On 29 April 2026, Rwanda gazetted Ministerial Order No. 004/26/10/TC, requiring non-resident providers of digital goods and services to charge and remit Rwanda's standard 18% VAT on supplies to Rwandan consumers. The order — grounded in the 2023 VAT Law and approved by Cabinet on 2 April — sweeps in streaming, cloud computing, online gaming, digital advertising, ride-hailing, web hosting and digital-marketplace services. Foreign suppliers get a three-month window to register for VAT or appoint a local tax representative; where they do neither, banks and payment processors must withhold and remit the tax at the point of payment (Fonoa).
The case for it is strong
Start with the strongest argument for the order, because it is a good one. When a Rwandan watches a film in a Kigali cinema, the ticket carries 18% VAT; when the same person streams that film on Netflix, until now it did not. As RRA Commissioner General Ronald Niwenshuti put it, "a foreign provider offering the same service to the Rwandan market should also be taxed" (The New Times). That is not protectionism — it is neutrality. A consumption tax that exempts foreign digital supplies quietly subsidises offshore platforms against local cinemas, software firms and broadcasters, and it leaks a growing share of the tax base as consumption moves online.
The design Rwanda chose is also, on its face, the right one. The OECD's VAT Digital Toolkit for Africa recommends exactly this architecture: a vendor-collection regime that puts the collection obligation on the non-resident supplier, a simplified online registration portal, and a place-of-taxation rule keyed to the customer's location (OECD). The 2023 VAT Law already established that taxable supplies include those provided online, and explicitly left the modalities to a ministerial order — so the April order is the rule-making that statute anticipated, not a surprise levy (Law N° 049/2023, RRA). South Africa has run a vendor-collection model since 2014 and Nigeria's equivalent took effect on 1 January 2026, so Rwanda is converging on a regional norm rather than improvising.
Where the design starts to wobble
The problem is not the VAT. It is what sits underneath it. In May 2025 Rwanda introduced a separate 1.5% Digital Services Tax on the gross Rwandan revenue of foreign platforms with a substantial national presence — a turnover tax layered on top of the 18% consumption tax now being collected (PwC Worldwide Tax Summaries). VAT and a gross-revenue DST are different instruments aimed at different things, and a platform can be liable for both on the same Rwandan transaction. VAT is at least creditable and ultimately borne by the consumer; a DST on gross revenue is not creditable, falls on margin, and for thin-margin services tends to be passed straight through as a price increase. Stacking the two is how a defensible neutrality measure turns into a regressive one.
That regressivity is not hypothetical. The clearest cautionary tale on the continent is Uganda's 2018 social-media excise: subscriptions fell, users migrated to VPNs, and the government missed its revenue target while degrading internet access (International Bar Association). Digital taxes that raise the headline price of connectivity and content reliably suppress the very usage they tax.
A second design flaw is the absence of a registration threshold for non-residents. Domestic businesses only enter the VAT net above RWF 20 million in annual taxable turnover (about RWF 5 million a quarter) (PwC). The April order sets out no equivalent de minimis for foreign suppliers, which means a small indie game studio or a niche SaaS tool with a handful of Rwandan subscribers faces the same registration, filing and representation obligations as Google. For marginal markets, the rational response is not compliance — it is geo-blocking Rwandan users, which shrinks consumer choice and undercuts the local startups that depend on foreign cloud and developer tools.
The withholding backstop is blunt
The payment-processor withholding fallback is pragmatic — it stops unregistered suppliers from escaping entirely — but it is a heavy instrument. Banks and mobile-money operators are not well placed to distinguish a taxable B2C digital supply from a zero-rated export, a B2B reverse-charge transaction, or a refund, and the order's own provision for consumer VAT refunds on cancelled subscriptions hints at the over-collection risk. Withholding at the rail is simple to administer and easy to get wrong, and the cost of getting it wrong lands on consumers and small merchants who must chase refunds.
A regional, not a Rwandan, fix
The deeper issue is fragmentation. Each African market that bolts a bespoke DST onto its VAT raises the fixed cost of serving the continent, and any single market is small enough that a global platform may simply decline. The proportionate path is the one the OECD toolkit and the East African Community's harmonisation agenda both point to: a single, well-designed vendor-collection VAT with a sensible threshold and a simplified portal — and the discipline to not run a gross-revenue DST in parallel once the VAT is live.
Rwanda got the harder half right. Its 18% digital VAT is principled, statutorily grounded and built on the model international bodies recommend. The unforced error is the 1.5% turnover tax sitting beneath it. Repeal or suspend the DST now that the VAT collects, add a non-resident threshold, and Rwanda would have one of the cleanest digital-tax regimes in Africa instead of a double-charged one.