November 12, 2026

A ‘non-toxic’ label is no defence when the sand tests positive for asbestos

Two young children playing with toys in a colorful outdoor sandpit on a sunny day.

The busiest trading weeks of the year are about to run on the weakest safety regime available. Consumer NZ is telling shoppers to keep any toy containing play sand out of Christmas stockings, after tremolite asbestos was found in Faber-Castell’s ‘Plant & Grow Woodland Forest’ kit, a product sold through Spotlight and the Warehouse and labelled ‘non-toxic’ and ‘safe for children’.

Faber-Castell initially told a concerned parent its products were thoroughly tested. It has since started a recall. For any importer or retailer, that sequence is the whole story. A global brand’s assurance, and the vetting of two major national chains, did not survive contact with an independent lab.

The rules arrive after the tills close

The new Unsafe Goods (Consumer Sand Products) Notice 2026 was gazetted on 23 October and runs for 18 months while MBIE works out a permanent approach. It requires testing by ISO/IEC 17025-accredited labs, with at least three unit samples per product variant analysed by electron microscopy, and paperwork no older than 12 months before import or manufacture.

But it does not bite until 1 February 2027. Christmas 2026 falls entirely under the old system, where compliance meant trusting the supplier.

The Commerce Commission’s principal compliance adviser Grant McIntosh says the regime will “help prevent sand products containing asbestos from reaching New Zealand shelves and homes”. That is the right goal. The timing is the problem. By the commission’s own count, 23 products had already been recalled covering roughly 45,000 items since the first detections in November 2025.

Christchurch mother Elle Chrisp, who received compensation from Kmart in 2025 after her twins were exposed, put it bluntly: “The urgency is still not there”.

Official advice was borrowed from a supplier

If businesses feel they were given mixed signals, they were. In July it emerged that the public messaging that asbestos fibres were ‘unlikely’ to be released unless the product was mechanically processed had come from a risk assessment commissioned by supplier Educational Colours, not independent work. An AUT study then found fibres could become airborne during ordinary simulated play.

That is a regulatory failure, and the industry is entitled to point it out. It is also a lesson that cuts against the same industry. If a regulator can be caught out relying on a supplier’s homework, so can a buyer.

Retail warned the timeline was unworkable

Retail NZ, which represents nearly 70% of domestic retail turnover, told MBIE in September that “three months is not sufficient time” to arrange testing, especially offshore, and asked for at least six months. More awkwardly, it said it understood no IANZ-accredited lab could yet test to ISO 22262-2:2026, one of the mandated standards. Importers that had been voluntarily testing to Australia’s AS 5370:2024 standard since late 2025 may find that work does not count.

Retail NZ also argued that without a defined limit of detection, enforcement could chase trace findings rather than real risk. That is a fair concern about proportionality. Toy industry groups have similarly argued for auditing suppliers’ due diligence instead of blanket testing.

The science complicates everyone’s position. Ben Alford, director of Auckland lab Aerum, which first identified the Faber-Castell contamination, notes a result only applies to the portion sampled, so no test can clear a whole batch. His conclusion is a blanket ban on coloured play sand, arguing “a lower concentration does not make its presence acceptable”.

Where the bill actually lands

Importing asbestos-containing products is already illegal. Unsafe goods already trigger Consumer Guarantees Act obligations, including refunds and testing and cleaning costs. From February, breaches of the notice carry Fair Trading Act penalties of up to $600,000 for companies and $200,000 for individuals.

The nearer cost is reputational. A recall in December lands when shelf space, staff time and customer goodwill are at their most valuable. Pulling stock, processing refunds and fielding anxious parents in peak season is far more expensive than a lab invoice in October.

The practical response is not complicated. Anything sand-based on the shelf this Christmas, from craft kits to decorative fills, should have independent test documentation from an accredited lab, not a supplier email saying it is fine. Where that paper does not exist, the product is a liability waiting to be named in a Consumer NZ warning.

The regulator moved late and leaned on the wrong evidence. Retailers can fairly complain about that. They cannot outsource the consequences of it. February’s deadline will formalise what the Faber-Castell recall has already made obvious: a safety claim is only as good as the lab that checked it.

Sources

Reader Poll · 5 questions

Do you agree or disagree with the following?

Community

Join the discussion

Add useful context, ask a good question, or challenge an idea — keep it specific and respectful.

Create a commenter account

Enter the name you want shown publicly and your email. We will email you a password-set link; you cannot comment until you use it.

Your email is used for sign-in and account security. It is not published with comments.

Subscribe for weekly news

Subscribe For Weekly News

* indicates required