Amica Mutual Insurance v. Fogel

656 F.3d 167, 2011 U.S. App. LEXIS 18623, 2011 WL 3930285

• Court: Court of Appeals for the Third Circuit
• Decided: September 8, 2011
• Docket: 10-3611
• Status: Published

Opinion

OPINION OF THE COURT

AMBRO, Circuit Judge.

The principal issue in this case is whether New Jersey or Pennsylvania law applies to an automobile insurance dispute between Arnica Mutual Insurance Company (“Arnica”) and Edward and Maureen Fogel, individually and as guardians ad litem of Marcy Fogel and Carrie Fogel, and as administrators of the estate of Melissa Fogel (collectively, the “Fogels”). Arnica issued the policy to the Fogels when they were residents of New Jersey. During its term, the Fogels moved to Pennsylvania, and made Arnica aware of their permanent relocation, before they were involved in a fatal traffic accident in Pennsylvania that triggered the policy claim to Arnica. The District Court granted declaratory relief to Arnica at the summary judgment stage, believing that New Jersey law applied to the contract. We conclude that Pennsylvania’s choice-of-law rules do not apply; instead, we look to New Jersey’s choice-of-law rules, and they point to Pennsylvania law as governing this dispute. We remand so that summary judgment may be entered for the Fogels on the choice-of-law issue. However, we affirm the District Court’s grant of summary judgment to Arnica on the Fogels’ counterclaim alleging that it engaged in a bad faith denial of insurance coverage.

I. Facts and Procedural History

A. Facts

Arnica insured the Fogels under an automobile insurance policy (the “policy” or “contract”) effective from December 1, 2007 through December 1, 2008. The Fogels lived in Howell, New Jersey when the policy was issued. It provided underinsured motorist coverage in the amount of $300,000 for each accident. The Fogels moved to Pottsville, Pennsylvania in August 2008.

The same month of their move, Mr. Fogel called Arnica to advise his insurer that he had moved from New Jersey to Pennsylvania with the intent to remain there permanently. An Arnica representative informed Mr. Fogel that his policy would need to be rewritten in Pennsylvania. A few days later, Mr. Fogel spoke with an Arnica senior account representative and asked that his policy be converted to, or reissued as, a Pennsylvania policy, due to his family’s relocation to Pennsylvania. The Arnica representative told Mr. Fogel that Arnica would not issue a Pennsylvania policy until he obtained a Pennsylvania driver’s license and registered his two vehicles in Pennsylvania, and that until that time his New Jersey policy would remain in effect. She also obtained Mr. Fogel’s new home address in Pennsylvania. Arnica began billing the Fogels for their insurance premiums at their residence in Pennsylvania in September 2008. Mr. Fogel did not obtain a Pennsylvania driver’s license or register his vehicles in Pennsylvania until early 2009.

In October 2008, Mr. Fogel and three children in the Fogel family were in a serious automobile accident in Pennsylvania. Them vehicle was struck head-on by a motorist who was allegedly intoxicated— one of the Fogel daughters died as a result of the accident and Mr. Fogel and the other two Fogel daughters were seriously injured. The driver of the vehicle that struck the Fogels had liability insurance with a limit of $100,000, which has been paid to the Fogels in settlement of their claims against him. At the time of the accident, the Fogels’ policy with Arnica had not been rewritten or reissued as a Pennsylvania policy.

After the accident, Arnica established a New Jersey personal injury protection (“PIP”) file for each of the four injured Fogel family members. It then processed their medical expenses pursuant to New Jersey PIP coverage, which, under New Jersey law, had limits of $250,000 per person per accident.

The Fogels, however, believed that they were entitled to more. In December 2008, their counsel wrote a letter to Arnica requesting payment up to the limits of the “stacked” underinsured motorist (“UIM”) benefits available under their policy as a result of the accident. Under Pennsylvania law, insureds whose policies cover more than one vehicle may “stack” UIM benefits to cover losses that result from a collision with an at-fault underinsured driver — that is, the insured may add together the policy limits for each of the covered vehicles even though the collision involved only one of the covered vehicles. Stacking is available in Pennsylvania unless the insured expressly waives that right in writing, but it is not permitted in New Jersey. Moreover, insurers in Pennsylvania may not offset UIM benefits by amounts insureds receive from other insurance sources, whereas such offsets are permitted in New Jersey.

Arnica’s position in response was “that ‘stacking’ [of UIM benefits on the Fogels’ two automobiles] is prohibited under the New Jersey policy.” Counsel for the Fogels again wrote to Arnica stating that, under Pennsylvania law, the Fogels are entitled to stacked UIM benefits of up to $600,000 (as their policy covered two cars with limits of $300,000 each). After receiving that letter, Arnica’s claims adjuster confirmed that the Fogels had been residing in Pennsylvania since August 2008. He also confirmed that Mr. Fogel communicated with Arnica before the accident regarding his relocation to Pennsylvania.

B. Procedural History

In January 2009, Arnica filed an action in the Superior Court of New Jersey seek ing a declaration that its obligations under the policy are only those that it owes under New Jersey law (not Pennsylvania law). The Fogels successfully petitioned for removal to the District Court for the District of New Jersey. On the District Court’s own motion, the action was transferred to the Middle District of Pennsylvania pursuant to 28 U.S.C. § 1404(a), which provides that “[f]or the convenience of parties and witnesses, in the interest of justice, a district court may transfer any civil action to any other district or division where it might have been brought.”

Arnica filed an amended complaint in July 2009 seeking a declaration that it owed no further UIM benefits to the Fogels. In the alternative, if the District Court applied Pennsylvania law, Arnica sought return of payments it had made for New Jersey PIP benefits in excess of the amount it would have provided under Pennsylvania law ($5,000 per person).

After discovery, the Fogels filed an amended answer and Arnica moved for summary judgment on the choice-of-law issue. The Fogels filed a cross-motion for summary judgment on the same issue and made a counterclaim alleging bad faith by Arnica’s claims adjuster under 42 Pa.C.S.A. § 8317. Arnica also moved for summary judgment on the Fogels’ bad faith claim.

The motions and counterclaims were submitted to a Magistrate Judge, who prepared a Report and Recommendation (“R & R”) for the District Court. The R & R applied Pennsylvania’s choice-of-law rules and concluded that New Jersey law applies to the automobile insurance policy. Thus, the R & R suggested that Arnica’s motion for summary judgment be granted and the Fogels’ cross-motion for summary judgment be denied. It also recommended that summary judgment be granted in favor of Arnica on the Fogels’ counterclaim for bad faith.

The District Court adopted the Magistrate’s R & R, granted Arnica’s motion for summary judgment, and denied the Fogels’ cross-motion for summary judgment. See District Court Memorandum of July 29, 2010 (“Dist. Ct. Memo”). 1 The Fogels appealed.

We. have jurisdiction under 28 U.S.C. § 1291. We review an order granting summary judgment de novo. Gardner v. State Farm Fire and Cas. Co., 544 F.3d 553, 557-58 (3d Cir.2008). We also exercise plenary review over a district court’s determination of which state’s substantive law governs in a civil action based on diversity of citizenship. Berg Chilling Sys., Inc. v. Hull Corp., 435 F.3d 455, 462 (3d Cir.2006) (citing Garcia v. Plaza Oldsmobile Ltd., 421 F.3d 216, 219 (3d Cir.2005)).

II. Analysis

A. Choice-of-Law Framework

In an action based on diversity of citizenship, a federal court generally applies the choice-of-law rules of the jurisdiction in which it sits. Klaxon Co. v. Stentor Elec. Mfg. Co., 313 U.S. 487, 496, 61 S.Ct. 1020, 85 L.Ed. 1477 (1941); Hammersmith v. TIG Ins. Co., 480 F.3d 220, 226 (3d Cir.2007) (applying Klaxon)-, Garcia, 421 F.3d at 219 (same). However, Van Dusen v. Barrack established that when a civil action is transferred from one district court to another pursuant to § 1404(a) on motion of the defendant, the transferee forum must apply the law of the initial forum. 376 U.S. 612, 639, 84 S.Ct. 805, 11 L.Ed.2d 945 (1964) (“[W]here the defendants seek transfer [under § 1404(a) ], the transferee district court must be obligated to apply the state law that would have been applied if there had been no change of venue”). The Supreme Court extended the Van Dusen rule to transfers initiated by the plaintiff in Ferens v. John Deere Co., 494 U.S. 516, 110 S.Ct. 1274, 108 L.Ed.2d 443 (1990). Thus, after transfer by either party, when “[fjaced with a choice-of-law question, federal courts in the district to which the case has been transferred under § 404(a) must apply the law of the transferor state.” Lafferty v. St. Riel, 495 F.3d 72, 76-77 (3d Cir.2007) (citing Van Dusen, 376 U.S. at 639, 84 S.Ct. 805, and Ferens, 494 U.S. at 527-28, 110 S.Ct. 1274).

Ferens also makes plain that the Van Dusen rule applies to sua sponte transfers, such as occurred here. “[T]he transferor law should apply regardless of who makes the § 1404(a) motion,” in part because a contrary rule “would leave unclear which law should apply when both a defendant and a plaintiff move for a transfer of venue or when the court transfers venue on its own motion.” Ferens, 494 U.S. at 530-31, 110 S.Ct. 1274 (emphasis added). We agree with the Ninth Circuit Court of Appeals that “[t]he language and reasoning in Ferens leave no doubt that the [Van Du- sen] rule equally is applicable where a district court transfers an action sua sponte.” Muldoon v. Tropitone Furniture Co., 1 F.3d 964, 966 (9th Cir.1993); see also In re Helicopter Crash Near Weaver-ville, California 8/5/08, 714 F.Supp.2d 1098, 1102 (D.Or.2010) (construing Ferens as “finding that whether the transferor court grants a [§ ] 1404(a) transfer sua sponte, at the defendant’s request, or at the plaintiffs urging, ‘the transferee court must follow the choice of law rules of the transferor court.’”). Thus, we hold that Ferens compels application of the Van Du-sen rule to sua sponte transfers pursuant to § 1404(a).

This civil action was brought in New Jersey state court. It was then removed to the District of New Jersey. That Court sua sponte transferred the case to the Middle District of Pennsylvania under § 1404(a). Likely because neither party referred to Van Dusen or Ferens, the District Court in the Middle District of Pennsylvania did not apply the choice-of-law rules of the transferor forum (New Jersey). Instead, it applied Pennsylvania’s choice-of-law rules rather than New Jersey’s. We proceed by applying New Jersey’s choice-of-law rules to determine whether Pennsylvania or New Jersey substantive law controls the merits of the dispute before us.

B. New Jersey Choice-of-Law Approach

In State Farm Mutual Automobile Insurance Company v. Estate of Simmons, 84 N.J. 28, 36-37, 417 A.2d 488 (N.J.1980) (“Simmons”), the New Jersey Supreme Court “rejected the mechanical and inflexible lex loci contractus [law of the place where the contract is made] rule in resolving conflict-of-law issues in liability-insurance contracts,” as it later explained in Gilbert Spruance Company v. Pennsylvania Manufacturers’ Ass’n Insurance Co., 134 N.J. 96, 629 A.2d 885, 888 (N.J.1993) (“Gilbert”). Since Simmons, “[New Jersey] courts have adopted a more flexible approach that focuses on the state that has

the most significant connections with the parties and the transaction .” Gilbert, 629 A.2d at 888 (citation omitted). Gilbert explained that approach, known as the “most significant relationship” test in the Restatement (Second) of Conflict of Laws (1971) (“Restatement ”), as follows:

[In Simmons, w]e held that because the law of the place of contract “generally comport[s] with the reasonable expectations of the parties concerning the principal situs of the insured risk,” ... that forum’s law should be applied “unless the dominant and significant relationship of another state to the parties and the underlying issue dictates that this basic rule should yield.” ... In making that determination, courts should rely on the factors and contacts set forth in Restatement sections 6 and 188.

Id. (citing Simmons, 417 A.2d at 492-93).

Simmons emphasized the importance of considering the parties’ expectations regarding the principal location of the insured risk as the rationale for looking to the place of contracting. However, it also recognized that “this choice-of-law rule should not be given controlling or dispositive effect.” Simmons, 417 A.2d at 493. That is, “[i]t should not be applied without a full comparison of the significant relationship of each state with the parties and the transaction.” Id. Moreover, “[t]hat assessment should encompass an evaluation of important state contacts as well as a consideration of the state policies affected by, and governmental interest in, the outcome of the controversy.” Id.

Though Simmons stopped short of adopting the Restatement expressly, Gilbert and a series of later New Jersey Appellate Division cases made clear that “the governmental interest test [dealt with below] and the ‘most significant relationship’ standard of the Restatement” provide the applicable choice-of-law framework in New Jersey courts. N.J. Mfrs. Ins. Co. v. MacVicar, 307 N.J.Super. 507, 512, 704 A.2d 1343, 1346 (N.J.Super.App.Div.1998) (describing Gilbert as “emphasizing again that[,] with respect to insurance contracts, the law of the place understood by the parties to be the principal location of the risk controls unless some other state has a more significant relationship”). See also Canal Ins. v. F.W. Clukey Trucking, 295 N.J.Super. 131, 684 A.2d 953 (N.J.Super.App.Div.1996); Hertz Claim Mgmt. v. Marchetta, 281 N.J.Super. 190, 656 A.2d 1298 (N.J.Super.App.Div.1995); Chalef v. Ryerson, 277 N.J.Super. 22, 648 A.2d 1139 (N.J.Super.App.Div.1994); Pittston Co. v. Allianz Ins. Co., 795 F.Supp. 678, 683 (D.N.J.1992) (collecting cases and explaining that “New Jersey Appellate Division decisions have made clear that the Restatement’s “most significant relationship” test is the law of New Jersey”).

Under the Restatement approach, “the general rule in contract actions is that the law of the state with the most significant relationship to the parties and the transaction under the principles stated in Restatement section 6 governs.” Gilbert, 629 A.2d at 888 (citing Restatement § 188). In determining which state has the most significant relationship, § 188 instructs courts to evaluate each state’s contacts, “according to their relative importance,” such as the place of contracting and performance, the location of the subject matter of the contract, and the domicile, residence, nationality, place of incorporation and place of business of the parties. Restatement § 188(2). Section 6 of the Restatement instructs courts to consider the following factors: “(a) the needs of the interstate and international systems, (b) the relevant polices of the forum, (c) the relevant policies of other interested states and the relative interests of those states in the determination' of the particular issue, (d) the protection of justified expectations, (e) the basic policies underlying the particular field of law, (f) certainty, predictability and uniformity of result, and (g) ease in the determination and application of the law to be applied.” Restatement § 6; see also Gilbert, 629 A.2d at 888; Simmons, 417 A.2d at 491.

Section 193 of the Restatement “provides guidance in applying section 188’s ‘relevant contacts’ to the special case of casualty-insurance contracts.” Gilbert, 629 A.2d at 888. As the more specific Restatement provision, § 193 is the starting point “in determining the choice-of-law rule to govern casualty-insurance contracts.” Id. at 893. It provides that “the law of the state that ‘the parties understood was to be the principal location of the insured risk [governs unless] some other state has a more significant relationship under the principles stated in § 6.’ ” Id. at 889 (alteration in original) (citing Restatement § 193). Comment b to Restatement § 193 explains that the location of the insured risk under the contract should generally be given “greater weight than any other single contact,” as Simmons and Gilbert also instruct. However, the importance of the principal location of the insured risk diminishes when the insured object is a moveable chattel, such as a motor vehicle. In those cases, “the significance of the state of the risk’s principal location diminishes with the length of time that it can be anticipated the chattel will be in other states during the term of the insurance.” Restatement § 193 cmt. b. Furthermore, on “occasions when following the issuance of the policy the principal location of the risk is shifted to some other state,” the “other state will have a natural interest in the insurance of the risk and it may be that its local law should be applied to determine at least some issues arising under the policy.” Id. cmt. d.

C. Application of New Jersey Choice-of-Law Rules to this Dispute

The first step in our analysis is to determine whether an actual conflict exists between the laws of New Jersey and Pennsylvania. As the Magistrate Judge’s R & R noted, the laws of these states conflict as follows in this case:

Pennsylvania law permits insureds to “stack” underinsured motorist benefits, whereas New Jersey does not. Additionally, ... Pennsylvania law does not permit insurers to offset UIM benefits against amounts received from other insurance sources, whereas New Jersey does. There is thus no question that the differences in each state’s laws with respect to automobile insurance — specifically, with respect to how much coverage is legally available to insureds for underinsured motorist benefits — give rise to a true conflict...:

R & R at 18. We agree, as did the District Court, with the Magistrate Judge’s conclusion that an actual conflict exists.

In this context, we follow the approach from Simmons, Gilbert, and subsequent New Jersey cases that apply the Restatement standards discussed above. Although the Fogels did not ask us to apply New Jersey choice-of-law rules to this dispute, they argue that we should follow the reasoning of MacVicar, a New Jersey Appellate Division case that is strikingly similar to the case before us. The District Court declined to follow MacVicar, as it applied Pennsylvania’s choice-of-law rules rather than New Jersey’s. However, once we recognize that New Jersey’s choice-of-law rules apply, it is clear that MacVicar is the most closely analogous case to the current dispute and we are bound by it as the law of the transferor forum.

MacVicar decided “whether the law of New Jersey or Pennsylvania govern[ed] the determination of defendant-insureds’ entitlement to stacking of the underinsured motorist benefits afforded by the automobile policy issued to them by [the] plaintiff [insurer,] New Jersey Manufacturers Insurance Company (NJM).” MacVicar, 704 A.2d at 1344. “The issue is simply one of choice of law....” Id. at 1346.

The critical facts of MacVicar mirror closely those of our case. Like the Fogels, the MacVicars lived in New Jersey when the disputed automobile insurance policy was issued for their family cars, all of which were garaged in New Jersey at the time. Id. at 1344. Also like the Fogels, the MacVicars moved to Pennsylvania during the term of their policy. Id. 2 In perhaps the most important resemblance between MacVicar and our case, the insurer received notice from the MacVicar family of their relocation to Pennsylvania by telephone before the accident that triggered possible application of the policy. Subsequent telephone communication between Mrs. MacVicar and the insurer, NJM, occurred regarding the process of converting the family’s New Jersey automobile insurance policy to a Pennsylvania policy. Id. Similarly, after Mr. Fogel’s initial call to Arnica providing notice of his family’s relocation and being told of the need to have the policy rewritten in Pennsylvania, he subsequently spoke with an Arnica senior account representative regarding how to complete that process.

Again tracking closely to the facts of the case before us, the MacVicars’ New Jersey policy was still in force when the family was involved in a tragic accident that killed one family member (Mr. MacVicar), seriously and permanently injured another (Mrs. MacVicar), and less seriously injured the third passenger (their daughter). Id. at 1345. As the Fogels’ lawyer did, the MacVicars’ “lawyer asserted that the family was entitled to stacked UIM coverage.” Id. In response “NJM asserted that since the New Jersey policy was in effect, New Jersey’s anti-stacking law applied,” id., as Arnica now asserts.

The procedural history in MacVicar unfolded in much the same manner as here, except that the case was never removed to federal court from New Jersey state court. The insurer, NJM, brought an action seeking a declaration that New Jersey law applied to the policy and that stacking was impermissible. Id. “On motion and cross-motion for summary judgment, the trial court agreed with NJM’s position,” and entered summary judgment for the insurer based on the choice-of-law issue. Id.

The New Jersey Appellate Division reversed the trial court’s grant of summary judgment for NJM, and determined that Pennsylvania law applied. It concluded that “the choice of Pennsylvania law is dictated by the well-settled principles underlying [New Jersey’s] conflicts of law jurisprudence in respect of insurance contracts.” Id. at 1346. It explained the New Jersey Supreme Court’s approach in Simmons and Gilbert, and went on to apply the Restatement to the facts before it, using § 193 as its starting point per Gilbert, 629 A.2d at 893. Our analysis, set forth below, tracks closely to that of MacVicar in its application of New Jersey’s settled choice-of-law principles in the context of insurance contract disputes.

Principal Location of the Insured Risk

MacVicar began its analysis under the Restatement by examining the parties’ justified expectations regarding the principal location of the insured risk. “Applying the test of § 193 of the Restatement,” the Court stated, “we think it plain that the parties understood that[,] as of the date of the MacVicars’ move to Pennsylvania, a date and an event' of which NJM indisputably had notice prior to the accident, Pennsylvania would be the principal location of the risk.” Id. at 1347. The reasons cited for this conclusion are virtually identical to the reasons that the Fogels now argue Arnica must have understood Pennsylvania to be the principal location of the insured risk once the Fogels made the insurer aware of their relocation: “The insureds all resided [in Pennsylvania]. The covered vehicles were all garaged there. The insureds would be doing all their local driving there. [The insurer], moreover, ... clearly acknowledged its understanding that the locus of the risk had been transferred to Pennsylvania....” Id.

For the same reasons expressed in MacVicar, we believe the justified expectations of the parties' shifted when Arnica was put on notice that the Fogel family had permanently relocated to Pennsylvania. Because the Fogels had notified Arnica of their move in August 2008, well before the accident, garaged and were primarily driving their cars in Pennsylvania, and were being billed by Arnica in Pennsylvania, we find it difficult to believe that any party could have believed the primary location of the insured risk was still New Jersey by the time the accident occurred. 3

While Arnica argues correctly that it advised the Fogels of steps they needed to take before it would rewrite their policy under Pennsylvania law, the key focus of our inquiry under Restatement § 193 is not whether the Fogels undertook these steps before the accident, but whether the parties understood that the principal location of the insured risk was Pennsylvania. In other words, the issue is not whether the policy was actually rewritten as a Pennsylvania policy before the accident— all parties acknowledge that it was not (and if it had been, there would be no dispute) — rather, the issue that the Restatement and New Jersey law instruct courts to consider is what the parties believed regarding where the insured risk would be located. We conclude that there was simply no justified expectation after August 2008 that the risk insured under the Fogels’ policy would be located in New Jersey.

MacVicar elaborated on its analysis under § 193 of the Restatement by noting that,

as a general proposition respecting the parties’ understanding of the location of the risk, we think it clear that every automobile insurer is on notice that because of the mobility both of automobiles and their owners, the circumstances of a particular loss may well result in the applicability of the law of a state other than that in which the policy was issued, and that is hence a predicate of the risks they underwrite.

704 A.2d at 1347.

MacVicar cited Parker v. State Farm Insurance Co., 543 F.Supp. 806 (E.D.Pa.1982), as “recognizing” this proposition. Id. The plaintiff in Parker was a resident of Chester, Pennsylvania who was involved in a motor vehicle accident with an uninsured motorist. She obtained UIM benefits under her own policy, and then also claimed UIM benefits under her husband’s insurance policies with State Farm. Parker, 543 F.Supp. at 807-08. Her husband had been a Maryland resident when the State Farm policies were issued and delivered to him, but he and the plaintiff subsequently established residency in Pennsylvania before the accident occurred. Id. at 808. Similar to the situation here, the husband’s insurance provider denied the plaintiffs attempt to recover under his plans because “stacking” was prohibited in Maryland.

Parker acknowledged that the State Farm policies, “as originally contemplated, have most of their relevant contacts with the State of Maryland.” Id. at 810. As was the case for the State of New Jersey in MacVicar and the case before us, in Parker “Maryland is the place where the parties resided or did business at the time of contracting, the place where the contract was negotiated and formed, and— perhaps most importantly — the state where the insured automobiles were expected to be located at the time of the policies’ creation.” Id.

However, Parker recognized that “a raw count of contacts cannot decide the question. The importance of the various contacts depends on the nature of the particular contract and issues involved.” Id. (citing Restatement § 188 cmt. b). Thus, based on weighing each state’s contacts with the policy and their underlying interests, Parker concluded that Pennsylvania law should apply. At the time of the accident “Pennsylvania had become the state of James Parker’s domicile and the state where the insured vehicles were principally garaged.” Id. at 810-11. The Court also noted, as MacVicar did, that in a § 193 analysis

the “location of the insured risk” is of less importance in the context of [an] automobile liability insurance policy than it would be in the context of, for example, a policy of fire insurance for a building. The mobility of the insured risk makes it possible that an accident covered by the policies will occur in a state other than the state where the automobile is principally garaged. Furthermore, in today’s mobile society, it is also foreseeable that a person covered by insurance- may himself marry and move from state to state, as plaintiff James Parker did here, and, thus, wholly change his domicile and the place where the automobile is garaged. Thus, the possibility that another state’s law may be applied to determine questions arising out of automobile liability insurance policies made in Maryland cannot be totally unforeseen by an insurer in State Farm’s position.

Id. at 810-11. As Parker noted, comment d to § 193 explains that on “occasions when following the issuance of the policy the principal location of the risk is shifted to some other state,” the “other state will have a natural interest in the insurance of the risk and it may be that its local law should be applied to determine at least some issues arising under the policy.” Restatement § 193 cmt. d. In the case of automobile insurance contracts specifically, the mobility of the insured risk creates a foreseeable potential for the principal location of the risk to shift during the term of a policy. In those situations, “application of the local law of the [state to which the insured risk has shifted] would hardly be unfair to the insurance company,” as the mobility of the insured risk itself provides reason to foresee “that there might be a shift to another state of the principal location of the risk.” Id.

The District Court here found Parker unpersuasive, essentially because it believed that Parker used a broad concept of foreseeability to read the “principal location of the insured risk” out of the § 193 analysis. It found Parker’s diminished reliance on the principal location of the insured risk problematic because the Restatement instructs that “[t]he location of the insured risk will be given greater weight than any other single contact in determining the state of the applicable law....” Restatement § 193 cmt. b; see also R & R at 25-27. However, the District Court’s analysis fails to take into account the reason why the principal location of the insured risk is such a weighty factor. As the Restatement comments state, courts focus on that factor because it ordinarily comports with the parties’ expectations regarding what law will apply to the insurance contract. See Restatement § 193 cmt. c (A primary “rationale” is that “it can often be assumed that the parties ... would expect that the local law of the state where the risk is to be principally located would be applied to determine many of the issues arising under the contract.”). The comments further explain that in the case of movable items, such as motor vehicles, the expectation that the location of the insured risk will remain constant is diminished, even if it is not entirely eliminated. See id. cmt. d; id. cmt. b (explaining that this factor “enjoys greatest significance when an immovable” object, such as a building, is the subject of the insurance policy). In sum, we do not view Parker as a dramatic departure from the Restatement approach, but simply as an application of the rationale expressed in the Restatement comments.

Moreover, even if we discounted the foreseeability of motor vehicle owners relocating to other states during the term of a policy, as the District Court did in rejecting Parker’s reasoning, we would still reach the result we do today. On both the facts of MacVicar and the case now before us, NJM and Arnica each had actual knowledge that the principal location of the insured risk had already shifted from New Jersey to Pennsylvania based on direct communication of that fact by the insureds in each case.

Thus, as in MacVicar, “[w]e are ... satisfied that no state had a more significant relationship with this risk than did Pennsylvania. New Jersey’s relationship as the place where the policy was initially issued plainly became tangential after the severance of all other ties and connections [in New Jersey].” 704 A.2d at 1347.

Governmental Interest Analysis

Although we have concluded that Pennsylvania has the most significant relationship with the Fogels’ policy, we need to examine the “state policies affected by, and governmental interest in, the outcome of the controversy.” Simmons, 84 N.J. at 37, 417 A.2d 488. Relying primarily on Parker and an earlier case in our Court, Travelers Insurance Co. v. Davis, 490 F.2d 536 (3d Cir.1974), MacVicar concluded that it was “clear that application of a governmental interest analysis points inexorably to Pennsylvania as well.” 704 A.2d at 1347. We agree, and conclude that Pennsylvania’s underlying policy interests predominate.

MacVicar explained that “Pennsylvania’s UIM stacking law is obviously predicated on its firm public policy of affording its residents the full advantage of all the insurance they have purchased.” Id. Indeed, its “commitment to that policy is ... so strong that it prescribes the exact verbiage that an insured must assent to in writing before he will be deemed to have waived the benefit of that law.” Id. (finding “no interest of New Jersey” that “could possibly override that of Pennsylvania” in this regard). Parker also attached great weight to Pennsylvania’s strong underlying interest in its UIM stacking policy. Parker, 543 F.Supp. at 811 (emphasizing importance of “Pennsylvania’s strong interest in protecting its citizens’ rights to the full benefit of insurance provisions agreed to and coverage paid for”) (internal citation omitted).

In Travelers we held as well that Pennsylvania law applied in a choice-of-law dispute based on the Commonwealth’s strong interest in its UIM stacking policy. Travelers involved a Pennsylvania citizen who was killed in Texas in an automobile accident with an uninsured motorist that occurred in the course of his employment. His employer’s automobile liability insurance policy had been issued in Massachusetts. As here, the insurance company sought a declaratory judgment that stacking was impermissible based on an argument that Pennsylvania law did not apply. The District Court dismissed the insurer’s complaint. In affirming the dismissal, we noted that while Massachusetts had some interest “because the contract was made there,” “Pennsylvania’s interest is the greatest because decedent and his executors are citizens of that state. Pennsylvania is vitally concerned with the administration of decedent’s estate and the well-being of the surviving dependents.” Travelers, 490 F.2d at 543 (internal citation omitted). Thus, we applied Pennsylvania law to determine the parties’ rights and obligations under the policy.

% Hí ❖ # & ‡

For these reasons, we conclude that although New Jersey has some interest in the Fogels’ policy — as the place of contracting and the initial location of both parties — after the Fogels notified Arnica of their relocation to Pennsylvania, there was no longer a justified expectation that New Jersey remained the principal location of the insured risk. All parties understood that the family’s cars, the insured risk, were to be garaged and primarily driven in Pennsylvania following their move. Moreover, following MacVicar, Parker, and Travelers, we conclude that Pennsylvania’s strong underlying policy interest outweighs that of New Jersey in this case. Hence we reverse the District Court’s grant of summary judgment for Arnica and remand so that summary judgment may be entered for the Fogels on the choice-of-law issue.

III. The Fogels’ Bad Faith Counterclaim

The Fogels allege in their counterclaim that Arnica acted in bad faith, under 42 Pa.C.S.A. § 8371, by failing to investigate adequately their claim and improperly denying them stacking benefits under their policy. Arnica moved for summary judgment on this counterclaim, arguing that it had a reasonable basis for all actions taken based on its belief that New Jersey law applies. It also notes that it had paid nearly $375,000 in claims to the Fogels under the policy as of the date it moved for summary judgment.

In response, the Fogels argue that Arnica’s claims adjuster assigned to their case did not complete a thorough investigation. They note that the adjuster was aware that the Fogels had moved to Pennsylvania and that they had been billed for their policy in Pennsylvania prior to the accident at issue, but that he nonetheless failed to “consider the length of time that the Fogels were residing in Pennsylvania prior to the accident,” and also failed to determine “where the Fogels’ vehicles were being garaged” or “where their local driving was taking place.... ” Fogel Br. 29. Finally, they argue that Arnica’s advice to the Fogels regarding Pennsylvania’s vehicle registration procedures was contrary to Pennsylvania law and its own claims manual. The District Court thoroughly examined the applicable law and rejected the Fogels’ counterclaim alleging bad faith.

In the primary case construing bad faith under 42 Pa.C.S.A. § 8371, Terletsky v. Prudential Property & Casualty Co., the Superior Court of Pennsylvania explained:

“Bad faith” on [the] part of [an] insurer is any frivolous or unfounded refusal to pay proceeds of a policy; it is not necessary that such refusal be fraudulent. For purposes of an action against an insurer for failure to pay a claim, such conduct imports a dishonest purpose and means a breach of a known duty (i.e., good faith and fair dealing), through some motive of self-interest or ill will; mere negligence or bad judgment is not bad faith.

437 Pa.Super. 108, 125, 649 A.2d 680, 688 (Pa.Super.Ct.1994) (quoting Black’s Law Dictionary 139 (6th ed.1990)). Terletsky held that, “to recover under a claim of bad faith,” the insured must show that the insurer “did not have a reasonable basis for denying benefits under the policy and that the insurer knew of or recklessly disregarded its lack of reasonable basis in denying the claim.” Id. Thus, an insurer may defeat a claim of bad faith by showing that it had a reasonable basis for its actions. Horowitz v. Federal Kemper Life Assurance Co., 57 F.3d 300, 307 (3d Cir.1995). Our Court has described “the essence of a bad faith claim” as “the unreasonable and intentional (or reckless) denial of benefits.” UPMC Health Sys. v. Metro. Life. Ins. Co., 391 F.3d 497, 506 (3d Cir.2004).

Bad faith “must be proven by clear and convincing evidence and not merely insinuated.” Terletsky, 649 A.2d at 688 (collecting cases). As the District Court noted, this heightened standard requires the insured to provide evidence “so clear, direct, weighty and convincing as to enable a clear conviction, without hesitation, about whether or not the defendants acted in bad faith.” Bostick v. ITT Hartford Grp., 56 F.Supp.2d 580, 587 (E.D.Pa.1999) (citations omitted). In deciding a motion for summary judgment, a court “must view the evidence presented in light of the [insured’s] substantive burden at trial.” Northwestern Mut. Life Ins. Co. v. Babayan, 430 F.3d 121, 137 (3d Cir.2005) (citations omitted).

Because the District Court believed that Arnica was correct in applying New Jersey law, it concluded that “it would be anomalous to find that the Fogels could nevertheless proceed on a claim of bad faith that is predicated in large part [on] the insureds’ contention that Arnica’s claims adjuster engaged in bad faith in handling the claims ... by construing the Policy under New Jersey law.” R & R at 36. Although we part with the District Court on its conclusion that New Jersey law applied, we believe that if reasonable judges could reach that conclusion, as the Magistrate Judge and District Judge did in this case, the conduct alleged simply does not amount to bad faith. See J.H. France Refractories Co. v. Allstate Ins. Co., 534 Pa. 29, 626 A.2d 502, 510 (1993) (noting that “it would be harsh indeed to attribute bad faith to parties which relied on the reasoning and approaches that other courts have found convincing”); cf. Babayan, 430 F.3d at 137 n. 22 (“an insurer’s denial of a claim does not constitute bad faith if it is based on a reasonable legal position in an unsettled area of the law”) (citing Terletsky, 649 A.2d at 690). Thus, we affirm the District Court’s grant of summary judgment for Arnica on the Fogels’ counterclaim alleging bad faith.

IV. Conclusion

By the language and logic of Ferens, the Van Dusen rule applies to sua sponte transfers pursuant to § 1404(a). In this case, the Middle District of Pennsylvania, the transferee forum, should have applied the choice-of-law rules that the District of New Jersey, the transferor, would have followed. New Jersey choice-of-law rules, laid out in Simmons and Gilbert and applied in the analogous case of MacVicar, leave no doubt that Pennsylvania law applies to this dispute. Thus, we reverse the District Court’s grant of summary judgment for Arnica on the choice-of-law issue and remand for summary judgment in favor of the Fogels on that issue. 4 We affirm, however, the District Court’s grant of summary judgment for Arnica on the Fogels’ counterclaim of bad faith under 42 Pa.C.S.A. § 8371.

1 . The District Court dismissed all of the Fogels’ objections to the R & R. The only objection it addressed in its opinion was the Fogels’ argument that it was not possible to meet Arnica’s requirement that an insured who moves to a new state obtain a driver's license and registration in the new state before Arnica will issue a new insurance policy in that state. This requirement was impossible to meet, the Fogels argued, because Pennsylvania requires proof of insurance prior to registering a vehicle in the Commonwealth. The District Court concluded that this sequencing problem did not make it impossible for the Fogels to follow Arnica’s procedures because the Pennsylvania vehicle registration office and insurers may " 'swap’ information simultaneously to meet both Arnica’s and the Commonwealth’s procedure.” Dist. Ct. Memo at 6-7.

2 . The MacVicar family moved to Pennsylvania approximately six months into the term of their policy with NJM, while the Fogels moved to Pennsylvania approximately eight months into the term of their policy with Arnica. The length of time remaining on the policy was not a consideration in MacVicar, nor do we consider it a critical fact here.

3 . The only discernible difference at this stage of the analysis between MacVicar and this case is that more time had passed between the Fogels’ communication with Arnica and the MacVicars’ communication with NJM when the accident in question occurred. It is inconceivable to us that the insurer’s “understanding that the locus of the risk had been transferred to Pennsylvania,” id., would be diminished by the passage of time, nor do we find highly probative the Fogels’ inaction in registering their vehicles in Pennsylvania during that time. Indeed, because Arnica had begun billing the Fogels on their policy in Pennsylvania before the accident (and the Fogels returned payment accordingly), if anyr thing the passage of time under these circumstances would solidify the understanding that Pennsylvania had become the location of the insured risk for the foreseeable future.

4 . As noted above, the R & R points out (and indeed the parties do not dispute) that, under Pennsylvania law, the Fogels are entitled to stacked UIM benefits under the policy. R & R at 18. Moreover, no dispute exists that, under Pennsylvania law, offsetting those benefits against amounts received from other insurance sources is not permitted. Id. ("[T]he parties agree, and the Court recognizes, that Pennsylvania law does not permit insurers to offset UIM benefits against amounts received from other insurance sources, whereas New Jersey does.”). What remains is the determination of the amount of relief afforded.