New research horizons

Global value chains and transaction-level customs data

Over 70% of global trade goes through global value chains (GVCs). For firms in developing countries, insertion into GVCs has been associated with economic and social benefits. Yet, critics also point to power imbalances between lead firms in developed countries and supplier firms in the Global South, with adverse consequences for workers and local communities.

In this context, a major limitation in GVC analysis has been the paucity of firm-level data to capture the compound effects of participation in GVCs on individual firms’ performance and labour conditions.

For traditional GVC analysis, economists tend to rely on aggregated data such as national gross trade flows and input-output tables with data on the share of value-added embedded in a country’s exports (see OECD 2013). Further methodological advancements also distinguish between an industry’s origin of value-added across more (and less) capital intensive activities (see Timmer et al. 2019). While this allows for a better understanding of countries’ positions in GVCs, the high level of aggregation of this data (e.g., by sector instead of firm) fails to capture the granularity of inter-firm interactions and their governance (Johnson 2018).

Some scholars of GVCs and global production networks (GPNs) have privileged qualitative methods along with survey data to provide a more nuanced analysis of firm-level dynamics (see Kano et al. 2020). Importantly, these methodologies are justified by the fact that firms in GVCs are not alike: ‘Chains often have a governor or lead firm that largely determines production parameters and wields power over the other firms’ (Navas-Aleman 2011). Large aggregate datasets, it is argued, mask these dynamics. 

An alternative data source

Another important source of quantitative firm-level data that has been largely overlooked in the study of GVCs is transaction-level customs data. This is the disaggregated trade information collected by national revenue offices whenever something is imported or exported from a country, inclusive of buyer’s and supplier’s information, the country of destination/origin, the type, quantity, and value of product traded, and the date of the transaction.

Ground-breaking research using this data has shown their potential to provide robust indicators of firm’s upgrading and value chain governance (see Dallas 2015). Furthermore, the recent 2020 World Development Report on GVCs called for increasing usage of transaction data which — unlike aggregated trade data — captures the heterogeneity of value chain linkages and allows ‘a finer understanding of how multinational firms organize their production networks’ (p.18).

Ground-breaking examples already exist

As part of the ‘Shifting South’ research project, researchers at the Global Development Institute (University of Manchester) also work with customs data collected by the Kenya Revenue Authority (KRA) to shed light on the differential implications that participation in North-South and South-South GVCs presents for local suppliers. Whilst research is still ongoing, a first example of how transaction-level customs data can be used to this purpose is presented in my newly-published article in the Journal of World Business.

The article tests whether links to North-South and South-South value chains affect suppliers’ improvement of product quality and value-added tasks over time (respectively defined as product and functional upgrading). The granularity of customs data is unpacked through a growth model using multilevel linear regressions to allow for nested levels of analysis that account for both intra- and inter-firm variation.

Results show that product quality and value-added tasks are higher for exports to the North than to the South, but that there is no systematic difference in product and functional upgrading between the two destinations. Digging deeper, however, reveals that value chains governed by Chinese lead firms present steeper functional upgrading than value chains governed by European firms.

Moreover, intra-Africa value chains emerge as platforms for small suppliers to specialize in higher value-added tasks. In other words, the article advances important conclusions on how certain Southern end-markets (i.e. China and the regional market) can be more conducive to learning and upgrading among Kenyan suppliers than traditional European markets.

Another example comes from Eswatini’s apparel industry. By combining customs data collected by the Eswatini Revenue Authority with open-ended interviews, the researchers explore the influence of private governance on suppliers’ economic up/downgrading before and during the COVID-19 lockdown. Suppliers selling through intermediaries suffered a more significant drop in prices than their larger competitors who had direct contracts with final retailers. Notably, the granularity of the data allowed the authors to explore the differential impact of COVID-19 for specific firms in the same country and sector, leading to targeted policy recommendations to support the employers and workers most affected by the crisis.

Other ongoing research within the ‘Shifting South’ project includes the use of customs data to evaluate the governance of regional value chains linking South African apple producers to Kenyan buyers, a comparative analysis of the impact of external shocks on Kenya’s exporting companies, and the modelling of the economic impact of co-participation in North-South and South-South value chains for Kenyan horticultural exporters.

Leveraging transaction-level customs data for future research

There are some limitations to this data. For example, they do not report domestic purchases within the country of reference — therefore downplaying the importance of domestic value chains. Moreover, especially in developing countries, they tend to misrepresent regional trade by not accounting for informal commerce with neighbouring countries. An additional hurdle to overcome is in accessing and harmonising this data across different national customs offices, including the matching of firms’ identifiers in importing and exporting countries. An on-going attempt to provide such a shared platform, however, is the World Bank’s Exporter Dynamics Database.

Despite these limitations, the research discussed above shows critical potential in leveraging customs data for research purposes within the study of GVCs. In sum, transaction-level customs data provides a critical instrument to observe and measure the interaction between lead firms and suppliers. Within the study of GVCs, this implies that aspects of economic upgrading and private governance can be quantified and compared across the wider universe of registered exporting and importing firms, rather than just inferred from smaller samples.

The views expressed in this piece are those of the author(s), and do not necessarily reflect the views of the Institute or the United Nations University, nor the programme/project donors.