When do you actually need a local entity?

Founders ask me about setting up a US Inc or an Australian Pty Ltd long before they have a customer in either country. It is one of the most expensive ways to feel like you are making progress.

An entity is a legal wrapper. It lets you employ people locally, contract as a local party, hold a local bank account, and change where your profit is taxed. That is the whole list. Selling is not on it, and neither is being taken seriously.

Almost everything founders believe the entity unlocks, they can already do from New Zealand. The question worth asking is not "should we set one up" but "what, specifically, is blocked without it".

You can sell, invoice and get paid without one

In Australia, a New Zealand company can register for GST as a non-resident without incorporating anything. The trigger is turnover, not presence: you must register once your GST turnover from sales connected with Australia reaches A$75,000 across twelve months (Australian Taxation Office).

The United Kingdom is stricter, in a way that catches people out. The GBP 90,000 VAT threshold that UK businesses enjoy does not apply to you. A business with no UK establishment has no threshold at all, and must register from its first taxable supply in the UK (HMRC). HMRC is also explicit that a registered, serviced or virtual office address does not by itself create a UK establishment, so the mail-forwarding trick changes nothing in either direction.

The United States has no federal sales registration to worry about. Since South Dakota v. Wayfair (US Supreme Court, 21 June 2018), a state can require a seller with no physical presence there to collect sales tax. South Dakota's own small-business exception was US$100,000 of sales or 200 transactions in a year, and every state now sets its own thresholds.

All three are registrations, not incorporations. You can carry a tax obligation in a market without owning a company in it, and most NZ tech exporters do exactly that for years.

What actually forces the decision

Four things genuinely move the answer from no to yes.

• Someone starts signing on your behalf. Inland Revenue's guidance on double tax agreements is clear that an agent who habitually exercises authority to conclude contracts for you creates a permanent establishment. Your first offshore salesperson closing deals is the trigger. Your first offshore customer is not.
• You want to employ, not contract. An employer of record carries the first hire or two without an entity, and stops being good value somewhere around the third or fourth.
• The buyer's procurement genuinely cannot contract offshore. Real, and common in government and large enterprise. Test it before you believe it. It is more often an assumption inside the deal team than a written policy, and the person who can confirm it is rarely the person who told you.
• You need local payment rails. Some buyers can pay an offshore account only by starting a process nobody in their finance team wants to own.

Notice what is missing from that list: credibility. No buyer has ever chosen a vendor because of where the company was registered. They choose on the problem you solve and the risk you carry for them.

So when do you actually need one?

Three questions, in order:

• Is someone in that market concluding contracts in your name, or about to be?
• Has a real buyer told you in writing that they cannot contract with a New Zealand company?
• Are you employing more people there than an employer of record can sensibly carry?

If all three are no, you do not need an entity. You need a tax registration where the thresholds bite, a workable way to get paid, and more customers.

The cost of being early

An entity you do not need is not free. It is annual filings, a second set of accounts, an accountant in a second jurisdiction, transfer pricing between the two companies, and a dormant shell that still has to be fed every year. Closing it later costs money too.

The bigger cost is attention. The weeks a founder spends on incorporation, bank onboarding and a registered agent are weeks not spent in front of buyers, and buyers are the only thing that was ever going to tell you whether the market is real.

Get the first five customers, then let them tell you which wrapper you need. What are you hoping the entity will prove that five signed contracts would not?

Sources. Australian GST: non-resident businesses must register once GST turnover from sales connected with Australia reaches A$75,000 (Australian Taxation Office, "GST registration for non-resident businesses," ato.gov.au, accessed 16 September 2026). UK VAT: the GBP 90,000 registration threshold applies to UK-established businesses (HMRC, "Register for VAT," gov.uk); a non-established taxable person has no threshold and must notify HMRC regardless of the value of the supply, and a registered, serviced or virtual office does not by itself create a UK establishment (HMRC, VAT Registration Manual VATREG37150 and VATREG37250, gov.uk). US sales tax: South Dakota v. Wayfair, Inc., 585 U.S. 162 (US Supreme Court, 21 June 2018), which upheld South Dakota's economic nexus law and its US$100,000 or 200-transaction small-seller exception; thresholds now vary state by state (Streamlined Sales Tax Governing Board, streamlinedsalestax.org). Permanent establishment: an agent who habitually exercises authority to conclude contracts on your behalf is a dependent agent and creates a permanent establishment (Inland Revenue, guidance on double tax agreements and permanent establishment, ird.govt.nz). The employer-of-record crossover point, the procurement test and the three-question sequence come from my own work with NZ tech companies entering Australia, the UK and the US, not from published data. This is general information, not legal or tax advice - thresholds and treaty positions change, and your facts matter. Take advice before you incorporate anything.

Nick Burns is a fractional CRO for New Zealand B2B tech companies, at home and expanding internationally. He co-founded Emendo and sold it to McKesson, then the 14th-largest company on the Fortune 500, and has since helped 60+ tech companies grow sales.