Ellingburg v. United States: Supreme Court to Hear Ex Post Facto Clause Challenge to Criminal Restitution Statute
September 23, 2025 (LSB11371)

On October 14, 2025, the Supreme Court is scheduled to hear oral argument in Ellingburg v. United States, a case presenting the question whether criminal restitution under the Mandatory Victim Restitution Act (MVRA) is a criminal punishment for purposes of the Constitution's Ex Post Facto Clause. This Legal Sidebar provides background on the Ex Post Facto Clause's prohibition on retroactive punishments, summarizes the litigation in Ellingburg, then discusses considerations for Congress related to the case.

The Ex Post Facto Clause

The Constitution imposes several limits on retroactive legislation. The general theory behind those limits is that people need to know what the law is in order to comply with it, and that is impossible when a law regulates actions taken before it was enacted. Two limitations on retroactive legislation are contained in Article I, Section 9, Clause 3, which provides that "[n]o Bill of Attainder or ex post facto Law shall be passed" by Congress. (Another constitutional provision prohibits the states from enacting bills of attainder and ex post facto laws, among other things. The constitutional prohibitions on bills of attainder and the state Ex Post Facto Clause are not at issue in Ellingburg.)

The phrase "ex post facto," Latin for "after the fact," refers to laws that apply retroactively. A law is deemed to be retroactive if it applies to actions taken before its enactment—the relevant point in time is when the regulated conduct occurred, not when an offense was discovered or when the defendant was arrested, tried, convicted, or sentenced.

While the Ex Post Facto Clause on its face might appear to bar all retroactive legislation, courts have applied the clause only to penal laws. In the 1798 case Calder v. Bull, Justice Samuel Chase stated that the Clause applies to any law that (1) renders criminal an action that was legal when it was taken, (2) makes an existing offense more severe, (3) increases the punishment for an offense, or (4) alters the applicable rules of evidence after a crime was committed to make it easier to convict. In 1990, in Collins v. Youngblood, the Supreme Court summarized the doctrine by explaining, "the constitutional prohibition on ex post facto laws applies only to penal statutes which disadvantage the offender affected by them."

Courts considering whether a law is penal for purposes of the Ex Post Facto Clause look first to the intent of the enacting legislature. If Congress intended for a law to impose punishment, the Clause applies, and any retroactive punishment will be struck down as unconstitutional. When a law is intended to be civil, the question of whether it is actually penal in nature depends on its substance, not its form. Thus, the Supreme Court has explained that "the ex post facto effect of a law cannot be evaded by giving a civil form to that which is essentially criminal." While courts will "ordinarily defer to the legislature's stated intent" to enact civil legislation, they may apply the Ex Post Facto Clause to a statutory scheme intended to be civil upon the "clearest proof" that the scheme is "so punitive either in purpose or effect as to negate that intention."

The Supreme Court has applied the Ex Post Facto Clause to strike down laws that increased the length of prison sentences for past offenses. By contrast, courts have upheld statutes that impose retroactive civil penalties against Ex Post Facto Clause challenges, even when the penalties at issue exceeded the amount of actual damages. The Supreme Court has upheld statutes that decrease the frequency of parole eligibility hearings and those that retroactively impose new collateral consequences for past criminal convictions, such as mandatory sex offender registration. It has also upheld changes to procedural rules found not to fall within any of the categories identified in Calder. The Court has held that the Ex Post Facto Clause prohibits legislatures from reviving statutes of limitations that have run, though it does not prohibit extending a statute of limitations before it expires.

The MVRA and Lower Court Proceedings in Ellingburg

In 1995, Holsey Ellingburg, Jr. and an accomplice robbed a bank. At the time of Ellingburg's offense, the Victim and Witness Protection Act (VWPA) gave courts discretion to order offenders to pay restitution as part of their criminal sentences and provided that restitution obligations expired twenty years from the entry of judgment.

In 1996, Congress enacted the MVRA as part of the Antiterrorism and Effective Death Penalty Act, amending the statutory restitution regime to make restitution mandatory for certain offenses. The MVRA also extended the obligation to pay restitution until "the later of 20 years from entry of judgment or 20 years after the release from imprisonment." The amendment provided that it applied "to the extent constitutionally permissible" to "sentencing proceedings in cases in which the defendant is convicted on or after the date of enactment of this Act."

After Congress enacted the MVRA, a federal jury found Ellingburg guilty of bank robbery and use of a firearm during a crime of violence. He was sentenced to 322 months in prison and ordered to pay $7,567.25 in restitution. He was released from prison in 2022, having paid $2,154 in restitution while incarcerated. Under the VWPA as in effect at the time of his offense, his restitution obligation would have expired in 2016. However, in 2023, he received notice from his parole officer stating that he owed $13,476 in restitution, including accumulated interest, and was required to make monthly $100 restitution payments.

Ellingburg, proceeding pro se, filed suit in the U.S. District Court for the Western District of Missouri challenging the continuing enforcement of his restitution obligation under the Ex Post Facto Clause. The district court noted that Ellingburg "appear[ed] to argue that the sentencing court wrongfully applied the MVRA in ordering restitution," but found that Ellingburg's restitution obligation was actually imposed under the VWPA. It thus construed Ellingburg's pleading to argue "that applying the MVRA's expanded liability period to the order of restitution in this case violates the Ex Post Facto Clause." The court rejected the challenge on the ground that the MVRA's extension of the time period during which restitution must be paid did not constitute an increase in criminal punishment.

The U.S. Court of Appeals for the Eighth Circuit (Eighth Circuit) affirmed on alternative grounds. While noting that a majority of federal courts of appeals have found that MVRA restitution is a criminal penalty, the court applied Eighth Circuit precedent holding that MVRA restitution is primarily remedial or compensatory rather than punitive. Two members of the three-judge panel joined a concurrence stating that, absent circuit precedent, they would have held that the MVRA was punitive in light of Paroline v. United States, a 2014 case in which the Supreme Court found that a different criminal restitution statute served both "remedial and penological purposes." However, the panel was bound by circuit precedent, which could only be overruled by the full Eighth Circuit sitting en banc. The remaining member of the Eighth Circuit panel concurred in the judgment and opined that the Eighth Circuit's precedent was consistent with relevant Supreme Court decisions.

Ellingburg, now represented by counsel, filed a petition for a writ of certiorari, and the Supreme Court granted review on the question of whether criminal restitution under the MVRA is penal for purposes of the Ex Post Facto Clause.

Supreme Court Briefing in Ellingburg

Before the Supreme Court, Ellingburg argues that the text, structure, and legislative history of the MVRA demonstrate that Congress intended for MVRA restitution to be punitive. Among other things, he points to the fact that restitution is imposed "as part of the criminal sentence for a criminal conviction," is imposed through criminal procedures rather than civil procedures, and is enforced via the threat of other criminal punishments such as revocation of supervised release or resentencing to imprisonment. He contends that Congress "literally wrote its penal goals into the statute," expressing its view of restitution "as serving punitive and deterrent purposes." He also points to Supreme Court opinions, representations from the Department of Justice, and legislative history to assert that "[w]hen the MVRA was enacted in 1996, it was well understood that restitution imposed at sentencing is criminal punishment." In holding otherwise, he contends, the Eighth Circuit "erred by fixating on restitution's compensatory aspects and ignoring more relevant evidence of Congress' intent" and the fact that "restitutionary payment to victims has always been a core aspect of criminal punishment."

Ellingburg further argues that, "[e]ven were [the] Court to conclude that Congress intended to create a civil remedy in the MVRA, restitution is nonetheless criminal punishment because it functions as criminal punishment in purpose and effect." Applying a seven-part test for punitiveness laid out in another context in the 1963 Supreme Court case Kennedy v. Mendoza-Martinez, he asserts that restitution (1) involves affirmative disabilities or restraints, (2) has historically been regarded as a punishment, (3) comes into play only on a finding of scienter, (4) promotes retribution and deterrence, the traditional aims of punishment, (5) applies only to criminal offenses, (6) does not rationally serve the non-punitive goal of victim compensation, and (7) is excessive with respect to that purpose.

The United States opposed Ellingburg's petition for a writ of certiorari. However, after the Supreme Court agreed to hear the case, the government filed a brief supporting vacatur of the Eighth Circuit's decision. On the question before the Court of whether restitution under the MVRA is a criminal penalty for purposes of the Ex Post Facto Clause, the government agrees with Ellingburg that restitution is punitive. The government contends that "the text and structure of the MVRA integrate restitution into the defendant's criminal sentence," which "indicates that restitution under that statutory scheme should be considered part of the punishment for a criminal conviction." In support, the government cites the ways in which restitution is imposed and enforced, as well as the legislative history of the MVRA.

While the government agrees that the MVRA is punitive, it also asserts that such a finding does not end the inquiry into whether the law violated the Ex Post Facto Clause as applied to Ellingburg. Noting that "many courts of appeals have rejected claims like petitioner's on the ground that altering the amount of time for paying off a restitution obligation is not . . . an increase [in punishment]," the government contends that the MVRA's effect is "analogous to the extension of a nonexpired limitations period for charging a criminal offense." The government maintains that "the court of appeals' judgment was ultimately correct because applying the [MVRA] did not result in an increase in punishment that would violate the Ex Post Facto Clause." However, because the government takes the position that the Eighth Circuit's decision rested on an incorrect application of law, it asks the Court to "follow its usual practice by remanding to allow the court of appeals to apply the correct legal standard in the first instance."

Because the government agrees with Ellingburg that the Eighth Circuit's decision should be vacated, the Court appointed an amicus curiae (non-party counsel participating as a "friend of the court") to argue in support of the Eighth Circuit's judgment. The amicus first argues that the Supreme Court should dismiss the case as improvidently granted because Ellingburg's "order of restitution was issued under the preexisting discretionary regime of the VWPA, not under the MVRA," meaning that the case does not actually implicate the question of whether the MVRA is punitive.

If the Court decides to reach the question presented, the amicus argues that the MVRA is not punitive because Congress did not intend for the MVRA to impose punishment and MVRA restitution is not punitive in form or effect. With respect to congressional intent, the amicus contends that precedent requires "unmistakable penal intent" and "overwhelming indications of punitive purpose" before a law may be deemed penal. He points to the MVRA's text, structure, and legislative history as well as Supreme Court precedent on restitution as evidence of Congress's non-punitive intent. He also argues that "the canon of constitutional avoidance counsels against construing the statute as punitive." With respect to the statute's form and effect, the amicus asserts in part that MVRA restitution rationally and proportionately serves the non-punitive purpose of compensating crime victims and that, historically, "compensatory remedies have been almost universally understood to be nonpunitive." He further argues that "MVRA restitution does not advance the traditional aims of punishment more than any other civil monetary remedy." The amicus asks the Court to dismiss the case as improvidently granted or, in the alternative, affirm the judgment of the Eighth Circuit.

The Supreme Court is scheduled to hear oral argument in Ellingburg on October 14, 2025, and likely will issue a decision in the case by early July 2025.

Considerations for Congress

A Supreme Court ruling in Ellingburg may or may not fully resolve the litigation over Ellingburg's individual restitution obligation, but will likely provide some guidance to Congress on the scope of the Ex Post Facto Clause. Any Ex Post Facto Clause limitations articulated by the Supreme Court cannot be altered by legislation, and federal and state laws that exceed such limitations are unconstitutional. However, the Ex Post Facto Clause applies only to retroactive punishments, so legislation that violates the clause when applied retroactively may be constitutional when applied to post-enactment offenses. Thus, a potential finding that Ellingburg's restitution obligation violates the Ex Post Facto Clause would not call into question the validity of restitution obligations related to crimes committed after the MVRA was enacted.

If the Court were to affirm the Eighth Circuit based on the reasoning of the circuit court majority opinion, it would end Ellingburg's case by establishing that the Ex Post Facto Clause does not apply to MVRA restitution because such restitution is not a criminal punishment. More generally, such a ruling would indicate that the Ex Post Facto Clause does not limit Congress's ability to require restitution for past offenses or extend the time for payment of restitution for past offenses as it did in the MVRA. (If the Court decided to dismiss the case as improvidently granted, as suggested by the amicus, the Eighth Circuit's decision would stand without a Supreme Court merits decision on the question presented, leaving the matter open for the Court's consideration in a future case.)

If the Court instead were to vacate the Eighth Circuit's ruling, the implications for Ellingburg's individual claim would be less clear. Even if the Court were to hold that MVRA restitution is a form of criminal punishment, it is possible that a court might uphold Ellingburg's restitution obligation on other grounds, as recommended by the government. For instance, on remand, the Eighth Circuit might join the majority of federal appeals courts that have considered the issue and hold that, even if restitution is a criminal punishment, the MVRA's extension of the time to pay restitution did not effect an ex post facto increase in punishment as applied to past offenses. A Supreme Court ruling that MVRA restitution is punitive would signal to Congress that the Ex Post Facto Clause might apply to retroactive legislation imposing or modifying restitution obligations. Congress would then need to take care that any such legislation does not retroactively increase punishment.

Beyond the context of the Ex Post Facto Clause, the litigation in Ellingburg may also be of general interest to legislators drafting or considering legislation that might raise constitutional questions. The MVRA's effective date provision makes the law applicable to post-enactment sentencing "to the extent constitutionally permissible." Ellingburg cites that language in his brief as evidence that Congress understood the MVRA to be punitive and thus to raise potential issues under the Ex Post Facto Clause if applied to cases where the offense was committed before enactment but sentencing occurred after enactment. Principles of judicial review dating back to Marbury v. Madison hold that, if a statute is challenged in court, the courts will only apply the statute to the extent it is constitutional. Language like the portion of the MVRA's effective date provision cited above therefore appears to be superfluous as a substantive matter, and may be invoked by challengers to argue that Congress itself doubted the constitutionality of legislation it enacted.