Do Exports from Developing Economies Still Matter in Global Value Chains? Evidence from Malaysia, Thailand and Vietnam.

• Author: Lee, Woocheol

1. Introduction

   Over the last two decades, the importance of global value chains (GVCs) in international trade and production has grown significantly. Today, a large number of economies and firms participate in such value chains. The main features of GVCs include the fragmentation and regionalization (Humphrey 2019) of production processes. Here, fragmentation does not simply mean the reallocation of some fabrication tasks but encompasses many other aspects of production such as research and development (R&D), design, sales and marketing, and consumption and recycling (Lee and Gereffi 2015). Likewise, regionalization refers to the concentration of production processes, particularly assembly tasks, into specific regions such as East Asia and East Europe. These features have been derived from strategies developed by multinational corporations (MNCs) aimed at revamping profits by saving costs and enhancing the incorporation of local indigenous firms (LIFs) in developing and emerging economies (DEEs) into GVCs. (1) A recent report by the Organization for Economic Co-operation and Development (OECD) states that, in 2014, 70 per cent of international trade was conducted through GVCs, and their foreign affiliates accounted for 31 per cent of global exports and 28 per cent of imports (Cadestin et al. 2019).

   Whether the growing participation of LIFs in GVCs has led to the upgradation of the production capabilities of these firms remains unclear. Two different perspectives on this topic have been debated extensively. The optimistic viewpoint held by advocates of free trade is that LIFs are likely to enhance and upgrade their capabilities via knowledge and technology spillovers when they collaborate with MNCs within GVCs (Pietrobelli and Rabellotti 2011). Over time, DEEs and their LIFs are supposed to carry out more sophisticated and value-creating tasks.

   The activist perspective, however, emphasizes that opportunities to upgrade production capabilities within GVCs are not readily available to DEEs without the intervention of the state. For example, the biological evolutionary game theory developed by Smith (1982) predicts that asset owners (MNCs in our case) win over intruders (LIFs) in an asymmetric game of achieving technological progress. NeoSchumpeterians and evolutionary economists argue that upgrading indeed demands a lot of time and combined innovation systems to embody transmitted knowledge (Nelson and Winter 1982; Fagerberg, Lundvall, and Srholec 2018). Furthermore, it is widely documented that MNCs strategically control their assets so as to not transfer more than what is required to partner LIFs (Hart-Landsberg and Burkett 1998; Bernard and Ravenhill 1995; Kaplinsky, Morris, and Readman 2002).

   One method to empirically address this contrast is to examine the types of products that DEEs export and determine the value-added in those exports captured by domestic LIFs. Note that some products create more value than others if they, say, include sophisticated tasks. Hence, if DEEs increase the export of these goods, we can assume that their production capabilities have improved. But it is also possible that the increase in high value-creation exports is due to the involvement of key MNCs in the export sector. It is therefore critical to calculate the domestic share of the value-added (DVA) and the foreign share of value-added (FVA) of those exports. An increasing DVA trend can be interpreted as an upgrade of the production capabilities of the local firms operating in developing countries.

   Several recent studies show that the share of value-added attributed to DEEs has increased over time. For instance, the share of global value-added trade captured by DEEs increased from about 20 per cent to over 40 per cent between 1990 and 2010 (UNCTAD 2013), while the value-added share of developed economies in global manufacturing fell from 74 per cent to 56 per cent during the same time (Timmer et al. 2014). Do these figures lend support to the optimistic perspective on the impact of GVC expansion? Yes and no. The increase in the shared value-added could be attributed to LIFs in DEEs conducting a higher volume of low value-added, unsophisticated tasks. This could imply that the engagement in GVCs does not lead to massive upgradation of production capabilities among domestic firms. Additionally, using aggregate level data is likely to cancel out some important changes that take place at the sectoral level; we must therefore carry out a definitive sectoral analysis to better understand the correlation between GVC participation and the upgradation of DEEs.

   Most existing studies on this subject focus exclusively on the supply side of GVCs, including only value-added data in the analyses. But what DEEs export and sell in the global market forms the demand side of the value chains and must also be considered. This paper attempts to focus on both. With respect to the supply side, the study collates the weighted sectoral domestic share of value-added (DVA) of Malaysia, Thailand and Vietnam from the OECD Trade in Value-added database (TiVA) 2018 edition. For the demand side, this paper estimates the weighted sectoral income elasticities of demand for exports (EEs) from the UN COMTRADE database over the period 1980-2017 for Malaysia and Thailand and 19972017 for Vietnam (different time frames due to data availability issues). Changes in income elasticities of a country's export demand over time indicate that structural changes from one sector to another, which can help us trace the process of industrial upgradation--as observed in Latin American countries (Gouvea and Lima 2010) and Vietnam (Lee 2021).

   Upon analysing the trade data and value-added data of the three developing economies in Southeast Asia, this paper finds mixed evidence on the relationship between upgrading technological capabilities and participating in GVCs. The demand side analysis highlights the rise of sectors that generate high value-added in the three economies, suggesting some form of transformation in the export sector--from involvement in the production of low value-added goods to high value-added ones. The supply side evidence, however, reports a significant divergence among countries. The DVA at the aggregate level increased for Malaysia and Thailand but fell in the case of Vietnam.

   The rest of the paper is structured as follows. The next section focuses on existing studies on the expansion of GVCs and their impact on economic growth and the improvement of production capabilities. The data set, estimation model and sectoral changes in the share of value-added form the bulk of the third section. The subsequent section covers the income elasticities of export demand and collated DVA in each of the three economies. The final section concludes.

2. Literature Review

   As mentioned earlier, the expansion of GVCs is characterized by unprecedented fragmentation and regionalization of production processes. So why has production been divided into fragments and across regions? One answer lies in the profit squeeze theory. The already intense focus on squeezing profits has been further strengthened due to the dominant idea of maximizing shareholder value in global capitalism (Lazonick and O'Sullivan 2000). Here, MNCs played a vital role. These companies sought ways to alleviate profits by cutting production costs. Outsourcing and offshoring were one of the many key strategies that resulted in the aforementioned fragmentation and regionalization of production processes (see, for example, Auvray and Rabinovich 2017; Bogliacino, Guarascio, and Cirillo 2018; Demir 2009; Milberg and Winkler 2013; Orhangazi 2008). (2) As more economies began participating in GVCs, the MNCs started to keep the core tasks such as R&D and design onshore while outsourcing and offshoring simple assembly tasks to DEEs--depending on the endowment of relevant resources in each county (Gereffi, Humphrey, and Sturgeon 2005; Lee and Gereffi 2015).

   However, DEEs and their LIFs would not join GVCs if they did not benefit. One advantage often pointed out was (and is) the enhanced access to advanced technology, product design and marketing strategies without any heavy investment or risk of failure (Baldwin 2013). This is a typical snapshot of a virtuous circle that mainstream neoclassical economic approaches emphasize, within which it is supposed that engaging in GVCs is a win-win game for all participants. Yet, in reality, we do not have many cases supporting this viewpoint. For example, Fagerberg, Lundvall, and Srholec (2018) in their study of 125 countries over the period 1997-2013 argue that participation in GVCs itself, measured by the foreign value-added in a country's exports, does not explain faster economic growth. GVCs have also been found to have a bias against the low-educated workers who are relatively abundant in low-income countries (Reijnders, Timmer, and Ye 2021).

   Furthermore, there are various obstacles that LIFs are likely to encounter over the course of the upgrade. Specifically, MNCs determine how financial, material and human resources are allocated within GVCs. Simultaneously, they also employ highly sophisticated strategies against LIFs by controlling firm-specific assets and by limiting technology diffusion (Hart-Landsberg and Burkett 1998; Bernard and Ravenhill 1995; Machado 1999; Kaplinsky, Morris, and Readman 2002). In terms of offshoring, MNCs can outsource activities where there is (or where they can create) competition among LIFs, particularly in buyer-driven sectors such as textiles and apparel (Pietrobelli and Staritz 2018; Gereffi 1994). These giants can also prevent LIFs from designing, marketing and branding by themselves as this may threaten the core competencies of MNCs in the future (Bazan and Navas-Aleman 2004; Giuliani, Pietrobelli, and Rabellotti 2005). LIFs are also likely to be locked in specific technology cycles, which could deter further development...