concurring.
I concur in Judge Boudin’s thoughtful opinion for the en banc court, which is consistent with the en banc decision in United States v. Jimenez-Beltre, 440 F.3d 514 (2006). I write separately because of my concern about the district court’s explanation of the reasons for its decision.
In Jimenez-Beltre, the defendant objected to the district court’s treatment of some factors the defendant cited at sentencing as reasons for a sentence below the guidelines range. The en banc court stated that “our emphasis in reviewing such claims will be on the provision of a reasoned explanation, a plausible outcome and — where these criteria are met — some deference to different judgments by the district judges on the scene.” Id. at 518. We stated further that “[wjhether the sentence falls inside or outside the applicable guidelines range, it is important for us to have the district court’s reasons for its sentence ... and this is even more important in the more open ended post-Booker world.” Id. The en banc court then added this important caveat: ‘Yet a court’s reasoning can often be inferred by comparing what was argued by the parties or contained in the pre-sentence report with what the judge did.” Id.
Now, in one of our first decisions post-Jimenez-Beltre, we rely on this caveat to affirm the decision of the district court. The defendant offered a number of mitigating arguments in support of his position that he should receive a sentence at the bottom of the guidelines range. He cited his thirty year diagnosis of bipolar disorder, his chronic medical problems, his acceptance of responsibility and cooperation with authorities, and the sentences imposed on two other defendants convicted in the New Hampshire federal district court of extensive fraud. After listening to these arguments and the counter-arguments of the government, the district court explained in detail the aggravating factors that, in its judgment, required a sentence considerably above the guidelines range. It said summarily that the mitigating factors advanced by the defendant had been “taken into account,” along with the sentencing factors set forth in 18 U.S.C. § 3553(a). Noting the “effective” arguments advanced by the government for rejecting the mitigating factors cited by the defendant, we say in our en banc decision that “it is fair to infer that the district court found them persuasive.” We add, citing Jimenez-Beltre: “That the district court did not elaborate on them — it said only that it took them into account — does not preclude the inference where the record explains it.”
Although this is an efficient approach to appellate review of sentencing, it has costs that we should acknowledge. If a sentencing court listens to the specific arguments of a defendant for leniency, and then rejects them with the summary statement that they have been taken into account in the final decision, a defendant naturally *97will be skeptical that this is so. That skepticism will be particularly acute if the court explains in detail, as happened here, its reasons for imposing a sentence considerably above the guidelines range while dismissing summarily the arguments for a lower sentence.
This is not a one-sided phenomenon. If the government advanced specific arguments for a particular sentence, and the court summarily rejected those arguments while explaining in detail its reasons for imposing a lesser sentence, the government would feel as the defendant feels here — that its arguments were not in fact heard and fully considered, and it would rightly complain on appeal. In either case — summary dismissal of a defendant’s arguments or summary dismissal of the government’s arguments — we can save the decision on appeal by inferring the court’s reasoning. But we will never convince the party whose arguments were summarily dismissed that these arguments were fully and fairly considered by the sentencing court.
There are also costs on appeal. There are two components of a reasonable sentence — the explanation for the choice of the sentence imposed and the sentence itself. If we have to infer a critical portion of the explanation, as we have done here, we are inescapably dealing with conjecture about it — reasonable conjecture to be sure, but conjecture nonetheless. Instead of focusing on our primary appellate responsibility, to subject the actual sentencing explanation of the district court to critical scrutiny, we must first articulate the explanation that the district court did not. That articulation will always be an approximation. We will never know what the district court actually thought. Moreover, to an extent difficult to quantify, we will mix our inferences about the district court’s reasoning with our own responses to the record of the sentencing proceeding. To that extent, we will, in effect, be doing the sentencing in the first instance — a confusion of roles.
Our willingness to infer the reasoning of the district court at sentencing also creates odd incentives for the district court. The judge who explains in detail the reasons for accepting or rejecting particular arguments advanced by the parties may make it easier for the appellate court to question the reasonableness of these explanations. An inferred explanation may be less vulnerable to that scrutiny. Hence, wittingly or unwittingly, district courts may be content to let the record speak for itself. That practice would be incompatible with the district court’s responsibility to sentence in the first instance.
I wish to be clear about the scope of my concern. I do not minimize the burdensome sentencing responsibilities of the district court. I am not suggesting that there is legal error if a district court does not explicitly address every sentencing factor and purpose set forth in 18 U.S.C. § 3553(a). There is no checklist that must be followed. But when the defendant or the government advances specific arguments for leniency or severity, grounded in the defendant’s history or the circumstances of the offense, it is reasonable to expect a district court to explain why those specific arguments are or are not persuasive. This responsiveness by the district court is not a matter of legal correctness. It is a matter of good practice. A sentencing decision that withstands appellate review is not necessarily the measure of good practice.
Here, a very able and thoughtful district court judge explained in detail the aggravating circumstances that required, in his view, a sentence considerably above the advisory guidelines range. It was a compelling explanation. The fraudulent con*98duct here was egregious. But it would have been better for us as an appellate court, and better for perceptions about the fairness of the process in the district court, if the judge had explained specifically why he rejected the defendant’s arguments in mitigation.2 Of all the important work done by the district courts, nothing is more important than their sentencing work. With so much at stake — for defendants, victims, prosecutors, and the public- — -the district courts should take the extra time, which will be minimal, to respond in some detail to the specific arguments of the government and the defendant. That will be time well spent.
. Subsequently, in a sealed statement of reasons attached to the judgment, the judge elaborated on the aggravating factors that required a sentence above the advisory guidelines range. He again said summarily that he took into account the arguments in mitigation presented by defense counsel.