Rules of Origin in a World of Global Value Chains: New Rules Required?
Rules of origin (“ROOs”), which determine whether goods are eligible for a reduced rate of import duties under a trade agreement, have always played an important role in global trade. But are they still fit for purpose? Do existing rules adequately address the challenges posed by increased interdependence among economies and the rise of Global Value Chains (“GVCs”)?
ROOs can be divided into non-preferential ROOs and preferential ROOs. Non-preferential ROO are not relevant when the most favoured nation obligation applies, but they are important in the context of, for example, imposing countervailing duties. They are governed by the WTO Agreement on Rules of Origin, through a set of general principles on their use. The general rule is that Members’ ROOs must be transparent and applied in a non-discriminatory manner.
Preferential ROOs are contained in preferential trade agreements (“PTAs”) and seek to ensure that goods from countries (C) which are not members of the PTA (A and B) do not pass through a low tariff member (A) to benefit from the preferential tariff treatment afforded to it by the other member (B). ROOs can however also be used restrictively, for protectionist purposes (being designed in such a way that only products of a certain origin qualify) or even to exclude products from the scope of PTAs (again, by defining them narrowly). Rules of origin in FTAs are very detailed and complex (for example, NAFTA’s annex covering ROOs is close to 200 pages long). The diversity of PTAs also results in a lack of uniformity in preferential ROO: there are over 300 PTAs in force worldwide and most have their own distinct set of origin rules (the EU has notably spearheaded an initiative for harmonising them in Europe and around the Mediterranean).
Despite the complex details, there are some common principles. The traditional rule is that the country of origin will be the last country where a “substantial transformation” took place. There are three main methods for determining this: a change of tariff classification (the most frequent method), percentage value added in a country, and the use of specific manufacturing process. The result is that origin is often determined by the location of manufacture and/or assembly.
This sounds reasonable, except that GVCs have fundamentally changed in recent decades. Assembly of a product is not the most valuable part of the production process any more. The problem is well illustrated by the Apple iPhone. The iPhone is famously designed in the United States and manufactured in China from components originating from other countries such as Japan and Korea. A ROO that is based on change in tariff classification (e.g. the one applicable to preferential trade between Australia and China) would clarify the origin of the iPhone as China. Yet manufacturing processes in China constitute approximately 2% of the final value of the end product. This means that only this 2% determines the origin, and therefore the applied tariff rate, while the remaining 98% of the final product’s value is apparently irrelevant.
To understand what is going on, we need to look more closely at modern GVCs. These broadly consist of four parts: (1) product development, research and design; (2) inputs; (3) production; and (4) marketing, sales and distribution. Most value is created in the first and last stages. The current approach, by focusing on the actual manufacturing or assembly of the final product, does not truly reflect the origin of a product (at least in terms of its value). This is closely linked to the increasing role which services play in the operation of GVCs. Manufacturing activities have become increasingly service-dependent, while services are embedded in manufacturing value chains from the pre-production stage up to the delivery of goods. In fact, services now account for the majority of trade in intermediate goods and almost half of world trade on a value-added basis. This integration of goods and services in GVCs, however, is not seriously taken into account by current ROOs.
Is this a problem? A “liberal” rule of origin means that more goods will enjoy preferential terms under an PTA – presumably a good thing. More negatively, it might bias the comparative advantage of poorer countries towards low-value assembly operations, and disincentivize developed countries from liberalising trade amongst themselves. Why lower tariffs for trade with the US, when most of your value-add can be routed through a developing country which itself receives preferential treatment from the US? At the very least, current ROOs are an opaque way of governing multilateral trading relations. The natural solution seems to be to confer origin on the country where the most value was added, which would cover value added services such as research and development, design, and marketing. While such an approach has its merits, it carries two risks. Firstly, a value-added test would likely generate substantial compliance costs for companies. Secondly, this approach will penalise less-developed countries whose primary comparative advantage is cheap labour and cheap materials. If services are taken into account, you would expect more products to have originated in developed countries.
A more radical solution advocated by some scholars is to implement a multiple country of origin rule. Indeed, the designation of one country of origin to a product is becoming increasingly outdated and misleading. Introducing the designation of multiple country of origin based on value added at different stages in different countries seems like a valuable alternative in principle, even though it is unrealistic for now.
Perhaps the only option currently available is to make wider use of cumulation amongst parties to multiple PTAs. Cumulation allows goods of third countries to count as though they were of domestic origin. It makes most sense where two parties each have a separate PTA with a third party. Such a course of action would have the added benefit of reducing some of the trade diversion normally associated with PTAs. Typically, however, cumulation is limited to countries which apply the same rules of origin in those PTAs as apply to the country the products of which are cumulated with these products.
Whatever happens, it is clear that the existing approach to ROOs needs to be reviewed in light of complex and rapidly changing supply chains, and possibly modernised. Alas, with the current crisis at the WTO, it appears that for now reforms can only be undertaken at bilateral level.
Written by Samuel Coldicutt and Dr Zvenyslava Opeida, Visiting Scholar at the University of Pittsburgh School of Law.