Ayestas v. Davis

LII note: The U.S. Supreme Court has now decided Ayestas v. Davis .


Is the Fifth Circuit’s “substantial need” test for awarding investigative resources to indigent defendants consistent with the requirements of 18 U.S.C. § 3599(f)?

Oral argument: 
October 30, 2017

Carlos Manuel Ayestas brought a state habeas petition after he was sentenced to death in Texas for murder. After the state denied his petition in 2008, Ayestas petitioned for federal habeas relief, alleging that he received ineffective assistance of counsel under the Sixth Amendment. Ayestas requested funding for “investigative, expert, or other services” under 18 U.S.C. § 3599(f) to help support his claim of ineffective assistance of counsel. Both the district court and the Fifth Circuit Court of Appeals dismissed Ayestas’s claim and denied his § 3599(f) motion, finding that he had not demonstrated a “substantial need” for investigative assistance. Ayestas now challenges this substantial need test on appeal, arguing that it is inconsistent with the text, history, and purpose of § 3599(f). The Director of the Correctional Institutions Division of the Texas Department of Criminal Justice, Lorie Davis, on the other hand, argues the test is proper in light of the requirements of the Antiterrorism and Effective Death Penalty Act. This case will allow the court to determine the appropriate statutory interpretation of § 3599(f), as well as its applicability to federal habeas proceedings. The case could have significant consequences for the resources available to capital defendants bringing ineffective assistance of counsel claims.

Questions as Framed for the Court by the Parties 

Whether the Fifth Circuit erred in holding that 18 U.S.C. § 3599(f) withholds “reasonably necessary” resources to investigate and develop an ineffective-assistance-of-counsel claim that state habeas counsel forfeited, where the claimant’s existing evidence does not meet the ultimate burden of proof at the time the § 3599(f) motion is made.


Petitioner Carlos Manuel Ayestas was convicted and sentenced to death for murder in 1995. After the Texas Court of Criminal Appeals affirmed his conviction and sentence, Ayestas petitioned for state habeas relief; a request that included a claim of ineffective assistance of trial counsel. The ineffective assistance of counsel claim rested on the fact that Ayestas’s trial attorneys had failed to contact his family, who could testify to his good character. The Texas state court denied Ayestas’s request for habeas relief, and the Texas Court of Criminal Appeals affirmed the denial in 2008.

In 2009, Ayestas obtained new counsel and applied for federal habeas relief under 28 U.S.C. § 2254. Ayestas’s habeas claim asserted that his trial counsel had been ineffective due to his failure to reasonably investigate potentially mitigating evidence. Ayestas claimed that his trial counsel had failed to investigate and present evidence of his struggles with schizophrenia and substance abuse. The district court held that by failing to raise this claim in his earlier state habeas proceeding, Ayestas had procedurally defaulted on it—forfeiting his right to argue it later. Because there was nothing that had prevented Ayestas’s state habeas attorneys from raising the ineffective assistance of counsel claim, the court did not excuse his default. In 2012, the United States Court of Appeals for the Fifth Circuit denied Ayestas’s request for a certificate of appealability, which would have allowed him to appeal the procedural default.

After the Supreme Court held in Martinez v. Ryan that a state habeas counsel’s failure to claim ineffective assistance of counsel may excuse a procedural default, Ayestas filed a motion with the Fifth Circuit requesting that it vacate the prior denial of appealability. The Fifth Circuit denied the motion and held that Martinez did not apply in Texas. Soon after, in Trevino v. Thaler, the Supreme Court explicitly extended Martinez to Texas and consequently vacated and remanded Ayestas’s case to the Fifth Circuit for consideration in light of Trevino. The Fifth Circuit then remanded to the district court, instructing it to reconsider Ayestas’s ineffective assistance of counsel claim.

On remand, Ayestas requested investigative assistance under 18 U.S.C. § 3599(f) to refine his ineffective assistance of counsel claim. In November 2014, the district court denied Ayestas § 3599(f) investigative resources, habeas relief, and a certificate of appealability. The district court held that, because neither Ayestas’s trial counsel nor his state habeas counsel had been ineffective, his ineffective assistance of counsel claim remained procedurally defaulted, and therefore investigative resources were not “reasonably necessary” as required by § 3599(f).

In 2015, before the Fifth Circuit, Ayestas argued that § 3599(f) entitled him to funds that would enable a mitigation specialist to develop his ineffective assistance of counsel claim by investigating his history of mental illness and substance abuse. The Fifth Circuit rejected this argument and affirmed the district court, agreeing that his trial counsel and state habeas counsel were not ineffective for failing to explore particular arguments concerning Ayestas’s mental health. The Fifth Circuit also reinforced its requirement that defendants show a “substantial need” for mitigation specialist assistance. Ayestas then petitioned the Supreme Court, which agreed to hear whether the Fifth Circuit had erred in withholding “reasonably necessary” § 3599(f) investigative resources to develop his ineffective assistance of counsel claim.



Ayestas argues that the Fifth Circuit’s substantial need test is incompatible with the text, history and purpose of § 3599(f). Section 3599(f) provides funding for “investigative, expert, or other services” that are “reasonably necessary” to represent criminal defendants. Ayestas claims that, contrary to the statutory language of “reasonably necessary,” the Fifth Circuit’s “substantial need” test denies defendants funding for representation-related services if it determines that their claim is not viable on the merits. Reading this “substantial need” requirement into § 3599(f), Ayestas contends, imposes a significant additional burden on capital defendants seeking funding under § 3599(f): it effectively requires them to demonstrate the merit and viability of their claims before receiving the funding needed to investigate those very claims. Ayestas also argues that the substantial need test contravenes Congress’ intent to create a statutory scheme that would improve the quality of representation available to capital defendants. Ayestas points to § 3599(f)’s legislative history to support this argument, noting that it grew out of the Criminal Justice Act. Ayestas argues that courts consistently interpreted the Criminal Justice Act’s funding provisions according to a reasonable necessity standard, which Congress then incorporated into § 3599(f) in 1988. Ayestas maintains that, unlike the substantial need test, the reasonable necessity standard did not require defendants to establish the viability of their claims in order to obtain financial assistance to investigate those claims. Finally, Ayestas points out the detrimental effects of the substantial need test on capital defendants: it invites courts to speculate on the merits and viability of a claim before the defendant has access to the funding necessary to properly develop it. According to Ayestas, the substantial need test puts § 3599(f) movants between a rock and hard place, forcing them to show the court that there is relevant evidence they could discover without having the resources to do so.

Ayestas argues that he, specifically, is entitled to funding for services under § 3599(f) because they are reasonably necessary to his representation. Ayestas contends that any reasonable lawyer would have investigated his history of mental illness and substance abuse to produce potentially mitigating evidence at trial. To that end, Ayestas points out that his narrow § 3599(f) motion requested only those services necessary to make these determinations. Further, Ayestas explains that his state habeas counsel ignored the advice of a mitigation specialist who recommended a thorough investigation into his personal background.

Respondent Lorie Davis (“Davis”), the Director of the Correctional Institutions Division of the Texas Department of Criminal Justice, counters that the substantial need test properly accounts for the limitations that the Anti-Terrorism and Effective Death Penalty Act (“AEDPA”) places on § 3599(f) funding in the context of federal habeas proceedings. Davis asserts that § 2254(e)(2) of AEDPA bars the defendant from introducing evidence in a federal habeas proceeding that was not already included in the state court record. Davis therefore concludes that § 3599(f) funding is never available to investigate such evidence. Davis further notes that when Congress passed AEDPA in 1996, it amended the statutory language in § 3599(f)’s precursor to give greater discretion to judges over funding decisions. Instead of ordering that judges “shall authorize” funding, Davis points out that Congress stated that judges “may authorize” it. Davis also asserts that circuit courts routinely deny funding to capital defendants based on a preliminary assessment of the merits of their claim, which, she observes, avoids channeling funds to frivolous investigations. In contrast, Davis argues that Ayestas’s proposed standard for § 3599(f) provides no limit on which habeas petitioners may obtain funding.

Thus, Davis suggests that Ayestas’s approach transforms the statutory scheme into an entitlement program for capital defendants to pursue evidence that may not exist.

Davis disputes the relevance of the evidence that Ayestas seeks to uncover. Davis observes that Ayestas’s post-trial diagnosis of schizophrenia does not demonstrate that his trial counsel should have known to investigate the possibility of mental illness during the trial process. Further, Davis asserts that Ayestas has not shown that such evidence would have changed the outcome of the trial. Davis argues that the evidence of substance abuse and mental illness that Ayestas claims an investigation will unearth might have a prejudicial rather than mitigating effect. In effect, Davis contends that Ayestas has not shown a substantial likelihood that the jury would have given him a life sentence rather than the death penalty had they been aware of his substance abuse issues and mental illness, especially given the violence of his crime.


Davis contends that the Supreme Court cannot review the Fifth Circuit’s denial of § 3599(f) funding to Ayestas because the denial is an administrative act which does not fall within the Court’s Article III jurisdiction. Davis notes that a judicial decision constitutes an administrative act when it is non-adversarial and subject to review outside of normal Article III procedures. Here, Davis asserts that the decision to deny funding is non-adversarial because the § 3599(f) motion was made ex parte, involving only Ayestas and the judge. Similarly, Davis observes that § 3599(f) decisions at the district court level are not subject to normal Article III review because a single circuit court judge can override them. Davis characterizes § 3599(f) as a government benefit program for criminal attorneys representing indigent defendants that the judiciary merely happens to administer. From this perspective, Davis argues that any decisions a judge might make about denying or granting funding under § 3599(f) are not reviewable exercises of Article III judicial power.

Ayestas contends that it is within the Supreme Court’s jurisdiction to review § 3599(f) determinations because circuit courts routinely review such determinations and other determinations made under the broader § 3599 provisions . Ayestas points out that the Supreme Court probably would not have accepted this case on the § 3599(f) question presented only to dismiss it for lack of Article III jurisdiction. Moreover, Ayestas argues that § 3599(f) decisions are not purely administrative in nature—and thus fall under the Court’s Article III jurisdiction—because the Fifth Circuit’s § 3599(f) order was an integral part of the final decision in his case, which is inherently an adversary proceeding. Ayestas notes that the State’s decision to contest the allocation of § 3599(f) funding makes the decision an adversarial one, as the State’s objections often extend to the underlying facts that support a request for relief. Thus, Ayestas contends that § 3599(f) asks the Court to apply a statutory standard to the same facts that support the defendant’s legal claim and that this is a quintessential exercise of Article III judicial power.



On behalf of Ayestas, the National Association of Criminal Defense Lawyers (“NACDL”) and the American Civil Liberties Union (“ACLU”) argue that the Fifth Circuit’s “substantial need” test severely limits the funding that Congress intended to provide to habeas petitioners, and thereby undermines the ability of habeas counsel to conduct potentially vindicating investigations. Pointing out that the statutory funding is necessary for indigent defendants to discover relevant evidence and therefore necessary to avoid death sentences, the NACDL and ACLU argue that the decision to award § 3599(f) funding often constitutes a matter of life and death. In support of these arguments, the organizations cite to a number of federal and state habeas cases in which indigent petitioners utilized government-funded investigative resources to develop meritorious claims, including several that were investigated for the first time in federal court. The Fifth Circuit’s “substantial need” test, the NACDL and ACLU therefore claim, prevents the discovery of critical evidence that could support a meritorious habeas claim. Blocking access to § 3599(f) funds may also discourage counsel from taking on habeas cases in the Fifth Circuit, the NACDL and ACLU argue, because counsel may anticipate a lack of resources necessary for investigations and experts. When attorneys do take on habeas cases in the Fifth Circuit, the organizations claim, their ability to effectively represent their clients will be impaired because, they will be forced to devote time and resources to litigating about access to funding for investigation as a threshold issue. If habeas attorneys are unsuccessful in obtaining such funding, the organizations contend, the attorneys will then be forced to personally conduct mitigation investigations, which will not be as successful at producing mitigating evidence because mitigation specialists possess skillsets beyond those of attorneys, and empirical research indicates that defense attorneys are not equipped to conduct effective mitigation investigations.

Arizona and fourteen other States (“the States”), writing in support of Davis, counter that awarding § 3599(f) resources without reference to the underlying merit of the claim produces large amounts of unnecessary litigation and contravenes the federal habeas regime established by Congress in AEDPA. The States argue that two core purposes of AEDPA were to reduce procedural delays in capital cases and ensure that federal courts respect the factual findings of state courts. The statute accomplishes these goals, the States explain, in part by severely restricting what kinds of evidence may be presented in federal habeas cases. According to the States, awarding § 3599(f) resources to investigate claims not raised in state court bypasses AEDPA’s prohibition on this type of evidence and therefore undermines the purposes of the statute. Petitioners such as Ayestas, the States argue, will be able to start new lengthy evidentiary investigations in federal court, effectively allowing them to stretch their capital sentences out into life sentences. The States argue that this would saddle them with an unnecessary financial burden as they are forced to litigate these new evidentiary disputes over defaulted claims in federal court for years or decades.

Edited by 


Additional Resources