[PUBLISH]
IN THE UNITED STATES COURT OF APPEALS
FOR THE ELEVENTH CIRCUIT
________________________
No. 17-11703
________________________
D.C. Docket No. 1:15-cv-04130-RWS
PRINCIPLE SOLUTIONS GROUP, LLC, Plaintiff – Appellee,
versus
IRONSHORE INDEMNITY, INC., Defendant – Appellant.
________________________
Appeal from the United States District Court for the Northern District of Georgia
________________________
(December 9, 2019)
Before WILLIAM PRYOR, TJOFLAT, and GILMAN, * Circuit Judges. WILLIAM PRYOR, Circuit Judge:
*
Honorable Ronald Lee Gilman, United States Court of Appeals for the Sixth Circuit, sitting by designation.
This appeal requires us to decide whether a loss of more than $1.7 million to scammers was covered under a commercial crime insurance policy. The loss stemmed from a sophisticated phishing scheme in which a scammer posing as an executive of Principle Solutions Group, LLC, persuaded an employee to wire money to a foreign bank account. After Principle discovered the fraud and determined that it could not recover the funds, it sought coverage under the “fraudulent instruction” provision of its policy with Ironshore Indemnity, Inc., which then denied Principle’s claim. Ironshore asserted that the scammer’s communications with the employee did not meet the conditions for a fraudulent instruction under the policy and that the loss did not result directly from the alleged fraudulent instruction, as the policy required. Principle filed a complaint against Ironshore to enforce the policy. The district court concluded that the policy covered the loss and granted summary judgment to Principle. Because we also conclude that the policy unambiguously covers Principle’s claim, we affirm.
I. BACKGROUND
On the morning of July 8, 2015, Principle lost over $1.7 million in a fraud
scheme. It began at 9:10 a.m., when Loann Lien, the controller for Principle, received an email purporting to be from Josh Nazarian, a managing director of Principle. The email informed Lien that Principle had been secretly working on a “key acquisition” and asked her to wire money “in line with the terms agreed . . .
3
as soon as possible.” As for the details of the wire transfer, the email told Lien to give her “full attention” to “attorney Mark Leach,” who would provide further information. Because the purported deal was not public, Lien was to “treat [the] matter with the upmost discretion and deal solely with” Leach. Lien responded to Nazarian’s purported email that she would give her “total attention” to Leach.
Lien received an email five minutes later from someone purporting to be Leach, a partner at the London-based law firm Bird & Bird. After Lien confirmed that Principle could wire the money, Leach sent Lien remittance details for a bank in China. Leach later reiterated to Lien over the phone that Nazarian approved the wire transfer.
Lien worked with another Principle employee to create and approve the transfer, but a fraud prevention service, Wells Fargo, asked for verification that the wire transfer was legitimate. Lien then confirmed with Leach that Nazarian had approved the transaction. Lien relayed this information to Wells Fargo, which released the hold. At 11:21 a.m., about two hours after Lien received the first email, Principle wired more than $1.7 million to the scammers.
Lien discovered that the request was fraudulent a day later when she spoke with Nazarian, who told her that he was not even in the office that day. Nazarian promptly called Wells Fargo to report the fraud, but neither Principle nor law enforcement could recover the funds.
4
Principle sought coverage for the loss under its insurance policy with Ironshore. The policy covered “[l]oss resulting directly from a fraudulent instruction directing a financial institution to debit [Principle’s] transfer account and transfer, pay or deliver money or securities from that account.” Ironshore denied coverage. It asserted that Nazarian’s purported email did not “direct[] a financial institution to debit [Principle’s] transfer account” because it only told Lien to await instructions from Leach. Ironshore also argued that the asserted loss did not “result[] directly from” a fraudulent instruction because Leach conveyed necessary details to Lien after the initial email and Wells Fargo held the transaction, both of which were intervening events between the instruction and the loss.
Principle filed a complaint against Ironshore in state court seeking payment under the policy and alleging that Ironshore had acted in bad faith. Ironshore removed the case to federal court based on diversity jurisdiction. 28 U.S.C. §§ 1332(a)(1), 1441(a). The parties filed competing motions for summary judgment. Although the district court concluded that the policy provision was ambiguous, it held that Georgia’s rule requiring construction of insurance policies in favor of policyholders required it to grant partial summary judgment to Principle on its coverage claim. The district court also granted partial summary judgment to Ironshore on Principle’s claim of bad faith. Only Ironshore appealed.
5
II. STANDARD OF REVIEW
We review a summary judgment de novo. Regions Bank v. Legal Outsource PA, 936 F.3d 1184, 1189 (11th Cir. 2019). Summary judgment is warranted “if the movant shows that there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.” Fed. R. Civ. P. 56(a).
III. DISCUSSION
Under Georgia law, which the parties agree governs, we use a three-step approach to interpret insurance policies. As with any contract, we first look to the text of the policy and, if it is “explicit and unambiguous, [our] job is simply to apply the terms of the contract as written, regardless of whether doing so benefits the carrier or the insured.” Ga. Farm Bureau Mut. Ins. Co. v. Smith, 784 S.E.2d 422, 424 (Ga. 2016) (citation and internal quotation marks omitted). But “if a provision of an insurance contract is susceptible of two or more constructions, even when the multiple constructions are all logical and reasonable, it is ambiguous,” and we must move to the second step of applying Georgia’s “statutory rules of contract construction.” Hurst v. Grange Mut. Cas. Co., 470 S.E.2d 659, 663 (Ga. 1996). And if the ambiguity “cannot be resolved through the rules of construction,” we may, at the third step, look to parol evidence. Coppedge v. Coppedge, 783 S.E.2d 94, 97 n.3 (Ga. 2016) (citation and internal quotation marks omitted). But “if the parol evidence is in conflict, the question of what the parties intended
6
becomes a factual issue for the jury.” Id. (citation and internal quotation marks omitted); see also Ga. Code Ann. § 13-2-1 (“The construction of a contract is a question of law for the court. Where any matter of fact is involved, the jury should find the fact.”).
Ironshore repeats its twin justifications for denying coverage: no communication between the scammers and Lien triggered the fraudulent- instruction provision, and the loss did not “result[] directly from” any alleged fraudulent instruction. Both fail. We address each argument in turn.
A. The Loss Involved a Fraudulent Instruction Directing a Financial Institution to Transfer Funds.
To receive coverage, Principle had to identify a “fraudulent instruction” that “direct[ed] a financial institution to debit [Principle’s] transfer account and transfer, pay or deliver money or securities from that account.” As relevant here, the policy defines a “fraudulent instruction” as an “electronic or written instruction initially received by [Principle], which instruction purports to have been issued by an employee, but which in fact was fraudulently issued by someone else without [Principle’s] or the employee’s knowledge or consent.” Ironshore contends that no communication satisfied both conditions of the coverage provision, but we disagree.
As Ironshore concedes, the email purporting to be from Nazarian, which informed Lien of the need to wire money and told her to await further instructions
7
from Leach, qualifies as a fraudulent instruction. It was, after all, a “fraudulently issued” “electronic . . . instruction” that “purport[ed] to have been issued by an employee . . . without [Principle’s] or the employee’s knowledge or consent.” But Ironshore asserts that the email instructed Lien only to work with Leach to wire funds later in the day, not to wire a specific amount of money to a specific recipient. So, it explains, the email did not “direct[]” Principle to pay money out of its accounts, as the coverage provision required.
This argument is unpersuasive. As an initial matter, we are hard pressed to construe the email as doing anything but “directing a financial institution to debit [Principle’s] transfer account and transfer . . . money . . . from that account.” The email told Lien, “I will need you to make the initial wire as soon as possible, for which you have my full approval to execute.” But even if we assume that the email needed additional details before we could fairly construe it as “directing” a wire transfer, a later email from Leach identified the amount of the wire transfer, the recipient bank, and the purported beneficiary of the transfer. That email remedied any possible lack of detail.
Ironshore agrees that Leach’s email was sufficiently detailed, but it contends that the email was not a “fraudulent instruction” under the policy because someone purporting to be an outside attorney, not a Principle employee, sent it. In other words, Ironshore contends that Leach’s email, though fraudulent and specific, was
8
not a “fraudulent instruction,” and that Nazarian’s purported email, though a “fraudulent instruction,” was not specific enough to meet the provision’s other requirements.
We disagree with Ironshore’s divide-and-conquer approach. Nothing in the policy language warrants the assumption that the two emails could not be part of the same fraudulent instruction. Although the policy defines a fraudulent instruction as a singular “electronic or written instruction,” Georgia adopts the longstanding rule of construction that the “singular or plural number each includes the other, unless the other is expressly excluded.” Ga. Code Ann. § 1-3-1(d)(6); see also Antonin Scalia & Bryan A. Garner, Reading Law: The Interpretation of Legal Texts § 14, at 130 (2012). And reading the emails together leaves no doubt that they were part of the same fraudulent instruction. Leach’s email supplemented the email purporting to be from Nazarian, which cloaked Leach with the authority to give additional details. That email told Lien to “deal solely” with Leach and to “give [Leach her] full attention” to make sure “the wire goes out today.” And Leach’s email informed Lien that the additional remittance details were “requested by” Nazarian. Viewing the emails together, the sole purpose of Leach’s email was to provide details to effectuate an explicit instruction to make a wire transfer. So the fraudulent instruction from the scammer purporting to be Nazarian unambiguously falls within the coverage provision.
9
B. The Loss Resulted Directly from a Fraudulent Instruction.
Ironshore next contends that Principle’s losses did not “result[] directly from” a fraudulent instruction as the policy required. Citing dictionary definitions, it interprets the policy’s use of “directly” to require an “immediate” link between a fraudulent instruction and a loss. Because the loss depended on Lien’s conversations with Leach and Wells Fargo, which occurred after Nazarian’s purported email told Lien to wire money, Ironshore concludes that no immediate link existed between the instruction and the loss.
We again disagree with Ironshore. Georgia gives terms in insurance contracts their ordinary meaning. See Ga. Code Ann. § 13-2-2(2). And the ordinary meaning of the phrase “resulting directly from” requires proximate causation between a covered event and a loss, not an “immediate” link. To be sure, the definitions that Ironshore cites reveal that “directly” can sometimes mean “immediately.” See, e.g., Directly, Black’s Law Dictionary (10th ed. 2014) (“immediately”). But discerning the ordinary meaning of a term requires more than uncritically citing dictionaries. By isolating “directly” from the surrounding language, Ironshore erroneously defined the term “without accounting for its semantic nuances.” Scalia & Garner, Reading Law app. A, at 418.
Situating “directly” within the phrase “resulting directly from” reveals a different meaning. When used with the preposition “from,” “resulting” means “to
10
proceed, spring, or arise as a consequence, effect or conclusion.” Result/-ing, Webster’s Third New International Dictionary (1993); see also Resulting, The American Heritage Dictionary (5th ed. 2012) (“[t]o happen as a consequence”). “‘Results from’ imposes, in other words, a requirement of actual causality.” Burrage v. United States, 571 U.S. 204, 211 (2014); see also Blockum v. Fieldale Farms Corp., 573 S.E.2d 36, 39 (Ga. 2002) (using “causes” and “resulting from” interchangeably). So reading the phrase “resulting directly from” as a whole requires us to define “directly” within the context of causation. And in that context, “directly” means “proximately.” See Directly, Webster’s Third New International Dictionary (“in close relational proximity”); Direct Cause, Black’s Law Dictionary (11th ed. 2019) (“proximate cause”); cf. Proximate Cause, Webster’s Third New International Dictionary (“a cause that directly or with no mediate agency produces an effect”). In short, the ordinary meaning of “resulting directly from” requires us to determine whether the fraudulent instruction here proximately caused Principle’s loss.
In Georgia, a “[p]roximate cause is not necessarily the last act or cause, or the nearest act to the injury.” Sprayberry Crossing P’ship v. Phenix Supply Co., 617 S.E.2d 622, 624 (Ga. Ct. App. 2005) (citation and internal quotation marks omitted). Instead, it encompasses “all of the natural and probable consequences” of an action, “unless there is a sufficient and independent intervening cause.” Cowart
11
v. Widener, 697 S.E.2d 779, 784 (Ga. 2010) (emphasis added). An intervening cause is not sufficient and independent if “its probable or natural consequences could reasonably have been anticipated, apprehended, or foreseen by the original wrong-doer.” Goldstein Garber & Salama, LLC v. J.B., 797 S.E.2d 87, 89 (Ga. 2017) (citation omitted).
Neither of the two “causes” that Ironshore asserts intervened between Nazarian’s purported email and Principle’s loss—Lien’s communications with Leach and Wells Fargo’s involvement—can sever the causal chain. Both were foreseeable consequences of the email. Nazarian’s purported email told Lien that Leach would contact her and provide further details on the wire request. And although Wells Fargo’s involvement was not inevitable, it was certainly foreseeable. The email proactively sought to avoid third-party interference by requiring Lien to “deal solely” with Leach. Because of this instruction, the scammers could circumvent Wells Fargo’s fraud-prevention process; through a series of phone calls and emails between Leach and Lien, they fabricated the precise information that Wells Fargo required to release the hold. So not only did the scammers foresee that a fraud prevention service might get involved, they put a system in place to circumvent that risk. Georgia courts have held that intervening acts were foreseeable in more attenuated circumstances. See, e.g., Imperial Foods Supply, Inc. v. Purvis, 580 S.E.2d 342, 345 (Ga. Ct. App. 2003) (holding that it
12
was foreseeable that the driver of a borrowed car would try to fix a broken door latch with a “makeshift, rigged mechanism” that did not perform reliably and caused injury).
For the same reason, we are not persuaded by the dissent’s enumeration of eleven events that occurred between the fraudulent instruction and loss, such as each call that Lien made to Leach and Lien’s approval of the wire transfer. Each of these events was either a necessary action to transfer the money or falls under one of the two ostensible intervening causes that Ironshore proposes—Lien’s communication with the scammers and Wells Fargo’s involvement. Although the dissent parses the events more finely than Ironshore, each of its eleven events was still a foreseeable consequence of the fraudulent instruction.
We also cannot agree with the dissent that Rustin Stamp & Coin Shop, Inc. v.
Ray Brothers Roofing & Sheet Metal Co., 332 S.E.2d 341 (Ga. Ct. App. 1985), controls this appeal. Rustin held that no proximate causation existed between a roofing company’s negligent decision to leave unmarked holes on a roof covered by fiberglass and rain damage to the interior of the building. Id. at 342–44. The court found that the causal chain was severed by the actions of an air-conditioning repairman, who exposed the building’s interior to rain by perforating the fiberglass covering and failed to notify the roofing company about the problem. Id. It concluded that the defendant could not foresee that someone would both expose
13
the interior of the building to outside rain and fail to notify anyone else about it. Id. at 344. But a fraudulent instruction necessarily contemplates that an unwitting employee will negligently transfer money. And unlike the intervening act in Rustin, Wells Fargo did not cause Principle’s loss, which would have occurred whether or not it became involved. Needless to say, something that is not a cause of damage cannot be an intervening cause. See Ga. Bank & Tr. v. Cin. Ins. Co., 538 S.E.2d 764, 765–66 (Ga. Ct. App. 2000) (holding that a financial loss did not “result[] directly from” a forged signature when the loss would have occurred “[e]ven if the signature . . . was authentic”).
Nor can we agree with the dissent’s argument that the ostensibly “suspicious nature of the entire transaction” severed the causal chain because it gave Principle “notice that the wire transfer may be fraudulent” and the “opportunity to prevent the loss.” Dissenting Op. at 27. To start, the record does not support the assumption that the entire transaction was “suspicious.” The scammers’ interactions with Lien revealed that they had a sophisticated knowledge of Principle’s operations and personnel. Without the benefit of hindsight, we cannot say that the transaction was inherently suspicious. And in any event, whether various red flags “arguably should have triggered a deeper investigation,” id. at 28, is not the relevant question. Instead, the relevant question is whether Lien’s failure to verify the transfer in the ways the dissent suggests was foreseeable. And that
14
failure was foreseeable: the scammers set up a system designed to prevent Lien from verifying the request, which means that they foresaw Lien’s failure.
Finally, we disagree that proximate causation is a question for the jury in this appeal. To be sure, “the issue of proximate cause is generally a question of fact for the jury.” Bennett v. Dep’t of Transp., 734 S.E.2d 77, 78 (Ga. Ct. App. 2012). But “it may be decided as a matter of law where the evidence is clear and leads to only one reasonable conclusion.” Id. And the evidence in this appeal leads to only one reasonable conclusion: No unforeseeable cause intervened between Nazarian’s purported email and Principle’s loss. The loss unambiguously “resulted directly from” the fraudulent instruction.
IV. CONCLUSION
We AFFIRM the summary judgment to Principle.
15
TJOFLAT, Circuit Judge, dissenting:
Principle suffered a loss of $1.717 million after it was duped into transferring funds to an international bank account. Principle now seeks recovery under its “Commercial Crime Policy” insurance agreement with Ironshore, which provides coverage for a “[l]oss resulting directly from a ‘fraudulent instruction’ directing a ‘financial institution’” to transfer or pay funds.
As the majority lays out, Principle’s controller transferred $1.717 million to an international bank account after receiving a spoofed email, purportedly sent by Principle’s Managing Director Josh Nazarian (hereinafter the “Nazarian email”). After receiving the Nazarian email, the controller received instructions from an “outside attorney” about how much to send, as well as where and to whom to send the money. The controller, Loann Lien, also helped initiate a wire transfer; approved the wire transfer; and verified the transaction after the fraud prevention department of Principle’s bank placed a hold on the transaction.
The majority claims that the insurance agreement unambiguously covers the loss because it resulted directly from the Nazarian email. I disagree with the majority’s conclusion that the Nazarian email unambiguously directed a “financial institution” to transfer funds. Furthermore, I conclude that the bank’s intervention presents a jury question for whether Principle’s loss resulted directly from this “fraudulent instruction.”
16
I.
A.
The majority lays out the relevant, and uncontested, facts for the events leading up to Lien’s initiation and approval of the wire transfer. However, Wells Fargo did more than “ask[] for verification that the wire transfer was legitimate.” Wells Fargo’s intervention arguably short-circuited the causal connection between the Nazarian email and Principle’s loss. Here is what Wells Fargo did:
Less than fifteen minutes after Lien approved the wire, Wells Fargo emailed her that the transaction was being temporarily held pending verification. The email clearly flagged the potential for fraud—it was sent from “CEOFraudPrevention@wellsfargo.com” and stated that the sender, Bryan Chu, identified as a Fraud Prevention Consultant, would call Lien. Chu assigned the transaction a case number as well as provided a phone number for the Fraud Prevention Department.
Beneath Chu’s signature line, the email included a note, which stated that “Imposter Fraud is on the rise” and identified three best practices to avoid this type of fraud, including, in underlined font, “[v]erify your requestor.” In bold face type, the note directed Lien to the Fraud Protection section of Wells Fargo’s website to learn more about Imposter Fraud. In addition, Chu attached a pdf entitled “Imposter Fraud Information.” The attachment cautioned that one version of
17
imposter fraud was when “[a] fraudster posing as an executive of your company . . . instructs you to make one or more payments outside of normal channels – usually by wire for a high dollar amount.” The attachment again provided a list of tips to “help reduce your risk,” including “[v]erify your requestor”—which in this case was Josh Nazarian—and recommended that clients “[s]et a policy requiring all requests for unusual payments made outside normal channels to be verified with a phone call to the requestor.”
Upon receipt of the email, Lien called the Wells Fargo Fraud Prevention Department. She spoke to an employee, identified only as Daniel, regarding the purpose and details of the wire. According to Lien, Daniel asked her “to verify with Mark Leach how he had received the wire instructions.”
While Lien was on the phone with Daniel, she received a voicemail and another email from Chu. The second email largely mirrored the first and stated that Chu had left Lien a voicemail regarding the transaction. The record does not reveal the contents of the voicemail.
Immediately after getting off the phone with Daniel and without listening to the voicemail from Chu, Lien emailed Leach asking him to call her so that she could verify how he had received the wire instructions. Leach called Lien and stated that he had received the wire instructions telephonically from Josh.
18
Lien then called Daniel at the Wells Fargo Fraud Prevention Department and relayed the details of her conversation with Leach. Wells Fargo released the wire.
B.
As relevant here, Principle’s policy covers “Computer and Funds Transfer Fraud,” which includes “[l]oss resulting directly from a ‘fraudulent instruction’ directing a ‘financial institution’” to transfer or pay funds. A “fraudulent instruction” is defined to include:
A computer, telegraphic, cable, teletype, telefacsimile, telephone, or other electronic or written instruction initially received by you, which instruction purports to have been issued by an “employee”, but which in fact was fraudulently issued by someone else without your or the “employee’s” knowledge or consent. 1
1 The policy also defines “fraudulent instruction” as:
(1) A computer, telegraphic, cable, teletype, telefacsimile, telephone or other electronic instruction directing a “financial institution” to debit your “transfer account” and to transfer, pay or deliver “money” or “securities”
from that “transfer account”, which instruction purports to have been issued by you, but which in fact was fraudulently issued by someone else without your knowledge or consent.
(2) A written instruction (other than those covered by Insuring Agreement A.2.) issued to a “financial institution” directing the “financial institution”
to debit your “transfer account” and to transfer, pay or deliver “money” or “securities” from that “transfer account”, through an electronic funds transfer system at specified times or under specified conditions, which instruction purports to have been issued by you, but which in fact was issued, forged or altered by someone else without your knowledge or consent.
Insuring Agreement A.2 pertains to forgery or alteration of “checks, drafts, promissory notes, or similar written promises, orders or directions to pay a sum certain.”
19
C.
The District Court found that the computer-and-funds-transfer-fraud provision was ambiguous. Principle argued that the provision provides coverage despite the existence of intervening events between the fraud and the loss. Ironshore argued that the term “directly” requires an immediate link between the fraud and the loss. The Court determined that both interpretations were reasonable. Under Georgia law, because the language was ambiguous, the provision must be construed in the light most favorable to the insured and therefore, the District Court held that “the loss resulted directly from an instruction that Ironshore admit[ed] was fraudulent.”
III.
The majority affirms on a different basis: that the insurance provisions are unambiguous, and we can simply apply the terms as written. I agree that the Nazarian email was unambiguously a “fraudulent instruction” and that it functioned to direct Wells Fargo to transfer funds. But the question of coverage does not end there. To be covered under the policy, Principle must have suffered a “[l]oss resulting directly from a ‘fraudulent instruction’ directing a ‘financial institution’” to transfer money. Whether Principle’s loss resulted directly from the Nazarian email is a question that the District Court impermissibly removed from the jury.
20
A.
The Nazarian email is a “fraudulent instruction.” It unambiguously falls within the explicit terms of the agreement: it was an “electronic . . . instruction initially received by” Lien, which was purportedly from Nazarian, but “which in fact was fraudulently issued by someone else” without Principle’s, Lien’s or Nazarian’s “knowledge or consent.” Furthermore, Ironshore conceded that the Nazarian email may be a “fraudulent instruction” within the meaning of the policy and, under Georgia law, we construe any ambiguity in favor of the insured. Ga. Farm Bureau Mut. Ins. Co. v. Smith, 784 S.E.2d 422, 425 (Ga. 2016).
B.
The contract is ambiguous as to how an instruction “initially received by [Principle]” can direct a financial institution 2 to transfer money. In contrast to the contract’s two other definitions of fraudulent instruction, there is no requirement that the “electronic or written instruction initially received by [Principle]” be issued to or directed toward a financial institution to qualify as a “fraudulent instruction.”3 Under the coverage provision, however, Ironshore will only pay for
2 “Financial institution” is defined as (1) A bank, savings bank, savings and loan association, trust company, credit union or similar depository institution;
(2) An insurance company; or (3) A stock brokerage firm or investment company.
3
See note 1, supra.
21
“[l]oss resulting directly from a ‘fraudulent instruction’ directing a ‘financial institution’” to transfer money. The duplicity of the phrase “direct a financial institution” in combination with the uncertain meaning of that expression renders these provisions ambiguous. Ga. Farm Bureau Mut. Ins. Co. v. Meyers, 548 S.E.2d 67, 69 (Ga. App. 2001) (“Ambiguity in an insurance contract is duplicity, indistinctiveness, uncertainty of meaning of expression, and words or phrases which cause uncertainty of meaning and may be fairly construed in more than one way.”).
“[I]f a provision of an insurance contract is susceptible of two or more constructions, even when the multiple constructions are all logical and reasonable, it is ambiguous,” Hurst v. Grange Mut. Cas. Co., 470 S.E.2d 659, 663 (Ga. 1996), and this provision is susceptible of at least three logical and reasonable interpretations. One interpretation, which Ironshore argues, is that the instruction received by Principle, must be sent as is to a financial institution. Another interpretation is that the instruction must include all necessary details, even if some internal processes, such as input into a system, are required to transmit the information to a financial institution. A third interpretation, which Principle advocates, is that the instruction direct or authorize someone within Principle to issue an instruction to a financial institution, regardless of whether the instruction initially received by Principle includes the necessary details. As the statutory rules
22
of construction require that we construe the ambiguous clause against the insurer, we adopt the interpretation least favorable to the insurer. Id.; OCGA § 13-2-2(5). Therefore, although the contract is ambiguous, the Nazarian email qualifies as a fraudulent instruction that directs a financial institution to transfer money as it, in combination with the information that Lien received from Leach, directed Wells Fargo to transfer money.
B.
I agree with the majority that directly must be defined in the context of causation. And, I agree that the ordinary meaning of “resulting directly from” requires a proximate cause analysis.
But our inquiry does not end there. Determinations of proximate cause can be made as a matter of law “only if reasonable persons could not differ as to both the relevant facts and the evaluative application of legal standards (such as the legal concept of ‘foreseeability’) to the facts.” Atlanta Obstetrics & Gynecology Grp., P.A. v. Coleman, 398 S.E.2d 16, 17 (Ga. 1990). Summary judgment should be granted only in “plain and undisputed cases.” Id. at 18. Given the intervention of the Wells Fargo Fraud Prevention Department, this is not such a case. Under Georgia law, whether Principle’s loss was proximately caused by the Nazarian email and therefore covered by the insurance agreement is a fact question that the District Court impermissibly removed from the jury’s consideration.
23
1.
Despite legal scholars’ and Georgia courts’ best efforts, there is “no satisfactory universal formula” to determine proximate cause. Id. at 17 (citing Prosser and Keeton on Torts, 5th ed., § 42 at pp. 276–279 (1984)). Instead, proximate cause “is always to be determined on the facts of each case upon mixed considerations of logic, common sense, justice, policy and precedent.” Id. Proximate cause requires an evaluative application of the law to the facts. Id. at 18. That is why, under Georgia law, what amounts to proximate cause is “undeniably a jury question.” Id. (quoting McAuley v. Wills, 303 S.E.2d 258, 260– 61 (Ga. 1983)). Proximate cause may be determined by the courts only “in plain and undisputed cases.” Ontario Sewing Mach. Co. v. Smith, 572 S.E.2d 533, 536 (Ga. 2002) (quoting Atlanta Obstetrics, 398 S.E.2d at 18).
At bottom, proximate cause is a “limit on legal liability.” Atlanta Obstetrics, 398 S.E.2d at 17. It represents a “policy decision” that the conduct at issue and the loss “are too remote for the law to countenance recovery.” Id. While proximate cause encompasses “natural and probable consequences,” it does not include those that are “merely possible.” Rustin Stamp & Coin Shop, Inc. v. Ray Bros. Roofing & Sheet Metal Co., 332 S.E.2d 341, 343–44 (Ga. Ct. App. 1985).
A determination that certain conduct is not the proximate cause of a loss is not to say that the conduct is not a “cause in fact” of the loss. Atlanta Gas Light
24
Co. v. Gresham, 394 S.E.2d 345, 347 (Ga. 1990). Something can furnish the condition or occasion of an injury without proximately causing the loss. Id.; see also Ovbey v. Cont’l Ins. Co., 613 F. Supp. 726, 728 (N.D. Ga. 1985) (“[A] situation which merely sets the stage for the later event is not regarded as being the proximate cause merely because it made possible the subsequent loss.” (quoting 10A Couch on Insurance 2d § 42:650, p. 690 (1982)), aff’d sub nom. Ovbey v. Cont’l Ins., 782 F.2d 178 (11th Cir. 1986).
Something also ceases to be a proximate cause when “several additional factors” must happen between the conduct at issue and the loss. See, e.g., Rustin Stamp, 332 S.E.2d at 344. In Rustin, an air conditioning repairman stepped on a concealed hole on the roof of a building and perforated the opening. Id. at 342. Rain later leaked into the opening, destroying some merchandise in Rustin’s store. Id. Rustin sought to recover from the roofer, Ray Bros, for its failure to warn of the hidden openings. Id. at 343. The Court determined that the causal connection between the negligent failure to warn and the resulting loss of merchandise was too remote to allow recovery. Id. The Court held that “several additional factors” had to happen between the negligence and the loss. Id. at 344. Although it was foreseeable that someone might step in the hidden opening, the Court held that in order for damages to arise, that event had to be “followed by inaction on the part of
25
all concerned in either recognizing the problem and correcting it or notifying Ray Brothers or anyone else for that matter of the situation” before it rained. Id.
2.
Whether the Nazarian email proximately caused Principle’s loss or just set the stage for that loss to occur is a jury question. Arguably, Principle did not suffer a loss “resulting directly” from the admittedly fraudulent Nazarian email because Wells Fargo intervened before Principle suffered a loss. The Nazarian email set the stage for Principle’s loss, but it did not directly cause it.
To illustrate, consider all the additional factors that happened between Lien’s receipt of the Nazarian email and Principle’s loss. 4 • Step 1: The fraudsters send the Nazarian email to Lien. Lien responds that she understands and will work with Leach.
• Step 2: Leach emails Lien. Leach asks her whether Principle can send a wire transfer in an international currency to an international bank.
• Step 3: Lien calls Wells Fargo to confirm that the account can send an international wire transfer in a different currency.
• Step 4: Lien confers with Leach, who indicates that per “Josh,” Lien should approve the wire.
Lien confirms that she can approve the transfer.
4 Temporally, Principle suffered the loss a little over two hours after receiving the Nazarian email. The entire scheme was accomplished between 9:10 a.m., when Lien received the Nazarian email, and 11:21 a.m., when the funds were successfully transferred. But just because the loss happened quickly does not mean that the loss was a proximate result of the Nazarian email. The intervention of Wells Fargo’s Fraud Prevention Department rendered the loss too “remote” to allow recovery.
26
• Step 5: Leach emails Lien the wire instructions. Leach also calls to say that he had spoken to “Josh,” and represents that “Josh” had given his full approval.
• Step 6: Andrea initiates the wire. Lien approves the wire.
• Step 7: Wells Fargo holds the wire and contacts Principle for a fraud investigation. Lien receives an email from a Wells Fargo Fraud Prevention Consultant named Bryan Chu with information on how to identify and prevent fraudulent transfers.
• Step 8: Following receipt of the email, Lien calls Wells Fargo Fraud Prevention Department and has a telephone conversation with an employee named Daniel. Lien receives a second email and voicemail from Chu while on the phone with Daniel.
• Step 9: Lien emails Leach and asks him to call. Leach immediately calls Lien and states that he verbally received the wire instructions from “Josh.”
• Step 10: Lien calls the Wells Fargo Fraud Prevention Department to approve the wire transfer.
• Step 11: Wells Fargo releases the hold and the money is wired.
There is no question that a fraudulent instruction, as defined in the policy, occurred in Step 1. Principle’s loss, however, did not occur until Step 11—when the transfer was complete. Several additional factors had to occur: one, Lien had to communicate with Leach to receive the instructions (steps 2-5); two, the wire had to be initiated and approved; and three, Lien had to override Wells Fargo’s fraud prevention hold (steps 7-11).
Did the Nazarian email proximately cause Lien to talk to Leach? It did.
Lien’s initial communications with Leach were a natural consequence of the Nazarian email. The email told Lien to expect a communication from Leach and
27
that he would explain the requirements. It also asked for her to “work with [Leach],” “to treat this matter with utmost discretion,” and give Leach her “full attention.” The communications with Leach were a “plain and undisputed” “natural and probable” consequence of the Nazarian email.
Did the Nazarian email proximately cause Lien to approve a wire transfer?
It did. Lien continued to follow the instructions given by the Nazarian email and reiterated by Leach. The Nazarian email also gave Lien “full approval to execute” the wire. Again, this was a “plain and undisputed” “natural and probable” consequence of the Nazarian email.
But did the Nazarian email proximately cause Principle’s loss? That is a jury question. As the majority concludes, Lein’s decision to call Leach may have been a “natural and probable” consequence of the Nazarian email. On the other hand, once Wells Fargo stepped in, Lien was no longer relying on the Nazarian email to override the hold and release the funds.
As soon as Wells Fargo put the hold on the funds, the link between the Nazarian email and the loss was arguably short-circuited. Principle was given notice that the wire transfer may be fraudulent, and it had an opportunity to prevent the loss. Multiple emails were sent from the Wells Fargo Fraud Prevention Department; the emails included information regarding imposter fraud and cautioned that management should verify the request with the requestor through a
28
different channel than the one by which they received the instruction. This notification arguably should have triggered a deeper investigation. Like in Rustin, in order for damages to arise, foreseeable consequences had to be “followed by inaction on the part of all concerned in either recognizing the problem and correcting it or notifying . . . anyone else . . . of the situation” before losses occurred. For example, Lien could have easily called Josh Nazarian to confirm the wire details. Or she could have talked to other members of management about the purported acquisition. Alternatively, Lien could have called a different number at Bird & Bird to confirm Leach’s representation of Principle or to ascertain more details about the international acquisition.
Given the suspicious nature of the entire transaction, the intervention of the Wells Fargo Fraud Prevention Department was arguably enough to stop Principle’s loss. The interruption in the process would have led a reasonable employee to question why Wells Fargo thought the transaction may be fraudulent and, in turn, to question the propriety of an unknown attorney providing international wire instructions for an American IT staffing company to pay over a million dollars to acquire a Chinese company. Instead, Lien called Leach, an unknown non- employee, to confirm the international wire instructions for an unspecified acquisition. It was her response to Wells Fargo’s fraud investigation—not the
29
Nazarian email—that directly caused the money to be released and Principle to suffer a loss.
The majority conclusory states that “although Wells Fargo’s involvement was not inevitable, it was certainly foreseeable.” But the majority fails to explain why overriding the fraud prevention process of Principle’s bank was a “natural or probable” consequence of the Nazarian email. The majority’s interpretation fails to draw a line for when resulting consequences are no longer natural or probable. At what point does the loss become merely a result of the fraudster’s conduct, instead of resulting directly from the fraudster’s conduct? Regardless of the level of intervention by a third-party, or even internally by Principle, the majority’s interpretation of proximate cause would conclude—as a matter of law—that it was foreseeable that Lien would ignore those interventions and talk only to Leach in the wake of a fraud investigation.
Concluding that the Nazarian email was, as a matter of law, the proximate cause of Principle’s loss provides no “limit on legal liability.” Atlanta Obstetrics, 398 S.E.2d at 17. The majority’s interpretation would always provide coverage, no matter how much notice the insured had that a scheme could be fraudulent, so long as the insured’s actions could, in some way, be traced to an initial fraudulent instruction. Such an interpretation renders the word “directly” meaningless. It allows remote losses to be traced back to one email without a jury’s consideration
30
of whether an intervening act broke the chain of proximate causation. A jury should decide whether Wells Fargo’s intervention broke the causation chain from the Nazarian email.
Principle’s loss did not occur until after Wells Fargo Fraud Prevention Department had contacted Lien, at which time she had a chance to verify the information, and then called the bank to release the hold. I would hold that the intervention of Wells Fargo Fraud Prevention Department presents a jury question regarding whether the Nazarian email proximately caused Principle’s loss. So long as reasonable minds could differ, it is no longer the “plain and undisputed” case that a court can decide as a matter of law. I respectfully dissent.