Work program, with our input,
and focused on members of legally mandated bipartite worker-management committees
in 96 export-oriented apparel factories that were part of the Better Work Indonesia
program. Generally speaking, members of the committees comprise an equal number
of managers and workers (both union and nonunion members). In Indonesia, the commit-
tees are mandated by law (under the name of Lembaga Kerja Sama Bipartit or LKSB) and
are tasked with discussing and addressing issues that pertain to working conditions.
While the management members are appointed by the employer, the law provides that
worker members are elected by workers through a democratic process. Research on
Better Work programs elsewhere suggests that committee members tend to be knowl-
edgeable of factories’challenges and buyers’codes/commitments, and that they play
an active role in advancing improvements (Pike, 2020; Afros, 2022). This makes our
sample particularly useful for assessing the influence of promises from distant brands/
buyers, since workers and managers in these factories would be generally aware of the
dynamics of corporate responsibility, whether through dissemination/posting of work-
place standards, training programs, or the experience of being audited. Still, because
our respondents are committee members in export-oriented factories producing at least
in part for reputation-sensitive brands/buyers that are members of Better Work (such
as Ralph Lauren, J. Crew, C&A, H&M), their views should not be taken as representative
Lupo and Bartley 9
of the industry as a whole. Nevertheless, our sample includes factories in five different
provinces in different parts of Indonesia, from the older manufacturing regions West
Java, Banten and the Greater Jakarta Area to the more recent frontiers in Central Java.
Our results should be interpreted with these contours and limits of the sample in mind.
From October to November 2020, workers and managers were asked to complete a
self-administered survey with remote support, delivered by Better Work Indonesia
field staff and a team of trained consultants. This resulted in a sample of 696 individuals
who responded to our question about the perceived utility of different types of interven-
tions. The respondents were roughly equally split between managers and workers, and
slightly more production workers than nonproduction workers. (See Table 1 for a full
set of descriptive statistics.) A second survey was conducted in November to
December 2021, yielding a sample of 850 individuals who responded to our question
gauging optimism/pessimism about factory-level reforms in general. While responses
were confidential, the involvement of Better Work staff in administering the surveys
could have limited contentious or socially undesirable responses, though the extent of
this is unlikely to vary across the sub-groups in our analysis or across the experimental
conditions, which were randomly assigned.
Table 1. Descriptive statistics.
Mean SD Min Max p50 p75 Observations
Perceptions of local interventions
Government support 4.08 1.05 1 5 4 5 696
Factory owner support 4.06 1.01 1 5 4 5 624
Workers’collective action 4.21 0.77 1 5 4 5 628
Labor flexibilization 3.26 1.34 1 5 4 4 559
Childcare support 3.63 1.27 1 5 4 5 589
Optimism about factory reforms 3.43 1.04 1 5 4 4 850
Buyer promises 0.46 0.50 0 1 0 1 780
Union 0.76 0.71 0 2 1 1 1660
Female 0.44 0.50 0 1 0 1 1651
Age 36.87 8.68 19 60 36 43 1648
Education 5.15 0.72 1 6 5 6 1443
Children 0.85 0.36 0 1 1 1 1549
Job position
Manager/supervisor 0.48 0.50 0 1 0 1 1563
Nonproduction worker 0.23 0.42 0 1 0 0 1563
Production worker 0.29 0.45 0 1 0 1 1563
Province
Banten 0.03 0.17 0 1 0 0 1619
Central Java 0.35 0.48 0 1 0 1 1619
Jakarta 0.03 0.16 0 1 0 0 1619
West Java 0.57 0.49 0 1 1 1 1619
Yogyakarta 0.02 0.15 0 1 0 0 1619
10 Journal of Industrial Relations 0(0)
To measure the perceived utility of various local reforms, respondents were
briefly reminded of the disruptions of COVID-19 and asked what should be done
to support workers in their factory. They were then shown five potentially relevant
interventions (listed below) and asked to rate each on a five-point scale from “Not
helpfulatall”to “Extremely helpful.”This scale resembles the widely used five-
point Likert scale of agreement-disagreement(Joshietal.,2015)butadaptsitto
fit our research question about the extent to which various interventions are per-
ceived as helpful in a period of crisis. Simple 5-point helpfulness scales have per-
formed reasonably well in other settings (Di Malta et al., 2023) and were used
here to keep response categories simple. Respondents were asked to rate the helpful-
ness of five types of interventions:
1. The government should take measures to prevent layoffs;
2. Factory owners should make sure that workers are paid even if they are laid off or
suspended;
3. Workers should work together to protect their rights as much as necessary;
4. The government should give factories more flexibility to hire and lay off workers as
needed; and
5. Workers with children should be able to go on paid leave to take care of them even
when production has restarted.
These were intended to cover a range of relevant demands directed at various power-
holders, including the government and factory owners. Statement (3) highlights collect-
ive action by workers themselves allowing us to examine perceptions of reform “from
below”relative to requests for help “from above.”This statement was phrased in a com-
prehensive manner, potentially encompassing a broad range of actions, from coopera-
tive to more contentious forms of collective action. Statement (4) focuses on making
labor markets more flexible, as was being pursued through the Omnibus Law.
Statement (5) specifically focuses on childcare and social reproduction, given the
high salience of these issues during the pandemic. Given the widespread nature of
layoffs and furloughs, the governments’project to push through labor market flexibili-
zation reforms, and the intensification of domestic care responsibilities, these state-
ments cover a range of locally-relevant reforms that should be salient to both
managers and workers.
In addition to being interested in perceptions of these interventions in general, we
developed a survey experiment to see whether these perceptions could be shaped by a
simple (and slight) cue about corporate responsibility by distant buyers. Prior to rating
the helpfulness of the five interventions mentioned above, all respondents read the fol-
lowing text:
The disruptions of COVID-19 have raised many new questions. As you think about the next two
months, what do you think should be done to support workers in your factory? Please tell us
whether each action below would be helpful or not.
Lupo and Bartley 11
For respondents that were randomly assigned to the treatment group, the first sentence
(“The disruptions of COVID-19 have raised many new questions.”) was followed by a
cue about promises of reform by distant buyers/brands:
In response, many international buyers have committed to helping factories and workers survive.
The introductory text then continued as with the control group.
Survey experiments have become a standard tool in the social sciences, including to
identify causal impacts of particular pieces of information or framings of a situation, and
unpack belief systems (Mutz, 2011; Wallander, 2009). Our simple framing experiment pro-
vided a small piece of information that we expect would evoke an additional set of ideas—
about CSR—that was not explicitly discussed elsewhere in the survey. In essence, our
survey experiment allows us to see if the mere mention of international buyers’promises
of support makes workers and managers in export-oriented Indonesian factories think dif-
ferently about what sorts of reforms should be going on within Indonesia.
While survey experiments cannot approximate the real experience of participating in
contentious political activities, they can reveal important assumptions of practical judg-
ment and subtle contours of collective perceptions that might otherwise be obscured. To
supplement these responses, we presented respondents (in 2021) with a statement about
factory managers, workers, and buyers taking steps to improve the treatment and well-
being of workers. We then asked, “When this happens, how often do you think it
leads to real improvements?”with responses measured on a 5-point scale from “None
of the time”to “All the time”(a popular variant of a Likert scale). This allows us to
measure the perceived effectiveness of factory-level reforms in general, independent of
any particular promises from brands or retailers.
Since our responses come from employees nested in factories, we use multi-level
models (Robson and Pevalin, 2015) when analyzing the nonexperimental effects. While
our variables of theoretical interest are at the individual level, multi-level models (with
random effects at the factory level) both control for nonindependence of observations
(among individuals within the same factory) and estimate the degree to which variation
in our dependent variable is clustered by factory. Previous research has similarly used
multi-level models to ensure accurate estimates with nested individual-factory data
(Kuruvilla et al., 2020). Additional information is available in the Appendices.
Results
We first examine workers’and managers’views of worker-friendly reforms in general
and then turn to perceptions of different types of local interventions and our survey
experiment of buyer/brand promises.
Perceptions of factory reforms
To what extent do different types of employees in export-oriented Indonesian apparel fac-
tories think that factory-level reforms are meaningful or merely illusory? As shown in
Table 2, there are notable cleavages in these perceptions. As expected, production
12 Journal of Industrial Relations 0(0)
workers are significantly less optimistic about factory-level reforms than are workers in
managerial/supervisory roles (scoring 0.31 points lower on our 5-point scale in the base-
line model on average). This difference persists when using multi-level models to capture
the clustering of perceptions within factories, which could capture prior accommodations
of factory owners. Similarly, union members are less optimistic than nonmembers (0.24
points lower on our 5-point scale on average). In line with labor regimes scholarship on
the patterning of antagonistic interests on the shop floor, these findings likely reflect pro-
duction workers’and union members’more demanding stances on factory conditions
relative to other groups of workers.
We also find evidence of a gendered sense of the prospects for reform, with women
reporting more optimistic answers than men, even when controlling for position and
Table 2. Perceptions of factory-level reforms.
(1)
Optimism about factory-level reforms
OLS Multi-level
Female 0.36∗∗∗ 0.33∗∗∗
(0.07) (0.07)
Age −0.00 0.00
(0.01) (0.01)
Education 0.00 0.02
(0.07) (0.06)
Children −0.04 −0.08
(0.14) (0.13)
Union −0.24∗∗
−0.24∗∗
(0.11) (0.11)
Manager/supervisor 0.00 Base category
Nonproduction worker −0.19∗
−0.16
(0.10) (0.10)
Production worker −0.31∗∗
−0.22∗
(0.12) (0.11)
Constant 3.52∗∗∗ 3.41∗∗∗ 3.40∗∗∗
(0.50) (0.05) (0.45)
Random-effects parameters
Constant 0.16∗∗∗ 0.13∗∗∗
(0.03) (0.04)
Residuals 0.92 0.89∗
(0.05) (0.06)
Observations (individuals) 694 843 694
Number of factories 96 96
ICC 0.15 0.13
R
2
0.06
Note. Standard errors in parentheses (OLS: clustered by factory, multi-level: robust).
∗p< .10, ∗∗p< .05, ∗∗∗p< .01.
Lupo and Bartley 13
union membership. While many accounts of women’s marginalization in the apparel
industry would suggest that women would be less optimistic about reform, we find
more optimism, perhaps indicating a higher degree of trust in factory leadership or
more constraints on their capacity to express dissatisfaction, as discussed above.
Our multi-level model reveals that optimism about reform is clustered by factory to
some degree (broadly similar to the findings about clustering in Kuruvilla et al., 2020).
A baseline intra-class correlation coefficient of 0.15 indicates that 15% of the variation
in perceptions is attributable to differences between factories. This may be based on dif-
ferent experiences of reform or different cultures of management and production. Once
we account for these average differences across factories, production workers (compared
to managers/supervisors) and union members remain less optimistic about factory-level
reforms. Yet nonproduction workers (e.g., clerical and security staff) no longer differ
from managers/supervisors, suggesting that their views might more closely align with
those of management once differences between factories are factored out.
Perceptions of local interventions
One way to buffer workers from disruptions is for governments to take measures to
prevent layoffs. The majority of workers and managers in our sample rated this sort of
intervention as quite helpful, with an overall average of 4.08 on our 5-point scale
(from “not helpful at all”to “extremely helpful”). There are no clear differences
among different types of workers in how government intervention was perceived. As
shown in the first column of Model 1, there are no statistically significant differences
between women and men; union members and nonmembers; or managers, production
workers, and nonproduction workers. This is contrary to the image of polarized percep-
tions in the labor regimes literature but perhaps understandable given the precarity of the
moment. Moreover, the multi-level model reveals little clustering by factory. A baseline
intra-class correlation coefficient of just .04 indicates that only around 4% of the variation
in perceptions is attributable to differences between factories. Put differently, the factory
where people work has little to do with their perceptions of local government
intervention.
Turning to accommodations by factory owners (to “make sure that workers are paid
even if they are laid off or suspended”), we see a similar level of support on average
(4.07 on our 5-point scale) but more structured variation. Union members and production
workers saw factory owner accommodations as significantly more important than others
did, as shown in Model 2. This is consistent with the idea that production workers and/or
union members are generally more demanding of factory owners than managers/supervi-
sors are, which can also be observed in some qualitative research on the garment industry
(Connor and Haines, 2013; Holzberg, 2023). It could also be that production workers
were more affected by the pandemic, which could have heightened their expectations
of factory owners. There is also more evidence of clustering by factory here, with 24%
of the variation in perceptions being attributable to differences between factories. This
may indicate that respondents’perceptions are linked to their sense of their own
factory owners’capacities and trustworthiness (Table 3).
14 Journal of Industrial Relations 0(0)
In Model 3, we look at perceptions of collective action by workers (that they should
“work together to protect their rights as much as necessary”). This was the most popular
form of intervention among our respondents, with an average rating of 4.21 on our
5-point scale. Views of collective action in this context and format were quite consensual,
with no significant differences between managers, production workers, and nonproduc-
tion workers; men and women; or union members and nonmembers. Note that our phras-
ing of this item encompasses a range of activities, and respondents may not have viewed
it as an endorsement of strikes or other contentious forms of collective action. So while
views of collective action were less sharply divided than most labor regimes scholarship
would expect, the pattern might be different if the question were phrased differently.
There is very little clustering by factory here (with an intra-class correlation of nearly
0), meaning that we do not see factories with significantly varying average perceptions
on this issue. The only significant effects here pertain to age, which slightly diminishes
support, and having children, which is associated with higher support.
A different sort of intervention involves making labor markets more flexible (giving
factories “more flexibility to hire and lay off workers as needed”). This was quite unpopu-
lar with our respondents (with a mean of 3.24 on our 5-point scale), likely reflecting the
controversies surrounding the Omnibus Law. Surprisingly, even those in managerial/
supervisory roles did not tend to see labor market flexibilization as helpful. There are
no significant differences based on position, union membership, gender, education, or
other factors in Model 4. We do see a bit more clustering by factory here, with 17% of
the variation being attributable to differences across factories. Nevertheless, this is a strik-
ingly nonpolarized set of reactions to labor market flexibilization, at least among workers
and managers involved with the Better Work program. It is possible that the context of the
survey (linked to the Better Work program) or de facto managerial strategies for flexibil-
ity in the apparel industry (even prior to the Omnibus Law) account for this pattern of
responses.
Finally, given how the COVID-19 pandemic disrupted children’s education and care,
we asked about whether special accommodations for workers with children (to “be able to
go on paid leave to take care of them even when production has restarted”) would be
helpful. This elicited a moderate degree of support (an average of 3.64 on our 5-point
scale), with significantly higher support from parents (not surprisingly), union
members, and production workers (compared to managers/supervisors), as shown in
Model 5. There is also a bit of clustering by factory (with 18%ss of the variation
between attributable to differences across factories), and also suggestive negative
effects of age and education once these factory-level effects are taken into account in
the multi-level models.
So far, these results portray some interventions—especially collective action and local
government support—as fairly popular among workers and managers with a variety of
characteristics. Special accommodations by factory owners are also popular, but more
so among union members and production workers than others. Notably, men and
women did not differ significantly in their perceptions of any of these interventions (at
least when controlling for these other factors), despite stark gender inequalities in
labor markets, familial expectations, and experiences of mistreatment on the job. This
Lupo and Bartley 15
Table 3. Perceptions of local interventions
(1) (2) (3) (4) (5)
Government support Factory owner support Workers’collective action Labor flexibilization Childcare support
OLS Multi-level OLS Multi-level OLS Multi-level OLS Multi-level OLS Multi-level
Female 0.10 0.10 −0.11 −0.08 0.02 0.02 0.15 0.04 −0.06 −0.11
(0.10) (0.09) (0.10) (0.10) (0.07) (0.06) (0.15) (0.14) (0.14) (0.12)
Age −0.00 −0.00 −0.00 −0.01∗
−0.02∗∗∗
−0.02∗∗∗
−0.00 −0.00 −0.01 −0.02∗∗
(0.01) (0.01) (0.01) (0.01) (0.01) (0.00) (0.01) (0.01) (0.01) (0.01)
Education 0.09 0.09 0.08 0.07 −0.01 −0.01 0.09 0.11 −0.09 −0.13∗
(0.06) (0.07) (0.06) (0.06) (0.05) (0.05) (0.11) (0.12) (0.08) (0.07)
Children 0.18 0.18 0.20 0.30∗∗∗ 0.23∗∗ 0.23∗∗ 0.25 0.24 0.42∗∗ 0.42∗∗
(0.15) (0.14) (0.15) (0.11) (0.11) (0.09) (0.21) (0.22) (0.19) (0.18)
Union −0.02 −0.02 0.22∗∗∗ 0.20∗∗∗ 0.06 0.06 0.07 0.06 0.27∗∗∗ 0.21∗∗
(0.08) (0.07) (0.07) (0.08) (0.06) (0.06) (0.12) (0.12) (0.10) (0.10)
Manager/
supervisor
Base
category
Base
category
Base
category
Base
category
Base
category
Base
category
Base
category
Base
category
Base
category
Base
category
Nonproduction
worker
−0.07 −0.08 0.02 −0.12 0.05 0.05 −0.14 −0.08 0.19 0.10
(0.13) (0.14) (0.13) (0.10) (0.09) (0.08) (0.19) (0.17) (0.17) (0.15)
Production
worker
−0.01 −0.01 0.30∗∗∗ 0.21∗0.05 0.05 0.00 0.06 0.39∗∗ 0.28∗
(0.14) (0.16) (0.12) (0.12) (0.10) (0.09) (0.19) (0.18) (0.16) (0.16)
Constant 3.63∗∗∗ 4.08∗∗∗ 3.64∗∗∗ 3.21∗∗∗ 4.07∗∗∗ 3.58∗∗∗ 4.84∗∗∗ 4.21∗∗∗ 4.84∗∗∗ 2.51∗∗∗ 3.24∗∗∗ 2.43∗∗∗ 3.71∗∗∗ 3.64∗∗∗ 4.36∗∗∗
(0.43) (0.05) (0.55) (0.46) (0.06) (0.43) (0.33) (0.03) (0.32) (0.75) (0.07) (0.82) (0.61) (0.08) (0.58)
Random-effects
parameters
Constant 0.05∗∗∗ 0.01 0.23∗∗ 0.27∗∗ 0.00 0.00 0.17∗∗∗ 0.26∗∗∗ 0.28∗∗∗ 0.31∗∗
(0.03) (0.03) (0.14) (0.15) (0.00) (0.00) (0.06) (0.11) (0.13) (0.14)
Residuals 1.06 1.09 0.74∗∗∗ 0.70∗∗∗ 0.60 0.56 1.62∗∗ ∗ 1.62∗∗∗ 1.29∗∗∗ 1.27∗∗∗
(0.08) (0.09) (0.07) (0.08) (0.57) (0.57) (0.11) (0.14) (0.09) (0.10)
Observations
(individuals)
481 692 481 433 620 433 436 624 436 387 556 387 412 585 412
Number of
factories
92 87 91 85 91 86 91 85 91 87
ICC 0.04 0.01 0.24 0.28 0.00 0.00 0.10 0.14 0.18 0.20
R20.01 0.04 0.06 0.01 0.06
Note. Standard errors in parentheses (OLS: clustered by factory, multi-level: robust).
∗p< .10, ∗∗p<.05, ∗∗∗p< .01.
16 Journal of Industrial Relations 0(0)
contrasts not only with expectations from the literature on gendered labor regimes but
also with other survey research on garment workers, which finds gendered differences
in rates of wage violations (Kuruvilla et al., 2020) and rights consciousness (Bartley,
2018). It is possible that the lack of gender differences in our study resulted from our
statements being too general to tap into gendered inequalities; further research with
more tailored questions is warranted.
Effects of buyer/brand promises
Our survey experiment was intended to assess whether perceptions of these types of local
interventions could be shaped by a small cue about promises of support from foreign
buyers. As discussed above, there are competing views about whether promises of
support from distant buyers should bolster or diminish support for local interventions
—and for whom these effects should hold. Our results provide no consistent evidence
of buyer promises affecting employees in Indonesian apparel factories across the
board. Put differently, there are no statistically significant aggregate differences
between those who saw the prompt about brand/buyer support and those who did not.
Buyer promises neither uniformly bolstered nor undermined support for local interven-
tions. Since the bipartite committee members who were our respondents are among the
most likely to have a good understanding of CSR and of its potential effects on the
ground (see Pike, 2020), it is notable that this cue had no uniform effect. This could
be driven by ambivalence about buyers’commitment, the subtlety of our cue, or hetero-
geneity in perceptions, as discussed below.
We do find some evidence of heterogeneous effects. As shown in Figure 1, distant pro-
mises of support sometimes reduced interest in local interventions among production
workers in particular (shown on the right panel of the figure). Perhaps most notably, pro-
duction workers’interest in collective action by workers was diminished by our simple
and slight cue about buyer support. Far from bolstering production workers’demands
or giving them confidence that collective action will not be punished, buyers’promises
did the opposite—seemingly making production workers less likely to express positive
views about engaging in collective action. This is discernibly different than the reaction
among those in managerial/supervisory positions, as confirmed by a statistically signifi-
cant interaction effect (‒0.27) in the regression analysis in Table 4. To the extent that pro-
duction workers were thinking of disruptive collective action, the involvement of buyers
may have been perceived as a threat, particularly in the context of canceled orders during
the pandemic.
Likewise, production workers were less demanding of childcare accommodations
when distant buyers/brands pledged their support for factories. Here too, the reaction
of production workers was significantly different than that of managers/supervisors (as
confirmed by the significant interaction effect of ‒0.62 in Table 4). Recall that this
was only a moderately popular form of intervention overall, with production workers
being more interested than managers/supervisors. The results here indicate that produc-
tion workers were less demanding when shown a slight and simple cue about buyer
support, perhaps because it pacified concerns about future instability.
Lupo and Bartley 17
When it comes to demanding help from factory owners, the pattern is slightly more
complex. Managers and production workers responded in significantly different ways
to the promise of support from distant brands/retailers (as shown in the significant inter-
action effect of ‒0.32 in Table 4), consistent with the idea of heterogeneous effects. Yet
while the groups tended toward opposite reactions (toward a bolstering effect for man-
agers/supervisors and an undermining effect for production workers), neither of these
group-specific effects was statistically significant. Put differently, a buyer promises gen-
erated different responses but not fully enough to decisively move either group from their
baseline.
Finally, when it comes to government support to prevent displacement or policies to
add flexibility to labor markets, there is no evidence of buyer promises making a differ-
ence—for either managers/supervisors or workers. While instability in orders from global
apparel buyers may have contributed to the rise of the Omnibus Law (as a way for firms in
Indonesia to cope with rapidly changing markets), and some buyers promised more sta-
bility in the wake of COVID-19, we see no evidence of our statement about buyers shift-
ing perspectives on this issue. Flexibility and short-term contracts have been contentious
issues in the Indonesian apparel industry for two decades (Amengual and Chirot, 2016),
and our findings suggest that voluntary commitments are unlikely to alter this debate.
Law and policy are more proximately related to labor contracts, and buyer commitments
may have been too shallow (see Anner, 2022) to shape perceptions.
Overall, we find no evidence to support the idea of a bolstering effect, in contrast to
expectations from previous research on reputation-sensitive firms and upgrading in
Figure 1. Effects of buyer promises on perceptions of local interventions.
18 Journal of Industrial Relations 0(0)
Table 4. Heterogeneity in the effects of buyer promises and perceptions of local interventions.
Government
support
Factory owner
support
Workers’collective
action
Labor
flexibilization
Childcare
support
Buyer promises 0.01 0.14 0.02 −0.23 0.19
(0.11) (0.11) (0.09) (0.18) (0.17)
Manager/supervisor
Buyer promises ×Manager/
supervisor
Base category Base category Base category Base category Base category
Nonproduction worker −0.02 0.05 0.18∗∗
−0.28 0.19
(0.15) (0.17) (0.09) (0.20) (0.20)
Buyer promises ×Nonproduction
worker
−0.03 −0.13 −0.13 0.26 −0.25
(0.17) (0.19) (0.13) (0.29) (0.27)
Production worker 0.03 0.36∗∗ 0.32∗∗∗
−0.33 0.76∗∗∗
(0.15) (0.18) (0.10) (0.21) (0.20)
Buyer promises ×Production
worker
−0.19 −0.32∗
−0.27∗0.41 −0.62∗∗∗
(0.19) (0.18) (0.14) (0.29) (0.23)
Constant 4.13∗∗∗ 3.97∗∗∗ 4.14∗∗∗ 3.42∗∗∗ 3.40∗∗∗
(0.09) (0.15) (0.06) (0.13) (0.16)
Observations 649 580 583 519 546
R20.00 0.01 0.02 0.01 0.03
Note. Standard errors clustered by factory in parentheses.
∗p< .10, ∗∗p< 0.05, ∗∗∗p< .01.
Lupo and Bartley 19
global value chains. While the factory managers in Vietnam studied by Malesky and
Mosley (2018) responded to hypothetical signals about the reputation-sensitivity of
buyers, the managerial employees in our study had no discernible reaction to our
buyer promise of support. The effects tended to be positive for managers/supervisors
(as shown in Figure 1), but they did not approach statistical significance. While our
subtle cue is a step away from the market profiles named by Malesky and Mosley
(2018) or the actual affiliations to the Bangladesh Accord and Alliance measured by
Frenkel et al. (2022), it could still have bolstered expectations if managers were especially
attuned to buyers’statements.
We do find some evidence that promises from global buyers could undermine local
demands, though not uniformly. Production workers’views of collective action and
demands for childcare accommodations were undercut to some degree by a promise of
buyer support/responsibility. The responses of production workers, then, may resemble
those of consumers, who have been found to be appeased by voluntary corporate pro-
mises (Malhotra et al., 2019). Yet the processes at play in our study are probably more
about a sense of shelter amidst severe precarity, as observed in Uddin et al.’s (2023)
study of garment workers in Bangladesh, or fears of brand oversight backfiring (as
noted by LeBaron and Lister, 2021; Anner, 2022) than about simply appeasement of con-
cerns or amelioration of guilt among affluent consumers.
Conclusions and implications
In this article, we have combined a focus on the Indonesian apparel industry during the
COVID-19 pandemic with an interest in how distant claims of responsibility-taking by
brands/buyers—along with other forms of intervention—are perceived by workers and
managers. Echoing labor regimes scholarship, we view these workers and managers as
embedded in social relations that reflect and refract antagonistic relationships between
capital and labor (Baglioni et al., 2022). In addition, we build on studies of strategic sig-
naling in global supply chains and production networks, which have shown how compan-
ies use responsibility-taking to deflect calls for more stringent forms of intervention
(LeBaron and Rühmkorf, 2019) and to water down negative claims made by activists
(McDonnell and King, 2013). At the same time, prior research suggests that even
severely limited corporate responsibility initiatives can improve some elements of
global supply chains (Distelhorst and Locke, 2018; Malesky and Mosley, 2018).
Our contribution is twofold. First, we have examined expectations and hopes for
reform among employees at the point of production in the midst of the COVID-19 pan-
demic. Rather than documenting the damage done by pandemic-era disruptions, we have
sought to understand (at least in ways that can be captured in a survey) how workers and
managers view a range of potential solutions broadly conceived. We have especially
focused on demands for local interventions—from factory owners, domestic govern-
ments, and workers themselves. Second, we have combined this focus on perceptions
at the point of production with an interest in how distant claims of support from
buyers might shift people’s expectations for reform. This extends a growing literature
on the political effects of CSR among consumer-citizens in affluent markets and a
20 Journal of Industrial Relations 0(0)
smaller amount of research on how CSR shapes the expectations of factory managers in
global supply chains. This study represents a novel first attempt to elicit the views of
workers themselves, as well as the structuring of perceptions among different groups
of workers. In doing so, we connect the literature on supply chains signals to a more
qualitative body of labor regimes studies, which center local and national struggles
(Baglioni et al., 2022) and “constrained labor agency”(Coe and Jordhus-Lier, 2011) in
questions about global standards.
Taken together, the interventions examined in this article—such as governments taking
measures to prevent layoffs, workers coming together to protect their rights, or making it
easier for managers to hire and fire workers—are not unique to the COVID context.
Arguably, they reflect perennial issues in apparel factories in Indonesia and beyond.
While our survey was done amidst COVID-era disruptions, our results are not necessarily
strongly tied to this period. Although we cannot assess their generalizability, they may tap
into more durable perceptions of how to address insecurity in apparel factories.
Our nonexperimental analyses reveal that some kinds of local reforms are viewed quite
similarly by managers, production workers, and nonproduction workers; by men and
women; and by union members and nonmembers, despite frequently antagonistic interests.
Moreover, while perceptions of factory owner interventions and childcare support were
clustered by factory, perceptions of government intervention, collective action, and labor
market flexibilization tended not to be clustered by factory. While labor regimes can be
identified at multiple scales, including the factory level, these results suggest that factories
matter for some kinds of reforms (perhaps those most closely tied to the actions of factory
owners themselves) more than others.
Turning to our survey experiment, we found no evidence that mentions of distant pro-
mises support bolster demands for local reforms. Instead, we found partial support for the
idea that distant responsibility-taking by brands can undermine demands for local reforms
—though this effect is only clear among production workers considering the utility of col-
lective action or accommodations for workers with childcare responsibilities. These
results suggest that the general bolstering and undermining expectations predicted by
the literature on supply chain signals might be overstated.
Nevertheless, these findings also point to vexing questions about the interactions between
different types of interventions to equalize apparel supply chains. Most studies of reform in
global supply chains point to the importance of combining responsible sourcing from above,
mobilization from below, and state intervention or capacity-building in exporting countries
(e.g., Locke, 2013). Yet if responsible sourcing initiatives undermine mobilization from
below, then combining them may be even more difficult than typically imagined. To be
sure, our results suggest a fairly modest and contingent effect on production workers’expec-
tations, but if these are indicative of larger dynamics of demobilization (which may be hard to
capture in a survey), then the crowding out/displacement problem may be a serious issue.
While broadening opportunities for solidarity in some cases, corporate promises may not
be as beneficial in sites of production as some scholars and policy-makers imagine.
Several limitations of this study should be kept in mind. First, workers and managers
in the factories surveyed are likely to have a greater level of awareness of and familiarity
with corporate responsibility due to their participation in the Better Work Indonesia
Lupo and Bartley 21
program. Our findings should not be taken as representative of the Indonesian apparel
sector as a whole. Instead, they shed light on the perceptions of locally engaged employ-
ees in factories producing for reputation-sensitive buyers—for whom promises form
distant buyers are arguably more important. Second, the local interventions we asked
about were phrased in minimally contentious ways, so they may not have tapped into
views of more challenging acts (such as strikes). At the same time, we do find divergent
perceptions of some interventions—such as special accommodations from factory
owners—so points of agreement should not be overstated. It could be that there are
shared goals for local interventions (e.g., preventing layoffs) but divergent assumptions
about reform that our survey could not pick up. Third, our survey experiment used a fairly
subtle cue in our hypothetical statement about brand/buyer commitments. It is plausible
that a louder or more extended signal (perhaps conveyed through a personal statement
from a valued buyer) could have provoked more discernible reactions.
For research on labor relations in Indonesia, this study contributes a methodologically
novel approach to understanding the intertwining of protective regulations, flexibilization
reforms, and global buyers, rather than focusing on just one of these arenas. Although our
sample is limited to one particular slice of the Indonesian apparel industry, our results
suggest some skepticism of flexibilization reforms among factory managers, which could
limit the impact of the Omnibus Law. On the other hand, the results suggest a tension
between promises of global buyers and the domestic politics of pro-worker reforms, in con-
trast to the synergies highlighted in other studies (e.g., Bartley and Egels-Zandén, 2016). The
evolution of labor politics in Indonesia leaves many questions to be answered about how man-
agers and workers are navigating supply chain structures, which are themselves evolving
toward regionalism and heightened regulatory scrutiny. Future research should seek to
measure workers’and managers’perceptions of reform in a fuller way, taking account of mul-
tiple dimensions of the surrounding context (i.e., perceptions of local governments, the
national government, brands/buyers, and foreign regulations). In addition, detailed studies
of activism and upgrading in situ may benefit from using comparative quasi-experimental
approaches to see how different kinds of distant promises and interventions bolster or
hinder improvements in labor conditions within factories.
Acknowledgments
We would like to thank Arianna Rossi, Jeff Eisenbraun, and the Better Work Indonesia program for
their support in realizing this study and the three anonymous reviewers for their careful and con-
structive comments. Special thanks also to all the survey respondents who took the time to partici-
pate in our study amidst the devastating impacts of the COVID-19 pandemic.
Declaration of conflicting interests
The authors declared the following potential conflicts of interest with respect to the research,
authorship, and/or publication of this article: Prior to beginning this research, Luisa Lupo was
employed by the Better Work program in Geneva between 2018 and 2020.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
22 Journal of Industrial Relations 0(0)
ORCID iD
Luisa Lupo https://orcid.org/0000-0003-3788-1422
Supplemental materials
Supplemental material for this article is available online.
References
Afros A (2022) Workers’voice within and beyond participation committees: A case study of two
Bangladeshi Garment Factories. Better Work Discussion Paper No. 46. Geneva:
International Labour Office.
Alford M and Phillips N (2018) The political economy of state governance in global production
networks: Change, crisis and contestation in the South African fruit sector. Review of
International Political Economy 25(1): 98–121.
Amengual M and Bartley T (2022) Global markets, corporate assurances, and the legitimacy of
state intervention: Perceptions of distant labor and environmental problems. American
Sociological Review 87(3): 383–414.
Amengual M and Chirot L (2016) Reinforcing the state: Transnational and state labor regulation in
Indonesia. ILR Review 69(5): 1056–1080.
Anner M (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.
Anner M (2022) Power relations in global supply chains and the unequal distribution of costs
during crises: Abandoning garment suppliers and workers during the COVID-19 pandemic.
International Labour Review 161(1): 59–82.
Baglioni E, Campling L, Coe NM, et al. (2022) Labour Regimes and Global Production. Newcastle
upon Tyne: Agenda Publishing.
Bair J (2022) Gendered labour regimes in global production. In: Baglioni E, Campling L, Coe NM
and Smith A (eds) Labour Regimes and Global Production. Newcastle upon Tyne: Agenda
Publishing, 29–44.
Barrientos S (2019) Gender and Work in Global Value Chains: Capturing the Gains? Cambridge:
Cambridge University Press.
Barrientos S (2022) Work, gender and discrimination in global production. In: Kimberly AE (ed)
Handbook on Globalisation and Labour Standards. Cheltenham: Edward Elgar Publishing,
64–83.
Bartley T (2018) Rules Without Rights: Land, Labor, and Private Authority in the Global Economy.
Oxford: Oxford University Press.
Bartley T, Amengual M and Koos S (2023) Perceptions of Supply Chain Capitalism: Advocacy,
Responsibility, and Regulation. Working paper presented at Earth Systems Governance
Conference.
Bartley T and Egels-Zandén N (2016) Beyond decoupling: Unions and the leveraging of corporate
social responsibility in Indonesia. Socio-Economic Review 14(2): 231–255.
Burawoy M (1985) The Politics of Production: Factory Regimes under Capitalism and Socialism.
London: Verso.
BW (Better Work) (2021) Indonesia updates. Available at: https://betterwork.org/2021/05/07/
indonesia-updates/ (accessed 10 August 2021).
Lupo and Bartley 23
Caraway TL (2008) Explaining the dominance of legacy unions in new democracies: Comparative
insights from Indonesia. Comparative Political Studies 41(10): 1371–1397.
Caraway TL (2023) Labor’s reversal of fortune: Contentious politics and executive aggrandizement
in Indonesia. Social Movement Studies 22(5–6): 689–705.
Caraway TL and Ford M (2022) Labor and Politics in Indonesia. Cambridge: Cambridge
University Press.
Coe NM and Jordhus-Lier DC (2011) Constrained agency? Re-evaluating the geographies of
labour. Progress in Human Geography 35(2): 211–233.
Connor T and Haines F (2013) Networked regulation as a solution to human rights abuse in global
supply chains? The case of trade union rights violations by Indonesian sports shoe manufac-
turers. Theoretical Criminology 17(2): 197–214.
Di Malta G, Cooper M, Bond J, et al. (2023) The patient-perceived helpfulness of measures scale:
Development and validation of a scale to assess the helpfulness of using measures in psycho-
logical treatment. Assessment 31(5): 994–1010.
Distelhorst G and Locke RM (2018) Does compliance pay? Social standards and firm-level trade.
American Journal of Political Science 62(3): 695–711.
Elson D and Pearson R (1981) ‘Nimble fingers make cheap workers’: An analysis of women’s
employment in third world export manufacturing. Feminist Review 7: 87–107. Available at:
https://www.fairwear.org/covid-19-dossier/worker-engagement-and-monitoring/country-specific-
guidance/covid19-indonesia/ (accessed 16 August 2021).
Enloe C (2000) Daughters and generals in the politics of the globalized sneaker. In: Aulakh PS and
Schechter MG (eds) Rethinking Globalization: Corporate Transnationalism to Local
Interventions. London: Palgrave Macmillan, 238–246.
Foley L and Piper N (2021) Returning home empty handed: Examining how COVID-19 exacerbates
the non-payment of temporary migrant workers’wages. Global Social Policy 21(3): 468–489.
Ford M (2009) Workers and Intellectuals: NGOs, Trade Unions and the Indonesian Labour
Movement. Honolulu: University of Hawaii Press.
Ford M, Gillan M and Ward K (2023) Beyond the brands: COVID-19, supply chain governance, and
the state–labor nexus. Industrial Relations: A Journal of Economy and Society 62(2): 172–188.
Ford M and Ward K (2021) COVID-19 in Southeast Asia: Implications for workers and unions.
Journal of Industrial Relations 63(3): 432–450.
Frenkel SJ, Rahman S and Rahman KM (2022) After Rana Plaza: Governing exploitative workplace labour
regimes in Bangladeshi garment export factories. Journal of Industrial Relations 64(2): 272–297.
Frenkel SJ and Schuessler ES (2021) From Rana Plaza to COVID-19: Deficiencies and opportun-
ities for a new labour governance system in garment global supply chains. International Labour
Review 160: 591–609.
FW (Fair Wear) (2021) COVID-19 impact and responses: Indonesia. Available at: https://www.
fairwear.org/covid-19-dossier/worker-engagement-and-monitoring/country-specific-guidance/
covid19-indonesia/ (accessed 16 August 2021).
GereffiG, Humphrey J and Sturgeon T (2005) The governance of global value chains. Review of
International Political Economy 12(1): 78–104.
Greenhill B, Mosley L and Prakash A (2009) Trade-based diffusion of labor rights: A panel study.
American Political Science Review 103(4): 169–190.
Holzberg B (2023) Local understandings of decent work and the legitimacy of global labour standards:
Insights from garment suppliers in Egypt and Jordan. Journal of Business Ethics 192: 689–712.
ILO (International Labour Organization) (2020a) The Supply Chain Ripple Effect: How COVID-19
is Affecting Garment Workers and Factories in Asia and the Pacific. Bangkok: ILO.
24 Journal of Industrial Relations 0(0)
ILO (International Labour Organization) (2020b) COVID-19: Action in the global garment indus-
try. Available at: https://www.ilo.org/global/topics/coronavirus/sectoral/WCMS_742343/lang–
en/index.htm (accessed 13 November 2023).
Joshi A, Kale S, Chandel S, et al. (2015) Likert Scale: Explored and explained. British Journal of
Applied Science & Technology 7(4): 396–403.
Judd J, Kuruvilla S and Jackson JL (2022) Security for Apparel Workers: Alternative Models. NCP
Working Paper No. 3, Cornell University. Available at: https://www.ilr.cornell.edu/sites/default/
files-d8/2022-05/NCP%20IndustriALL%20Severance%20042222.pdf (accessed 13 November
2023).
Kabeer N, Huq L and Sulaiman M (2020) Paradigm shift or business as usual? Workers’views on
multi-stakeholder initiatives in Bangladesh. Development and Change 51(5): 1360–1398.
Kabeer N, Razavi S and Rodgers YM (2021) Feminist economic perspectives on the COVID-19
pandemic. Feminist Economics 27(1–2): 1–29.
Keahey J and Murray DJ (2017) The promise and perils of market-based sustainability. Sociology of
Development 3(2): 143–162.
Kolcava D, Rudolph L and Bernauer T (2019) Voluntary business initiatives can reduce public pres-
sure for regulating firm behaviour abroad. Journal of European Public Policy 28(4): 591–614.
Kuruvilla S, Liu M, Li C, et al. (2020) Field opacity and practice-outcome decoupling: Private regu-
lation of labor standards in global supply chains. ILR Review 73(4): 841–872.
Kusumaningrum S, Siagian C and Beazley H (2022) Children during the COVID-19 pandemic:
Children and young people’s vulnerability and wellbeing in Indonesia. Children’s
Geographies 20(4): 437–447.
LeBaron G and Lister J (2021) The hidden costs of global supply chain solutions. Review of
International Political Economy 29(3): 669–695.
LeBaron G and Rühmkorf A (2019) The domestic politics of corporate accountability legislation:
Struggles over the 2015 UK modern slavery act. Socio-Economic Review 17(3): 709–743.
Lee CK (2007) Against the Law: Labor Protests in China’s Rustbelt and Sunbelt. Berkeley:
University of California Press.
Li C. and Kuruvilla S. (2023) Corporate codes of conduct and labour turnover in global apparel
supply chains. British Journal of Industrial Relations, 61, 481–505.
Locke RM (2013) The Promise and Limits of Private Power: Promoting Labor Standards in a
Global Economy. Cambridge: Cambridge University Press.
Mahler D.G., Lakner C., Castañeda Aguilar R.A., H., et al. (2020). The impact of COVID-19
(Coronavirus) on Global Poverty: Why Sub-Saharan Africa Might Be the Region Hardest
Hit. World Bank Blog. Washington, DC: World Bank.
Malesky EJ and Mosley L (2018) Chains of love? Global production and the firm-level diffusion of
labor standards. American Journal of Political Science 62(3): 712–728.
Malhotra N, Monin B and Tomz M (2019) Does private regulation preempt public regulation?
American Political Science Review 113(1): 19–37.
Mcdonnell MH and King B (2013) Keeping up appearances: Reputational threat and impression
management after social movement boycotts. Administrative Science Quarterly 58: 387–419.
Mezzadri A and Majumder S (2020) Towards a feminist political economy of time: Labour circu-
lation, social reproduction & the ‘afterlife’of cheap labour. Review of International Political
Economy 29(6): 1804–1826.
Mutz DC (2011) Population-Based Survey Experiments. Princeton: Princeton University Press.
Oka C (2010) Accounting for the gaps in labour standard compliance: The case of Cambodia’s
garment sector. European Journal of Development Research 22(1): 59–78.
Lupo and Bartley 25
Oka C (2012) Does Better Labour Standard Compliance Pay? Linking Labour Standard
Compliance and Supplier Competitiveness. Better Work Discussion Paper Series: No. 5,
International Labour Office, Geneva.
Pike K (2020) Voice in supply chains: Does the better work program lead to improvements in labor
standards compliance? ILR Review 73(4): 913–938.
Pun N (2005) Made in China: Women Factory Workers in a Global Workplace. Durham: Duke
University Press.
Robson K and Pevalin D (2015) Multilevel Modeling in Plain Language. Thousand Oaks: Sage.
Siegmann KA, Merk J and Knorringa P (2017) Positive class compromise in globalized produc-
tion? The freedom of association protocol in the Indonesian sportswear industry.
International Labour Review 156: 345–365.
Steinberg J (2023) Structure and context: A multi-level approach to supply chain governance.
Annual Review of Political Science 26: 411–429.
Tejani S and Fukuda-Parr S (2021) Gender and COVID-19: Workers in global value chains.
International Labour Review 160(4): 649–667.
Trappmann V, Umney C, Neumann D, et al. (2022) Labour protests during the pandemic: the case
of hospital and retail workers in 90 countries. ILO Working Paper, November 7.
Uddin S, Ahmed MS and Shahadat K (2023) Supply chain accountability, COVID-19, and viola-
tions of workers’rights in the global clothing supply chain. Supply Chain Management 28(5):
859–873.
UN Women (2020) Whose Time to Care? Unpaid Care and Domestic Work during COVID-19.
New York: UN Women.
Wallander L (2009) 25 Years of factorial surveys in sociology: A review. Social Science Research
38(3): 505–520.
Wickramasingha S (2023) Geographies of dissociation: Informality, ethical codes and fragmented
labour regimes in the Sri Lankan apparel industry. Journal of Economic Geography 23(6):
1191–1211.
Worker Rights Consortium (2021) Covid-19 tracker: Which brands acted responsibly toward sup-
pliers and workers? Available at: https://www.workersrights.org/issues/covid-19/tracker/
(accessed 13 November 2023).
Biographical notes
Luisa Lupo is PhD Candidate in International Relations/Political Science and a researcher at the
Gender Centre for the project Gendering Survival from the Margins at the Geneva Graduate
Institute (Switzerland). She holds degrees in Development Studies (MA) from the Geneva
Graduate Institute and in Economics (BSc Hons) from the University of Rome Tor Vergata, Italy.
Her areas of research include issues related to gender, social reproduction, and the everyday life of
households, as well as rights and contestations in global supply chains and production networks.
Tim Bartley is Professor in the Earth Commons Institute and Department of Sociology at
Georgetown University. His previous work includes Rules without Rights: Land, Labor, and
Private Authority in the Global Economy (Oxford University Press, 2018), “Global Markets,
Corporate Assurances, and the Legitimacy of State Intervention”(with Matthew Amengual,
American Sociological Review, 2022), “Raising the Floor: New Directions in Public and Private
Enforcement of Labor Standards in the United States”(with Janice Fine, Journal of Industrial
Relations, 2019), and “Beyond Decoupling: Unions and the Leveraging of Corporate Social
Responsibility in Indonesia,”(with Niklas Egels-Zandén, Socio-Economic Review, 2015).
26 Journal of Industrial Relations 0(0)