Carlson Marketing Group, Inc. v. Royal Indemnity Co. (517 F. Supp. 2d 1089, 2007)

MEMORANDUM OPINION AND ORDER

PATRICK J. SCHILTZ, District Judge.

Plaintiff Carlson Marketing Group (“Carlson”) filed this action against two of its excess insurers — defendant Royal Indemnity Company (“Royal”) and defendant National Union Fire Insurance Company of Pittsburgh (“National Union”) — to determine to what extent each insurer is liable to Carlson for expenses that Carlson incurred in defending and settling two patent-infringement lawsuits. Currently before the Court are five summary-judgment motions, one each by Carlson and Royal, and three by National Union. This is two motions too many by National Union, as the Court has explained elsewhere.

Nonetheless, the Court will address all of the arguments raised by the parties.

The five motions raise only four issues. The main issue is how to allocate certain losses suffered by Carlson among various insurance policies and policy periods. A second issue (closely related to the first) is whether the expenses incurred by Carlson in defending and settling one of the patent-infringement suits should be treated as a single loss or — because the suit involved two patents and (arguably) multiple accused products — as multiple losses. The third issue is whether some of Carlson’s losses are really punitive damages and therefore not covered or, if covered, uninsurable under Minnesota law. The final issue is whether the legal expenses incurred by Carlson in defending the underlying patent-infringement lawsuits count against (“erode”) the applicable policy limits or instead are payable in addition to the policy limits.

For the reasons given below, the Court rules as follows:

First, all of the expenses incurred by Carlson in defending and settling the Maritz lawsuit are attributable to the 1997-1998 policy period. The Court will therefore grant Carlson’s motion for summary judgment on this issue, deny National Union’s motion for summary judgment on “trigger and allocation,” and deny as moot Royal’s motion for summary judgment as it relates to this issue.

Second, all of the acts committed by Carlson which gave rise to the Maritz lawsuit “involved] the same intellectual right and covered Product” and therefore must be aggregated into a single wrongful act. The Court will therefore grant Carlson’s motion for summary judgment on this issue.

Third, the $5-million portion of the Meridian settlement designated as “enhanced compensatory damages” by Carlson and its insurers is neither covered under Carlson’s insurance policies nor insurable under Minnesota law. The Court will therefore grant National Union’s motion for summary judgment on this issue.

Finally, under the excess-insurance policy issued by Royal, Carlson’s defense costs are payable in addition to the policy limits. The Court will therefore grant National Union’s motion for summary judgment on this issue and deny Royal’s motion for summary judgment as it relates to the same issue.

I. BACKGROUND

A. Underlying Patent Litigation

Carlson seeks indemnification from Royal and National Union for losses Carlson incurred in defending and settling two separate patent-infringement lawsuits, referred to by the parties as the Maritz and Meridian lawsuits. For the most part, the parties do not dispute that these losses are the type of losses covered by the relevant insurance policies. They also do not dispute the essential facts related to those suits.

1. The Maritz Lawsuit

In 2002, Maritz, Inc. and a unit of American Express (collectively “Maritz”) sued Carlson in federal district court in Missouri for infringing two Maritz patents, U.S. Patent Nos. 5,689,100 (the TOO patent) and 5,956,695 (the ’695 patent). Maritz, Inc. v. Carlson Cos., No. 4:02-CV-161 (E.D.Mo.); Pl. Mem. Supp. Mot. Partial S.J. (“Carlson SJM” [Docket No. 83]) App. O. The '100 patent was issued on November 18, 1997, Carlson SJM App. L; the ’695 patent was issued on September 21, 1999, Carlson SJM App. M. The earliest possible date on which Carlson could be liable for infringing either patent is therefore November 18, 1997.

Maritz alleged that Carlson infringed its patents by operating incentive programs on behalf of various employers under which the employers rewarded their employees with stored-value cards, such as gift cards and prepaid debit cards. Carlson SJM App. O. Both of Maritz’s patents cover systems and methods for limiting the use of such stored-value cards to particular merchants or merchandise — for instance, limiting a prepaid Visa or MasterCard to use at Neiman-Marcus only, rather than to all locations that accept Visa or MasterCard.

Carlson operated a number of different stored-value-card incentive programs, but only one was named in the Maritz complaint: the “TravPass” program, under which the award cards could be used only at travel- and dining-related merchants. Carlson SJM App. 0 at 3; id. App. V. Nonetheless, both Maritz and Carlson also litigated over whether a different Carlson program — the “Best of Everything” (“BOE”) program, under which the award cards could be used only at high-end merchants, Carlson SJM at 4; id. App. V—infringed the Maritz patents. Id. App. X at 5; Pl. Reply Mem. Supp. Mot. Partial S.J. (“Carlson Reply” [Docket No. 160]) App. B.

In November 2004, Carlson and Maritz settled the lawsuit. Carlson SJM App. P. Carlson agreed to pay a running royalty of one percent on the amount of money loaded on stored-value cards issued under incentive programs, plus a lump-sum payment of $12 million. Id. at 9, 12. By the time Carlson settled the suit, it had incurred $3,385 million in defense costs. Carlson SJM at 13; id. App. F ¶ 18. Neither National Union nor Royal has yet indemnified Carlson for any of these defense or settlement costs. Carlson SJM at 13.

2. The Meridian Lawsuit

In 2002, Meridian Enterprises Corporation sued Carlson in federal district court in Missouri for infringing U.S. Patent No. 5,025,372 (the ’372 patent). Meridian Enters. Corp. v. Carlson Mktg. Group, Inc., No. 4:01-CV-1955 (E.D.Mo.); see Nat’l Union Mem. Partial S.J. Enhanced Damages (“NU SJM Enh. D.” [Docket No. 92]) App. A. Like the patents in the Maritz suit, the ’372 patent covers certain aspects of stored-value-card incentive programs. NU SJM Enh. DApp. A at 1-2.

The case was tried to a jury in February 2004. The jury found in favor of Meridian and awarded it $10.5 million in compensatory damages. Id. at 4. The jury also found that Carlson’s infringement was willful under 35 U.S.C. § 284. NU SJM Enh. D.App. A at 4. This finding of willfulness authorized the district judge, Judge Catherine D. Perry, to award up to treble damages. In May 2004, Judge Perry awarded enhanced damages equal to the jury’s $10.5 million award, for a total award of $21 million. Id. at 33. She also entered a permanent injunction against Carlson. See id. at 38.

Carlson appealed to the Federal Circuit. Meridian Enters. Corp. v. Carlson Mktg. Group, Inc., Nos. 04-1401 & 04-1554 (Fed.Cir.); see Pl. Mem. Opp. NU SJM Enh. D. (“Carlson Opp. Enh. D.” [Docket No. 116]) App. T. On appeal, Carlson argued, among other things, that the Federal Circuit should vacate the judgment because of Judge Perry’s failure to recuse herself despite a purported conflict of interest. See Carlson Opp. Enh. D.App. V at 3-4. The Federal Circuit held that Carlson was sufficiently likely to prevail on that argument that the district court’s injunction should be stayed during the pendency of the appeal. Id. at 4.

The Federal Circuit never ruled on the merits of the appeal. In May 2005, just before the case was scheduled for oral argument, Carlson and Meridian settled for $17 million. Id. App. W at 7; Carlson Opp. Enh. D. at 5. At the same time, Carlson entered into an agreement with Royal and National Union related to Carlson’s settlement with Meridian. Id. App. X. The two insurers authorized Carlson to settle with Meridian for up to $18 million, but specified that any amount over $16 million would be paid by Carlson alone. Id. at 2. As to the remaining $16 million, Carlson and its insurers agreed to litigate in this' action who was responsible for what portion of that amount. Id. at 1-3. They also agreed to designate $11 million as “compensatory damages” and the other $5 million as “enhanced compensatory damages.” Id. at 3.

B. Insurance Coverage

The details of the parties’ coverage disputes will be addressed at great length below. At this point, the Court merely provides a brief introduction to the insuranee policies involved in this case and the specific coverage provided under those policies for patent infringement.

1. Insurance Policies

Between July 1997 and June 2003 (the period relevant to this action), Carlson purchased three layers of insurance coverage: one primary layer and two excess layers. For its primary coverage, Carlson purchased a Commercial General Liability (“CGL”) policy with various endorsements. One of those endorsements, the Intellectual Property Endorsement (“IPE”), covered losses resulting from patent infringement by Carlson. Joint Subm. Stip. Authenticity Policies (“Policy Stip.”) Tab 1 at P00015; Tab 2 at P00096; Tab 5 at P00512; Tab 6 at P00636; Tab 8 at P00760.

The policy years generally ran from July 1 of each year to June 30 of the next. From July 1997 through June of 2000— that is, for the policy years 1997-1998, 1998-1999, and 1999-2000 — Carlson’s primary coverage was provided under policies issued by Reliance National Indemnity Company (“Reliance”). Policy Stip. Tabs 1-3. Coverage for the 1999-2000 policy year was originally provided by Reliance, but responsibility for that policy year was transferred to Lumbermens Mutual Casualty Company (“Lumbermens”) as a result of a “novation” endorsement executed by Carlson, Reliance, and Lumbermens in November 2000. Id. Tab 4. The novation did not change the terms of the Reliance policy in effect during the 1999-2000 policy year; instead, the novation simply transferred responsibility for claims under that policy from Reliance to Lumbermens. Neither Reliance nor Lumbermens is a party to this suit.

From July 2000 through April 2003— that is, for the policy years 2000-2001 and 2001-2002 and most of the policy year 2002-2003 — Carlson’s primary coverage was provided under policies issued by Lumbermens. Id. Tabs 5-7. Finally, from May through June 2003, Carlson purchased its primary coverage from National Union. Id. Tab 8. Thus, for a very short period, National Union was both Carlson’s primary insurer and one of Carlson’s excess insurers.

Carlson purchased excess insurance for the entire period from July 1997 through June 2003 from defendants Royal and National Union. Royal provided $5 million per year in first-layer excess coverage through an umbrella policy (the “Big Shield Commercial Catastrophe Liability Policy”) and associated endorsements. Id. Tab 9. National Union provided $44 million per year in second-layer excess coverage through an insurance contract that expressly “follow[s] the terms, definitions, conditions and exclusions” — in insurance lingo, “follows the form” — of Royal’s first-layer policy. Id. Tab 10 at NU02.

For any particular covered loss, the primary policy must respond first. If that policy’s limits are exhausted, Royal’s policy must then respond. Finally, if a covered loss exceeds the limits of both the primary policy and Royal’s policy, National Union’s policy must respond.

2. Coverage for Patent Infringement

Before July 1997, the primary policy did not cover losses related to patent infringement. Effective July 1, 1997, however, the primary policy included the IPE, under which Carlson was covered for losses resulting from alleged patent infringement. Policy Stip. Tab 1 at P00015; Tab 2 at P00096; Tab 5 at P00512; Tab 6 at P00636; Tab 8 at P00760. Specifically, Carlson was covered under the IPE for losses resulting from certain IP-related “wrongful acts,” defined to include “any violation of a legal right or rights associated with patents.” E.g., id. Tab 1 at P00017. The IPE was amended from time to time; some of those amendments will be addressed below.

The IPE provided $1 million in coverage per wrongful act, subject to a $1 million deductible. E.g., id. Tab 1 at P00018. This creates what is known in the insurance industry as a “fronting” arrangement: Carlson is technically covered under the primary policy for patent infringement, but because the deductible equals the policy limit, Carlson cannot actually recover any proceeds from the primary insurer. Instead, Carlson will be indemnified — if at all — by its excess insurers once Carlson exhausts the $1 million limit/deductible of the underlying policy. The parties do not dispute that, under the primary policy, defense costs incurred by Carlson in connection with a covered patent-infringement suit erode the policy limits and count against the deductible. Nat’l Union Mem. Partial S.J. Def. Costs at 8-10 (“NU SJM Def. Costs” [Docket No. 63]); Royal Resp. Opp. Nat’l Union Mot. Partial S.J. Def. Costs at 6 (“Royal Opp. Def. Costs” [Docket No. 102]); Pl. Mem. Opp. Nat’l Union Mot. Partial S.J. Def. Costs at 2-3 (“Carlson Opp. Def. Costs” [Docket No. 117]).

II. ANALYSIS

A. Standard of Review and Applicable Law

A party is entitled to prevail on a motion for summary judgment “if the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any, show that there is no genuine issue as to any material fact and that the moving party is entitled to a judgment as a matter of law.” Fed.R.Civ.P. 56(c). In considering a motion for summary judgment, a court must assume that the nonmoving party’s evidence is true and draw all justifiable inferences arising from the evidence in that party’s favor. Taylor v. White, 321 F.3d 710, 715 (8th Cir.2003).

The parties seem to agree that Minnesota law governs interpretation of the policies in this case. The Court sees no reason to doubt that conclusion. See Nat’l Union Fire Ins. Co. of Pittsburgh v. Terra Indus., Inc., 346 F.3d 1160, 1164 (8th Cir.2003) (state law governs insurance policies). Under Minnesota law, the interpretation and construction of an insurance policy is a matter of law for the court. Watson v. United Servs. Auto. Ass’n, 566 N.W.2d 683, 688 (Minn.1997); Caledonia Cmty. Hosp. v. St. Paul Fire & Marine Ins. Co., 307 Minn. 352, 239 N.W.2d 768, 770 (1976).

Insurance policies are interpreted according to the same principles that govern the interpretation of contracts generally. Progressive Specialty Ins. Co. v. Widness ex rel. Widness, 635 N.W.2d 516, 518 (Minn.2001); Nathe Bros., Inc. v. Am. Nat’l Fire Ins. Co., 615 N.W.2d 341, 344 (Minn.2000). Thus insurance policies must be interpreted to give effect to the parties’ intent. Nathe Bros., 615 N.W.2d at 344. Further, insurance policies must be construed as a whole. Haarstad v. Graff, 517 N.W.2d 582, 584 (Minn.1994). All terms of an insurance policy must be given effect if possible. Bobich v. Oja, 258 Minn. 287, 104 N.W.2d 19, 24 (1960); Steele v. Great W. Cas. Co., 540 N.W.2d 886, 888 (Minn.Ct.App.1995).

If the language of an insurance policy is clear, the policy language must be given its “usual and accepted meaning.” Widness, 635 N.W.2d at 518. Moreover, the meaning of policy language is assessed against an objective standard: The court asks what a “reasonable person” in the insured’s position would have understood by the language. Canadian Universal Ins. Co. v. Fire Watch, Inc., 258 N.W.2d 570, 572 (Minn.1977); Soo Line R.R. Co. v. Brown’s Crew Car of Wyo., 694 N.W.2d 109, 113 (Minn.Ct.App.2005).

Whether an insurance policy is ambiguous is a question of law for the court. Columbia Heights Motors, Inc. v. Allstate Ins. Co., 275 N.W.2d 32, 34 (Minn.1979); Brault v. Acceptance Indem. Ins. Co., 538 N.W.2d 144, 147 (Minn.Ct.App. 1995). Policy language is ambiguous if it can reasonably be interpreted more than one way. Am. Commerce Ins. Brokers, Inc. v. Minn. Mut. Fire & Cas. Co., 551 N.W.2d 224, 227 (Minn.1996). Ambiguous policy language will be construed against the insurer. Widness, 635 N.W.2d at 518. Put another way, all doubts (or, at least, all reasonable doubts) about policy interpretation will be resolved in favor of the policy holder. Haarstad, 517 N.W.2d at 584 (“A ‘policy should be construed as a whole with all doubts concerning the meaning of language employed to be resolved in favor of the insured.’ ” (quoting Canadian Universal, 258 N.W.2d at 572) (emphasis added)); Brault, 538 N.W.2d at 147 (“The general rule provides that any reasonable doubt as to the meaning of the language of an insurance policy is resolved in favor of the insured.” (emphasis added)).

B. Allocation of Maritz Losses

The most difficult and consequential issue before the Court is how to allocate among the various policies and policy periods the losses that were sustained by Carlson in defending and settling the Maritz suit — losses that relate to seven years of alleged infringement. The Maritz suit involved alleged infringement by Carlson from November 1997 (when the patent was issued) through November 2004 (when the case was settled). The IPE in the primary policy was in effect from July 1997 through June 2003' — -that is, for all but the last 17 months of the seven-year period of infringement. How, then, should losses caused by that infringement be allocated? The three parties take three different positions.

1. The Parties’ Contentions

Carlson moves for summary judgment on this issue, in effect asking this Court to rule that all of the losses that Carlson incurred in defending and settling the Maritz lawsuit should be allocated to a single policy period. Carlson SJM at 28-29. (The question of which policy period is a separate issue and is addressed below.) On this theory, Carlson would be obligated to pay only one deductible, and Carlson’s excess insurers (Royal and National Union) would each be obligated to pay up to only one policy limit. Responsibility for the almost $15.4 million that Carlson spent to defend and settle the Maritz lawsuit thus would be allocated as follows: Carlson would be responsible for $1 million (a single $l-million deductible), Royal would be responsible for $5 million (a single $5-million policy limit), and National Union would be responsible for the rest (about $9.4 million).

Royal agrees with Carlson on this issue. Royal Resp. Pl. Mot. Partial S.J. at 2 (“Royal Resp. Carlson SJM” [Docket No. 112]). Royal also asserts that, entirely apart from the merits of Carlson’s arguments, Royal’s exposure for the Maritz suit is limited to one $5-million annual policy limit based on a non-cumulation clause in the Royal umbrella policy. Id. at 2-3. Royal has moved separately for summary judgment on this issue. See Royal Mem. Supp. Mot. Partial S.J. Policy Limits at 20-26 (“Royal SJM” [Docket No. 87]).

Not surprisingly, National Union — who, under Carlson’s and Royal’s theories, gets stuck paying most of the Maritz losses— opposes Carlson’s and Royal’s motions. National Union has filed its own summary-judgment motion, asking this Court to rule that Carlson’s losses must be allocated across several policy periods. Nat’l Union Mem. Supp. Mot. Trigger & Alloc. (“NU SJM Trigger”) [Docket No. 75]. According to National Union, for each period during which Carlson engaged in loss-producing activity (i.e., patent infringement), the primary policy that was in effect during that period — and, if reached, the excess policies that were in effect during that period — must respond and cover the losses that resulted from Carlson’s activities during that period. On this theory, Carlson would pay multiple $l-million deductibles — one for each policy period. To the extent that more than $1 million in losses occurred during a particular policy period, Royal would then be responsible up to its policy limit of $5 million. National Union would not have any liability unless losses attributed to infringing activity during a single one-year policy period exceeded $6 million. The bottom line is that, under National Union’s theory, Carlson and Royal would each be responsible for roughly half of the Maritz losses, and National Union would pay nothing.

National Union’s argument is based primarily on Minnesota law governing generic CGL insurance. Carlson’s and Royal’s arguments, by contrast, are based on the specific language of the relevant insurance policies. Because that language is the primary source of the rights and duties of Carlson and its insurers, the Court begins its analysis there.

2. Policy Language Regarding Aggregating “Wrongful Acts”

The primary policy covers patent-infringement liability through the IPE. During all but one of the policy periods (2002-2003), the language of the key provisions of the IPE was identical. The Court will focus on this language, and then address the 2002-2003 policy period.

Two provisions of the IPE are at issue. The first, Subsection A, generally sets out the primary insurer’s coverage obligations as follows (relevant definitions from other sections of the IPE are quoted in brackets):

We shall reimburse the Insured for those sums which the insured becomes legally obligated to pay and shall have paid as “Damages” resulting from any “Claim” or “Claims” [“a demand for Damages”] made against the Insured for any “Wrongful Act” [“any violation of a legal right or rights associated with patents”] caused by [the] manufacture, use, development, distribution, advertising or sale of a “Covered Product” [“any product ... sold or any process used ... by the Insured”] committed by the Insured ... occurring within the term of this policy.

Policy Stip. Tab 1 at P00015-17; Tab 2 at P00096-98; Tab 5 at P00512-13; Tab 8 at P00760-61.

According to National Union, this clause is unambiguous. Nat’l Union Mem. Opp. PI. Mot. Partial S.J. at 8-15 (“NU Opp. Carlson SJM” [Docket No. 108]); see also NU SJM Trigger at 18-19. By its terms, Subsection A covers damages resulting from wrongful acts (like patent infringement) “occurring within the term of this policy.” Thus, each annual primary policy (e.g., the policy covering 1998-1999) covers only wrongful acts that occur during the term of that policy (e.g., during 1998-1999). Put differently, each wrongful act in a multi-year series of wrongful acts is covered by one and only one policy: the annual policy that was in effect when the wrongful act occurred.

If the IPE included only Subsection A, the Court would agree with National Union. But National Union ignores another provision of the IPE — Subsection E. Under Subsection E, at least some wrongful acts are aggregated with other wrongful acts for coverage purposes. Subsection E provides:

Regardless of the year in or policy under which “Claims” ... are made, any covered “Wrongful Act” which is the subject of a “Claim” ... involving the Insured and involving the same intellectual right and covered Product as any other “Wrongful Act” [sjhall be considered as a single ‘Wrongful Act[.]”

Policy Stip. Tab 1 at P00016; Tab 2 at P00097; Tab 5 at P00512; Tab 8 at P00760. All parties agree that under Subsection E, some wrongful acts must be aggregated, but they disagree as to which ones.

National Union focuses on the word “covered.” NU Opp. Carlson SJM at 10. Because Subsection E applies to “any covered Wrongful Act,” the Court must (on this argument) look back to Subsection A to determine which wrongful acts are “covered.” Since Subsection A covers only wrongful acts “occurring within the term of the policy,” Subsection E can aggregate only wrongful acts occurring during a single period. Thus, according to National Union, Subsection E mandates that all related wrongful acts that occurred during the policy period 1998-1999 are aggregated into a single wrongful act, and all related wrongful acts that occurred during the policy period 1999-2000 are aggregated into a single wrongful act, but the wrongful acts that occurred during the policy period 1998-1999 are not aggregated with the wrongful acts that occurred during the policy period 1999-2000.

The problem with National Union’s argument is that it begs the question — that is, it assumes as true what it ostensibly proves. National Union’s argument depends on the notion that only acts occurring within the policy period are “covered.” But if the very function of Subsection E is to consolidate extra-period acts with intraperiod acts, the consolidated extra-period acts would become, constructively, intraperiod acts — and thus be “covered” under the policy.

Carlson does not make this point. Instead, Carlson downplays the word “covered” in favor of Subsection E’s introductory prepositional phrase, “Regardless of the year in or policy under which ‘Claims’ ... are made.... ” Carlson SJM at 26-27; Carlson Reply at 4. Relying on this language, Carlson characterizes Subsection E as an “anti-stacking” or “deemer” clause. Carlson SJM at 8.

An anti-stacking clause of the type at issue in this case is designed to assign losses that occur over several policy periods to a single period for coverage purposes, thus preventing “stacking” of policy limits. See 12 Couch* on Ins. § 169:5 (3d ed.2005) (discussing stacking the limits of policies covering different time periods). Frequently such a clause will protect insurers, not policy holders, because it reduces an insurer’s total exposure to a single policy limit. In practice, however, an anti-stacking clause could also benefit the policy holder because it reduces the number of deductibles that the policy holder must pay. Which party will benefit from an anti-stacking clause — the insurer or the policy holder — will depend on the size of the covered loss, the amount of the deductibles, and the policy limits for each possibly relevant period.

Reading Subsection E as an anti-stacking clause, Carlson argues that all related wrongful acts, “[r]egardless of the year in or policy under which ‘Claims’ ... are made,” must be grouped together and considered a single wrongful act. In other words, as Carlson reads it, Subsection E mandates that a wrongful act that occurred during the policy period 1998-1999 is aggregated not only with related wrongful acts that occurred during that policy period, but also with related wrongful acts that occurred during the 1999-2000 policy period, the 2000-2001 policy period, and every other relevant policy period. Thus, Carlson is required to pay only a single $l-million deductible, Royal is required to pay only a single $5-million policy limit, and National Union is required to pay only a single $44-million policy limit. Carlson SJM at 26-28.

The Court finds the language of both Subsection A and Subsection E to be ambiguous. Indeed, because Subsection E was carelessly drafted, it is literally meaningless. Subsection E’s purpose is undoubtedly to aggregate multiple wrongful acts, but grammatically it fails to do so. The clause provides that “any covered Wrongful Act” — singular—“shall be considered as a single Wrongful Act” — also singular. Apparently “any covered Wrongful Act” should have been written “all covered Wrongful Acts. ”

Subsection E would require substantial redrafting to conform unambiguously to either party’s proposed interpretation. If National Union is right, Subsection E should read:

All Wrongful Acts that occur during the term of this policy and that involve the same intellectual right and covered product shall be considered a single Wrongful Act.

If Carlson is right, Subsection E should read:

All Wrongful Acts that involve the same intellectual right and covered product, whether those acts occur during or outside the term of this policy, shall be considered a single Wrongful Act.

But Subsection E says neither of these things. And when Subsection E is read together with Subsection A, the IPE is ambiguous enough to support either National Union’s interpretation or Carlson’s interpretation. Because the IPE is ambiguous, the Court must turn to Minnesota law on interpreting ambiguous insurance contracts.

3. The Contra Proferentem Canon

The general rule in Minnesota is that ambiguity in an insurance contract is construed against the insurer- — the contra proferentem canon. See, e.g., Widness, 635 N.W.2d at 518; St. Paul Fire & Marine Ins. Co. v. MetPath, Inc., 38 F.Supp.2d 1087, 1091-92 (D.Minn.1998). In an alternative formulation, ambiguity is construed in favor of finding coverage. Nordby v. Atl. Mut. Ins. Co., 329 N.W.2d 820, 822 (Minn.1983). In general, both formulations of the rule have the same effect. But in a case such as this — involving, as it does, anti-stacking clauses — construing ambiguity “against the insurer” can lead to a different result than construing ambiguity “in favor of coverage.” For that reason, it is difficult to apply the contra proferentem canon in a principled way in this (or any similar) case.

To illustrate: Suppose an insurance policy has a $4-million limit and a $2-million deductible. If the policy holder suffers losses of $1 million per year for four consecutive years ($4 million total), the insurer would prefer that the losses be allocated to each policy year and that the policy limits for each year be stacked. In that event, the insurer would never be liable. Although the total loss across all four years is $4 million, the loss for any particular year is only $1 million, all of which the policy holder would absorb through the yearly deductible. The policy holder, by contrast, would prefer that all losses be allocated to a single policy year, because after payment by the policy holder of a single $2-million deductible, the insurer would be on the hook for the remaining $2 million in damages.

Suppose, though, that under the identical policy, the policy holder instead suffers losses of $4 million per year for four consecutive years ($16 million total). Now the policy holder would prefer stacking while the insurer would prefer that all losses be allocated to a single policy year. If losses were allocated to one year, the insurer would pay $2 million (its $4-million policy limit less the $2-million deductible), while the policy holder would absorb $14 million (the $2-million deductible plus the $12 million in losses exceeding the policy limit). But if losses were allocated to all four policy years, the insurer would pay $8 million ($2 million per year for four years) as would the policy holder (a $2-million deductible per year for four years).

What would it mean to construe such a hypothetical policy “against the insurer”? There is no way to know in advance. Only after the size and timing of the losses were known would it be possible to determine whether the insurer would prefer stacking. But this would mean that identical language in the hypothetical policy would be construed differently in two cases presenting different facts.

This possibility of varying outcomes is defensible if the purpose of the contra proferentem canon is to place the risk of sloppy drafting (or strategic ambiguity) on the insurer. See Safeco Ins. Co. v. Lindberg, 380 N.W.2d 219, 222 (Minn.Ct.App.1986) (“The insurer drafted this policy. It had the opportunity to clearly identify coverage and exclusions. It did not do so. Thus, it must bear the consequences”), aff'd 394 N.W.2d 146 (Minn.1986); see also Richard A. Posner, The Law and Economics of Contract Interpretation, 83 Tex. L.Rev. 1581, 1608 (2005) (“The doctrine of contra proferent[e]m may still be a sensible tiebreaker, on the ground that the party who drafted the contract was probably in the better position to avoid ambiguities.”). Cf. Michael B. Rappaport, The Ambiguity Rule and Insurance Law: Why Insurance Contracts Should Not Be Construed Against the Drafter, 30 Ga. L.Rev. 171 (1995). At the same time, though, the possibility of such variance conflicts with the general principle that the same contractual language should be interpreted consistently across cases. See Restatement (Second) of Contracts § 211(2); Rappaport, 30 Ga. L.Rev. at 180 n. 17; see also Mich. Chem. Corp. v. Am. Home Assur. Co., 728 F.2d 374, 380 n. 7 (6th Cir.1984) (“[0]nce courts establish a legal rule, such as how the number of occurrences [under an insurance policy] is to be determined, any party is entitled to rely upon that rule in future litigation.”); Am. Commerce Ins. Brokers, 551 N.W.2d at 229-30 (criticizing on public-policy grounds a proposed interpretation of insurance-policy language about aggregating acts of embezzlement where the interpretation would have benefitted the policy holder in that case but harmed policy holders given different facts).

Moreover, construing this hypothetical policy in favor of coverage could, depending on what “coverage” means, result in a different outcome than construing the policy against the insurer. At least one court has held, in a case involving multi-year losses, that construing an ambiguous policy “in favor of finding coverage” meant allocating those losses to multiple policy periods — that is, stacking. Olin Corp. v. Certain Underwriters at Lloyd’s London, 468 F.3d 120, 129-32 (2d Cir.2006). The court in Olin observed that stacking under the circumstances of that case disadvantaged the insured, but the court did not think that this militated against its application of New York’s contra proferentem canon requiring that ambiguity be construed to “find coverage.” Id. at 129 n. 6 (“We acknowledge that generally the policy of construing provisions in favor of coverage benefits the insured and not the insurer. We find nothing in the case law to suggest, however, that we should employ a different rationale simply because of the anomalous circumstances in this case that result in the insurer defendant benefitting from finding coverage.” (citation omitted)). But it is hard to understand why an ambiguous policy should be construed in favor of “coverage” (taken to mean “number of available policy limits”) if that construction harms the policy holder, who is the victim, not the author, of the ambiguity.

In this case, applying the contra proferentem canon — whether that means construing the policy against the insurer or in favor of coverage — is difficult for the additional reason that this case involves both primary and excess coverage. It is the IPE that is ambiguous. But the IPE is part of the primary policy, and the primary policy provides only “fronting” coverage (because Carlson’s deductible equals its policy limit). In other words, the primary carrier' — which is responsible for drafting the ambiguous IPE — will never actually have to indemnify Carlson for losses under that IPE, no matter how the IPE is construed. To some extent, then, the primary insurer is indifferent as to whether IP damages are allocated to a single policy period or spread across several periods.

The primary carrier is not entirely indifferent, though. It would likely prefer stacking — that is, it would prefer that multi-year losses for patent infringement be counted against multiple policies. In that way, the IP-related losses absorbed by Carlson would erode more of the primary carrier’s aggregate policy limits. In other words, if seven years of infringement by Carlson were treated as a single wrongful act, the primary carrier’s aggregate exposure would be reduced by only $1 million. But if the same seven years of infringement were treated as seven wrongful acts (one for every policy year), then the primary carrier’s aggregate exposure would be reduced by $7 million. Either way, the primary carrier would not have to pay any money to Carlson, but the latter interpretation leaves the carrier in a better position than the former. So to the extent that the contra proferentem canon applies to the “dispute” between Carlson (who is a party) and the primary carrier that drafted the ambiguous IPE (who is not a party), the canon favors Carlson’s position that the Maritz losses should be allocated to a single policy period.

That said, the Court is reluctant to rule in Carlson’s favor based solely on the con tra proferentem canon, given the difficulties of applying that canon under the circumstances of this case. But, as the Court will now explain, when the canon is considered together with evidence of how the language of the IPE changed over time, it becomes clearer that Carlson’s losses should be allocated to only one policy period.

4. Changes in the IPE

Lumbermens became Carlson’s primary carrier in 1999. For the 1999-2000 policy year, Lumbermens provided coverage under the terms of a policy that had been issued by Reliance. Beginning in July 2000, Lumbermens covered Carlson under Lumbermens-issued policies, but the IPE in those policies was almost identical to the IPE that had appeared in the Reliance policies. That changed in July 2002, when Lumbermens amended the IPE significantly.

First, Lumbermens amended Subsection A. Instead of covering losses from “any ‘Wrongful Act’ ... occurring within the term of this policy,” Subsection A was amended to cover losses from “any ‘Wrongful Act’ ... which first occur [sic] during the term of this policy.” Policy Stip. Tab 6 at P00636 (emphasis added). Second, Lumbermens made a corresponding amendment to Subsection E. Lumbermens retained all of the language in the Reliance IPE (some typos included) but added a clause at the end of the section as follows (new language in italics):

Regardless of the year in or policy under which “Claims” ... are made, any covered “Wrongful Act” which is the subject of a “Claim” ... involving the insured and involving the same intellectual right and “Covered Product” as any other “Wrongful Act” [s]hall be considered as a single “Wrongful Act” and shall be covered only under the policy covering the period during which the “Wrongful Act” first occurred.

Policy Stip. Tab 6 at P00636 (emphasis added).

All of the parties agree that the 2002 Lumbermens version of Subsection E— whether it is called an anti-stacking or deemer or telescope clause (see supra note 5) — aggregates acts from multiple policy periods into a single act covered only under the policy in effect when the first of the aggregated acts took place. National Union argues that the Lumbermens amendment reveals that Carlson and its insurers knew how to aggregate acts across policy periods when they wanted to. NU Opp. Carlson SJM at 14. Thus, according to National Union, the fact that Subsection E of the 2002 Lumbermens IPE is so clearly an anti-stacking clause is compelling evidence that Subsection E of the Reliance IPE (which, as noted, appeared in Reliance and Lumbermens policies from July 1997 until July 2002) is not an anti-stacking clause.

The Court disagrees. In the Court’s view, the changes that Lumbermens made — and, more importantly, did not make — to the Reliance IPE hurt National Union more than they help. First, the grammatical subject of the sentence making up Subsection E of the Lumbermens IPE is “any covered ‘Wrongful Act’ ”— exactly as it is in Subsection E of the Reliance IPE. Yet according to National Union, the same adjective “covered” in this same noun phrase in the Reliance IPE dictates that Subsection E cannot operate to aggregate extra-period and intra-period acts. The better view is that the phrase “any covered “Wrongful Act’ ” does not rule out aggregating extra- and intra-period acts under the Reliance IPE any more than it does under the Lumbermens IPE.

Second, given how slightly Subsection E was changed by Lumbermens, it is hard to see how Subsection E of the Lumbermens IPE could be an anti-stacking clause (as the parties agree it is) if Subsection E of the Reliance IPE was not an anti-stacking clause (as National Union argues). Lumbermens changed Subsection E in only one respect: Lumbermens designated a particular policy period — the one during which the first wrongful act occurred — as the policy period to which the aggregated wrongful acts should be assigned. But the aggregating is accomplished by language that is carried over unchanged from the Reliance IPE. This is strong evidence that the Reliance IPE did aggregate intra-period and extra-period acts into one period but simply failed to specify which period. Lumbermens corrected that omission in 2002.

National Union, however, argues that a clause cannot be an anti-stacking clause if — like Subsection E of the Reliance IPE — the clause fails to specify to which policy period the aggregated losses should be assigned. NU Opp. Carlson SJM at 12-13. According to National Union, one of the “hallmarks” of anti-stacking clauses is that they specify which policy period should receive aggregated losses. Id. at 12. Such specificity may be a hallmark of better-drafted, complete anti-stacking clauses. But other endorsements from the Reliance policy itself indicate that insurers sometimes draft incomplete anti-stacking clauses.

The 1997 Reliance policy includes, in addition to the IPE, an endorsement that provides professional-liability insurance. Policy Stip. Tab 1 at P0023-26. A clause in Section IV of that endorsement governs “related acts claims.” That clause defines “interrelated and/or related wrongful acts” and provides that “whenever occurring [they] shall be considered as one (1) wrongful act....” Id. at P00026. That same clause also says that claims based on related wrongful acts “shall be deemed made within the policy period in which the earliest of such wrongful acts occurred .... ” Id. There can be no doubt, then, that Section IV of this endorsement includes a classic anti-stacking clause.

The 1998 Reliance policy includes an endorsement for employment-practices liability not found in the earlier policy. Policy Stip. Tab 2 at P00104-06. Like the professional-liability endorsement in the 1997 policy, the employment-practices-liability endorsement in the 1998 policy includes a section titled “related acts claims.” In that section, “interrelated and/or related wrongful acts” are defined exactly as they are defined in the 1997 professional-liability endorsement, down to the provision that such related acts “whenever occurring shall be considered as one (1) wrongful act[ ]....” Id. at P00106. But the 1998 employment-practices-liability endorsement omits the language from the 1997 professional-liability endorsement about when events should be deemed to occur. Given that both endorsements include identical language that indisputably requires the aggregation of related acts, only one explanation makes sense: The 1997 professional-liability endorsement includes a complete anti-stacking clause, while the 1998 employment-practices-liability endorsement includes an incomplete anti-stacking clause.

In short, a comparison between the 1997 professional-liability endorsement and the 1998 employment-practices-liability endorsement demonstrates that Reliance inserted incomplete anti-stacking clauses into its policies. It is thus not startling that the Reliance IPE also contained an incomplete anti-stacking clause — that is, a clause that did aggregate related acts (both intra-period and extra-period) into a single wrongful act, but that failed to assign the resulting single wrongful act to a particular policy period. Moreover, the incomplete anti-stacking clause in the employment-practices-liability endorsement— just like Subsection E of the IPE' — was revised by Lumbermens in 2002 to cure this omission by including explicit “deem-er” language. Id. Tab 6 at P00639.

Finally, recall that National Union took over from Lumbermens as Carlson’s primary insurer in May 2003. Subsection E of the IPE in the National Union policy is identical to Subsection E in the Reliance policy. Compare id. Tab 1 at P00016 with id. Tab 8 at P00760. In other words, the “deemer” language that was added to the Reliance IPE by Lumbermens in 2002 was stripped out by National Union in 2003. But National Union has presented no evidence that it intended to provide substantively different IP coverage under its 2003 policy than Lumbermems did under the 2002 policy. Occam’s razor counsels us to look for the simplest explanation. Here, the simplest explanation for the change is that National Union considered the Lumbermens language and the Reliance language essentially equivalent. It therefore saw no reason to carry the Lumbermens language forward.

5. Royal’s Non-Cumulation Clause

Royal argues that, entirely apart from how the IPE should be construed, its excess-insurance policy includes a non-cumulation clause that caps Royal’s liability at a single policy limit. Royal SJM at 20-26; Royal Resp. Carlson SJM at 2-3. The clause, entitled “Prior Insurance and NonCumulation of Liability,” reads:

It is agreed that if any loss is also covered in whole or in part under any other excess policy issued to the Insured prior to the inception date hereof, the Company’s limit of liability as stated in the Declarations shall be reduced by any amounts due to the Insured on account of such loss under such prior insurance.

Policy Stip. Tab 9 at R105.

If Royal had issued a series of one-year policies each including this clause, the Court would be inclined to agree with Royal. But Royal initially issued its policy for the three-year “policy period” from July 1, 1992 to July 1, 1995. Id. at R001. It appears that the policy was then renewed, rather than reissued, through 2003. See id. at R001 (showing policy number HN 005285 on original declarations form), R002 (listing policy number RHN005285 on declarations form effective July 1, 2002).

By its terms, the non-cumulation clause applies only if losses were covered under an excess policy issued “prior to the inception date hereof.” Royal argues, in effect, that each annual renewal created a new inception date. Royal Reply Supp. Mot. Partial S.J. Policy Limits at 9-15 [Docket No. 158]; S.J. Hr’g Tr. at 22:18—25:2, Oct. 10, 2006 (rough draft, attached hereto as Appendix A). But the Minnesota Court of Appeals rejected a similar argument in Safeco Insurance Co. v. Lindberg, 380 N.W.2d 219 (Minn.Ct.App.1986). The insurer in that case argued that “inception refers to the first day of a renewal period.” Id. at 221. The court disagreed, after reading the policy at issue in light of the general rule that “ ‘[w]hether the renewal of a policy constitutes a new and independent contract or continuation of the original contract primarily depends upon the intention of the parties as ascertained from the instrument itself.’ ” Id. (quoting 18 Couch on Ins. § 68.40 (2d ed.1983)).

Here the “instrument itself’ — the Royal policy — indicates that each renewal was the continuation of the original policy, not a new policy. The policy does not define the term “policy period,” but all of the declarations from various years feature policy number RHN005285 (if they feature a number at all), the same number that appears on the original declarations page. See, e.g., Policy Stip. Tab 9 at R001, R034, R054, R093. Moreover, not only do the “Schedule of Underlying Insurance” pages from 1995 through 2002 bear the same policy number (RHN005285), but the schedules for these different years are numbered sequentially as “Revision 1” through “Revision 9.” Id. at R002-18.

The Court finds, therefore, that Royal’s policy has only one inception date — July 1, 1992 — and the non-cumulation clause does not cap Royal’s liability at a single policy limit because the Maritz-related losses are not “covered in whole or in part under any other excess policy issued to the Insured prior to [July 1, 1992].” Nonetheless, Royal’s exposure is limited to a single policy limit in light of the Court’s ruling with respect to the IPE.

In sum, the Court agrees with Carlson that, under the IPE, the Maritzrelated losses must be aggregated into a single policy period. Carlson is responsible for paying a single $l-million deductible per related wrongful act; it more than satisfied this responsibility by paying $3,385 million in defense costs. As to the $12 million Carlson paid to settle the Maritz lawsuit (i.e., setting aside defense costs), Royal is responsible for paying a single $5-million policy limit per related wrongful act, and National Union is responsible for the rest. The Court will therefore grant Carlson’s motion for summary judgment on this issue, deny National Union’s motion for summary judgment on “trigger and allocation,” and deny as moot Royal’s motion for summary judgment as to this issue.

6. Assigning a Policy Period

Having found that the Maritz-related losses must be aggregated into a single policy period, the Court must now decide to which policy period the losses must be assigned.

The 2002-2003 period can quickly be eliminated as a candidate. The Lumbermens IPE — as amended in 2002 — plainly excludes coverage, since the series of acts underlying the Maritz suit (and aggregated by virtue of the IPE) “first occurred” in 1997.

As to the remaining periods, Carlson’s position is not entirely clear. On the one hand, Carlson argues in one of its briefs both that the Maritz losses “trigger[] a single limit and deductible” and that “six policy periods are triggered and damages can be allocated accordingly.” Carlson Reply at 8. The Court fails to understand what it would mean to allocate damages across several periods while applying only a single limit and deductible. On the other hand, at the summary-judgment hearing, Carlson seemed to say that because both Royal and National Union provided the same coverage with the same policy limits throughout the period from July 1997 to June 2003, it makes no difference to which policy period the Maritz losses are allocated. S.J. Hr’g Tr. at 42:11-16, Oct. 10, 2006 (rough draft, attached hereto as Appendix B) (“Carlson doesn’t care what policy they put it in. Carlson doesn’t say it should be allocated across the six policies ____ [T]he court doesn’t have to reach that issue in this case.”). The Court, however, has interpreted the IPE to aggregate losses in part because the Court concluded that the 2002 Lumbermens IPE made explicit what was already implicit in the Reliance IPE. Moreover, the Court is obligated under Minnesota law to construe the policy in light of how a “reasonable person in the position of the insured would have understood” the policy. Canadian Universal, 258 N.W.2d at 572; Soo Line, 694 N.W.2d at 113.

Accordingly, the Court holds that, based on the language of the IPE, an insured would reasonably understand that losses related to a series of wrongful acts would be allocated to the policy period in which the first of those acts occurred. Because the infringing acts challenged in the Maritz suit allegedly began in November 1997, the Court holds that the losses incurred by Carlson in defending and settling that suit are attributable to the 1997-1998 policy period.

C. Nature and Number of “Wrongful Acts” at Issue in Maritz

To this point, the Court has held that Subsection E of the IPE aggregates related wrongful acts — both those that occur during the policy period and those that occur outside of the policy period— and that losses occasioned by a series of related wrongful acts are assigned to the policy period in which the first of those acts occurred. The Court must now determine how related two or more wrongful acts must be before they can be aggregated under Subsection E. Carlson asks this Court to rule that all of the acts that gave rise to the Maritz suit were sufficiently related to be aggregated into one “wrongful act” under Subsection E. National Union naturally resists Carlson’s characterization, as the more deductibles that Carlson must pay — and the more times that Royal’s first-level excess policy must respond — the less money that National Union will have to pay under its second-level excess policies.

Subsection E of the IPE aggregates only those wrongful acts that “involvfe] the same intellectual right and covered Product.” See, e.g., Policy Stip. Tab 1 at P00016. Whether wrongful acts are sufficiently related to be aggregated under Subsection E thus depends on the answers to two questions: (1) Do the wrongful acts involve the same “intellectual right”? and (2) Do the wrongful acts involve the same “covered product”? The Court will address each question in turn.

1. Number of “Intellectual Rights”

No one disputes that the Maritz suit involved two patents. Carlson, however, argues that those two patents covered only one “intellectual right.” Carlson S JM at 18-22; Carlson Reply at 14 n. 11. The Court agrees.

The IPE does not define “intellectual right.” But at least with respect to patent infringement, the scope of the term seems fairly clear. A single suit for infringing two patents covering very different inventions (say, a heating element and an electrical plug) would involve two different intellectual rights, even if a single product (say, a space heater) infringed both patents. At the same time, a suit for infringing a single patent would involve only one intellectual right, even if multiple claims of that single patent were asserted against the accused product (as they typically are). This is so even though infringement must be assessed claim by claim (and, for each particular claim, limitation by limitation). Based on the nature of patents and patent litigation, no policy holder and no insurer would expect that each separate claim in a single asserted patent would be treated as a different “intellectual right” under the IPE.

In the Maritz suit, one of the two patents asserted against Carlson (the ’695 patent) was a continuation of the other (the TOO patent) — -that is, both patents resulted from the same initial application. Carlson SJM App. M at CMG02500. The two patents share the identical written description and drawings; they differ only in their claims. Carlson SJM at 19. Such continuation patents are common. See Manual of Pat. Exam. Proc. § 201.07.

The Court is not prepared to — and need not — declare that, for purposes of the IPE, every continuation patent necessarily involves the same “intellectual right” as a related patent. In the Maritz suit, however, the patentee agreed to file a so-called terminal disclaimer with respect to the ’695 patent. Carlson SJM App. R at 10-11. Indeed, the patentee argued that a de facto terminal disclaimer already existed. Id. Based on the patentee’s willingness to file a terminal disclaimer as to the ’695 patent, the Court finds that the TOO patent and the ’695 patent covered the same intellectual right.

Terminal disclaimers are a means of overcoming what is called “non-statutory double patenting,” a doctrine under which a continuation patent can be held to be obvious (and therefore invalid) in light of a related patent that issued earlier from the same application as the continuation patent. See Geneva Pharms., Inc. v. GlaxoSmithKline PLC, 349 F.3d 1373, 1377-78 (Fed.Cir.2003). By means of a terminal disclaimer, the patentee gives up (“disclaims”) any portion of the continuation patent’s term that exceeds the term of the related patent. Id.; 37 C.F.R. 1.130; Manual of Pat. Exam. Proc. § 718. As a result, the two patents are effective for the exact same length of time. A terminal disclaimer is a concession by the patentee that two patents that resulted from the same application should be treated as a single patent, effective for a single patent term. Put differently, the claims of the continuation patent are treated as i/they appeared in the first patent.

If the continuation claims had appeared in the first patent, no one could dispute that all of the claims in that one patent would have represented a single “intellectual right” under the IPE. Because a terminal disclaimer constructively consolidates two patents, the result should be no different where a patentee files, or is prepared to file, a terminal disclaimer as to a continuation patent. Accordingly, the claims of the TOO patent and the claims of the ’695 patent together constitute a single “intellectual right” under the IPE.

2. Number of “Covered Products”

Under Subsection E of the IPE, only wrongful acts that involve “the same ... covered Product” can be aggregated into a single wrongful act. See, e.g., Policy Stip. Tab 1 at P00016. The IPE defines a covered product as “any product ... sold or any process used” by the policy holder. Id.

Carlson argues that the Maritz suit involved only one “covered Product.” Carlson SJM at 22-26. Carlson concedes that the TravPass incentive program and the BOE incentive program may have appeared to be separate products to Carlson’s customers. But Carlson relies heavily on the fact that the IPE defines a “covered Product” to include not just “any product ... sold,” but also “any process used.” Id. at 24; Carlson Reply at 14 & n. 12. Carlson argues that the TravPass and BOE incentive programs used the same patented “process” and therefore are constructively the same “covered Product” for purposes of the IPE. Id.

National Union disagrees. According to National Union, there is a factual dispute over whether TravPass and BOE involve different infringing processes. NU Opp. Carlson SJM at 22-26. The evidence on which National Union relies, however, fails to create any genuine issue of fact.

First, National Union emphasizes that TravPass and BOE were operated on two different computer systems. NU Opp. Carlson SJM at 22-23. But this is not true in any meaningful way. If TravPass operated on one system and BOE operated on another, that would be some evidence tending to show that they used different processes. In fact, however, they both operated on a first computer system from 1998 to 2000, then they both operated on a second computer system (called “Card-Link”) from 2000 on. Id. at 23. Far from suggesting that TravPass and BOE used different processes, this suggests that they used the same process — and that, if that process changed over time, it changed in the same way for both TravPass and BOE.

Second, National Union relies on differences between TravPass and BOE identified in a Carlson document that outlines the “processes” involved in each of the different award programs (TravPass, BOE, and others) running on Carlson’s CardLink computer system. Id. at 23-26. These differences, however, are irrelevant to the question of coverage under the IPE.

The IPE must be interpreted in light of its function. As it relates to this case, the function of the IPE is to provide coverage for patent infringement. Under the IPE, infringement losses arising from a single “process used” by the policy holder are treated as arising from a single “wrongful act.” The key question, then, is whether the policy holder’s activities involve the same infringing process, regardless of what other processes those activities might also involve.

To illustrate: Suppose that a policy holder who owns a restaurant — call it Eggcorp — infringes a method patent having two steps. Step 1 reads, “mix eggs and milk.” Step 2 reads, “scramble mixture.” Eggcorp serves the resulting infringing scrambled eggs in three ways: (1) by themselves; (2) with ketchup; and (3) with cheese. In the second and third cases, the scrambled eggs involve an additional step not mentioned in the patent (adding ketchup or adding cheese). But all of the scrambled-egg dishes — the plain eggs, the eggs with ketchup, and the eggs with cheese — infringe the same patent. In other words, Eggcorp practices the “same process” — steps 1 and 2 of the method patent — in making all three dishes, even though making the eggs with ketchup also involves a third step (adding ketchup), as does making the eggs with cheese (adding cheese).

A similar point can be made here. The processes identified in the CardLink document on which National Union relies are like adding ketchup or cheese to our hypothetical scrambled eggs. National Union has failed to show how the “process” differences identified in the CardLink document are relevant to the patented process that both TravPass and BOE infringed. The mere fact that certain differences between TravPass and BOE are associated with what are labeled “proc