New Zealand's Leaky Homes Crisis

When it ran. The "Leaky Home Era" has been claimed to start around 1993, with the large-scale use of fibre-cement panels and other "monolithic" (smooth, seamless) claddings on domestic dwellings, and to run to 2004, when the new building code (the revised E2/AS1 compliance document, taking effect in 2005) introduced the drained-cavity requirement for at-risk claddings. Between those dates, tens of thousands of New Zealand homes were built with monolithic cladding fixed directly onto timber framing — with no cavity behind it to let water drain or dry.

How many homes were expected to be affected, versus how many actually claimed. Newspaper and industry reporting at the time predicted a catastrophe of enormous scale:

In the end, just 7,382 claims covering 12,815 properties were ever lodged with the government's Weathertight Homes Resolution Service (the window closed permanently on 31 December 2021), and only 2,300 of those claims were ever recorded as resolved.

The class actions against James Hardie. Three separate cases were filed in 2015 against James Hardie and its NZ subsidiary Studorp over its fibre-cement claddings (Harditex, Monotek, Titan Board). Together they sought hundreds of millions of dollars; not one homeowner recovered a cent:

Across all three cases, the confirmed amount James Hardie paid to homeowners: $0. The confirmed minimum paid to James Hardie by litigants and their funder: $3.35m+.

Did the High Court say James Hardie's products were defective? No. In Cridge v Studorp the High Court held James Hardie did owe homeowners a duty of care — but found Harditex was not flawed in design, was adequately tested, carried independent endorsement, and was fit for purpose when installed and maintained correctly. The water damage was attributed to building and installation errors, not the product. The Court of Appeal upheld that finding in 2024.

Did the High Court say the design of monolithic homes was defective? No. The homeowners argued eight specific design flaws; the High Court — and later the Court of Appeal — rejected all eight. Direct-fixed, no-cavity construction was the then code-accepted standard, and in every one of the eight test houses examined in detail, the failures were traced to construction defects and non-compliance with installation instructions, not to the design itself. (The building code still changed — a drained cavity became mandatory in 2005 — but that was a regulatory response to risk, not a judicial finding that the design was defective.)

Have cavity systems stopped homes from leaking? No. Cavity construction is a genuine improvement — water that gets past the cladding can now drain and dry instead of soaking the framing. But the leak-cause data on this page shows failures concentrate overwhelmingly at junctions, windows, balconies and penetrations rather than the wall field, and a cavity does nothing to fix failed mitre-joint sealants, missing flashings, or degraded balcony membranes. Our field experience says the same: every home leaks eventually, usually where materials meet — regardless of what cladding system or cavity it has. The Canadian survey data on this page confirms it: windows and flashings, not the basic wall assembly, cause the large majority of envelope failures.

Are news stories saying there will be a new "leaky home era"? Yes. Recent reporting (NZ Herald, 2025) carries warnings from leaky-building advocate John Gray and surveyors at Prendos that a new wave of leaky buildings has already begun — in homes and apartments barely five to ten years old. The reasons given echo the 1990s almost exactly: lightweight construction put up at speed, skill shortages in the trades, weak or overwhelmed council inspection (the Herald reported roughly 37% of residential buildings failing their final inspections), and complex multi-unit designs. Current examples in the news include the Kingsland apartments and Auckland's St Lukes Garden Apartments, facing a repair bill of more than $240m across 243 owners.

Did the homes in the class actions have a Certificate of Completion (Code Compliance Certificate) from the council? The public judgments don't record the Code Compliance Certificate status of each individual property, so there is no verified answer to whether all, some, or none had final council sign-off. What is documented is this: in Cridge v Studorp, all eight test houses examined in forensic detail had significant building defects, and none had been built in compliance with James Hardie's own installation instructions — so whatever council sign-off those homes carried, it did not reflect code-compliant construction. More broadly, it is well established that many leaky-era homes did receive CCCs: territorial councils ended up bearing roughly 25% of the Financial Assistance Package repair costs precisely because they had signed off buildings that later leaked.

Every figure above is sourced in full in the sections below.

James Hardie litigation — sued for hundreds of millions, paid out $0

Predicted scale vs. what MBIE actually recorded

MBIE's published statistics (year-end 2020, out of 7,382 total claims lodged) split claims into three buckets: Resolved (2,300 — 31%), Closed (4,636 — 63%), and Open (446 — 6%). "Closed" is a catch-all covering claims that were discontinued, found ineligible, withdrawn, not progressing, transferred, or ended by a change in property ownership — MBIE's public data doesn't break out exactly how many of those 4,636 were specifically withdrawn versus closed for another reason.

What the government set aside for repairs, versus what it actually paid

Where the leaks actually happen (ranking exactly as recorded in the source data, unmodified)

Leak-cause percentages are calculated from the approximate BRANZ/Prendos chart values used elsewhere on this page (total pool ≈1,424 recorded leak instances across 18 defect types) — see "What specific defects cause the most leaks?" below for the full ranking.

There are really two separate "discoveries" here — one on the ground, years before anyone official was listening, and one that made it official.

On the ground: Building surveyor Philip O'Sullivan, of Auckland firm Prendos, is widely credited as the person who first raised the alarm. Cutting into buildings during routine inspections in the 1990s, O'Sullivan was shocked by the extent of rot he found behind seemingly sound cladding. He warned about it publicly for years afterward through "Dr Rot" columns in the trade press, but for a long time few in the industry wanted to listen — in 2002, when Prendos went public with the scale of the problem, a master builder publicly accused the firm of exaggerating both the leak problem and its likely repair costs.

Officially: The Government's response came later that same year. In February 2002, the Building Industry Authority appointed a Weathertightness Overview Group, chaired by former State Services Commissioner Don Hunn (with co-author David Kernohan), to investigate. Their findings — published 31 August 2002 as the "Report on Weathertightness of Buildings to the Building Industry Authority," universally known as the Hunn Report — became the first official confirmation that the problem was systemic rather than a series of isolated incidents. It made 25 recommendations and directly led to the Weathertight Homes Resolution Services Act later that year, which established the formal claims process.

In 2009, the Government commissioned PricewaterhouseCoopers (PwC) to produce a report titled "Weathertightness – Estimating the Cost." Its job was to re-estimate the scale of the leaky homes problem for dwellings built between 1992 (when the Building Act 1991 took effect) and July 2008, including how many homes were affected, how many had been repaired, and who was bearing the costs.

The 42,000 figure was a "consensus forecast" within a wider estimated range of 22,000 to 89,000 homes. That range reflected genuine uncertainty in the data — PwC itself acknowledged the true number could be significantly higher. The 11% figure comes from applying that consensus forecast against the roughly 401,000 dwellings built nationwide during the 1992–2008 period.

PwC put the total economic cost of remediating all 42,000 affected dwellings at $11.3 billion (in 2008 dollars), including repair costs, legal costs, transaction costs, and the cost of running the Weathertight Homes Resolution Service (WHRS).

Owners carried the largest share mainly because they paid their own transaction costs, and because failures discovered after the 10-year legal limitation period became solely the owner's responsibility.

No — building industry experts, including the Home Owners and Buyers Association and the Institute of Building Surveyors, argued the Government had underestimated the problem. Their view: up to 89,000 homes could fail, at a cost of almost $23 billion. PwC's report labelled this the "extreme view" and considered it likely an overstatement at the time.

PwC's report included survey data on when owners first noticed evidence of a leak, measured in years since the home was built or renovated. Failures weren't concentrated in one narrow window — they were spread fairly evenly across years 1 through 10, with each individual year accounting for roughly 7% to 13% of all reported failures.

The single riskiest year was year 8 after construction, when about 13% of all reported failures first became evident — the tallest bar in the survey. Year 9 eased back to around 7%, year 10 ticked up slightly to about 8%, and then the rate fell away sharply: by year 11 it had dropped to under 4%, roughly half of where it sat at years 8–10, and from year 12 onward each year accounted for 1% or less, tapering to almost nothing by year 14–15.

Source: PwC, "Weathertightness – Estimating the Cost" (2009), Figure 20, based on the Weathertight Claimant and Respondent Survey 2009. Bar values are read directly from the published chart.

Monolithic cladding is an exterior wall finish designed to look like one continuous, seamless surface — no visible boards, laps, joints, or overlapping sheets, just a smooth, uninterrupted plane. It was popular in New Zealand through the 1990s and early 2000s because it suited the "Mediterranean style" architecture in fashion at the time: flat or low-pitched roofs, minimal eaves, plastered walls, and clean, modern lines.

There were three main types used in NZ, all of which PwC's report specifically named as high-risk:

Solid plaster/stucco — a cement-based plaster applied directly over a rigid backing sheet, finished smooth or textured.
EIFS (Exterior Insulation and Finish System) — a polystyrene foam board fixed to the framing, then coated with a thin, flexible plaster-like finish over a reinforcing mesh. Lightweight and good for insulation, but the foam itself has almost no structural rigidity and no ability to let trapped water escape.
Flush-finished fibre-cement — fibre-cement sheets butted together and finished with jointing compound so the seams disappear, rather than the sheets being overlapped or capped.

The critical difference from traditional claddings like weatherboard or brick veneer: those systems expect some water to get past the outer surface, and they're built with overlaps, gaps, and — especially with brick — a drained and vented cavity so any water that gets in can run down and out, or dry out before it does damage. Monolithic cladding, as it was built in New Zealand during that era, was typically fixed directly onto the timber framing with no cavity at all.

This "direct fix" method — building the cladding straight onto the framing with no drainage cavity behind it — was the standard, code-accepted construction practice across New Zealand until the Building Act 2004 and the revised E2/AS1 compliance document took effect in 2005, at which point a drained cavity became mandatory for at-risk claddings going forward.

MBIE's Weathertight Homes Resolution Service (WHRS) — the official claims body — recorded 7,382 claims covering 12,815 properties by the time claim numbers had essentially flatlined (late 2019). The window to lodge a new claim closed permanently on 31 December 2021.

That's a small fraction of the 42,000 "consensus" homes PwC identified — let alone the 89,000–174,000 some experts believed were affected.

The 12,815 properties with a formal claim represent about 30% of PwC's own 42,000 consensus figure, roughly 14% of PwC's 89,000 "extreme view," and just over 7% of the 174,000 some independent experts believed were affected.

MBIE's public monthly claims statistics don't actually publish this breakdown. Their reports only split claims by status (closed / open / resolved) and by council area — the only reference to dwelling type is a footnote explaining that a multi-unit complex is allowed to be lodged as a single claim.

The closest real figures come from PwC's original 2009 report, but only for a small sample used in risk modelling: of an 857-claim sample assessed at the time, 625 were single-unit failures and 228 were multi-unit — roughly a 73% / 27% split. That was based on claims lodged up to 2009 (3,967 at the time), not the final ~7,385.

For the final claim count, we can estimate using a different route: the final ratio of 12,822 properties across 7,385 claims (1.74 properties per claim on average), combined with FAP data showing multi-unit claims covered roughly 31 units each on average (from the $3.2 million average multi-unit repair cost divided by the $101,912 per-unit cost).

The striking part: multi-unit claims were a small minority by count, but because each one could represent dozens of individual apartments or townhouses, they're estimated to account for something like 40–45% of all properties covered by every claim ever lodged.

This is a derived estimate built from a proxy ratio in the FAP dataset, not a figure MBIE has directly published. A precise breakdown would require an Official Information Act request to MBIE.

Launched in May 2010, the Financial Assistance Package (FAP) split agreed repair costs 25% government / 25% council / 50% owner. Officials predicted 23,500 eligible households, with 70% uptake — about 16,450 homes fixed — at a cost to taxpayers of roughly $1 billion over five years.

By the time the FAP closed in 2016, total spending across all parties was $134.8 million: government $33.7 million, councils $13.9 million, and owners $87.2 million. On every measure — homes helped, and money spent — actual results came in at roughly a tenth of what was originally forecast.

Two different "predictions" were made at two different points, and it's worth keeping them separate:

PwC's 2009 prediction (scale of the whole problem): 42,000 homes affected, $11.3 billion total economic cost — most of it (69%) expected to be paid by owners directly, outside any government program.
Officials' 2010 prediction (for the Financial Assistance Package specifically): 23,500 eligible households, 16,450 expected to be fixed (70% uptake), costing taxpayers around $1 billion over five years.

What actually happened, as far as it can be measured by 2025:

On top of that, councils have kept paying out on claims lodged before the 2021 deadline — Auckland Council alone disclosed roughly $315 million in weathertightness settlements between 2018 and 2023, and budgeted over $260 million a year for further claims into 2024–25.

There is no single official "total spent nationally by 2025" figure, because most of PwC's $11.3 billion estimate was private owner spending that was never centrally tracked. The chart below compares what was predicted against the parts of the picture that are publicly measurable — the formal government/council programs.

Yes. Even though the WHRS claims window closed in 2021, councils are still paying out on claims lodged before the deadline. Auckland Council alone paid roughly $315 million in weathertightness-related settlements between 2018 and 2023, and was budgeting more than $260 million a year for further claims in 2024–25. Industry commentary through 2024–2026 has also raised concerns about newer buildings showing similar defects, suggesting some of the underlying causes — weak inspection, skill gaps, unfamiliar materials — were never fully resolved.

No. The deadline to lodge a new WHRS claim with MBIE expired on 31 December 2021. Separately, the 10-year "longstop" limitation period under the Building Act 2004 has now run out for almost all homes built during the affected era (roughly 1988–2004), closing off most legal avenues even where problems are only now surfacing. Homeowners with an already-open claim can still be supported through to resolution.

Yes — three separate class actions were brought against James Hardie and its NZ subsidiary, Studorp Limited, over its fibre-cement cladding products: Harditex, Monotek, and Titan Board, all monolithic-style claddings installed on New Zealand homes roughly between 1983 and 2010.

The "White" litigation (Harditex) — filed 2015, collapsed mid-trial in 2021. Filed by lead claimant Karen White and originally covering claims over Harditex, Monotek and Titan Board cladding on nearly 400 buildings, this was described at the time as "a David and Goliath battle … one of the largest class action lawsuits in New Zealand history." Technically it wasn't a class action at all — every one of the plaintiffs was individually named, rather than represented by a few people on behalf of an unnamed group, a distinction that turned out to matter a great deal (see below). In early 2021 the Titan and Monotek claims were dropped, leaving 376 Harditex-clad buildings and roughly $230 million in claimed damages. The case was funded throughout by a UK litigation funder, Harbour Fund II, which had spent $17.1 million on it by mid-2021.

The Stage 1 trial began 17 May 2021 in the Auckland High Court, set down for 16 weeks. Six weeks in, lead trial counsel Simon Hughes QC told the legal team the case "was not going well": a key expert witness, whose evidence was meant to help prove systemic defects across a sample of properties, had — under cross-examination — effectively assumed the very thing he was supposed to prove, rather than establishing that the water damage was caused by defective cladding rather than poor building work. His written advice on 7 July 2021 rated the plaintiffs' prospects as now "significantly below 50/50." Within days, Harbour told the plaintiffs' lawyers it wanted out.

A settlement was struck: the plaintiffs would drop all claims and receive no compensation, and in exchange Harbour (the funder) would pay James Hardie NZ$1.25 million. That's the unusual detail — money flowing toward the defendant — and the QC opinion explains why: because these plaintiffs were all individually named (not protected class members), they were personally exposed to an adverse-costs bill from James Hardie estimated at $3-4 million if the case were simply abandoned. Paying $1.25 million capped that risk at a much smaller, known amount, funded by Harbour rather than the homeowners themselves.

Because some plaintiffs objected to settling for nothing, the funding agreement required an independent King's/Queen's Counsel review of whether the deal was "reasonable in all the circumstances." Philip Skelton QC took on that role, inviting all ~400 plaintiffs to email their views. Over 100 responded — a very high response rate — splitting roughly 45 in favour, 35 opposed, and 25 neutral. Many opposing plaintiffs argued Harditex must have been defective (since James Hardie had discontinued it) or wanted to wait for the outcome of the separate Cridge case. Skelton QC ultimately found the settlement reasonable — "not without hesitation" — calling it "a disastrous result" for both the plaintiffs and the funder, but concluded there was no better realistic alternative once Harbour pulled its funding. Along the way, he also caught and fixed two drafting problems that would otherwise have unfairly stripped some plaintiffs of rights: the settlement was narrowed so that owners with Titan or Monotek cladding (whose claims had been dropped from this case) kept the right to sue James Hardie separately, and so that claims over other Harditex-clad properties not part of this specific case (raised by Harvey Norman Properties, among others) were also preserved.

The contrast with how this case was originally pitched is striking. When recruiting plaintiffs, Adina Thorn Lawyers described the claim's strength as "distilled from more than 1,600 owners and numerous body corporates who have registered with us," having "exceeded the NZ$100 million claim target," with the backing of "one of the world's leading funders of such actions." By the time it reached trial, that had narrowed to 376 properties and roughly $230 million sought — and it ended with the litigation funder paying James Hardie money, not the other way around.

The "Waitakere" litigation (Harditex, Monotek and Titan Board) — settled just before trial, April 2023. Scheduled for a 32-week trial starting 8 May 2023 in the Auckland High Court, this case never got there. On 25 April 2023, James Hardie announced it had ended via a negotiated commercial agreement, discontinuing the litigation entirely. Terms were kept confidential — again, no admission of liability from either side.

Cridge v Studorp Limited (Harditex) — the one that actually went to judgment. Lead plaintiffs Tracey Cridge and Mark Unwin, representing 144 Harditex-clad homeowners, sued over a Wellington home with widespread internal water damage. The High Court ruled against the homeowners in August 2021 — but on causation, not duty of care: the Court actually found that James Hardie did owe homeowners a duty of care, it just held the homeowners hadn't proved Harditex was the cause of their water damage. On appeal, the Court of Appeal (2024) dismissed the homeowners' case again on the same causation grounds — but confirmed, as an appellate court after a full trial, that a cladding manufacturer does owe homeowners a duty of care in negligence. James Hardie tried to appeal that specific finding to the Supreme Court, which declined to hear it. So the precedent (duty of care exists) now stands for future cases — but on these particular facts, James Hardie was found not to have caused the damage, and the 144 homeowners lost.

Across all three cases, James Hardie never lost on the substance. Two settled on confidential, liability-free terms before reaching a verdict, and the one case that ran to full judgment and appeal ended in James Hardie's favour — though it left behind a legal precedent (manufacturers owe homeowners a duty of care) that could matter for future cladding-defect claims against other manufacturers.

The High Court issued an official media release alongside its judgment in Cridge and Unwin v Studorp Limited; Fowler and Woodhead v Studorp Limited and James Hardie New Zealand Limited [2021] NZHC 2077, embargoed until 9am Thursday 12 August 2021. It's a plain-language summary written by the Court itself to explain the ruling — not part of the legal reasoning, but a useful window into how the judges themselves framed the decision.

Background, in the Court's own words: "From 1987–2005 James Hardie manufactured and sold a fibre-cement wall cladding called Harditex. Its principal use was on residential homes. In this proceeding, 144 owners of homes clad in Harditex, and which have suffered watertightness-related damage, sued James Hardie." Harditex was one of the earlier fibre-cement products to go asbestos-free, replacing asbestos with cellulose fibres from wood.

What the homeowners argued: that Harditex was insufficiently tested before release and didn't meet durability standards; that James Hardie's claim the sheet wouldn't rot was wrong, and its moisture absorbency was an inherent flaw; that James Hardie gave builders inadequate installation guidance for a genuinely new product, and that the sheet was too difficult for an average builder to install correctly; and that by the year 2000 James Hardie knew or should have known about these flaws and had a duty to warn the public.

What the Court examined: whether Harditex met accepted moisture management standards, its durability (including resistance to microbiological attack), its ability to cope with building movement once installed, its susceptibility to mould, and the various complaints about installation. Both sides called expert witnesses from New Zealand and overseas — the Court said it "preferred by a significant margin" the evidence led by James Hardie, describing many of its experts as world leaders in their field. The Court also weighed physical test walls built specifically for the case, and detailed water-damage evidence from eight of the claimants' own houses.

The High Court's conclusions: the Court held that James Hardie did owe the homeowners a duty of care — but ruled the homeowners had not proved the Harditex product actually caused their weathertightness problems. Assessed against established building science, the product was found not to be flawed in design, to have been adequately tested, and to carry independent endorsement. James Hardie's technical guidance was found sufficient and correctly pitched at a reasonably competent builder, and Harditex was not shown to be materially harder to install than the sheet products that came before it.

On the eight houses examined in detail, the Court repeatedly found fundamental building errors and non-compliance with James Hardie's installation instructions — errors serious enough that the homeowners couldn't isolate a product flaw as an independent cause of the damage. Both the negligence claims and a related set of Fair Trading Act claims were dismissed on that basis.

The document the Court released is titled a "media release," not a press release, and it's dated 2021, not 2020 — the case is otherwise an exact match for James Hardie's Harditex litigation. Source: High Court of New Zealand media release, 12 August 2021, contact Liz Kennedy, Senior Judicial Communications Advisor.

The homeowners appealed the High Court's 2021 loss. On 26 September 2024, the Court of Appeal handed down its judgment — Cridge v Studorp Limited [2024] NZCA 483 — and dismissed the appeal outright: "The appeal is dismissed." The 144 homeowners lost for a second time.

On duty of care — a homeowner win, but not a decisive one. The Court confirmed this was "the first time that a New Zealand court has held following trial that the manufacturer of a building product intended to be a key component in a building owes a tortious duty of care to the building's owner." It rested this on a 2016 Supreme Court case (Carter Holt Harvey Ltd v Minister of Education) that had only found such a duty "arguable" at a preliminary strike-out stage — Cridge was the first time it was actually tested and confirmed at a full trial, then upheld on appeal. James Hardie tried to get this finding overturned too; the Court of Appeal rejected that argument and kept the duty of care intact.

But on the facts, the homeowners still lost. The Court's own summary of its reasoning: its assessment of the expert evidence confirmed the trial judge's view that James Hardie's experts had "greater expertise" and gave more measured, less partisan evidence than several of the homeowners' experts. Critically, none of the eight test properties had been built in compliance with James Hardie's own installation instructions, and all had significant building defects — meaning they didn't provide a fair test of whether Harditex itself was flawed. The Court reasoned that if any class member's home had suffered water damage despite being built correctly to James Hardie's instructions and the building code, that property would have been the one selected for detailed testing — and none was.

What the judgment reveals about the Harditex "monolithic" design. The judgment describes Harditex in detail as a direct-fixed panel cladding system — rectangular fibre-cement sheets (7.5mm, later up to 9mm thick) nailed straight onto the timber frame with dozens of nails, with only a breathable building wrap behind them and no drained cavity. Joints between sheets were sealed using a PVC "h-mould" flashing, then the whole surface was coated to produce the smooth, seamless "monolithic" look, with moulded polystyrene trim added around windows, arches, cornices and columns for architectural detailing. The homeowners argued this design was inherently flawed in eight specific ways — moisture absorbency, water entry at joints and penetrations, poor drainage and drying, inability to handle normal building movement, lack of durability, and inadequate installation instructions. The Court rejected all eight claims, finding Harditex was "fit for purpose" and worked as intended when built and maintained correctly — the water damage in the test properties was attributed to construction defects and poor texture coating, not the cladding design itself.

Net effect: the 2024 ruling is a mixed result. It's a genuine legal landmark — cladding manufacturers in New Zealand now owe homeowners a confirmed duty of care, which matters for future cases against other manufacturers. But for the 144 Harditex homeowners themselves, and for Harditex as a product, the courts have now twice found the direct-fix monolithic system itself was not to blame — poor installation was. Source: Cridge v Studorp Limited [2024] NZCA 483, Court of Appeal of New Zealand, judgment of French, Brown and Gilbert JJ.

All three cases were filed in 2015. All three took somewhere between six and ten years to resolve. And in every single one, James Hardie did not lose — and came out financially ahead, not behind, despite being sued for hundreds of millions of dollars combined.

White v James Hardie — filed 2015, sought roughly $230 million on behalf of up to ~400 Harditex-clad properties. Collapsed in the sixth week of a 16-week trial in mid-2021 after a key expert witness's evidence didn't hold up under cross-examination. Settled on "drop hands" terms: the homeowners received nothing, and their litigation funder, Harbour, paid James Hardie NZ$1.25 million — a payment the independent QC review explained was there to cap the homeowners' exposure to an estimated $3-4 million adverse-costs bill they'd otherwise have faced personally.

Cridge v Studorp — filed 2015, brought by 144 homeowners seeking repair costs, special damages, general damages, post-remediation damages and expert costs (the total amount claimed was never made public). Lost at the High Court in 2021, lost again at the Court of Appeal in 2024, and James Hardie's own bid to challenge the duty-of-care finding at the Supreme Court was abandoned in March 2025 — meaning the whole ten-year process ended exactly where the 2021 judgment left it. In September 2022, James Hardie was awarded $2.1 million in costs against the homeowners, on top of them recovering nothing.

The Waitakere litigation — filed 2015, covering Harditex, Monotek and Titan Board claims. Scheduled for a 32-week trial in May 2023, it settled just before trial via a confidential negotiated agreement, with no admission of liability by either side. Because the terms were never disclosed, it's not possible to say whether any money changed hands here or in which direction — but going by the pattern of the other two cases, there's no public record of James Hardie paying any homeowner compensation.

Add it up and the picture is stark: across three separate cases that together took the better part of a decade to run, and that collectively sought several hundred million dollars, not one homeowner recovered a cent from James Hardie. Meanwhile, at least $3.35 million moved in the opposite direction — from the litigants and their funders, to James Hardie — before accounting for whatever the confidential Waitakere settlement may have added to that total. Years of litigation, an enormous legal spend on both sides, and the manufacturer finished not just undefeated, but paid.

Yes — a separate representative claim, run by the same law firm behind the James Hardie cases, against a different manufacturer and a different product. It followed a similar arc: years of preparation, a funder pulling out, and no compensation to homeowners.

Paine v Carter Holt Harvey Ltd was a representative proceeding brought by Adina Thorn Lawyers (ATL) — the same firm behind the White v James Hardie Harditex litigation — over "Shadowclad," a textured plywood cladding made by Carter Holt Harvey, alleged to fail when installed without a drainage cavity behind it. It was funded by Harbour Fund III, part of the same UK litigation-funding group (Harbour Litigation Funding) that funded the James Hardie case, though a different fund within it.

In July 2019, Carter Holt tried to get the case thrown out as an abuse of process, arguing ATL had made misleading public statements about the claim's size and needed the Court's permission before filing. The High Court largely rejected that application ([2019] NZHC 1614) — it found the plaintiffs didn't need prior permission because everyone represented had individually consented, and most of the disputed statements weren't misleading. The judge did find that one 2017 comment overstated the claim's value, and a later ruling noted a "whiff of overstatement" in a comparison Thorn drew to a separate, much larger Ministry of Education claim against Carter Holt over school cladding ($1.1–1.3 billion, settled confidentially, and not part of this homeowner case).

Having survived that challenge, the case was heading toward a four-month trial — then, in June 2021, the funder unexpectedly withdrew from the arrangement. With no funder, the claim was discontinued. Carter Holt made no concessions, no compensation was paid to homeowners, and Carter Holt was awarded more than half a million dollars in security-for-costs money that had been paid into court. Three years of litigation, one failed strike-out attempt by the defendant, and the case still ended with homeowners recovering nothing.

The Shadowclad case also independently confirms a detail from the James Hardie litigation covered above: a building surveyor named John Dalton appears throughout the Paine judgment as ATL's expert witness, inspecting properties and giving evidence on the alleged defect. He's the same expert whose name surfaces in connection with the White v James Hardie case — suggesting ATL relied on him across more than one representative cladding claim, not just one.

Combining the broad water-source categories with the detailed breakdowns for Wall, Capillary, and Other leak sources — all data attributed to BRANZ via a Prendos Limited presentation:

Balcony Leaks and Window installation stand out as the two largest single causes, together accounting for over 40% of everything on this list. Note that Balcony Leaks and Roof Leaks are broader top-level categories (no further sub-breakdown was available in the source material), while everything from rank 4 down is a specific, granular defect type.

Figures are approximate, read from bar-chart heights in the source material (no exact data labels were available). We could not verify the original BRANZ publication or dataset behind these numbers — they appear consistent with material Prendos, a leaky-building expert-witness firm, has used in industry presentations.

Everything above this point in the FAQ is drawn from court judgments, government reports, and published industry data. This section is different — it's our own professional opinion, built from more than 2,500 monolithic home inspections and 35 years working in the insurance industry, not from a published source. It's worth reading alongside the research, but it should be read as field experience and judgment, not as a cited fact.

All homes leak, eventually. In our experience, every home leaks at some point over its lifetime, usually because a material wears out or ages, rather than because of one dramatic failure. Since 2009 we've used thermal imaging as a matter of routine, testing every window and door on every inspection. Using that method, we can find evidence of moisture entry around windows in the majority of dwellings that face into the prevailing wind — and that holds true regardless of cladding type, not just monolithic homes.

Why monolithic cladding gets the blame. We think there are three reasons monolithic-clad homes turn up more detected moisture than older homes, and none of them are really about the cladding material itself. First, people look harder for it — the reputation of the product means inspectors, buyers and lawyers specifically go looking for moisture in a monolithic home, in a way they never did with older cladding types. Second, the sensors are better — thermal imaging technology available today is far more sensitive than what was used to inspect homes in the 1990s and early 2000s, so more genuine moisture is being detected now than would have been found at the time. Third, there's a financial incentive to find it — litigation, insurance claims and property transactions all create a reason to look for and document moisture that might once have gone unnoticed.

Before monolithic cladding became common, most New Zealand homes were clad in brick, concrete or timber weatherboard — materials that absorb moisture rather than shed it. That absorbency masked a lot of what was actually happening behind the cladding: water got in, sat in a material that could hold it, and often dried out again before it caused visible damage or was ever detected. Monolithic cladding changed that. It's a hard, largely non-absorbent surface, so once water finds a way in, there's nothing left to hide it or soak it up. It shows up faster and more visibly — which is part of why this one cladding type became so strongly associated with the crisis.

The mitre joints matter more than people think. Aluminium window frames expand and contract daily with the weather. The physics holds up: aluminium's coefficient of thermal expansion is about 23 micrometers per metre per °C, and dark or mid-tone anodised or powder-coated frames commonly run well above ambient air temperature in direct sun. A 25-30°C daily swing — from around 35-40°C in direct sun down to 15°C overnight — is entirely plausible for New Zealand conditions, and drives real, repeated dimensional movement in the frame. The failure point isn't the aluminium extrusion itself flexing — it's the mitre joints, the corners where two frame lengths meet. Corners are typically mechanically fixed (crimped, screwed, or corner-keyed) with a sealant filling the joint. Repeated expansion-contraction cycling stresses that seal over years; sealants have a rated movement capability, and once cycling exceeds it, you get adhesive failure (the sealant pulling away from the frame) or cohesive failure (the sealant tearing internally). Either opens a capillary path exactly where you'd expect: at the corner, which is also usually near a critical junction with the cladding.

Wind adds a second, separate mechanism. Ordinary rain relies on gravity — water runs down a window and off the sill, and a well-designed seal only has to stop it pooling or seeping in under its own weight. Wind-driven rain is different: high wind pushes water sideways and even upward against the face of a window, forcing it into gaps and joints under real pressure rather than just resting against them. That pressure can drive moisture past seals that would comfortably handle still-air rain. Once it's in, that water usually doesn't find its way back out through the window's own drainage system — weep holes are designed for gravity drainage, not to expel water that's been forced in under wind pressure. Instead it tends to track sideways or downward inside the frame and exit into the building envelope — the wall cavity and framing behind the cladding — which is exactly where you don't want it.

At what wind speed does this actually start? Under NZS 4211, a new window's water penetration rating is tested at 30% of its rated positive design wind pressure, held for 15 minutes of controlled water spray. Converting typical test pressures for low-to-high wind zone joinery (roughly 100-200 Pa) into an equivalent wind speed, using the standard dynamic-pressure relationship, works out to somewhere around 45-65 km/h — a strong breeze to near-gale on the Beaufort scale, not an extreme event. That's the threshold for a brand-new window with intact seals. It is not the threshold that matters most in the field. Once a mitre joint's sealant has fatigued from years of thermal cycling, as described above, the effective water-penetration threshold for that joint drops — sometimes well below the window's original as-tested rating. In our experience, aged single-glazed aluminium joinery on a wall facing the prevailing wind can show evidence of moisture entry in conditions that wouldn't have troubled it when the seals were new. This is a general engineering approximation, not a lab measurement of any specific window — actual thresholds vary with orientation, gusting, rain intensity, and the individual joint's condition.

How long does that seal actually last? We've seen windows fail to perform in as little as 5 years on exposed hill sites in Wellington — some of the harshest wind-loading conditions in the country. Our own sense, from the properties we've inspected, is that 8-12 years is a typical performance span for window and door seals in high-wind, marine coastal environments generally, well short of the building's overall design life. That's a seal-and-joint problem, not a materials problem: the aluminium extrusion itself, left alone, could plausibly last 100 years or more. It's the sealant at the mitre joints, and the compression gaskets around the glazing, that wear out from constant thermal cycling and salt-laden coastal wind long before the aluminium frame does. A window can be structurally sound and completely weathertight-failed at the same time.

What we call, in the insurance industry, a leaky window or a leaky door, gets called a leaky home once it reaches a real estate listing or a court filing. The underlying physical problem — a joint, a seal or a flashing detail failing under thermal or wind stress — is often small and localised. It's the label, and the financial and legal consequences that come with it, that scale up to "leaky home."

Like the rest of this section, this is practical field guidance rather than a cited standard — a starting checklist, not a substitute for a proper inspection.

A standard pre-purchase building report usually isn't enough on its own. Most are a visual, non-invasive walk-through — exactly the kind of inspection that misses a problem hidden behind an intact-looking cladding surface. On a monolithic-clad home, especially one built before the mid-2000s, that's the home's defining risk, so a generic building report is checking the wrong thing. A specialist weathertightness inspection — thermal imaging plus moisture meter follow-up — is a different, more targeted exercise, and worth the extra cost on any monolithic home, particularly if it's on an exposed hill or coastal site, or is already past the 8-12 year seal-performance span mentioned above.

Visual red flags anyone can check, no equipment needed: cracking or gapping in the sealant along window and door mitre joints; staining, discolouration or a slightly different paint texture on the cladding below windows, balcony edges, and penetrations (pipes, meter boxes); ponding water or debris buildup on balcony or deck surfaces instead of a clear fall to a drain; cladding that sits hard against paving, decking, or garden beds with no visible gap (ground clearance is meant to stop capillary action drawing water up into the cladding); and a soft, spongy, or slightly flexible feel when you press on cladding near ground level or below a window.

What thermal imaging adds. Moisture sitting inside a wall cavity changes how that section of wall holds and releases heat, showing up as a temperature anomaly on a thermal camera that's invisible to the naked eye. It works best after a run of wet weather, when there's a real temperature differential to detect, and it takes a trained eye to tell a genuine moisture signature apart from other causes of a thermal anomaly — a gap in the insulation, or framing timber sitting behind the wall, can look similar on camera to someone without the experience to tell them apart. A thermal scan flags where to look closer; a moisture meter (and, where warranted, a small invasive test) is what actually confirms it.

Priority areas, in order of what we'd check first: window and door junctions on the elevation facing the prevailing wind, balcony membranes and the junction where the balustrade or door threshold meets the deck, ground-level cladding clearance all the way around, and any pipe, vent, or fixing penetration through the cladding face. Those four line up closely with the top causes in the leak-cause table above — balconies, window installation, and penetrations account for most of what actually goes wrong.

For existing owners, the practical takeaway is maintenance, not panic: resealing mitre joints and glazing gaskets on a roughly 8-12 year cycle in high-wind or coastal conditions is cheap, routine upkeep. Treated that way, it's a maintenance line item. Left unaddressed until it shows up as visible damage or a musty smell, it becomes a much larger and more expensive repair.

Even the Court of Appeal, in Cridge v Studorp [2024] NZCA 483, described the underlying cladding system in mechanical terms that matter here: a "direct-fixed panel cladding system" — sheets "nailed straight onto the timber frame … with only a breathable building wrap behind them and no drained cavity," the joints "sealed using a PVC 'h-mould' flashing," then coated over "to produce the smooth, seamless 'monolithic' look." That's the judge's own description, not ours — and it's the whole reason detection matters so much here. With no cavity to drain into and no absorbent surface to mask it, any moisture that gets past the cladding has nowhere to go but into the framing, silently, until someone specifically looks for it. That makes the quality of the looking — the equipment and the person using it — a bigger factor in the outcome than most homeowners realise.

Handheld sensors are indicative, not absolute. The two tools in routine use each have real limitations. Non-invasive moisture meters read an electromagnetic field through the surface and are sensitive to depth, material density, and what's actually behind the wall — a metal fixing, a stud, or a patch of denser material can all produce a reading that looks like moisture but isn't. Invasive pin-type meters measure conductivity directly between two pins, which is more reliable, but the result still depends on pin depth, contact pressure, and calibration to the specific material being tested. Thermal cameras are arguably the most operator-dependent of all: the image depends on recent weather (a wall that's been in full sun an hour earlier can throw off a reading), the angle and distance the camera is used at, and — most of all — the experience to tell a genuine moisture signature apart from a stud, an insulation gap, or ordinary thermal bridging, all of which can look similar on screen to an untrained eye. None of these tools are lab instruments. Used well, they're a way of narrowing down where to look. Used carelessly, or read by someone without the experience to interpret them, they can just as easily produce a false alarm as a real finding.

How the sensor is held matters as much as which sensor is used. With a pin-type moisture meter, the angle of insertion, how hard the pins are pressed in, and whether they're driven with or against the grain of the timber can all change the reading — the same damp stud can read dry or wet depending on technique alone. With a non-invasive meter, holding the sensor face flat and fully in contact with the surface gives a different result than holding it at a slight angle or with light, inconsistent pressure, because the field it reads through the material changes with contact quality. Thermal cameras have the same issue at a distance: shooting a wall square-on gives an accurate surface temperature, but shooting it at an oblique angle picks up reflected heat from the surroundings rather than the wall's own temperature, which can manufacture an anomaly that isn't really there, or mask one that is. None of this is exotic — it's basic technique — but it means two competent-looking readings, taken thirty seconds apart by two different people at the same spot, can genuinely disagree. That's exactly why a single reading shouldn't be treated as the final word.

A good report shows its working. Because technique affects the result this much, the report itself should show, for each place tested, what was actually done to get that reading — which tool was used, whether it was a surface or invasive test, and ideally a photo of the sensor in contact with that specific spot, not just a plan marked with a dot and a pass/fail symbol. That level of detail lets a second person check whether the test was done properly, rather than just trusting the conclusion. If a report can't show how a given point was tested, there's no way to tell whether an isolated "wet" reading reflects real moisture in the wall, or just how the sensor happened to be held that day.

Liability changes behaviour, and not always in the homeowner's favour. Inspectors now operate in an environment shaped by exactly the litigation history covered elsewhere in this FAQ — building professionals who miss a defect and get it wrong can face real professional and financial consequences, sometimes years later. That creates an asymmetric incentive: flagging an ambiguous reading as a possible problem costs the inspector very little, while clearing a home that later turns out to have an issue can cost them a great deal. In our experience, that pushes some assessments toward caution by default — recommending invasive testing, or reporting "moisture present" off a single ambiguous reading, not because the evidence clearly supports it, but because it's the lower-risk call for the person writing the report. That's a completely understandable response to the environment inspectors work in. It doesn't mean it always produces an accurate picture of the actual house.

What actually gets you a fair, accurate assessment. A few things matter more than the brand of equipment used. First, ask for the underlying readings and photos, not just a pass/fail conclusion — a transparent report can be sanity-checked; a verbal "there's moisture here" can't. Second, ask whether the inspector compared a suspect reading against a known-dry reference point elsewhere on the same wall, on the same day, in the same conditions — a single absolute number means far less than a comparison against a same-building baseline. Third, treat a single sensor hit as the start of an investigation, not the end of one: a genuine finding should be corroborated by more than one method (thermal plus moisture meter, and invasive testing where the finding is significant enough to justify it) before it's treated as confirmed. And finally, weigh any finding against context — a home's age, orientation, and exposure. A moisture reading near a window on a 15-year-old, west-facing home in a high-wind location is consistent with the ordinary seal-wear timeline described above, and may be a manageable maintenance item rather than a structural emergency. The same reading on a 3-year-old home in a sheltered location is a different and more serious conversation.

None of this means moisture findings should be dismissed — plenty are real, and the leak-cause data elsewhere in this FAQ shows how common they are. It means a finding is only as good as the process behind it: the equipment, the operator's skill, and an honest read of what the incentives in the room actually are.

Who is paying the inspector changes what gets reported, and it can pull in opposite directions depending on which side of the sale that person is on. None of what follows is an accusation against any individual — it's the kind of incentive analysis a fraud investigator or forensic accountant would apply as a matter of routine, asking who benefits financially from a given outcome, before asking whether that benefit actually shaped the result in any specific case.

The incentive to talk a vendor's price down. A buyer commissioning a pre-purchase inspection has an obvious use for a negative finding: leverage. A moisture reading — real, marginal, or ambiguous — can be the difference between paying full asking price and negotiating a meaningful discount, adding conditions, or walking away entirely. That creates a financial incentive, for the buyer and by extension for an inspector who wants repeat business from buyers or the agents who refer them, to report ambiguous findings in the most negative available light. It doesn't require dishonesty to work: an inspector inclined to flag every marginal reading as "moisture present, recommend further invasive investigation" produces exactly the report a buyer negotiating on price would want to receive.

The incentive to go light and help a sale along. The same pressure runs the other way for a vendor. A vendor who commissions their own pre-sale report, or steers a prospective buyer toward a particular inspector, benefits from a clean result — a negative finding can cost tens of thousands off the sale price, delay settlement, or kill the deal. Real estate agents, paid on a percentage of the final sale price, share that same incentive: a report that surfaces a serious moisture problem is bad for everyone getting paid on the transaction closing at asking price. Again, none of this requires bad faith. It's enough that a vendor and a buyer, choosing inspectors separately, each naturally gravitate toward the one whose reputation matches what they want the report to say.

The moral risk this creates for each side. For a buyer, the risk is over-reliance on the idea that if something's wrong, someone else — a manufacturer, a council, the government's assistance scheme — will ultimately cover it, which can quietly reduce the incentive to commission genuinely thorough, independent due diligence before purchase. The numbers already in this FAQ suggest that reliance is largely misplaced: across three major James Hardie cases, homeowners recovered nothing, and government compensation ultimately covered roughly 11% of what was set aside. For a vendor, the risk runs the other way — a light-touch inspection lets a sale proceed without the vendor ever having to formally know about a problem, and under ordinary caveat emptor principles, what a vendor didn't know (or can't be shown to have known) is far harder to hold them liable for later. Neither side is necessarily acting badly. Each is responding rationally to how the costs and the risks are actually distributed.

Standing back, what would an investigator make of the whole crisis? Widen the lens from a single transaction to the entire history covered in this FAQ, and a pattern emerges: at nearly every stage, the party best placed to know the truth had a financial incentive that didn't clearly point toward disclosing it to the homeowner who ultimately carried the cost. Litigation funders backed cases on a purely commercial basis — and in the largest one, ended up paying the defendant money rather than losing it, while the homeowners they'd funded recovered nothing. Claims were promoted to prospective plaintiffs using numbers (participant counts, claim targets) that framed the case's strength in the most compelling available terms, and those numbers didn't end up matching the case that actually reached trial. Government compensation was capped and administered through a claims process that, by its own published figures, paid out a small fraction of what was originally set aside. None of this proves wrongdoing in any individual instance — courts examined several of these exact questions directly and largely didn't find misconduct. But the consistent direction of the incentives, stage after stage, is the kind of pattern an investigator looks at structurally: not one bad actor, but a chain of participants each responding rationally to their own financial position, in a system where the homeowner at the end of that chain had the least information and bore the largest share of the actual cost.

Yes — New Zealand wasn't the first, and wasn't the last. Very similar crises have hit at least three other countries, all driven by the same underlying pattern: synthetic or "monolithic" claddings applied without a drainage cavity, over timber framing, during a building boom with under-experienced trades and light-touch inspection.

United States (mid-1990s): The first of these crises to surface. Building inspectors in Wilmington, North Carolina discovered widespread rot behind EIFS ("synthetic stucco") cladding in August 1995 — nearly all of the area's 3,200 EIFS-clad houses were affected. A class-action lawsuit followed, and by the end of 1995 the problem — and the litigation — had spread nationwide. One North Carolina attorney alone won roughly $150 million in claims against EIFS manufacturers. This predates New Zealand's own crisis by around seven years, and involved some of the exact same cladding types later found responsible in NZ (EIFS was one of the three "monolithic cladding" types PwC identified here too).

Who was sued — and who wasn't — is the interesting part. In the United States, the litigation was aimed squarely at the EIFS manufacturers — the companies making the polystyrene-foam-and-plaster cladding systems themselves. James Hardie was not a defendant in the US cases. That's not because James Hardie escaped the US market; it's because James Hardie does not manufacture EIFS claddings at all. Its product line is fibre-cement panel systems (such as Harditex, the direct-fixed fibre-cement sheet at the centre of the NZ litigation), which is a completely different cladding technology from EIFS. So in the USA, a fibre-cement panel maker like James Hardie was never dragged into the EIFS litigation.

New Zealand ran the mirror image of the US story. Here, it was James Hardie that was taken to court — over its fibre-cement panels (Harditex, Monotek, Titan Board) — while no EIFS manufacturer was taken to court in New Zealand at all, even though EIFS was one of the same three "monolithic cladding" types PwC singled out as high-risk here, and even though the identical EIFS systems that failed and were litigated in the USA were being sold and installed in New Zealand over the same period. So the same cladding systems were on the market in both countries at the same time, with the same underlying defect (a sealed, non-draining monolithic face over timber framing), yet the litigation went in opposite directions in each place: EIFS manufacturers in the US, a fibre-cement manufacturer in NZ.

Why that's worth noticing. James Hardie is the biggest manufacturer of fibre-cement panels in the world. That exact product — fibre-cement panels — was litigated against in New Zealand (across three cases, covered earlier on this page) but was not litigated against in the United States, where the legal fire instead fell on the EIFS manufacturers. One product, one global manufacturer, two countries, the same era — and only one of those countries saw fit to take fibre-cement panels to court. Whether that reflects a genuine difference in how the product performed in each market, a difference in building practice (installation, cavities, detailing) between the two countries, or simply a difference in who the legal system and the litigants chose to aim at, is a question the public record doesn't fully answer. But the asymmetry itself — EIFS sued in the US, fibre-cement sued in NZ, neither product sued in the other country — is one of the more striking features of the whole crisis.

Canada — British Columbia (from the late 1990s): Known as the "leaky condo crisis," this centred on multi-unit condominium buildings in Vancouver and coastal BC built from the late 1980s to early 2000s. Estimates put damage at $4–5 billion across more than 900 buildings and 31,000 individual units — at the time, the most extensive and costly housing remediation in Canadian history. Some studies found 45% of all strata units built in BC between 1985 and 2000 leaked, and nearly 90% of three- and four-storey Vancouver strata complexes built 1980–1995 had serious problems.

Where the water actually got in (CMHC's own findings). Canada's housing agency, the Canada Mortgage and Housing Corporation (CMHC), investigated the failures in detail, and its findings put windows at the top of the list of water-entry locations. The specific defect set it identified reads almost like a checklist of the same monolithic-construction failures that drove the NZ crisis: leaking corner and frame joints; inadequate sill and head flashing; poor integration between window flanges and the building paper; missing or discontinuous sealant; no drained sub-sill pan; and water trapped at stucco returns. Crucially, CMHC reported that approximately 90% of the problems it investigated related to interfaces and penetrations — joints, windows, doors, and the points where different building elements meet — while only about 10% were attributed directly to the basic wall assembly itself. In other words, the wall cladding material was rarely the primary culprit; it was the detailing at openings and junctions. And while all cladding types experienced some failures, stucco walls showed substantially more reported problems and higher average repair costs than the alternatives — the same pattern New Zealand saw, where the monolithic, non-absorbent claddings failed more visibly and more expensively once water reached the framing.

Source: Canada Mortgage and Housing Corporation (CMHC), Keeping Walls Dry — the agency's research into the BC leaky condo crisis, on the dominance of window/interface failures (~90% of problems at interfaces and penetrations vs ~10% the basic wall assembly) and the higher failure rate and repair cost for stucco walls. The window defect list (corner and frame joints, sill/head flashing, flange-to-building-paper integration, discontinuous sealant, no drained sub-sill pan, water trapped at stucco returns) is drawn directly from CMHC's findings.

Where the problems were actually found — CMHC's field survey. The most useful Canadian field evidence is CMHC's Survey of Building Envelope Failures in the Coastal Climate of British Columbia. It covered multiple cladding systems rather than weatherboards alone, but it provides a valuable indication of where envelope problems occurred across the surveyed buildings:

Source: Canada Mortgage and Housing Corporation (CMHC), Survey of Building Envelope Failures in the Coastal Climate of British Columbia — distribution of recorded building-envelope problems by location. Note the same pattern as the NZ leak-cause data and the CMHC Keeping Walls Dry finding: windows and horizontal flashings together account for the largest share (~45%), decks and balconies add another ~17%, and the basic cladding/weather-barrier/sheathing itself is only ~8%. It is overwhelmingly the junctions and penetrations that fail, not the wall field.

Australia (more recent, still emerging): Concerns have grown particularly around apartment and strata buildings, compounded by a performance-based building code that varies by state, under-resourced enforcement, and a private-certification system that critics say creates gaps similar to those seen in NZ and Canada. It's viewed by some building-industry commentators as a crisis still unfolding rather than one already resolved.

This is a bigger, longer-running story than a single lawsuit — it's a recurring pattern across nearly every major US window manufacturer, spanning almost 40 years, and it's mechanically the same defect as EIFS and New Zealand's monolithic cladding: a sealed exterior layer trapping water against wood with no way for it to drain or dry.

Who was affected: Homeowners across the US over multiple decades, not one company or region. Separate class actions hit Pella, Marvin, Lincoln Wood Products, Weather Shield, Ply Gem, and — more recently — JELD-WEN.

What was affected: Almost always the same failure point — aluminum-clad or vinyl-clad wood windows and doors, where the cladding was meant to protect the wood core but instead sealed moisture in rather than out. Once water got behind the cladding it had nowhere to go, and the wood frame rotted from the inside, often invisibly, until it failed structurally or let water into the surrounding wall.

Pella Corporation ran two separate rounds. Its ProLine Series casement windows (aluminum-clad wood) were the subject of Saltzman v. Pella Corp, filed 2006. A $90 million settlement was reached in 2012, but the 7th Circuit Court of Appeals overturned it in 2014 (the appeal is known as Eubank v. Pella Corp) in a scathing opinion by Judge Richard Posner, who found class counsel had "sold out the class" — attorneys stood to collect $11 million in fees while class members would get at most $8.5 million, likely far less due to a deliberately confusing claims process. It took until 2019, 13 years after filing, to reach a fairer, reduced $26 million settlement. Separately, Pella's Architect and Designer Series windows were the subject of another case that settled for $35 million.

Marvin Windows and Doors settled a class action over windows and doors made 1985–1989 that rotted due to a defective wood preservative. Notably, the remedy wasn't cash — affected customers got 38–58% discounts on replacement products, a weaker outcome than Pella's cash settlements.

Lincoln Wood Products, Weather Shield, and Ply Gem each faced their own class actions over the same core defect pattern — inadequately sealed cladding (aluminum or vinyl) letting water reach and rot the wood core, or pool and transfer moisture into sills, frames, and surrounding walls. JELD-WEN is the most recent, with litigation still building toward possible class certification as of the early 2020s.

The Pella case became a landmark in US class-action law — Posner's 2014 ruling is still cited nationwide as the textbook example of a self-dealing settlement, reshaping how courts scrutinise attorney fees and claims-form complexity in consumer class actions generally. But the bigger pattern is the recurrence itself: the same defect kept resurfacing across different manufacturers and materials for four decades — the same systemic-failure story New Zealand, Canada, and Australia each went through in their own building industries.

No. Direct-fixed cladding over building paper is not inherently a defective system. It can perform extremely well when the cladding profile, building shape, exposure, underlay, flashings and workmanship collectively allow water to drain and the wall to dry.

The important distinction is between traditional direct-fixed overlapping cladding, such as timber weatherboards, and direct-fixed face-sealed cladding, such as flush-finished fibre-cement sheets or plaster systems. Both may be called "direct-fixed", but their moisture performance is fundamentally different. The sections below trace how each approach evolved — and why one endured for three centuries while the other became the centre of a national crisis.

Wooden clapboards, bevel siding and overlapping boards were used across Britain, Scandinavia and colonial North America, commonly fixed directly to framing, boarding or timber sheathing. Modern building paper was not generally present. These walls survived because:

Boards overlapped and shed most rain · buildings had steep roofs and generous eaves · window and door heads were protected · walls were relatively simple · air moved readily through the construction · solid timber could absorb limited moisture and subsequently dry · and there was no impermeable internal lining trapping moisture.

This was a drain-and-dry assembly, even though it did not contain a formally designed cavity. Traditional clapboard has been used from the eighteenth century through to the present.

The US National Park Service records clapboard as a common exterior finish from the 1700s onwards, and notes the continuing survival of historic timber siding.

Rosin-sized and tar-treated papers developed during the nineteenth century, with rosin paper recorded as being used as building paper by approximately 1850. Paper behind siding improved resistance to wind-driven rain, draughts, dust, water passing through weatherboard laps, and air movement through boarded walls.

Crucially, the paper was not expected to make the exterior cladding waterproof. Instead it acted as a secondary drainage plane — water reaching it was supposed to run downward and discharge outside.

During the early-to-mid twentieth century, asphalt-saturated felt or kraft building paper became common behind timber siding, stucco, shingles and other claddings in New Zealand, Australia, Canada and the United States.

New Zealand timber weatherboards were typically nailed through the building paper to studs. There was no deliberate 20 mm cavity, but the system still possessed some drainage and drying capacity because:

Weatherboards overlapped · the rear surfaces were not perfectly flat · each horizontal lap interrupted water movement · timber could temporarily store and release moisture · and traditional homes generally had simpler walls and larger eaves.

Many New Zealand houses built between the 1920s and 1980s demonstrate that this arrangement can last for many decades.

Synthetic house wraps gradually replaced traditional bituminous papers. A house wrap is intended to resist liquid water while remaining sufficiently vapour-permeable to allow moisture to escape. It is not insulation, and it is not intended to remain exposed as the primary exterior surface.

At the same time, buildings became more complicated. Eaves became smaller or disappeared · upper-storey walls became more exposed · internal gutters and parapets increased · windows became flush with the cladding · insulation reduced the heat passing through walls and therefore reduced drying · interior linings became more airtight · untreated framing became common · and flush-finished fibre-cement and EIFS walls relied heavily on coatings and sealants.

This combination severely reduced the system's tolerance of small errors.

A properly built traditional wall contains several layers of defence:

The weatherboards are not assumed to be perfectly watertight. Their shape, laps, flashing details and ability to dry make the assembly resilient.

A drained cavity adds another layer by physically separating the cladding from the underlay and framing. It provides a clearer drainage path, improves ventilation, and reduces capillary transfer.

Flush-finished fibre-cement sheets could be installed directly over building paper and framing, with joints reinforced and plastered to make the wall appear seamless. That created several weaknesses:

1. A sheet could lie tightly against the underlay. Water did not necessarily have a clear path downward.
2. Large sheets moved differently from timber framing. Thermal movement, moisture movement and structural movement could crack stopped joints.
3. Windows interrupted the drainage plane. Without effective head flashings and sill trays, water could enter behind the sheet.
4. The coating became the principal defence. Once a crack, penetration or failed sealant admitted water, the wall had little redundancy.
5. Moisture could become trapped. Insulation, internal linings and direct contact between layers reduced drying.
6. Untreated framing had little survival capacity. A small leak could progress into extensive concealed decay.

MBIE explains that moisture becomes particularly damaging when it enters behind certain claddings and there is no drainage or ventilation space between the cladding and the framing (building.govt.nz — signs of a leaky home).

Therefore, the critical problem was not simply "building paper plus direct fixing". It was direct fixing combined with a face-sealed surface, complex architecture, weak junction details and limited drying capacity.

Direct-fixed cladding remains in use internationally, particularly for timber lap siding, vinyl siding, fibre-cement weatherboards, metal siding with profiled drainage paths, shingles and shakes, and some stucco assemblies using multiple layers of building paper.

However, wet or highly exposed regions increasingly favour drained and ventilated rainscreens. Formal rainscreen research developed in Norway and Canada during the 1960s, recognising that controlling pressure, drainage and drying was more reliable than attempting to make the exterior face completely waterproof.

Direct fixing is more defensible where the building has:

One or two simple storeys · adequate eaves · moderate wind exposure · overlapping cladding · proper sill, jamb and head flashings · effective ground clearance · a correctly lapped and undamaged wall underlay · drainage at the bottom of the wall · durable or appropriately treated framing · and accessible junctions with regular maintenance.

A cavity becomes increasingly important where there are high wind pressures, multiple storeys, parapets, enclosed decks, complex junctions, minimal eaves, or absorbent and face-sealed claddings.

Direct-fixed construction is an old, proven method — but only when the wall is understood as a water-management system rather than a waterproof surface.

Traditional weatherboards worked because they shed, interrupted, drained, stored and released moisture. Problematic monolithic walls attempted to create a seamless waterproof face. When that face failed, the wall behind it often had no effective second or third line of defence.

A cavity does not excuse poor workmanship, and direct fixing does not automatically mean failure. The cavity simply gives the building greater tolerance when materials move, sealants age, or a builder makes a small mistake. In exposed New Zealand conditions, that additional tolerance is extremely valuable.