Sanford J. Berger; Lulu Wills; Philip & Eleanor Rees; Andgerrit Joosten, Plaintiffs-appellants, v. Samuel Pierce; et al., Defendants-appellees, 983 F.2d 1065 (6th Cir. 1992)

In this appeal, we revisit a question concerning the interpretation to be given a 1973 amendment to the National Flood Insurance Act, 42 U.S.C. §§ 4001-4129 ("the Act"). We addressed that amendment and remanded this cause to the district court for further fact finding in Berger v. Pierce, 933 F.2d 393 (6th Cir. 1991) ("Berger I ").

The Act created a federally subsidized program that provides flood insurance. Congress created the program because private insurance companies had been unable to write flood insurance policies that were economically feasible. In 1973, Congress broadened the scope of the Act by including within the definition of "flood" the "collapse or subsidence of land along the shore of a lake or other body of water as a result of erosion or undermining caused by waves or currents of water exceeding anticipated cyclical levels." 42 U.S.C. § 4121(c).

Plaintiffs are property owners along Lake Erie, in Willoughby, Ohio. They were issued standard flood insurance policies by the Federal Insurance Administration ("FIA"). The shore side of the land upon which plaintiffs' homes were located overlooked the beach and was bounded by a cliff. Under normal conditions, the beach prevented lake waters from touching the base of the cliff. Plaintiffs contend that during 1986 and into 1987, Lake Erie continually exceeded its anticipated level to such an extent that the beach was flooded over and waves from the lake eroded and undermined the cliff supporting their homes. When their homes became uninhabitable during 1987, they sought recovery under their flood insurance policies. When the FIA denied their claims, they filed this lawsuit seeking a declaratory judgment that certain provisions of the insurance contract upon which their claims were denied are contrary to the Act.

The FIA, citing legislative history, had argued before the district court that

Congress intended that erosion only be covered when it was caused by a sudden and unexpected force of nature as opposed to the normal action of tides. The words "exceeding anticipated cyclical levels" are indicative of an intent to limit recovery under the [flood insurance program] to losses caused by abnormal erosion.

Berger I, 933 F.2d at 399. The district court agreed and dismissed plaintiffs' claims on the basis that their loss was caused by "gradual erosion," which was not covered under the flood insurance policies. The district court stated that " [i]t is clear from the [regulations'] definition that flood related erosion is a covered loss. Gradual erosion, however, is not." Id.

We held that the district court had erred:

Congress did not create an exception for gradual erosion. Congress directed that policy coverage turn on "waves" or "currents" that "exceed [ ] anticipated cyclical levels." Until there is an examination of the facts with attention to whether the "erosion or undermining" was caused by "waves or currents of water exceeding anticipated cyclical levels," there can be no determination as to whether or not the property was damaged in a manner compensable under the statute or the contract.

....

... While the FIA's argument, and the district court's analysis of it, are not entirely clear, they appear merely to beg the question. What Congress intended to cover was (1) damage caused by shoreline erosion, (2) resulting from cyclical water levels exceeding those anticipated.... We agree with the FIA that Congress intended to draw a distinction between usual and unusual events. However, that distinction is entirely satisfied when it is determined whether or not the erosion which caused the insured's loss was caused by water "exceeding anticipated cyclical levels." Nothing more sudden or unusual is necessary.

Id. at 398-99.

Since it was undisputed that the policyholders' loss was the result of shoreline erosion, the district court's task upon remand was to determine whether that shoreline erosion resulted from cyclical water levels exceeding those anticipated. Upon remand, however, the FIA once again asserted what it said was the intent of Congress, which was to require proof of "waves or currents which are severe or of unanticipated force ... an unusual occurrence, more than high lake water levels." The district court once again accepted these arguments, holding that the erosion damage must be "caused by an unusually high water level ... accompanied by a severe storm or an unanticipated force of nature." Berger v. Pierce, No. 1:88CV1147 at 11 (N.D. Ohio Aug. 30, 1991).

Both the FIA and the district court improperly relied upon legislative history¹  to reach a conclusion contrary to the law of this case, as can be seen from our opinion in Berger I quoted above. Plaintiffs were not required to prove that a severe storm or other unanticipated force of nature was a cause of their erosion loss, so long as the erosion was occasioned by the natural action of the waves or currents in the lake at a time when its water²  exceeded anticipated cyclical levels.

The factual determinations we sought from the district court were (1) the "anticipated cyclical level" of Lake Erie's waters during the time the policyholders suffered the shoreline erosion damage (i.e., April and May of 1987), and (2) the actual level of Lake Erie's waters during that same time.

Although the district court received evidence from both sides, it did not identify either the anticipated or actual lake levels for April and May of 1987. Nevertheless, it concluded that the policyholders had not proved that their erosion damage had been caused by waves or currents of water exceeding anticipated cyclical levels. We are unable to find any evidence in the record that would support such a conclusion.

Ordinarily, such error would require yet another remand to the district court. However, after thoroughly reviewing the record before us and the arguments of counsel, we conclude that all the evidence in the record demonstrates that Lake Erie did exceed its anticipated cyclical level at the time it caused the policyholders' erosion damage.

The policyholders presented the district court with the Monthly Bulletin of Lake Levels for the Great Lakes, a detailed report from the United States Army Corps of Engineers. This report averaged data from 1900 to 1986 to show the historic average level of Lake Erie during each month of the year. It also reported the actual lake levels for April and May of 1987. This data demonstrated that the actual water levels for April and May of 1987 exceeded the 1900-1986 historic averages for April and May by 2.1 and 1.85 feet, respectively. The policyholders cite this report as proof that Lake Erie exceeded its anticipated cyclical levels during this time.

The FIA does not dispute the accuracy of this data. Instead, it argues that the benchmark for "anticipated cyclical levels" should not be long-term historical averages, but data from 1973, the year that the amendment to the Act was enacted. When pressed on this point during oral argument however, counsel for the FIA could cite no evidence in the record establishing 1973 as the benchmark for anticipated cyclical levels. In fact, the record shows that in the nineteen years since Congress enacted the amendment, neither the FIA nor FEMA have ever authoritatively defined "anticipated cyclical levels."

The FIA's position that a single year's water levels should be viewed as "anticipated cyclical levels" merely on the basis that legislation was passed during that year, is not persuasive. Since 1973 was a year of unusually high water levels, we doubt that Congress intended those levels to serve as the benchmark for anticipating future norms. Moreover, data for a single year provide only a statistical "snapshot," rather than the broad overview that one would expect to be necessary in order to anticipate future events, such as "cyclical levels," with confidence.

The 1900-1986 averages, on the other hand, do indicate average water levels over a long period of time. In addition, the policyholders introduced several FEMA policy statements which referred to long-term lake levels while addressing inquiries about anticipated cyclical levels. For these reasons, we conclude that all the evidence in the record demonstrates that the computation by the Army Corps of Engineers of 1900-1986 average lake levels provides the appropriate benchmark for determining the "anticipated cyclical levels" of Lake Erie in April and May of 1987.

Since it is undisputed that the actual water levels in Lake Erie exceeded the 1900-1986 averages by 2.1 feet in April 1987 and 1.85 feet in May, the record is equally clear that the water level in Lake Erie exceeded anticipated cyclical levels at the time when shoreline erosion caused the policyholders' damage.

Given the circumstances of this case as well as the facts present in the record, the only conclusion available to us is that the policyholders have indeed proved that their erosion damage was caused by "waves or currents of water exceeding anticipated cyclical levels." 42 U.S.C. § 4121(c). They are therefore entitled to compensation under the terms of their standard flood insurance policies.

The order of the district court is reversed, and this cause is remanded to the district court with instructions to enter a declaratory judgment in favor of the policyholders to the effect that the FIA interpretation of 42 U.S.C. § 4121(c) and of the exclusion provision in its policy relative to gradual erosion are contrary to law, and are not a valid basis for denying the policyholders' claims under the circumstances of this case. The district court also is directed to reconsider the policyholders' request for attorneys' fees, in light of our decision today.