concurring in part and dissenting in part:
I concur in Parts I and II of the court’s opinion, but dissent from Part III. I would hold that the district court abused its discretion in departing upward regarding Connelly’s criminal history, and would therefore vacate that part of his sentence and remand for resentencing.
As the majority recognizes, the guidelines provide that when prior independent crimes are “consolidated for trial or sentencing,” the resultant sentences are counted as a single sentence for the purpose of criminal history calculations. USSG § 4A1.2, App. Note 3. Consolidated sentences are a common occurrence, and, by explicitly determining that they should be treated as single sentences, it is apparent that the Sentencing Commission intended upward departures on that basis to be the exception rather than the rule. The guidelines expressly state that prior, consolidated sentences must “significantly underre-present” the defendant’s criminal history or his propensity for recidivism to warrant an upward departure from the defendant’s “heartland” criminal history category. USSG § 4A1.3.
It follows directly from this “significant underrepresentation” requirement that consolidated sentences do not provide permissible grounds for upward departures unless they were imposed for “serious” offenses. In fact, Application Note 3 to guideline § 4A1.2 says as much. It explains the criminal history points assigned to a consolidated sentence “may not adequately reflect the seriousness of the defendant’s criminal history ..., [f]or example, if [the] defendant was convicted of a number of serious non-violent offenses committed on different occasions.” (emphasis added). This qualifier is reiterated in both other places the guidelines address the issue. First, guideline § 4A1.3, the controlling *986guideline in this case, provides that district courts may base an upward departure on “prior sentenee(s) of substantially more than one year imposed as a result of independent crimes committed on different occasions.” USSG § 4A1.3(b) (emphasis added). Second, § 4A1.3’s policy statement states that an upward departure would be permissible when a defendant “had received a prior consolidated sentence of ten years for a series of serious assaults.” (emphasis added).
Our prior cases also emphasize § 4A1.3’s “seriousness” requirement. In United States v. Brady, 928 F.2d 844 (9th Cir.1991), overruled in fart on other grounds by Nichols v. United States, 511 U.S. 738, 114 S.Ct. 1921, 128 L.Ed.2d 745 (1994), we reversed an upward departure based on two prior, uncounted convictions that resulted in sentences of less than 30 days. We explained that:
[T]he five examples listed by the [Sentencing] Commission as possible grounds for departure all make reference to previous “serious ” or “large-scale ” offenses which have not been included in the criminal history calculation. Neither of Brady’s tribal assault convictions, both sentences of less than 30 days, represent this level of seriousness.
Id. at 853. We similarly refused to permit an upward departure based on a prior, uncounted auto burglary conviction in United States v. Carrillo-Alvarez, 3 F.3d 316, 321-22 (9th Cir.1993), reasoning that “[h]is offense was neither ‘serious’ nor ‘large-scale.’ ” Id. at 322. The Tenth Circuit — the only other circuit of which I am aware that has directly addressed the issue — has also held that the guidelines’ provisions governing criminal history upward departures “require[ ] that our focus remain on distinguishing [uncounted] offenses to be regarded as ‘serious ’ from within the realm of all criminal behavior.” United States v. Wyne, 41 F.3d 1405, 1409 (10th Cir.1994) (emphasis added). Thus, although a couple of our cases have upheld departures without, in the words of the majority, “an explicit showing of seriousness,” maj. op. at 984,1 think it plain that both the text of the guidelines and precedent require that circumstance to be present.1
Applying this rule to the present case, I believe that the district court erred in basing an upward departure on each of Connelly’s prior, consolidated sentences. First, it seems obvious that Connelly’s prior, consolidated probationary period for forging a $2000 check is not serious enough to support a departure. Because the district court expressly increased Connelly’s criminal history category in part on the basis of that sentence, this “improper consideration” alone requires that we vacate and remand. See United States v. Donaghe, 50 F.3d 608, 611—12 (9th Cir.1994) (holding that “[f]or the upward departure to be valid, all [of the] factors [considered] must be proper bases for departure” and if one factor is invalid we must “ “vacate the sentence and remand for resentencing”’) (quoting United States v. Montenegro-Rojo, 908 F.2d 425, 428 (9th Cir.1990)) (emphasis added). A probationary period is less than the 30-day sentence we found inadequate in Brady, and is nowhere near the statute’s guideline that the prior sentence be “substantially more than one year imposed as a result of independent crimes committed on different occasions.” USSG § 4A1.3(b) (emphasis added).2
*987Second, Connelly’s prior sentence of 18 months for prior frauds involving between $ 2,500 and $28,000 is also not evidence of sufficiently “serious” prior criminal conduct to warrant a departure from the guidelines’ heartland. While the length of the prior sentence may not be entirely dispositive on this point, I do believe that it provides significant guidance, both because the guidelines themselves repeatedly refer to sentence length in measuring seriousness and because sentences are generally designed to reflect the seriousness of criminal conduct. An 18-month consolidated sentence for a few small-time frauds is not one that is “substantially more than one year” pursuant to “serious” offenses and is a far cry from § 4A1.3’s example — a 10-year sentence for a series of serious assaults. See Wyne, 41 F.3d at 1408 (reversing an upward departure based on prior, uncounted sentences of 13 months or less because, given the guidelines one-year guidepost, such crimes were not “evidence of ‘serious’ criminal conduct”). Nor, in my opinion, does failing to account for such a sentence “significantly underrepresent” the criminal history of a defendant such as Con-nelly who is already at a criminal history level of IV.
Finally, I disagree with the majority’s assertion that the district court’s departure is “justified purely on the basis of Defendant’s likelihood of recidivism.” Maj. op. at 985. While the text of section § 4A1.3 indicates that likelihood of recidivism alone might support an upward departure, the guidelines’ examples (which are mentioned above) do not differentiate between recidivism and underrepresented criminal history. This structure strongly suggests that both of the bases of § 4A1.3 departures must be supported by serious, uncounted criminal conduct.
But even assuming that an unusually high potential for recidivism not evidenced by serious, uncounted prior consolidated sentences could, in itself, support an upward departure, this surely is not such a case. Connelly has three uncounted prior offenses. Given that the guidelines expressly contemplate counting consolidated sentences together, I cannot see how such a small number of uncounted sentences removes him from the heartland of category TV, which is already a highly recidivist category; and neither the district court nor the majority points to any evidence that indicates that three “uncounted” sentences is unusual for a defendant in category IV. See Donaghe, 50 F.3d at 611 (holding that on review we do “‘not search the record for permissible reasons for departure; instead, [we] analyzef ] the reasons actually given by the district court’”) (quoting Montenegro-Rojo, 908 F.2d at 428). Nor does anything in the district court’s decision, or even the record as a whole, suggest that Connelly has been deterred by his prior sentences to a lesser degree than the typical defendant who has amassed a category IV criminal history. See id.
That leaves the majority’s final basis for affirmance: the similarity of Connelly’s uncounted conduct to his offense conduct. First, because the district court did not rely on this reason, the majority plainly errs in doing so. See id. While the majority acknowledges the existence of our rule requiring reversal in this circumstance, see maj. op. at 983, it simply fails to consider or apply it here, choosing instead simply to ignore the plain facts. Second, even if the district court had relied on this justification, a departure on this ground makes little, if any, sense. Someone who is in category IV is likely to have committed several offenses of the same general type. Here, the general type of offense identified by the majority is “stealing from people who trusted him,” maj. op. at 985 — in other words, non-violent economic offenses. It is true that individuals who commit fraud and earn enough criminal history points to achieve classification as category TV offenders usually have not varied their criminal careers by committing rape, murder, or mayhem. Thus, I would suspect that the *988defendant is a fairly typical economic crime or fraud offender of the category IV class. That is, of course, a reason for non-enhancement, not a justification for enhancing — and especially not a justification for enhancing in order to be able to impose the length of sentence the court has predetermined is appropriate when it is dissatisfied with the prescribed guideline sentence. And that, indeed, is what occurred here. The district court deemed in advance that in its view Connelly’s criminal conduct did not place him in “an acceptable range for sentencing” under the guidelines. Therefore, it decided to enhance. Yet neither the district court’s reliance on Connelly’s prior, uncounted sentences nor the majority’s improper addition of other inappropriate reasons provides a legal basis for an upward departure in Con-nelly’s case on the basis of potential recidivism.
Accordingly, I would vacate the upward departure and remand for resentencing.
. I do not understand the majority’s claim that the question of "seriousness” is an "unbriefed issue.” Connelly argued repeatedly in his brief that departure was warranted only "when a defendant's criminal history category significantly under-represents seriousness of his criminal history or likelihood that defendant will commit further crimes." Appellant's Brief at 20 (emphasis added); see also id. (arguing that departures are proper "only when Guidelines do not adequately reflect seriousness of an offender’s past criminal conduct”) (emphasis added). Thus, this issue is clearly before us.
. United States v. Myers, 41 F.3d 531 (9th Cir.1994), upon which the majority relies and which it asserts is our "most closely analogous” case on this point, is actually not analogous at all. In that case, the defendant’s "uncounted” criminal history did not even arise from a prior sentence. Rather, the uncounted conduct was a wire fraud that the defendant perpetrated after the conviction at issue, while awaiting sentencing. It was that critical and unique fact (i.e., the timing of events) — not the gravity of the fraud — that led us to conclude that "Myers’ post-offense criminal conduct is not only serious; it constitutes the strongest possible evidence of a likelihood that she will continue to commit similar crimes in the *987future.” Id. at 534; see also id. (emphasizing that unlike the Sentencing Commission’s explicit consideration of uncounted sentences prior to conviction, “[n]othing in the Guidelines indicates that the Sentencing Commission adequately considered a defendant's post-offense conduct as aggravating sentence warranting an upward departure"). Hence, Connelly’s prior probation for his fraud is decidedly different than Myers' because it had occurred several years before the instant sentencing and had already been completely adjudicated.