Place of Supply

Business customers: taxed where the business is

Ireland’s Revenue states the EU general rule plainly: for business to business services, “the place of supply is the place where the business receiving the services is established. This applies whether the business customer is in the European Union (EU) or not.” A business with a fixed establishment elsewhere can be treated as receiving the service at that establishment instead.

The practical consequence is who pays. Revenue’s own example is a US company supplying a service to an Irish company: the place of supply is Ireland, and the Irish customer “must self account for VAT on the service.” The seller charges no VAT, and the customer reports it on its own return. That mechanism is the reverse charge, and it is why a valid business tax ID on the customer record usually turns a taxable cross-border sale into a zero-tax invoice.

Telling business customers from consumers is the seller’s job. Revenue says a business customer’s status “is often verified through the provision of the VAT number by the customer.” If there is no number and the seller is unsure, it may treat the customer as a consumer and charge VAT, then adjust if a valid number arrives within a reasonable time and the customer was in business when the sale was made.

Consumers and digital services: taxed where the customer lives

For services sold to consumers, the EU general rule taxes the supply where the seller is established. Telecommunications, broadcasting and electronically supplied services are carved out. Revenue’s guidance on those services says the place of supply “is the place where the customer is established, has a permanent address, or usually resides.” For a seller outside the EU, that means the consumer’s member state every time.

An electronically supplied service, in Revenue’s words, is one delivered over the internet that is “essentially automated, involves minimal human intervention and in the absence of information technology does not have viability.” A SaaS subscription or an on-demand streaming plan usually qualifies. Using email to deliver a consultant’s advice doesn’t turn the advice into an electronic service, and services that mix electronic and human elements are judged case by case.

One exception applies only to EU sellers. A business established in a single member state whose cross-border sales of these services and intra-EU distance sales of goods stay at or below 10,000 EUR, excluding VAT, in the current and previous calendar year can charge its home VAT instead. The threshold doesn’t exist for a seller established outside the EU.

Other countries apply the same destination idea with their own evidence rules. Peru treats a consumer as resident if any one of four facts points there: the IP address, the SIM country code, a card issued by a Peruvian financial entity, or the address the customer registered. Chile presumes use in Chile when at least two of a similar list of four facts are true. The principle travels. The proof requirements don’t.

Proving where a consumer is: the EU evidence rules

Revenue’s guidance sets out the EU approach in two layers. First come specific presumptions for situations where the delivery channel identifies the customer: a service supplied through a mobile phone is located by the country code of the SIM card, a service through a fixed landline by where the line is, and a service at a Wi-Fi hotspot or hotel lobby by where that place is. The seller must use a specific presumption unless it holds three pieces of non-contradictory evidence pointing elsewhere.

Most digital services fall outside those channels, because the seller doesn’t know how the customer connected. Then the general rule applies: the seller “must determine the place of supply using two pieces of non-contradictory evidence.” Revenue lists the kinds of evidence that count: the billing address, the IP address, bank details such as the location of the bank, the SIM country code, the location of a fixed landline, and “any other commercially relevant information,” including information from third-party payment service providers.

Where evidence conflicts, Revenue says the seller “should use their best judgement in selecting the most reliable indicators.” It also says systems should be reviewed regularly, because complete accuracy is hard and the duty is one of best efforts. In practice that means keeping the evidence, not just the conclusion: the billing country, the card’s issuing country and the IP country at the time of sale.

How Stripe Tax decides the customer’s location

Stripe Tax makes the location decision for you, from one address. Its customer locations guide says it “uses a single address as your customer’s location when calculating tax,” and adds that this address “isn’t always the final place of taxation.”

Which address depends on how you charge. Checkout and Payment Links use the address collected in the session. For Invoicing, Subscriptions and custom integrations, Stripe takes the first viable item in a fixed order: the customer’s shipping address, then the billing address, then the billing details of the most specific payment method (the invoice’s, then the subscription’s, then the customer’s default), then the IP address. If billing details are incomplete, Stripe tries to assemble a fuller address, for example by combining the card issuer’s country with the postal code.

Two behaviors in that hierarchy matter for the books. A shipping or billing address that is present but invalid doesn’t fall through to the next item: Stripe raises a customer_tax_location_invalid error instead. And when a subscription has no recognized location at all, its invoices “continue to finalize automatically but without calculating taxes,” with automatic tax switched off on the subscription and the invoice. The customer pays, and no tax is recorded. Setting tax.validate_location to immediately on the customer catches the problem when the address is saved, not months later in a tax report.

Stripe also says IP addresses are “more likely to resolve to an accurate address” in regions like the EU that don’t impose local taxes, while it doesn’t recommend relying on an IP address for US customers. Comparing the two systems: Stripe needs one address and the EU general rule asks for two consistent pieces of evidence. For most consumers they agree. Where they might not, keep the second piece of evidence yourself.

Excluded territories and other places that look taxable

A country’s VAT doesn’t always cover all of its map. Stripe’s guide gives the example that the Italian postal code 00120 “identifies Vatican City, where Italian VAT doesn’t apply,” and says to collect the customer’s postal code or state so Stripe Tax can tell when a customer is in an excluded territory. Without the postal code, a Vatican customer looks Italian.

The opposite case is also worth knowing: a customer established outside every country you are registered in. Stripe says that without a registration in the customer’s location, “the calculation returns zero tax.” That zero is a statement about your registrations, not a ruling that the sale is untaxed. A sale whose place of supply is in a country where you should be registered and aren’t still carries that country’s tax, whatever the invoice shows.

Why the place of supply shows up in your QuickBooks file

Every tax line on a Stripe invoice reflects a place of supply decision, and those decisions are what your liability account is made of. A US company selling digital services abroad usually ends up with US sales tax, EU VAT and other countries’ taxes in one liability account, cleared by payments to several tax authorities on different calendars. Tying that account back means knowing which country each tax amount was charged for, which is the place of supply Stripe decided.

Stripe’s itemized tax export carries that country on each row, along with a taxability reason that records why a line was taxed, zero rated or reverse charged. Filtering the export by country is how you build each return and the matching slice of the liability account. When a customer’s location was wrong, the fix is a credit note and a corrected invoice in Stripe, so the correction shows up in Stripe’s tax reports, which is where each return is built, rather than as an adjustment made only in QuickBooks.