Water Exclusion Clause

 

July 13, 2015

 

In Property Insurance Forms

 

Summary: A standard feature of property insurance forms is the exclusion of various categories of loss associated with water. The water exclusion clause in the commercial property program (Insurance Services Office forms CP 10 10 10 12, CP 10 20 10 12, and CP 10 30 10 12) and in the ISO homeowners policy excludes loss caused by flood, surface water, water that backs up from a sewer or drain, and water under the ground surface. Though the exclusionary language is quite broad, disputes continue to arise when the language is applied to particular loss situations. The controversies created by differing views have led to several court decisions that are discussed in this article. Although some of the cases cited in this article are old, all have been verified as being valid case law.

Concurrent Causation

 

The introductory paragraph to the water damage exclusion says that the policy will not pay for a loss caused directly or indirectly by the excluded peril, regardless of any other cause or event that contributes concurrently or in any sequence to the loss. These words were introduced in 1986 to address the problem of concurrent causation. Briefly, the theory of concurrent causation was used by claimants to argue that if, under an open perils policy, there were two causes of a loss, one excluded and one not, the loss would be covered. For example, if faulty construction of a dam caused it to burst, flooding land downstream, property damage to property downstream caused by the flood should be covered in spite of the flood exclusion because the faulty construction of dams was not excluded.

 

Because the catastrophic nature of a flood, hurricane, or tidal wave means that not one or two, but possibly hundreds or thousands of units in a general area are damaged, and given the fact that only persons in flood-prone areas will purchase coverage against the exposure, such risks are uninsurable through normal channels. Proponents of the concurrent causation theory argued that it is precisely the catastrophic exposure—that represented by natural and inevitable events rather than accidental or artificial causes—at which the water exclusion, theoretically, is aimed, and not a loss caused by faulty construction of a dam.

 

This wording must therefore be kept in mind when reading about the cases cited in this section since most of them occurred while the theory of concurrent causation was still operative. For a full discussion of concurrent causation,  see Concurrent Causation and Efficient Proximate Cause.

 

After Hurricane Katrina, an anti-concurrent cause exclusion used by State Farm came into question. The court found that “storm surge” is “little more than a synonym for a 'tidal wave' or wind-driven flood,” both of which are perils excluded by the State Farm policy. The water damage exclusion did not render the policy ambiguous or allow the insured party to recover for losses caused by the storm surge associated with Hurricane Katrina. [Tuepker v. State Farm Fire & Cas. Co. 507 F.3d 346 (2007).]

 

Courts and insurers had great difficulty with the concurrent cause doctrine. Eventually, the courts took on the issue and many have eliminated the concurrent cause doctrine and replaced it with one determining coverage by a requirement that insurers and courts determine the efficient proximate cause of the loss.

 

In Sabella v. Wisler, 377 P. 2d 889 (Cal. 1963) and affirmed in Garvey v. State Farm Fire & Casualty Co., 770 P. 2d 704 (Cal. 1989) and State Farm Fire & Casualty Co. v. Von Der Lieth, 820 P. 2d 285 (Cal. 1991), the concept of “all risk” was clarified. The court in Von Der Lieth explained:

 

[T]he scope of coverage under an all-risk homeowner's policy includes all risks except those specifically excluded by the policy. When a loss is caused by a combination of a covered and specifically excluded risks, the loss is covered if the covered risk was the efficient proximate cause of the loss. [T]he question of what caused the loss is generally a question of fact, and the loss is not covered if the covered risk was only a remote cause of the loss, or the excluded risk was the efficient proximate, or predominate cause … [I]f third party negligence is not excluded under such a policy, it is a covered peril.

 

A case that brought to light the concurrent causation doctrine is Safeco Ins. Co. v. Guyton, 692 F.2d 551 (9th Cir. 1982). Following heavy rains in September, 1976, a system of flood control channels, levees, and dikes in the Palm Desert area of California failed to contain the accumulation of water. A number of homeowners insureds sought recovery under their all risk policies for the ensuing flood damage, partly on the grounds that their loss was proximately caused by the water district's negligence in maintaining the flood control system. The insurers invoked the standard homeowners water exclusion— flood or surface water—and argued that the flood itself had preceded in the chain of causation any negligence that might have been a contributing or intervening cause of the damage.

 

A California district court agreed with the insurers but was overruled when the case was appealed to the Ninth Circuit Court of Appeals. That court held that “two independently created conditions [the flood and the negligently maintained dike] interacted to cause [the] loss” and that an all risk homeowners policy, since it provides coverage for damage caused by third party negligence, would respond to the insureds' loss.

 

The Supreme Court of Colorado reached a different decision in Kane v. Royal Ins. Co. of America, 768 P.2d 678 (Colo. 1989). The Lawn Lake Dam in Rocky Mountain National Park failed. The water swept downhill into the Fall River, which overflowed its banks and flooded many homes and businesses. The insureds' argument for coverage took two forms. The first was that the term “flood” was ambiguous since the policy made no distinction between naturally and artificially caused floods. The second argument was that the flood was caused by negligence, which was not an excluded cause of loss. In a decision that seemed to forecast the policy language in use today and the Garvey decision, the judge found the arguments did not hold water and that the term “flood” was not ambiguous. Referring to Webster's Collegiate Dictionary, Ninth Edition, the judge in the Kane decision stated that a “flood” is the “rising and overflowing of a body of water especially onto normally dry land.” The definition made no distinction between naturally and artificially occurring flooding. The judge found, in ruling against the second argument, that the policy language precluded coverage because loss “caused by, resulting from, contributed to or aggravated by [a flood]” was excluded. Although the loss might have been caused by negligence, it was also caused by the natural conditions—the amount of rainfall—that preceded the flood.

 

In California and those states that follow Garvey, the concurrent causation doctrine applies only to third-party liability cases—it does not apply to first-party policies. When evaluating coverage under a third-party liability policy, if a proximate cause of the loss that is not excluded occurs with a proximate cause of the loss that is excluded, coverage applies. In a first-party policy, on the other hand, under Garvey and its progeny, it does not matter how many causes of loss concur to cause the loss. Coverage is determined by the efficient proximate or moving cause of loss.

 

In Vision One, LLC v. Philadelphia Indem. Ins. Co., 276 P.3d 300 (Wash. 2011), the court concluded that damages associated with collapse of a concrete slab were covered under the ensuing loss clause in the exclusion for faulty workmanship; that when an excluded peril sets in motion a causal chain that includes covered perils, the efficient proximate cause rule does not mandate exclusion of the loss, and the insurer could not rely on the “sequence of events” causation clause in the policy to deny coverage.

 

In Western Nat. Mut. Ins. Co. v. University of North Dakota, 643 N.W.2d 4 (ND 2002), the North Dakota Supreme Court found that a boiler and machinery insurer sued its insured for a declaratory judgment that its flood exclusion barred coverage for damage to boiler and machinery equipment after sanitary sewer lift stations were shut down during a river flooding. The court found that (1) concurrent causation language of flood exclusion was unenforceable to the extent it purported to exclude coverage in a manner contrary to statutes adopting the efficient proximate cause doctrine, and (2) evidence supported conclusion that sewer backup, not flood, was the efficient proximate cause.

 

In Julian v. Hartford Underwriters Ins. Co., 35 Cal.4th 747 (2005), following heavy rains, a slope failed above the West Hills, California home of Frank and Carole Julian. The slope failure led to a landslide, which caused a tree to crash into the Julians' house. Soon thereafter, the Julians presented a claim for the resulting damage to their insurer, Hartford Underwriters Insurance Company. The Julians had a standard form homeowners insurance policy with Hartford. The California Supreme Court looked at claims of efficient proximate cause and reasoned as follows:

 

[A]n insurance company can limit the coverage of a policy issued by it as long as such limitation conforms to the law and is not contrary to public policy.” (Lumberman's Mut. Cas. Co. v. Wyman (1976) 64 Cal.App.3d 252, 259, 134 Cal.Rptr. 318.) “An insurance policy may exclude coverage for particular injuries or damages in certain specified circumstances while providing coverage in other circumstances.” (Frank and Freedus v. Allstate Ins. Co. (1996) 45 Cal.App.4th 461, 471, 52 Cal.Rptr.2d 678.) It follows that an insurer is not absolutely prohibited from drafting and enforcing policy provisions that provide or leave intact coverage for some, but not all, manifestations of a particular peril. This is, in fact, an everyday practice… For example, a policy might exclude losses caused by freezing to plumbing, but provide coverage for other types of freezing, or vice versa. The fact that the exclusion does not apply to all types of freezing does not, by itself, render it invalid. Likewise, an insurance policy can provide coverage for weather conditions generally, but exclude coverage for specific weather conditions such as hail, wind, or rain. The fact that hail, wind, and rain are types of weather conditions does not bind the insurer to insure against all weather conditions, or none at all. A reasonable insured would readily understand from the policy language which perils are covered and which are not. Similar logic applies where the limitations of our language require an insurer to describe a specific peril in terms of a relationship between two otherwise distinct perils (e.g., rain and landslide) in order to plainly and precisely communicate an excluded risk.

 

The peril of rain inducing a landslide is a genuine one. Rain inducing a landslide is a commonly understood risk of loss and the frequent and direct causal relationship between rain and landslide is widely and easily understood. The Hartford engineer's report attested that the type of slope failure involved in this case was always caused by water. The landslide here was not an independent causal agent in the Julians' loss; by all accounts it was dependent on the weather condition of heavy rains. Accordingly, to the extent the weather conditions clause excluded the specific peril of rain inducing a landslide, there was no violation of the concurrent cause doctrine.

 

The California Supreme Court concluded:

 

We hold, in sum, that the weather conditions clause excludes the peril of rain inducing a landslide and that as applied here the clause does not violate section 530 or the efficient proximate cause doctrine. Because the policy effectively excludes the perils of earth movement, third party negligence, and rain inducing a landslide, and the Julians produced no evidence that a different peril was the efficient proximate cause of their loss, we agree with the Court of Appeal that the trial court did not err in granting Hartford summary judgment.

 

The current policy language seeks to eliminate any possibility of coverage, as has already been stated, through the wording “such loss is excluded regardless of any other cause or event contributing concurrently or in any sequence to the loss,” and through the addition of exclusions for weather conditions and acts or decisions, including those of governmental bodies. That intent was confirmed by the California Supreme Court in Julian v. Hartford Underwriters Ins. Co.

 

Flood Insurance

 

Flood insurance, of course, is available for both residential and commercial risks through the National Flood Insurance Program. In the flood insurance policy, “flood” means either the overflow of inland or tidal waters or the unusual and rapid accumulation or runoff of surface waters from any source. A flood occurs when these waters invade land areas that are normally dry and inflict a condition—that is both general and temporary—of partial or complete inundation. Also included as flood are mud slides caused by flooding. Thus, flood insurance aims to cover the traditionally uninsurable natural forces (whether or not set in motion by man) that result in widespread damage verging on or equaling the catastrophic.

 

Although such events as water flowing from a ruptured city water line have been termed a “flood,” generally courts have held that a flood is a large-scale inundation or deluge from a body of water. The court in Kane, discussed previously, for example, found that a water main was not a body of water, so water from a ruptured main did not constitute a flood.

 

In Mt. Zion Missionary Baptist Church of Cedar Rapids v. Church Mut. Ins. Co. 860 N.W.2d 923 (Table) (Iowa App. 2014), the court concluded that a flood exclusion applied. The court stated that the policy language was not ambiguous and that water that backs up due to an event away from the property is not covered if the backup was caused by flood, surface water, or overflow of any body of water.

 

In Wallis v. Country Mut. Ins. Co., 723 N.E.2d 376 (Ill. App. 2000), the court found that a manmade watercourse was a body of water, and the water exclusion precluded coverage for damage caused when it flooded. Larry and Mae Wallis owned a home with the Rob Roy creek flowing about 150 feet away. The creek was not natural but was a manmade irrigation channel used by nearby farms.

 

Severe storms struck the area and the creek rose out of its banks, eventually reaching the Wallis home and flooding the crawlspace underneath the house, the garage, and the living quarters, causing extensive damage. Their homeowners insurer, Country Mutual, denied their claim based on the water damage exclusion.

 

The homeowners argued that the damage was not caused by a flood, surface water, or back-up from a sewer or drain, which is what the water exclusion addressed. They also asserted that the terms were ambiguous because more than one interpretation could be made. The court, though, quoting Shakespeare, said that “plaintiff's argument is 'much ado about nothing.'”

 

The insureds said that the damage did not result from a flood because the water overflowed from a manmade source, which, they claimed, was not how the term “flood” had been interpreted in the past. In contrast, the court said, “The common meaning of a flood is 'a rising and overflowing of a body of water that covers land not usu[ally] under water.” The creek was a body of water, the insureds' home was not usually under water—ergo, a flood occurred. The court found no credence in the argument that a manmade body of water cannot cause a flood. The court stated, “The creek is a permanent watercourse. The creek has a defined bed, visible banks, and a recurrent flow of water.” That the creek was manmade did not matter as long as it retained the other characteristics listed. Because flood was part of the water exclusion, the loss was not covered.

 

The Chicago flood of 1992 brought another dimension to the term's meaning. In that event the freight tunnels beneath the Chicago River were breached, allowing some 300 million gallons of water to flood the downtown office area. The ISO commercial forms exclude flood and water under the ground surface “flowing or seeping through foundations, walls, floors or paved surfaces; basements. . . or doors, windows or other openings,” which would preclude coverage for property. Cases involving loss of business income, however, alleged the losses were the product of a civil evacuation order and not flood.

 

Surface Water and Sewer Back-Up

 

An often-cited precedent for construing the water exclusion clause's reference to surface water was established in Richman v. Home Ins. Co. of New York, 94 A.2d 164 (Pa. Super. 1953). The court stated that an examination of the various definitions of “surface water” shows that the term is commonly understood to mean water on the surface of the ground usually created by rain or snow that is of a casual or vagrant character, following no definite course and having no substantial or permanent existence.

 

In Neal v. Ohio River R. Co., 34 S.E. 914 (W.Va. 1899), surface water was defined as “water of casual, vagrant character, oozing through the soil, or diffusing and squandering over or under the surface, which, though usually and naturally flowing in known direction, has no banks or channel cut in the soil; coming from rain and snow, and occasional outbursts in time of freshet, descending from mountains and hills, and inundating the country; and the moisture of wet, spongy, springy, or boggy land. For obstructing or diverting surface water, though damaging another, the party is not liable.”

 

In Salem United Methodist Church of Cedar Rapids, Iowa v. Church Mut. Ins. Co. No. 13-2086, 2015 WL 1546431 (Iowa App. April 8, 2015), the court found that the exclusion for flood damage was clear and the policy unambiguously excluded coverage for damages that were concurrently caused by a covered cause—such as sewer back up—and an uncovered cause—such as flooding—by its language, “Such loss or damage is excluded regardless of any other cause or event that contributes concurrently or in any sequence to the loss.” The policy plainly covered “damage to Covered Property caused by back up of water or sewage through sewers or drains only…when the damage is not caused by flood.” The policy reiterated that “[s]ewer or water damage occurring as a result of, either before or after, the excluded flood…is not covered.” And, in its general exclusion section, the policy separately excluded loss or damage caused directly or indirectly by flood.

 

Similarly, in Durrett v. Nationwide Property and Cas. Ins. Co., No. A-14-CA-167-SS, 2015 WL 1564783 (W.D. Tex. April 7, 2015), the only admissible evidence indicated the sewage water that backed up into the insured's house came down from the city sewer line. Based on this finding, the insurer properly applied the exclusion for damage caused by “water or water-borne material which backs up through sewers or drains from outside dwelling's plumbing system.”

 

This understanding of the meaning of “surface water” may be seen in Georgetowne Square v. United States Fidelity and Guaranty Co., 523 N.W.2d 380 (Neb. App. 1994). A retaining wall on the insured's property was damaged by water and water pressure from an underground pipe that drained water from the roof of a neighbor's building. The insurer denied the claim based on a policy provision excluding coverage for damages resulting from flooding, which the policy defined to include “run-off” and “surface water.” The court found that surface water is water “which is diffused over the surface of the ground, derived from falling rains or melting snows, and continues to be such until it reaches some well defined channel in which it is accustomed to flow and does flow with other waters.” The court further found that surface water and run-off both lost their characteristics as surface water when diverted into artificial channels.

 

With respect to damage from water that backs up through sewers or drains, distinctions have been made between cases involving a high water table and those confined to sewer blockage—where the sewer fails to accept the water from within the structure's plumbing system and that water spills over the structure. While it is apparent that the exclusion applies in the case of the high water table, it is not so clear that it applies to the second situation. In the case of sewer blockage where the sewer fails to accept the water from within the plumbing system, the water has never been in the sewer system. For a discussion of cases involving the back-up of sewers and drains exclusion as well as the sewer back-up endorsements, see Coverage for Back-Up of Sewers and Drains. These endorsements agree to insure “direct loss caused by water which backs up through sewers or drains and which is not caused by the negligence of the insured.” The endorsements do not exclude sewer back-up caused by the influx of surface water through the drainage system as may occur after a cloudburst.

 

Courts have also ruled that raw sewage is “water” for purposes of the water exclusion. For example, in Citrano v. Hingham Mut. Fire Ins. Co., 788 N.E.2d 975 (Mass. App. Ct. 2003), an insured incurred damage when waterborne sewage backed up through a basement toilet due to heavy rains. The homeowners insurer, Hingham, denied coverage based on the water damage exclusion.

 

The insureds argued that the damage was not caused by water but by sewage. The court, though, stated that the water referred to in the exclusion was “not pure tap water or rain water, but only water that backs up from sewers (or drains).” The court found that common usage of the word “sewage” “implies waste borne in water.” The court said even if a difference existed between sewer water and the sewage carried in it, the exclusion would still apply because water, which backed up in the sewer line and transported the raw sewage into the basement, was a contributing cause of loss.

 

Similarly, in Rodin v. State Farm Fire & Cas. Co., 844 S.W.2d 537 (Mo. App. 1992), a Missouri appellate court held that “whether the loss was caused by pure water or by the pollutants contained in the sewage acting concurrently with water, it is excluded from coverage by clear, unambiguous policy language.”

 

Underground Water versus Accidental Discharge

 

Over the years, courts have tended to find some basis for limiting the scope of the underground water exclusion in an attempt to resolve the apparent conflict between it and the named peril of accidental discharge from plumbing systems. The accidental discharge peril covers accidental discharge or leakage from the breaking or cracking of any part of a system or appliance containing water or steam. Although many court decisions have established that coverage under the accidental plumbing discharge peril applies only to damage arising from the insured's own plumbing system, the current homeowners edition leaves this issue open to question, as will be seen later.

 

It should be noted that the cases discussed here involve losses of a sudden and accidental nature. When damages result from a slow leak or build-up of water over a period of time, many property forms provide insurers with some protection by excluding loss from continuous or repeated seepage or leakage. For example, the ISO commercial property forms exclude coverage for loss or damage caused by “continuous or repeated seepage or leakage of water that occurs over a period of 14 days or more.”

 

In Fine v. Underwriters at Lloyd's of London, 239 F.2d 362 (3rd Cir. 1956), water leaked from the plumbing system in an adjoining building and seeped through the walls of the insured's building. The court held that the exclusion of seepage applied, even though the water originated from a leak in a plumbing system. A later Pennsylvania court decision took the Fine case as its precedent is Kozlowski v. Penn Mutual Insurance Co., 441 A.2d 388 (Pa. Super. 1982). When a water main curb valve on a neighbor's property burst, water from the leak seeped into the insured's basement. Although the no-coverage ruling of the court was based upon its finding that the homeowners accidental plumbing discharge peril applies only to the insured's own plumbing system, the court went on to a more general consideration of the underground water exclusion in relation to the accidental discharge peril. “Giving effect to all of [the policy] provisions,” the court said, “we hold that where . . . the cause of the discharge or overflow of water is in the insured's own plumbing system, the [underground water] exclusion does not apply; but that if the cause is outside that system, then the exclusion is applicable.”

 

As noted previously, the current ISO homeowners forms present another facet. For example, suppose a neighbor's plumbing ruptures, sending a cascade of water into the adjoining home. Historically the water damage exclusion for surface water has generally been interpreted to mean water from natural sources. Therefore, in this instance, would the “off premises” limitation apply? In the HO 00 02 form the accidental discharge peril, with the off-premises exclusion, applies to all section I coverages. But under form HO 00 03, coverages A and B are covered for open perils and there is no off-premises accidental discharge exclusion. The accidental discharge off-premises exclusion appears only with respect to coverage C. If the off-premises exclusion was meant to clarify the water damage exclusion, the proper place for the clarification would have been within the water damage exclusion itself rather than in the accidental discharge peril.

 

In Liberty Mut. Fire Ins. Co. v. Martinez, 157 So.3d 486 (Fla. App. 2015), a pool and deck popped out of the ground. The court ruled that damage to the insured's pool deck, rock garden, and waterfall was not an ensuing loss. The loss was directly due to subsurface water pressure and was excluded as such. The ensuing loss provision did not apply.

 

In a unique case that combined aspects of underground water with accidental discharge, damage caused by underground water was covered because it flowed into and was discharged from an appliance of the insured. In Stone v. Royal Ins. Co., 511 A.2d 717 (N.J. Super. 1986), the insured had a portable sump pump in a sump pit in the basement of his home; its purpose was to remove subsurface water when it reached the level of the sump pit. A water damage loss occurred when the hose of the pump—connected to a drain in the basement—ruptured.

 

The insurer denied the claim, stating that the loss was due to underground water and was thereby excluded. The trial court concurred. On appeal, however, a New Jersey superior court concluded that the sump pump, being portable and plugging into a standard wall outlet, was “no less an appliance than a dehumidifier or a washing machine.” Considered in that fashion, coverage from the accidental discharge peril could apply to the water damage caused by the underground water released from the sump pump's ruptured hose.

 

In the Stone case, the underground water, an excluded peril, started the loss-producing chain of causation, but the last event, the ruptured hose on the appliance, was a covered risk.

 

However, current wording in the ISO homeowners forms, wherein the accidental discharge peril is revised to state that a “plumbing system or household appliance does not include a sump, sump pump or related equipment,” would seem to preclude coverage in all but a few instances. Even when an additional modification in the water exclusion is examined—this one removing coverage for water that overflows from a sump (defined as a pit or reservoir)—there is still potential coverage under the homeowners policy for accidental discharge or overflow losses that result from rupture of a sump pump hose.

 

As regards leakage from plumbing systems located below the surface of the ground, the following cases illustrate that courts have generally found coverage under the accidental discharge peril rather than applying the underground water exclusion.

 

One case, Cantanucci v. Reliance Ins. Co., 364 N.Y.S.2d 890 (1974), involved damage to a foundation wall from a ruptured underground sewer line that caused the foundation to sag and crack. The policy covered loss by accidental discharge of water from within a plumbing system but excluded loss by water below the surface of the ground. The court agreed with the insured that the sewer line was a part of the plumbing system but felt the water exclusion precluded coverage. On appeal, however, the decision was reversed. The appellate court reasoned that by rejecting coverage the insurer is in effect insuring a leak from a plumbing system above ground but excluding coverage for a leak below the ground. Had such a distinction been intended, the court said, the insurer could have used specific language to that effect.

 

In Hartford Accident and Indemnity Co. v. Phelps, 294 So. 2d 362 (Fla. App. 1974), damage resulted from an underground leak in a water pipe under the concrete slab floor of the insured's house. This claim involved a homeowners “open perils” policy, and a Florida appeals court determined that the loss was covered as an accidental discharge from within a plumbing system rather than excluded as below-surface water or settling.

 

In a similar case, the Mississippi Supreme Court found coverage for damage to the floors and walls of a house caused by water escaping from a pipe beneath the terrazzo floor. Here again, the policy was open perils homeowners, and the court held that the water exclusion clause was limited to natural underground water and did not apply to below-surface plumbing systems. The case is The New Hampshire Ins. Co. v. Robertson, 352 So. 2d 1307 (Miss. 1977).

 

And in Gatti v. The Hanover Ins. Co., 601 F. Supp. 210 (D.C. Pa. 1985), a large amount of water leaked into the ground from the underground pipes of an apartment complex after it passed through the insured's water meter. The insured submitted a claim against its special businessowners policy for loss totaling $39,523.44, the amount necessary to repair the pipes and pay the cost and interest charges to the city of Philadelphia for the water used.

 

The insurer attempted to deny the claim based on the following points: (1) pecuniary detriment is not the same as physical loss; (2) water is not personal property of the insured; and (3) water below the surface of the ground emanating outside the insured's own system is excluded. The court rejected the insurer's contentions stating first that if property damage includes loss of use, then the physical loss of the water in this case certainly also includes damages from its loss of use. Further, the water itself became the property of the insured when it passed through the water meter on the insured's premises and fit the definition of personal property of the businessowners policy in that it was used for the “service of the building.”

 

And finally, once the water passed through the water meter, it entered the insured's own plumbing system. Since the leakage was in the insured's own plumbing system, the loss was not reached by the underground water exclusion.

 

Note that “water” has since been added to the list of excluded property on ISO commercial property forms and homeowners policies.

 

Natural versus Artificial

 

A number of court decisions regarding the water exclusion are significant because of the distinction made between damage caused by water occurring in a natural state and that caused by water directed by accidental or artificial means. Many courts conclude that the exclusion is intended to preclude coverage only for loss caused by natural circumstances, not by artificial means. For example, in Ferndale Development Co., Inc. v. Great American Ins. Co., 527 P.2d 939 (Colo. App. 1974), damage was caused by water from a bursting water main. The insurer declined coverage on the basis of that part of the exclusion relating to flood, surface water, and the like. The court rejected this reasoning and concluded that the terms “flood” and “surface water, as used in this policy, do not include water escaping from a burst water main. This case was used by the insureds in Kane, discussed previously, as a reason to provide coverage. However, as noted, the judge distinguished between a flood as being a large-scale overflow from a body of water and the escape of water from a man-made object.

 

Slightly different reasoning was used by the Texas court of appeals in Adrian Associates General Contractors v. National Surety Corp., 638 S.W.2d 138 (Tex. App. 1982). An underground water main owned by a municipality ruptured causing water to migrate underground and undermine a concrete slab Adrian Associates had poured as the foundation for a warehouse. The slab settled to the extent that it had to be pulled out and replaced. Settlement of a foundation was excluded but not if the settlement was caused by an insured peril.

 

The court reasoned that the water exclusion was broken into three parts. The first part excluded flood, surface water, waves, tides, tidal waves, overflow of any body of water, or their spray, all whether driven by wind or not; and water that derives from natural sources. The second part excluded water that backs up from a sewer or drain and water that derives from an artificial source. The third part excludes water under the ground surface pressing on, or flowing or seeping through various parts of a building without specifying its derivation.

 

The insurance company agreed that the policy was “designed to exclude certain water of natural origin and certain water of artificial origin” in the first two parts. It followed, according to the court that since the third part did not specify the origin of the water, it was ambiguous and should be interpreted in favor of the insured. “Although the insurance company contends that the exclusion applies to all water whatever the source, we conclude that if it was intended to do so, then the insurance company would have said so.”

 

In Jones v. Columbia Mut., 700 S.W.2d 187 (Mo. App. 1985), the insured's pipes froze and broke, releasing water that damaged not only the interior of the house but also the foundation and carport. The insurer agreed to pay for the interior water damage but refused payment for the damage done by the water that went underground.

 

The trial court agreed with the insurer, but a Missouri court of appeals reversed the decision after examining the exclusions regarding water damage and the coverage provided by the freezing of plumbing peril. The court felt that “the only reasonable way left to reconcile [the water damage exclusion] with the peril of freezing of a plumbing system is to find that they mean that any water damage directly resulting from such freezing is covered, and that [the underground water exclusion] applies only to damage resulting from natural water, or perhaps other sources.” This case also illustrates that a similar conflict exists between the underground water exclusion and the named peril of freezing of a plumbing system as exists between that exclusion and the accidental discharge peril, discussed earlier.

 

There are many cases where a water damage exclusion was held not to apply, and there are many cases reaching the opposite result. See, McDonough v. Hardware Dealers Mutual Fire Ins. Co., 448 F.2d. 870 (1st Cir.1971) (water used in fighting fire held a natural consequence of a fire, and thus covered); but see Bergeron v. State Fire & Casualty Co., 766 A.2d 256 (N.J. 2000)(failure of dammed pond due to defective design held not a covered event). The critical question is whether there was a reasonable expectation that an “all risk” homeowner's policy would include flood insurance. In the absence of evidence that it is typical to include such coverage, the circuit court concluded that it was not, and the appellate court agreed.

 

Water versus Ice

 

An additional distinction has been made in some courts between water in its natural liquid state and water that has been frozen into solid ice. In Nationwide Ins. Co. v. Warren, 675 S.W.2d 402 (1984), the insured suffered damage to a warehouse floor caused by ice and a frost heave. The appeals court invoked the rule of ejusdem generis (a general term that is found to be ambiguous should be interpreted in light of the specific terms preceding [or following] it), reasoning that because ice does not flow, seep, or leak, it is not of the same general class as water.

 

The court further said, “Ice and water…are not the same thing. It was not water in a natural, liquid state which caused the damage to the warehouse; it was ice, creating pressure, which caused the damage. The exclusion makes no reference to ice, although it could easily have done so.”

 

In Ercolani v. Excelsior Ins. Co., 830 F.2d 31 (3d Cir. 1987), the foundation of Ercolani's home yielded to hydrostatic pressure from underground water that had frozen during extremely cold weather. The court found that coverage for the subsequent collapse of the house was not barred by the water exclusion because the normal hydrostatic pressure of water against the foundation was increased by the expansion of the ice as the ground water froze. The court cited Warren and referred to the American College Dictionary definition of “water” as “1. The liquid which in a more or less impure state constitutes rain, oceans, lakes, rivers, etc., and which in pure state is a transparent, odorless, tasteless liquid, a compound of hydrogen and oxygen, H2O, freezing at 32 degrees F. or 0 degrees C.” The court added, “Applying the water damage exclusion to water in its liquid state and not to its expanded frozen state gives effect, it seems to us, to the 'objectively reasonable expectations' of an insured.”

 

Water Exclusion Endorsement

 

In 2008 ISO filed a water exclusion endorsement, CP 10 32 08 08, for use with the commercial property coverage part and the standard property policy. ISO considered issues raised in litigation following Hurricane Katrina surrounding the definition of flood when it created the endorsement.

 

ISO also cited Sher v. Lafayette Ins. Co., 988 So.2d 186 (La.2008), as playing a part in the creation of the endorsement. The court in that case stated, “The plain, ordinary and generally prevailing meaning of the word 'flood' is the overflow of a body of water causing a large amount of water to cover an area that is usually dry. This definition does not depend on locality, culture, or even national origin – the entire English speaking world recognizes that a flood is the overflow of a body of water causing a large amount of water to cover an area that is usually dry land. Contrary to the court of appeal's reasoning, this definition does not change or depend on whether the event is a natural disaster or a manmade one— in either case, a large amount of water covers an area that is usually dry.”

 

The endorsement states that the listed water losses are excluded “regardless of whether any [of the causes of loss] is caused by an act of nature or is otherwise caused.” An example of a dam or levee breaking and failing to contain water is included to illustrate a situation to which the exclusion applies. The form also adds tsunami and storm surge to the list of excluded causes of loss to reinforce the exclusion of such events. The provisions of the endorsement were incorporated into the causes of loss forms in the 2012 revision.

 

This premium content is locked for FC&S Coverage Interpretation Subscribers

Enjoy unlimited access to the trusted solution for successful interpretation and analyses of complex insurance policies.

  • Quality content from industry experts with over 60 years insurance experience, combined
  • Customizable alerts of changes in relevant policies and trends
  • Search and navigate Q&As to find answers to your specific questions
  • Filter by article, discussion, analysis and more to find the exact information you’re looking for
  • Continually updated to bring you the latest reports, trending topics, and coverage analysis