February 04, 2021
Regional Trade Agreements with Global Value Chains
François de Soyres, Julien Maire and Guillaume Sublet1
The rise of Global Value Chains (GVCs) is a dominating feature of the recent evolution in the structure of international trade. In the OECD countries, the import content of exports increased by 63% between 1995 and 2011, reaching a value of 24.3% on average.2 This internationalization of production through GVCs calls for a reevaluation of the effects of regional trade agreements (RTA) on trade flows and their implications for the debate on regionalism and multilateralism.
Since Viner (1950), the effects of RTAs have been framed in terms of two concepts: trade creation and trade diversion. An increase in trade within the agreement zone is called trade creation. In turn, a decrease in trade between the agreement zone and the rest of the world is called trade diversion.3 Trade creation and trade diversion are key considerations for the desirability of RTAs for the countries involved (see Panagariya (2000) for a survey). This view is mostly inspired by a world of trade in final goods where consumers shop for the lowest prices, inclusive of tariffs, for their purchases. However, it misses a key feature of today’s production process: with GVCs, trade agreements also facilitate firms’ access to lower-priced inputs, which in turn impacts their competitiveness both at home and in other markets.
In de Soyres et al. (2020), we investigate how GVCs alter the consequences of RTAs. This note summarizes the main results of the paper, drawing out the most policy-relevant implications. Our first main finding is that RTAs increase intra-regional trade and decrease imports from the rest of the world to the region, which is in line with the existing literature. More surprisingly, we also find strong evidence of an increase in exports from the region to the rest of the world. To understand this finding, we then refine our empirical analysis to account for participation in GVCs. Results show that production linkages are a key determinant of the increase in outflows from members of the RTA to the rest of the world.
In a second part, we augment an intra-industry trade model with trade in intermediate inputs. Our model corroborates our empirical findings, which cannot be rationalized in a mode focusing on trade in final goods only. We then use our model to investigate the implications of our empirical and quantitative findings for regionalism and multilateralism. Regionalism refers to the fragmentation of the world into preferential trade regions while multilateralism refers to the indiscriminately lowering of trade barriers on a world scale. Our analysis highlights that, in the presence of GVC, regional integration increases incentive for the rest of the world to increase its tariffs. Moreover, production integration within the agreement zone creates strategic complementarities in tariffs. While this fosters cooperation on trade policy, it also exacerbates the negative consequences of a breakdown in trade policy cooperation. In this sense, regional agreement with GVCs can be a stumbling block for multilateralism.
Empirical Findings: GVCs are associated with an increase in exports from the agreement region to the rest of the world
Using more than 220 trade agreements over the past 60 years, we separately identify the changes in trade within the region, the changes in imports from the rest of the world to the region, and the changes in exports from the region to the rest of the world. Our identification strategy exploits the panel structure of the data to control for country-pair and country-year fixed effects. Importantly, taking the direction of trade into account allows for the separate identification of the change in inflows to, and the change in outflows from, the region.4
As expected, we find evidence of an increase in intra-regional trade, which is in line with the existing literature. More interestingly, we also find strong evidence of a decrease in inflows to the region and an increase in outflows from the region. The evidence of strong opposite effects on inflows and outflows is, to the best of our knowledge, new to the literature.
We then refine our analysis to account for participation in GVCs. To that end, we build two indices measuring the degree of production integration across countries capturing (1) the bilateral production linkages (between the exporter and the importer) and (2) the production integration with the agreement region as whole. We find that accounting for GVCs matters most for the increase in outflows from the region. In particular, production integration within the agreement zone accounts for the positive effect of RTAs on outflows from the region (Figure 1).
Indeed, countries that rely on the regional value chain to source their inputs are the ones that benefit the most from the RTA. By decreasing the cost of production input, an RTA reduces marginal cost for firms in the agreement zone. As a result, the boost in industry performance fosters more exports from the region to the rest of the world. These findings can be interpreted as a manifestation that globalization benefits industry performance through regional input-output networks.
Model of international trade with Global Value Chains
Motivated by these new facts, we then augment a model of intra-industry trade à la Chaney (2008) with GVCs. Our goal is to assess if the addition of GVC into an otherwise standard model can rationalize the our empirical findings.
In our setup, firms use labor and intermediate inputs to produce. Firms decide whether to export to other countries, which endogenously determines the global network of value chains. Importantly, the marginal cost of production depends on a firm's access to intermediate inputs, which is determined endogenously by the global network of value chains. Each firm's marginal cost is endogenously determined as a result of all other firms' pricing and export decisions. Numerical simulations show that, following a RTA, our setup with GVCs generates an increase in outflows from the region while standard models of trade in final goods do not. This result highlights that adding GVCs to standard frameworks of trade policy analysis is key to reconcile the data and model predictions.
Policy implications: Regionalism and Multilateralism
Finally we use our model to investigate the implications of our empirical and quantitative findings for regionalism and multilateralism. Regionalism refers to the fragmentation of the world into preferential trade regions. Preferential trade regions fall in two categories depending on whether members of the region set common trade barriers with the rest of the world (a customs union), or not (a free trade agreement). Multilateralism refers to the indiscriminatory lowering of trade barriers on a world scale.
Is regionalism a building block, or a stumbling block, for multilateralism? We say that regionalism is a building (resp. stumbling) block for multilateralism if the incentive for protectionism is dampened (resp. heightened) by a preferential trade agreement (Frankel, (1997), Krishna (2005)). The WTO puts restrictions on members of a preferential trade region to not increase their external trade policies. Hence the answer to the question above cannot be read directly from data on external tariffs set by members of a preferential trade region. This is where our model proves particularly useful. We simulate the counterfactual scenario where countries set their preferred tariffs following a regional trade agreement. We evaluate the incentives for trade protection by simulating preferred tariffs before and after the formation of a preferential trade region. If the incentives for protectionism are heightened, it suggests that countries will be less inclined to reduce their trade barriers as a result of the preferential trade region.
In short, GVCs make countries’ preferred tariffs strategic complements: an increase in tariffs by the rest of the world heightens the members’ incentives to raise their tariffs and vice versa. Strategic complementarities in tariff policies tie countries’ incentives for trade protection: an increase in tariffs by the rest of the world heightens the members’ incentives to raise their tariffs and vice versa. We simulate a breakdown in trade policy cooperation, also known as a tariff war, as the (Nash) equilibrium resulting from countries pursuing their preferred tariff policy.
We find that the type of preferential trade region is a key determinant of its members incentive for protectionism. Figure 2 summarizes our findings qualitatively:
- With GVCs, the rest of the world has an incentive to increase its tariffs vis-a-vis the agreement zone. This is a response to the surge in outflows from the region, which increases the rest of the world’s incentives for protectionism.
- Members of a customs union have an incentive to set a higher tariff on imports from the rest of the world. In contrast, members of a free trade region compete for trade with the rest of the world which results in dampened incentives for protectionism.
- We conclude that customs unions are a stumbling block whereas free trade areas are a building block for multilateralism.
It is insightful to contrast our finding on the incentives to set tariffs with the tariffs that would result from compliance with WTO rules. Article XX of the WTO imposes that members of a preferential trade region do not increase their trade barriers, which constrains all members of a customs union to level their external trade policies in line with the most favorable trade policy of its members. However, the heightened incentive for trade protection for members of a customs union makes it harder to make the next step towards multilateralism. To summarize, WTO rules induce customs unions to make a bigger step towards multilateralism than do free trade agreements. Yet, in terms of incentives for further trade liberalization, customs unions become a stumbling block on the path to multilateralism whereas free trade agreements facilitate the next step and are thus a building block for multilateralism.
Antràs, P. (2020). “De-Globalization? Global Value Chains in the Post-COVID-19 Age”, Working Paper.
Carrere, C. (2006), “Revisiting the effects of regional trade agreements on trade flows with proper specification of the gravity model”, European Economic Review ,50(2):223–247.
de Soyres, F., J. Maire and G. Sublet (2020), “Regional Trade Agreements with Global Value Chains”, Working Paper.
Frankel J. (1997). “Regional Trading Blocs in the World Economic System”. Peterson Institute for International Economics.
Krishna, P. (2005). “Trade Blocs: Economics and Politics”. Cambridge University Press.
Magee, C. S. (2008). “New measures of trade creation and trade diversion”, Journal of International Economics,75(2):349–362
Panagariya A. (2000), “Preferential trade liberalization: the traditional theory and new developments”, Journal of Economic Literature, 38, pp. 287-331
Viner, J. (1950). “The customs union issue”. Carnegie Endowment for International Peace, New York.
1. François de Soyres is a Senior Economist in the Division of International Finance of the Federal Reserve Board of Governors. Julien Maire is a Research Assistant at the Peterson Institute for International Economics. Guillaume Sublet is an Assistant Professor of Economics at the Université de Montréal and a member of CIREQ. The views expressed are solely the responsibility of the authors and should not be interpreted as reflecting the views of the Board of Governors of the Federal Reserve System or of any other person associated with the Federal Reserve System. Return to text
2. Despite the slowdown of international trade flows relative to GDP over the past few years, Antras (2020) argues that global value chains will likely remain prominent on the international stage given the large economies of scale that firms benefit. Return to text
3. Trade creation is thought to be associated with resources being shifted from relatively inefficient domestic suppliers towards more efficient regional suppliers. In contrast, trade diversion may shift resources from efficient extra-regional suppliers towards inefficient regional suppliers. Return to text
4. Our empirical approach is most closely related to Magee (2008), which estimates the effects of trade agreement using a panel of 133 countries from 1980 to 1998. Also closely related is Carrere (2006) who uses a gravity model to assess trade creation and trade diversion effects. Our contribution comes from the broader coverage of RTAs and countries, our focus on the effect of RTA on inflows to and outflows from the region, and on our account of the role of GVCs. Return to text
de Soyres, François, Julien Maire, and Guillaume Sublet (2021). "Regional Trade Agreements with Global Value Chains," FEDS Notes. Washington: Board of Governors of the Federal Reserve System, February 04, 2021, https://doi.org/10.17016/2380-7172.2841.
Disclaimer: FEDS Notes are articles in which Board staff offer their own views and present analysis on a range of topics in economics and finance. These articles are shorter and less technically oriented than FEDS Working Papers and IFDP papers.