12 SEP 2026 — ASEAN's Digital Economy Framework Agreement is described as a two trillion dollar opportunity. A survey released at the American Chamber of Commerce Singapore's regional economic conference on 8 September asked business leaders what they had done about it. Four per cent have an action plan.
Thirty-five per cent said their organisation was unfamiliar with DEFA entirely. Forty-nine per cent were aware of it but had not assessed what it means for them. Twelve per cent had assessed its relevance.
Read the sample before the finding
The 2026 ASEAN Business Outlook Survey was run by AmCham Singapore with Accenture and Google. It drew 184 responses across eight markets: Cambodia, Indonesia, Malaysia, Myanmar, the Philippines, Singapore, Thailand and Vietnam.
The survey sample of 184 firms is small, and it is not representative. An American Chamber of Commerce survey reaches US multinationals, not ASEAN businesses in general. These are the companies with the most legal resources and the staff paid to read trade agreements.
That makes the 84 per cent harder to wave away. If 84 per cent of well-resourced multinationals have not heard of DEFA or have not looked at it, the figure for a mid-sized Indonesian logistics firm is not going to be better.
Awareness is not the worrying part
An unsigned agreement that nobody has assessed is not obviously a failure. DEFA is not in force, its text is still being negotiated, and a company that waits to read the final version before committing planning time is behaving sensibly rather than negligently.
The concern is about sequencing. Harmonising data rules is worth two trillion dollars only if firms restructure their operations around the new framework. Restructuring data flows and contracts takes longer than ratifying a treaty. A business that starts assessing on the day the agreement is signed is starting years late.
That makes the 49 per cent who are aware but have not assessed the group worth watching, more than the 35 per cent who have never heard of it.
The two trillion is not a measurement
The figure attached to DEFA in almost every account of it, including the first line of this one, is an estimate produced in advocacy for the agreement. It projects the value of a harmonised regional digital economy against a counterfactual in which harmonisation does not happen.
That is a legitimate way to size an opportunity, but a poor way to judge progress. Signing the agreement produces none of the two trillion by itself. That depends on firms restructuring afterwards, which is slower and far less certain than ratification. A survey showing that 4 per cent of firms have a plan is not a small problem. It is a direct measurement of the adoption the two-trillion-dollar figure depends on.
The part that is already operational
One line in the survey's framing deserves more attention than the headline percentage. Organisations deploying agents that move data between markets, make automated decisions, or interact with customers across borders will face direct operational implications.
That describes software many firms in this region are shipping now. An AI agent that reads a customer record in Singapore, calls a model hosted elsewhere and writes back a decision is a cross-border data flow with an automated decision in the middle, and both halves are exactly what a digital economy agreement governs.
The gap that matters runs between the systems being deployed this year and the rules still being drafted for them. That gap widens on its own while 49 per cent of companies wait.
We have been tracking the supply side
This publication has covered DEFA from the negotiating side repeatedly — the signing timetable, and the pattern of members continuing to write their own national rules while the regional rulebook is drafted.
This survey is the first demand-side measurement we have seen. The supply and demand sides point the same way. The agreement is being negotiated by governments whose own members are still writing divergent national laws, for a business population that has largely not looked at the draft. Neither of those is fatal on its own. Together they describe a framework whose benefits depend on adoption that nobody has yet started.
What would change the number
A ratified text with a compliance date. Assessment is cheap to defer while the obligations are hypothetical and expensive to defer once they are dated, and the 49 per cent are waiting for precisely that signal.
The other thing that would move it is a national implementation that bites. Regional agreements become real to a business when a member state writes them into its own law with a deadline attached, and that is also the point at which divergence between members stops being a drafting problem and becomes an operational one.
What the 49 per cent can do now
You do not need the final text to start an assessment. Map your cross-border data flows, identify which ones carry personal data, and flag any that involve automated decisions about people. That inventory is useful regardless of what DEFA says, because every plausible version of it governs those three things.
That inventory is also the input for complying with national laws already in force. A firm that cannot answer those three questions is already behind on the rules in force today, whatever DEFA eventually says.
