One Region, Many Rulebooks

Nikita Melashchenko · (14 June 2026) · opinion

This week the EU and Korea signed a standalone Digital Trade Agreement, the EU’s second after Singapore. China, meanwhile, is reviving the FTAAP idea as host of APEC 2026, twenty years after it was first floated. The WTO’s plurilateral e-commerce agreement waits in the wings, and the General Council is trying to bottle the MC14 momentum. All of this lands in an Asia-Pacific that already has the CPTPP electronic commerce chapter, RCEP’s chapter 12, the DEPA and its accession queue, the Australia–Singapore DEA, a scatter of bilateral digital economy agreements, and the IPEF’s remains.

The scholarship has settled on a name for this, regime complexity, the worry that overlapping, non-hierarchical rulebooks governing the same domain, with no umpire among them, weaken obligations and make compliance harder. Mira Burri’s survey of the resulting framework is blunt, neither coherent nor coordinated, messy and fragmented in rules and membership alike.^[1] The literature also documents what practitioners already know. The texts are siblings but not twins. RCEP’s data-flow obligations resemble the CPTPP’s until you reach the exceptions, where RCEP’s self-judging security clause quietly swallows the rule.^[2] The DEPA was designed as a template, modular by construction, and it has indeed been borrowed, but borrowing with local amendments produces divergence at scale. Whether this dense web amounts to an alternative Asia-Pacific model of digital integration, as Neha Mishra and Ana Maria Palacio Valencia suggest, or just an unusually well-documented mess, is a live question.^[3]

Here is my reflection on it, wearing my dataset-explorer hat. The regime complex is not primarily a diplomatic problem. It is an information problem. New Zealand alone is simultaneously party to the CPTPP, RCEP and the DEPA, and the divergence between them is not evenly three-way. The DEPA borrows the CPTPP’s data-flow obligation, its self-judging security exception, and its definition of “electronic transmission” almost word for word. RCEP is the outlier on all three counts. Its data-flow article is framed as a prohibition rather than a duty to allow, it embeds a second self-judging security carve-out inside that same article instead of leaving security to the general exceptions chapter, and its electronic commerce chapter never defines “electronic transmission” at all. No trade ministry, let alone a mid-sized exporter, holds the interaction of those texts in working memory. Every new agreement, including friendly, well-drafted ones like the EU–Korea DTA, increases the compliance surface faster than it increases the liberalisation.

The optimistic reading of treaty proliferation has always been convergence-through-copying. If everyone borrows the same clauses, the complex tidies itself. The Allee–Elsig finding that treaties are largely copy-pasted lends this some support.^[4] But copying with strategic edits converges the boilerplate while diverging exactly the provisions that matter most, the exceptions and definitions, because that is where negotiators spend their deviation budget. The result is a region where the agreements agree on everything except the parts you would litigate.

Which is why I keep insisting, that coding these agreements into structured, comparable data is not academic housekeeping. It is the precondition for policy and legal scrub. A regime complex you have not mapped is a regime complex you navigate by anecdote. And in a year when new rulebooks are being added faster than old ones are read, anecdote is losing ground.