concurring in the judgment.
There can be no doubt, based upon the way in which he treated his victims, that *247John Jarvis is a despicable person who deserved to have “the book thrown at him.” It therefore seems difficult, at first blush, to take issue with Judge Fuentes’ forceful opinion affirming the upward adjustments to Jarvis’ base offense level under the United States Sentencing Guidelines. But the Guidelines are a carefully reticulated set of regulations whose animating goal is the elimination of the former regime in which a judge could react to such terrible conduct by simply imposing a harsh sentence or, conversely, reward a felon with an otherwise exemplary background by “giving him a break.” The Guidelines establish instead a regime under which: (1) harms are quantified through careful legal definition; and (2) a range of punishments derived from those definitions is prescribed, subject to a variety of guided departures that depend on objective judicial findings that must be consonant with the Guidelines’ terms. Most importantly, sentencing judges are not free to ignore the strictures of the Guidelines, however untoward they deem the result.
Principally at issue here is Guideline § 5K2.3, a guided departure provision, which authorizes the sentencing court to increase the sentence above the normal Guideline range if a victim suffered “psychological injury much more serious than that normally resulting from commission of the offense.” The section goes on to explicate this standard as follows:
Normally, psychological injury would be sufficiently severe to warrant application of this adjustment only when there is a substantial impairment of the intellectual, psychological, emotional, or behavioral functioning of a victim, when the impairment is likely to be of an extended or continuous duration, and when the impairment manifests itself by physical or psychological symptoms or by changes in behavior patterns.
U.S.S.G. § 5K2.3.
The Guidelines text imposes a rigorous standard of proof, and for good reason. Fraud offenses that involve duping people out of their life’s savings will usually cause psychological injury; the greater the loss to the victim, the greater the probable injury. The Guidelines capture most of the harm from fraud offenses by incrementally increasing the sentencing range in accord with the amount of money involved, see U.S.S.G. § 2Fl.l(b)(l), or by enhancing the sentencing range for vulnerable victims, see U.S.S.G. § 3Al.l(b). Thus, in § 5K2.3 the Commission apparently wanted to be quite specific as to the showing needed for a psychological injury enhancement so as not to duplicate these other factors.
I do not believe that the evidence upon which the District Court imposed the substantial upward adjustment for psychological harm meets the rigorous standard of the Guidelines, which in essence requires reliable evidence. More specifically, I believe that neither the uncorroborated letter from Mr. & Mrs. Hager that they are on depression medication and are “seeing a shrink,” nor the uncorroborated and vague letter from Anna Marie Kmonk relating that she and her husband “have had health problems ... I believe brought on by this stressful situation” provides the reliable evidence of psychological injury “much more serious than that normally resulting” from the offense that is required by § 5K2.3. Furthermore, these letters are the strongest pieces of evidence in the Government’s case; there is nothing comparable respecting the other victims.
The majority’s conclusion that these letters meet the § 5K2.3 test is grounded on our opinion in United States v. Astorri, 923 F.2d 1052 (3d Cir.1991), in which, based on *248quite similar evidence, the panel affirmed the District Court’s upward adjustment for psychological harm. Because I agree with the majority that Astorri controls, I am constrained to join in the judgment of the Court.1 I write separately because I believe that Astorri was wrongly decided; my hope is to persuade the Court to take up this case en banc to overrule Astorri, or better still, to convince the Sentencing Commission to revise § 5K2.3 so as to clarify it and make clear that cases of this genre do not justify an upward adjustment.
Astorri involved a fact pattern very similar to that in the case at bar: the defendant was convicted of defrauding unsuspecting investors, including many elderly persons who thereby lost much or all of their life savings.2 As in this case, the district court adjusted upward the base offense level for fraud to account for the amount involved in the fraud under U.S.S.G. § 2F1.1, and added a two-level enhancement for vulnerable victims under § 3A1.1. The important adjustment in Astorti for our purposes was the district court’s two-level upward departure under § 5K2.3 for the extreme psychological injury inflicted on the victims of the fraud. When the defendant appealed, the panel majority upheld the departure based on two factors: (1) the district court had gauged the effects on the victims from its observations at trial; and (2) two of the fraud victims had suffered physical and behavioral manifestations of their psychological injury, thus meeting the requirements of § 5K2.3. As the Court put it: “Mrs. Needles has been forced to seek treatment for high blood pressure as a result of Astorri’s scheme. She continues to be under a doctor’s care.... Record evidence reveals that Mr. Taylor, already in poor health, displayed adverse physical and behavioral effects from those dealings.” 923 F.2d at 1059. The basis for these conclusions about the physical effects felt by the victims was two uncorroborated letters, one written by Mr. Taylor’s attorney, and one written by the Needles themselves.
I believe that this upward adjustment in Astorri was improper because it was not founded on reliable evidence, and was not demonstrably justified by psychological injury “much more serious than that normally resulting from commission of the offense.” U.S.S.G. § 5K2.3 (emphasis added). My point of departure is the dissenting opinion in Astorri of my late respected colleague, Judge William D. Hutchinson. Indeed, the best way to make my point is to quote (at some length) Judge Hutchinson’s words:
The evidence about the effect Astor-ri’s fraud had on the Taylors and the Needles’ health is insufficient to support the district court’s conclusion that some of the victims suffered the kind of substantial and permanent physical, intellectual or behavioral impairments that Guidelines § 5K2.3 requires before an upward departure for extreme psychological injury is authorized. These unsupported lay statements are not reliable evidence of the kind required to support enhancement of a guidelines sentence. See, e.g., United States v. Sciarrino, 884 F.2d 95 (3d Cir.) (while hearsay is permissible in determining a guidelines sentence, it must have some degree of reliability), cert. denied, 493 *249U.S. 997, 110 S.Ct. 553, 107 L.Ed.2d 549 (1989).
Likewise, I think the sentencing judge’s own observations that the psychological trauma naturally resulting from the economic losses Astorri’s fraud visited upon his victims and his profound betrayal of the Kronyaks and their daughter is insufficient to show objective symptoms of substantial and continuous intellectual, psychological, emotional or behavioral impairment. Those observations are conclusions that must be founded on rehable evidence under the guidelines. They are not themselves evidence.
Perhaps determinations of crime’s effect on its victims would have been better left to the observations and sound discretion of the sentencing judge, but Congress has decided otherwise. See Mistretta v. United States, 488 U.S. 361, 109 S.Ct. 647, 652, 102 L.Ed.2d 714 (1989). The Sentencing Commission has acted to implement Congress’s decision when it confined the sentencing judge to a relatively narrow sentence range objectively determined on the basis of reliable evidence that particular effects accompany a particular crime. Even in departures, where a fairly large element of discretion is retained, facts grounded on rehable evidence must show that one of the reasons for departure is present.
I do not doubt that a person suffers psychologically when he loses his life’s savings, let alone his home. However, I believe any economic loss a victim suffers is otherwise adequately taken into account under Guidelines § 2Fl.l(b)(l)(H), adjusting the offense level for the amount of monetary loss. Likewise, I believe that the age of the victim is taken into account under Guidelines § 3A1.1, relating to vulnerable victims, a section which the district court correctly applied to enhance As-torri’s sentence.
Astorri’s conduct demonstrates a heartless willingness to trade on the affection of the woman he had promised to marry and the trust she and her parents placed in him while he secretly plundered the savings her parents had reserved against their old age. The common understanding of men and women of every time and place condemns Astor-ri as a despicable cad. However, the Sentencing Commission has taken As-torri’s truly outrageous and cynical manipulation of his fiancee’s family for his own private gain into consideration in Guidelines § 3A1.1, relating to vulnerable victims, and Guidelines § 3B1.3, relating to abuse of trust.
Guidelines § 5K2.3 focuses elsewhere in permitting enhancement for extreme psychological suffering. There is no reliable evidence in this record to show such an injury. Although the evidence here is sufficient to support the district court’s finding that Astorri’s victims were vulnerable and to show that his scheme included the elements of an abuse of trust, it was not sufficient to show any of Astorri’s victims suffered “extreme psychological injury” and so permit enhancement of his sentence under § 5K2.3.
923 F.2d at 1061-62 (Hutchinson, J., dissenting).
I agree. Although Astorri controls the outcome of this case because of the similarity of the facts and evidence used to support the § 5K2.3 departure, the arguments in Judge Hutchinson’s dissent ring true. I do not believe that, under the correct interpretation of § 5K2.3, the evidence before the District Court in this case was sufficiently reliable to support the necessary finding of “psychological injury much more serious than that normally re-*250suiting from commission of the offense”; the evidence simply does not sufficiently support a finding of “substantial impairment of the intellectual, psychological, emotional, or behavioral functioning of a victim ... [that] manifests itself by physical or psychological symptoms or by changes in behavior patterns.”
Both here and in Astorri, the District Court relied upon the conclusory lay statements of the victims (or their lawyers) and the court’s own observations of the victims to support the departure. But the Guidelines (and due process) generally require that evidence used in sentencing be reliable. See, e.g., United States v. Sciarrino, 884 F.2d 95 (3d Cir.1989). I agree with Judge Hutchinson that the type of evidence employed in Astorri (and in this case) is insufficiently reliable to use as a basis for an upward departure under § 5K2.3. More specifically, I believe that Judge Hutchinson was correct that, because the Sentencing Guidelines generally require an objective basis for a departure or enhancement, something more than conclusory, unsupported lay statements and the District Court’s “eyeballing” of the victim should be required to show the requisite impairment of physical, psychological, or behavioral functioning and the comparative severity of psychological injury.
I do not suggest that expert medical testimony is a prerequisite to a § 5K2.3 departure (although such testimony would certainly suffice as objective evidence). I do, however, believe that medical evidence is preferable and that, in the absence of detailed and truly compelling lay reports, some sort of medical evidence from an expert should be required — e.g., an affidavit or even a signed letter from a health care provider, or the victim’s medical records. In short, the basis for the departure should be more than the naked claims of the victim set forth in a letter.
I believe that we should take up this case en banc to overrule Astorri. This issue arises with some degree of regularity and surely presents an important question. Alternatively, I suggest to the Sentencing Commission that it alter § 5K2.3 (thus effectively overruling Astorri) by adding to the end of that Guideline something like the following:
In the absence of detailed and truly compelling lay testimony from the victim, a departure by the sentencing court under this section should be based upon objective evidence such as an affidavit or signed letter from the victim’s health care provider or a verified copy of the victim’s medical records.
An amendment along these lines would provide for an objective basis for upward departures under § 5K2.3 that would make that section consonant with the Guidelines as a whole. The clerk will send a copy of this opinion to the Chair and General Counsel of the Sentencing Commission.
With these thoughts, I join in the judgment of the Court.
. I note in this regard my agreement with the segment of Judge Fuentes’ opinion dealing with the upward adjustment for knowing endangerment of the victim(s) solvency.
. In another — and odd — bit of similarity between these two cases, both Astorri and Jarvis dated daughters of their fraud victims.