Treadstone Associates
Case File · Latent Defects

A material latent defect nobody disclosed

Anonymised, illustrative composite. A failed sale two years earlier left a paper trail that turned an ordinary caveat-emptor argument into a documented actual-knowledge case.

Treadstone Associates · Updated 2026

At a glance

  • • Ontario resale, $540,000 purchase price; a foundation crack, cosmetically patched, went undetected by the buyer’s own home inspector.
  • • Two years earlier, a different buyer’s inspection on the same property had flagged the same crack in a written report the seller received and kept.
  • • RECO Bulletin 7.4 turns disclosure on whether a defect is patent (detectable on reasonable inspection) or latent and known to the seller.
  • • Two structural-repair quotes came in at $34,500 (helical piers) once the buyer had the wall opened up for assessment.

The situation

A buyer purchased a resale bungalow for $540,000 after a standard home inspection turned up nothing structurally concerning. Eighteen months later, cracking reappeared along an interior basement wall in the same spot the original paint job had covered. A structural engineer traced it to a foundation issue that had been patched cosmetically — parged over and painted — rather than actually repaired, well before this buyer’s purchase. The buyer’s own inspector had photographed that same wall and noted nothing more than “minor cosmetic hairline cracking, monitor,” which is precisely what a competent patch job is designed to produce on a visual inspection.

The problem

The buyer’s lawyer discovered, through the seller’s own real estate file produced during the dispute, that a different buyer had made an offer on the same property roughly two years earlier, backed out after a home inspection flagged the identical crack, and that the seller had received and kept that inspection report. Nothing about that history had been mentioned to this buyer. Under RECO Bulletin 7.4, a patent defect — one “detected by a potential home buyer or their home inspector by reasonable observation and inquiry” — carries no disclosure duty. A latent defect, one that “would make a property unfit for habitation, dangerous, or potentially dangerous” and is not apparent on a reasonably careful inspection, must be disclosed. A cosmetically patched crack that this buyer’s own inspector missed reads as latent, not patent — and the seller’s prior inspection report made the actual-knowledge question straightforward instead of speculative.

The numbers

$540,000 purchase price. Once the wall was opened for proper assessment, two structural firms quoted $34,500 for helical-pier underpinning to correct the underlying settlement — a figure specific to this property’s foundation and soil conditions, not a general repair-cost benchmark.

The rule that decided it

The doctrine that governs here is caveat emptor’s actual-knowledge exception: a seller has no general duty to volunteer everything they know, but that changes once the seller has actual knowledge of a defect that makes the property dangerous or unfit for habitation. “The deciding factor is the seller’s actual knowledge, not merely whether the defect was hidden” is the precise formulation the doctrine turns on, and it is why the two-year-old inspection report mattered so much more than the patched wall itself: a patch job alone might be explained as a cosmetic touch-up nobody thought twice about, but a patch job made after receiving a written report naming the exact defect is very hard to explain as anything other than actual knowledge, concealed rather than disclosed.

The outcome

Faced with the earlier inspection report in the seller’s own file, the seller’s position shifted quickly from “I didn’t know” to negotiating a settlement that covered the $34,500 repair cost. The case never needed to argue what a reasonable seller ought to have discovered — the documentary record answered the actual-knowledge question directly, which is a substantially stronger position for a buyer than the ordinary “the seller should have known” argument most latent-defect disputes are stuck arguing.

The seller’s own agent from the prior, failed transaction was still at the same brokerage, and confirmed in a written statement that the inspection report had in fact been forwarded to the seller at the time — removing any remaining room to argue the report had simply never been seen.

What it would have cost otherwise

Without that paper trail — if no prior inspection report had ever existed, and the crack had simply been patched by the seller on their own initiative years earlier without any documented flag from anyone — this buyer’s claim would have collapsed to circumstantial inference about what the seller must have known, a considerably weaker position under a doctrine that hinges on actual, provable knowledge rather than what a careful seller arguably should have noticed on their own property.

The tell

The buyer’s lawyer asked, as a matter of routine file review, whether the property had ever previously been under a firm agreement that fell through — a question that is easy to skip when a deal closes cleanly the first time it’s tried. A prior failed sale is exactly where an inspection report naming a defect is most likely to already exist on file somewhere.

Takeaways

  • • RECO Bulletin 7.4 draws the disclosure line at patent versus latent: detectable on reasonable inspection carries no duty; not apparent, and dangerous or unfit for habitation, does.
  • • A seller’s actual knowledge is the trigger, not whether a defect happened to be hidden — and a prior inspection report is direct, documentary proof of that knowledge.
  • • A cosmetic patch made after receiving a report naming the exact defect is a very different fact pattern from an old repair nobody thought twice about.
  • • Ask whether a property was ever under a prior firm agreement that fell through — the paper trail from that deal often already answers the actual-knowledge question. See a companion file where a permit history did the same evidentiary work.

See where AI pays off first in your business.

A 30-minute call is enough to tell you whether AI pays for itself here.