Allan E. RUD, Plaintiff-Appellant, v. LIBERTY LIFE ASSURANCE COMPANY OF BOSTON, Defendant-Appellee, 438 F.3d 772

POSNER, Circuit Judge. Allan Rud, a factory worker employed by Andersen Windows, Inc., fell and seriously injured his back. Andersen, which sponsors and administers a welfare benefits plan for its employees, had bought an insurance policy from Liberty Life, whereby the latter would process and pay disability claims under the plan. Rud submitted to Liberty Life a claim for permanent disability benefits.

The insurance policy provided that an employee would be deemed permanently disabled for the 24 months following the onset of the disability if he couldn’t do his former work, but that to obtain permanent-disability benefits beyond that period he had to be unable to perform the duties of any job. Rud received benefits for the full 24 months and indeed for five more months as the insurance company considered his entitlement to additional benefits.

But eventually the company concluded that although Rud couldn’t return to his former, strenuous work (which is why he had received permanent disability benefits for the first 24 months after his fall), he could perform the duties of a variety of other jobs, of a light or sedentary character. Rud disagreed, and filed suit against Liberty Life in a state court, charging breach of contract under state law but also violation of ERISA. He did not sue Andersen Windows. Liberty Life removed the suit to federal district court. The judge gave

summary judgment for Liberty Life, and Rud appeals. There is no doubt that Liberty Life’s determination that Rud was capable of performing light or sedentary work despite his back problem was reasonable; that is, the determination was not so off the wall that it could be adjudged “arbitrary and capricious.” And that is the correct standard when an ERISA plan provides that the benefits determination is within the discretion of the ERISA plan’s administrator or some other ERISA fiduciary, Firestone Tire & Rubber Co. v.

Bruch, 489 U.S. 101, 111-14, 109 S.Ct. 948, 103 L.Ed.2d 80 (1989), unless the administrator or other fiduciary has a conflict of interest. Id. at 115, 109 S.Ct. 948; Ladd v. ITT Corp., 148 F.3d 753, 753-54 (7th Cir.1998). (These cases say not that the conflict alters the standard of review, but that it affects how the standard is applied; but that comes to the same thing.) Rud argues that Liberty Life, as an insurance company, is not the plan’s administrator or any other species of ERISA fiduciary, and that in any event it had a profound conflict of interest.

On either ground, he concludes, the district court should have given plenary rather than deferential review to the company’s denial of his application for benefits. The plan’s administrator was Andersen Windows, not Liberty Life. But if Liberty Life was not an ERISA fiduciary too, there is no basis for Rud’s ERISA claim, Reich v. Continental Cas.

Co., 33 F.3d 754, 756-57 (7th Cir.1994), and then all he would have would be a simple state-law breach of contract suit, though one removable to federal court because the requirements of diversity jurisdiction (including, in the case of removal, that the defendant be a citizen of a state other than the one in which the federal district court to which he is removing it is located) are satisfied. One cannot bring an ERISA claim against someone who is not a fiduciary just because he happens to have a contract with an ERISA plan.

Rud could not sue FedEx under ERISA for failing to deliver a benefits check fedexed to him by an ERISA fiduciary. See Smith v. Provident Bank, 170 F.3d 609, 617 (6th Cir.1999); Arizona State Carpenters Pension Trust Fund v. Citibank (Arizona), 125 F.3d 715, 724 (9th Cir.1997). But Liberty Life is an ERISA fiduciary, defined as an entity that has discretionary authority over assets of an ERISA plan. 29 U.S.C. § 1002(21)(A); Farm King Supply, Inc. Integrated Profit Sharing Plan & Trust v. Edward D. Jones & Co., 884 F.2d 288, 292 (7th Cir.1989).

The policy that it issued to Andersen states that the insurance company “shall possess the authority to construe the terms of this policy and to determine benefit eligibility hereunder. Liberty’s decisions regarding construction of the terms of this policy and benefit eligibility shall be conclusive and binding.” Plan assets were used to buy the policy. The administration of the policy, and thus the disbursement or refusal to disburse plan assets, was confided to the discretion of the insurance company. This combination makes Liberty Life an ERISA fiduciary. Mers v.

Marriott International Group Accidental Death & Dismemberment Plan, 144 F.3d 1014, 1019-20 (7th Cir.1998); Cozzie v. Metropolitan Life Ins. Co., 140 F.3d 1104, 1107 (7th Cir.1998). It may seem odd that Liberty Life should be administering the plan, when Andersen is the plan administrator. The oddness is dissipated by recognizing that administration is divided. Andersen decides who is eligible to participate in the plan and explains the plan to its employees, but the determination of eligibility to receive benefits under the plan is confided to Liberty Life.

This division gives the insurance company discretionary authority over claim applications, making it an ERISA fiduciary. We are mindful of the circuit split over the question whether there can be, alongside the official plan administrator, a “de facto” administrator. Hall v. Lhaco, Inc., 140 F.3d 1190, 1195 (8th Cir.1998). The courts that say “no” are worried about the confusion that can result if decisions are being made by someone (usually the employer) who is not designated as the plan administrator. Our court, while aware of the potential for confusion, see Riordan v.

Commonwealth Edison Co., 128 F.3d 549, 551 (7th Cir.1997), has suggested that equitable estoppel might sometimes justify treating someone else as the plan administrator. “We can imagine a case in which the plan sponsor would be estopped to deny that it was the administrator; the district judge may have thought this such a case.

If UOP’s legal department had told Jones’s lawyer to forget about the Committee and direct all his document requests to the legal department, and if in reliance on this advice the lawyer had forgone an opportunity to obtain the documents from the plan administrator and Jones had suffered a harm as a result, the elements of equitable estoppel would be present.” Jones v. UOP, 16 F.3d 141, 144 (7th Cir.1994). But maybe it’s wrong to get hung up on who is (are) the plan administrator(s).

Maybe the right question to ask is whether the particular defendant made a discretionary determination concerning the plaintiffs entitlement to plan benefits. A genuine oddity is that the plan is not in the record — only irrelevant fragments of the plan

summary, plus the insurance policy. The proximate cause of the omission is that Rud did not sue Andersen. But of course he has or can obtain a copy of the plan; and Liberty Life must have a copy. The plan’s absence, however, is of no consequence to the appeal. No one is questioning that there is a plan and that the plan confides the making of benefits determinations to Liberty Life. The policy is the plan, Bidlack v.

Wheelabrator Corp., 993 F.2d 603, 615 (7th Cir.1993), so far as disability benefits are concerned, and the recital in the policy that we quoted thus determines the scope of judicial review— unless Rud is correct that the insurance company has a conflict of interest sufficient to defeat the recital. He argues that a conflict of interest exists because any money Liberty Life pays to a claimant reduces its profits. The ubiquity of such a situation makes us hesitate to describe it as a conflict of interest.

There is no contract the parties to which do not have a conflict of interest in the same severely attenuated sense, because each party wants to get as much out of the contract as possible. How serious the conflict is depends on circumstances. See, e.g., Leahy v. Raytheon Co., 315 F.3d 11, 15-16 (1st Cir.2002). If Liberty Life refuses to honor meritorious claims, it will obtain windfall profits in the short run, assuming that the premium that Andersen paid it was calculated’ on the expectation of a normal claims experience.

But Andersen will be dismayed — it has no interest in conferring such profits on Liberty Life, thereby incurring its employees’ ill will with no offsetting financial benefit to itself — and so may refuse to renew the policy when it expires, or demand a much lower premium.

The latter option, however, suggests a theoretical basis for suspecting a long-run conflict of interest: the chintzier the insurance company is in responding to benefits claims, the lower (given a competitive insurance market) the premium that Andersen will have to pay, whether to Liberty Life or to a competitor of Liberty Life, to obtain insurance.

This kind of thinking has persuaded several circuits that there is a conflict of interest, requiring more searching judicial review of a denial of benefits than the “arbitrary and capricious” standard implies, whenever an insurer is being asked to dip into its own pocket to pay a claim for benefits. Pinto v. Reliance Standard Life Ins. Co., 214 F.3d 377, 387-89 (3d Cir.2000); Armstrong v. Aetna Life Ins. Co., 128 F.3d 1263, 1265 (8th Cir.1997); Brown v. Blue Cross & Blue Shield of Alabama, Inc., 898 F.2d 1556, 1561-62 (11th Cir.1990). Our court has rejected that position. Shyman v. Unum Life Ins.

Co., 427 F.3d 452, 455 (7th Cir.2005); Leipzig v. AIG Life Ins. Co., 362 F.3d 406, 408-09 (7th Cir.2004); Perlman v. Swiss Bank Corp. Comprehensive Disability Protection Plan, 195 F.3d 975, 980-81 (7th Cir.1999); Mers v. Marriott International Group Accidental Death & Dismemberment Plan, supra, 144 F.3d at 1020-21; Chalmers v. Quaker Oats Co., 61 F.3d 1340, 1344 (7th Cir.1995). We have pointed out that given reasonably well-informed employees, an employer cannot reap a long-run benefit from reducing welfare benefits, whether directly or by delegating administration to a hard-nosed insurance company.

The employee’s total compensation package includes benefits as well as wages; reducing any component reduces the total. If the employer could have met its labor needs with a cheaper compensation package, it would have paid lower wages or promised lower benefits from the start. There would have been no need to incur the bother and uncertainty of trying to “steal” contracted-for benefits through the back door.

This analysis can be criticized as reflecting too sunny a view of the operation of labor markets, assuming as it does that employees have such good information that they cannot be fooled in the fashion suggested and that employers are always in for the long haul and therefore always cultivate a reputation for fair dealing with employees. E.g., Pierre Cahuc & André Zylberberg, Labor Economics, ch. 4 (2004).

Employees, the argument continues, might be less quick to blame their employer for an adverse benefits decision by an insurance company, albeit a company that had been retained by the employer, than they would the employer itself, if the employer were the plan administrator. “[W]hile in a perfect world, employees might pressure their companies to switch from self-dealing insurers, there are likely to be problems of imperfect information and information flow.

Employees typically do not have access to information about claim-denying by insurance companies, and the relationship between employees and insurance companies is quite attenuated; so long as obviously meritorious claims are well-handled, it is unlikely that an insurance company’s business will suffer because of its client’s employees’ dissatisfaction.” Pinto v. Reliance Standard Life ins. Co., supra, 214 F.3d at 388.

There is doubtless some truth in these critiques, but their acceptance would destabilize large reaches of contract law, of which ERISA is, after all, a part, since it neither requires employers to establish welfare and pension plans nor prescribes the terms of such plans. When the Supreme Court in Carnival Cruise Lines, Inc. v.

Shute, 499 U.S. 585, 111 S.Ct. 1522, 113 L.Ed.2d 622 (1991), held enforceable a forum-selection clause printed on the back of a cruise ticket, it brushed aside the arguments that many consumers don’t read the fíne print in their contracts and do not appreciate the significance, or perhaps even the meaning, of a forum-selection clause. See also IFC Credit Corp. v. Aliano Bros. General Contractors, Inc., 437 F.3d 606, 610 (7th Cir.2006). Of course ERISA is a paternalistic statute in a number of respects, notably in its vesting rules.

But it is hard to see why, if the plan unequivocally authorizes the insurance company to make the conclusive determination of eligibility, the courts should rewrite the provision. Firestone, it is true, did so, to a degree, by allowing “arbitrary and capricious” review. This was an implicit rejection of the model of a perfectly competitive labor market, for in such a market there would be no reason to look behind plan language that granted unreviewable discretion to the administrator. The assumption would be that the employee had been compensated for surrendering his right to judicial review.

But whether or not Firestone goes too far, courts that go further by enlarging the scope of judicial review whenever the plan administrator has a financial interest in the eligibility determination make even deeper inroads into freedom of contract, since, as we said, so attenuated a “conflict of interest” is found in every contract. As noted in Pinto v. Reliance Standard Life Ins. Co., supra, 214 F.3d at 379, a number of cases adopt a “sliding scale” approach that permits the intensity of judicial review of benefits determinations to vary with circumstances. They include two of our cases. Manny v.

Central States, Southeast & Southwest Areas Pension & Health & Welfare Funds, 388 F.3d 241, 242-43 (7th Cir.2004), and Van Boxel v. Journal Co. Employees’ Pension Trust, 836 F.2d 1048, 1052-53 (7th Cir.1987); see also Vega v. National Life Ins. Services, 188 F.3d 287, 296-97 (5th Cir.1999); Chambers v. Family Health Plan Corp., 100 F.3d 818, 825-27 (10th Cir.1996). This approach enables the court to distinguish between two classes of case.

The first is where (1) benefits are paid from a trust fund that has ample assets that cannot be transferred back to the employer, so that the employer’s stake in a particular benefits determination, especially if the amount of the benefits is small, is negligible, or (2), as in Leahy v. Raytheon Co., supra, the plan is voluntary and employees can withdraw without penalty, or (3) the plan uses an insurance policy that is retrospectively rated, meaning that the insurer charges a premium equal to its disbursements plus an administrative fee, so that the less it pays out on claims the less it takes in in premiums.

The second type of case is where the trust fund is underfunded, the benefits claim is large, the employer will have to transfer assets to the plan if the claim is honored, and the employer is planning to leave the industry, a decision that eliminates concern with reputation as a constraint against sharp dealing with employees. Rud has presented no evidence to show where between these end points in the spectrum of potential conflicts of interest his case lies. The absence of evidence is fatal.

The critical question in the application of the sliding-scale approach is whether evidence shall be required or conflicts of interest presumed. If no evidence is required, the sliding-scale approach can allow a conflict of interest to be attributed on the basis of the fundamental tug-of-war character of every contract. Our cases, such as Shyman and Leipzig, by rejecting the automatic assumption that an insurer has a conflict of interest that justifies departing from the “arbitrary and capricious” standard, implicitly reject a sliding-scale approach conceived in such abstract terms.

But our cases permit the plaintiff to show that the particular circumstances of his case demonstrate the existence of a real and not merely notional conflict of interest, the sort of thing the parties would not have foreseen when they agreed that the plan administrator’s determinations would be conclusive. Mers v. Marriott International Group Accidental Death & Dismemberment Plan, supra, 144 F.3d at 1020-21. We find this approach in cases from other circuits as well. See Atwood v. Newmont Gold Co., 45 F.3d 1317, 1322-23 (9th Cir.1995), and cases cited in Pinto v. Reliance Standard Life Ins.

Co., supra, 214 F.3d at 385. It strikes the right balance, consistent with Firestone, between freedom of contract and a realistic sense of its limits. Rud’s claim against Liberty Life cannot survive as a state law claim for breach of contract, he being a third-party beneficiary of the insurance policy, a contract between Andersen and the insurance company. A suit to enforce a claim for benefits under an ERISA plan can be brought only under ERISA; parallel state law remedies are preempted. 29 U.S.C. § 1144(a); Rush Prudential HMO, Inc. v. Moran, 536 U.S. 355, 392, 122 S.Ct. 2151, 153 L.Ed.2d 375 (2002).

It is not as if, to return to an earlier example, Liberty Life were merely a contractor with an ERISA administrator or fiduciary. It is an ERISA fiduciary, and Rud’s complaint is that it has failed to comply with the duties that the plan imposed on it. Affirmed.