What standard should govern the analysis in a habeas proceeding of the government’s knowing or reckless presentation of perjured testimony at a trial? Is it, as the Ninth Circuit has held, simply a materiality determination (using the lower materiality standard for perjury) or is it a two-step process requiring both a materiality determination, followed by a harmless error analysis? The Eighth Circuit opted for the latter formulation in United States v. Clay, issued on August 3, 2013, and, as a result, affirmed the district court’s denial of Clay’s petition. The ruling is a notable one because it concerned perjured testimony by the government’s lead witness on a key contested issue at trial, a fact that prompted the government to offer Clay’s more culpable co-defendant a very favorable plea bargain prior to his severed trial.
Clay was convicted of fraud based on his alleged pocketing of funds earmarked for renovations to be completed by Clay’s construction company, but which were never in fact done. At trial, Clay asserted he had merely rented his construction license to his business partners, Neely (whose trial was severed) and McCuien, now chief government witness, who Clay understood had experience in construction and contracting work and would complete the renovations as required. McQuien, testified, however, that he had never held himself out to the public as a contractor, had never engaged in either construction work or real estate transactions, and that Clay was well aware that no actual renovation work would ever take place. Clay was convicted. Post-trial, evidence emerged that McCuien had perjured himself on this issue (he had in fact extensive experience in construction), and the government subsequently offered a plea deal to Neely of “misprision of a felony” because “the prosecutors had learned that Donny McCuien had lied to them.” Clay filed a habeas petition, asserting in part that his conviction was obtained through the use of perjured testimony from McCuien and that the Government knew, or should have known, that McCuien perjured himself. The district court denied the petition, holding on the false testimony claim that whether or not the government had knowingly used perjured testimony, “Clay could not show prejudice because there was ‘no reasonable likelihood that the perjured testimony would have affected the jury’s verdict.’”
In cases on direct review, the Supreme Court has held that “a conviction obtained by the knowing use of perjured testimony is fundamentally unfair, and must be set aside if there is any reasonable likelihood that the false testimony could have affected the judgment of the jury.” United States v. Agurs, 427 U.S. 97, 103(1976). Clay argued that the district court incorrectly assessed “reasonable likelihood” in this case by examining the sufficiency of the non-tainted evidence. The Eighth Circuit disagreed. Assuming that the government had indeed knowingly or recklessly used perjured testimony by its main witness, it held that the Agurs materiality standard was “only part of the equation.” Here, on habeas review, the government’s use of perjury was a “trial error” to which harmless error analysis applied. This required Clay to establish that the perjury error “had substantial and injurious effect or influence in determining the jury’s verdict.” Clay had failed to meet that burden where “the undisputed circumstantial evidence create[d] an overwhelming impression that Clay was not an innocent pawn in the scheme,” and, in any event, “McCuien was impeached so thoroughly at trial.”
In Hayes v. Brown, 399 F.3d 972 (9th Cir 2005) (en banc), a habeas case similarly involving false testimony that was mentioned but not followed in Clay, the Ninth Circuit held that where the Supreme Court has declared a materiality standard, as in the context of false testimony, ”there is no need to conduct a separate harmless error analysis.” Implicitly, the Ninth Circuit took the position that if there is a reasonable likelihood that a prosecutor’s deliberate elicitation of false testimony affected the jury’s judgment, then, necessarily, the error had a substantial and injurious effect on the verdict. In Hayes, as in Clay, the false testimony came from the prosecutor’s primary witness, whose very presentation represents an affirmation by the prosecutor of his accuracy and reliability, and who is often the lynchpin of the prosecutor’s summation. Notably, in Hayes, the court’s analysis was infused with references to the “special role” occupied by the prosecutor in our criminal justice system:
One of the bedrock principles of principles of our democracy implicitly in any concept of ordered liberty is that the State may not use false evidence to obtain a criminal conviction . . . Indeed, if it is established that the government knowingly permitted the introduction of false testimony, reversal is virtually automatic . . . There is nothing redemptive about the sovereign’s conspiring to deceive a judge and jury to obtain a tainted conviction . . . Our criminal justice system depends on the integrity of the attorneys who present their cases to the jury. When even a single conviction is obtained through perjurious or deceptive means, the entire foundation of our system of justice is weakened.