Susan Hess worked for Fleet Mortgage Company as a loan officer until she became disabled in April 1996. Throughout her tenure at Fleet, Hess was covered by long-term disability insurance under a group policy provided by Hartford Life and AccidentInsurance. When she became disabled, Hess filed for benefits under the Hartford plan. Hartford agreed to begin paying benefits, but at a level lower than the amount to which Hess thought she was entitled. Hess brought this suit under the Employee Retirement Income Security Act of 1974, 29 U.S.C. sec. 1001 et seq., commonly known as ERISA. The district court, ruling on stipulated facts, held that Hartford's calculation of Hess's benefits was arbitrary and capricious and ordered Hartford to pay benefits at the full rate Hess sought. We see no error in the district court's analysis of the case and therefore affirm its judgment.

The long-term disability policy that Hartford sold to Fleet, which governed Hess's benefits, provided that a participant who became disabled while insured under the plan would be paid a monthly benefit calculated, in relevant part, by multiplying the participant's "Monthly Rate of Basic Earnings" (MRBE) at the time of disability by a benefit percentage, which for Hess was 66.67%. The term "Monthly Rate of Basic Earnings" was defined as "your regular monthly pay, from the Employer, not counting: (1) commissions; (2) bonuses; (3) overtime pay; or (4) any other fringe benefit or extra compensation." Thus, the benefit to which Hess was entitled depended on what qualified as her "regular monthly pay" at the time she became disabled.

Hess's compensation as a loan officer at Fleet was based entirely on commissions. This might make one think that she had no MRBE for purposes of the long-term disability plan, but that is not how things worked. Instead, prior to 1996, Fleet paid Hess a biweekly "draw" against her commissions and then paid her the remaining commissions in a monthly lump sum. Her 1995 contract with Fleet stated that her level of benefits, including long-term disability benefits, would be determined according to her "base compensation," which the contract defined as the amount of her draw. Hess's draw for 1995 amounted to $23,400. Fleet withheld insurance premiums for the Hartford policy from Hess's draw payments.

In 1996, Fleet decided to phase out the draw system for fully-commissioned employees like Hess. Hess's 1996 contract with Fleet stated that the draws would end as of April 5, 1996. The contract also stated that benefit levels, including those for the long-term disability plan, would be based on Hess's "base salary and/or draw." Although the contract did not define "base salary," Fleet stipulated that it calculated that amount by averaging the employee's total commissions over the previous two years. Using this definition, Hess's base salary in 1996 was a little over $45,000. For reasons unexplained, Fleet did not phase out the draw on April 5, 1996, as it was contractually obligated to do; the phase-out was extended until June 1, 1996. On that date, Fleet increased the amount of the insurance premiums it withheld from Hess's monthly compensation to reflect the change from the draw system to the base salary system of calculating benefits.

Hess became disabled on April 19, 1996. When she filed her claim under the long-term disability policy in October 1996, Hartford had on file a document from Fleet stating that her annual base salary was $23,400, which was the draw she had received under her 1995 contract. Hartford calculated her benefits based on this amount and awarded her a monthly benefit of $1,300. Hess objected. She reasoned that she did not become disabled until after April 5, which was when her contract with Fleet stated that Fleet would phase out her draw and begin basing her benefits on the average of her total commissions over the past two years. On this basis, she appealed the benefits decision within Hartford.

The Hartford claims examiner assigned to Hess's case looked into the situation and determined (1) that Hess's pay had been based entirely on commissions, (2) that (in his view) the language of the policy excluded commissions from the MRBE calculation, but (3) that Hess had paid premiums for disability insurance, and (4) that Fleet intended for Hess to be covered. The examiner testified that, in his view, Hartford had three options: first, it could find that Hess was not entitled to any benefits because her pay was based entirely on commissions; second, it could continue to pay benefits based on the draw amount notwithstanding her date of disability; or third, it could pay Hess benefits based on her "base salary," which was in turn a function of her total commissions. The examiner chose the second option, to continue paying benefits based on the old draw amount. Before making this decision, he spoke with a Fleet representative who told him that Fleet intended for Hess to be covered and that it had viewed the draw amount as Hess's base pay. The Fleet representative told the examiner that Fleet had switched from the draw system to a system based on total commissions on June 1, but because this date fell after Hess became disabled, the examiner disregarded this information. The examiner had received a letter from Hess's lawyer explaining that, according to Hess's contract with Fleet, the key date for the change-over should have been April 5, not June 1, and Hess's payments should therefore have been based on her total commissions. Although the letter made explicit reference to Hess's contract with Fleet and quoted the relevant portions, the examiner did not read the letter. One consequence of his omission was that he remained ignorant of the contents of Hess's contract. Despite the fact that the Hartford policy documents explicitly stated that "[i]f [Fleet] gives The Hartford any incorrect information, the relevant facts will be determined to establish if insurance is in effect and in what amount," the examiner made no effort to investigate the discrepancy between the Fleet representative's claim that the switch from the draw system to the average commission system occurred on June 1 and Hess's claim that it occurred on April 5.

After Hartford informed her of its decision, Hess filed this ERISA claim in the district court seeking benefits based on her average annual commissions. Finding that Hartford's handling of Hess's claim was arbitrary and capricious, the district court ordered Hartford to pay $50,927.10 in back benefits and to begin paying Hess's monthly benefits at a rate of $2,512.55. The court also awarded Hess a little over $40,000 in attorneys' fees. In this appeal, Hartford challenges both the judgment against it and the award of attorneys' fees. In addition, Hartford argues that even if the judgment against it was correct, the district court erred in ordering it to pay benefits at the higher level rather than remanding the case to Hartford to allow it to recalculate the benefits.

The parties disagree over the standard of review that should govern our decision. The district court entered its judgment after receiving a stipulation of the facts that made up the administrative record from the parties. Hartford urges that a judgment on stipulated facts is akin to a summary judgment, and accordingly that our review of the district court's decision should be de novo. It is true that the facts, to the extent they are stipulated, will not be in dispute, but we think that the procedure the parties followed here is more akin to a bench trial than to a summary judgment ruling. See May v. Evansville-Vanderburgh Sch. Corp., 787 F.2d 1105, 1115-16 (7th Cir. 1986) (where parties agree to a judgment on stipulated facts, "[i]n effect the judge is asked to decide the case as if there had been a bench trial in which the evidence was the depositions and other materials gathered in pretrial discovery."). The standard of review that governs is therefore the one found in Fed. R. Civ. P. 52(a). As we would after a bench trial, we will review the district court's legal conclusions de novo and review any factual inferences the district court made from the stipulated record as well as its application of the law to the facts for clear error. See Johnson v. West, 218 F.3d 725, 729 (7th Cir. 2000).

The parties agree that Hartford's policy gave Hartford discretion to interpret its terms. As the district court recognized, this meant that it was required only to determine whether Hartford's conclusion was arbitrary and capricious. See Herzberger v. Standard Ins. Co., 205 F.3d 327, 329 (7th Cir. 2000). This is, of course, a deferential standard of review. Under the arbitrary and capricious standard, a plan administrator's decision should not be overturned as long as (1) "it is possible to offer a reasoned explanation, based on the evidence, for a particular outcome," (2) the decision "is based on a reasonable explanation of relevant plan documents," or (3) the administrator "has based its decision on a consideration of the relevant factors that encompass the important aspects of the problem." Exbom v. Central States, Southeast and Southwest Areas Health and Welfare Fund, 900 F.2d 1138, 1142-43 (7th Cir. 1990) (citations omitted). Nevertheless, "[d]eferential review is not no review," and "deference need not be abject." Gallo v. Amoco Corp., 102 F.3d 918, 922 (7th Cir. 1996). In some cases, the plain language or structure of the plan or simple common sense will require the court to pronounce an administrator's determination arbitrary and capricious. Id. We agree with thedistrict court that this is such a case.

Hartford's arguments in favor of its examiner's decision fall into three broad categories. First, Hartford points out that the policy documents exclude "commissions" from the benefit base and suggests that, on this basis, it could have justified denying Hess benefits entirely. This argument appears to imply that Hartford's decision to grant Hess some benefits in the face of the exclusion for commissions must therefore have been fair and reasonable. Second, Hartford argues that the district court should not have taken Hess's contract with Fleet into consideration, both because it was not before the examiner and because it is not a plan document. Finally, Hartford argues that it made a reasonable decision in treating June 1, rather than April 5, as the date on which Fleet switched Hess's benefit base from the draw to her average commissions. We will consider each of these arguments in turn.

We agree with the district court that Hartford's argument that it could have denied Hess benefits entirely because her compensation was based solely on commissions ignores both the plain language of the policy and the intent of the parties. As the district court noted, the Hartford policy based an employee's benefits on her "regular monthly pay . . . not counting: (1) commissions; (2) bonuses; (3) overtime pay; or (4) any other fringe benefit or extra compensation." In Hess's case, her "regular monthly pay" was determined based on a formula derived from her commissions; she received no other form of compensation. Unless the plan was to be read as creating a class of employees who had no regular monthly pay at all, which was at a minimum not Fleet's position, the notion of a regular monthly pay that was dependent in some sense on commissions was the only interpretation consistent with the contract. On this reading, employees who had both a regular monthly pay plus extra commissions would not have been entitled to credit for the commissions. The district court held that Hess's commissions were not this kind of "extra" compensation, and therefore were not subject to the exclusion quoted above. To read the contract as Hartford urges, so that commissions are excluded regardless of whether they are "extra" compensation or are an employee's only compensation, renders the word "extra" in the exclusion superfluous. In addition, we note that Fleet withheld insurance premiums from Hess's pay throughout her tenure at Fleet, and Hartford concedes that Fleet intended for fully-commissioned employees such as Hess to be covered by the insurance policy. Thus, Hartford's argument that it could reasonably have denied Hess all benefits is unavailing. (Also, if this was indeed what the plan meant, Hartford would have had no business paying Hess even the amount it was willing to give her; plan administrators cannot randomly pay benefits to individuals not entitled to them.)

Turning to Hartford's contention that we should disregard Hess's contract with Fleet, we find Hartford's arguments on this point to border on the frivolous. Hartford is, of course, correct that in evaluating a plan administrator's decision under an arbitrary and capricious standard of review, we should consider only the evidence that was before the administrator when it made its decision. See Trombetta v. Cragin Fed. Bank for Sav. Employee Stock Ownership Plan, 102 F.3d 1435, 1437 n.1 (7th Cir. 1996). Hartford argues that its examiner never saw Hess's contract and that we therefore may not consider it. That characterization is not, however, a fully accurate account of the information before the examiner. In fact, the examiner had before him a letter from Hess's attorney that made explicit reference to the contract and even quoted the relevant portions. The contract itself apparently had been attached to the letter at some point, although it had been removed before the letter reached the examiner's desk. Nevertheless, the examiner had been alerted not only to the contract's existence, but also to the very language on which Hess was relying; he easily could have obtained a complete copy through a simple phone call to Hess's lawyer or to Fleet. The fact that the examiner did not bother to read pertinent evidence actually before him cannot shield Hartford's decision from review. To the contrary, this court has noted that the fact that an administrator blatantly disregards an applicant's submissions can be evidence of arbitrary and capricious action. See Perlman v. Swiss Bank Corp. Comprehensive Disability Protection Plan, 195 F.3d 975, 982 (7th Cir. 1999) (discussing hypothetical situation in which "application was thrown in the trash rather than evaluated on the merits"). Similarly, we are not persuaded by Hartford's argument that the contract is irrelevant because it is not a plan document but an independent agreement between Hess and Fleet, to which Hartford was not a party. The policy Hartford sold to Fleet explicitly based Hess's benefit level on Hess's compensation, and Hess's contract with Fleet is the best evidence of what that compensation was. Far from being irrelevant, the contract was the most critical evidence of the benefits to which Hess was entitled.

Having determined that Hess's 1996 employment contract is properly a part of the administrative record the district court was entitled to consider, we must next decide whether Hartford could reasonably have determined that Hess's benefits as of April 19, 1996, should have been based only on her 1995 draw amount. Like the district court, we cannot read the contract that way. Hess's 1996 contract clearly states that the draw system was to be phased out as of April 5. The contract also specifies that her benefits, including long-term disability benefits, would be calculated based on her "base salary and/or draw." (We note in passing that the phrase "and/or" has its critics. Bryan A. Garner reports in A Dictionary of Modern Legal Usage 56 (2d ed. 1995), that "and/or has been vilified for most of its life-- and rightly so." He goes on to say, however, that the expression, while "undeniably clumsy, does have a specific meaning (x and/or y = x or y or both)." Id.) Here, this would mean that Hess could have her benefits calculated on the basis of her base salary, or her draw, or both. In the context of Fleet's transition away from a draw system, the only reasonable interpretation of this provision was that the benefits would be based on the draw while it was in effect and on the base salary thereafter. As of April 5, Hess was thus contractually entitled to a benefits package based on her base salary--that is, based on the average of her previous two years' commissions. The fact that Fleet may have breached the contract (or been slow in implementing its details) by failing to move from the draw system to the base salary system until June 1 does not change the package of compensation and benefits to which Hess was contractually entitled. Nor could the fact that Fleet failed to inform Hartford about the date the change-over was to have occurred affect Hess's benefit amount. The Hartford policy states that "[i]f [Fleet] gives The Hartford any incorrect information, the relevant facts will be determined" to establish the correct benefit amount. Once informed by Hess's attorney that Hess believed the information Fleet provided Hartford was incorrect, it was incumbent on the examiner to refer to Hess's employment contract to determine her actual regular monthly pay. Had he done so, he would have seen that Hess became entitled to the higher level of benefits on April 5, two weeks before her disability. The district court therefore did not err when it concluded that Hartford's failure to consider the contract was arbitrary and capricious.

Having determined that Hartford's calculation of Hess's benefits was arbitrary and capricious, the court ordered Hartford to pay Hess benefits based on the average of her last two years' commissions. Hartford complains that the district court should not have ordered it to pay a specific amount but instead should have remanded the case to Hartford for reconsideration of the appropriate benefit. We do not dispute the fact that such a remand is sometimes the appropriate step to order. See Gallo, 102 F.3d at 923. Remand is unnecessary, however, when "the case is so clear cut that it would be unreasonable for the plan administrator to deny the application for benefits on any ground." Id. Here, only two levels of benefits were possible: either Hess was entitled to benefits based on her draw, or she was entitled to benefits based on her base salary, which Fleet calculated by averaging her previous two years' commis sions. Once the district court determined that the latter benefit level was the correct one, there was nothing left for Hartford to calculate. The benefit amount was clear cut, and the district court made no error in ordering Hartford to pay that amount.

Nor did the district court err in awarding attorneys' fees to Hess. ERISA permits the district court to award a reasonable attorneys' fee to either party, see 29 U.S.C. sec. 1132(g)(1), and there is a modest presumption that the prevailing party in an ERISA case is entitled to a fee. See Bowerman v. Wal-Mart Stores, Inc., 226 F.3d 574, 592 (7th Cir. 2000). Although we have formulated the test for when attorneys' fees should be awarded under ERISA in various ways, we have recently noted that the various formulations boil down to the same bottom-line question: "Was the losing party's position substantially justified and taken in good faith, or was that party simply out to harass its opponent?" Id. at 593. We review the district court's grant of fees only for abuse of discretion, see Wyatt v. UNUM Life Ins. Co. of Am., 223 F.3d 543, 548 (7th Cir. 2000), and we find no abuse in this case. In deciding to award fees, the court stressed that Hartford's examiner had failed to read the letter from Hess's lawyer outlining Hess's arguments for the higher benefit level and had failed to consider Hess's employment contract, which was the most reliable evidence as to what her compensation was. Had the examiner considered this information, the court believed, this litigation could have been avoided. As we discussed above, we share the district court's view of the examiner's lackadaisical approach in this case, and we find that the court was within its discretion in awarding fees. Compare Filipowicz v. American Stores Benefit Plans Comm., 56 F.3d 807, 816 (7th Cir. 1995) (approving award of fees where plan administrator failed to learn "what plan documents were actually in ef fect" at the relevant time).

The judgment of the district court, both with respect to liability and fees, is Affirmed