Welcome back to SCOTUS Oral Arguments and Opinions.

Today we're diving into Ellenberg versus United States.

@ its core, this case asks a deceptively simple question.

Is criminal restitution under the Mandatory Victim Restitution act actually criminal punishment for purposes of the ex post facto clause?

And that question matters because if restitution is criminal punishment, then the government can't apply new restitution rules retroactively to make things worse for defendants.

But if it's just a civil remedy to compensate victims, then the ex post facto clause doesn't apply at all.

The implications go far beyond just this one defendant.

We're talking about how courts interpret decades of federal restitution law and whether thousands of people convicted before 1996 are still on the hook for for debts that might have expired.

Before we go too far, I wanted to highlight this as another example where the government switched sides after the change in administration.

The Solicitor General's office now agrees with the petitioner that it was wrong to enforce restitution retroactively.

So the court had to appoint a special lawyer, called an amicus curie, just to defend the lower court's decision.

This lawyer is named John F. Bash, a former lawyer in the Solicitor General's office.

Alright, let's dive in.

The court granted certiorari to decide whether criminal restitution under the Mandatory Victim Restitution act is penal for purposes of the ex post facto clause.

Now, for listeners who might not be familiar, the ex post facto clause is in Article 1, Section 9 of the Constitution.

It says, no ex post facto law is shall be passed.

Expost facto is Latin for after the fact.

The clause prohibits the government from retroactively changing the rules to make things worse for criminal defendants.

So you can't pass a law today that increases punishment for crimes committed yesterday.

But here's the catch.

The ex post facto clause only applies to criminal punishment, not to civil remedies.

So if restitution is criminal punishment, the clause protects defendants.

If it's civil, it doesn't.

And that brings us to the key statutory text everyone's fighting over.

The Mandatory Victim Restitution act, or MVRA for short, was enacted in 1996.

Section 3663A says when sentencing a defendant convicted of an offense, the court shall order in addition to or in the case of a misdemeanor, in addition to or in lieu of any.

Any other penalty authorized by law that the defendant make restitution to the victim.

Let's break that down when sentencing a defendant convicted of an offense that's linking restitution to criminal sentencing in addition to any other penalty.

That's treating restitution like one penalty among others, alongside things like prison time and fines.

The petitioner seizes on that language, arguing it proves Congress treated restitution as as criminal punishment.

The amicus defending the lower court says that language is misleading.

Just because something happens at sentencing doesn't make it criminal punishment.

And we'll see how each side uses statutory interpretation to support completely opposite conclusions about what Congress intended.

Let's talk about how this case got to the Supreme Court, because the facts matter for understanding the legal fight.

In 1995, Halsey Ellingberg Jr.

Robbed a bank in St. Louis, Missouri.

The next year, in 1996, he was sentenced in federal court to 322 months in prison.

That's almost 27 years, plus five years of supervised release.

The court also ordered him to pay $7,567 in restitution to the victim.

Now here's where the timing gets Congress had just passed the MVRA in April 1996 before Ellingberg's sentencing.

But the sentencing court didn't impose restitution under the new mvra.

Instead, it used the old law that was in place when Ellingberg committed the crime, something called the Victim and Witness Protection act, or vwpa.

And that distinction becomes huge later in the case.

Under the VWPA rules in effect when Ellingberg committed his crime, his restitution obligation would have expired in November 2016, 20 years after the judgment.

Fast forward to 2022.

Ellingberg gets released from prison and tries to rebuild his life in Missouri.

He's working a minimum wage job at a warehouse, living with his fiance, and his earnings barely cover his bills.

But in early 2023, he starts getting text messages from his probation officer demanding monthly $100 restitution payments.

And the government now claims he owes $13,476, almost double the original amount because of mandatory interest and the extended collection period under the mvra.

Wait, so under the old law, his Debt expired in 2016, but the government is trying to collect in 2023 using the new law's provisions?

Exactly.

The MVRA extended the collection period from 20 years to the defendant's lifetime, and it also imposed mandatory interest on unpaid restitution.

So Ellingberg goes to court pro se, meaning representing himself, and files a motion arguing this violates the ex post facto clause.

The district court denied his motion.

The court said that even if the MVRA applies retroactively, it's just procedural and doesn't increase punishment, so there's no ex post facto violation.

The 8th Circuit affirmed, but on totally different reasoning.

The appeals court said restitution under the MVRA is, isn't criminal punishment at all.

It's a civil remedy.

So the ex post facto clause doesn't even apply in the first place.

The 8th Circuit relied heavily on its own precedent from a 2005 case called Carruth, which held that MVRA restitution is designed to make victims whole, not to punish perpetrators.

So it is essentially a civil remedy created by Congress and, and incorporated into criminal proceedings for reasons of economy and practicality.

Two judges on the panel wrote concurring opinions, basically saying, look, we think this might be wrong after the Supreme Court's decisions in Pasquantino and Parolin, but we're bound by circuit precedent.

Then Ellingberg petitions the Supreme Court, and here's where it gets wild.

After the court grants cert, the Trump administration takes over and the new Solicitor General files a brief saying, actually, we agree with Ellingberg.

The 8th Circuit got it wrong.

MVRA restitution is criminal punishment.

So now both parties are on the same side.

The court has to appoint an amicus curie, a friend of the court, to defend the 8th Circuit's judgment.

Otherwise there'd be no one arguing the other side.

The Supreme Court granted Certore in April 2025 to resolve this question, which has divided the federal courts of appeals.

Most circuits have held that MVRA restitution is criminal punishment and subject to the ex post facto clause.

But a few circuits, including the seventh and eighth, have said it's civil.

So there's a clear circuit split that needs resolution.

Alright, let's dive into the arguments.

Ellingberg's legal team makes three main points.

First, they say the statutory text and structure prove Congress intended restitution to be criminal punishment.

The language is pretty clear.

Section 3663A says restitution is ordered when sentencing a defendant convicted of an offense.

It has to be imposed in addition to or in lieu of any other penalty authorized by law.

Think about what in addition to any other penalty means.

The statute groups restitution with fines and imprisonment, things everyone agrees are criminal punishments.

If you can impose restitution instead of prison time for a misdemeanor, how is that not punishment?

Plus, the MVRA is codified in Title 18 of the U.S. code.

That's the federal criminal code.

Section 3556, which requires courts to order restitution is in Chapter 227 literally titled sentences.

And Section 3611 says a person who is sentenced to pay a fine, assessment or restitution shall pay it.

The statute treats restitution the same as fines, which are unquestionably criminal penalties.

Ellingberg also points to the procedures.

Restitution follows all the same procedures as other criminal penalties.

The probation officer prepares information for the court, criminal sentencing rules apply, and the Federal Rules of Criminal Procedure govern the whole process.

And get this, a sentence imposing restitution is a final judgment that can be appealed under section 3742, which is the statute governing appeals of criminal sentences.

So the petitioner's first argument is basically look at what Congress wrote.

They integrated restitution into criminal sentencing.

And at every turn, you can't read this statute without concluding Congress treated restitution as part of the criminal punishment.

The second major argument focuses on enforcement and express penal purpose.

Ellingberg argues restitution is backed by the threat of imprisonment, the most paradigmatic form of criminal punishment.

Right.

If you don't pay restitution, courts can revoke your probation or or supervised release and send you back to prison.

And here's the kicker.

They can resentence you up to the statutory maximum for your original crime, all without any new indictment, prosecution or conviction.

That's significant because it means the enforcement mechanism treats non payment not as a simple debt collection problem, but as a violation of your criminal sentence worthy of incarceration.

And payment of restitution is a mandatory condition of probation and supervised release.

You're under constant supervision by probation officers who are enforcing this as part of your criminal sentence.

But it gets even more explicit.

Congress actually wrote punishment into the statute.

Section 3614 says courts can resentence offenders to prison if alternatives to imprisonment are not adequate to serve the purposes of punishment and deterrence.

Wait.

So Congress explicitly used the words punishment, punishment and deterrence in describing restitution's purposes?

That's exactly right.

Ellingberg quotes this language prominently, arguing, how much clearer can Congress be?

They expressly linked restitution to the traditional penal goals of punishment and deterrence.

There's also a really interesting point about Congress's own concerns.

The MVRA includes language saying it applies only to the extent constitutionally permissible.

That suggests Congress itself worried about ex post facto problems.

Right?

If Congress thought restitution was just a civil remedy, why would they include that limitation?

The petitioner says this shows Congress's own understanding that MVRA restitution could trigger ex post facto concerns.

Now let's talk about the third major Supreme Court precedent and historical practice.

Ellingberg relies heavily on what the Supreme Court has said about restitution in other contexts.

The biggest case is Pasquantino vs United States from 2005, the court said, and I'm quoting directly here, the purpose of awarding restitution under the MVRA is to mete out appropriate criminal punishment.

That's pretty unambiguous.

Mete out appropriate criminal punishment.

Those are the Supreme Court's words describing MVRA restitution.

Then there's Parolene versus United States from 2014.

The court was looking at a related restitution statute for child pornography offenses, and it recognized restitution's penological purposes.

The court said that while restitution's primary goal is remedial or compensatory, it also serves punitive purposes because because it's imposed by the government at the culmination of a criminal proceeding and requires conviction of an underlying crime.

There's also historical practice.

After the VWPATHE, MVRA's predecessor, was enacted in 1982, defendants challenged restitution orders under the seventh amendment, arguing they had a right to a jury trial to determine the amount.

And courts of appeals uniformly rejected those challenges by holding that restitution was a criminal penalty, not a civil remedy.

So there's no Seventh Amendment jury trial right.

The Third Circuit, in a case called Palma, said courts uniformly held that an order of restitution imposed under the VWPA is a criminal rather than civil penalty.

So when Congress enacted the MVRA in 1996, it did so against this backdrop of courts consistently treating restitution as criminal punishment.

And here's something that really strengthens Ellingberg's case, the government's own historical position.

Right after the MVRA was enacted, the Solicitor General directed U.S. attorneys offices nationwide not to apply it retroactively.

The Solicitor General's position was consistent with the view that restitution under the MVRA is a penalty which, if applied to offenses occurring before the enactment of the act, would retrospectively increase punishment for the crime and in violation of the ex post facto clause.

And for decades, the Department of Justice maintained this position.

In Pasquantino, the Solicitor General told the Supreme Court that MVRA restitution is a criminal punishment that is imposed as part of the sentence for an offense.

Turning to the government's brief, the government now says the 8th Circuit got it wrong and the case should be vacated and remanded.

But they make slightly different arguments than Ellingberg.

So let's walk through them, the government's first main point is about statutory construction.

They emphasize that whether something is criminal punishment is principally a question of statutory construction.

Citing Kansas vs Hendricks, the government argues, you have to look at what Congress actually wrote.

And here the text and structure of the MVRA integrate restitution into the defendant's criminal sentence.

They quote section 3663A's language about ordering restitution when sentencing a defendant convicted of an offense in addition to or in lieu of any other penalty authorized by law.

The government emphasizes the codification structure that restitution is in the sentencing chapter alongside all other criminal penalties.

Section 3556 is in Chapter 2, 127, titled Sentences, right next to provisions authorizing imprisonment and fines.

They also point to section 3551, which identifies authorized sentences for defendants found guilty of an offense and specifies that courts may impose sanctions authorized under section 3556, which is the restitution provision.

The government stresses the procedural mechanisms.

They say restitution is imposed pursuant to procedures similar to those for other penalties imposed during sentencing.

The probation officer prepares reports, criminal procedure rules apply, presentence investigations happen, and appeals follow criminal sentencing appeal rules.

So the government's argument is that when you do straightforward statutory construction looking at text and structure, the conclusion is inescapable.

Congress intended restitution to be part of criminal punishment.

The second major government argument focuses on precedent.

They walk through the historical development, starting with how courts uniformly treated VWPA restitution as criminal punishment.

The government points out that when defendants challenged VWPA restitution under the Seventh Amendment, courts of appeals uniformly held that an order of restitution imposed under the VWP A is a criminal rather than civil penalty.

They cite Kelly vs. Robinson, where the Supreme Court called restitution a criminal sanction with a deterrent effect that serves as an effective rehabilitation penalty.

The government emphasizes what the court said in Kelly that because criminal proceedings focus on the state's interests in rehabilitation and punishment, restitution imposed during sentencing must be understood in that penal context.

Then they cite Pasquantino, where the court said it would be passing strange to think tort law provides the only model for restitution imposed by the government at the end of a criminal proceeding against a defendant convicted of a criminal offense and Paralyne, where the court explained that while restitution under the statute is paid to a victim, it is imposed by the government at the culmination of a criminal proceeding and requires conviction of an.

Underlying crime the government notes that the majority of the courts of appeals to address the issue have recognized that the ex post facto clause applies to restitution imposed under the MVRA.

They cite the 3rd, 5th, 6th, 9th, and 11th circuits.

Alright, so now we have a court appointed amicus defending the 8th Circuit's judgment.

And they make some pretty sophisticated arguments about why restitution isn't criminal punishment.

The amicus's first major argument is about the level of proof required.

They say that under the framework from Kennedy vs. Mendoza Martinez, you need conclusive evidence of congressional intent to find something.

Is criminal punishment.

What does conclusive evidence mean?

According to the amicus, courts must find unmistakable penal intent and overwhelming indications of punitive purpose.

The standard requires that the objective manifestations of congressional purpose indicate conclusively that the provisions in question can only be interpreted as punitive.

That's a really high bar.

The amicus is saying it's not enough to show that Congress might have intended punishment or even probably intended punishment.

You need overwhelming conclusive evidence.

And the amicus grounds this in separation of powers principles when a law might violate the Constitution if deemed criminal.

Attributing punitive intent to Congress is tantamount to finding that Congress intended to test constitutional limits.

So courts should be cautious about finding punitive intent because that interpretation could invalidate the statute, and courts are supposed to avoid constitutional problems when possible.

Now, applying that framework, the amicus says Congress didn't conclusively intend MVRA restitution to be punitive.

First look at the label.

The statute doesn't use the term criminal restitution or explicitly call it a criminal sanction.

The amicus points out that Congress used the label criminal in other statutes when it wanted to be clear, but it didn't use that label here.

It just says restitution, which ordinarily signifies a civil remedy.

Then there's the structure.

The amicus emphasizes that restitution under section 3663 is generally mandatory, depriving courts of the case, specific discretion ordinarily involved in criminal sentencing.

What do they mean by that?

When Section 3663A applies, courts must award full restitution for the victim's losses, no more, no less.

The court is not permitted to consider the defendant's culpability or economic circumstances, the nature of the offense, or basic pentological goals like deterrence and just punishment.

So the argument is that criminal punishment usually involves judicial discretion to tailor the penalty to the offender's culpability and the crime's severity.

But MVRA restitution is just a math problem.

Calculate the victim's loss and order that amount exactly.

The emicus says this looks more like civil damages, which aim to make victims whole, than criminal punishment, which aims to punish and determine.

The amicus also emphasizes where the money goes.

MVRA restitution goes directly to the victim in most cases, and it is never paid to the federal government as the prosecuting sovereign.

And when the United States is also a victim, other victims get paid first.

Plus, victims can enforce restitution just like any other civil creditor.

They can get liens and use civil collection procedures.

The amicus argues these features show Congress designed the MVRA to serve civil and remedial ends by compensating crime victims not to impose criminal punishment.

So the amicus's first argument is essentially Congress didn't conclusively intend this to be criminal punishment.

The label is civil, the structure is compensatory, and the payment scheme looks like a civil remedy.

The amicus's second major argument addresses the Mendoza Martinez factors, the seven factor test courts use when congressional intent isn't conclusive.

The amicus says that even if intent is ambiguous, restitution isn't so punitive in purpose or effect as to be criminal.

They start with traditional civil restitution doctrine.

The amicus argues that as an equitable remedy, traditional civil restitution focuses on the defendant's gain and requires disgorgement of that gain.

I But they note restitution under the MVRA is tied to the victim's losses even when the defendant never received a corresponding benefit.

So it looks more like civil damages to compensate victims than like criminal punishment.

On the affirmative disability or restraint factor, the amica says MVRA restitution itself involves no such disability.

Yes, failing to pay can result in revocation of probation, but but that doesn't make the restitution itself punitive.

The amicus makes this on that logic, a tax could be deemed penal because a person's failure to pay it might result in imprisonment.

The consequences of non payment don't determine whether the obligation itself is punitive on the scientific factor, whether it requires criminal intent.

The amicus acknowledges restitution requires a criminal conviction, but they say that consideration has little weight when the criminal conviction is necessary to serve a non punitive purpose like victim compensation.

They point out that lots of civil consequences follow from criminal convictions like loss of voting and firearm rights and changes to immigration status.

Those aren't criminal punishments you just because they stem from convictions on whether restitution.

Implicates the traditional aims of punishment retribution and deterrence.

The amicus says it doesn't because courts can't consider penological objectives when calculating restitution.

They favorably cite the 7th Circuit's decision in Newman, which said the MVRA does not directly promote the traditional aims of punishment, retribution, and deterrence.

So the amicus's second argument is that even if you go through all seven Mendoza Martinez factors, restitution looks more like civil compensation than criminal punishment.

Now here's the amicus's third argument, and it's a threshold issue that could dispose of the case without reaching the merits.

The amicus says the petition should be dismissed as improvidently granted.

What does that mean?

It means the Supreme Court made a mistake in agreeing to hear this case because it doesn't actually present the question the the court thought it was deciding.

The amicus argues that the question presented asks about restitution under the mvra.

But Ellingberg's restitution wasn't actually imposed under the mvra.

Wait, I thought we said the MVRA applies to him.

Here's the nuance.

The sentencing court imposed Ellingberg's restitution under the pre MVRA version of the VWPA.

Only the MVRA's extended collection period and interest provisions applied to him retroactively.

So the actual restitution order came from the VWPA, but the government is trying to collect under the MVRA's extended timeline.

Exactly.

And the amicus argues that answering whether MVRA restitution is criminal punishment would yield an advisory opinion to Because Ellingberg's restitution order came from a different statute, the.

Amicus says a retroactive extension of petitioner's restitution term would implicate the ex post facto clause only if his restitution obligations imposed under the VWP A qualify as criminal punishment.

The nature of MVRA restitution is irrelevant.

So if the court takes this argument seriously, it might dismiss the case without deciding the merits at all.

That would be a pretty anticlimactic ending.

After all this briefing, though, it's worth noting that neither the petitioner nor the government really engaged with this argument in their briefs, which is why the court appointed the amicus in the first place to make sure all possible arguments got a fair hearing.

So, looking ahead to oral arguments, what are you going to be listening for?

I'll be really interested in how the justices react to the statutory construction question.

Do they think the text clearly indicates punishment, or do they see ambiguity that requires looking at the Mendoza Martinez factors?

I'm also curious Whether the Justices find the government's position switch persuasive or problematic, Will they credit the government's representations about its historical position, or will they be skeptical about the timing of this change?

The Pascantino and Parylene precedents are going to be huge.

The petitioner and government rely heavily on language from those cases calling restitution criminal punishment.

But the amicus will argue those cases weren't deciding ex post facto questions, so the language was dicta, meaning non binding observations.

I'll be watching for questions about the practical implications.

If MVRA restitution is civil, does that mean it's not subject to other constitutional protections that apply to criminal punishment?

What about statutes of limitations?

What about the Sixth Amendment?

The alternative ground for affirmance could be really important, too.

Even if the justices agree restitution is criminal punishment, they might find that extending the collection period doesn't actually increase punishment.

Right.

Some justices might say the punishment was the $7,567 debt period.

The deadline for collecting it is just a procedural matter, not part of the.

Punishment itself, though Ellingberg would argue that's like saying it's not punishment to triple someone's prison sentence.

The original sentence was 10 years, and we're just extending the time horizon for serving it.

The amicus's threshold argument about improvident grant could come up if the Justices realize Ellingberg's restitution came from the vwpa, not the mvra.

They might question whether they can even reach the question presented, though practically speaking, the legal analysis would be very similar for VWPA restitution versus MVRA restitution.

So the Court might decide to reach the issue anyway.

I'm also curious about how the Justices view the mandatory versus discretionary distinction.

Does the fact that courts have no discretion make restitution less punitive, as the amicus argues?

Or does it not matter to the statutory construction question?

And watch for questions about congressional intent versus practical effects.

Some Justices might focus heavily on what Congress meant to do, while others might care more about how restitution actually functions in practice.

So why does Ellenberg versus United States matter beyond just this one defendant?

The immediate impact affects potentially thousands of people convicted of federal crimes before 1996.

If restitution is criminal punishment and the ex post facto clause applies, many people whose restitution debts expired under old law might not owe anything under new law.

But the broader implications go to how courts interpret federal criminal statutes.

This case is fundamentally about statutory construction.

When does Congress intend something to be criminal punishment versus a civil remedy, and.

It could affect restitution enforcement going forward.

If MVRA restitution is criminal punishment, it might be subject subject to other constitutional protections we haven't even thought about yet the Excessive Fines Clause, for example.

The case also highlights tensions between different models of criminal justice.

Is restitution primarily about punishing offenders and deterring crime, or is it about compensating victims and making them whole?

The answer isn't necessarily either, or.

Restitution can serve multiple purposes, but for constitutional purposes, courts need to categorize it as either criminal or civil.

One thing that makes this case fascinating is the government's position switch.

It's incredibly rare for the Solicitor General's office to confess error and switch sides after the court grants cert.

That tells you how strong they think the legal arguments are on one side.

The current administration clearly believes the 8th Circuit got it wrong and that correcting this error is is worth the unusual step of abandoning the government's original position.

We'll be covering the oral arguments as soon as they're scheduled, so make sure you're subscribed.

This is going to be a fascinating argument to listen to.

Thanks for joining us for this deep dive into Ellingberg versus United States.

It's a great example of how seemingly technical questions about statutory interpretation can have enormous practical consequences for for real people's lives.

In the next episode, we'll cover Bowe vs United States, a case that involves a nuanced question of habeas corpus.

If you like the content, please be sure to subscribe, rate and share.

Talk to you soon.