Fives Clarify Harmlessness Analysis of Guidelines Calculation Error

United States v. Ibarra-Luna, No. 09-40768 (5th Cir. Dec. 22, 2010) (Higginbotham, Clement, Owen)

Been wondering just when an error in calculating the advisory Guidelines range requires reversal?  Today is your lucky day:

Today we address whether a sentence may be upheld if an error in the calculation is shown to be harmless. We hold that under the discretionary sentencing regime of Booker and progeny, the harmless error doctrine applies only if the proponent of the sentence convincingly demonstrates both (1) that the district court would have imposed the same sentence had it not made the error, and (2) that it would have done so for the same reasons it gave at the prior sentencing. On the facts before us, we conclude that this high hurdle has not been cleared and remand for resentencing.

Guess we better look at the facts here, then.
 
 
Ibarra, a Mexican citizen, was removed from the United States after picking up a Texas conviction for delivery of a controlled substance.  He returned the U.S., sans permission.  He then shot a man to death during a high-speed car chase.  Ibarra pleaded guilty to murder in state court, where he was sentenced to 22 years' imprisonment.  The feds then went after him for illegal reentry.
 
The PSR hit Ibarra with a 12-level drug-trafficking enhancement, under guideline §2L1.2(b)(1)(B), on the basis of his Texas delivery conviction.  The district court recognized that Ibarra was not subject to that enhancement, because the record of conviction did not exclude the possibility that it could have been based on a bare offer to sell.  The Government then argued that the conviction triggered an 8-level "aggravated felony" enhancement.  Ibarra said it only qualified for the 4-level "any other felony" enhancement.  The court went with the 8, which produced a Guidelines range of 12 to 18 months.  But the court said that wasn't enough time, because Ibarra might serve as little as 11 years of his 22-year state murder sentence.  Given the circumstances of the murder and the need to "be certain that Mr. Ibarra doesn't return to society in this country," the court sentenced Ibarra to 36 months' imprisonment.

Ibarra challenged the district court's Guidelines calculation on appeal.

Of course, the intial question was whether the calculation was wrong.  The court noted that §2L1.2's "drug-trafficking" definition has since been amended to include a bare offer to sell.  But it did not consider whether that new definition applies retroactively because the Government did not cross appeal.  So the court moved on to the 4 or 8 question.  Answer: 4.  As the Government conceded, "a mere offer to sell, without evidence of possession or transfer, is tantamount to solicitation and is not proscribed by the Controlled Substances Act[,]" meaning that it's not an aggravated felony.  So the correct range was 6 to 12 months, not 12 to 18 months.

Now, in previous post-Booker harmlessness cases involving Guidelines calculation error, the district court had also considered and rejected the correct range.  But the district court did not do that here.  So how does the harmlessness analysis play out in this situation?

Drawing on the teachings of Gall, the court explained the nature of the problem: "Treating the Guidelines as mandatory or failing to consider the Guidelines at all is reversible error. Without attempting to identify the applicable Guidelines range, a district court cannot meet its obligation to explain why a sentence in that range would not be adequate[,]" frustrating appellate review and the goal of national consistency in sentencing.

As a result, an incorrect Guidelines calculation will usually invalidate the sentence, even when the district court chose to impose a sentence outside the Guidelines range. Although it may well be that the same explanation the court gave for imposing a sentence outside the miscalculated range could also support a sentence outside the correctly calculated range, the harmless error doctrine requires the proponent of the sentence to convincingly demonstrate that the court actually would have followed the very same reasoning absent the error.  This is a heavy burden.

Heavy, but not your brother insurmountable.

Had the Guidelines range been correctly calculated, the range would have been 6 to 12 months. Yet if a sentence of 12 to 18 months is not long enough, then plainly a 6-to-12-month sentence also would not be long enough, for precisely the same reason that the district court gave.

"But," as the court says, "there is more."

Our past cases also make clear that a sentencing error may not be found harmless unless the proponent of the sentence “proffer[s] sufficient evidence to convince the appellate court that the district court would have imposed the same sentence, absent the error.” To satisfy that burden, the proponent “must point to evidence in the record that will convince us that the district court had a particular sentence in mind and would have imposed it, notwithstanding the error.”

Even when the district court ultimately decides to impose a sentence outside the Guidelines range, an error in its Guidelines calculation may still taint the non-Guidelines sentence. For instance, the district court might settle upon a particular non-Guidelines sentence by doubling the maximum Guidelines range, or by starting with the Guidelines range and adding or subtracting a fixed number of years. In such cases it may be clear that the district court’s reasons for rejecting a sentence in the Guidelines range are unaffected by the error, but the error nevertheless is not harmless because the district court would not have imposed the very same sentence.

The burden of showing that a particular sentence was not derivative of the Guidelines recommendation is a difficult burden if the district court fails to indicate why it selected a sentence of a particular length. It may be prudent when imposing a non-Guidelines sentence to state whether the magnitude of that sentence rests on factors independent of the Guidelines, thereby permitting a reviewing court to consider claims of harmless error.

So how about here?  It was clear from the record that the district court would have varied upward, even if it had correctly calculated the Guidelines range.  But,

[w]e cannot state with the requisite certainty . . . that the district court would have imposed precisely the same sentence. The district court did not indicate how it selected a sentence of 36 months, and it did not state whether this sentence was influenced by its Guidelines calculations or based instead on independent factors. Notably, the district court was careful to insist that “I need to understand where he comes out on the Guidelines, and then make my variance,” which suggests that, quite properly, the Guidelines recommendation affected the sentence it selected. We also note that, compared to the 12-to-18-month range the court did consider, the 36-month sentence it imposed is exactly double the Guidelines maximum and exactly triple the Guidelines minimum.

On these facts, the government has not met its burden to convincingly demonstrate that the court would have imposed the very same sentence if it had not made an erroneous calculation. We must therefore VACATE Ibarra’s sentence and REMAND to the district court for resentencing.