On 25 September 2026, Indonesia began collecting value-added tax on foreign digital purchases through its banks. The system is called SPP-TDLN. It rests on Finance Minister Regulation No. 49 of 2026 (PMK 49/2026), and the Directorate General of Taxes (DJP) set the start date in announcement PENG-6/PJ/2026. The DJP's own SPP-TDLN page says the system introduces no new tax and no new VAT rate. It changes only how existing VAT on these transactions is collected.
The strongest case for the rule
The revenue case is plain. Indonesia already requires large foreign platforms that register as PMSE collectors to charge VAT. Smaller foreign sellers, and sellers that never register, sit outside that net. Their Indonesian customers pay no VAT, while domestic competitors do. Chasing thousands of offshore merchants one by one is impractical. Collecting at the payment layer, where the Indonesian consumer's money moves, reaches sellers that Indonesia could never audit. DJP Director General Bimo Wijayanto described the logic as asserting taxation rights (Investortrust) "directly from the payment gateway." That is a coherent response to a real compliance gap, and a government is entitled to tax consumption inside its borders.
Why this design is the better path
What matters for innovation policy is the instrument chosen. A digital services tax (DST) is levied on gross revenue from specific digital activities. It falls on big foreign firms, can be passed on to local businesses and consumers, and invites retaliation. VAT is a general consumption tax, applied under the same rules to domestic and foreign sellers. SPP-TDLN extends that logic; it does not create a special levy on a disfavoured class of companies.
The design also fits the trade context. The US-Indonesia Agreement on Reciprocal Trade was signed on 19 February 2026. Digital Policy Alert reports that its Article 3.1 commits Indonesia not to impose digital services taxes or similar measures. I could not verify the full clause text, because the tracker's excerpt is truncated and the White House fact sheet does not mention DSTs. A neutral, non-discriminatory VAT is the kind of measure such a clause is meant to leave alone. A nondiscriminatory consumption tax is a poor target for any treaty complaint, and it spares Indonesian users the risk of tariff retaliation tied to a DST dispute.
Where the risks lie
The rule is sound in principle, but its operation is where it could fail.
- Incomplete coverage at launch. According to DDTC, citing system operator PT Jalin Pembayaran Nusantara, collection initially runs only through state-owned Himbara banks and LinkAja, with others to follow once integrated. The same report says full collection depends on support from international card principals. If Visa and Mastercard transactions are not captured, much of the foreign digital spend escapes the net, and the system rewards consumers who pay by card network rather than local rails.
- Price effects without clear notice. The DJP says foreign sellers must include Indonesian VAT in the price customers pay, and that the amount a seller receives may be reduced accordingly. Sellers need clear checkout signals. Otherwise small developers and creators, who often sell at fixed global price points, may see revenue fall without understanding why.
- Double collection. The DJP states that sellers already appointed as PMSE collectors keep the old mechanism and are outside SPP-TDLN. Both systems run side by side, so the boundary between them must be reliable. A mistake means a consumer is taxed twice, or a seller is told it owes tax it already remitted. The DJP has said the two regimes should not produce double collection (DDTC); the real test will be how refunds and disputes are handled in practice.
- Data minimisation. Routing tax through payment institutions means a system operator learns what Indonesians buy abroad. DDTC reports that transaction data confidentiality is protected under the rule, but the DJP's public SPP-TDLN page does not address the point. Purchases of news subscriptions, VPN services, health apps, or political content are sensitive. Regulators should publish what data fields are shared with the tax authority, how long they are kept, and who audits access. Tax collection should not become a visibility layer over lawful online activity.
What a proportionate rollout looks like
Indonesia should publish a transparent coverage report after the first quarter: which payment methods are captured, which are not, and how much was collected. It should avoid widening the tool beyond VAT. Using the same payment-channel plumbing for other levies, blocking decisions, or data requests would turn a tax-administration fix into something far less benign. And it should keep the commitment visible that no separate digital tax is being layered on top.
The wider lesson is that consumption-based taxation of digital trade can work without a DST. Countries weighing a DST, with its trade-war exposure and distortions, can look at VAT collection through payment channels as a less disruptive alternative. Whether Indonesia proves it depends on two things: card-network cooperation and privacy discipline.