United States v. Garcia

IN THE UNITED STATES COURT OF APPEALS FOR THE FIFTH CIRCUIT ___________________ No. 96-20149 Summary Calendar UNITED STATES OF AMERICA, Plaintiff-Appellee, versus CESAR JAVIER GARCIA, also known as Cowboy, Defendant-Appellant. ________________________________________________ Appeal from the United States District Court for the Southern District of Texas (CR-H-94-288-1) ________________________________________________ July 15, 1997 Before GARWOOD, DeMOSS and STEWART, Circuit Judges.* GARWOOD, Circuit Judge: Defendant-appellant Cesar Javier Garcia (Garcia) appeals his conviction and sentence for conspiracy to possess with intent to distribute more than one hundred kilograms of marihuana. We affirm. Facts and Proceedings Below * Pursuant to 5TH CIR. R. 47.5, the Court has determined that this opinion should not be published and is not precedent except under the limited circumstances set forth in 5TH CIR. R. 47.5.4. On February 3, 1994, Alex Mata (Mata) entered into a written agreement with the Harris County District Attorney’s Office and Houston Police Department (State) whereby Mata——who had a criminal history and had pending at the time a state felony charge for delivery of marihuana——agreed to assist the State with the arrest and charging of Garcia for aggravated robbery or conspiracy to commit aggravated robbery.1 To accomplish this goal, the State along with federal authorities set up a “sting” operation in which Mata would play the role of a middleman between Garcia (the buyer) and Crisanto Perez (Perez), an undercover special agent with the Bureau of Alcohol, Tobacco, and Firearms who posed as a marihuana seller from the Rio Grand Valley in South Texas. The plan called for Garcia and Mata to arrange a sham marihuana buy with Perez so that Garcia could find out where the marihuana was being stored, allowing him to steal the drugs for resale. As planned, in late April of 1994 Mata approached Garcia and told him that Perez had 1,000 pounds of marihuana and was looking for a potential buyer. Garcia told Mata that he was interested in working with him and provided Mata with his telephone number and pager number. Beginning in May of 1994, Mata began secretly recording conversations he had with Garcia over the telephone and in person. In these conversations, Garcia continued to express 1 In exchange for Mata’s assistance, the State agreed to dismiss Mata’s pending marihuana charge and agreed not to revoke his parole. Also, prior to Garcia’s trial, the State paid Mata’s relocation expenses totaling approximately $3,500. 2 interest in the sham transaction and indicated that his “compadre” Jose Alcantar (Alcantar), also known as Joe Diamond or Diamond Joe, would assist in the operation.2 On June 30, 1994, Mata and Perez met with Alcantar at Michael’s International Nightclub in the southwest area of Houston. Also present with Alcantar at the nightclub were Garcia and several other men. Mata went over to the table at which Alcantar, Garcia, and others were sitting and, a short while later, returned to Perez’s table with Alcantar. Alcantar told Perez that he was interested in purchasing 250 pounds of marihuana, and Alcantar and Perez agreed on a price of $550 per pound. In accordance with Garcia’s instructions, Alcantar then followed Perez to the warehouse where the marihuana was being stored so that Alcantar could inspect the drugs. After testing the marihuana, Alcantar approved of its quality and requested that the transaction take place the next day.3 Over the next several days Garcia, Alcantar, and Mata remained in contact with one another and had numerous discussions pertaining primarily to the location of the marihuana, the amount of marihuana that remained in the warehouse, and the timing of the robbery. At 2 Mata also secretly recorded conversations he had with Alcantar. 3 As part of the undercover operation, police officers had placed 1,000 pounds of marihuana in a warehouse so that Alcantar could inspect the drugs. They also set up hidden video cameras to record the meeting and any subsequent robbery attempts. 3 one point during this time, Mata informed Garcia that the quantity of marihuana stored in the warehouse had increased to 1,400 pounds. On July 20, 1994, Mata and Alcantar tried unsuccessfully to contact Garcia. Alcantar called Mata that same day and told Mata that he wanted to go ahead with the robbery without Garcia. Apparently, Alcantar “said he already talked to Cesar and it was no problem with Cesar.” In the early morning hours of July 21, 1994, Mata, Alcantar, and several other men drove to the warehouse with the intention of stealing the marihuana. Alcantar and two of his cohorts, Julian Mata, Jr. and Norberto Coronado, drove in one vehicle while six other men followed in a grey van.4 Unbeknownst to the would-be thieves, the warehouse was then empty, as the police had previously transported the drugs to a different location. When they arrived at the warehouse, one of the men in the van hooked up a steel chain to the garage door of the warehouse and, with the other end of the chain connected to the van, proceeded to pull the door off the warehouse. Much to their dismay, the entire load of marihuana was gone. As they attempted to flee, SWAT officers swarmed the area and arrested the six men who had arrived in the van.5 Four weapons were seized from the van, including 4 The individuals in the van were Robert Luis Gonzalez, Rogelio Mata, Juan Mata, Adrian Mata, Joe Luis Lerma, and Raymond Palomo Trevino. 5 Hidden video cameras captured the botched robbery on tape. 4 three 9mm pistols, a .38 caliber pistol, and several rounds of ammunition. Alcantar and the two men in his car, who had left the area before the SWAT team arrived, were all arrested a few days later. On July 24, 1994, Mata and an acquaintance were driving on a Houston road when Garcia, who was driving his family to a shooting range, pulled up next to Mata’s vehicle, pointed a gold-plated .45 caliber semi-automatic pistol at Mata, and shouted a threat. Mata’s vehicle accelerated quickly away. Mata provided the police with a description of Garcia’s vehicle and weapon. On July 25, 1994, police officers arrested Garcia after they spotted him driving a vehicle that matched Mata’s description. The officers found in the glove compartment of his car the gold-plated pistol that Mata alleged Garcia had pointed at him. On December 2, 1994, Garcia was charged by indictment with conspiracy to possess with intent to distribute marihuana (count one) in violation of 21 U.S.C. §§ 841(a)(1), (b)(1)(B), and 846, and aiding and abetting the use and possession of a firearm during and in relation to a drug trafficking offense (count two) in violation of 18 U.S.C. § 924(c).6 On October 19, 1995, a jury found Garcia guilty on count one but not guilty on count two. On February 5, 1996, Garcia was sentenced to 160 months in prison, 5 6 Charged in the same indictment were codefendants Alcantar, Julian Mata, Jr., Norberto Coronado, Robert Luis Gonzalez, Rogelio Mata, Juan Mata, Adrian Mata, Joe Luis Lerma, and Raymond Palomo Trevino. All defendants except Garcia pleaded guilty. 5 years of supervised release, and fined $5,000. Discussion Garcia raises numerous arguments on appeal. Specifically, he alleges that (1) the method of selecting jurors in the Houston Division of the Southern District of Texas systematically excludes Hispanics; (2) the agreement between the government informant and the State violated due process; (3) the evidence was insufficient to support his conviction for conspiracy to possess with intent to distribute marihuana; (4) the district court improperly instructed the jury on his withdrawal defense; (5) the district court abused its discretion by declining to instruct the jury on multiple conspiracies; and (6) the district court erred when it calculated the amount of marihuana attributable to him for sentencing purposes and when it increased his offense level based on obstruction of justice and his leadership role in the conspiracy. We address each argument in turn below. 1. Jury Selection Process Garcia contends that the method of selecting jurors in the Houston Division of the Southern District of Texas systematically excludes Hispanics in violation of the Sixth Amendment to the United States Constitution and the Jury Selection and Service Act of 1968, 28 U.S.C. §§ 1861-1878 (Jury Act). Under the current system employed by the Houston Division, the jury wheel from which both grand and petit jurors are selected is generated by selecting 6 names exclusively from voter registration lists. Garcia contends that because at least 13% of the Hispanics in the Division are eligible to serve as jurors but only 7.5% of Hispanics actually register to vote, the current system does not produce grand and petit juries that represent a fair cross-section of the community. Garcia believes that the Division has available and should use a list of registered voters and licensed drivers that would, according to Garcia, more accurately reflect the racial composition of the Division.7 Under both the Sixth Amendment and the Jury Act, litigants are entitled to have their grand and petit juries drawn from a fair cross-section of the community in the district or division where the court convenes. United States v. McKinney, 53 F.3d 664, 670-71 (5th Cir.), cert. denied, 116 S.Ct. 261 (1995). In order to make a prima facie showing that his right under the Sixth Amendment to have a jury drawn from a fair cross-section has been violated, 7 In ruling on Garcia’s Motion to Dismiss Because of Improper Selection of Grand and Petit Jury, the district court below adopted the conclusions of law and findings of fact entered in a similar case, United States v. Rodriguez, Cr. No. H-94-216 (S.D. Tex.). The parties in this case submitted all of the evidence and the transcript of the hearing in the Rodriguez case as evidence to the district court. In addition to the evidence from the Rodriguez case, the court below also considered the testimony of Ray Hardy (Hardy), a former District Clerk of Harris County who was involved in the development of Texas’s new jury selection system, which now includes drivers license lists. Hardy testified that the main motivation behind this change was a perception that many people were purposefully avoiding jury service by choosing not to register to vote. 7 Garcia must demonstrate that (1) the group he claims is being excluded is a “distinctive” group within the community; (2) the group’s representation in venires from which juries are selected is not fair and reasonable in relation to the number of such persons in the community; and (3) this underrepresentation is caused by systematic exclusion of the group in the jury selection process. Duren v. Missouri, 99 S.Ct. 664, 668 (1979). The distinctive group consists of the pool of individuals in that division who are eligible to serve as jurors and not the group’s total population in the community.8 United States v. Fike, 82 F.3d 1315, 1321 (5th Cir.), cert. denied, 117 S.Ct. 241 (1996). A trial court’s factual determination that there is no systematic exclusion of minority members from the venire is reviewed for clear error. United States v. Sotelo, 97 F.3d 782, 790 (5th Cir. 1996), cert. denied, 117 S.Ct. 1002 (1997). In United States v. Maskeny, 609 F.2d 183 (5th Cir.), cert. denied, 100 S.Ct. 3010 (1980), we held that in order to determine whether there has been a violation of the Sixth Amendment we must look to the “absolute disparity” between the proportion of members of an identifiable class in the community and its representation in 8 In order to serve on a jury, one must be a citizen of the United States, at least 18 years old, be able to understand, read, speak, and write English, not suffer from a mental and physical infirmity, and not have a charge pending against him or her or have a conviction of a crime punishable by imprisonment of more than one year. 28 U.S.C. § 1865(b). 8 venires from which juries are selected.9 Id. at 189-90. In so holding, we rejected the defendants’ argument that we should base our Sixth Amendment analysis on data derived from other statistical methods, such as comparative disparity and standard deviation. Id. at 190. The Court went on to conclude that because the absolute disparity “between the percentage of each allegedly ‘distinctive’ group in the community and the percentage of that group . . . ending up on the qualified wheel is less than ten percent,” the defendants did not make out a Sixth Amendment violation. Id.; see also United States v. Butler, 611 F.2d 1066, 1069-70 (5th Cir.), cert. denied, 101 S.Ct. 97 (1980). Since Maskeny, many other courts have similarly held that an absolute disparity of less than 10% alone is not enough to demonstrate underrepresentation under the Sixth Amendment. See, e.g., United States v. Joost, 94 F.3d 640 (tab.), No. 95-2031, 1996 WL 480215, at *8 (1st Cir.), cert. denied, 117 S.Ct. 408 (1996); United States v. Grisham, 63 F.3d 1074, 1078-79 (11th Cir. 1995), cert. denied, 116 S.Ct. 798 (1996); United States v. Ashley, 54 F.3d 311, 313-14 (7th Cir.), cert. denied, 116 S.Ct. 232 (1995); see also United States v. Hawkins, 9 Absolute disparity measures the difference between the percentage of a distinctive group in a certain population and the percentage of that group in a subset of that population. In the jury selection context, this figure is generally achieved by subtracting the percentage of a group on the jury wheel from the percentage of persons within that group who are eligible to serve as jurors. See United States v. Esquivel, 88 F.3d 722, 726 (9th Cir.), cert. denied, 117 S.Ct. 442 (1996). 9 661 F.2d 436, 442 (5th Cir. 1981). We reiterated our less-than-ten-percent standard in United States v. Butler, 615 F.2d 685 (5th Cir. 1980) (per curiam) (denying petitions for rehearing and rehearing en banc). In Butler, we clarified our prior panel opinion by explaining that “[w]e did not wish to imply that the absolute disparity method is the sole means of establishing unlawful jury discrimination. However, given the small absolute disparities proven and the fact that a ‘less-than-10 percent minority’ was not at issue, we did not feel consideration of other statistical methods was necessary in this case.”10 Butler, 615 F.2d at 686. 10 Garcia asserts that this Court should abandon the absolute disparity test and should adopt either the comparative disparity model or the disparity of risk model. Under the comparative disparity model, which focuses on the percentage difference between the proportion of the distinctive group eligible to serve as jurors and the shortfall in that group’s representation, the disparity would amount to approximately 42%. See United States v. Hafen, 726 F.2d 21, 23-24 (1st Cir.), cert. denied, 104 S.Ct. 2179 (1984). Garcia contends that the comparative model is especially important in situations, such as in this case, where the minority group is so small that it would be difficult to meet the 10% minimum under the absolute disparity model. The disparity of risk model measures the statistical frequency of underrepresentation of the distinctive group on juries, which in this case would be approximately 35%. Under either of these alternative models, the disparity would be substantially greater than the 4.8% disparity under the absolute disparity model. See infra. As a threshold matter, we do not agree with Garcia’s assertion that the comparative disparity model provides a more accurate assessment of underrepresentation than does the absolute disparity model. See Hafen, 726 F.2d at 23-24 (explaining that “the smaller the group is, the more comparative disparity figure distorts the proportional representation”). Moreover, most courts have rejected requests by litigants to abandon the absolute disparity model in favor of one of the alternative disparity models. See, e.g., Joost, 1996 WL 480215, at *8; Esquivel, 88 F.3d at 726; Ford v. Seabold, 841 F.2d 677, 684 & n.5 (6th Cir.), cert. denied, 109 S.Ct. 315 (1988); United States v. Rodriguez, 776 F.2d 1509, 1511 (11th Cir. 1985); Hafen, 726 F.2d at 23. However, even assuming, 10 See also Rodriguez, 776 F.2d at 1511 n.4 (stating that “[a]lthough the absolute disparity method is not the sole means of establishing unlawful jury discrimination, where small absolute disparities are proven, as in this instance, and the minority group involved exceeds ten percent of the population, which is also the case in this challenge, it is not necessary to consider other statistical methods”). The parties do not dispute the fact that the percentage of Hispanics eligible to serve as jurors in the Division exceeds 10%; thus, we need only consider the absolute disparity between the percentage of Hispanics eligible to serve on juries and the percentage of Hispanic registered voters in the Division.11 arguendo, that we considered the comparative disparity or disparity of risk model to be superior to the absolute disparity model, we would nevertheless be precluded from adopting either of these alternative models today, as we are bound by Maskeny and Butler absent a Supreme Court decision or an en banc decision by this Court indicating otherwise. See United States v. Pettigrew, 77 F.3d 1500, 1511 n.1 (5th Cir. 1996). 11 Neither party opposes using the percentage of adult Hispanic citizens as a substitute for the percentage of Hispanics eligible to serve on juries, even though jury service requires more than simply being an adult citizen (e.g. a degree of proficiency in English). Also, both parties wholly adopt the data produced in Rodriguez despite the fact that the Rodriguez case used statistical evidence gathered on both the Houston and Galveston Divisions, while this case only involves a challenge to the selection system in the Houston Division. Moreover, as discussed in footnote 9, supra, absolute disparity figures are generally calculated by taking the difference of the percentage of persons within a group who are eligible to serve as jurors and the percentage of that group on the jury wheel. The court below, as did the district court in Rodriguez, calculated absolute disparity by subtracting the percentage of adult Hispanic 11 According to two of the experts who provided jury data analysis to the district court in Rodriguez, Professors John R. Alford and Robert M. Stein of Rice University, 7.7% of the registered voters in the Houston and Galveston Divisions are Hispanic, while the percentage of Hispanic adult citizens in the same Divisions is 12.4%.12 Another expert, Professor Kent L. Tedin of the University of Houston, opined that the percentages of Hispanic registered voters and adult Hispanic citizens are closer to 7.6% and 12.8%, respectively. The court found——and we do not consider the finding to be clearly erroneous——that the 7.6% and 12.4% figures best reflect the actual percentages, and concluded that the 4.8% disparity unequivocally demonstrates that the jury selection system does not result in an underrepresentation of Hispanics on the jury wheel.13 Thus, because the percentage of Hispanics eligible to serve as jurors in the Division clearly well exceeds 10% and the absolute disparity between the percentage of Hispanic registered citizens from the percentage of Hispanic registered voters. The parties did not object below nor do they object on appeal to using percentages of registered voters instead of percentages of Hispanics on the jury wheel to determine underrepresentation. In any event, nothing in the record indicates that the percentage of Hispanics who register to vote is any different from the percentage of Hispanics who appear on the jury wheel. 12 Professors Alford and Stein also found that 7.46% of the persons called to serve on grand juries were Hispanic, 7.28% of the persons who appeared were Hispanic, and 7.34% of the persons who were selected were Hispanic. 13 The court decided to use the 7.6% figure because it was more favorable to Garcia. 12 voters and the percentage of Hispanics eligible to serve on juries is substantially less than 10%, we hold that Garcia has not made a prima facie case of underrepresentation under the Sixth Amendment.14 Likewise, Garcia’s Jury Act challenge also lacks merits. To show that the method of selecting jurors in the Houston Division is not in conformity with the requirements of the Act, Garcia must establish that there is a substantial failure to comply with the Act’s provisions. 28 U.S.C. § 1867. Technical violations that do not affect the random nature or objectivity of the selection process do not constitute a substantial failure to comply. See United States v. Brummitt, 665 F.2d 521, 528 (5th Cir. 1981), cert. denied, 102 S.Ct. 2244 (1982). Garcia has failed to make the necessary showing. The Jury Act itself expressly authorizes the adoption of a jury plan under which potential jurors are selected exclusively from voter registration lists; additional sources are only required if necessary to protect rights secured by the statutory scheme involved. See 28 U.S.C. § 1863(b)(2). The legislative history of the Act “suggests that Congress, rather than disapproving the element of nonrandomness implicit in any form of self-screening, wanted people who lacked a sense of civic obligation not to serve on federal juries, unless 14 Moreover, even if we were to accept Garcia’s exaggerated figures of 7.5% and 13%, the absolute disparity would still amount to only a 5.5% difference, far shy of the 10% minimal disparity needed under Maskeny and Butler. 13 the number of ‘no-shows’ was so great that the qualified jury wheel could not be filled up. In that event, but only in that event, the clerks could be expected to use the coercive powers that the Act gave them.” United States v. Gometz, 730 F.2d 475, 480 (7th Cir.), cert. denied, 105 S.Ct. 155 (1984). Hence, the “failure of an identifiable group to register and vote does not render invalid the selection of jurors from a voter registration list.” Brummitt, 665 F.2d at 529; see also United States v. Apodaca, 666 F.2d 89, 92-93 (5th Cir.), cert. denied, 103 S.Ct. 53 (1982); United States v. Arlt, 567 F.2d 1295, 1297 (5th Cir.), cert. denied, 98 S.Ct. 2250 (1978). We do not consider the district court’s finding that the current selection system “is advantageous because it results in the selection of individuals over 18 years and those who are citizens of the United States, both of which are qualifications for jury service,” to be clearly erroneous. Additionally, “[t]he use of voter lists as a source insures that individuals who are interested enough in their community to register to vote will be able to serve as jurors.” Nothing in the record compels us to conclude that the court clearly erred in its determination that the current system “is unbiased at each of its selection points” and that “[t]he selection process is racially and ethnically neutral and is not susceptible to being used as a tool of discrimination.” Furthermore, Garcia has not explained how a jury selection 14 system based on both voter registration and drivers license lists would produce a jury wheel that would be more reasonable than that generated under the current system. Indeed, as evidenced by the testimony of Hardy given at the pre-trial hearing, such a system would likely be overly inclusive, as many people who would otherwise qualify for a drivers license would not be eligible to serve on juries. According to Hardy, Texas’ drivers license selection scheme so far has proven to be largely ineffective primarily because persons who have felony convictions or are not United States citizen——qualities making them ineligible for jury service——routinely are included on the list used by the County Clerk to compose the jury wheel. Conversely, with the one obvious exception that citizens who are unable to read, write, and speak English can vote but cannot serve as jurors, the qualifications needed to vote and to serve on juries are substantially similar. Although arguably under certain circumstances it may be appropriate to supplement voter registration lists with some other source, such as drivers license lists, in order to protect the rights secured by the Jury Act, this is plainly not such a case. Garcia does not allege, nor is there any evidence tending to show, that the jury list is selected in a nonrandom manner or that there has been any finagling at any stage of the selection process. The district court in essence found that this was not the situation, and nothing in the record suggests error in that finding. For these reasons and the reasons stated earlier, we conclude that 15 Garcia’s jury selection challenges fail on both Sixth Amendment and Jury Act grounds.15 2. Mata’s Agreement With the State Garcia claims that his due process rights were violated by the State’s use of Mata, who was an informant and testified as a government witness at Garcia’s trial. Garcia insists that the “contingency fee” agreement resulting in Mata’s testimony offended the Due Process Clause and that the district court erred by denying his motion to dismiss the indictment or suppress evidence gathered by Mata. In support of his argument, Garcia relies chiefly on Williamson v. United States, 311 F.2d 441 (5th Cir. 1962), cert. denied, 85 S.Ct. 1803 (1965), in which this Court established a per 15 In the district court below, Garcia objected to the jury selection process on Fifth Amendment grounds in addition to his Sixth Amendment and Jury Act challenges. On appeal, however, he challenges the selection process exclusively on Sixth Amendment and Jury Act grounds. His failure to brief his Fifth Amendment challenge constitutes a waiver of that argument. See Graef v. Chemical Leaman Corp., 106 F.3d 112, 115 n.2 (5th Cir. 1997). However, even if we were to consider the merits of a Fifth Amendment challenge to the jury selection process, we would nevertheless conclude that such a claim would fail, as Garcia has offered nothing but statistics to support his allegations of discrimination, and these statistics are alone insufficient to support a claim of intentional discrimination under the Fifth Amendment. See Brummitt, 665 F.2d at 527 (stating that “[a] prima facie case of discrimination cannot rest merely on statistics. The fact that an identifiable minority group votes in a proportion lower than the rest of the population and is therefore underrepresented on jury panels presents no constitutional issues”) (citations omitted). Moreover, Garcia has provided no evidence of discriminatory intent in the selection process, and we regard the lower court’s finding that the selection process is “unbiased” and “racially and ethnically neutral” as well supported and not even approaching clearly erroneous. 16 se rule that an informant paid a contingency fee is not a competent witness. There are numerous problems with Garcia’s argument. First, Mata’s agreement was with the State of Texas, not with the federal authorities who prosecuted Garcia. Indeed, Mata had no agreement whatsoever with the federal government. Second, Mata’s contingency fee agreement consisted of neither a contingency nor a fee. Under the agreement, the State agreed not to revoke Mata’s parole, to dismiss his marihuana charge, and to pay for relocation expenses, which totaled approximately $3,500, in exchange for Mata’s assistance in the successful “arrest and charging” of Garcia.16 The benefits of the agreement were not contingent upon the successful prosecution of Garcia. Also, this agreement did not involve payment of any fees as consideration for Mata’s services; rather, the consideration was for the State to drop the felony charge and not revoke his parole. The only fees received by Mata——the relocation expenses——were not paid as consideration for his services and certainly were not contingent upon the government’s getting a conviction against Garcia, as evidenced by the fact that Mata was paid prior to Garcia’s trial. Lastly, Garcia faces one final insurmountable hurdle: in United States v. Cervantes-Pacheo, 826 F.2d 310 (5th Cir. 1987) (en banc), cert. denied, 108 S.Ct. 749 (1988), we overruled Williamson, 16 Mata was relocated due to threats to his safety. 17 holding that “the credibility of the compensated witness, like that of the witness promised a reduced sentence, is for a properly instructed jury to determine. Accordingly, we overrule Williamson and its per se exclusionary rule.” Id. at 316. As we explained in Cervantes-Pacheo, an informant who is promised a contingency fee or other benefits is not automatically excluded from testifying at trial; rather, the matter should be left to the jury to consider in weighing the credibility of the witness-informant. Id. at 315. Certain procedural safeguards, however, must be employed in connection with the use of the informant’s testimony: “The government must not use or encourage the use of perjured testimony; the government must completely and timely disclose the fee arrangement to the accused . . .; the accused must be given an adequate opportunity to cross-examine the informant and government agents about any agreement to compensate the witness; and the trial court should give a special jury instruction pointing out a suspect credibility of paid witnesses.” United States v. Bermea, 30 F.3d 1539, 1552 (5th Cir. 1994), cert. denied, 115 S.Ct. 1113 (1995). See also United States v. Pruneda-Gonzalez, 953 F.2d 190, 197-98 (5th Cir.), cert. denied, 112 S.Ct. 2952 (1992). In the case sub judice, all of these procedural safeguards have been satisfied. Garcia does not claim nor does the evidence show that the government knowingly used or encouraged perjured testimony; the government timely disclosed the agreement to the defense; Garcia thoroughly cross-examined Mata at trial about his agreement with the State; and the district court cautioned the jury 18 with a special instruction (as to which no complaint is made) concerning Mata’s suspect credibility as a compensated or immunized government informant. Reversal is not warranted based on Mata’s involvement in this case. 3. Sufficiency of the Evidence Next, Garcia argues that the evidence was insufficient to support his conviction for conspiracy. Specifically, he contends that he had no discussions with any of his codefendants, except for Alcantar; the only evidence of those discussions was “puffing” on his taped calls with Mata or between Mata and Alcantar; in the days preceding the attempted robbery, he repeatedly failed to return calls from Mata and Alcantar; and he did not go with his codefendants to the warehouse. In considering Garcia’s sufficiency of the evidence claim, we view the evidence presented and all inferences that may be drawn therefrom in the light most favorable to the jury’s verdict, and decide whether any rational trier of fact could have found each element of the crime beyond a reasonable doubt. United States v. Resio-Trejo, 45 F.3d 907, 910-11 & n.6 (5th Cir. 1995). In a drug conspiracy prosecution under 21 U.S.C. § 846, the government must prove beyond a reasonable doubt (1) the existence of an agreement between two or more persons to violate the narcotics laws; (2) that the defendant knew of the agreement; and (3) that the defendant voluntarily participated in the agreement. United States v. 19 Limones, 8 F.3d 1004, 1009 (5th Cir. 1993), cert. denied, 114 S.Ct. 1562 (1994); United States v. Maltos, 985 F.2d 743, 746 (5th Cir. 1992). A conspiracy can be established by circumstantial evidence, that is, the jury frequently may infer its existence from a defendant’s concert of action with others. United States v. Cardenas, 9 F.3d 1139, 1157 (5th Cir. 1993), cert. denied, 114 S.Ct. 2150 (1994). The agreement and the defendant’s knowledge and participation in the conspiracy may be inferred from the “development and collocation of circumstances.” Maltos, 985 F.2d at 746 (internal quotations and citation omitted). The evidence shows, primarily through taped conversations and Mata’s testimony, that Garcia repeatedly agreed to enter into the conspiracy to steal amounts of marihuana ranging from 1,000 to 1,400 pounds; that Garcia recruited Alcantar to assist in the scam; that Garcia scheduled the meeting between Alcantar and the seller; that Garcia conferred with Alcantar prior to going to the warehouse to inspect the marihuana; and that Garcia continued to show interest in the days immediately before the attempted robbery.17 Viewing the evidence and all reasonable inferences therefrom in the light most favorable to the verdict, we hold that there is amply sufficient evidence for a reasonable jury to conclude that Garcia actively participated in the conspiracy. 17 Other than through cross-examination of government witnesses, the defense presented no evidence apart from a stipulation that there was an outstanding arrest warrant for Mata. 20 4. Withdrawal Defense Garcia next contends that the district court erred by refusing to give a correct jury instruction on his withdrawal defense. He complains that the instruction given unnecessarily narrowed the jury’s consideration of activities that could have constituted withdrawal. Also, he argues that the district court erred by denying his request that the jury be instructed that if he proved he withdrew from the conspiracy prior to the robbery he should be found not guilty. The actual instruction, according to Garcia, left open the possibility that the jury could have found him guilty even if they believed that he withdrew from the conspiracy. We review the lower court’s refusal to give a requested jury instruction for abuse of discretion. United States v. Branch, 91 F.3d 699, 711 (5th Cir. 1996), cert. denied, 117 S.Ct. 1466 (1997). We may reverse only if the requested instruction is substantially correct, was not substantially covered in the charge as a whole, and concerns an important point in the trial, the omission of which materially impaired the defendant’s ability to present a given defense effectively. United States v. Tannehill, 49 F.3d 1049, 1057-58 (5th Cir.), cert. denied, 116 S.Ct. 167 (1995). The instruction given by the court on withdrawal was a correct statement of the law and appears to have provided an adequate basis for the defendant to argue that he was entitled to be found not guilty based on that defense. Further, the instruction immediately 21 preceding the withdrawal instruction directed the jury that the defendant must be a member of the conspiracy at the time that the offense is committed in order to be responsible for the offense. Hence, the jury was instructed that if it determined that Garcia had withdrawn from the conspiracy at the time that the offense was committed, it could not find him guilty of the conspiracy offense. Furthermore, there was no evidence of withdrawal, and certainly not of withdrawal before any overt act in furtherance of the conspiracy. United States v. Nicoll, 664 F.2d 1308, 1315-16 (5th Cir.), cert. denied, 102 S.Ct. 2929 (1982). The district court did not reversibly err in refusing to give the requested instruction. 5. Multiple Conspiracy Garcia argues that the district court erred in not giving the jury his requested instruction on multiple conspiracies. He asserts that the evidence “arguably” supports the existence of two conspiracies, the first occurring before July 20 and the second being the July 21 attempted robbery. Garcia contends that his requested instruction, which stated that “to prove Count One, the prosecution must prove beyond a reasonable doubt the conspiracy alleged in the indictment . . . [and not a] different conspiracy,” should have been given to the jury. In considering whether a district court properly refused such 22 a requested jury instruction, this Court determines whether the requested instruction was supported by the evidence. United States v. Asibor, 109 F.3d 1023, 1036 (5th Cir. 1997). In determining whether a multiple conspiracy instruction should be given, we consider “the times, places, persons, offenses charged, and the overt acts involved.” Id. (internal quotations and citations omitted). The evidence shows that Garcia and Mata had several conversations in which Garcia expressed his intent to steal the entire load of marihuana from the warehouse, and although certain minor details of the drug heist were unsettled, throughout the entire conspiracy the single primary purpose of the conspiracy never changed. Alcantar and the other codefendants all fully participated in this same conspiracy, and nothing in the record indicates that any of these codefendants were operating under a different conspiracy. Accordingly, the district court did not err in refusing to give the requested instruction because the evidence does not reflect the existence of more than one conspiracy. 6. Sentencing Challenges Garcia’s final argument on appeal is a challenge to his sentence on three separate grounds. First, he argues that the district court erred in determining the quantity of marihuana attributable to him, as various quantities of marihuana were discussed during the taped conversations ranging from 200 to 1,400 pounds. He contends that he should not be held accountable for the 1,500 pounds anticipated to be obtained during the July 21 robbery 23 attempt, but instead should be accountable for only 600-800 pounds. This Court reviews a district court’s factual findings concerning the quantity of drugs attributable to a defendant for clear error. United States v. Maseratti, 1 F.3d 330, 340 (5th Cir. 1993), cert. denied, 114 S.Ct. 1096 (1994). A factual finding is not clearly erroneous if it is plausible in light of the record of the case as a whole. Id. During a taped telephone conversation 2 days before the robbery, Mata told Garcia that the seller had brought in an additional 800 pounds of marihuana and that the seller now had 1,400 pounds of marihuana. Garcia responded that was fine and later that evening indicated that he had everything ready to go for the next day. Thus, there was reliable evidence in the record to support a finding that Garcia was personally aware that the conspiracy involved 1,400 pounds of marihuana. The 100 pound difference was inconsequential with respect to the calculation of Garcia’s base offense level. See U.S.S.G. § 2D1.1(c)(6). That is, the same guideline range would have been produced by 1,400 (or 1,000) pounds as by 1,500 pounds. The court’s finding was not clearly erroneous. Second, Garcia contends that the district court erred in denying his objection to the two-level upward adjustment for obstruction of justice based on Mata’s allegation that on July 24, a few days after the attempted robbery, Garcia pointed a pistol at him and shouted a threat. Garcia asserts that he never threatened Mata with the pistol. 24 The district court’s determination that a defendant obstructed justice within the meaning of U.S.S.G. § 3C1.1 is reviewed for clear error. United States v. Bethley, 973 F.2d 396, 402 (5th Cir. 1992), cert. denied, 113 S.Ct. 1323 (1993). In addition to hearing the testimony of Mata at trial concerning the alleged threat, the district court also heard the testimony of Garcia’s wife at the sentencing hearing, who stated that she was with Garcia during the alleged incident and that he never pointed the gun or made any threats towards Mata. The court ruled against Garcia, stating that it was making a credibility determination based on what it had observed at trial and at the sentencing hearing. There is no basis for finding that the district court’s credibility determination was clearly erroneous. United States v. McAfee, 8 F.3d 1010, 1018 (5th Cir. 1993). Thus, the district court did not clearly err in making the adjustment for obstruction of justice. Finally, Garcia complains that the district court erred in denying his objection to the four-level upward adjustment for his role as an organizer in criminal activity involving five or more participants. He contends that he did not exercise decision-making authority such as setting the quantity and prices and that his only participation, besides his telephone conversations with Mata, was his introduction of Alcantar to Mata. He also argues that he did not recruit the other codefendants and there was no evidence that he was involved in the planning of the robbery. This Court reviews a district court’s determination that a 25 defendant held a supervisory role in an offense under U.S.S.G. § 3B1.1 for clear error. United States v. Musquiz, 45 F.3d 927, 932- 33 (5th Cir.), cert. denied, 116 S.Ct. 54 (1995). A defendant’s offense level is increased by four levels if the defendant was an organizer or leader of any criminal activity involving five or more participants or was otherwise extensive. U.S.S.G. § 3B1.1(a). Proof that the defendant supervised only one other culpable participant is sufficient to make the defendant eligible for this enhancement. United States v. Washington, 44 F.3d 1271, 1281 (5th Cir.), cert. denied, 115 S.Ct. 2011 (1995). The evidence shows that Garcia recruited Alcantar to carry out the scheme and continued to oversee his activities. Alcantar recruited at least six other men to participate in the robbery. The evidence was sufficient to support the district court’s imposition of the adjustment. Conclusion Based on the foregoing, we AFFIRM Garcia’s conviction and sentence. 26