Rules of Origin (ROO) are among the most controversial and complicated aspects of international trade law. This is especially so in a world in which multilateralism is in decline, paving the way for trade based on the rules found in preferential bilateral or regional free trade agreements (FTAs). The intricate nature of ROO is further exacerbated by intangible nature of the digital economy, where a significant portion of global GDP and world trade is now generated (roughly 15 per cent and 25 per cent respectively).
Preferential FTAs are the key exception to the World Trade Organization (WTO) principle of Most Favoured Nation (MFN), itself found in Article I of the General Agreement on Tariffs and Trade (GATT) and Article II of the GATS (General Agreement on Trade in Services). Preferentialism allows for better treatment to goods and services originating from FTA party countries, typically in the form of lower tariffs, than is accorded to the rest of the WTO community. In the case of goods, this treatment is contingent on the relevant good actually originating from the partner country. The good is not legally entitled to the lower tariff if it is not “from” the partner country, but is instead merely shipped through it.
The WTO Agreement on Rules of Origin usefully sets out standardized procedures for how origin of goods is calculated, requiring that all WTO members apply their ROO impartially, transparently, and consistently, aiming to ensure that ROO do not restrict, distort, or disrupt international trade. More importantly, each FTA specifies what specific percentage of a given product is required to be “from” the partner country, or regional grouping, for the purposes of satisfying the conditions for preferential treatment. These levels are a vital element of trade negotiations, currently featuring prominently in the CPTPP review and USMCA renegotiations for example. ROO can be complicated for composite goods like automobiles, which are manufactured and assembled across of range of jurisdictions in complex value chains. The burden of complying with ROO is thought to be so onerous that some companies choose to forego their preferential entitlement, trading instead on MFN terms.
Difficulties with fulfilling and verifying ROO are even more complicated in the context of digital trade. Cross border flows of data not only travel much faster than goods, they are difficult to track because data consists of bytes which can be duplicated almost without cost and sent anywhere simultaneously. As such, determining the “origin” of data is problematic. Traditional ROO, designed for physical goods, are ill-suited to their digital equivalent, which under digital trade agreements (DTAs) can attract preferential treatment such as zero customs duties or exemptions from data localization obligations. If the current global moratorium on customs duties on electronic transmissions were to end, as it may do in the near future, the issue of verifying the origin of data would become all the more urgent because such preferential treatment would likely only be accorded to digital goods from DTA partners, rather than on an MFN basis.
Technology might provide a way for assessing data sources as worthy of preferential treatment, a kind of ROO proxy. Digital firewalls, such as that operated (in)famously by China, are capable of evaluating the origin of data flows and restricting them accordingly, although virtual private networks can frustrate this strategy. Blockchain might also work, operating as a record of a transaction and evidence of traceability, complete with a digital certificate of origin. Some countries, including the UK and Singapore are already experimenting with this approach.
A multilateral framework for digital ROO, perhaps devised via the WTO, would be helpful in smoothing out the procedure for assessing the origin of data, even as it remains up to each state to decide what percentage of foreign-ness is acceptable to attract preferential treatment, whether this takes the form of zero customs duties or precludes the obligation to establish data storage within the jurisdiction. An agreement on digital ROO could attribute origin based on the residency of key service providers, the location of software development, or the jurisdiction hosting the data storage hardware. Additional factors like total software development hours or proprietary algorithm contributions could also fit into a ROO assessment and offer a greater degree of fairness, potentially advantaging developing countries. Based on standardized rules such as these, DTA parties would be free to set digital ROO thresholds, and also link them to social values such as minimum wages (as in case of the USMCA) or carbon content of energy consumption, along the lines of the EU’s CBAM.
Leave a Reply