concurring.
I agree that the district court erred in ruling that defendant’s sixth bank robbery had not been “considered” under the Sentencing Guidelines when § 3D1.4 provides explicitly that five levels are to be added for “more than 5” units of offenses. (Section 3D 1.4 reflects a declining punishment scheme: 4 levels are added for to 5 units; 3 levels are added for 23é to 3, etc.). I write separately, however, because I disagree with what I understand to be the majority’s conclusion that a sentencing court should not use § 5K2.0 for upward departure in a case like this unless the court finds first that enhancement of the defendant’s criminal history is not appropriate. To the contrary, I think § 5K2.0 is available and that a sentencing court could look to § 3D1.4 for guidance in calculating a reasonable upward departure under § 5K2.0.
Section 5K2.0 is a policy statement on grounds for departure. It incorporates the intent of 18 U.S.C. § 3553(b) that sentencing courts may impose sentences outside the ranges established by the sentencing guideline, if the court finds “that there exists an aggravating or mitigating circumstance of a kind; or to a degree, not adequately taken into consideration by the Sentencing Commission....” The majority suggests that the use of § 5K2.0 is not appropriate in this case because the aggravating circumstance is not related to the offense of conviction. In my opinion, the number of times a person commits the same offense is a factor closely related to the underlying offense. See United, States v. Chase, 894 F.2d 488, 491 (1st Cir.1990) (large number of robberies committed by defendant is aggravating circumstance of type contemplated in 18 U.S.C. § 3553(b)). Section 5K2.0 contemplates departure even where a factor has been taken into account if the “factor is present to a degree substantially in excess of that which ordinarily is involved in the offense.”
A review of the guidelines shows that the Sentencing Commission did not take into consideration the kind of crime spree perpetrated by this defendant. Section 3D 1.4 makes no provision for 17 armed robberies: the table provides for adding only one level for any number “more than 5.” The commission acknowledges that a court would have to depart; it notes explicitly that a departure may be warranted “in the unusual case where the additional offenses resulted in a total significantly more than 5 Units.” See, e.g., United States v. Chase, 894 F.2d 488 (proper to depart upward using § 3D1.4 as guide where defendant committed 15 bank robberies).
The majority objects to using § 3D1.4 to make an offense level adjustment using conduct to which the defendant has not pleaded, and adds that the additional robberies are not the sort of uncharged conduct that would result in an offense level adjustment, because the Sentencing Commission treats armed robberies and other crimes of violence as separate offenses and does not group them with other relevant conduct as it does crimes involving money or drugs. I have no quarrel with either of these propositions. I simply do not agree that a sentencing court would be making an offense level adjustment if it used § 3D1.4 as a guide to structuring an upward departure, once it has determined that a departure is proper. Then, § 3D1.4 would be an apt source of guidance for dealing with numerous contemporaneous offenses such as the additional armed robberies at issue in this case. The manner in which the commission weighs such offenses offers sen*467tencing courts useful guidance in determining how to depart once the decision has been made to do so.
My reading of United States v. Schmude, 901 F.2d 555 (7th Cir.1990), is that this court approved the extrapolation of criminal history categories as one way of fashioning an upward departure but that it did not rule out every other approach. Indeed, this court said that “the question of degree of departure is solely one of reasonableness,” to be afforded a deferential standard of review, id. at 560, and added that reasonableness implies an effort “to fashion the degree of departure to correspond to the number and the nature of the factors which warrant departure.” Id.
The majority implies that reference to the offense table in § 3D1.4 for calculating upward departures would work to the detriment of the defendant. In fact, if done properly, it may have the opposite effect. According to U.S.S.G. Part D, Introductory Commentary, the table in § 3D1.4 is designed to produce incremental punishment for significant additional criminal conduct, with the amount of additional punishment declining as the number of crimes increases. (This is why the table assigns a three level increase for 2$ to 3 units of offense; a four level increase for to 5 units of offense; and only five levels for more than 5.) A court calculating an upward departure should not depart upward the equivalent of one entire offense level for each additional offense but should try to conform its departure to the declining punishment scheme in the table. Under this scheme, five additional offenses could translate reasonably into a two level upward departure. Were the district court to use this approach and consider only the additional federal robberies, the proper sentence would be in the range of 210-262 months. Starting with the offense level of 28 calculated by the probation office and a criminal history category VI, the court would add the equivalent of two additional criminal history categories for defendant’s inadequate criminal history and the equivalent of two offense levels for the aggravating circumstance of the five additional federal bank robberies that were not taken into consideration by the Sentencing Commission.
Alternatively, the district court could use the methodology approved in Schmude, 901 F.2d 555, which is based on inadequacy of criminal history, § 4A1.3. Schmude contemplates that once a court has determined that an upward departure is warranted, the court will assess additional criminal history categories for the factors justifying departure. If the defendant is already in the highest category, then the court will extrapolate a criminal history category. Because each increase in a category represents roughly a ten to fifteen percent increase in the sentencing range, the sentencing court can depart upward to the range that corresponds to the hypothetical category it has selected.
It would be premature to say whether it would be reasonable for the district court to assess defendant one criminal history category increase for ea'ch of the five additional federal bank robberies to- which he did not plead.1 If the court did make such an assessment, defendant might be facing a maximum guideline sentence of 292-365 months. His adjusted criminal history category would be the equivalent of a category XIII, after starting with his original category of VI and adding the two-category increase for the consolidated state court sentences and a five-category increase for the federal bank robberies. At least in theory, defendant could be worse off under the Schmude methodology. However, it is arguable that it would be a violation of the spirit of the declining punishment policy of the guidelines to apply that methodology on a one-category increase for each crime basis in a situation like this where the court is dealing with closely related crimes.
*468In summary, I agree with the majority that this case must be remanded to the district court for resentencing, although I disagree with the implication that the district court cannot depart upward under § 5K2.0 unless it first determines that enhancement of criminal history is inappropriate. Also, I believe it would be proper for the district court to look to § 3D1.4 for guidance in structuring a reasonable upward departure for this defendant, taking into consideration the declining punishment scheme reflected in that section.
. It is arguable that assessing defendant one criminal history category increase for each federal bank -robbery is not reasonable, given the incremental punishing approach followed throughout the Sentencing Guidelines. Not only is this true of § 3D 1.4; it is reflected as well in § 4A1.1 (f) (criminal history). Under § 4Al.l(f), a sentencing court is limited to 3 points when adding points for prior sentences resulting from a conviction of a crime of violence that did not receive points under other sections of § 4A1.1 - because the sentence was considered related to another one resulting from a conviction of a crime of violence.