¶9 The question of law presented to this court is whether a circuit court’s imposition of a sentence using inaccurate information that the defendant was subject to a mandatory minimum five-year period of confinement is structural error or subject to the application of harmless error analysis…. If the latter, the question is whether the error in the present case was harmless.
¶10 We conclude that imposing a sentence under the erroneous belief that the defendant was subject to a five-year mandatory minimum period of confinement is an error subject to a harmless error analysis. The error is not a structural error, as the court of appeals stated. We further conclude that the error in the present case was not a harmless error. We affirm the decision of the court of appeals, but on different grounds, and remand the matter for resentencing.
From the beginning of the case through sentencing, everyone–prosecutor, defense counsel, judge–mistakenly believed Travis was subject to a mandatory minimum penalty of five years of confinement. (¶¶6, 12-13, 26). The error was spotted by Travis’s sharp-eyed postconviction lawyer, who moved for resentencing under State v. Tiepelman, 2006 WI 66, 291 Wis. 2d 179, 717 N.W.2d 1, arguing the circuit court relied on the inaccurate belief there was a mandatory minimum in sentencing Travis. (¶¶6, 19). The prosecutor conceded the charging documents cited the wrong statute and that the minimum did not apply; the circuit court agreed, but concluded the error was harmless and rejected Travis’s argument that the inaccuracy so fundamentally affected the structure of the entire proceeding that it should be considered per se prejudicial. (¶¶2, 7, 40). The court of appeals reversed and ordered resentencing, concluding the error was structural. 340 Wis. 2d 639, ¶¶21-24.
The supreme court agrees Travis is entitled to resentencing, but rejects the court of appeals’ structural error conclusion. Structural errors “seriously affect the fairness, integrity or public reputation of judicial proceedings and are so fundamental that they are considered per se prejudicial.” (¶54, quoting State v. Ford, 2007 WI 138, ¶42, 306 Wis. 2d 1, 742 N.W.2d 61). Few errors meet that standard. (¶56). Travis analogized the error in his case to the structural errors in State v. Shirley E., 2006 WI 129, 298 Wis. 2d 1, 724 N.W.2d 623 (denial of counsel); State v. Goodson, 2009 WI App 107, 320 Wis. 2d 166, 771 N.W.2d 385 (biased tribunal); and Sullivan v. Louisiana, 508 U.S. 275, 281-82 (1993) (deprivation of right to verdict beyond a reasonable doubt), but the supreme court concludes that inaccurate information at sentencing does not resemble those cases and so declines to extend the limited class of structural errors to the unusual situation in this case. (¶¶52-65).
The court also concludes the error in this case was not harmless. (¶86). The sentencing court repeatedly reminded itself and the parties about the mandatory minimum period of confinement. (¶¶74, 76). It also believed it was required to impose at least a five-year period of confinement; indeed, if the court believed the minimum penalty was mandatory but had not taken it into consideration, it was arguably in error. (¶¶78-79). Moreover, everyone’s mistaken belief about the minimum penalty skewed the sentence recommendations they made to the court. (¶¶81-82). Thus, “the error about the mandatory minimum period of confinement permeated the entire sentencing procedure,” (¶85), and Travis is entitled to resentencing.
Few errors are classified as structural, so the rejection of the court of appeals’ holding is not too surprising. Perhaps more important is the court’s Tiepelman analysis, under which the court determines, first, whether there was inaccurate information (basically undisputed at this point, with one exception noted below) and, second, whether the sentencing court relied on the inaccurate information; if those facts are established, the state has to prove the error was harmless. (¶¶21-23). Here are the conclusions that might be useful for analyzing (and making arguments in) other cases:
First, the court’s discussion of the reliance prong emphasizes the language of Tiepelman: “explicit attention” and “specific consideration” of the information showing it “formed part of the basis for the sentence.” (¶28). Here, the circuit court’s reference to the inaccurate penalty information was “explicit and repetitive.” (¶32). The lack of “magic words” connecting the inaccurate information to the sentence imposed isn’t dispositive (“For a reviewing court to conclude there was actual reliance by the circuit court, a circuit court need not have stated, ‘Because of the existence of this [inaccurate information], you are sentenced to X years of imprisonment.'”) (¶30). And the fact that other information might have justified the sentence is irrelevant when the court has relied on inaccurate information as part of the basis of the sentence. (¶47).
Next, the court’s harmless error analysis discounts the circuit court’s postconviction ruling that the sentence would have been the same even without the inaccurate information. A reviewing court is not bound by the circuit court’s retrospective review, so the focus must be on the sentencing transcript, not the circuit court’s assertions during the postconviction hearing or speculation about what a circuit court would do upon resentencing. (¶¶73, 77). Also, that Travis was sentenced to more than the minimum sentence is not determinative of harmless error: It is not the actual sentence that determines the constitutionality of the sentencing procedure, but whether the sentence “is based on a foundation of such materially inaccurate information that the proceedings are lacking in due process.” (¶84).
A final note: Justice Roggensack dissents, concluding Travis was subject to the minimum penalty. (¶¶91-115). The state made the very same argument in the court of appeals (despite the DA’s concession at the postconviction hearing), but the court of appeals rejected it. (¶3; see also 2012 WI App 46, ¶¶15-19, and here). The state didn’t renew the argument in its petition for review, so (as noted here) it was not before the supreme court, and the court of appeals’ rejection of the claim should retain its precedential value.