Downey v. State Farm Fire & Casualty Co.

Sep 17, 2001
00-3473
Reporters:
,
Before:
Easterbrook, Rovner, Diane P. Wood

Before Easterbrook, Rovner, and Diane P. Wood, Circuit Judges.

Easterbrook, Circuit Judge.

Michael Downey lives on a hill in Peoria, Illinois. His back yard runs downward at a 35 angle, creating a dangr of soil erosion that could compromise the foundation of his house. A retaining wall supported the soil, but in February 1997 heavy rain washed away the wall and much of the soil that it had been retaining. This in turn caused the house‘s foundation to shift and become unstable. Fortunately (or so he thought) Downey had purchased flood insurance. State Farm Fire & Casualty Co., from which Downey bought the policy, paid to fix cracks in the foundation but denied indemnity for the expense of stabilizing the house to ward off collapse. Injury caused by the failure of the retaining wall, State Farm asserted, is excluded from coverage. State Farm lost in the district court and on appeal challenges the district judge‘s interpretation of the policy. Before reaching the merits, however, we must consider both subject-matter and appellate jurisdiction.

Our ears pricked up at the assertion that this suit belongs in federal court, for a contract dispute between two private parties typically does not arise “under the Constitution, laws, or treaties of the United States“, 28 U.S.C. sec. 1331, and the complaint does not allege diversity of citizenship. State Farm‘s brief asserts that federal-question jurisdiction exists but does not explain why; Downey concurred in State Farm‘s presentation. Because the presentations at oral argument were unilluminating, we directed the parties to file supplemental memoranda addressing jurisdictional issues. Their responses focus on the nature of the insurance. Downey bought his policy through the National Flood Insurance Program (nfip), codified at 42 U.S.C. sec. sec. 4001-4129. This national system of flood insurance for residents of high-risk areas regulates the transactions between Downey and State Farm, and the parties offer three reasons why the upshot is a question within federal jurisdiction.

State Farm points to an explicit grant of jurisdiction in 42 U.S.C. sec. 4053:

[U]pon the disallowance by any such company or other insurer of any such claim . . . the claimant, within one year after the date of mailing of notice of disallowance or partial disallowance of the claim, may institute an action on such claim against such company or other insurer in the United States district court for the district in which the insured property or the major part thereof shall have been situated, and original exclusive jurisdiction is hereby conferred upon such court to hear and determine such action without regard to the amount in controversy.

Although Downey is a “claimant” and State Farm an “insurer“, Downey‘s action against State Farm is not a “claim” under sec. 4053--State Farm looked at the wrong part of the statute. When Congress created the nfip it gave the program‘s administrator two ways to execute the program and discretion to choose between them. The first method, the “Industry Program,” allows a pool of private insurers to underwrite flood insurance with financial backing from the government. See 42 U.S.C. sec. sec. 4051-56. The “Government Program,” the second option, allows the government to run the nfip itself--offering federally underwritten policies-- with the potential for administrative assistance from private insurers. See 42 U.S.C. sec. sec. 4071-72. See Edward T. Pasterick, The National Flood Insurance Program, in Howard Kunreuther & Richard J. Roth, Sr., eds., Paying the Price: The Status and Role of Insurance Against Natural Disasters in the United States (1998), for an explanation of the nfip. In 1977 the Secretary of Housing and Urban Development, who ran the nfip at the time (it has since been taken over by the Federal Emergency Management Agency), decided that the Industry Program was unworkable and ended it. He then implemented the Government Program, which has continued to the present. Section 4053, the grant of jurisdiction to which State Farm points, enables only claims brought “under this part“--the nfip as run under the Industry Program. State Farm, then, is 24 years too late to take advantage of sec. 4053. Courts occasionally make the same error. See Froelich v. Catawba Insurance Co., 10 F. Supp. 2d 597 (W.D. Va. 1998); Gagliardi v. Omaha Property & Insurance Co., 952 F. Supp. 212 (D. N.J. 1997). Like the eleventh circuit, Newton v. Capital Assurance Co., 245 F.3d 1306 (2001), we avoid this pitfall.

Downey observes that the Government Program has its own jurisdictional provision, 42 U.S.C. sec. 4072:

In the event the program is carried out as provided in section 4071 of this title, the Director [of fema] shall be authorized to adjust and make payment of any claims for proved and approved losses covered by flood insurance, and upon the disallowance by the Director of any such claim . . . the claimant . . . may institute an action against the Director on such claim in the United States district court for the district in which the insured property or the major part thereof shall have been situated, and original exclusive jurisdiction is hereby conferred upon such court to hear and determine such action without regard to the amount in controversy.

Yet this section allows only “an action against the Director“. Downey sued State Farm. He might have thought that State Farm is the only proper defendant: In 1983 fema created the Write-Your-Own Program (wyop), which allows private insurers to issue and administer flood-risk policies under the Government Program. The private insurers also defend suits arising from the policies. 44 C.F.R. sec. 62.23(d). Perhaps Downey figured that, because he contracted with State Farm, he had to sue State Farm. Neither party appears to have noticed 44 C.F.R. sec. 62.22, which permits suits against the Director of fema arising from decisions made by wyop insurance companies. This regulation effectively allows a direct action against the person who is ultimately responsible, rather than the wielder of delegated authority. But Downey sued State Farm rather than the Director and is stuck with that choice. Section 4072 does not mention the wyop or indicate that anyone other than the Director may be sued under this grant of jurisdiction.

Nonetheless, Downey insists, with the support of Van Holt v. Liberty Mutual Fire Insurance Co., 163 F.3d 161 (3d Cir. 1998), that, because “a suit against a wyo company is the functional equivalent of a suit against fema“, we should look past the caption of this case and pretend that the Director is the defendant. This position is not without force: fema provides a standard text for all nfip policies and forbids wyop companies from making changes; fema‘s interpretations of the policy bind all wyop participants; fema decides what rates may be charged; all premiums are remitted on to fema (minus a small fee); if wyop companies pay out on a claim they get reimbursed by fema; likewise with litigation costs. See generally 42 U.S.C. sec. 4081; 44 C.F.R. sec. sec. 62.23-62.24. So although private insurers issue the policies, fema underwrites the risk. The insurance companies handle administrative business for fema by selling policies and processing claims but do little else (unlike the Industry Program, where the private companies underwrite the risks). Arrangements like this make sense. fema likely is unsuited to tasks such as selling insurance and collecting fees, and even less adept at processing individual claims for flood damage. By purchasing the services of a more efficient claims processor, fema saves money. We see a similar structure in the Medicare program: Health care providers seek reimbursement, not directly from the government but from “fiscal intermediaries“--usually private insurance companies--that act as the claims processor in the government‘s stead. See Your Home Visiting Nurse Services, Inc. v. Shalala, 525 U.S. 449 (1999).

In a sense, then, State Farm is a place holder for fema, but does this fact have jurisdictional significance? Downey might have something if for jurisdictional purposes courts typically look to see who will be affected by a decision; but we don‘t. This would be clear enough in an ordinary tort dispute between two Illinois citizens. If the plaintiff in such a suit agreed to pay any proceeds from the judgment to an out-of-state insurance company (who, let‘s say, in return agreed to pay his medical bills), would a court peek behind the formality of the non-diverse parties and recognize that those who truly have something to gain or lose--the insurance company and the Illinois defendant--are diverse? Surely not. Nor do courts look past corporate form to the citizenship of the shareholders or other investors. There is a special rule for administrators of estates, 28 U.S.C. sec. 1332(c)(2), but normally the status of the named litigant governs--provided that the litigant is an entity rather than a name for an unincorporated association such as a partnership. See Carden v. Arkoma Associates, 494 U.S. 185 (1990); Indiana Gas Co. v. Home Insurance Co., 141 F.3d 314 (7th Cir. 1998). For example, the citizenship of a trustee rather than the trust beneficiary is dispositive under sec. 1332. Navarro Savings Association v. Lee, 446 U.S. 458 (1980). Downey cannot escape the same conclusion: Although a judgment against State Farm may come out of the federal treasury--creating a federal interest--the only litigants are in the private sector. Because we see no good reason to disregard not only the identity of the litigants but also the fact that sec. 4072 is limited to suits against the Director, we decline to adopt Van Holt‘s reasoning. (For purposes other than jurisdiction the economic incidence of the decision sometimes matters. Compare Montana v. United States, 440 U.S. 147 (1979) (for purposes of claim preclusion courts may look behind the nominal parties to a suit), with Regents v. Doe, 519 U.S. 425 (1997) (for purposes of sovereign immunity courts should look exclusively at the parties).)

This is not the end of the jurisdictional inquiry, however. Sometimes the federal interest in a controversy is so dominant that federal law applies--activating federal-question jurisdiction under sec. 1331--even if the national government is not a party. See National Farmers Union Insurance Cos. v. Crow Tribe of Indians, 471 U.S. 845 (1985). The fifth circuit has held, see West v. Harris, 573 F.2d 873 (1978), that because the nfip is a federal program, uniform judicial interpretations of the standard insurance policies are necessary. Every other circuit that has considered this issue has followed West‘s approach either explicitly or implicitly. See Linder & Associates, Inc. v. Aetna Casualty & Surety Co., 166 F.3d 547 (3d Cir. 1999); Flick v. Liberty Mutual Fire Insurance Co., 205 F.3d 386 (9th Cir. 2000); Newton, supra (11th Cir.). Cf. Atlas Pallet, Inc. v. Gallagher, 725 F.2d 131 (1st Cir. 1984); Leland v. Federal Insurance Administrator, 934 F.2d 524 (4th Cir. 1991); Berger v. Pierce, 933 F.2d 393 (6th Cir. 1991); Nelson v. Becton, 929 F.2d 1287 (8th Cir. 1991). We assumed the same result in Sodowski v. National Flood Insurance Program, 834 F.2d 653 (7th Cir. 1987), and now so hold.

In 1978, when West was decided, most judges assumed a nation-wide program automatically leads to federal common law. Atherton v. FDIC, 519 U.S. 213 (1997), has complicated matters. Atherton considered the question whether federal law provides the standard of care in derivative litigation involving the directors and officers of a federally chartered lending institution. Although a federal statute was also at issue, the Court first considered and rejected the proposition that federal common law could displace the state rule for the internal affairs of corporate entities, holding that, although an imperative of uniformity is enough to call for federal common law, the federal nature of a program alone does not demonstrate a need for uniformity. That analysis applies equally here. Neither the parties to this appeal nor any case we could find articulate any reason for uniformity of interpretation other than the very one Atherton rejected. None even takes notice of Atherton.

Not proscribed by Atherton is a narrower ground for applying federal common law. Clearfield Trust Co. v. United States, 318 U.S. 363 (1943), establishes that, when the duties or rights of the United States are at stake under a federal program, that federal interest requires the application (and if necessary the creation) of federal law. For example, United States v. Kimbell Foods, Inc., 440 U.S. 715 (1979), considered what law governs the priority of liens that secure federal agencies’ loans to private parties. The Court reasoned, citing Clearfield Trust, that the liens implicated the agencies’ rights under federal programs created by “specific Acts of Congress“, which meant that federal law had to apply--though the Court then borrowed that federal law from state law. See also United States v. Little Lake Misere Land Co., 412 U.S. 580 (1973). Typically the United States is a party to disputes that call for federal law under the approach of Clearfield Trust. See, e.g., Priebe & Sons v. United States, 332 U.S. 407 (1947); United States v. Standard Oil Co., 332 U.S. 301 (1947). But there is no reason to view that condition as a necessity. If fema were the defendant in our case, we would have no doubt that federal law applied: Just like the agencies in Kimbell Foods, fema runs a federal program, and because it bears the risk on all nfip contracts, fema‘s duties are at issue whenever an nfip policy is interpreted. Replacing “fema” with “State Farm” in the caption of the case changes nothing; a judgment against State Farm and a judgment against fema have identical effects: fema pays. And though we were concerned with the formal parties to this action when we interpreted sec. 4072, here we are concerned with the federal interest invoked by the dispute‘s subject matter.

The federal interest is no less here than in Turner/ Ozanne v. Hyman/Power, 111 F.3d 1312 (7th Cir. 1997), which applied federal common law to a dispute between two contractors in a construction project for the United States Postal Service. Hyman/Power, the general contractor, agreed in its contract with the usps to indemnify the agency for any on-the-job injuries. Turner/Ozanne, the usps‘s on-site overseer for the project, then sought indemnification from Hyman/Power on the theory that as a “representative” of the usps it should be treated as the usps. Although the dispute affected only private rights--the Postal Service had nothing to lose or gain--we held that, because Turner/Ozanne invoked a protection it had negotiated with the United States and because both parties were involved in an ongoing federal project, a federal question was present. Today‘s disagreement presents both characteristics-- Downey contends that he bought protection from the government through a federal program--and here the government coffer is at risk. That is enough to justify the application of federal law to the dispute, which means that subject-matter jurisdiction arises under sec. 1331.

Having assured ourselves that this dispute was properly before the district court, we now must inquire whether State Farm can ask us for relief. In two orders the district court held that the policy covers Downey‘s claim. At this point only the calculation of damages remained for the district court to accomplish. State Farm then offered to allow judgment in Downey‘s favor in the amount of $186,360.54. See Fed. R. Civ. P. 68. Downey accepted, and at the parties’ joint request the district court then entered a final judgment against State Farm expressly reserving “State Farm‘s right to pursue an appeal from the [liability] orders of this Court“.

An agreement among the parties to enter a judgment may create nothing adverse from which to appeal. How can State Farm contend that it is aggrieved by a judgment that it consented to? Appeals are taken not from issues but from judgments. See California v. Rooney, 483 U.S. 307 (1987). Parties often stipulate to issues such as damages once the district court resolves liability, but an agreement on a specific issue differs from asking the court to enter judgment, which winds up the case itself. In criminal cases courts allow conditional guilty pleas (with the district judge‘s consent) followed by appeal on a reserved issue, because Fed. R. Crim. P. 11(a)(2) expressly allows such a tactic, but the civil rules do not have a parallel provision. One might think, therefore, that if a judgment is not contested in the district court then the adversarial process has ended and the court of appeals has no role to play. Yet for jurisdictional purposes there is no distinction between “consent” and “adversarial” judgments. Judgments are judgments, and any party can appeal as of right from a final decision adverse to his interests. So says 28 U.S.C. sec. 1291, which allows appeal from “all final decisions of the district courts“. Finality is the necessary and sufficient condition. Distinguishing between final judgments entered with the consent of both parties and final judgments entered against one party‘s wishes would create an extrastatutory condition on appeal. This has little to recommend it, and the possibility has been rejected by the Supreme Court. See Pacific R.R. v. Ketchum, 101 U.S. 289, 295 (1880). Ketchum interpreted an earlier version of the statute, but the critical language has survived. See INB Banking Co. v. Iron Peddlers, Inc., 993 F.2d 1291 (7th Cir. 1993), applying Ketchum to the current version of sec. 1291.

State Farm is not home free, however. Although the Supreme Court has held that “consent judgments” are final and appealable under sec. 1291 (so appellate jurisdiction is secure) the Court has added that the act of giving consent usually waives the consenting party‘s right to review, leading to affirmance “without considering the merits of the cause.” Nashville, Chattanooga & St. Louis Ry. v. United States, 113 U.S. 261, 266 (1885). See also Swift & Co. v. United States, 276 U.S. 311 (1928); United States v. Babbitt, 104 U.S. 767 (1882); Association of Community Organizations for Reform Now v. Edgar, 99 F.3d 261 (7th Cir. 1996). Waiver affects, not a court‘s power to hear the case, but whether as a practical matter it has any job to do. So did State Farm waive its right to appellate consideration? Both the Offer of Judgment and the district court‘s judgment reserved State Farm‘s right to challenge the liability determination. A reservation of rights is incompatible with waiver. See Cutting v. Jerome Foods, Inc., 993 F.2d 1293 (7th Cir. 1993); Hudson v. Chicago Teachers Union, 922 F.2d 1306 (7th Cir. 1991). Almost every circuit that has considered the issue has held that an express reservation of the right to appeal avoids waiver of contested issues that had been resolved earlier in the litigation. See BIW Deceived v. Local S6, 132 F.3d 824 (1st Cir. 1997); Keefe v. Prudential Property & Casualty Insurance Co., 203 F.3d 218 (3d Cir. 2000); Cohen v. Virginia Electric & Power Co., 788 F.2d 247 (4th Cir. 1986); Slaven v. American Trading Transportation Co., 146 F.3d 1066 (9th Cir. 1998); Mock v. T.G. & Y. Stores Co., 971 F.2d 522 (10th Cir. 1992); Shores v. Sklar, 885 F.2d 760 (11th Cir. 1989) (en banc). Only the fifth circuit gives no effect to an express reservation of appellate rights. See Amstar Corp. v. Southern Pacific Transport Co., 607 F.2d 1100 (5th Cir. 1979). Amstar, however, offered no explanation of its holding and so gives us no reason to doubt our own conclusion: State Farm preserved its rights, and we may reach the merits.

The parties came to blows over many issues in the district court, but on appeal only one dispute remains: Does Downey‘s nfip policy cover the cost of shoring up the soil around his house after the retaining wall failed at the task? The 1997 storm (which the parties now agree caused a “flood” under the policy) washed away the retaining wall and a lot of soil. According to David Maurer, a structural engineer whose opinion State Farm does not contest, this removal of soil from the northwest corner of the house led to cracks in the foundation and caused the western exterior wall to tilt outward. Because that wall bears the weight of the second story bedroom, the tilt created, in Maurer‘s opinion, a “danger of partial collapse unless the movement [was] stopped.” Maurer recommended the changes that Downey implemented and for which he now seeks reimbursement-- installing gabion baskets and rocks in the hillside and injecting grout into the ground underneath the house. State Farm paid to fix the cracks in the house but insists that rendering the house stable and safe for occupancy is outside the scope of the contract.

The nfip policy covers only Downey‘s “dwelling” and explicitly excludes from coverage “[f]ences, retaining walls, seawalls, bulkheads, wharves, piers, bridges, and docks.” (Emphasis added.) State Farm argues that this clause relieves it of obligation to indemnify not only damage to a retaining wall but also damage to a house caused by damage to a retaining wall. The district court found little to support this reading. Neither do we. The policy covers “any loss [to the dwelling] in the nature of actual loss of or physical damage, evidenced by physical changes, to the insured property . . . which is directly and proximately caused by a flood“. (Emphasis deleted.) State Farm does not contend that the “physical changes” to the house were not symptomatic of “physical damage” and likewise makes no argument that the flood was not a proximate cause of that damage. End of story. If there is physical damage to the house, and that damage was caused by a flood, how can the policy not provide coverage? State Farm does not contend that the repairs Downey made were unjustifiable or excessive (for the parties settled all disputes about the amount of indemnity, if any is available). The retaining-wall exclusion is irrelevant. It puts no limitation on what types of flood damage to a house are covered. Had Downey installed a new retaining wall, the policy would not cover the expense. But he didn‘t; he fixed his house.

This understanding still leaves meaning in the retaining-wall exclusion: If the flood caused damage to the retaining wall with no loss of stability to the house, the policy would not cover the loss. If the retaining wall had supported a barn rather than Downey‘s bedroom, the policy would not cover the loss. Remember that this policy is a standard form for use by all nfip participants. It must, therefore address as many complications as possible; not all provisions will be relevant to every property owner. So the exclusion has plenty of work to do--just not in this situation. There is, however, no more work for us to do: the repairs Downey undertook to stabilize his house are covered by the policy, and State Farm must pay up. The parties settled all other differences in the district court.

Affirmed