# United States v. Lawrence Neadle, Jr.

> Court of Appeals for the Third Circuit · January 29, 1996 · 72 F.3d 1104

URL: https://www.frixlaw.com/law-library/cases/9488779

## Case

- **Full name:** UNITED STATES of America v. Lawrence NEADLE, Jr., Appellant
- **Court:** Court of Appeals for the Third Circuit
- **Decided:** January 29, 1996
- **Citations:** 72 F.3d 1104; 1995 WL 749562
- **Precedential status:** Published
- **Opinion:** Concurring in part by Becker
- **Judges:** Becker, Nygaard, Roth
- **Cited by:** 20 later opinions in the Frix Law Library

## Citator (automated)

- No negative treatment found by the automated citator. That is not the same as a confirmation that the case is good law; read the citing cases.
- Full citator and citing cases: https://www.frixlaw.com/law-library/cases/9488779

## Opinion text

BECKER, Circuit Judge,
concurring and dissenting.
This appeal presents an important question regarding the definition of “loss” under the fraud section of the United States Sentencing Guidelines (“Guidelines” or “USSG”), USSG § 2Fl.l(b). We must determine whether this definition incorporates a causation requirement, and if so, what that requirement entails. Because the majority fails to explicitly address this issue, and because I disagree with the majority’s conclusion as to the determination of loss, I do not join in Parts III and V of the majority opinion. I do join, however, in Parts I, II, and IV.
*1113 The majority affirms the district court’s calculation of loss without giving sufficient attention to the nexus between Lawrence Neadle’s illegal conduct — misrepresentations that AMP AC had complied with the requirement that insurance companies have $700,000 in unencumbered assets — and the $20 million in unpaid AMP AC claims. This inattention is not surprising given the Guidelines’ lack of guidance on the definition of loss. Indeed, USSG § 2F1.1 seems to envision loss as a readily apparent financial harm existing independent of legal definitions.
While the amount of unpaid AMPAC claims may, at first glance, seem to be the relevant financial harm, the severe consequences of criminal penalties impose on courts the duty to undertake a more searching inquiry. Legal concepts must be susceptible to definition, and it is our duty to explicate these definitions. We have performed this task before, see, e.g., United States v. Kopp, 951 F.2d 521 (3d Cir.1991) (defining loss under USSG § 2F1.1), and it is now necessary to do it again. The Guidelines’ language, its commentary, the caselaw, and sound sentencing policy all lead me to conclude that a financial harm is loss under section 2F1.1 only if it was caused by the defendant’s illegal conduct. This causation requirement demands, at the least, that the defendant’s conduct be a, “cause in fact” of the harm at issue, i.e., that the harm would not have occurred but for the defendant’s conduct.
I dissent because, applying the “but for” standard, the record contains insufficient evidence to find that Neadle’s fraud was a cause of the $20 million in unpaid claims. Because of this error in loss calculation, I would vacate and remand for resentencing.
I. Causation as an Element of Loss
In my view, the Guidelines require a finding of causation before a harm may be used as loss for purposes of section 2F1.1. The causation requirement, pervasive in the criminal law, see infra part II, is made explicit in the language of the Guidelines and is buttressed by caselaw and policy considerations.
A. The Language of the Sentencing Guidelines
The plain language of the Guidelines dictates that courts must make a finding of causation before assigning some harm as loss under section 2F1.1. Although the text of section 2F1.1 contains no definition of the specific offense characteristic loss, the Guidelines provide for such deficiencies by establishing default rules that govern, inter alia, what conduct and harms are relevant to determining specific offense characteristics. One of these default rules makes clear that courts may consider only those harms that were caused by the defendant’s conduct. Because no other provision of the Guidelines provides instructions to the contrary, this default rule governs the determination of loss under section 2F1.1.
Because the loss table in effect at the tíme of sentencing — the 1993 loss table — would provide for a greater penalty than would the loss table in effect at the time of the offense — the 1988 loss table — the 1988 Guidelines apply in their entirety. 1 Thus, except
*1114 where noted, my discussion is based on the 1988 Guidelines.
1. Section 1B1.3
The text of section 2F1.1 provides no definition of the term loss. It simply states that “[i]f the loss exceeded $2,000, increase the offense level as follows,” and then provides a table of sentencing increases based on different amounts of loss. See USSG § 2Fl.l(b). Thus, we must look elsewhere for the definition of loss.
The first place to look is Chapter 1, Part B of the Guidelines, which provides guidance on how to interpret the Guidelines’ sometimes sparse provisions. Within this chapter, USSG § 1B1.3 specifies what information courts may consider in determining, inter alia, specific offense characteristics such as loss under § 2F1.1. See USSG § lB1.3(a). The information specified in section lB1.3(a) is the only information relevant to determining specific offense characteristics “[u]nless otherwise specified” by another provision of the Guidelines. Id.; see also USSG § 1B1.3 Background (“Subsection (a) establishes a rule of construction by specifying, in the absence of more explicit instructions in the context of a specific guideline, the range of conduct that is relevant to determining the offense level....”).
Section lB1.3(a) establishes several important interpretive rules. Subsection (a)(1) states that specific offense characteristics will be based only on the following conduct: acts and omissions committed, aided and abetted by the defendant, or for “which the defendant would be otherwise accountable” (defined as conduct counseled, commanded, induced, procured, or willfully caused by the defendant, reasonably foreseeable acts in furtherance of a conspiracy, or, conduct underlying a conviction for solicitation, misprision or accessory after the fact that reasonably should have been known by the defendant) 2 “that occurred during the commission of the offense of conviction, in preparation for that offense, or in the course of attempting to avoid detection or responsibility for that offense, or that otherwise were in furtherance of that offense.” USSG § lB1.3(a)(l). Except for a special class of offenses — “offenses of a character for which § 3D1.2(d) would require grouping of multiple counts” — this is the only conduct courts may consider absent an expression of contrary intent. 3
More importantly, subsection (a)(3) establishes a causation requirement with respect to harms. Unless otherwise specified, the only harm that is to be taken into account in determining a specific offense characteristic is harm “that resulted from ” the acts and omissions identified above “if the harm ... was caused ” with the requisite level of intent, and harm that was “the object of such acts and omissions.” USSG *1115 § lB1.3(a)(3) (emphasis added). 4 That the harm must be caused with some bad intent necessitates that the harm be caused in the first place. Furthermore, the plain meaning of “resulted from” connotes causation. Indeed, Webster’s Third New International Dictionary defines the verb “result” as follows: “to proceed, spring, or arise as a consequence, effect, or conclusion.” Webster’s Third International Dictionary of the English Language 1937 (Philip Babcock Gove ed., 1966). Thus, absent an expression to the contrary, a court may take into account only harm that “arose as a consequence” of the defendant’s conduct — and only if that harm was “caused” with some bad intent— and (perhaps for offenses that were thwarted prior to their completion) harm that was “the object” or purpose of the defendant’s conduct. 5
Because loss under section 2F1.1 is clearly a “harm” within the meaning of section 1B1.3, see USSG § 1B1.3 Application Note 3 (defining “harm” as “bodily injury, monetary loss, property damage and any resulting harm” (emphasis added)), section lB1.3’s requirement of causation applies. Thus, if no other Guideline provisions addressed the definition of loss, the bare statement in the text of section 2F1.1 to increase the defendant’s sentence on the basis of loss must be read to include a requirement of causation.
2. Other Guidelines Provisions
Turning now to other provisions of the Guidelines that address the issue of loss, the question is not whether they explicitly state a causation requirement, but whether they contradict the rule of causation established by section 1B1.3. I conclude that the Application Notes to section 2F1.1, and the materials referenced therein, only reinforce § lB1.3’s causation requirement.
Application Note 7 to section 2F1.1 deals specifically with the definition of loss, but it provides little guidance. It states as follows:
Valuation of loss is discussed in the Commentary to § 2B1.1 (Larceny, Embezzlement, and Other Forms of Theft). In keeping with the Commission’s policy on attempts, if a probable or intended loss that the defendant was attempting to inflict can be determined, that figure would be used if it was larger than the actual loss. For example, if the fraud consisted of attempting to sell $40,000 in worthless securities, or representing that a forged check for $40,000 was genuine, the “loss” would be treated as $40,000 for purposes of this guideline.
USSG § 2F1.1 Application Note 7.
The Commentary to section 2B1.1 (the Guideline for “Larceny, Embezzlement, and Other Forms of Theft”), referenced above, provides slightly more guidance:
“Loss” means the value of the property taken, damaged or destroyed. Ordinarily, when property is taken or destroyed the *1116 loss is the fair market value of the particular property at issue. Where the market value is difficult to ascertain or inadequate to measure harm to the victim, the court may measure loss in some other way, such as reasonable replacement cost to the victim. When property is damaged, the loss is the cost of repairs, not to exceed the loss had the property been destroyed.
USSG § 2B1.1 Application Note 2. Thus, for theft and similar crimes, the Guidelines state that “ ‘[floss’ means the value of the property taken, damaged, or destroyed.”
This definition certainly does not negate section lB1.3’s background rule of causation. If anything, the definition itself suggests a causation requirement. Implicit in this definition is that loss means the value of the property taken, damaged, or destroyed by the defendant. The terms “taken,” “damaged,” and “destroyed” have meaning only insofar as there is a subject — a taker, damager, or destroyer. In this context, the subject must be the defendant — the person being sentenced for taking, damaging, or destroying some property. In other words, loss for purposes of section 2B1.1 is the value of the property that the defendant caused to be taken, damaged, or destroyed. 6
Another provision touching on the concept of loss is Application Note 8 to section 2F1.1, which deals with loss estimation. It states:
The amount of loss need not be precise. The court is not expected to identify each victim and the loss he suffered to arrive at an exact figure. The court need only make a reasonable estimate of the range of loss, given the available information. The estimate may be based on the approximate number of victims and an estimate of the average loss to each victim, or on more general factors, such as the nature and duration of the fraud and the revenues generated by similar operations. Estimates based upon aggregate “market loss” (e.g., the aggregate decline in market value of a stock resulting from disclosure of information that was wrongfully withheld or misrepresented) are especially appropriate for securities cases. The offender’s gross gain from committing the fraud is an alternative estimate that ordinarily will underestimate the loss.
USSG § 2F1.1 Application Note 8. This provision’s, discussion of loss estimation does not undercut the background requirement of causation.
The only provision that explicitly mentions the issue of causation is Application Note 11. It states:
In a few instances, the total dollar loss that results from the offense may overstate its seriousness. Such situations typically occur when a misrepresentation is of limited materiality or is not the sole cause of the loss. Examples would include understating debts to a limited degree in order to obtain a substantial loan which the defendant genuinely expected to repay; attempting to negotiate an instrument that was so obviously fraudulent that no one would seriously consider honoring it; and making a misrepresentation in a securities offering that enabled the securities to be sold at inflated prices, but where the value of the securities subsequently declined in substantial part for other reasons. In such instances, a downward departure may be warranted.
USSG § 2F1.1 Application Note 11.
The phrase “when a misrepresentation is ... not the sole cause of the loss” might suggest that the loss figure can include harms not caused by the defendant. But this is a misreading of the Application Note. That the defendant’s conduct need not be the sole cause of the loss in no way excuses that conduct from being a cause. 7 The availabili *1117 ty of a downward departure for situations in which a misrepresentation is not the “sole cause” of the loss simply accounts for the fact that some events have multiple causes. If causes beyond the defendant’s control played a large role in the loss, a downward departure may be justified. If anything, the Note’s discussion of a misrepresentation that is not the “sole cause” of the loss implies that such misrepresentation is a cause of the loss, and the Note thus supports the need for a finding of causation. See United States v. Kopp, 951 F.2d 521, 531 (3d Cir.1991) (“Application Note 11 made it clear that actual loss was how much better off the victim would be but for the defendant’s fraud. To the extent actual loss had other, more proximate causes, a discretionary downward departure ... might be appropriate.” (emphasis added)). 8
In the end, the Application Notes to section 2F1.1, and the materials referenced therein, only reinforce § lB1.3’s causation requirement. Reading these materials together with the background rule of causation, the Guidelines seem to contemplate the following approach: The court should first make an estimate of the loss, i.e., the financial harm caused by the defendant. If the harm caused by the defendant was also caused by other factors beyond the defendant’s control, the court should consider making a downward departure to better reflect the seriousness of the defendant’s offense.
*1118 B. Caselaw
This Court’s caselaw also suggests the presence of a causation requirement in the definition of loss. In United States v. Kopp, 951 F.2d 521 , we exhaustively analyzed the definition of loss under section 2F1.1. We concluded that the definition of loss for fraud sentencing under section 2F1.1 is not the same as section 2Bl.l’s “amount taken” rule even though section 2F1.1 makes reference to section 2Bl.l’s definition of loss. See id. at 529 (“Application Note 7 to USSG § 2F1.1 does not say that the definitions of ‘loss’ for theft and fraud cases are identical, just that ‘[vjaluation of loss is discussed in the Commentary to § 2B1.1...In Kopp , the defendant procured a bank loan by means of fraudulent misrepresentations. The district court calculated loss as the full amount of the loan, despite the fact that the victim of the fraud — the bank — recovered most of the loan amount by selling the property securing the loan. In rejecting the government’s argument that loss was appropriately calculated because the face value of the loan was the amount “taken,” USSG § 2B1.1 Note 2, we sought to develop a sensible definition of loss for the fraud context. Thus, we defined “fraud ‘loss’ as ... the amount of money the victim has actually lost.” Id. at 536. 9
Although we were not directly confronted with the causation issue we face here, we suggested that our definition of fraud loss as “the amount of money the victim has actually lost” incorporates a causation requirement. We stated, in response to the government’s argument, that “Application Note 11 [of the original Guidelines] made it clear that actual loss was how much better off the victim would be but for the defendant’s fraud.” Id. at 531 (emphasis added). This “but for” statement is a classic articulation of the causation requirement. See infra part II. More importantly, the fairness concerns that drove Kopp are very much present in this case. In Kopp , we were concerned with the district court’s assignment as loss a financial harm that had not occurred. Although the issue here — the district court’s assignment as loss a financial harm that did occur but which may not have been caused by the defendant — is perhaps analytically distinct, it shares the same basic problem: punishing the defendant for harm that he or she did not cause.
I also note that this Court and others have repeatedly suggested the need for finding causation in making a loss determination. See, e.g., United States v. Daddona, 34 F.3d 163 , 170 (3d Cir.) (remanding for resentenc-ing because the record did not contain any indication that the loss figure used by the district court “was due to the fraud of the appellants”), cert. denied, — U.S. -, 115 S.Ct. 515 , 130 L.Ed.2d 421 (1994); United States v. Marlatt, 24 F.3d 1005, 1007 (7th Cir.1994) (applying proximate cause analysis in assessing the district court’s determination of loss); United States v. Harper, 32 F.3d 1387, 1392 (9th Cir.1994) (“What we do insist upon, however, is use of a realistic, economic approach to determining what losses he truly caused or intended to cause_”), cert. denied, — U.S. -, 115 S.Ct. 1162 , 130 L.Ed.2d 1118 (1995); United States v. Wilson, 980 F.2d 259, 262 (4th Cir.1992) (“When the offense involves making a false statement, the inquiry to determine loss must focus on the amount of loss related to the false statement.”).
C. Policy Considerations
Finally, good sentencing policy requires that there be a causal link between the defendant’s crime and the harm on which his or her sentence is based. This causation requirement effects two fundamental Guidelines sentencing policies: First, that courts sentence defendants according to “the nature and degree of the harm caused by the[ir] offense[s],” United States v. Kopp, 951 F.2d 521, 529 (3d Cir.1991) (quoting 28 U.S.C.A. § 994 (c)(3) (West Supp.1991)) (emphasis added); and second, that courts sentence similarly situated defendants similarly, see id.; USSG Part A, section 3, p. 1.2 (1988) (“Congress sought uniformity in sentencing by narrowing the wide disparity in sentences *1119 imposed by different federal courts for similar criminal conduct by similar offenders.”). Sentencing defendants on the basis of fortuitous harm that they in no sense caused would thwart both of these policies.
D. Conclusion
In summary, I am convinced that a harm cannot be considéred loss under section 2F1.1 unless the defendant’s conduct was a cause of that harm. The causation requirement emanates from the interpretive principles of section 1B1.3, the provisions of the fraud and theft guidelines, and is supported by caselaw and policy considerations. However, because of the sparse and sometimes confusing guidance provided by the Application Notes, I believe that the Sentencing Commission should articulate a comprehensive approach to determining loss. Until that time, I believe that my approach best harmonizes the many provisions at issue and best accords with fundamental sentencing policy.
II. The Content of the Causation Requirement
Having established that the Guidelines contemplate a causation requirement in its definition of loss, I must now address the more difficult question of the content of the causation requirement. I need not, however, establish a comprehensive definition of causation under the Guidelines, for the figure approved by the majority — the $20 million in unpaid claims — fails even the most minimal causation test. In addition, it would be unwise to embark on such an ambitious task when subsequent cases faced with concrete problems can better resolve the various issues raised.
The notion of causation runs throughout the law — including the criminal law — and it is generally understood to encompass two concepts. A defendant’s conduct must generally be both the “cause in fact” and the “proximate cause” of some harm before liability is imposed. See Wayne R. LaFave & Austin W. Scott, Jr., 1 Substantive Criminal Law § 3.12, at 393-99 (1986); Richard A. Epstein, Cases and Materials on Torts 363-64 (5th ed. 1990); 4 Fowler V. Harper et al., The Law of Torts § 20.2, at 89-90, § 20.4, at 130-33 (2d ed. 1986) (“Harper, James, & Gray ”); W. Page Keeton et al., Prosser and Keeton on The Law of Torts § 41, at 263-65 (5th ed. 1984) (“Prosser & Keeton ”). While neither of these concepts are susceptible to uncontroversial definitions, a working understanding of them is useful.
The requirement of cause in fact purports to be an empirical test of whether the defendant’s conduct was a necessary antecedent to the harm at issue. See, e.g., Harper, James, & Gray, supra § 20.2, at 89-91. The most common expression of this test is the “but for” formulation: the defendant’s conduct is a cause in fact of some harm if the harm would not have occurred but for the defendant’s conduct. See LaFave & Scott, supra, at 393-94; Harper, James, & Gray, supra, § 20.2, at 91; Prosser & Keeton, supra, at 265-66. LaFave and Scott note that the Model Penal Code and several state codes put forth the “but for” test explicitly. See LaFave & Scott, supra, at 394 n. 11 (stating that “Model Penal Code § 2.03(1) declares that conduct ‘is the cause of a result when ... it is an antecedent but for which the result in question would not have occurred’ ” and citing similar state codes).
Courts sometimes apply a different test— the “substantial factor” formulation — to special causation problems not adequately addressed by the “but for” test. For example, when the conduct of each of two actors acting independently could have caused the harm at issue, and thus neither is the “but for” cause, courts generally hold both actors hable under the substantial factor test. See, e.g., Prosser & Keeton, supra, at 268.
Proximate cause, on the other hand, is a more explicitly policy-based determination of whether an actor’s conduct, despite its being a cause in fact, is too tenuously linked to the injury, to hold the actor liable. See, e.g., Prosser & Keeton, supra, § 42, at 272-73 (5th ed.1984). Courts have used various different tests to address this issue. See, e.g., id. at 273-80.
In this case, we are concerned only with *1120 the requirement of cause in fact. 10 Because the facts do not present any special causation problems, I need only apply the standard “but for” test. I now turn to a review of the facts and application of the “but for” test to them.
III. Application of the Causation Requirement
A. The Misrepresentation
I take issue with the majority’s assertion that on the facts of this case, there is a “clear causal connection between the fraud and the policy holders’ losses.” Neadle’s only fraud was in misrepresenting that he had $700,000 of unsecured assets backing his insurance venture. In order to conclude, as the majority and district court do, that this fraud was a “but for” cause of the over $20 million in unpaid claims, one must make the following inferences: first, that had Neadle not committed fraud, he would not have obtained an insurance license; second, that the AMPAC insureds would have thus purchased insurance from other insurance companies; and third, that these insurance companies would have paid all of their Hurricane Hugo related claims. 11 While these inferences might be plausible in theory, there is no evidence in the record to support them. In fact, record evidence undercuts the plausibility of such inferences.
First, there is no evidence that Neadle would have been unable to obtain an insurance license without committing fraud. Given his ability to post $500,000 in cash when his surety bond was called into question, he might have found backers for another $700,-000. If Neadle had obtained a license legitimately, he could have issued insurance to the same policy holders, and AMPAC still would have been unable to meet the $24 million in Hurricane Hugo claims. At best, the policyholders would have obtained an additional $700,000. 12
Second, no record evidence supports the inference that had AMPAC not been in the insurance business, AMPAC policy holders would have purchased insurance from other companies. This is a striking omission given the recent history of the unavailability of insurance in the Virgin Islands. Indeed, the record shows that only local companies wrote policies for the same kinds of customers and coverage that AMPAC did, and that there were few local companies in operation. Dev-erita Sturdivant, then director of Banking and Insurance, testified before the grand jury as follows:
Considering the fact that we had very few companies operating in the territory at the time that [Neadle] came on board, as I said, we had a mass exodus of companies from the territory in ’85 and we had chaos, basically with unauthorized companies in ’87. And so by the time Mr. Neadle and a couple of other domestics came on the scene in ’88, late ’87, we felt that perhaps the market was turning around. The availability problem was no longer acute because people were receiving cover [sic]. And oftentimes the small domestic companies were willing to provide the kind of coverage that many of the larger companies would not provide. [The larger companies] often could pick and choose their clients.
*1121 Thus, one cannot assume that the AMPAC policy holders would have found insurance elsewhere.
Finally, the record is devoid of evidence supporting the inference that other insurance companies would have met all of their Hurricane Hugo claims. Rather, the record shows that at least one insurance company that did not misrepresent the amount of its assets, American Alliance, was unable to pay all of its Hurricane Hugo claims. The record also shows that most of the small domestic insurance companies — possibly the only companies covering the same kinds of risks and policyholders as AMPAC — lacked the assets necessary to cover the huge Hurricane Hugo losses. Sturdivant testified before the grand jury that “just about all of the companies that came on board, the small domestic companies, came with the minimum statutory capital and surplus requirements. They came with $1.2 million and that’s it.” If AMPAC had not obtained a license to sell insurance, its customers may have purchased insurance from American Alliance or other companies unable to meet Hurricane Hugo claims.
On this record, then, I cannot conclude that Neadle’s fraud was a cause in fact of AMP AC’s inability to pay the over $20 million in claims. In summary: (1) had Neadle met the $700,000 requirement, AMPAC would still have been unable to pay the Hugo losses; (2) the district court’s (and the majority’s) intimation that other companies would have written the AMPAC policies is purely speculative and unsupported by the record; (3) if other companies were willing to provide the coverage AMPAC did, they may have been similarly undercapitalized; and (4) at least one other company that met the $700,-000 requirement was unable to pay its Hugo claims.
B. United States v. Robichaux
United States v. Robichaux, 995 F.2d 565 (5th Cir.), cert. denied, — U.S. -, 114 S.Ct. 322 , 126 L.Ed.2d 268 (1993), which the majority cites to support its decision, is distinguishable. In that case, the defendant misrepresented to an auditing firm hired by the Louisiana Insurance Commission that securities he had assigned to an insurance company were unencumbered. The firm relied on the defendant’s statement and issued a favorable audit. Two years later, the company was declared insolvent. The Fifth Circuit upheld the district court’s conclusion that the defendant was responsible for all of the losses attributable to the insurance company’s failure.
' Notably, the Fifth Circuit applied a cause in fact analysis similar to the one I describe in this opinion. However, it found that the district court’s inferences were “plausible in light of the record read as a whole”:
It is not clearly erroneous to assume that if [the auditor] had not issued a favorable audit for [the insurance company], which only occurred because of [the company’s] fraudulently inflated balance sheet, the Commission would have acted to liquidate the firm at an earlier date and minimized the losses.
Id. at 571 .
The Fifth Circuit’s holding that the district court’s inferences were “plausible in light, of the record read as a whole” says nothing about this case, where the record casts serious doubt on the plausibility of the district court’s inferences. To the extent Robichaux supports a relaxed inquiry into causation— which I do not think is the case — I disagree with it for the reasons expressed in Part I.
C. The Majority’s Other Arguments
The remainder of the majority’s reasoning fares no better. The majority asserts that “the deceitful and slipshod way in which Nea-dle ran the business” — i.e., failure to keep proper books and misrepresentations in quarterly reports — “supports the district court’s attribution of responsibility to him for the [$20 million] losses.” This assertion suggests two possible arguments, neither of which is convincing.
First, the majority might mean that Nea-dle’s poor record-keeping was a cause of the unpaid AMPAC claims. Evidence in the record suggests that beyond the minimum capital requirements, the Insurance Division policed the funds of insurance companies to ensure that companies maintained adequate *1122 reserves in proportion to the coverage they wrote. Thus, if we assume that AMPAC was inadequately capitalized, 13 it is possible that if Neadle had kept accurate books, the Insurance Division might have discovered the inadequacy and taken remedial measures.
However, even if Neadle’s poor accounting was, in some sense, a cause of the unpaid AMPAC claims, the Guidelines do not allow Neadle to be sentenced on the basis of this harm. Under section 1B1.3, only certain harms sufficiently connected to the defendant’s criminal conduct can be used to calculate a specific offense characteristic. See supra part I. On a different record, it might be argued that Neadle’s poor record-keeping was a means of hiding his misrepresentations and thus that harms caused by the poor record-keeping “occurred ... in the course of attempting to avoid detection or responsibility for th[e] offense [of conviction],” USSG § lB1.3(a)(l). But here no record evidence suggests that Neadle’s lax accounting was a means of avoiding detection.
Second, the majority might be asserting that Neadle meant to use AMPAC essentially to rob his clients, and thus that he intended to cause some significant loss. Courts may sentence defendants on the basis of “intended loss” if that figure is greater than actual loss. USSG § 2F1.1 Application Note 7; see also USSG § lB1.3(a)(3) (stating that courts may consider harm that was “the object of’ the defendant’s conduct).
However, once again, there is no record evidence that Neadle intended any loss. In fact, the record is replete with evidence to the contrary. Neadle purchased $4 million worth of reinsurance, an amount, that according to his uncontroverted testimony, professionals had advised was sufficient. The Virgin Islands did not require that companies have any reinsurance, and no one in the Virgin Islands government told Neadle how much reinsurance AMPAC should carry. Until Hurricane Hugo, AMPAC paid claims promptly, and was never reprimanded or sanctioned by the Insurance Division for wrongful denial or failure to pay claims.
D. The Proper Calculation of Loss
Having concluded that the bulk of the unpaid AMPAC claims was not caused by Nea-dle’s misconduct and thus cannot be assigned as loss under section 2F1.1, I feel obligated to suggest how loss might appropriately be calculated. In so doing, I question the basic model employed by the district court and the majority — that loss is based on the unmet claims of the policyholders.
One reasonable estimation of loss in this case is the $700,000 that Neadle misrepresented as unencumbered. Under the majority’s model, $700,000 is a sensible loss determination on several grounds. Had Neadle not committed fraud and still obtained an insurance license, AMPAC would probably have $700,000 more in assets than it does now, and the AMPAC insureds would be that much better off. Or, had the AMPAC insureds bought insurance from other legitimately formed insurance companies, the evidence suggests that these companies may have been similarly capitalized.
A different model for measuring loss that is consistent with the text and structure of section 2F1.1 would also generate a figure of $700,000. The indictment charged Neadle with misrepresenting the status of the $700,-000 to the Virgin Islands government. If we focus on the Virgin Islands as the victim of the fraud, the loss is the governmental or societal loss of the $700,000 reserve.
Alternatively, we might measure loss in terms of what Neadle obtained by virtue of his fraud. See USSG § 2F1.1 Application Note 8; see also Kopp, 951 F.2d at 530 (stating that the defendant’s gain can be used as a measure of loss when there is some loss but it is not measurable). Here, Neadle obtained the premiums of the policyholders to whom he falsely represented were validly insured. While I cannot derive this sum from the record, I surmise that it approaches $700,000 annually.
I do not deny that a sentence based on my suggested loss calculations might give Nea- *1123 die less time than he deserves. To rectify this problem, the district court could depart upward if the case meets the departure standard (although upper calibration would have to start at the proper base offense level). See United States v. Kikumura, 918 F.2d 1084 (3d Cir.1990). If the resulting sentence is still too lenient, that is a problem that only the Commission can address.
IV. Conclusion
For the reasons I have stated, the Guidelines require a finding of causation before some harm is deemed loss under section 2F1.1. This requirement demands, at the least, that the defendant’s conduct be a cause in fact of the harm. Because the record contains insufficient evidence to find that Neadle’s fraud was a cause in fact of the $20 million in unpaid claims, I respectfully dissent.
. USSG § 1B1.11(b)(1) (1994) states that "[i]f the court determines that use of the Guidelines Manual in effect on the date that the defendant is sentenced would violate the ex post facto clause of the United States Constitution, the court shall use the Guidelines Manual in effect on the date that the offense of conviction was committed.” USSG § IB 1.11 (b)(2) (1994) then states that "[t]he Guidelines Manual in effect on a particular date shall be applied in its entirety.” See generally United States v. Corrado, 53 F.3d 620 (3d Cir.1995) (explaining this "one book rule”). Because the loss table in effect at the time of sentencing provides for more jail time per dollar of loss than do earlier versions of the loss table, see majority opinion note 3, the Guidelines applicable at the time of the offense apply in their entirety.
The question then is: what version of the Guidelines was in effect at the time of Neadle’s offense? The majority and the district court apply the 1988 edition of the Guidelines, which incorporates amendments effective October 15, 1988. They do this despite the fact that there is a strong argument that the offense for which Neadle was convicted occurred on January 5, 1988. Pursuant to a plea agreement, Neadle was convicted of only the mail fraud charge alleged in count two of the indictment. Count two charges that the crime of mail fraud occurred on January 5, 1988, when Neadle mailed a letter to the Insurance Division stating that its assets were unencumbered. Nevertheless, the district court's *1114 judgment states that the offense was concluded on October 31, 1989. The district court’s finding as to the duration of the offense seems justified by the fact that Neadle concealed his initial mail fraud by continuously filing false quarterly reports. Although it is unclear whether Neadle accepts the district court’s finding as to the duration of his offense, he seems to acquiesce in the use of the 1988 Guidelines. Because of Neadle's acquiescence and the agreement of the government, the district court, and the majority that the 1988 Guidelines apply, I will use the 1988 Guidelines aá well.
. Application Note 1 defines "conduct 'for which the defendant is otherwise accountable’ " as follows:
Conduct "for which the defendant is otherwise accountable," as used in subsection (a)(1), includes conduct that the defendant counseled, commanded, induced, procured, or willfully caused. If the conviction is for conspiracy, it includes conduct in furtherance of the conspiracy that was known to or was reasonably foreseeable by the defendant. If the conviction is for solicitation, misprision or accessory after the fact, it includes all conduct relevant to determining the offense level for the underlying offense that was known to or reasonably should have been known by the defendant.
USSG § IB 1.3 Application Note 1 (citations omitted). This definition has been, for the most part, incorporated into the text of the current version of section 1B1.3. See USSG § lB1.3(a)(l) (1994). We may consider subsequent amendments to the Guidelines "to the extent that such amendments are clarifying rather than substantive changes.” USSG § IB.11(b)(2) (1994).
. "Solely with respect to offenses of a character for which § 3D 1.2(d) would require grouping of multiple counts,” courts may also take into account "all such acts and omissions that were part of the same course of conduct or common scheme or plan as the offense of conviction.” USSG § IB 1.3(a)(2).
. In full, Section 1B1.3(a)(3) states that courts may consider
all harm or risk of harm that resulted from the acts and omissions specified in subsections (a)(1) and (a)(2) above, if the harm or risk was caused intentionally, recklessly or by criminal negligence, and all harm or risk that was the object of such acts or omissions.
USSG § lB1.3(a)(3).
. I recognize that subsequent versions of the Guidelines have deleted the requirement that the harm be caused with bad intent, as well as the "risk of harm” language. In the 1994 Guidelines, for example, section lB1.3(a)(3) states that courts may consider “all harm that resulted from the acts and omissions specified in subsections (a)(1) and (a)(2) above, and all harm that was the object of such acts and omissions.” The Committee explained these changes as follows:
The purpose of this amendment is to delete language pertaining to "risk of harm” and "state of mind" as unnecessary. Cases in which the guidelines specifically address risk of harm or state of mind are covered in the amended guideline under subsection (a)(4) [formerly subsection (a)(5) ]. In addition, the amendment deletes reference to harm committed “intentionally, recklessly, or by criminal negligence” as unnecessary and potentially confusing.
USSG Appendix C, Amendment 76, p. 86 (1994). To the extent these changes can be characterized as "clarifying rather than substantive changes,” which we are invited to take into account in understanding an earlier version of the Guidelines, see USSG § IB.11(b)(2) (1994), they do not change my view that the Guidelines establish a causation requirement. Section lB1.3(a)(3) retains the "resulted from” language, which, as I state in the text, connotes causation.
. As I explain below, see infra section I.B, we have held that loss for purposes of section 2B1.1 is not identical to loss for purposes of section 2F1.1, despite the latter’s reference to the former's definition. My point here is simply that nothing in the commentary to section 2B1.1 undercuts the rule that only harms in some sense caused by the defendant can be assigned as loss under section 2F1.1.
. My view is not changed by subsequent amendments to the Application Notes to section 2F1.1. Even if these amendments are given weight as “clarifying” changes, I do not find them sufficient to overcome the background rule of causation. In lieu of Application Note 11, Application Note 7 to section 2F1.1, which deals specifically with the definition of ''loss,” now contains an additional paragraph:
*1117 There are, however, instances where additional factors are to be considered in determining the loss or intended loss:
(b) Fraudulent Loan Application and Contract Procurement Cases
In fraudulent loan application cases and contract procurement cases, the loss is the actual loss to the victim (or- if the loss has not yet come about, the expected loss). For example, if a defendant fraudulently obtains a loan by misrepresenting the value of his assets, the loss is the amount of the loan not repaid a the time the offense is discovered, reduced by the amount the lending institution has recovered (or can expect to recover) from any assets pledged to secure the loan. However, where the intended loss is greater than the actual loss, the intended loss is to be used.
In some cases, the loss determined above may significantly understate or overstate the seriousness of the defendant's conduct. For example, where the defendant substantially understated his debts to obtain a loan, which he nevertheless repaid, the loss determined above (zero loss) will tend not to reflect adequately the risk of loss created by the defendant's conduct. Conversely, a defendant may understate his debts to a limited degree to obtain a loan {e.g., to expand a grain export business), which he genuinely expected to repay and for which he would have qualified at a higher interest rate had he made truthful disclosure, but he is unable to repay the loan because of some unforeseen event (e.g., an embargo imposed on grain exports) which would have caused a default in any event. In such a case, the loss determined above may overstate the seriousness of the defendant’s conduct. Where the loss determined above significantly understates or overstates the seriousness of the defendant's conduct, an upward or downward departure may be warranted.
USSG § 2F1.1 Application Note 7 (1994). In the second example in this note — the grain exporter example — the Guidelines seem to assign as "loss” a harm that would have occurred regardless of the defendant's fraud. The example assumes that the grain exporter would have (1) obtained the loan whether or not he committed fraud; and (2) defaulted on the loan whether or not he committed fraud. Nevertheless, the example suggests that the lack of causation is a ground for downward departure, and thus, the argument goes, the lack of causation should not be factored into "loss" in the first instance.
I reject this argument for ihe same reasons that I state in the text. In addition, I note that the grain exporter example is stated under the heading "Fraudulent Loan Application and Contract Procurement Cases,” of which the present case is neither.
. The statements from Kopp quoted in the text shed light on some other confusing language from that case. Language in Kopp arguably suggests that Application Note 11 mandates taking account of the lack of causation only by way of mtiking a downward departure, and not by adjusting the amount of loss:
The government is correct on one point, however: Application Note 11 definitively rejected adjusting the “loss” itself downward to reflect other causes beyond the defendant’s control. As an example of when the dollar loss may overstate the seriousness of the defense and hence a downward departure may be appropriate, Application Note 11 included situations where the "misrepresentation ... is not the sole cause of the loss."
951 F.2d at 531 . However, this language was followed immediately by the statements quoted in the text, which make clear that we were speaking of situations in which the defendant caused the loss, but where other factors also played a causal role.
. The current Guidelines incorporate this definition of fraud loss. See USSG § 2F1.1 Application Note 7 (1994).
. It may ultimately be appropriate to hold that only harm proximately caused by the defendant's conduct can be deemed "loss.” See United States v. Marlatt, 24 F.3d 1005, 1007 (7th Cir.1994) (applying proximate cause analysis to a district court's determination of loss). However, it is unnecessary to reach this question on the facts of this case.
. These inferences were explicit in the district court's opinion:
If the defendant had not obtained his license by fraud, he would not have been in a position to issue insurance coverage and would not have had the customers or policyholders he did have. Therefore, the policyholders would have obtained coverage from other insurance companies, companies that did meet Hurricane Hugo claims.
.Even if AMPAC had initially held an additional $700,000 in unencumbered assets, it is unclear that the government or the policyholders would have obtained these funds. An insurance company, once qualified, is not required to keep the full $700,000 as part of its capital structure. Subject to certain limitations, an insurance company can "make use of its surplus for the development of its business.” 22 V.I.Code § 465.
. This assumption is probably not justified given the $4 million in reinsurance that AMPAC maintained. See infra.

---

Source: Frix Law Library, https://www.frixlaw.com/law-library/cases/9488779. Public record. Not legal advice.