The Concrete Was Failing From Within. The Collapse Clause Paid Only at Imminent Danger of Falling Down.

The cracks arrive in an order that homeowners in north central Connecticut learned to read like a diagnosis. First a horizontal line near the top of the basement wall. Then a spiderweb spreading down. Then concrete soft enough to scrape with a screwdriver.

The house is coming apart. The house is also standing perfectly straight.

That combination, disintegrating and upright, is the exact thing thousands of homeowner policies were not built to answer. The fight over who paid came down to one word printed in the coverage grant. The word was “collapse.”

The mineral inside the stone

The cause is pyrrhotite, an iron mineral that swells when it meets water and air.

It did not get poured in on purpose. It rode in inside the aggregate, the crushed stone that goes into every batch of concrete. In the Karas case, the Connecticut Supreme Court record traced the concrete to J.J. Mottes Concrete Company, whose quarry was mined from 1983 to at least 2010.

The reaction is slow and quiet. A foundation can look ordinary for years and then begin failing from the inside out, long after any builder’s warranty has expired and long after anyone remembers which truck delivered the pour.

State and regional bodies have put roughly 35,000 homes in the north, east and central parts of Connecticut inside the potentially affected zone. The Capitol Region Council of Governments uses that figure, and Claims Journal reported it in 2019 as homes built with suspect concrete between 1983 and 2015.

Nobody knows which of those 35,000 will fail. The Office of Legislative Research, citing the Capitol Region Council of Governments, put the potentially affected homes in at least 36 Connecticut towns.

What the policy actually promised

The Karas policy did not cover a foundation for wearing out. Its collapse language said in terms that collapse “does not include settling, cracking, shrinking, bulging or expansion,” and it excluded loss to a foundation unless that loss came directly from a building’s collapse.

What it did include was an additional coverage for collapse.

That single grant became the entire case. If the slow destruction of a foundation counted as a collapse, the carriers owed a new foundation, a job the state fund now caps at $205,000 in allowable costs. If it did not, the answer was zero.

Gail and Steve Karas found their cracks in Vernon in a house built in 1984. Liberty Insurance Corp., a Liberty Mutual company, denied the claim under the exclusions. The dispute moved through federal court, which asked Connecticut’s highest court to say plainly what “collapse” means.

The word had two different meanings, and the policy chose

Which definition a policy printed decided which families got paid.

Connecticut did not have one collapse standard in 2019. It had two, and which one applied depended on whether a homeowner’s own policy bothered to print a definition.

Where the policy left “collapse” undefined, Connecticut’s older case law read it generously, as a substantial impairment of structural integrity rather than a pile of rubble. Where the policy printed its own definition, the definition governed.

On November 12, 2019, the Connecticut Supreme Court decided Karas v. Liberty Insurance Corp. together with two companion cases, Vera v. Liberty Mutual Fire Insurance Co. and Jemiola v. Hartford Casualty Insurance Co.

In Jemiola, the policy defined collapse as “an abrupt falling down or caving in of a building or any part of a building” such that it “cannot be occupied for its current intended purpose.” The court held that language was clear. A cracking but standing foundation is not an abrupt falling down. No coverage.

In Karas, where the term was undefined, the court did not simply hand the homeowners the older, friendlier standard. It tightened it. Substantial impairment of structural integrity now requires evidence that the building is in imminent danger of falling down.

A foundation that is quietly disintegrating, in a house that is safe to sleep in tonight, does not clear that bar.

The gap was never in the science

Nobody in these cases disputed that the concrete was failing. The chemistry was not the argument.

The argument was about grammar, and the homeowners lost it in two directions at once. One group lost because their carrier had written a tight definition years earlier. The other group lost because the court read a loose one more strictly than the old cases had.

The Claims Journal headline on November 14, 2019 put the result flatly: “Conn. Supreme Court Finds Insurers Not Liable for Crumbling Foundations.”

The state built its own carrier instead

Connecticut did not wait for the courts. The legislature had already moved.

Under Public Act 17-2 of the June 2017 special session, the state created the Connecticut Foundation Solutions Indemnity Company, known as CFSIC. It is a captive insurer, meaning an insurance company created to cover exactly one thing, in this case failing foundations. It was set to open for applications on January 10, 2019, ten months before the Supreme Court ruled.

CFSIC is funded from two places. One is state bonding. The other is a $12 surcharge on Connecticut homeowner insurance policies, created by Public Act 18-160 and running through December 31, 2029, with 85 percent of the revenue set aside for crumbling foundations.

So the money that replaced the denied coverage came partly from every other homeowner in the state, twelve dollars at a time.

What the fund’s own pages show

CFSIC publishes its own numbers.

On August 24, 2026, the fund announced that it had completed its 1,300th foundation. In a June 22, 2026 update it put the average replacement value of the 125 foundations it had finished in South Windsor at $167,418, about $37,500 under its $205,000 cap. Its own FAQ says it will stop taking new applications at 5:00 p.m. on June 30, 2030.

Set 1,300 against the 35,000-home exposure estimate and the arithmetic speaks without help. Fewer than four homes in a hundred of the estimated zone have been through the fund, which is our division of those two numbers.

The dockets and the fund’s pages tell the same story from two directions. The private contract said the loss was not the kind of loss it had promised to pay. The public fund said someone still had to pay it, and picked a different pocket.

The Hurwitz Fine summary of the ruling puts the undefined-collapse standard in the court’s words: coverage requires proof the structure faces “imminent danger of falling down.” The opinions and the fund’s own pages are linked in the Sources block below.