thefact that Citibank, because it enjoyed the benefit of titleinsurance, never ran a risk of losing anything.......

United States Court of Appeals

For the First Circuit











No. 98-1976



UNITED STATES OF AMERICA,



Appellee,



v.



JOSE E. ALEGRIA,



Defendant, Appellant.







APPEAL FROM THE UNITED STATES DISTRICT COURT



FOR THE DISTRICT OF PUERTO RICO



[Hon. Juan M. P‚rez-Gim‚nez, U.S. District Judge]









Before



Selya, Boudin and Lipez,



Circuit Judges.







Alan M. Dershowitz, with whom Nathan Z. Dershowitz, Amy

Adelson, and Dershowitz & Eiger, P.C. were on brief, for appellant.

Jorge E. Vega-Pacheco, Assistant United States Attorney, with

whom Guillermo Gil, United States Attorney, and Nelson P‚rez-Sosa,

Assistant United States Attorney, were on brief, for appellee.











September 30, 1999









 SELYA, Circuit Judge. This appeal requires us, inter

alia, to explore the circumstances in which the government may be

compelled to move for a downward departure under USSG 5K1.1. We

conclude that the district court did not err either in refusing to

force the government to take such action or in any other material

respect. Consequently, we affirm.

I

A federal grand jury indicted defendant-appellant Jos‚ E.

Alegr¡a on sixteen counts of filing false statements with financial

institutions, 18 U.S.C.  1014, and bank fraud, 18 U.S.C.  1344.

He entered into a plea agreement with the government (the

Agreement), pled guilty to all charges, and met twice with

government agents pursuant to a promise to cooperate. We have no

detailed account of these debriefing sessions, but the appellant

states in a declaration (filed below in connection with his motion

for an evidentiary hearing) that he furnished the government with

whatever information he possessed concerning wrongdoing at the

financial institutions with which he was associated.

Despite the appellant's cooperation, the prosecutor

elected not to file a downward departure motion. The appellant

asserted that the prosecutor's decision contravened the Agreement

and, in the bargain, violated due process. The sentencing court

rejected these animadversions, see United States v. Alegr¡a, 3 F.

Supp. 2d 151 (D.P.R. 1998), denied the appellant's motion for an

evidentiary hearing, and proceeded to impose a 30-month

incarcerative sentence. In this forum, the appellant continues to

press his claim that the government wrongly refused to file a

downward departure motion and embellishes it with a challenge to

the lower court's calculation of his guideline sentencing range.

II

We start with the appellant's major premise: that the

government obligated itself to file a downward departure motion by

virtue of promises it made during the negotiations that led up to

the execution of the Agreement and in the Agreement itself. For

argument's sake, we take the facts from the appellant's

declaration.

After the indictment was returned and the appellant

entered a "not guilty" plea, the parties began discussing the

possibility of a plea bargain. In his declaration, the appellant

states that he had misgivings about whether the United States

Attorney's office would reward cooperation with a favorable

sentencing recommendation (he traces these misgivings to a previous

case in which the United States Attorney allegedly made similar

overtures to another bank executive, but subsequently reneged), and

therefore arranged to meet personally with Guillermo Gil, the

United States Attorney for the District of Puerto Rico, prior to

settling upon a course of action. According to the appellant, Gil

assured him (in the presence of his then-counsel) that if he would

"tell the truth, be available, and cooperate," the government would

move for a departure under USSG 5K1.1 (permitting a sentencing

court to depart downward on the prosecution's motion, based on a

defendant's "substantial assistance"). The appellant asserts that

this specific representation persuaded him to sign the Agreement

and change his plea. Hence, he asks that we hold the government to

Gil's word.

As a general rule, nothing precludes a prosecutor from

bargaining away something over which he has discretion in return

for promises extracted from a criminal defendant. See United

States v. Doe, 170 F.3d 223, 226 (1st Cir. 1999); United States v.

Hernandez, 17 F.3d 78, 82 (5th Cir. 1994). Relatedly, a binding

prosecutorial representation that is accepted by a defendant and

becomes the basis for a change of plea must be performed. See

Santobello v. New York, 404 U.S. 257, 262 (1971) (holding that

"when a plea rests in any significant degree on a promise or

agreement of the prosecutor, so that it can be said to be part of

the inducement or consideration, [the] promise must be fulfilled").

In the appellant's view, these uncontroversial axioms carry the

day.

But this conclusion depends entirely on the assumption

that what Gil allegedly said has legal force and that assumption

stands on shaky ground because the Agreement, which purports to

encompass the sum and substance of the arrangement between the

parties, was signed after Gil allegedly made the crucial

representation and contains no reference to it. The essential and

logically prior question, then, is whether the representation, even

if made, survives execution of the Agreement.

A

Courts customarily treat plea agreements, for purposes of

construction, more or less in the same manner as they do contracts.

See United States v. Atwood, 963 F.2d 476, 479 (1st Cir. 1992);

United States v. Anderson, 921 F.2d 335, 337-38 (1st Cir. 1990).

We say "more or less" because this analogy has its limitations.

See United States v. Hogan, 862 F.2d 386, 388 (1st Cir. 1988)

(observing that plea agreements are similar to commercial

contracts, but only "in certain respects"). Thus, although

contract law supplies a useful reference point for construing plea

agreements in federal criminal cases, such agreements are not

governed by the law of contracts. See United States v. Kelly, 18

F.3d 612, 616 (8th Cir. 1994).

If a plea agreement unambiguously resolves an issue, that

usually ends the judicial inquiry. See id.; Anderson, 921 F.2d at

338. If, however, a plea agreement lacks clarity or is manifestly

incomplete, the need to disambiguate may justify resort to

supplementary evidence or other interpretive aids. See Anderson,

921 F.2d at 338. We examine the text of the Agreement in light of

this dichotomy.

The appellant calls our attention to paragraph 8 of the

Agreement, which embodies his pledge to "cooperate fully and

truthfully with the United States." After spelling out the

elements of this cooperation which include the typical assurances

that the appellant will remain available for debriefing, appear as

a witness, speak truthfully, provide documents, and so forth the

paragraph explains that he is not expected to "make a case" against

anyone. This is a shorthand way of saying that the government's

obligations under the Agreement are not conditioned upon the

achievement of any particular objective (e.g., the conviction of

some other person), but, in the language of the Agreement, "only

upon [Alegr¡a] providing full, complete and truthful cooperation."

The appellant maintains that this phraseology imports the United

States Attorney's oral representation into the Agreement. We think

that this is a more ambitious reading of the passage than either

the text or the surrounding circumstances allow.

USSG 5K1.1 provides the background understanding against

which the parties signed the Agreement and through which their

arguments must be filtered. See United States v. Huang, 178 F.3d

184, 187-89 (3d Cir. 1999). The appellant's construction

contemplates an equivalency between "full, complete and truthful

cooperation," on the one hand, and "substantial assistance," on the

other. But the language and structure of section 5K1.1 belie the

idea that "full, complete and truthful cooperation" necessarily

constitutes "substantial assistance." The guideline suggests five

non-exclusive factors that a court should consider when deciding

whether it will grant a prosecutor's motion for a downward

departure predicated on a defendant's substantial assistance. See

USSG 5K1.1(a)(1)-(5). Full, complete and truthful cooperation

corresponds to only one of these five factors. See id.

5K1.1(a)(2). The others include things well beyond the purview of

cooperation per se, such as the significance and utility of the

information provided, id. 5K1.1(a)(1), the nature and extent of

the defendant's assistance, id. 5K1.1(a)(3), and the timeliness of

the proffer, id. 5K1.1(a)(5). In short, full, complete and

truthful cooperation, in and of itself, is not coextensive with the

substantial assistance of which the sentencing guidelines speak.

In the case at bar, the Agreement, read as a whole,

plainly was meant to be understood in terms of the general approach

limned in section 5K1.1. Although conditioned on the appellant's

conformance with paragraph 8, nothing in the text of the Agreement

suggests that the parties agreed either to collapsing the

substantial assistance determination into the relatively narrow

confines of paragraph 8 or to some other special definition of

substantial assistance. This point is made pellucid by paragraph

11, which memorializes the appellant's express agreement that "the

United States' decision whether to file a motion based on

'substantial assistance' as that phrase is used in Rule 35(b) of

the Federal Rules of Criminal Procedure and Section 5K1.1 of the

Sentencing Guidelines and Policy Statements . . . rests in the sole

discretion of the United States," and further provides that

disputes about that decision will not be referred to the district

court.

The obvious implication of this explicit reference to

section 5K1.1 is that the prosecutor will take into account all the

factors delineated in that guideline when determining whether to

move for a downward departure and those factors, as we have

noted, go well beyond full, complete and truthful cooperation. In

this way, the Agreement makes it quite clear that compliance with

the covenants contained in paragraph 8 constitutes a necessary, but

not an independently sufficient, precondition to the filing of a

section 5K1.1 motion. Moreover, the reference in paragraph 11 to

Fed. R. Crim. P. 35(b) bolsters, rather than weakens, this

conclusion: with regard to the meaning of "substantial

assistance," Rule 35(b) and USSG 5K1.1 are birds of a feather.

See United States v. Gangi, 45 F.3d 28, 30 (2d Cir. 1995).

The appellant's exhortation that our decision in Doe, 170

F.3d 223, points in a different direction is easily dispatched.

Although the Doe court spoke of internal "tension" within the

contours of a plea agreement, that tension related to a completely

different problem: the plea agreement purported to preserve the

government's absolute discretion in regard to section 5K1.1

motions, but at the same time stated that "the defendant's failure

to 'make a case' shall not relieve the government of exercising its

discretion" under section 5K1.1. Id. at 226. The government

decided not to file a downward departure motion because Doe's

assistance "came too late." Id. In that situation, we suggested

that the government's performance arguably conflicted with the

assurance contained in the plea agreement. Concerned that the

government may have declined to file a section 5K1.1 motion because

Doe's help had come "too late" to permit it to "make a case"

against a third party, we found some tension between what the

government said it would do (i.e., not peg the substantial

assistance determination on whether Doe had made a case against

someone) and what it actually did. See id.

There is a critical difference between this case and Doe.

The Agreement sub judice does not tie the government's exercise of

its section 5K1.1 discretion to whether the defendant has (or has

not) made a case against a third party, and thus does not contain

the language that caused the contretemps in Doe. What is more, Doe

does not in any way intimate that the mere inclusion of a

cooperation clause in a plea agreement somehow circumscribes ex

proprio vigore the government's discretion anent the filing of such

motions. Indeed, Doe never argued (as Alegr¡a does) that

cooperation, if rendered, mandates the filing of a section 5K1.1

motion.

We have said enough on this score. The short of it is

that the concepts of "full, complete and truthful cooperation" and

"substantial assistance" are neither congruent nor interchangeable,

and the plain text of paragraph 11 refutes the appellant's

contention that the parties expressly modified this basic

understanding. Consequently, the government's election not to file

a section 5K1.1 motion did not violate the stated terms of the

Agreement.







B

The appellant has a fallback position. He invites us to

supplement the Agreement by engrafting onto it the oral

representation allegedly made by the United States Attorney during

the pre-plea negotiations. We decline the invitation.

The appellant's position flies in the teeth of paragraph

22 of the Agreement, which states flatly that the written document

constitutes the complete agreement between the parties and that the

"United States has made no promises or representations except as

set forth in writing in this plea agreement and deny [sic] the

existence of any other terms and conditions not stated herein."

Where, as here, an unambiguous plea agreement contains an

unqualified integration clause, it normally should be enforced

according to its tenor. That means, of course, that an inquiring

court should construe the written document within its four corners,

"unfestooned with covenants the parties did not see fit to

mention." Anderson, 921 F.2d at 338.

In United States v. Burns, 160 F.3d 82, 83 (1st Cir.

1998), we decisively rejected a similar attempt by a defendant to

read into a written plea agreement an implied constraint on the

government. Burns, in the course of appealing the district court's

enhancement of his sentence under a guideline provision, argued

that a clause in the plea agreement which restrained the government

from recommending such an increase "at sentencing" implied a duty

not to oppose Burns's effort to set aside the increase on appeal.

See id. at 83. We rejected this argument, explaining that the

government's promise simply did not go so far. See id. Moreover,

we admonished that "significant plea-agreement terms should be

stated explicitly and unambiguously." Id. Alegr¡a, in effect,

asks us to ignore the obvious wisdom of the Burns court's

admonition and to read into the Agreement a representation that

nowhere appears in the text. We are unwilling to freelance in this

fashion. See id. (warning that the defense, like the prosecution,

"must be alert to the need for clear and explicit articulation of

all pertinent terms in any plea agreement").

Two other considerations buttress the government's

position that the Agreement should be read as written. In the

first place, soft-pedaling the integration clause and slipping an

antecedent oral promise into the text would render other of the

Agreement's relevant passages, such as paragraph 11, entirely

nugatory. And this would contravene the rule that plea agreements,

like contracts generally, should be construed where possible to

give effect to every term and phrase. See Feinberg v. Insurance

Co. of N. Am., 260 F.2d 523, 527 (1st Cir. 1958) ("In construing a

contract, we must give reasonable effect to all terms whenever

possible.").

In the second place, inserting a new promise into the

Agreement would turn the change-of-plea colloquy into a farce. At

that time, the district court placed the appellant (who was

assisted by counsel and does not question their effectiveness)

under oath and carefully questioned him. See generally Fed. R.

Crim. P. 11(e). The appellant assured the court that the

prosecution had made no promises to him apart from those that were

written explicitly into the Agreement. Although Alegr¡a's

appellate counsel tries to slough this off as a legal fiction and

admonishes us that all defendants prevaricate during change-of-plea

colloquies, courts cannot operate on the assumption that parties

feel free to lie with impunity in response to a judge's

interrogation. We believe, therefore, that a defendant who asserts

a fact in answer to a judge's question during a change-of-plea

proceeding ought to be bound by that answer, absent exceptional

circumstances (say, for example, the emergence of newly discovered

evidence that places what was reasonably thought to be a fact in a

different light). See, e.g., United States v. Doyle, 981 F.2d 591,

594 (1st Cir. 1992); United States v. Butt, 731 F.2d 75, 80 (1st

Cir. 1984). We see no reason either to deviate today from this

salutary rule or to give the appellant the benefit of the long-odds

exception to it.

The appellant attempts in several ways to denigrate the

effect of the integration clause and the other circumstances we

have mentioned. Citing United States v. Rounsavall, 128 F.3d 665,

668-69 (8th Cir. 1997), he argues that oral representations

routinely are used to augment written plea agreements. That case

(in which the precise wording of the plea agreement is never

discussed) simply does not stand for the proposition that prior

oral representations may trump unambiguous language in a plea

agreement that expressly purports to be integrated. Moreover, the

same court elsewhere has indicated that it will rely on extrinsic

evidence only when, after considering a plea agreement as a whole,

the parties' intent remains ambiguous. See Kelly, 18 F.3d at 616.

The appellant also cites United States v. Leonard, 50

F.3d 1152 (2d Cir. 1995), for the same proposition. The case makes

no such holding. While the plea agreement there apparently

contained an integration clause, the only relevant issue before the

appellate court concerned a much different question: the good

faith vel non of the government's decision not to move for a

downward departure. See id. at 1157-58. So, too, for obvious

reasons, we find inapposite the appellant's citations to a line of

cases in which courts have deemed terms outlined in transmittal

letters accompanying plea agreements to be part and parcel of those

agreements. See, e.g., United States v. Garcia, 956 F.2d 41, 44

(4th Cir. 1992); United States v. Melton, 930 F.2d 1096, 1098-99

(5th Cir. 1991).

When all is said and done, the Agreement's integration

clause paragraph 22 withstands the appellant's bombardment.

Accordingly, we hold that it is not reasonable for Alegr¡a to seek

the benefit of a prior oral representation by the government after

he signed a fully integrated writing that did not contain the

claimed representation, and expressly affirmed to the district

court in the change-of-plea colloquy that he had not been

influenced by extrinsic representations of any kind.

C

Although the plain language of the Agreement provides no

succor and the effort to supplement it fails, the appellant tries

an end run. He posits that, in construing plea agreements, courts

should imply a duty of good faith in performance. Thus, even

though a plea agreement states unambiguously as this one does

that the government retains absolute discretion with respect to the

filing of a section 5K1.1 motion, the accused is entitled to expect

that the government will honestly evaluate the appropriateness of

seeking a downward departure. In his peroration, the appellant

asserts that the government thwarted this expectation and that the

district court erred by not holding an evidentiary hearing.

This argument is not new. United States v. Garcia, 698

F.2d 31 (1st Cir. 1983) not cited to us by either the appellant

or the government deals effectively with it. That case arose

before the sentencing guidelines (and, hence, section 5K1.1) went

into effect. It involved a written, fully integrated plea

agreement in which the government promised, in its discretion, to

make a lenient recommendation at sentencing if the defendant's

cooperation were complete and truthful. See id. at 35 n.3. We

concluded that allowing the government to retain absolute

discretion in these circumstances would "render a significant

element of the consideration for appellant's change of plea

illusory." Id. at 36. We lent substance to this element by

requiring the government "to show a good faith consideration of

[the defendant's] cooperation," that is, "to set forth in the

record sufficient reasons for its belief that [the defendant] has

not cooperated fully and that . . . a recommendation [of probation]

is not proper." Id. at 35 (quoting decision below).

To be sure, given the passage of time, the emergence of

the federal sentencing guidelines, and the Court's decision in Wade

v. United States, 504 U.S. 181 (1992), Garcia is arguably

distinguishable. But we think that its central concept that the

government must perform in good faith the discretionary obligations

that it affirmatively undertakes in a plea agreement remains good

law. Of course, this does not mean that courts can add material

conditions to plea agreements. See, e.g., Garcia, 698 F.2d at 36

(emphatically eschewing such a course). Nor does it mean that

every challenge to the government's good faith necessitates

protracted proceedings. The government's burden of showing good

faith is only a burden of production, not of persuasion. As long

as the government satisfies this modest burden, the trial court

need go no further unless the defendant makes a substantial

threshold showing that the government acted in bad faith. See

Kelly, 18 F.3d at 618; United States v. Khan, 920 F.2d 1100, 1106

(2d Cir. 1990); cf. United States v. Catalucci, 36 F.3d 151, 154

(1st Cir. 1994).

A myriad of practical, commonsense considerations

recommend this approach. We mention five of them. First, for a

court to inquire into the adequacy of a defendant's performance

under a plea agreement and assess the good faith of the

prosecutor's evaluation, it likely will need to delve into

sensitive matters a course that ineluctably will have a

disruptive effect on the prosecutorial function. Second, the

quantum of knowledge about ongoing investigations that is necessary

to make an informed decision often may be very high and the process

of acquiring that knowledge may be very time-consuming. Third, an

uncontrolled good faith exception will provide criminal defendants,

after the fact, with virtual carte blanche. Many of them, having

little to lose, will depict their performance glowingly, inviting

district courts to regard the prosecution's contrary statements as

pretextual (thus prompting further inquiry). We doubt that a

proliferation of such collateral litigation would square either

with the Supreme Court's decision in Wade, 504 U.S. at 185-87, or

with the orderly administration of the criminal justice system.

Fourth, recognizing a duty of good faith on the prosecutor's part

creates possibilities for opportunism in a manner that threatens to

countermand the goals of the criminal law and these possibilities

multiply as the ground rules for such challenges become more lax.

We made this point emphatically in Doe, when we explained that

"[d]efendants, asked for information to incriminate others, have

good reasons to fear for their safety and, unless the prosecutor

holds the whip hand, the defendant may offer up some information

and hold back the more vital balance in the hope that the court

will find the government 'unreasonable' and infer 'bad faith.'"

Doe, 170 F.3d at 225. Thus, diminishing the bite of the whip

through overzealous enforcement of the duty of good faith would be

counterproductive. Finally, the path that we have mapped out

comports with our oft-stated belief that, in criminal cases,

evidentiary hearings should be the exception, not the rule:

We have repeatedly stated that, even in the

criminal context, a defendant is not entitled

as of right to an evidentiary hearing on a

pretrial or posttrial motion. Thus, a party

seeking an evidentiary hearing must carry a

fairly heavy burden of demonstrating a need

for special treatment.



United States v. McGill, 11 F.3d 223, 225 (1st Cir. 1993)

(citations omitted); accord United States v. Isom, 85 F.3d 831, 838

(1st Cir. 1996).

In this case, the pertinent portion of the plea agreement

is the government promise to consider whether the appellant had

rendered substantial assistance, and, thus, merited a section 5K1.1

motion. See supra Parts II(A)-(B). This commitment carried with

it an obligation to evaluate the appellant's assistance in good

faith (although the "sole discretion" language in which the promise

was couched informed the nature of the obligation). The government

proffered facially adequate reasons for its conclusion that the

appellant had failed to achieve the substantial assistance

benchmark: the supplied information was "[on] occasions . . .

hearsay and on others . . . just too meager," and also included

"self-serving rationalizations" (a characterization that the

government punctuated with a telling example). This rejoinder

satisfied the government's burden of production. Taken at face

value, the appellant's counter-proffer showed that he attended two

debriefing sessions with FBI agents and that he was responsive and

truthful. He described in some detail information that he gave

regarding alleged kickbacks received by a certain bank officer, and

referred the government agents to a company that had been

transferring large sums of money between Puerto Rico and the

Dominican Republic under suspicious circumstances.

We do not believe that the district court erred in

deeming this proffer insufficient to warrant further proceedings.

Although the appellant professes to be sanguine about the value of

the information that he furnished, the record contains no

indication that any of it was useful to the government. By like

token, the appellant wholly fails to explain how this information

relates to any ongoing criminal investigation. The lower court

therefore lacked any kind of reliable framework within which it

could even begin to assess whether this "assistance" helped the

government to any degree, let alone whether it proved

"substantial."

In a last gasp, the appellant calumnizes the government's

failure to respond to information contained in a supplementary

letter that he transmitted. We need not linger over this

correspondence. The appellant again fails to show how that

information was any more useful than the material cited in his

declaration. In addition, we explained in Doe that the

government's failure to pursue such information, without more,

amounts at most to carelessness and does not suffice to make out a

case of bad faith. See Doe, 170 F.3d at 225-26.

To say more on this point would be supererogatory. We

review the district court's decision as to whether to go further,

that is, whether to convene an evidentiary hearing on the issue of

the government's good faith, for abuse of discretion. See David v.

United States, 134 F.3d 470, 477 (1st Cir. 1998). Stripped of

rhetorical flourishes, the appellant offers a wealth of conclusory

assertions, but no persuasive evidence of either substantial

assistance or bad faith. On this gossamer record, we cannot

conclude that the sentencing court abused its discretion in ruling

that the appellant failed to attain the requisite threshold. It

follows inexorably that the government did not breach the duty of

good faith in performance that due process imposes.

D

The appellant's final departure-related argument is that,

if the absence of a government motion places section 5K1.1 beyond

his reach, the district court, given his cooperation, nonetheless

should have departed downward under the general departure

guideline, USSG 5K2.0. See generally Koon v. United States, 518

U.S. 81, 94-95 (1996) (discussing the circumstances in which

departures under 5K2.0 are proper); United States v. Dethlefs, 123

F.3d 39, 44 (1st Cir. 1997) (same). We need not linger over this

importuning. The three courts of appeals that have addressed the

question since Koon agree that section 5K1.1 occupies the field and

that departures for substantial assistance, however labeled, are

available only under section 5K1.1. See In re Sealed Case, 181

F.3d 128, 140-42 (D.C. Cir. 1999); United States v. Solis, 169 F.3d

224, 227 (5th Cir. 1999), petition for cert. filed, ___ U.S.L.W.

___ (U.S. June 3, 1999) (No. 98-9623); United States v. Abuhouran,

161 F.3d 206, 213 (3d Cir. 1998), cert. denied, 119 S. Ct. 1479

(1999).

We had left the question open in a pre-Koon case. See

United States v. Romolo, 937 F.2d 20, 25 (1st Cir. 1991). We now

answer it, adopt the reasoning of our sister circuits, and hold

that a defendant's assistance to the prosecutor cannot serve as the

basis for a section 5K2.0 departure. By necessary implication,

then, the district court did not err in refusing to depart downward

based on Alegr¡a's cooperation.

III

We turn last to the calculations underpinning the

appellant's sentence. For economic crimes like bank fraud, amount

of loss is a critical component in formulating the guideline

sentencing range. See, e.g., United States v. Rostoff, 53 F.3d

398, 407-08 (1st Cir. 1995); United States v. Tardiff, 969 F.2d

1283, 1285 (1st Cir. 1992). The appellant claims that the district

court erred in calculating the pecuniary losses caused by his

conduct. Because this challenge cannot be divorced from the facts,

we sketch the contours of the offenses of conviction. As is the

custom in sentencing appeals, we draw our factual insights from the

change-of-plea colloquy, the presentence investigation report, and

the transcript of the disposition hearing. See, e.g., United

States v. Dietz, 950 F.2d 50, 51 (1st Cir. 1991). In this case,

moreover, we also have the benefit of our opinion in related

litigation. See Sheils Title Co. v. Commonwealth Land Title Ins.

Co., 184 F.3d 10 (1st Cir. 1999).

The appellant served as the president of Bankers Finance

Mortgage Corporation (BFMC). In the ordinary course of its

business, BFMC made residential mortgage loans. Many of these

loans were refinancings. In such a refinancing, BFMC's business

plan called for it to pay off the borrower's existing first

mortgage and make a new loan secured by a new first mortgage. It

then bundled groups of these loans and peddled the packages to

large financial institutions (most prominently, Citibank), in each

case representing that the purchaser would receive the functional

equivalent of a first mortgage, viz., an assignment of BFMC's first

mortgage. When a loan was sold, BFMC would send the purchaser an

assignment and a certificate of title insurance. Simultaneously,

BFMC would notify the borrower to make future payments directly to

the purchaser. The purchaser would then pay the agreed purchase

price to BFMC.

The major snag in this scenario was that, in certain

instances, BFMC never paid off the original mortgages. Thus, the

purchasers held second rather than first mortgages. Moreover, the

mortgagors went into default on the original mortgages,

notwithstanding their payments to the purchasers. BFMC's chicanery

came to light when some of the original lenders initiated

foreclosure actions. The purchasing institutions were in many

instances left holding an empty bag.

The indictment focused on BFMC's transactions with

Citibank, which lost approximately $3,100,000 as a result of the

scheme. Citibank managed to recoup some two-thirds of this amount

from BFMC and the appellant, reducing its net loss to roughly

$1,200,000. This amount was reimbursed by the title insurer. See

Sheils, 184 F.3d at 12-13 (describing scheme and recounting details

of title insurer's involvement).

Against this mise-en-scene, the district court concluded

that the essence of the criminal conduct more closely resembled

theft than simple fraud (in that the appellant took value from

Citibank without intending to give anything of value in return).

See United States v. Orton, 73 F.3d 331, 334 (11th Cir. 1996)

(explaining difference between theft and simple fraud); United

States v. Smith, 951 F.2d 1164, 1167 (10th Cir. 1991) (similar);

United States v. Kopp, 951 F.2d 521, 528-29 (3d Cir. 1991)

(similar); see also United States v. Flowers, 55 F.3d 218 (6th Cir.

1995) (declining to treat check-kiting cases like fraudulent loan

application cases); United States v. Frydenlund, 990 F.2d 822, 825-

26 (5th Cir. 1993) (similar); cf. United States v. Schneider, 930

F.2d 555, 558 (7th Cir. 1991) (distinguishing between fraud in

which the fraudfeasor intends to deprive the victim of the entire

value of an object and fraud in which the fraudfeasor returns

something of value to the victim). Having reached this conclusion,

the court ignored the title insurer's payments and fixed the amount

of loss attributable to the scheme at $1,200,000. See USSG

2F1.1(b). This, in turn, dictated the guideline sentencing range

and influenced the length of the prison sentence that the court

imposed.

We review sentencing determinations under a bifurcated

standard. Quintessentially legal questions, including

determinations as to the meaning and application of particular

guidelines, engender de novo review. See United States v. St. Cyr,

977 F.2d 698, 701 (1st Cir. 1992). The sentencing court's

factfinding, however, is reviewed deferentially and will be

disturbed only if it is shown to be clearly erroneous. See id. We

assume here, favorably to the appellant, that the de novo standard

of review obtains.

The appellant's principal objection to the district

court's loss calculation is that it failed to take into account the

fact that Citibank, because it enjoyed the benefit of title

insurance, never ran a risk of losing anything on the mortgage

transactions. In the appellant's view, the title insurance

functioned essentially as pledged assets, see USSG 2F1.1, comment.

(n.8(b)); the transactions between BFMC and Citibank thus were

equivalent to fraudulent loan transactions, see id.; and,

therefore, any amounts recovered by Citibank from the title insurer

must be offset in computing the amount of loss.

Although the parties debate longiloquently the question

of whether the appellant's conduct was tantamount to a series of

fraudulent loan transactions, we need not force our way into

Procrustean taxonomies to resolve the underlying dispute. Cf.

United States v. Riley, 143 F.3d 1289, 1291-92 (9th Cir. 1998)

(endorsing economic reality approach to sentencing). Even assuming

that this scheme is best characterized as involving fraudulent

loans, we find no error in the court's decision not to shrink the

amount of loss to reflect the receipt of title insurance proceeds.

Insurance, unlike pledged assets, does not diminish the impact of

the fraud. Rather, insurance simply shifts the loss to another

victim (the insurance company), so it is irrelevant in calculating

the amount of loss for sentencing purposes. See United States v.

Daniels, 148 F.3d 1260, 1262 (11th Cir. 1998) (per curiam).

We need go no further. There is no question but that

the appellant engaged in willful misconduct. The mere fact that

his victim was insured puts him in a worse, not a better, position

from the standpoint of the criminal law: he not only committed

fraud, but his knowledge that Citibank had title insurance

permitted him to gamble with other people's money. It would be

perverse to hold that criminals need not account for fraudulent

losses because they know that, regardless of their machinations,

their principal victim will be made whole by an insurance company.

The sentencing guidelines surely do not compel such a conclusion.



Affirmed.