State of the New Zealand company register

What the register actually is

The register is much larger than the economy it records: most entities ever incorporated have since been removed, and not all of those that remain are active businesses. These figures set the scale for what follows.

“Active” means currently on the register; “removed” covers companies struck off or dissolved over its whole history. GST registration and a website are shown only where the register records them — plenty of trading companies have neither on file, so read those two as a floor, not a full count.

Not all of these are trading businesses

Around one active company in ten appears to be a holding vehicle — the kind a law or accounting firm registers for a family trust or a nominee arrangement, rather than an operating business. Set alongside how few carry an outward sign of trading, it’s a reminder that the headline company count isn’t a count of businesses.

Formations hold steady; insolvent failures have climbed

New incorporations run at a fairly flat 12,000–16,000 a quarter. Insolvent liquidations — companies wound up because they can’t pay their debts — are a much smaller number on a different scale, which is why they sit in their own panel below rather than sharing an axis. We count only insolvent liquidations here: solvent members’ liquidations, where a healthy company is wound up deliberately, are excluded, because lumping them in overstates distress.

Most companies aren’t liquidated — they’re quietly struck off

Insolvent liquidation is only a small part of how companies leave the register. Each year tens of thousands are removed, but only a couple of thousand are wound up as insolvent — the rest are struck off, usually for not filing after they’ve stopped trading. The failure figures on this page cover roughly one company in twenty that leaves; the other nineteen go without any insolvency process.

Which industries fail most

Not every industry carries the same risk. Setting insolvent liquidations against the number of active companies in each industry gives a rate that’s comparable across sectors of very different sizes. Hospitality and construction come out where you’d expect — several times the rate of professional or financial services.

One-director companies fail at more than twice the rate

Board size turns out to be one of the clearest structural signals in the register. Among limited companies incorporated over 2010–2022, those with a single director are currently in liquidation at around two and a half times the rate of those with two or more — and the rate keeps falling as directors are added. A second name on the board tends to mean a co-owner, an outside investor or simply someone else watching the books.

Failures peak at four to five years, not year one

You might expect most companies to fail in their first year or two. Not by liquidation: a business that folds that fast is usually just struck off, never formally wound up. Insolvent liquidations peak at four to five years — long enough to have taken on staff, leases and credit — then tail off.

The five-year failure rate has been drifting down

Against the recent rise in the raw number of liquidations, the underlying five-year failure rate has been drifting down. Of companies incorporated in 2001, about one in forty was insolvent-liquidated within five years; for the 2018 cohort it was closer to one in a hundred. Following whole cohorts from birth removes the survivorship problem that dogs most register statistics.

Half of every cohort is gone within a decade — almost all of it quietly

The charts above each take one slice of failure. This one follows every limited company incorporated since 2001 from the day it was registered, and asks what has become of it at each age. By five years about seven in ten are still on the register; by ten, only half. Of those that have gone, though, the great majority were struck off rather than liquidated — the thin band along the bottom is formal insolvency, and it never climbs past a few percent. And most of those quiet exits aren’t failures at all: the owner retires, the company has finished the one job it was set up for — holding a property that’s since been sold, say — or the business simply stops filing. The kind of failure the rest of this page measures is the rare way out.

Method and limits

• Sources. The Companies Register and the Insolvency Register, both from the New Zealand Companies Office. No other agency’s data is used on this page.
• Failure means insolvent liquidation. We use the insolvency record, not company removals — most companies leaving the register are struck off quietly, not liquidated, so removals are a poor measure of failure. Solvent members’ liquidations are excluded throughout.
• One case, not one appointment. A liquidation with more than one appointed practitioner is counted once, so joint appointments don’t inflate the totals.
• The failure series starts in 2018. The register’s electronic insolvency record only becomes reliable around 2001, and we show recent years where the picture is clearest.
• The survival curve follows whole cohorts. It tracks every limited company incorporated since 2001 from birth and splits each age between liquidation, removal and still-registered using a standard competing-risks method (Aalen–Johansen), which accounts properly for companies too young to have reached the older ages — so it carries no survivorship bias. The split of the quiet-removal exits by industry is deliberately not shown: the register drops a company’s industry code when it’s removed, so that breakdown can’t be measured.
• Aggregates only. Everything here is a register-wide total or rate. Nothing on this page scores or judges an individual company, and no figure is drawn from what Legible’s own users look up.
• Updated monthly from each new register snapshot. The as-at date above tells you which one.

Legible is a research aid built on the public registers. It doesn’t provide credit scores, financial advice, or AML/CFT certification. See our service disclosure for the data-use commitments behind figures like these.