Labor Provisions in Trade Agreements (LABPTA): Introducing a New Dataset
| DOI | http://doi.org/10.1111/1758-5899.12577 |
| Date | 01 November 2018 |
| Author | Dora Sari,Damian Raess |
| Published date | 01 November 2018 |
Labor Provisions in Trade Agreements
(LABPTA): Introducing a New Dataset
Damian Raess
University of Bern and University of Reading
Dora Sari
University of Geneva and Harvard Law School
Abstract
Global labor policy through trade has begun to receive growing attention with the inclusion of labor provisions in preferential
trade agreements (PTAs). Until recently there has been a shortage of available data that would adequately capture the varia-
tion that exists with respect to the scope and stringency of labor provisions, preventing scholars and practitioners from
addressing key questions about the design and effects of the trade-labor linkage. This paper introduces a new dataset cover-
ing 487 PTAs from 1990 to 2015 coded against 140 distinct items pertaining to six main categories, presenting –to our knowl-
edge –the most rigorous and fine-grained mapping of labor provisions. It also offers the first systematic description of key
trends in the design and occurrence of those commitments. Our study shows that labor provisions have not only expanded in
terms of their content and participating countries but that labor provisions have, although to a varying degree, also become
more stringent over time. The provisions that have across all PTAs increased most steadily are the ones related to the institu-
tional framework set up for the monitoring and implementation of labor commitments, becoming more specialized and more
inclusive of third party involvement over time.
Policy Implications
•Policy makers should actively invest in high-quality datasets and quantitative studies particularly regarding the socio-eco-
nomic impact of labor provisions to design better targeted and more effective labor provisions in their trade agreements.
Such datasets also provide a new tool for knowledge sharing and the training of relevant constituencies so that they
develop a better understanding of existing approaches and practices for their ongoing or future trade negotiations.
•In line with the growing recognition of the importance of ex post evidence-based assessment of the effectiveness of cer-
tain mechanisms, policy makers should also commit to the inclusion of provisions establishing regular impact assessment
of the agreed labor provisions and their implementation.
•While the Singapore Ministerial Declaration (1996) of the WTO designated the ILO as the competent body to deal with
labor standards in the context of international trade, the role of the ILO in this area remains circumscribed. Given its
capacities as standard setter, provider of technical assistance, and holder of key expertise and initiatives in this area, policy
makers should consider reinforcing the role of the ILO in the trade-labor linkage.
•Labor provisions in trade agreements have gained momentum in recent years, including in countries once staunchly
opposed to their introduction, a movement that is likely to be reinforced under the UN Sustainable Development Agenda,
particularly SDG goal 8 on inclusive growth and decent work. In view of the growing proliferation of labor provisions and
arising new challenges such as parallel jurisprudence, institutional overlap, lack of coordination, WTO members should
consider the renewed discussion and possible introduction of a labor clause at the multilateral level.
The pendulum is swinging back, but not where the master
expected it –and for good reason. Since the time Polanyi
wrote his masterpiece The Great Transformation (2001
[1944]), capitalism has gone global again. While domestic
labor market liberalization has proceeded unabated world-
wide in recent decades, international labor regulation has
made significant progress often beyond the headlines. The
trade-labor linkage is a case in point. Since the early 1990s,
preferential trade agreements (PTAs) increasingly incorporate
labor provisions (LPs).
1
These LPs link the benefits of
enhanced market access to, for instance, the (effective)
enforcement of internationally recognized labor rights. But
there is much variation across time and space with respect
to the scope and stringency of these LPs. While some PTAs
include far-reaching and highly enforceable LPs, others only
make scant or even no references to labor standards.
The objectives of this article are twofold: one is to intro-
duce a new dataset on the design of Labor Provisions in
Trade Agreements (LABPTA); the other is to document the
evolution of the content and stringency of these LPs over
time. The central idea is to provide a detailed and up-to-
date picture for the better assessment of the trade-labor
Global Policy (2018) 9:4 doi: 10.1111/1758-5899.12577 ©2018 University of Durham and John Wiley & Sons, Ltd.
Global Policy Volume 9 . Issue 4 . November 2018 451
Research Article
nexus on a global scale. To our knowledge, LABPTA is the
most fine-grained and comprehensive mapping of the con-
tent of LPs in PTAs, allowing for extended analysis of the
design and effects of (varying) LPs in PTAs.
While researchers have started to focus their attention on
explaining the causes and consequences of variation in the
design of PTAs, LPs have until recently either been
neglected or not been the focus of systematic investigation
(e.g. D€
ur and Elsig, 2015). Existing studies have tended to
focus on a single case, be it a particular PTA or global player
(e.g. Giumelli and van Roozendaal, 2016; Nyland and
O’Rourke, 2005), or on a small-N comparative analysis of the
United States and the European Union (e.g. Hafner-Burton,
2009; Kerremans and Gistelinck, 2009), all too often using
and reproducing the narrow statement that US PTAs are
enforceable with respect to labor standards whereas EU
agreements are not (e.g. Postnikov and Bastiaens, 2014),
which does not do justice to the broadening scope of
enforceability in US PTAs and the enforceability, yet with
variable geometry, of EU PTAs. The first broader comparative
analysis was done by the International Labour Organization
(ILO) in 2009, an assessment that was extended in subse-
quent reports (ILO 2013,2016, 2017). The ILO’s pioneering
work in this domain, while providing detailed insights into
the design and effectiveness of LPs in PTAs, on a method-
ological basis remains limited in two important ways: on the
one hand it fails to introduce a comprehensive template for
the mapping of LPs in PTAs; on the other hand it fails to
carry out a systematic coding of the content of LPs in the
available largest possible number of PTAs based on an inte-
grated and encompassing conceptual and analytical
approach.
Existing mapping exercises tend to focus on LPs as part
of a broader objective to map nontrade issues (NTIs) in
PTAs. Researchers engaged in such endeavor almost invari-
ably face a trade-off between breadth and depth of issues
covered that comes at the expense of a higher level of
granularity in the coding of a single issue. Milewicz’s (2016;
Milewicz et al., 2018) dataset codes labor rights alongside
human rights, environment, corruption, security and democ-
racy for 522 PTAs signed over the period 1951–2009. Mile-
wicz’s mapping of LPs carries important limitations not only
because it adopts a broad definition of labor rights that
includes social security rights dealing with sickness, invalid-
ity, old-age, industrial accident and unemployment, but also
because the coding of the different NTIs, in spite of the dis-
tinction between provisions in the preamble and the main
text of the agreement, remains very crude.
Lechner’s (2016) dataset of NTIs design provides a novel
coding for 474 PTAs signed since 1990, covering civil and
political rights, environmental protection, and economic and
social rights, the latter category including rights in the
realms of development, education, health, labor, and social
security. It is a significant improvement on Milewicz’s data-
set in that it covers a range of rights coded over the three
dimensions of obligation, precision and delegation inherent
to the legalization approach (Abbott et al., 2000). Lechner,
while using a broader definition of LPs than we do, codes
25 items narrowly relating to labor protection (substantive-
related commitments as well as institution and cooperation-
related commitments) out of a total of 72 in the economic
and social rights category. Yet, her coding scheme remains
limited to the extent that the coding of obligation and dele-
gation (i.e. issues of bindingness, enforeceability, as well as
third party involvement) is carried out in relation to eco-
nomic and social rights in general, preventing us from
knowing whether obligation and delegation apply effectively
to LPs. Also, labor-related cooperation commitments are
coded in a crude fashion with a single category.
Kamata (2016) examines the impact of labor clauses in
PTAs on conditions of work for a sample of 223 PTAs from
1995 to 2011. The classification of PTAs with LPs builds on
two criteria: (1) provisions that demand, urge, or expect the
signatory countries to harmonize their domestic labor condi-
tions and regulations with the internationally recognized
standards; and (2) provisions that stipulate items over which
the signatory countries will cooperate and the procedures
for dispute settlement on labor issues. While an improve-
ment on the binary coding of LPs, his classification remains
too rudimentary to adequately capture the large variation in
the design of LPs in PTAs. Among the shortcomings is the
lack of details as to whether commitments are specificto
any particular standards, the conflation between soft and
hard mechanisms such as cooperation and enforceability,
respectively, and the failure to distinguish between various
degrees of enforceability.
Finally, Hofmann et al. (2017) constructed a novel data-
base with detailed assessment of the content of 279 PTAs
notified to the World Trade Organization (WTO) signed
between 1958 and 2015. Their mapping covers 52 policy
areas with a separate category for ʻLabour Market Regula-
tionʼ. It refers to provisions that pertain to the regulation of
the national labor market and the affirmation of ILO com-
mitments (a single item coded in a binary mode), and the
enforcement of such provisions (coded on a 0-2 scale). In
spite of their focus being more specific to labor rights, their
methodological approach suffers from similar limitations as
Kamata’s approach.
In short, a rigorous and fine-grained approach to mapping
LPs in PTAs is overdue. Our LABPTA coding scheme and
dataset provide just that.
1. Conceptualizing and measuring LPs in PTAs
Given that there is no single approach as to what consti-
tutes a ʻlaborʼprovision in PTAs, we opted for a narrow def-
inition which has the advantage of drawing clear
distinctions between labor and (broader) social provisions
on the one hand, and between labor protection provisions
and (broader) labor market regulation provisions on the
other. In our conceptualization, LPs refer to rules and regula-
tions that aim to protect and/or promote workers’rights
and working conditions. In the context of PTAs, a labor issue
is considered to be ʻcoveredʼby an agreement if the agree-
ment contains a provision (i.e. chapter, article, paragraph or
sentence) providing for some commitment in this field. In
©2018 University of Durham and John Wiley & Sons, Ltd. Global Policy (2018) 9:4
Damian Raess and Dora Sari
452
To continue reading
Request your trial