Petition for Writ of Certiorari — Sanchez-Aran v. United States (Nos. 07-518, 07-528, 07-530)

Supreme Court brief2007

Ask Donna

What actually matters in this document.

Text

es OR ILE D

nt 07-5360CT 12 2007

IN THE

Supreme Court of the Anited States

Dr. FRANCISCO SANCHEZ-ARAN,

Petitioner,

Vv.

UNITED STATES,

Respondent.

On Petition for a Writ of Certiorari to the

United States

Court of Appeals for the First Circuit

PETITION FOR WRIT OF CERTIORARI

Thomas C. Goldstein* Mark J. Rochon

Patricia A. Millett MILLER & CHEVALIER

Duncan N. Stevens CHARTERED

AKIN GUMP STRAUSS 655 15th St. NW

HAUER & FELD LLP Suite 900

1333 New Hampshire Ave. Washington, DC 20005

Washington, DC 20036 (202) 626-5800

(202) 887-4000

*Counsel of Record

Attorneys for Petitioner

October 18, 2007

_*~__ OFFICE OF THE CLERK

WicsSon-EPES PRINTING Co., INC. — (202) 789-0096 -— WASHINGTON, D. C. 20002

QUESTION PRESENTED

Whether the court of appeals erred in holding, in

conflict with the decisions of the Second, Third,

Seventh and D.C. Circuits, that a failure to instruct

a jury on the enactment date of a criminal statute

was not error, when the evidence permitted the jury

to rest its verdict solely on pre-enactment conduct.

LIST OF PARTIES TO THE PROCEEDING

The names of all parties to the proceeding in the

United States Court of Appeals for the First Circuit

are:

Lorenzo Munoz-Franco

Dr. Francisco Sanchez-Aran

Ariel Gutierrez-Rodriguez

Wilfredo Umpierre-Hernandez

The United States

TABLE OF CONTENTS

QUESTION PRESENTED cecscscsccoscscescsovesoersenssessooess i

LIST OF PARTIES TO THE PROCEEDING....... ii

SAA nale GPW CAREC BEES BO rectecoececconsenvedencsocosvonsseenenee ill

TAB OF AU THORET IES cccvecosessovelecsescvassscovcensios v

FF RPI SEIU concesconsnscosersuvormnivcotoomisariniecoeiesnine 1

PUI BP intstctcinstntsnsmnnivwinsianeiniaminiwersansatienese 1

CONSTITUTIONAL AND STATUTORY

PV Be BEEF OIG LG cecscccctciascscceiscevenecescnceeesses 1

STATEMENT OF THE CASE .............ccccecccreceessrones 2

REASONS FOR GRANTING THE PETITION ....7

A. THE CIRCUITS HAVE SPLIT ON AN

IMPORTANT CONSTITUTIONAL ISSUE........ 8

1. Many Circuits Vacate Continuing-

Offense Convictions If It Is Reasonably

Possible That An Ex Post Facto Clause

Ve So cisiccatricccntecestetentcniatadsniccnesnees g

2. Four Circuits Have Drawn Inferences

In The Government’s Favor. ............cesecceceeee 15

B. UNDER THE SECOND, THIRD,

SEVENTH AND D.C. CIRCUITS’ STANDAR!?;,

PETITIONER’S CONVICTIONS MUST BE

Ve I ialiisciihinsdapusiindsinausininitipndiaaatabiaiaheisatinitminnieians 18

1. The First Circuit Did Not Follow The

Second, Third, Seventh and D.C. Circuit’s

PG hiseinsns cnnsissciinieniniiitieebeniaiiieemmminiieinnaniunsiiis 19

2. There Is A Reasonable Possibility That

The Jury Relied Exclusively On Pre-

Enactment Evidence in Convicting

I. siiictesscaatinviiciakesdenbenieaniianitnenncaseninioreses 24

C. THE FIRST CIRCUIT’S DECISION

CONFLICTS WITH THIS COURT’S

FT Geniccristaadeconsibiainbecnbitiamecasestonceosee 26

1. This Court Has Articulated a

Demanding Standard of Review for

Constitutional Errors. .ccescccosccesseccssssesssoveccosces 27

2. The First Circuit’s Ex Post Facto

Clause Analysis Is Inconsistent With This

Court’s Approach to Error Review. ............. 30

D. THIS CONSTITUTIONAL QUESTION IS

IMPORTANT AND RECURRING, AND IS

APPROPRIATELY RESOLVED IN THIS

Ils sisbnpentnigintuvnisocnhciovesexesacabieneaiiecidiinnnbeseniansboain 32

ATTY eiehinsesscvecnend beiinindininnsaiemnavianteninesiehoiias 34

Appendix A:

Opinion of the Court of Appeals for the First

BIE siisnstcicesvenincncnneveckensisssianniiibiiecinaibinaalaisiiiads la

Appendix B:

Opinic= and Order of the U.S. District Court

for the District of Puerto Rico ...............000« 92a

Appendix C:

Order of the Court of Appeals for the First

Circuit Denying Rehearing and Rehearing

BE HINES vic cnicceseacsniniscnsinecigiunainisaiiadinidbisesbaiiei 117a

TABLE OF AUTHORITIES

Federal Cases

Blakely v. Washington, 542 U.S. 296 (2004)............ 29

Boyde v. California, 494 U.S. 370 (1990) ................ 27

Calder vu. Bull, 3 U.S. SSC C1 TGS) .ccvccccccieseccccoscecsssees OO

Carmell v. Texas, 529 U.S. 513 (2000) ..................... 33

Chapman v. California, 386 U.S. 18(1966)...... passim

Clemons v. Mississippi, 494 U.S. 738 (1990) ........... 27

Collins v. Youngblood, 497 U.S. 37 (1990)................. 8

County Court v. Allen, 442 U.S. 140 (1979) . 27, 29, 31

Cramer v. United States, 325 U.S. 1 (194d) ............. 29

Cunningham v. California, 127 S. Ct. 856 (2007) ... 29

Dennis v. United States, 384 U.S. 855 (1966) .......... 16

Dobbert v. Florida, 432 U.S. 282 (1977)..............0.08. 33

Fahy v. Connecticut, 375 U.S. 85 (1968) ................4. 28

Francis v. Franklin, 471 U.S. 307 (1985)...........0..... 27

Marks v. United States, 430 U.S. 188 (1977) ........... 29

Miller v. Florida, 482 U.S. 423 (1987)...........0..... 8

Mills v. Maryland, 486 U.S. 367 (1988) .................4. 27

Olden v. Kentucky, 488 U.S. 227 (1988)...............006. 31

Rose v. Clark, 478 U.S. 570 (1986) ................00scceerees 27

Sandstrom v. Montana, 442 U.S. 510 (1979)..... 28, 30

Shepard v. United States, 544 U.S. 13 (2005).......... 29

Street v. New York, 394 U.S. 576 (1969) .......00..... 28

Stromberg v. California, 283 U.S. 359 (1931).......... 29

Sullivan v. Louisiana, 508 U.S. 275 (1993) ..... passim

United States v. Bagley, 473 U.S. 667 (1985)........... 29

United States v. Booker, 543 U.S. 220 (20085) .......... 29

United States v. Brown, 555 F.2d 407 (5th Cir. 1977).

ELEN SSIatS HPN RAEN TO! PRR nor Cee MR en Reer passim

United States v. Calabrese, 825 F.2d 1342 (9th Cir.

RR EM SCA alas eae arn CDS NCVER: bre RB DL PN passim

United States v. Cortez, 757 F.2d 1204 (11th Cir.

SE ESSN oe ea LE DR AON tne ate te ee ET passim

United States v. Duncan, 42 F.3d 97 (2d Cir. 1994)... 8

United States v. Harris, 79 F.3d 223 (2d Cir. 1996)....

ERG PAG MEER TOE NUR E NT LA BOT TERRE Nr NOT eNO 10

United States v. Henson, 848 F.2d 1374 (6th Cir.

STEARNS Nb Ons ae MAE a BY Bie bee Seems eRe PC Rr 16

United States v. Julian, 427 F.3d 471 (7th Cir. 2005),

cert. denied, 546 U.S. 1220 (2006).................... passim

United States v. Mitchell, 49 F.3d 769 (D.C. Cir.

PEE 5. iaicckiivtidcnacedaetinidibi tiene eisai oen 13, 18

United States v. Monaco, 194 F.3d 381 (2d Cir. 1999)

piikcbiieid vesiebiphautucniaopisiadind ioundelciidgttdeegtas dinuiaAeeameut bodes 10

United States v. Munoz-Franco, 487 F.3d 25 (1st Cir.

IN isis. ects vcidnechesiniighitdeietatiplchitnteaauanin vacate l

United States v. Olano, 507 U.S. 725 (1993).... passim

United States v. Scates, No. 86-5621, 818 F.2d 30

(table), 1987 WL 37328 (4th Cir. May 6, 1987)........ 17

United States v. Todd, 735 F.2d 146 (Sth Cir. 1984)...

inhale ibbaig dengtin dats cemented adieeneilgatisnia dada hicinls edeicamea aaahiaaiained dal passim

United States v. Torres, 901 F.2d 205 (2d Cir. 1990)

cilia elo psuenghacsu a haassessssd oat sihalobaiuaeidinpeasdbabamai tanto haat passim

United States v. Tykarsky, 446 F.3d 458 (3d Cir.

I ei hice issnch nnd Arbesdhsw cdlahicdadesesoedstibiasnedemeaioldibas passim

United States v. Williams-Davis, 90 F.3d 490 (D.C.

RN II ssinsbeds ccnwdeestacueshecnadicsocsuckcasabembass 13, 18, 20

United States v. Yates, 354 U.S. 298 (1957)....... 28, 29

Williams v. North Carolina, 317 U.S. 287 (1942).... 28

State Cases

Knowles v. State, 708 So. 2d 549 (Miss. 1998)... 13, 14

People v. Graham, 876 P.2d 68 (Colo. App. 1994) ... 14

People v. Kyle, 111 P.3d 491 (Colo. App. 1994)........ 14

People v. Luman, 994 P.2d 432 (Colo. App. 1999) ... 14

State v. Aho, 975 P.2d 512 (Wash. 1999).................. 14

State v. Hudspeth, 821 P.2d 547 (Wash. App. 1992) ...

Mintiaudiasrcrecenetonbateyahi dengan ueidbanemabadie 14

State v. Ricci, & A.2d 362 (N.J. Super. Ct. App.

Be SOURED va daisies, wdcidlaxvisa hice cu aictaacetea uaa ieee 14

Constitutional Provisions

es Rn OE: FED OE: SB iiiiticccciddimenninn 1

Statutes

ee ee O00, osc en ee 1

- Ax 8 ON Gs. SEER REREDyR Mine mrp eye Ropes pear nCcr mentee mean Te 1

Sas Oe BO, BOGS faisiiceisssscc ore 2

Se BR a iibiiccccevuistsvaiveensestee cnn passim

SE EE bisis ac cseuianstartiiabelsestieese teas tea 1

Pe eae ae he Utica acecsdissniseencd oe cee 5

Wee Te ca eee 5, 7

Bk 0 I savivsiisdsstanicienineaanaaes 9

Rules

ee Ue CA Oe eosin ae ee ae 18

OPINIONS BELOW

The opinion of the United States Court of

Appeals for the First Circuit is reported at United

States v. Munoz-Franco, 487 F.3d 25 (Pet. App. A).

The opinion of the United States District Court for

the District of Puerto Rico, denying Petitioner's

motion for new trial, is unreported (Pet. App. B).

JURISDICTION

The district court had jurisdiction over this

criminal case under 18 U.S.C. § 3231. The court of

appeals had jurisdiction to review the final judgment

of the district court pursuant to 28 U.S.C. § 1291.

The First Circuit decided this case on May 22, 2007.

A timely rehearing petition was denied by the First

Circuit on July 20, 2007 (Pet. App. C). The

jurisdiction of this Court is invoked under 28 U.S.C.

§ 1254(1).

CONSTITUTIONAL AND STATUTORY

PROVISIONS INVOLVED

1. The Ex Post Facto Clause of the Constitution,

U.S. Const. art. 1, § 9, cl. 3, states as follows:

No Bill of Attainder or ex post facto Law shall be

passed.

2. The federal bank fraud statute, 18 U.S.C.

§ 1344(a), states as follows:

Whoever knowingly executes, or attempts to

execute, a scheme or artifice--

(1) to defraud a federally chartered or

insured financial institution; or

(2) to obtain any of the moneys, funds,

credits, assets, securities or other property

owned by or under the custody or control of a

federally chartered or _ insured financial

institution by means of false or fraudulent

pretenses, representations, or promises, shall be

fined not more than $ 10,000, or imprisoned not

more than five years, or both.

STATEMENT OF THE CASE!

The First Circuit ruled that a district court’s

failure to instruct a jury on the enactment date of a

criminal statute, when the conduct in question began

well before that enactment date, does not violate the

Ex Post Facto Clause as long as the majority of the

evidence presented at trial related to conduct that

occurred after the statute was enacted. That holding

deepens a split among the circuits on the proper

standard for addressing Ex Post Facto Clause

instructional errors, and improperly resolves against

criminal defendants the ambiguities created by

constitutional errors at trial. The rule adopted by

the Second, Third, Seventh, and D.C. Circuits, and

the rule that is consistent with this Court’s

precedents, is to vacate a conviction if there is a

reasonable possibility that the jury relied on conduct

predating the criminalizing statute. This Court

should grant review to resolve the conflict among the

circuits on this important question of constitutional

law.

1. The bank fraud statute, 18 U.S.C. § 1344,

became law on October 12, 1984. Prior to that, there

was no federal criminal prohibition on bank fraud

itself. Federal law, however, did prohibit mail and

wire fraud. 18 U.S.C. §§ 1341, 1343.

The indictment alleged two bank fraud schemes,

one lasting from June 1980 to May 1990, and the

' Petitioner hereby adopts and joins the petitions of the other

defendants in this matter.

other lasting from December 1981 to May 1990. The

indictment did not allege mail fraud, wire fraud, or

any other general fraud offense that was on the

books prior to October 12, 1984. Section 1344 was

the only fraud provision cited.

2. Throughout the 1980s, Lorenzo Munoz-

Franco and Petitioner Dr. Francisco Sanchez-Aran

were the president and executive vice president,

respectively, of Caguas Central Federal Savings &

Loan, a Puerto Rico thrift.2 Ariel Gutierrez and

Wilfredo Umpierre were officers of Transglobe

Corporation and Modules Manufacturing, Inc.,

companies that received substantial construction

loans from Caguas in the early 1980s and thereafter

acted as contractors for other borrowers. Caguas

also executed construction loans to companies owned

by Francisco Mirandes throughout the 1980s.

One of Caguas’s first loans during the indictment

period to companies led or controlled by Gutierrez

and Umpierre was executed on June 25, 1980 for the

La Marina project.? Another project, Levittown

Plaza, was the subject of a Caguas loan executed on

January 30, 1981, and another loan for the Quintas

de Country Club project was executed on March 31,

1982.4. On these loans, the construction schedule

was not met, and interest payments were taken from

loan proceeds to maintain the repayment schedule.®

In addition, monies were transferred to and from

other projects, and the transferred amounts were

2 JA-T-2387. References are to the record filed with the First

Circuit.

3 JA-T-0079-80.

4 JA-T-0143-44, -0174.

5 JA-T-0085-86, -0089, -0241-42.

used for loan repayments.® The loans were repaid on

October 29, 1984, by the DOW Group (not owned or

controlled by Gutierrez or Umpierre), which obtained

new loans from Caguas and closed out the Gutierrez

loans.? After that date, Gutierrez and Umpierre

(and affiliated companies) had no construction loans

with Caguas; they served as contractors on projects

financed through other borrowers through 1985.

Effective December 31, 1985, Gutierrez sold Modules

to Sergio Camero but remained as an employee of

Modules; he and Umpierre left Modules in December

1986.8

Beginning in the early 1980s, Caguas also

executed loans to companies held by Francisco

Mirandes. For example, loans to Deproco

Corporation, a Mirandes-held company, were

executed for the Reparto Valenciano project on

December 23, 1981, and for the Villas de Gurabo I

project on May 15, 1984.9 Both of those projects fell

behind schedule, and multiple transfers were made

to and from those loans and other Mirandes-held

projects. !°

At various points in 1982 and 1983, Caguas

executed two-party checks payable to Transglobe

Corporation and to various third parties (contractors

and government agencies) in connection with the La

Marina and Quintas de Country Club projects. The

checks in question were ultimately deposited in

Transglobe’s accounts, endorsed by the third-party

payees. It was later alleged that those endorsements

6 See, e.g., JA-T-0093-94, -0162-63, -0178-79.

7 JA-T-0096-99, -0159-60, -0243-44.

§ JA-T-1692-93, -2125-27, -3386-87.

¥ JA-T-0508-10, -0529.

0 JA-T-0517-18, -0533-35.

had been forged on a few checks.'! Also in 1983, a

Caguas employee determined that a Gutierrez

company had written checks from a Caguas account

for deposit in another bank, and had also written

checks from the other bank for deposit at Caguas—a

"19

practice he characterized as “check kiting.

Following the passage of the Financial

Institutions Reform, Recovery and Enforcement Act

of 1989, Caguas fell out of capital compliance and

was seized in May 1990.'3

3. On November 22, 1995, Petitioner Sanchez-

Aran was indicted in the U.S. District Court for the

District of Puerto Rico (along with Munoz-Franco,

Gutierrez, and Umpierre) on charges of bank fraud

under 18 U.S.C. § 1344, conspiracy under 18 U.S.C.

§ 371 to commit, inter alia, bank fraud, and

misapplication of bank funds under 18 U.S.C. § 657.

Three superseding indictments were returned in

1997 and 1998. The final indictment charged two

separate conspiracies and schemes (one involving

Gutierrez and Umpierre and one involving Francisco

Mirandes, who pleaded guilty before trial) and

alleged acts extending from June 25, 1980 through

May 25, 1990. 22 of the 104 overt acts alleged in the

final indictment occurred prior to October 12, 1984.

At trial, the prosecution offered ample testimony

on the pre-October 1984 conduct described above.

The first several weeks of trial, for instance, were

devoted entirely to the La Marina, Levittown Plaza,

and Quintas de Country Club projects; there was no

testimony about any other projects until more than

1) See, e.g., SA-T-1549-52, -1584-86.

12 JA-T-1774-75.

13 JA-T-2784-87.

two weeks after trial began. Several other witnesses

discussed those projects at length later in the trial.

Much of the one-day testimony of Mirandes

concerned activity on Reparto Valenciano between

1981 and 1984. A Caguas employee testified for

several days about the alleged check-kiting in 1983.

Ten witnesses, in over two weeks of testimony,

testified that their signatures endorsing Caguas

checks over to Gutierrez-related corporations had

been forged, and virtually all of those alleged

forgeries predated 1984.

The jury returned guilty verdicts on all counts

for four of the five defendants. It had not been

instructed on the enactment date of the bank fraud

statute. Petitioner retained new counsel for

sentencing and appeal. His new counsel discovered

that the bank fraud statute had not been enacted

until October 12, 1984, and sought a new trial,

arguing that Petitioner’s convictions violated the Ex

Post Facto Clause. The district court denied the

motion, holding that Petitioner had waived the issue

and that there was no error because post-enactment

evidence was presented. Pet. App. 10la-104a. The

court sentenced Petitioner to 46 #£=months’

imprisonment on conspiracy and bank fraud and a

concurrent 60 months on misapplication. Pet. App.

10a.

4. The First Circuit affirmed. It noted that the

government had conceded most of the “plain error”

analysis, and acknowledged that the courts of

appeals were divided on the Ex Post Facto Clause

issue. Pet. App. 58a-61a. Under the approach taken

by the Second, Third and Seventh Circuits, a

conviction cannot stand if it is reasonably possible

that the jury relied exclusively on pre-enactment

conduct. Pet. App. 60a-6la. By contrast, under the

approach followed by the Ninth Circuit and one Fifth

Circuit panel, a conviction may stand as long as it is

supported by substantial post-enactment evidence.

Pet. App. 58a-59a. The court ostensibly took no

position on this split, opining that Petitioner could

not prevail even under the standard most favorable

to defendants (that of the Second, Third and Seventh

Circuits). Pet. App. 61a. The court then held,

echoing the Ninth Circuit, that the error did not

affect Petitioner’s “substantial rights” because there

was “considerable evidence” of post-enactment

events and because the “majority” of the overt acts in

the indictment were post-enactment. Pet. App. 62a.

The court also found no “transformative event”

distinguishing pre-enactment from post-enactment

conduct—analysis not followed by any other circuit—

and held on that basis that the jury could not have

relied solely on pre-enactment acts. Pet. App. 62a-

63a.

The defendants sought rehearing, but the First

Circuit denied the petitions on July 20, 2007. See

Pet. App. 118a. Petitioner sought a stay of the

mandate pending a certiorari petition, which was

granted.

REASONS FOR GRANTING THE PETITION

This Court should grant review to resolve a

broad and entrenched conflict among the courts of

appeals. As the First Circuit acknowledged (see Pet.

App. 58a-61a), there is substantial disagreement

among the courts of appeals on the standard for

reviewing a district court’s failure to instruct a jury

on the enactment date of a criminal statute, as

required by the Ex Post Facto Clause. The Second,

Third, Seventh and D.C. Circuits, and the Brown

panel of the Fifth Circuit, require reversal if it is

reasonably possible that the jury relied solely on pre-

enactmient evidence and thereby violated the

defendant’s Ex Post Facto Clause rights. By

contrast, the Fourth, Sixth, Ninth and Eleventh

Circuits, and the Todd panel in the Fifth Circuit,

would uphold a conviction if there is substantial

post-enactment evidence, even if the jury was not

instructed on the enactment date. Had this case

arisen in the Second, Third, Seventh, or D.C.

Circuits, the court would have vacated Petitioner’s

convictions because the record establishes a

reasonable possibility that the jury relied exclusively

on pre-enactment evidence. Tracking the Ninth

Circuit's approach, however, the court of appeals

here affirmed the convictions. That decision conflicts

with this Court’s precedents disfavoring judicial

factfinding to uphold a conviction ar.d mandating

that ambiguities in jury verdicts be resolved against

the government. The confusion and uncertainty

arising from this split, together with the importance

of this constitutional question, warrant a grant of

certiorari.

A. THE CIRCUITS HAVE SPLIT ON AN

IMPORTANT CONSTITUTIONAL ISSUE.

The Ex Post Facto Clause of the Constitution

prohibits Congress from criminalizing conduct after

it has occurred, Collins v. Youngblood, 497 U.S. 37,

42 (1990), or from increasing the punishment for a

crime after it is committed. Miller v. Florida, 482

U.S. 423, 429 (1987). The lower courts agree that

the Ex Post Facto Clause is not necessarily violated

by the prosecution of an offense that begins before

but continues after the enactment date of the

statute. See, e.g., United States v. Duncan, 42 F.3d

97, 104 (2d Cir. 1994). In such cases, however, the

jury must rely on post-enactment evidence to find all

elements of the crime; exclusive reliance on pre-

enactment evidence to find any element violates the

Ex Post Facto Clause.

Ordinarily, juries are instructed on the relevant

enactment date, and decide whether the offense is

fully proven post-enactment. When, as in this case,

the jury is not so instructed, courts must determine

whether the error could have resulted in a verdict

that violates the Ex Post Facto Clause. A

widespread conflict has emerged among the courts of

appeals regarding the proper’ standard for

determining whether a jury verdict runs afoul of the

Ex Post Facto Clause’s protections.

1. Many Circuits Vacate Continuing-

Offense Convictions If It Is Reasonably

Possible That An Ex Post Facto Clause

Violation Occurred.

Four circuits have adopted a standard for

reviewing Ex Post Facto Clause instructional errors

that asks whether it was reasonably possible that

the jury relied exclusively on that evidence and

thereby violated the Ex Post Facto Clause.

The Second Circuit’s decision in United States v.

Torres, 901 F.2d 205, 224 (2d Cir. 1990), dealt with

this question in the context of the “continuing

criminal enterprise” sentencing enhancement

enacted on October 27, 1986. That provision

requires life imprisonment for a _ “principal

administrator, organizer, or leader” of a criminal

enterprise that obtains gross receipts exceeding $10

million in a_ twelve-month period. 18 U.S.C.

§ 848(b)(1). In Torres, as here, the district court did

not instruct the jury on the enactment date. Rather,

it told the jury that it could convict the defendants if

they acted as administrators between June 24, 1986,

and June 23, 1987, and if the enterprise grossed $10

million in that period.'* In reviewing for plain error,

the Second Circuit acknowledged that it was “quite

unlikely that the jury would have found a significant

difference in the character of these appellants’

relationship to, and leadership of, the [enterprise]

before and after October 27, 1986.” 901 F.2d at 228.

Indeed, the only evidence of the enterprise’s

violations was post-enactment; all of the acts

presented at trial occurred in 1987. Id. at 228-29.

Nevertheless, the court held, as long as it was

possible that the jury relied on pre-enactment

evidence for the “administrator” finding, reversal

was required no matter how unlikely such a finding

was, or how likely it was that the jury, suitably

instructed, would have found that the defendants

acted as “administrators” after October 27, 1986. Id.

at 229. Accordingly, the court vacated the

conviction. Id.; cf. United States v. Harris, 79 F.3d

223 (2d Cir. 1996) (where statute required that jury

find “series” of violations and only one of nineteen

acts predated the relevant enactment date, “no

possibility” of exclusive reliance on pre-enactment

conduct); United States v. Monaco, 194 F.3d 381, 386

(2d Cir. 1999) (“A conviction for a continuing offense

straddling enactment of a statute will not run afoul

of the Ex Post Facto clause unless it was possible for

14 The First Circuit stated that Torres was distinguishable

because the defendants there preserved their objection. In fact,

the Torres court noted that the defendants’ objection below

“{did] not suffice to preserve an objection for appeal, absent

plain error.” Pet. App. 60a n.34.

the jury, following the court's instructions, to convict

‘exclusively’ on pre-enactment conduct”; no violation

because jury was properly instructed).

The Third Circuit has followed the Second

Circuit’s approach. In United States v. Tykarsky, 446

F.3d 458 (3d Cir. 2006), the defendant was convicted

under a solicitation statute, the minimum sentence

for which was increased on April 30, 2003; of the

nine acts presented at trial, seven occurred after that

date. The court held that, if “a reasonable jury could

have concluded only that the attempted persuasion

or enticement continued past April 30,” then the

conviction could stand. /d. at 480-81. Because it was

possible that the jury did not find the post-April 30

conversations to constitute “enticement = or

persuasion,” there was “at least a possibility that the

jury convicted Tykarsky based solely on _ pre-

[enactment] conduct. ... The most that can be said

here is that it is improbable, rather than impossible,

as a factual matter, that the jury convicted Tykarsky

exclusively on the basis of pre-April 30 acts.” Id. at

482 (emphasis in original). Hence, even though the

majority of the conduct was post-enactment, the

court found plain error and vacated Tykarsky’s

sentence. The court also rejected the government’s

argument that that “judicial fact-finding” was

permissible, holding that the jury, not the judge, was

responsible for finding a _ violation after the

enactment date. /d. at 480 n.18.

The Seventh Circuit has applied the same

analysis. In United States v. Julian, 427 F.3d 471

(7'h Cir. 2005), cert. denied, 546 U.S. 1220 (2006), the

court held that the failure to instruct the jury as toa

statutory enactment date “implicated Julian’s ex

post facto rights.” 427 F.3d at 482. Julian was

—

convicted of conspiracy under a provision whose

maximum penalty was increased on October 30,

1998. The jury was never instructed to consider

whether the conspiracy extended past that date, and

the court found that “[i]f a jury, properly instructed

on this point, might have found that the conspiracy

had come to an end before the increased penalty took

effect or that Julian had withdrawn from the

conspiracy before that date,” vacatur was required.

Id. at 482. If “a reasonable jury could only have

concluded that the conspiracy continued beyond the

effective date of the new statute and that Julian

remained a member of the conspiracy beyond that

date,” however, no violation occurred. Id. at 482-83.

There was no dispute that the conspiracy had

continued past October 30, and nothing in the record

indicated an affirmative act by Julian to withdraw

from the conspiracy prior to that date. Jd. at 483.

Hence, “no reasonable jury” could have found

withdrawal, and there was no plain error.!5

'5 The First Circuit attempted to distinguish the Second and

Third Circuits’ decisions from the Seventh Circuit’s holding, see

Pet. App. 61a-62a, but in substance those circuits all conducted

the same analysis The Seventh Circuit looked to the fourth

Olano factor, asking whether permitting the error to stand

would “implicate[] the fairness, integrity, or public reputation

of the judicial process,” whereas the other circuits focused on

the third factor (harm to the defendant's “substantial rights”).

Julian, 427 F.3d at 482. But all three courts asked the same

question: whether the jury could have relied solely on pre-

enactment acts in convicting. Compare Julian, 427 F.3d at 482

(“If a jury, properly instructed on this point, might have found

that the conspiracy [came] to an end before the increased

penalty took effect .. . then the error is one that implicates the

fairness, integrity, or public reputation of the judicial process.”)

with Tykarsky, 446 F.3d at 480 (“To affec{t) substantial rights.

the error must have been prejudicial. Tykarsky has been

The D.C. Circuit, in United States v. Mitchell, 49

F.3d 769 (1995), has also looked to the sufficiency of

pre-enactment evidence. At issue in Mitchell was

the same sentencing enhancement considered in

Torres; because all of the violations submitted to the

jury were committed after the relevant enactment

date, the court had “no doubt” that the jury had

found him to have led the criminal enterprise based

on post-enactment acts, and hence found no plain

error. Jd. at 781; see also United States v. Williams-

Davis, 90 F.3d 490, 511 (D.C. Cir. 1996) (discussing,

in preserved-error context, whether’ exclusive

reliance on pre-enactment evidence was “possible”;

no violation because conviction on pre-enactment

evidence required “bizarrely configured jury finding”

the likelihood of which was “zero’”).

One Fifth Circuit decision has adopted the same

legal standard for reviewing Ex Post Facto Clause

appeals. In United States v. Brown, 555 F.2d 407

(5th Cir. 1977), the court addressed a conspiracy that

lasted from 1966 to 1974 and a statute enacted in

October 1970. Even though the majority of the acts

in furtherance of the conspiracy had been committed

after the enactment date, see id. at 412 n.4, the court

found plain error and vacated the convictions. The

court held that assuming that the jury did not rely

on pre-enactment conduct would amount to

impermissibly directing a verdict for the prosecution.

Id. at 421.

Finally, two state supreme courts have also

adopted this approach. In Knowles v. State, 708 So.

prejudiced if there is a reasonable possibility that a jury,

properly instructed on this point, might have found Tykarsky

guilty based exclusively on acts that occurred before the

increased penalty took effect.”) (citation omitted).

2d 549 (Miss. 1998), the statute was amended on

July 1, 1995 to delete a required element of the

crime. The defendant was convicted of an offense

that ran from May 1993 to December 1995. On

appeal, he argued that the failure to instruct on the

enactment date violated the Ex Post Facto Clause

(an objection apparently preserved below), and the

Mississippi Supreme Court vacated the conviction.

The court concurred with other state courts’ holdings

that a failure to instruct the jury on the enactment

date of a continuing offense violates the Clause, even

if “some of the unlawful conduct occurred after the

new law became effective.” Jd. at 554-56. In State v.

Aho, 975 P.2d 512 (Wash. 1999), the relevant statute

was enacted on July 1, 1988, and the defendant was

convicted for conduct running from January 1987 to

December 1982 for one offense and to August 1995

for another. The Washington Supreme Court

vacated the conviction (despite the defendant’s

failure to preserve the issue), holding that the failure

to instruct the jury on the significance of the July 1,

1988 enactment date made it “possible that Aho has

been illegally convicted based upon an act or acts

occurring before the effective date.” Id. at 516.'6

16 See also People v. Kyle, 111 P.3d 491, 506 (Colo. App. 1994)

(sentencing enhancement improperly applied when jury could

have relied on pre-enactment conduct); People v. Luman, 994

P.2d 432, 437 (Colo. App. 1999) (same); People v. Graham, 876

P.2d 68, 72 (Colo. App. 1994) (same); State v. Ricci, 593 A.2d

362, 364-65 (N.J. Super. Ct. App. Div. 1991) Gury should have

been instructed that July 1, 1987 was effective date of “kingpin”

statute enhancing penalties, since defendant was charged with

crime lasting from August 1986 through August 1987; “we can

not be sure” that jury would have found that defendant led

conspiracy after July 1, and error was plain); State v. Hudspeth,

821 P.2d 547, 547 (Wash. App. 1992) (“[BJecause the jury was

Accordingly, in the Second, Third, Seventh and

D.C. Circuits, and in Mississippi and Washington,

courts have set a clear standard: a conviction must

be vacated if the elements of the offense were

established by pre-enactment evidence, such that the

jury could have returned a conviction based solely on

that evidence. In those jurisdictions, courts do not

require further proof that such reliance was the more

likely outcome, or that post-enactment acts were

qualitatively different from pre-enactment conduct.

Only when it is not reasonably possible for a jury to

have rendered its decision based exclusively on pre-

enactment evidence will those courts uphold a

verdict. Accordingly, had Petitioner’s case arisen in

any of these jurisdictions, his conviction would have

been vacated on appeal.

2. Four Circuits Have Drawn Inferences In

The Government’s Favor.

The Fourth, Sixth, Ninth and Eleventh Circuits

have taken the opposite tack. Those courts look to

the nature and volume of post-enactment events and

uphold convictions whenever that evidence was

sufficient to support a jury verdict, regardless of

whether the verdict could reasonably have rested

exclusively on pre-enactment evidence.

In United States v. Calabrese, 825 F.2d 1342

(1987), the Ninth Circuit upheld a conviction based

on evidence post-dating the relevant enactment date.

The defendant had been convicted of conspiracies

that began in October 1982 and November 1983;

not instructed to indicate when the offenses occurred, we

cannot reject the possibility that Hudspeth was convicted of

violating a statute that was not in effect at the time the

offenses occurred”; vacating though objection not preserved).

both ended in November 1984. /d. at 1346. The

relevant statute was enacted on October 12, 1984,

and the Ninth Circuit held that there was

“substantial evidence” to support a jury finding that

the conspiracy extended beyond that date, without

discussing whether it was possible that the jury did

not so find. Jd. (One defendant had raised the issue

at sentencing, but not at trial. Jd.)

The Sixth Circuit adopted a similar test in

United States v. Henson, 848 F.2d 1374 (1988),

where it concluded that, although all of the overt

acts set forth in the indictment occurred before a fine

enhancement provision went into effect on January

1, 1985, the provision was nonetheless applicable

because “there was evidence of other acts that were

performed in furtherance of the conspiracy after

December 31, 1984.” Id. at 1386. Like the Ninth

Circuit, the Sixth Circuit in Henson offered no

reason to believe that the jury must have relied—or,

even, was more likely to have relied—on the evidence

of post-enactment conduct. The existence of that

evidence in the record was sufficient.!?

The Eleventh Circuit has embraced the same

test. In United States v. Cortez, 757 F.2d 1204 (11th

Cir. 1985), the court considered a conspiracy that

‘7 Both the Sixth Circuit in Henson and the Ninth Circuit in

Calabrese also suggested that the problem can be assumed

away, reasoning that a guilty verdict on a continuing offense

necessarily means a conclusion that the offense lasted for the

full duration stated in the indictment. 848 F.2d at 1385; 825

F.2d at 1346. No other circuits appear to have endorsed that

view, and it is inconsistent with this Court’s holdings that the

government need not prove all of the overt acts alleged in an

indictment to obtain a conspiracy conviction. See, e.g., Dennis

uv. United States, 384 U.S. 855, 863 n.8 (1966).

began in the fall of 1979 and extended through

November 1980 and a statute that was enacted on

September 15, 1980. That court held that “[e]nough

activities of the conspirators continued after § 955a’s

enactment so that even if the jury had been

instructed by the trial court as to the date of

enactment and to consider only activities thereafter,

there is no doubt that the jury would have decided

the case the same way.” Id. at 1207.

The Fourth Circuit followed this approach in

United States v. Scates, No. 86-5621, 818 F.2d 30

(table), 1987 WL 37328 (May 6, 1987), holding that,

where one out of three acts occurred after the

effective date of a sentencing enhancement, there

was no Ex Post Facto Clause violation even though

the jury was not instructed on that enactment date.

The court noted Scates’s argument (not preserved

below) that “the jury may have convicted him solely

on the basis of events occurring prior to the effective

date of the act,” 1987 WL 37328 at *3, but affirmed

his conviction without addressing that possibility.

Finally, one panel of the Fifth Circuit adopted

this standard in United States v. Todd, 735 F.2d 146

(1984). The defendants there were convicted of a

conspiracy that began before the effective date of a

sentencing enhancement. Id. at 149. The jury was

not instructed on the significance of the date, but the

court found no plain error as to the Ex Post Facto

Clause violation because “[mlost of the evidence

focused on events” occurring after that statutory

date. Id. at 150. Notably, the same was true of the

evidentiary record for the Fifth Circuit’s earlier

decision in Brown, supra. Yet the panel in Brown

found plain error. The Fifth Circuit thus has a

longstanding and unresolved internal conflict on this

question.

Unlike the Second, Third, Seventh and D.C.

Circuits, therefore, the Fourth, Sixth, Ninth and

Eleventh Circuits do not consider whether the jury

verdict could reasonably have rested on _ pre-

enactment evidence. In the latter courts, the

existence and volume of post-enactment evidence is

controlling.

This conflict is entrenched. The First and D.C.

Circuits have denied requests for rehearing en banc,

and the split has persisted for more than twenty

years. All of the circuits with criminal dockets other

than the Eighth and Tenth Circuits have now

addressed the issue. This Court’s review is therefore

needed to resolve the dispute.

B. UNDER THE SECOND, THIRD, SEVENTH

AND D.C. CIRCUITS’ STANDARD,

PETITIONER’S CONVICTIONS MUST BE

VACATED.

The First Circuit acknowledged the split of

authority among the circuits, but concluded that

Petitioner would fail under either standard. In so

holding, the court misapprehended the test followed

by the Second, Third, Seventh and D.C. Circuits and

its effect in this case. In those circuits, if there is a

reasonable possibility that the jury relied solely on

the pre-enactment acts alone, the conviction must be

vacated. That standard was met here.!8

'8 Trial counsel in this case did not request an instruction on

the enactment date, and the First Circuit accordingly reviewed

for plain error. Virtually every significant decision on this

issue has been in that context, as set forth above

(unsurprisingly, as it would be unusual for a district court to

1. The First Circuit Did Not Follow The

Second, Third, Seventh and _ D.C.

Circuit’s Analysis.

The First Circuit’s analysis relied on the relative

quantities of pre- and post-enactment evidence,

mirroring the analysis of the Fourth, Sixth, Ninth,

and Eleventh Circuits. The differences between the

First Circuit’s approach and that of the Second,

Third, Seventh, and D.C. Circuits are clear and

sharp.

In Torres and Tykarsky, the Second and Third

Circuits unequivocally held that a failure to instruct

refuse to instruct the jury on a statutory enactment date). The

government conceded three of the four elements of plain error

on appeal: error occurred, it was “clear or obvious,” and letting

it go uncorrected would “seriously impair the fairness,

integrity, or public reputation of judicial proceedings.” See

United States v. Olano, 507 U.S. 725, 732-36 (1993). The only

question in dispute here was whether the error influenced the

verdict and thereby affected Petitioner’s “substantial rights.”

See id. at 734. The other plain error decisions on this issue

have likewise turned on whether the failure to instruct affected

the verdict. Accordingly, this Court need only resolve the

dispute over when an uninstructed jury’s verdict violates

“substantial rights” to harmonize the law in this area.

Even in the rare case where an instruction on the

enactment date is requested but not given, and the review is

thus for harmless error rather than plain error, the required

analysis is substantively the same. ‘The “substantial rights”

inquiry under Olano is identical to the harmless-error inquiry

prescribed by Federal Rule of Criminal Procedure 52 for

preserved errors. Compare Mitchell, 49 F.3d at 781 (in plain-

error inquiry, discussing whether jury “could have” convicted

based solely on pre-enactment acts) with Williams-Davis, 90

F.3d at 511 (in preserved-error setting, discussing whether

exclusive reliance on pre-enactment conduct was “possible” and

following Mitchell’s analysis). The only difference is that,

under plain error, the defendant bears the burden of

persuasion. Olano, 507 U.S. at 734.

the jury on the Ex Post Facto Clause cannot be held

harmless under plain-error analysis merely on the

ground that most of the evidence post-dated the

statutory enactment date. Yet that was precisely the

reasoning offered by the First Circuit. See Pet. App.

62a. In Torres, for instance, the question for

sentencing purposes was whether the defendant

acted as a leader or organizer of a criminal

enterprise after the relevant date. All of the

enterprises acts in question. occurred post-

enactment, and the Second Circuit found it “quite

unlikely” that the defendant assumed such a role

and then abandoned it before the enterprise

committed any criminal acts. 901 F.2d at 228.

Nevertheless, because it was “possible” that the

jury’s verdict rested on such a finding, the sentence

was vacated on appeal. Id.

Similarly, in Tykarsky, seven of the nine acts

presented at trial post-dated the statute, yet the

Third Circuit found a “reasonable possibility” that

the jury relied exclusively on one of the two pre-

enactment acts and vacated the conviction. 446 F.3d

at 482. Like the Second Circuit, the Third Circuit

found the Ex Post Facto Clause error to be plain even

though most of the evidence supported a conviction

on proper grounds: “The most that can be said here

is that it is improbable, rather than impossible, as a

factual matter, that the jury convicted Tykarsky

exclusively on the basis of pre-April 30 acts.” Id.

(emphasis in original). See also Julian, 427 F.3d at

482 (Seventh Circuit holding that, “fiJf a jury,

properly instructed on this point, might have found

that the conspiracy had come to an end before the

increased penalty took effect or that Julian had

withdrawn from the conspiracy before that date,” the

ond

conviction must be vacated) (emphasis added);

Williams-Davis, 90 F.3d at 511 (D.C. Circuit,

addressing whether it is “possible” that verdict

rested on pre-enactment conduct); Brown, 555 F.2d

at 412 n.4, 421 (Fifth Circuit, vacating conviction

even though majority of overt acts occurred after the

enactment date).

The First Circuit’s analysis cannot be reconciled

with the holdings of those Circuits. The court was

only able to uphold the conviction by holding that the

“majority” of the evidence presented at trial and the

“bulk” of the allegations considered in its sufficiency

analysis were post-enactment, and that there was

“considerable” evidence of post-enactment conduct.

Pet. App. 62a. The court further noted that

“relatively few” (22 out of 104) of the acts set forth in

the indictment occurred before October 12, 1984. /d.

On that basis, the court concluded that the jury did

not rely exclusively on pre-enactment evidence. Id.

The contrast between this case and Tykarsky, in

which the Third Circuit vacated a sentence even

though seven of the nine acts were post-enactment,

is stark. Although the ratio of pre-enactment to

post-enactment acts was virtually identical in these

two cases, the courts reached opposite results. See

446 F.3d at 482. The Second Circuit’s holding in

Torres is even more inconsistent with the First

Circuit’s approach, since all of the evidence in Torres

post-dated the enactment. 901 F.2d at 229.

Accordingly, that the majority of the evidence in this

case related to post-enactment conduct was not by

itself sufficient, under those circuits’ standard, to

support a finding that the jury’s exclusive reliance on

pre-enactment conduct was “impossible.” See, e.g.,

446 F.3d at 482.

- The First Circuit also noted that it saw “nothing

to differentiate appellants’ pre-enactment conduct

from subsequent conduct,” because certain key

government witnesses testified about both pre- and

post-1984 transactions and there was. no

“transformative event” prior to October 12, 1984.19

Pet. App. 62a. Again, whatever the merits of this

view, it cannot be reconciled with the approach

adopted by the Second, Third, Seventh, and D.C.

Circuits, who do not require a “transformative event”

in their plain-error analysis.2° Indeed, in Torres, the

Second Circuit specifically noted the absence of such

an event, finding it “quite unlikely that the jury

would have found” a meaningful distinction between

pre- and post-enactment conduct. 901 F.2d at 229.2!

19 That many government witnesses testified about both pre-

and post-enactment conduct did not make it “implausible that

the jury would find such testimony compelling only for” pre-

enactment acts. Pet. App. 62a. The jury need not have

believed any witness’s testimony for one period and not for

another because those witnesses were testifying about different

transactions in the different time frames.

20 The Seventh Circuit did focus on withdrawal from the

conspiracy in Julian, finding no withdrawal and affirming on

that ground. 427 F.3d at 483. The duration of the continuing

offense in that case was not at issue, however. The only dispute

was whether the jury could have found that the defendant

withdrew from the conspiracy. Jd. To show withdrawal, a

defendant must prove an “affirmative act to defeat or disavow

the criminal aim of the conspiracy.” In light of that “high

evidentiary threshold,” the court concluded that “no reasonable

jury” would have convicted the defendant based on his pre-

enactment conduct. Jd. Outside the factual context of that

case, therefore, the Julian decision does not dictate a

“transformative event” analysis.

21 Even if the First Circuit’s “transformative event” analysis

were appropriate, reversal would still be required because there

were, in fact, “transformative events” near or prior to the

Hence, the First Circuit’s analysis conflicts with

the standard adopted by the Second, Third, Seventh

and D.C. Circuits. The First Circuit did not

demonstrate that a jury verdict based solely on pre-

enactment evidence was “impossible,” or even that

such a _ verdict was particularly “improbable.”

Rather, the court reasoned that such an outcome was

less hkely than a verdict based on post-enactment

evidence—a holding that mirrored the Fourth, Sixth,

Ninth and Eleventh Circuits’ approach. See, e.g.,

Pet. App. 62a (“considerable” post-enactment

evidence); compare Calabrese, 825 F.2d at 1346

(“substantial evidence”); Cortez, 757 F.2d at 1207

(“[e]nough activities of the conspirators’); Todd, 735

F.2d at 150 (“most of the evidence”). Like those

courts, the First Circuit focused on the volume and

sufficiency of the post-enactment evidence and did

not address whether’ pre-enactment' evidence,

standing alone, could have supported the convictions.

It also assumed, unsupported by precedent, that the

jury could not have distinguished pre-enactment

from post-enactment conduct because there was no

pre-enactment “transformative event.” The First

Circuit’s decision therefore deepened and

underscored the split among the Circuits.

enactment date that could easily have “differentiated” pre-

enactment from post-enactment conduct in the jury’s view,

notably the end of Gutierrez’s direct construction-loan

borrowing relationship with the bank.

2. There Is A Reasonable Possibility That

The Jury Relied Exclusively On Pre-

Enactment Evidence in Convicting

Petitioner.

Had the First Circuit followed the Second, Third,

Seventh and D.C. Circuits’ analysis, it would have

vacated Petitioner’s convictions. There was more

than enough pre-enactment evidence in the record to

pose a reasonable possibility that the jury relied

exclusively on such evidence.

The prosecution spent significant amounts of

time at trial presenting evidence of acts prior to

October 12, 1984. In particular, three of the seven

construction loans that were the subject of the

Gutierrez bank fraud count—Transglobe La Marina,

Levittown and Country Club—were _ executed

between 1980 and 1982 and were paid off in October

1984. There was extensive discussion of each of

those loans at trial. Government witnesses testified

at length about lack of progress on the construction

projects, payments of interest from loan proceeds,

and transfers of funds from other projects into and

out of the loans.22. The prosecution made it even

more likely that the jury would rely on this evidence

by emphasizing it repeatedly in its brief closing.”

22 See, e.g., JA-T-0083-97, -3173D-31731 (excerpts from La

Marina testimony); JA-T-0143-59, JA-T-2886-2903 (Levittown);

JA-T-0174-0244, -1196 (Country Club).

23 See, e.g., JA-T-3593 (defendants “failed to build the required

number of houses” on La Marina, Levittown and Country Club

projects), -3594 (“No houses were ever built or installed” on La

Marina and “the Levittown project, Transglobe phase, in which

very little houses were built, millions were disbursed by

Caguas”), -3595 (La Marina. Country Club and Levittown

houses “not built, even though millions of dollars were

disbursed”), -3678.

On the Mirandes bank fraud count, while there

was less pre-enactment activity, the jury could have

relied on that activity to return a guilty verdict. In

particular, there was extensive testimony about pre-

enactment conduct relating to the Reparto

Valenciano loan, executed in December 1981."4

There was similar testimony about the Villas de

Gurabo loan, executed early in 1984.25

Significant amounts of trial time were devoted to

other allegations of pre-enactment conduct. In

particular, the prosecution called ten witnesses,

spending more than two weeks of trial, to suggest

that the defendants had forged _ third-party

endorsements of two-party checks. Virtually all of

that evidence predated October 12, 1984.26 Another

prosecution witness claimed that the defendants

were responsible for “check kiting” in 1983.27 Again,

the prosecutor amplified the likelihood of the jury

relying on these pre-enactment acts by emphasizing

them at closing.?8

Moreover, the prosecution itself directed the

jury’s attention to conduct that predated the fraud

statute’s enactment. The prosecutor told the jury

24 See, e.g., JA-T-0507-19 (pre-1984 disbursements and

transfers), -2290-96 (1981 discussions between Mirandes and

bank officers about the handling of the loan), -1328-29

(increases to and disbursements from the. loan prior to

enactment).

25 See, e.g., JA-T-0529-33.

*6 See, e.g., JA-T-1550-59 (1982), -1583-86 (1982), -2013-18

(1982-83), -2382-83 (1981), -2343-44 (1982), -2067-71 (1982).

27 See, e.g., JA-T-1768-86.

28 See, e.g., JA-T-3634-35 (“In the years of 1982 and 1983...

there has been evidence with regard to forged endorsements on

checks missing from La Marina and Levittown Plaza projects.

Transglobe phases.”), -3678-80, -3575-76, -3687.

during rebuttal that “it is sufficient [for a guilty

verdict] that the evidence in the case establish[es]

beyond a reasonable doubt that the offense was

committed on a date reasonably near” the inception

of the alleged conspiracy.?? In that light, and in light

of the pre-enactment evidence presented throughout

the trial and emphasized by the prosecution at

closing, it is likely that the jury relied on that

evidence; the only question is whether it also relied

on post-enactment evidence in returning bank fraud

convictions. Nothing in the record indicates that it

did, and there is a reasonable possibility that it did

not.

Hence, fairly viewed, the record contains more

than enough pre-enactment conduct for the jury to

have rested solely on such conduct in returning its

verdicts. The Second, Third, Seventh, and D.C.

Circuits would to be plain, and would have vacated

Petitioner’s convictions.

C. THE FIRST CIRCUIT'S DECISION

CONFLICTS WITH THIS COURT'S

PRECEDENTS.

The court in this case, like the Fourth, Sixth,

Ninth and Eleventh Circuits, deemed constitutional

error harmless (under the plain-error standard) by

relying on the volume of properly admitted evidence

and resolving doubts in favor of the government.

This Court has repeatedly criticized that approach,

holding that the volume of legitimate evidence in the

record is irrelevant and that ambiguities arising

from constitutional error should be resolved against

the prosecution. The court’s willingness. to

substitute itself for the finder of fact also runs afoul

*9 JA-T-3667-68.

of this Court’s precedents cautioning against judicial

factfinding.

1. This Court Has Articulated a

Demanding Standard of Review for

Constitutional Errors.

Under both the plain and harmless error

standards, a conviction must be vacated if there is a

“reasonable possibility’ that constitutional error

affected the verdict.2° Chapman uv. California, 386

U.S. 18, 23 (1967). The Court has often applied the

Chapman analysis, or a variation thereof, in

analyzing error in the context of improper, or

improperly withheld, instructions. See, e.g., Clemons

uv. Mississippi, 494 U.S. 738, 753 (1990); Boyde v.

California, 494 U.S. 370, 380 (1990); Mills uv.

Maryland, 486 U.S. 367, 376 (1988); Rose v. Clark,

478 U.S. 570, 584 (1986); Francis v. Franklin, 471

U.S. 307, 320 (19885).

When constitutional error occurs, the volume of

properly admitted evidence is ordinarily immaterial.

The Court “has held it irrelevant in analyzing a

mandatory presumption ... that there is ample

evidence in the record other than the presumption to

support a conviction.” County Court v. Allen, 442

U.S. 140, 160 (1979). Courts may not deem errors

30 As noted above, virtually every case addressing this issue has

done so in the plain-error context and has found that the

analysis turns on whether the error affected the verdict, an

inquiry that the Court in Olano characterized as the “affecting

substantial rights” prong of plain-error review. 507 U.S. at 738.

Because the “affecting substantial rights” analysis is

substantively the same as the harmless-error test developed by

this Court, see Olano, 507 U.S. at 734, the Court’s harmless

error precedents are fully relevant.

harmless simply on grounds that there was sufficient

legitimate evidence to support the verdict:

The inquiry, in other words, is not whether, in a

trial that occurred without the error, a guilty

verdict would surely have been rendered, but

whether the guilty verdict actually rendered in

this trial was surely unattributable to the error. .

.. The most an appellate court can conclude [in

this case] is that a jury would surely have found

petitioner guilty beyond a reasonable doubt - not

that the jury's actual finding of guilty beyond a

reasonable doubt would surely not have been

different absent the constitutional error. That is

not enough.

Sullivan v. Louisiana, 508 U.S. 275, 279-80 (1993);

see also Fahy v. Connecticut, 375 U.S. 85, 86-87

(1963) (“We are not concerned here with whether

there was sufficient evidence on which the petitioner

could have been convicted without the evidence

complained of. The question is whether there is a

reasonable possibility that the evidence complained

of might have contributed to the conviction.”), The

mere existence of properly admitted evidence in the

record, therefore, even in substantial quantities, is

not sufficient to excuse constitutional error.

Where one of several possible grounds for a

conviction is legally defective, this Court has held

that the conviction cannot stand. See, e.g.,

Sandstrom v. Montana, 442 U.S. 510, 526 (1979)

(‘(W]hen a case is submitted to the jury on

alternative theories the unconstitutionality of any of

the theories requires that the conviction be set

aside.”); Street v. New York, 394 U.S. 576, 586-87

(1969) (“[A]ppellant's conviction must be set aside if

we find that it could have been based solely upon his

words and that a conviction resting on such a basis

would be unconstitutional.”); Yates v. United States,

354 U.S. 298, 312 (1957) (vacating conspiracy

conviction because “it is impossible to tell which

ground the jury selected”); Cramer v. United States,

325 U.S. 1, 36 n.45 (1945); Williams v. North

Carolina, 317 U.S. 287, 292 (1942); Stromberg uv.

California, 283 U.S. 359, 368 (1931)

Finally, the Court has repeatedly disapproved

judicial displacement of the jury’s factfinding role.

In recent years, it has repeatedly done so in the

sentencing context. See Cunningham v. California,

127 S.Ct. 856, 871 (2007); Shepard v. United States,

544 U.S. 13, 25-26 (2005); United States v. Booker,

543 U.S. 220, 246-47 (2005); Blakely v. Washington,

542 U.S. 296, 309 (2004). At other times, the Court

has expressed concern about appellate courts’ finding

of substantive facts in order to uphold a jury verdict.

See, e.g., Sullivan, 508 U.S. at 280 (“The Sixth

Amendment requires more than _ appellate

speculation about a hypothetical jury's action,”);

United States v. Bagley, 473 U.S. 667, 707 n.7 (1985)

(harmless error analysis may not “substitute the

reviewing court's judgment of the facts .. . for that of

the jury”); Marks v. United States, 430 U.S. 188, 196

n.12 (1977) (appeals court’s factual judgment as to

element of offense “not an adequate substitute for

the decision in the first instance of a_ properly

instructed jury’).

This Court, in short, has repeatedly emphasized

that harmless constitutional errors are the

exception, not the rule. Under Chapman,

constitutional error is not harmless if there is a

“reasonable possibility” that it affected the verdict;

under Allen and Sullivan, courts cannot simply point

to substantial legitimate evidence in the record and

assert on that ground that constitutional errors are

harmless; and under Yates and Sandstrom, when one

among many justifications for a jury verdict is

defective, that is sufficient to overturn a conviction.

Generally, doubts are to be resolved in the

defendant’s favor and judicial factfinding on appeal

disfavored; this Court has been reluctant to assume

that the jury ignored the presumption and relied

only on lawful grounds, or that a significant volume

of proper evidence justifies an assumption that the

jury did not rely on the improper evidence.

2. The First Circuit’s Ex Post Facto Clause

Analysis Is Inconsistent With This

Court’s Approach to Error Review.

The analysis embraced by the First Circuit here,

and by the Fourth, Sixth, Ninth and Eleventh

Circuits, conflicts with these precedents.

Specifically, contrary to Sullivan, the focus in those

decisions on whether the post-enactment evidence

was sufficient to support the jury’s verdict inverts

the “affecting substantial rights” inquiry under

Olano. Rather than vacating convictions due to the

possibility that a violation occurred, those decisions

affirm if it was possible that the violation did not

occur. See, e.g., Pet. App. 62a (“considerable” post-

enactment evidence); Calabrese, 825 F.2d at 1346

(“substantial”); Cortez, 757 F.2d at 1207 (“enough”).

That might be appropriate in a sufficiency-of-the-

evidence analysis, but it is inappropriate for

considering constitutional error, even in the plain-

error setting. Nor, under this Court’s precedents, is

it proper in this setting for lower courts to weigh the

relative volume of pre-enactment and_ post-

enactment evidence and conclude that the jury must

have relied on the larger pile of documents. See Pet.

App. 62a (“relatively few” of the overt acts in the

indictment preceded October 12, 1984); id. (“(T]he

bulk of our sufficiency analysis details conduct

occurring after the enactment date.”); Todd, 735 F.2d

at 150 (“[mjost of the evidence” post-enactment);

compare Allen, 442 U.S. at 160 (that there is “ample

evidence in the record other than the presumption to

support a conviction” is irrelevant).

To determine whether a constitutional violation

is harmless (or “affects substantial rights” under

Olano), courts must examine the record to assess the

likely impact of the violation. See, e.g., Olden uv.

Kentucky, 488 U.S. 227, 232-33 (1988). Here,

determining whether the jury could have relied on

pre-enactment evidence required an analysis of what

that evidence was, how it related to other evidence in

the record and functioned within the prosecution's

case, and how it was treated at trial. The First

Circuit failed to undertake that analysis. It asserted

in cursory fashion that Gutierrez and Umpierre

“remained employed by the company” after the

enactment date and that “their culpable conduct .. .

continued uninterrupted after the sale of the

company. Pet. App. 63a. In other words, rather

than explain why there was no “reasonable

possibility” of exclusive reliance on pre-enactment

acts, the court simply restated the undisputed fact

that the conduct continued past the enactment date.

Id. Such “analysis” is insufficient to meet the

Chapman standard. It is significant as well that the

First Circuit never concluded beyond a reasonable

doubt that the jury did not rely exclusively on pre-

enactment evidence, as Chapman requires. 386 U.S.

at 24.

EE NN OL ae NN,

Nor did the First Circuit’s assertion that “the

government presented overwhelming evidence of

appellants’ conduct, the majority of which occurred

after October 12, 1984,” support a finding that the

error did not affect Petitioner’s substantial rights.

Pet. App. 62a. Even on its face, the decision did not

establish that there was “overwhelming” post-

enactment evidence. The court opined that all of the

conduct taken together was overwhelming, but it

found only that the majority of the evidence post-

dated the enactment date. Furthermore, the court

misapplied this Court’s “affecting substantial rights”

analysis, repeatedly suggesting that the jury would

have convicted the defendants even without the pre-

enactment evidence. That is not the relevant

inquiry. The Court held in Sullivan that the

question is “not whether, in a trial that occurred

without the error, a guilty verdict would surely have

been rendered, but whether the guilty verdict

actually rendered in this trial was_ surely

unattributable to the error.” 508 U.S. at 279.

The First Circuit’s approach to this issue,

especially its focus on the sufficiency of post-

enactment evidence, conflicts with many of this

Court’s precedents. Those conflicts warrant this

Court’s review.

D. THIS CONSTITUTIONAL QUESTION IS

IMPORTANT AND RECURRING, AND IS

APPROPRIATELY RESOLVED IN THIS

CASE.

The question of how to review general jury

verdicts resting on evidence that spanned the

enactment of a pertinent statute is one of paramount

importance. The issue implicates the first and most

fundamental aspect of Ex Post Facto Clause

prohibitions—the bar to prosecution for acts that

were not criminal when committed. Calder v. Bull, 3

U.S. 386, 391-92 (1798). The Framers viewed that

protection as essential. TId.; see also Carmell uv.

Texas, 529 U.S. 513, 521-25 (2000). The question

thus goes to the heart of the “substantial personal

rights” protected by the Clause. Dobbert v. Florida,

432 U.S. 282, 293 (1977).

The Court has not yet addressed the Ex Post

Facto Clause’s applicability to a law that is enacted

in the middle of a course of conduct. The question

arises often, however. The modern criminal code is

replete with offenses that can continue over a course

of years, including the various conspiracy and fraud

statutes, and it is not uncommon for the relevant law

to change over those extended periods. When juries

are not informed about the change (either because

the parties neglect to do so or because the changed

law related to a sentencing enhancement that was

not considered at the trial), the problem of

ambiguous verdicts arises.

This case is an appropriate vehicle for resolving

this important question. Most of the circuits have

now addressed this question, so the issue is ripe for

this Court’s review. The First Circuit was fully

aware of the split, discussed it at length, and

explained the basis for its holding. Substantial

amounts of pre-enactment evidence were presented

at trial, so there is no question that the verdict could

have violated the Ex Post Facto Clause—and if the

First Circuit's approach was improper, the result

must be the vacatur of Petitioner's convictions.

Defendants’ protections against Ex Post Facto

Clause violations should not depend on where the

prosecution takes place.

CONCLUSION

For the foregoing reasons, the petition for a writ

of certiorari should be granted.

Respectfully submitted,