996 F.2d 1541
UNITED STATES of America
v.
Ramon Enrique PAULINO, a/k/a "Rafael", a/k/a Ramon Suarez,
a/k/a "Ramon E. Gomez", a/k/a "Chino", a/k/a "Chukin",
Ramon Enrique Paulino, Appellant in No. 92-1748.
UNITED STATES of America
v.
Luis PAULINO, Appellant in No. 92-1576.
UNITED STATES of America
v.
Eduardo Burgos LOPEZ, a/k/a "Louie", a/k/a "Gigio Paulino",
Eduardo Burgos Lopez, Appellant in No. 92-1580.
UNITED STATES of America
v.
Luis Hugo GARCIA a/k/a "Cujo",
Luis Hugo Garcia, Appellant in No. 92-1583.
UNITED STATES of America
v.
Jose Joaquin RODRIGUEZ, a/k/a Jose Joaquin, Luis Joaquin,
Louie, Geovanni,
Jose Joaquin Rodriguez, Appellant in No. 92-1587.
UNITED STATES of America
v.
Danilo Jacinto LEONARDO, a/k/a "Freddie Leon",
Freddie Leon, Appellant in No. 92-1622.
Nos. 92-1748, 92-1576, 92-1580, 92-1583, 92-1587 and 92-1622.
United States Court of Appeals,
Third Circuit.
Argued Feb. 25, 1993.
Decided June 28, 1993.
Sur Petition for Rehearing
July 27, 1993 in No. 92-1748.
Mark C. Levy (argued), Saul, Ewing, Remick & Saul, Philadelphia, PA, for appellant in No. 92-1748.
Jack L. Gruenstein, Philadelphia, PA, for appellant in No. 92-1576.
Louis J. Ruch, Ruch & Gontram, Philadelphia, PA, for appellant in No. 92-1580.
Jules Epstein (argued), Kairys, Rudovsky, Kalman & Epstein, Philadelphia, PA, for appellant in No. 92-1583.
William T. Cannon, Philadelphia, PA, for appellant in No. 92-1587.
Diane V. Elliott (argued), Easton, PA, for appellant in No. 92-1622.
Walter S. Batty, Jr., Asst. U.S. Atty., Chief of Appeals, John C. Dodds (argued), Michael M. Baylson, Office of U.S. Atty., Philadelphia, PA, for appellee.
Before: MANSMANN, SCIRICA and FEINBERG,* Circuit Judges.
OPINION OF THE COURT
MANSMANN, Circuit Judge.
Before us are the appeals from criminal judgments entered against several participants in a drug conspiracy which operated out of the Saloon Bar in Reading, Pennsylvania. Although the defendants have raised multiple individual grounds for appeal,1 we address only the issue common to the convicted co-conspirators2 concerning the district court's calculation of the quantity of drugs attributed to the conspiracy.
The arrest of the defendants did not yield a substantial seizure of cocaine or records evidencing the amount of cocaine flowing through the conspiracy. Although the indictment charged the distribution of over five kilograms of cocaine, the quantity specifically proven at trial was the much smaller amount of 389.2 grams. Thus, for purposes of sentencing, the district court was required to approximate, from trial evidence regarding the level of cocaine activity at the Saloon and corroborating testimony from federal agents at the sentencing hearing, an amount of cocaine which would accurately reflect the extent of the illegal enterprise of the co-conspirators.
Because the district court's evidentiary basis for its drug quantity determination met the minimum standard of reliability required at sentencing, and because its findings, based upon the evidence as testified to at trial and sentencing were not clearly erroneous, we will affirm the judgments of the district court.
I.
The evidence at trial, viewed favorably to the government, showed that cocaine activity began in approximately April of 1988 when Luis Garcia, a/k/a "Cujo," with the help of Luis Paulino and others, established the Saloon Bar as a distribution point for cocaine. Although Garcia was providing the financial backing for the opening of the bar, his interest in the real estate transaction was disguised by naming his mother as the lessor of the property.
In February 1988, George Manning was hired to help with renovations at the Saloon. Shortly thereafter, Manning was approached by Red Nuss, who offered Manning the opportunity to run the bar as Manning's own in exchange for having the liquor license put in Manning's name. Manning agreed.
Shortly after the signing of the lease, the sale of cocaine began from a room on the second floor of the Saloon. Concerned that he might be implicated in these activities, Manning met with Nuss, Garcia and Luis Paulino and threatened to pull out of the deal. Garcia and Paulino assured Manning that the drug sales were temporary and that the cocaine operation would soon be moved elsewhere. Contrary to these assurances, after Manning officially opened the bar, the drug sales upstairs increased.
Access to the upstairs area was restricted. To assure security, a steel door, steel-reinforced wall and an intercom system were installed. The room above the Saloon was manned by sellers working in shifts from mid-morning usually until midnight. The organization communicated through codes sent by way of telephone, pagers or beepers--common tools of the drug trade. Ordinarily, when a customer ordered cocaine from the room above the Saloon, a phone call was made and a "runner" would bring the drug.
Manning became a witness for the prosecution and testified at trial that the drug operation was characterized by a high level of activity and quality goods. He personally bought cocaine upstairs approximately 50 times from at least five or six different people. The purchases took place at various times of the day and night.
In addition to the numerous buys of Manning, another government witness, Gary Nesbitt, a/k/a "Jamaica," testified that he bought cocaine upstairs at the Saloon as often as twice a day, four times a week from 1988 until the Saloon closed in late 1990.
A number of law enforcement agents participated in an undercover operation at the Saloon. Detective Anthony Kerr, a Berks County Detective, frequented the Saloon posing as a narcotics buyer. At trial, Detective Kerr testified that on one occasion when he bought cocaine at the Saloon, he waited upstairs for his delivery with at least 12 or 13 other customers. Appellant Ramon Paulino, one of Garcia's "workers," apologized to Kerr for the delay, explaining that "they were really busy today."
Regarding the profits garnered from the drug trade at the Saloon, the evidence indicated that they were substantial. In March of 1990, one of Garcia's "workers," Jose Luis Novoa, complained to Luis Matos,3 a narcotics crony of Garcia, that although he made $12,000 for Garcia in one night, he had to wait one week for his $500 fee.
There was evidence that Garcia laundered drug profits through attorney Robert Van Hoove, a former District Attorney of Berks County. Van Hoove took over the legal operations of the Saloon when its former counsel and Van Hoove's law partner, Paul Schlippert, was imprisoned on a drug charge. Van Hoove's secretary, Carol Young, testified that on one occasion she saw piles of cash stacked on a table when Garcia and Van Hoove met privately. When she walked in unexpectedly, Garcia and Van Hoove pushed the money off the table into a duffle bag. Young later saw Garcia carry the duffle bag out empty.
In December 1990, 19 defendants were charged, in a 23-count indictment, with conspiracy to distribute over five kilograms of cocaine, various counts of distributing cocaine, and aiding and abetting in violation of 21 U.S.C. §§ 846 and 841(a)(1) and 18 U.S.C. § 2. In January of 1991 the grand jury returned a superseding indictment adding another defendant.
On motion of the government, the district court dismissed the indictment as to one of the 20 defendants. Another defendant pled guilty before trial and three of the defendants were fugitives at the time of trial.
On the eve of trial, defendant Luis Paulino was severed from the remaining 15 defendants because his trial counsel became unavailable. Fourteen of the remaining defendants were tried during June and July of 1991. Paulino was tried alone in November of 1991.
Seven of the initial 14 defendants were convicted at the conclusion of the first trial and Paulino was convicted after his separate trial. Of the appellants here, all but Danilo Jacinto Leonardo, a/k/a "Freddie Leon," were convicted of conspiracy to distribute cocaine. Leon was convicted of one count of distributing cocaine. In addition to the conspiracy, defendants Lopez and Rodriguez were also convicted of the substantive crime of distributing cocaine.
As previously mentioned, the investigation that resulted in the indictment did not lead to significant seizures of cocaine and no fiscal records of the alleged conspiracy were recovered. Accordingly, the trial court was required to estimate the amount of drugs sold and/or intended to be sold by the conspirators in order to properly fix the sentencing range under the Sentencing Guidelines. Examining the trial record and relying upon certain corroborative testimony from the sentencing hearing, the district court calculated that the amount of cocaine sold during the conspiracy fell within a range of 127 to 147 kilograms. This amount provided the base level for the offense. The district court then made certain adjustments based upon the individual participation of each defendant and sentenced each accordingly. This appeal followed.
We have jurisdiction to review the judgments in these criminal cases under 28 U.S.C. § 1291. Review of the appellants' sentences is also granted by 18 U.S.C. § 3742.
As identified at the outset, the paramount issue confronting the parties at sentencing was the quantity of drugs distributed by the conspiracy. Concerning facts relevant to sentencing, the government bears the burden of proof by a preponderance of the evidence. United States v. McDowell, 888 F.2d 285, 290 (3d Cir.1989).4 We review the district court's findings of fact concerning the drug quantity to determine whether they are clearly erroneous. United States v. Miele, 989 F.2d 659, 663 (3d Cir.1993).
II.
In imposing a sentence under the Sentencing Guidelines in a narcotics case, the district court relies chiefly upon the quantity of drugs involved in the offense. U.S.S.G. §§ 1B1.3(a)(2); 3D1.2(d). The relevant amount of drugs establishes the base offense level, see § 2D1.1(a)(3). Then, the sentencing judge may adjust upward or downward to reflect other factors and decides upon a sentence which fits in the appropriate Guideline range.
We have recognized that in calculating the amount of drugs involved in a particular operation, a degree of estimation is sometimes necessary. Often the covert nature of the drug trade precludes seizure and precise measurement of the drugs that flow through a drug distribution conspiracy. United States v. Collado, 975 F.2d 985, 998 (3d Cir.1992). The Sentencing Guidelines recognize the difficulties posed and authorize the sentencing court's approximation of unseized drug quantities:
Where there is no drug seizure or the amount seized does not reflect the scale of the offense, the sentencing judge shall approximate the quantity of the controlled substance. In making this determination, the judge may consider for example, the price generally obtained for the controlled substance, financial or other records, similar transactions in controlled substances by the defendant....
U.S.S.G. § 2D1.4, application note 2.
This recognition of the need to estimate, however, is not a license to calculate drug amounts by guesswork. Instead, the sentencing court must carefully scrutinize the government's proof to ensure that its estimates are supported by a preponderance of the evidence. Collado, 975 F.2d at 998.
For example, in Collado, we overturned a finding that telephone calls regarding heroin transactions must have referred to a sale of at least 62.5 grams because the defendants dealt in quantities of at least that amount on other occasions. We upheld, as not clearly erroneous, the district court's acceptance of the government's estimate that a drug transaction involved 125 grams based upon use of the term "one". This finding was supported by the testimonies of a government expert that co-conspirators had previously referred to one-eighth of a kilogram of heroin as "one", and of a conspirator, testifying for the government, that the conspiracy's code word for specific quantities of drugs was "one." Id. at 999.
III.
The present situation likewise exemplifies the difficulty of determining the quantity of cocaine involved in a particular drug operation. The indictment charged that the defendants conspired to distribute over five kilograms of cocaine during a 30-month period extending from April of 1988 through October 1990. When, at sentencing, the government was called to proof concerning the quantity of cocaine, it began by moving to incorporate the trial record. The government emphasized the sophistication and the duration of the Saloon drug operation: the steel-reinforced wall, intercom and runner system; and the hours of operation--10 a.m. to midnight, seven days a week. Also noted was the testimony of the undercover officer that on one occasion, he waited with 12 or 13 other customers for delivery of his cocaine. The testimony of Manning, who made fifty purchases, and of Nesbitt, who made as many as eight buys a week for two years, were also offered as proof of quantity. The most specific evidence of volume presented by the government, however, was deduced from the testimony at trial that Jose Luis Novoa complained to Luis Matos that he had to wait to be paid $500 from Garcia when in one shift Garcia took in $12,000 from the sale of cocaine from the Saloon.
The government then presented an experienced Drug Enforcement Agency agent, Ellis Hershowitz, who was familiar with the price of cocaine in the Reading area. Hershowitz testified that cocaine was selling in Reading for $1,000 an ounce during the period of the conspiracy. Thus, the $12,000 figure was converted into a specific amount of cocaine to determine the amount transacted on a daily basis at the Saloon. The appellants have not challenged Hershowitz' estimate of the per ounce value of the cocaine.
The government then presented two additional witnesses: Special Agents Sharon Rueppel of the Drug Enforcement Agency and Ernest Gesko of the Immigration and Naturalization Service. Special Agent Rueppel related facts elicited from a debriefing of a confidential informant who, living in the closely knit community of the defendants, claimed knowledge of how the various defendants and their associates obtained and distributed five to ten kilograms of cocaine weekly. Rueppel also testified concerning amounts and frequency of drug deliveries to the participants in the Saloon Bar conspiracy which had been obtained by another DEA agent who had debriefed two other unidentified informants. Neither the informants nor the debriefing agent testified at sentencing; nor had Special Agent Rueppel met the two informants whose information she testified to at sentencing.
Special Agent Gesko testified regarding his debriefing of a codefendant named in the indictment, Fernandez-Minalla, who did not testify at sentencing. Gesko testified that Minalla informed him that a person named "Aguja" recently delivered 10 kilograms of cocaine to Garcia and Leon. The informant stated that Aguja was accompanied by Chalupa Rodriguez, Paulino's father-in-law. Minalla also told Gesko that Garcia and Leon were the largest drug dealers in Reading.
Based upon the evidence presented, the district court made the following findings regarding the amount of cocaine: the distributions attributed to the defendants who were convicted of the conspiracy totalled 14.2 grams; additional evidence showed that Garcia distributed a total of 375 grams of cocaine to Matos. Thus, the distribution proven at trial within the ambit of the conspiracy totalled 389.2 grams.
The district court concluded, however, that based upon trial evidence corroborated by testimony at the sentencing hearing, the total quantity of cocaine involved in the conspiracy far exceeded the weight of the cocaine for which the defendants had been convicted.5 The court then calculated the total volume of cocaine, relying primarily on the testimony of Novoa as to the dollar amount of cocaine sold on one particular night of business on the second floor of the Saloon. The district court accepted the testimony of Special Agents Rueppel and Gesko regarding their own informants,6 solely as "corroborative of its finding, independently based on the trial evidence, that there was a high level of cocaine sales on a regular basis by those found guilty of the conspiracy." The court made specific note that it considered the agents' testimony because they were supported by the required indicia of reliability.
On appeal, the appellants argue that the district court improperly relied upon Special Agents Rueppel's and Gesko's inadmissible hearsay evidence in determining the drug quantity involved. To the contrary, under the applicable standard of review for admission of evidence at sentencing, the hearsay was properly heard and considered.
At sentencing, "a convicted criminal is entitled to less process than a presumptively innocent accused." United States v. Mobley, 956 F.2d 450, 455 (3d Cir.1992). Thus, a sentencing judge may consider information, "largely unlimited" as to kind or source, that would be inadmissible at trial. Roberts v. United States, 445 U.S. 552, 556, 100 S.Ct. 1358, 1362, 63 L.Ed.2d 622 (1980).
Prior to the Sentencing Guidelines, the principle that sentencing judges could consider evidence at sentencing that would not be admissible at trial was firmly established. See Williams v. New York, 337 U.S. 241, 246-47, 69 S.Ct. 1079, 1082-83, 93 L.Ed. 1337 (1949) (judge has broad discretion as to sources and types of information relied upon at sentencing). Regardless of this discretion to discover a broad range of information, however, the introduction of evidence at sentencing is subject to a due process standard of reliability. Townsend v. Burke, 334 U.S. 736, 741, 68 S.Ct. 1252, 1255, 92 L.Ed. 1690 (1948).
We examined the due process constraints on admission of evidence in United States v. Baylin, 696 F.2d 1030 (3d Cir.1982), a pre-Guidelines case. In Baylin, we remanded a case for resentencing where the sentencing court had inferred the defendant's involvement in a crime solely from the fact that the government had promised not to prosecute another crime. We held that "[i]t is impermissible in sentencing to rely upon such an inference, since it does not meet the minimum reliability standard we have enunciated." Id. at 1042. We nevertheless established a broad standard for admission of evidence at sentencing. We held that, as a matter of due process, facts relied upon at sentencing must contain "minimum indicia of reliability beyond mere allegation." Id. at 1040.
In addressing this admissibility issue, the Sentencing Commission issued a policy statement recommending a standard of reliability stricter than that which we articulated in Baylin. Section 6A1.3 of the Guidelines, while not establishing a specific standard of proof, suggests that information relied upon at sentencing should have "sufficient indicia of reliability to support its probable accuracy." U.S.S.G. § 6A1.3(a). We have specifically held that while "the guidelines sentencing limits the exercise of judicial discretion, ... it does not follow that due process requires a different rule with respect to hearsay." United States v. Sciarrino, 884 F.2d 95, 97 (3d Cir.1989). See also United States v. Torres, 926 F.2d 321, 324 (3d Cir.1991) (evidence suppressed at trial in violation of the Fourth Amendment admitted at sentencing).
Recently, in United States v. Miele, 989 F.2d at 664, we observed that this "sufficient indicia of reliability" standard of admissibility, adopted by several of our sister circuits, "should be rigorously applied." Id.7
In Miele, an informant, who was also a drug addict, made inconsistent statements at sentencing about various amounts of narcotics involved in a conspiracy. No other witnesses testified as to specific drug quantities and there was no other corroborative evidence to support the estimate. The district court, in setting the base offense level, opted to follow the informant's higher estimate of the quantities involved without explanation. We held that admission of the informant's testimony was not supported by any indicia of reliability sufficient to determine the base offense level for sentencing purposes. Id. at 665. In so holding we distinguished our decision in Sciarrino, where we upheld the district court's decision to discredit a co-conspirator's testimony at sentencing and to rely instead on that witness' prior inconsistent and hearsay statement as a basis for the court's quantity finding. 884 F.2d at 97. We found the hearsay in Sciarrino sufficiently reliable to support the drug quantity finding because of substantial corroborating testimony from other witnesses. Id. at 97.
In this case, the sentencing court properly considered the testimony of the agents, despite the hearsay (and, in the case of Special Agent Rueppel, the layers of hearsay) present in the reports to which they testified. First, the totality of the circumstances described by the agents shows the hearsay declarations to be reasonably trustworthy. An agent's recital of events concerning discussions with a reliable confidential informant, and recollection of a conversation relayed to another federal agent working on the same case, meet the standard of minimum indicia of reliability. We emphasize, too, that, unlike the Miele case but similar to Sciarrino, there was ample evidence from other witnesses, including specific drug quantity estimates, that corroborated the hearsay statements.
The hearsay statements of the agents thus had sufficient indicia of reliability to support the court's drug quantity finding here and were properly admitted and considered by the district court.
We turn now to the question of whether the district court's final approximation of the amount of cocaine attributable to the conspiracy was clearly erroneous.
IV.
The cornerstone of the district court's drug quantity calculation was the testimony from one of Garcia's workers, Novoa, complaining that although the cocaine enterprise made $12,000 on a particular night, he waited a week to be paid his $500 salary. The court decided that, although the government had proved by a preponderance of the evidence per the testimony of Agent Hershowitz that 12 ounces, at $1,000 an ounce, represents at least the amount of cocaine sold on a good day on a regular basis, the government failed to prove that this amount was sold every day for two years. In order to take into account the days in which sales were not that high or days in which no sales were made, the court reduced by 50 to 55% the government's proposed 255 kilogram estimate based on 12 ounces a day for the period of the conspiracy. The district judge concluded that, by his most satisfactory estimate, the amount of cocaine sold during the scope of the conspiracy was thus within a range of 127 to 140 kilograms.
The district court's calculation was squarely based on the quantifiable trial evidence--Novoa's statement concerning the day when sales in the amount of $12,000 were transacted. The halving of this amount, rather than being arbitrary, is, instead a reasonable calculation by the district court, erring on the side of caution, to take into consideration "off" days and days in which perhaps lesser sales occurred.
The defendants challenge the district court's quantity approximation based on United States v. Reyes, 930 F.2d 310, 316 (3d Cir.1991). In Reyes, the trial court estimated the amount of cocaine based solely on the testimony of the single prosecution witness. After review of the trial record, the government acknowledged on appeal that the testimony of this witness standing alone could not support the court's finding. Accordingly, we remanded the case to the district court for resentencing.
Obviously, the facts in Reyes and in this case are different. First, there is no concession by the government as to the unreliability of its evidence. Two, there is corroborative evidence of the quantity of cocaine, albeit hearsay, but nonetheless admissible, concerning the quantity. Indeed, in Reyes, we expressly stated that the length of a drug conspiracy together with other evidence may provide a basis for a satisfactory estimate of the quantity of drugs involved. 930 F.2d at 315.
This is not a case like Reyes where the limited testimony relied upon was not reliable, nor is it like Miele where the district court credited unsubstantiated and questionable testimony. Here, there was no challenge to the reliability of Novoa's statement and the agents' testimony at sentencing readily met the "sufficient indicia of reliability" standard.
Finally, we note that the district court's finding of 127 to 140 kilograms of cocaine distributed by the conspiracy placed the defendants' base level at 36--the offense level applicable to drug quantities ranging from 50 to 150 kilograms. U.S.S.G. § 2D1.1(c)(4). Thus, to affect the sentences, i.e., to move the base offense down to the next lowest level, the district court's approximation would have to be erroneous by 77 grams, more than 50% off the calculation. Given this leeway, we are convinced the district court's findings as to the amount of cocaine are not clearly erroneous.
V.
We will, therefore, affirm the judgments of sentence in this matter.
Present: SLOVITER, Chief Judge, BECKER, STAPLETON, MANSMANN, GREENBERG, HUTCHINSON, SCIRICA, COWEN, NYGAARD, ROTH, LEWIS and FEINBERG,** Circuit Judges.
SUR PETITION FOR REHEARING
July 27, 1993.
The petition for rehearing filed by appellant in the above entitled case having been submitted to the judges who participated in the decision of this court and to all other available circuit judges of the circuit in regular active service, and no judge who concurred in the decision having asked for rehearing, and a majority of the circuit judges of the circuit in regular active service not having voted for rehearing by the court in banc, the petition for rehearing is denied. Judges Becker and Stapleton would have granted rehearing.
Honorable Wilfred Feinberg of the United States Court of Appeals for the Second Circuit, sitting by designation
We will affirm, without discussion, the district court's resolution of the other issues: whether a co-conspirator's statement was properly admitted; whether evidence of a prior conviction should have been excluded; whether evidence that a co-defendant was selling heroin as well as cocaine was properly excluded; whether the evidence was sufficient to support Luis Paulino's conviction; and, whether impeachment evidence should have been allowed. Other sentencing issues raised are whether the amount of cocaine involved in the conspiracy was reasonably foreseeable to some of the defendants; the "roles" of various defendants; and whether a particular kilogram of cocaine was properly included as an amount "under negotiation" for purposes of sentence calculation
After due consideration, we determine that none of these issues has merit.
The co-conspirators, the appellants here, are: Luis Hugo Garcia, Luis Paulino, Eduardo Burgos Lopez, Jose Joaquin Rodriguez and Ramon Enrique Paulino
Luis Matos ran a smaller cocaine business in Reading. On one occasion Matos approached Garcia about a loan to open a dress shop. Garcia offered alternatively to supply Matos with cocaine. According to Matos' trial testimony as a witness for the government, he received cocaine from Garcia on four occasions in the amounts of 125 grams, 62 grams, 62 grams and 125 grams
In United States v. Kikumura, 918 F.2d 1084 (3d Cir.1990), we cautioned that, under certain circumstances, a higher standard of proof may be required. Id. at 1098. We announced that in the case of an extreme departure from the Guidelines, (where the sentencing hearing does, in fact, become the " 'tail that wags the dog of the substantive offense' ") a clear and convincing standard of proof is required. Id. at 1101, quoting McMillan v. Pennsylvania, 477 U.S. 79, 88, 106 S.Ct. 2411, 2417, 91 L.Ed.2d 67 (1986). The appellants today urge that a heightened clear and convincing standard is applicable here. However, the present fact-finding dispute does not arise in a Kikumura significant-departure context. Thus, we are not compelled to utilize the clear and convincing standard and instead review the government's proof under the preponderance of the evidence standard
Even the appellants conceded that the amount of cocaine involved in the distribution counts did not exemplify the scope of the conspiracy
The court did not mention Rueppel's testimony regarding the informant she did not personally debrief
Akin to their argument, based upon Kikumura, that the government was required to prove the amount of cocaine under the heightened clear and convincing standard, the appellants claim that a heightened standard of admissibility of hearsay should be employed, i.e., "the court should examine the totality of the circumstances, including other corroborating evidence, and determine whether the hearsay declarations are reasonably trustworthy," 918 F.2d at 1103. As previously noted, we are not confronted here with a substantial departure from the Guidelines and thus are not compelled to follow the dictates of Kikumura in this instance
Honorable Wilfred Feinberg of the United States Court of Appeals for the Second Circuit, as to panel rehearing only