UNITED STATES OF AMERICA, APPELLEE v. TARKARA COOPER AND BRIAN BRYANT, APPELLANTS
No. 17-3057
United States Court of Appeals FOR THE DISTRICT OF COLUMBIA CIRCUIT
February 11, 2020
Argued October 28, 2019
Consolidated with 18-3021
Appeal from the United States District Court for the District of Columbia (No. 1:15-cr-00152-3) (No. 1:15-cr-00152-5)
Dennis M. Hart, appointed by the Court, argued the cause and filed the briefs for appellant Tarkara Cooper.
Brian J. Young, appointed by the Court, argued the cause and filed the briefs for appellant Brian Bryant.
Katherine M. Kelly, Assistant U.S. Attorney, argued the cause for appellee. With her on the brief were Jessie K. Liu, U.S. Attorney, and Elizabeth Trosman and John P. Mannarino, Assistant U.S. Attorneys.
Before: SRINIVASAN and RAO, Circuit Judges, and RANDOLPH, Senior Circuit Judge.
Opinion for the Court filed by Senior Circuit Judge RANDOLPH.
Antonio Cooper, a bus driver and the hub in the wheel in this conspiracy, entered a plea of guilty and testified for the prosecution against others, including his niece, Tarkara Cooper. The jury convicted her and Brian Bryant of theft of public money and conspiracy to defraud the United States. See
The evidence showed that Antonio Cooper gathered names, birth dates and Social Security numbers from friends, family members, strangers, and “wherever [he] could.” Using this information, he filed more than a thousand fraudulent federal income tax returns seeking refunds. The
For his part, Bryant helped Mr. Cooper cash the refund checks and received half of the face value of those checks in exchange. Brianna Turner, a former Bank of America employee, testified that she helped Bryant deposit some of these checks. She said that Bryant expressed a willingness to compensate her in exchange for violating the Bank‘s rules on third-party deposits. Bryant paid Turner between $300-$600 per check deposited. He controlled dozens of bank accounts into which refund checks were deposited.
At some point in 2010, Postal Inspector Maria Couvillion opened an investigation after detecting what appeared to be fraudulent tax returns being sent through the mail. Not until six years later did a grand jury return indictments against Antonio Cooper, Tarkara Cooper, Bryant and others. In the interim, the conspirators carried on, filing more fake returns and cashing more refund checks.
I.
We take up first the district court‘s rejection of Tarkara Cooper‘s pretrial motion to suppress statements she made to federal agents. The evidence at the suppression hearing showed that before dawn on December 1, 2010, twelve law enforcement officers – eleven federal agents and one from the local police arrived at the door of Cooper‘s home in a “multi unit residential complex” in the District of Columbia.1 The officers were there to execute a search warrant. An agent knocked and announced their presence. Cooper opened the door and some of the agents entered. Inside were Cooper‘s six-to-seven year old daughter and an adult female friend of Cooper‘s. After being interviewed, the friend left.
About an hour into the search, two agents began interviewing Cooper in her living room. At some point in the questioning, she admitted receiving in the mail at her address one to five refund checks each week, and that her uncle paid her when she turned over the checks to him. She argues here, as she did in the district court, that the government could not introduce these statements at trial because the agents did not first give her Miranda warnings. See Miranda v. Arizona, 384 U.S. 436 (1966).
The district court denied Cooper‘s motion. The court found that Cooper “was not coerced into answering questions” and that
she was not in custody within the meaning of Miranda. At trial, an agent who interrogated Cooper testified about her statements.
A suspect is “in custody” if the circumstances of the questioning “present a serious danger of coercion.” Howes v. Fields, 565 U.S. 499, 508-09 (2012). To determine whether such a danger existed, courts first consider whether a reasonable person in the suspect‘s position would have felt that “she was not at liberty to terminate the interrogation and leave.” Id. at 509 (internal quotation marks omitted). Relevant factors in this assessment include the location of the questioning, statements made during the interview, the presence of any physical restraints, and whether the interviewee was released once the interrogation ended. Id.
Because “[n]ot all restraints on freedom of movement amount to custody for purposes of Miranda,” a finding that a person in the suspect‘s shoes would not have felt free to leave does not end the inquiry. Id. Rather, courts must then ask “the additional question whether the relevant environment presents the same inherently coercive pressures as the type of station house questioning at issue in Miranda.” Id. We review de novo the district court‘s custody determination, and we review the underlying factual findings for clear error. United States v. Brinson-Scott, 714 F.3d 616, 621 (D.C. Cir. 2013).
Examination of the evidence before the district court at the suppression hearing and at trial reveals no clearly erroneous factual findings.2 Instead, the evidence amply supports the conclusion that Cooper was not in custody when she admitted that she was aware of, and participated in, her uncle‘s fraud.
The agents questioned Cooper in her living room. When an interview takes place in a suspect‘s home, that circumstance usually weighs against finding the kind of custodial situation that merits a Miranda warning. See, e.g., Beckwith v. United States, 425 U.S. 341, 342, 347-48 (1976); United States v. Luck, 852 F.3d 615, 621 (6th Cir. 2017); 2 Wayne R. LaFave et al., Criminal Procedure § 6.6(e) (4th ed. 2014).
Before the interview began, the agents told Cooper that she was a subject of an investigation and described “the voluntary nature of the interview.” They asked her “if she would agree to” answer their questions. She agreed.3 The evidence shows, in other words, that Cooper‘s statements were given freely and voluntarily. Such statements “remain a proper element in
In addition, no weapons were brandished and no handcuffs were used. Cooper was “cooperative,” and the agents employed
a “professional and cordial tone.” At no point did Cooper ask to end the questioning. And once the interview was over, the agents left without arresting her. Put simply, these facts do not portray an environment presenting a meaningful danger of coercion.
Cooper‘s argument to the contrary focuses on a fifteen-minute break during the questioning. She needed to take her daughter to school, and so the agents drove them to the school, dropped her daughter off, and returned to the house with Cooper to resume the interview. Cooper alleges that the agents refused to let her leave the house unaccompanied, and that this refusal illustrates the custodial nature of the questioning.
Not so. Cooper offers no evidence that she asked to take her daughter to school alone but was prevented from leaving. For safety reasons, if a person leaves her residence during the execution of a search warrant, she is not typically allowed to return until the search is completed. It is thus as likely as not that the agents gave Cooper a ride so that she could return to her house while the search was ongoing. And while she was in the agents’ car, Cooper was not asked any questions.
These circumstances do not amount to custody within the meaning of Miranda. Indeed, to hold otherwise would be akin to suggesting that voluntary statements uttered during the execution of a valid search warrant are inadmissible at trial absent the issuance of a Miranda warning. The Supreme Court has rejected such an expansive view. See Andresen v. Maryland, 427 U.S. 463, 475 (1976). The district court thus correctly concluded that Cooper‘s statements could be used at trial.
II.
Cooper and Bryant also challenge the use of Special Agent LaRose as a summary witness. Before trial, the government asked the court to impose what it called the “Rule on Witnesses.” Although the government did not mention it,
Both IRS agents attended the entire trial. Milne sat at counsel table. LaRose sat in the audience. The prosecution apparently decided not to use Milne as a witness. The prosecution called LaRose toward the end of its case-in-chief
without attempting to qualify him as an expert witness under
LaRose‘s lengthy testimony on direct examination dealt with extensive and detailed charts listing bank records, tax returns, refund checks, and more connecting the defendants with the tax fraud. He testified, for example, that after December 1, 2010, when the agents searched Cooper‘s residence, 65 refund checks totaling $151,853.24 were mailed to her home. Of those, 59 checks were negotiated, totaling $141,333.49.
Cooper and Bryant complain that the district court erred in allowing LaRose to testify about matters other than those contained in the charts. Their complaint greatly exaggerates the record.
A prosecutor‘s “non-expert summary witness can help the jury organize and evaluate evidence which is factually complex.” United States v. Lemire, 720 F.2d 1327, 1348 (D.C. Cir. 1983). Putting it in terms of “helping the jury” may suggest that prosecutors call summary witnesses for some neutral, educational trial purpose. That of course is not true. Prosecutors call witnesses, including summary witnesses, to prove their case – to help convince the jury that the defendant is guilty as charged.
It follows that there must be limitations on the sort of “help” a government summary witness may provide. One of the most important is this: the witness may not usurp the jury‘s fact-finding function by summarizing or describing not only what is in evidence but also what inferences should be drawn from that evidence. United States v. Hampton, 718 F.3d 978, 983 (D.C. Cir. 2013). Another danger to be guarded against is that the jury will treat summary testimony “as additional evidence or as corroborative of the truth.” Lemire, 720 F.2d at 1348. See also Lauren Weiser, Requirements for Admitting Summary Testimony of Government Agents in Federal White Collar Cases, 36 AM. J. CRIM. L. 179, 181-82 (2009).
LaRose, the defendants claim, crossed these lines. They argue that the government functionally introduced the agent as though he were an expert witness by reviewing at length his training and professional background. LaRose then appeared to present an expert opinion, suggesting that based on his fifteen years of experience as an investigator, it is not uncommon
The defendants also argue that LaRose discussed evidence that was, at the time of his testimony, not yet admitted. He testified before two of the government‘s witnesses, Chyna Watkins and Vincent Short, took the stand. Watkins was Bryant‘s girlfriend, and she told the jury that she “gave him access to her bank account and allowed him to deposit thousands of dollars into it. She later withdrew the money for him in exchange for a series of small payments. Short was one of the scheme‘s victims. He testified that he lost his identification card, that he did not recognize a tax return filed in his name listing Cooper‘s address, and that he did not sign a check made payable to Bryant that appeared to contain his signature.
Before Watkins and Short offered this testimony, LaRose stated that Watkins‘s account was under Bryant‘s control. He also referred to Short‘s “purported” signature on the check to Bryant. According to the defendants, by doing so he improperly bolstered the credibility of Watkins‘s and Short‘s subsequent testimony.
Cooper and Bryant also take issue with LaRose‘s count-by-count synthesis of the government‘s evidence against them. They argue that the government used the agent to present conclusory testimony that effectively served as a second closing argument.
But even if the district court erred in allowing LaRose to stray beyond a narrow summary of the evidence reflected in the charts, Cooper and Bryant have not shown that they were substantially prejudiced by this error. See United States v. Miller, 738 F.3d 361, 372 (D.C. Cir. 2013); Kotteakos v. United States, 328 U.S. 750, 776 (1946). LaRose‘s testimony was extensive, consuming 150 pages of transcript. The defendants point to a minuscule portion of what he said. The alleged errors
were, in other words, quite minor. Without LaRose‘s testimony, the record still contains overwhelming evidence of the defendants’ guilt. Any error was thus harmless.
Consider the evidence against Cooper. She confessed that she had an agreement with her uncle to receive payment in exchange for giving him the checks that were mailed to her home. The hundreds of checks mailed to her address amply support the finding that she was a knowing participant in the scheme. Her uncle also testified that he told Cooper the tax checks she was receiving were not “legit.” She “knew about everything.” In addition, Cooper was identified in video surveillance footage interacting with a bank teller who helped the participants perpetrate the fraud. The strength of this evidence makes it very unlikely that the portion of LaRose‘s testimony Cooper complains about materially influenced the jury‘s verdict.
The evidence against Bryant was also compelling. Antonio Cooper testified that Bryant helped him cash the refund checks and received half of the face value of those checks in exchange. Brianna Turner, the former bank teller, admitted that she helped Bryant deposit some of these checks in exchange for compensation.
The government presented evidence of multiple bank accounts controlled by Bryant into which dozens of refund checks were deposited. For example, Bryant deposited 26 refund checks totaling more
Further minimizing the effect of any possible error was the district court‘s limiting instruction, which informed the jury that LaRose‘s summaries “are not in and of themselves proof of the facts,” and that his testimony was to be used “only as a matter
of convenience.” This instruction, virtually identical to the one issued by the trial court in Lemire, helped lower the risk that the jury would take the summary testimony for more than it was worth. See Lemire, 720 F.2d at 1348 n.32. Juries “are presumed to follow their instructions,” and the record offers no reason to doubt the validity of that presumption here. Zafiro v. United States, 506 U.S. 534, 540 (1993) (internal quotation marks omitted).
More still, the defendants had an opportunity to cross-examine LaRose. They did so, eliciting testimony that LaRose is not a hand-writing expert and that he lacked personal knowledge about whether Bryant controlled Watkins‘s bank account. In sum, Watkins‘s and Short‘s testimony, as well as the record evidence, fully corroborated LaRose‘s summary of the investigation. We therefore find that his testimony did not meaningfully prejudice the defense.
III.
We have reviewed and reject the defendants’ remaining arguments. Cooper moved for a mistrial during the government‘s rebuttal. She alleged that the prosecutor improperly lumped her together with co-defendants whose participation in the scheme was much more substantial. The district court denied this motion. It did not err by doing so. The prosecutor did not misstate the evidence applicable to Cooper, and the occasional inadvertent references to “these defendants” when discussing acts not attributable to Cooper were quickly remedied. See United States v. Gartmon, 146 F.3d 1015, 1026 (D.C. Cir. 1998).
The defendants also challenge the sentences the district court imposed, suggesting that the loss amounts attributed to Cooper and Bryant were too high. The court attributed a loss to
the Treasury of about $650,000 to Bryant‘s participation in the scheme, and a loss of around $4.6 million to Cooper‘s participation. These findings of fact must be upheld unless they are clearly erroneous. United States v. Day, 524 F.3d 1361, 1367 (D.C. Cir. 2008). They are not.
The value of the bank accounts reasonably connected to Bryant‘s participation in the scheme was at least $528,252.94. And the government presented evidence of checks attributable to co-conspirators’ home addresses that were not attributed to particular bank accounts. The value of such checks was well over a million dollars, and at least some are reasonably connected to Bryant‘s role in the fraud.
Similarly, the amount attributed to Cooper—the entire intended loss attributable to her uncle‘s scheme was not clearly erroneous. See United States v. Seiler, 348 F.3d 265, 268 (D.C. Cir. 2003) (noting that, in determining a sentence in a conspiracy case, the district court takes into account “all reasonably foreseeable acts and omissions” committed by others in furtherance of the conspiracy) (internal quotation marks omitted). Because, as the district court noted, Cooper‘s participation in the scheme was “neither minor nor minimal,” it was not clear error to conclude that her co-conspirators’ actions were reasonably foreseeable to her.
Cooper also challenges the restitution the district court ordered her to pay.
For these reasons, the defendants’ convictions and the sentences they received are affirmed.
So ordered.
