People v. Gutierrez

278 Cal. Rptr. 748 | Cal. Ct. App. | 1991

227 Cal.App.3d 1634 (1991)
278 Cal. Rptr. 748

THE PEOPLE, Plaintiff and Respondent,
v.
MICHAEL S. GUTIERREZ, Defendant and Appellant.

Docket No. A049255.

Court of Appeals of California, First District, Division Two.

March 1, 1991.

*1635 COUNSEL

Alison Hardy, under appointment by the Court of Appeal, for Defendant and Appellant.

*1636 John K. Van de Kamp and Daniel E. Lungren, Attorneys General, Richard B. Iglehart, Chief Assistant Attorney General, John H. Sugiyama, Assistant Attorney General, Morris Beatus and Matthew R. Golde, Deputy Attorneys General, for Plaintiff and Respondent.

OPINION

BENSON, J.

Michael Gutierrez appeals under Penal Code section 1237, from the sentence imposed on him in case No. 16778-C for violations of Penal Code[1] sections 459 (residential burglary), 496 (receiving stolen property), and in case No. 16135-C for violation of section 666 (theft). Appellant was found guilty of residential burglary and receiving stolen property in a jury trial, and in a separate and unrelated case, pled guilty to the theft charges. The sole issue on appeal is whether the trial court's failure to state reasons for imposing a consecutive sentence for the unrelated theft offense constitutes harmless error. Appellant also requests that the abstract of judgment be modified to reflect the correct amount of credit for time served as shown by the reporter's transcript of the sentencing hearing.

Statement of the Case

On January 9, 1990, following a plea of guilty, appellant was sentenced to four years in state prison for violation of section 666 with a one-year enhancement imposed under section 667.5, subdivision (b), for a prior prison sentence. This theft offense had occurred on January 21, 1989.

On December 12, 1989, an information was filed in the Sonoma County Superior Court charging appellant with two counts of burglary in violation of section 459, two counts of possession/receiving stolen property in violation of section 496, and alleging one prior prison term within the meaning of section 667.5, subdivision (b). The burglary occurred on November 3, 1989. The stolen property was found in appellant's car on November 9, 1989.

On January 19, 1990, appellant filed a motion to suppress evidence gathered in two searches of his car. This motion was denied.

Appellant waived his right to a jury trial on the truth of the prior prison term, and his trial was bifurcated. After a six-day jury trial, appellant was convicted of one count of residential burglary and one count of receiving *1637 stolen property. He was acquitted of the other charges. The trial court found true the charged prior prison term.

Sentencing

On March 26, 1990 appellant was sentenced on the burglary and receiving stolen goods charges, and resentenced pursuant to section 669 on the prior theft charge.

The prosecutor requested that the court follow the suggestion of the probation department and sentence appellant to eight years and four months in state prison. The defense requested an alternative sentence of a middle term for the first count, the residential burglary; a year for the prior prison term; eight months concurrent or consecutive for the receiving stolen property; and eight months consecutive for the prior theft.

The court denied probation under section 462, subdivision (a), finding no unusual circumstance and noting that appellant's prior performance on probation had been unsatisfactory. The court also found appellant's "prior convictions as an adult and adjudications, and commissions of crimes as a juvenile, are numerous." The court then stated, "And based upon that the Court finds that the aggravated term for Section 459 is appropriate," sentencing appellant to six years on that count. The court then sentenced appellant to one third the midterm, eight months, to run concurrently on the receiving stolen property count. (This sentence was later amended to two years, concurrent and stayed.) The court determined the term would run concurrently because the crimes were so close in time as to indicate a single period of aberrant behavior. The court sentenced appellant to one year under section 667.5 for the prior prison term, for a total of seven years.

The court then stated "[a]nd then in file number 16135, the 666, the court orders one third the consecutive time period; so that would be seven years, eight months total. Further the court finds that 666 is the — that the 666 is the subordinate term, that the principal term is to be the 459." No further statements were made with regard to the sentence imposed.

Discussion

I. Whether the Judge's Failure to State Reasons For Imposing a Consecutive Sentence Was Harmless Error

As we previously noted, in January 1990, appellant was sentenced to four years in prison in case No. 16135-C for theft. In imposing a different sentence in March 1990, the judge was required to specify how the second *1638 sentence should run with respect to the prior sentence. Section 669 allows the court the option of running a sentence concurrently or consecutively. If imposing a sentence consecutively, the court was required by section 1170.1 to aggregate the sentences into one term of imprisonment. This aggregate term is to be composed of the sum of a principal term, a subordinate term, and any enhancements imposed under section 667.5. The subordinate term for a consecutive offense is to consist of one-third of the middle term of imprisonment prescribed for that offense.

Here, the judge determined that the principal term would be the residential burglary, and the subordinate term would be the prior theft. However, the court did not state reasons for deciding to run the theft sentence consecutively instead of concurrently.

(1) "A decision to impose consecutive sentences is a sentencing choice for which a statement of reasons is required. [Citations omitted.]" (People v. McLeod (1989) 210 Cal. App.3d 585, 590 [258 Cal. Rptr. 496].) This holds true even where there have been two separate trials, and the defendant has already been sentenced in one of the cases. (People v. Stone (1981) 117 Cal. App.3d 15, 21-22 [172 Cal. Rptr. 445].)

(2) However, "a failure to state reasons is not prejudicial error per se: If the error is harmless the matter need not be remanded for resentencing." (People v. McLeod, supra, 210 Cal. App.3d at p. 590.)

"In order to determine whether error by the trial court [in making a sentencing choice] requires remanding for resentencing `the reviewing court must determine if "it is reasonably probable that a result more favorable to the appealing party would have been reached in the absence of error."'" (People v. Avalos (1984) 37 Cal.3d 216, 233 [207 Cal. Rptr. 549, 689 P.2d 121].)

(3a) The question in the present case is whether it is reasonably probable that the trial court would run the theft sentence concurrently if the matter were remanded.

California Rules of Court, rule 425 provides the criteria affecting a decision for imposing consecutive rather than concurrent sentences. Under rule 425(a) the court considers facts relating to the crimes, including in part: (1) whether or not the crimes and their objectives were predominately independent of each other; and, (2) whether the crimes were committed at different times or separate places, rather than being committed so closely in time as to indicate a single period of aberrant behavior.

*1639 The presence of both these factors suggests imposition of a consecutive sentence in the present case. The theft occurred on January 21, 1989; the residential burglary and receiving stolen property crimes occurred on November 3 and 9, 1989. The January and November crimes were obviously independent of each other, and were committed at different times and places.

Under California Rules of Court, rule 425(b), the court may also consider any circumstances in aggravation or mitigation in making a consecutive/concurrent decision. In the present case however, there was no mitigating factor. The probation reports from both cases, No. 16778-C and No. 16135-C, list numerous factors in aggravation, and "none noted" under factors in mitigation.

In a situation such as this, where the defendant was sentenced in one case months earlier, where that crime occurred at a different time and place, and where there was no mitigating factor, the court had obvious reasons for imposing the sentence consecutively, based on the factors listed in California Rules of Court, rule 425(a).

(4) In People v. Callahan (1983) 149 Cal. App.3d 1183, 1187 [198 Cal. Rptr. 12], the court held that the same fact may not be used to impose an upper term and a consecutive term, even when there is more than one sentencing proceeding. Where the court does not state reasons for imposing a consecutive sentence, a reviewing court must be sure that the sentencing judge did not use a fact to impose a consecutive sentence that was used earlier to impose an upper term. But, as the court points out in Callahan, this problem does not occur where the reasons for imposing a consecutive term could not "conceivably" have been used to impose an upper term. (Ibid.)

(3b) Here, the facts possibly used for imposing a consecutive sentence, could not have been used to also impose an upper term. The obvious reasons for imposing a consecutive sentence here were that the crimes were independent of each other and occurred at different times and places. In imposing an upper term for the section 459, the court stated that the decision was based on appellant's prior criminal history and poor performance on probation. In imposing an upper term for the theft at the earlier hearing, the sentencing court there could not have considered these factors since the present case had not yet resulted in conviction. Thus, the rule prohibiting the dual use of facts could not have been violated.

Appellant cites People v. Edwards (1981) 117 Cal. App.3d 436 [172 Cal. Rptr. 652] for support. In Edwards the court remanded for resentencing where the trial court did not state reasons for imposing a consecutive sentence. The defendant had been sentenced to two years for a Madera County *1640 burglary conviction. When sentencing the defendant in a later case, the court stated that the sentence for the Madera conviction would be the subordinate term and would run consecutively to the newly imposed sentence. The court reduced the Madera sentence from two years to eight months. The appellate court remanded because there was nothing in the record to indicate that the sentencing judge was cognizant of any circumstances of the Madera offense upon which he could rely to impose a consecutive sentence. (Id. at p. 450.)

The present case is distinguishable. Here, the probation report, read and signed by the sentencing judge, referred to the prior theft sentence. Also, the judge here had a copy of the probation report from the prior case. This earlier probation report related all the details of the theft offense, described appellant's extensive criminal history, and listed numerous circumstances in aggravation and none in mitigation. From this, the sentencing judge was obviously cognizant of sufficient circumstances from the prior sentence upon which he could rely to impose a consecutive sentence. The factors that concerned the court in Edwards do not cause concern in the case at hand.

Given that appellant had already been sentenced to an upper term for the earlier crime, that the two cases were based on incidents occurring months apart, that appellant's sentence for the theft has been drastically reduced, and that the judge had obvious reasons for imposing a consecutive sentence, we conclude it is not reasonably probable that resentencing would result in a more favorable disposition for appellant. Consequently, the court's error in failing to state reasons for imposing the sentence consecutively, was harmless.

II. Whether the Abstract of Judgment Accurately States the Credit for Time Served as Ordered by the Court

At the sentencing hearing the court stated that appellant's total credit would be 93 days. Appellant pointed out that he should also receive credit for time already served in case No. 16135-C. The court ordered a total of 208 days credit for time served in that case. The abstract of judgment correctly reflects the 208 days of credit in case No. 16135-C, and also shows 93 days credit for time served in case No. 16778-C. The abstract of judgment correctly reflects the credits ordered at the time of the hearing. Appellant's request for modification is denied.

Disposition

The judgment is affirmed.

Peterson, J., concurred.

*1641 KLINE, P.J., Concurring.

I write separately for two reasons. First, I do not believe the case law supports application of the standard of reversal articulated by the majority. The rule which seems to be reflected in the cases is that the failure to state reasons for a sentencing choice requires reversal if there is any reasonable possibility a statement of reasons would have produced a more lenient sentence or would have disclosed reversible substantive error. I concur in the result because, applying this standard to the facts of this case, I perceive no reasonable possibility the trial court would have imposed concurrent sentences under the facts presented, or that a statement of reasons would have disclosed reversible substantive error in the court's sentencing calculus.

Second, although I feel compelled by existing case law to apply some harmless error rule in this context, I seriously question the wisdom of doing so. For the reasons stated in part III below, I believe the better rule would call for reversal in all cases where this error occurs.

I.

A basic theme running through the determinate sentencing law, Penal Code sections 1170 et seq. (DSL),[1] is that the trial court's discretion in sentencing is to be limited and controlled. Subdivision (a)(1) of section 1170 declares that the DSL's mechanism — proportional and determinate sentences — is best achieved through the exercise by trial judges of "specified discretion." Determinate sentencing arose from the widespread perception that "significant disparities in the punishment imposed upon like individuals committing like offenses was a pernicious evil endangering the very integrity of the criminal justice system.... `Numerous empirical studies revealed that the chief explanation for sentencing disparity was not the differences in defendants but the differences in judges.... Thus the movement to promote uniformity in sentencing ... was in no small part a movement to diminish judicial discretion.'" (People v. Martin (1986) 42 Cal.3d 437, 442-443 [229 Cal. Rptr. 131, 722 P.2d 905].)

One of several statutory mechanisms for controlling judicial discretion is the requirement in question here — that the trial court state reasons for its "sentence choice." (§ 1170, subd. (c); see §§ 1170, subd. (a)(2), 1170.1, subd. (d); Cal. Rules of Court, rules 405(f), 425, 433(c)(3), 443.) The Legislature undoubtedly believed that such a statement would (1) help to ensure meaningful review; (2) act as a guard against careless decisions; (3) encourage the trial judge to make sentencing choices systematically, in full recognition of the applicable considerations; and (4) "preserv[e] public *1642 confidence in the decision-making process by helping to persuade the parties and the public that the decision-making is careful, reasoned and equitable." (People v. Martin, supra, 42 Cal.3d at p. 450; see In re Podesto (1976) 15 Cal.3d 921, 937 [127 Cal. Rptr. 97, 544 P.2d 1297]; People v. Brandt (1987) 191 Cal. App.3d 143, 147 [236 Cal. Rptr. 258]; People v. Whitehouse (1980) 112 Cal. App.3d 479, 486 [169 Cal. Rptr. 199] [absence of statement obstructs goal of uniformity]; People v. Ibarra (1982) 134 Cal. App.3d 413, 426 [184 Cal. Rptr. 639] [obstructs effective review]; People v. Enright (1982) 132 Cal. App.3d 631, 636 [183 Cal. Rptr. 249] [adequate statement enables parties to correct misconceptions at time of sentencing].)[2]

II.

The majority opinion declares that the failure to state reasons constitutes reversible error only if the reviewing court finds it "reasonably probable that a result more favorable to the appealing party would have been reached in the absence of error." (Maj. opn., ante, at p. 1638.) For this proposition the majority cites People v. Avalos (1984) 37 Cal.3d 216, 233 [207 Cal. Rptr. 549, 689 P.2d 121]. The relevant discussion in Avalos is anything but clear. In the sentence quoted by the majority, the court ostensibly adopted the familiar test adopted in People v. Watson (1956) 46 Cal.2d 818, 837 [299 P.2d 243], i.e., reversal is warranted only where it is "reasonably probable" that a more favorable result would have been reached in the absence of the error. In the next paragraph, however, the court said that reversal is required if the reviewing court "cannot determine whether the improper factor was determinative for the sentencing court." (Ibid.) If reversal follows where the court "cannot determine" the likelihood of a more favorable result, i.e., where a more favorable result does not appear "probable," then something other than a true Watson standard is at work.

Nor is Avalos on point. Neither it nor any other Supreme Court decision addresses the standard of reversal for an erroneous failure to state reasons under the DSL. In Avalos the error was a "dual use of facts," i.e., citing the same sentencing factors to impose both an aggravated term and a consecutive sentence. (37 Cal.3d at p. 233.) The Supreme Court was able to detect this error only because the trial court had fully stated the reasons for its choice. This also permitted the court to conclude that "the improper dual use of facts was not determinative." (Ibid.)

*1643 Part of the vice in a complete failure to state reasons is that any attempt to ascertain its effect calls for compound speculation and subjective assumptions about the trial court's mental processes. (See People v. Reiley (1987) 192 Cal. App.3d 1487, 1491 [238 Cal. Rptr. 297] [refusing to "`speculat[e] on the probability the court will impose the same sentences on remand'"].) Of course, this is precisely what the statement of reasons is supposed to avoid. Moreover, there will always appear to be at least an even chance that the court would have imposed the same sentence even without the error. If the appellant must show that a more favorable result was "probable," one can hardly imagine a record on which an appeal could succeed. In effect, the error becomes self-insulating: the trial court's silence prevents effective review; the appellate court's inability to review the error precludes reversal. Literal application of Watson would thus eviscerate a legislative safeguard central to the DSL.

Although Watson states the general rule, it is not universally applicable. Some errors are serious enough to require reversal regardless of the likelihood of a more favorable result. (See People v. Watson, supra, 46 Cal.2d at p. 835; 6 Witkin & Epstein, Cal. Criminal Law (2d ed. 1989) Reversible Error, §§ 3303-3311, pp. 4084-4095.) And there appears to be at least one intermediate standard. As suggested in Avalos, supra, reversal is sometimes mandated when the reviewing court cannot determine the effect of the error. And elsewhere the Supreme Court has suggested that reversal may be warranted based on a "reasonable possibility" that the error affected the outcome. (See People v. Heishman (1988) 45 Cal.3d 147, 101 [246 Cal. Rptr. 673, 753 P.2d 629], cert.den. 488 U.S. 948 [102 L.Ed.2d 369, 109 S.Ct. 380] [failure to state reasons for denying automatic death penalty modification motion].)

The Fifth District recently adopted the "no reasonable possibility" standard in connection with the error involved here, i.e., failure to state reasons under the DSL. (People v. May (1990) 221 Cal. App.3d 836, 839 [270 Cal. Rptr. 690].) I believe this holding correctly reflects the de facto standard applied in the vast majority of published cases. It is true that a number of courts have cited Watson in this setting, but a great number of these cases have reversed and remanded for resentencing. As previously noted, I believe such a result is virtually precluded by a Watson standard. Nor have any of these cases actually declared that a more favorable result was "probable." Instead, they require resentencing when the reviewing court finds "enough likelihood ... that correction of the trial court's ... errors would result in [a] more lenient sentence." (People v. Jackson (1987) 196 Cal. App.3d 380, 392 [242 Cal. Rptr. 1].) Harmless error will not be found "in the face of mitigating factors which present a basis for a less severe sentence choice." (3 Witkin & Epstein, Cal. Criminal Law (2d ed. 1989), Punishment for Crime, *1644 § 1455, p. 1724; see People v. Burke (1980) 102 Cal. App.3d 932, 945 [163 Cal. Rptr. 4] [sentence was "supportable" but "not a foregone conclusion"]; People v. Anjell (1979) 100 Cal. App.3d 189, 204 [160 Cal. Rptr. 669] [court could not ascertain why consecutive sentences imposed in face of mitigating circumstances]; People v. Bobb (1989) 207 Cal. App.3d 88, 97 [254 Cal. Rptr. 707], review den. [presence of mitigating factors reflected in selection of lower term]; In re Spears (1984) 157 Cal. App.3d 1203, 1211-1214 [204 Cal. Rptr. 333], review den. [remand required, particularly where record reflects mitigating factors which would have supported more lenient sentence and court arguably relied on impermissible factor].)[3]

The vast majority of cases affirming despite the error are consistent with a rule under which the error would be reversible if there were a reasonable possibility of a more lenient sentence. One court found not only that a more favorable result was improbable but that there was "no reasonable possibility" of a less severe sentence. (People v. Preyer (1985) 164 Cal. App.3d 568, 577 [210 Cal. Rptr. 807].) Another found it a "virtual certainty" that the same sentence would be entered on remand. (People v. Bravot (1986) 183 Cal. App.3d 93, 98 [227 Cal. Rptr. 810].) In the widely cited case of People v. Blessing (1979) 94 Cal. App.3d 835, 839 [155 Cal. Rptr. 780], the court found that the failure to state reasons had no effect on the sentence because "defendant's crimes met every criteria [sic] listed in the rule and there is a total absence of any mitigating circumstances." Other cases finding harmless error have noted the presence of many aggravating factors and few or no mitigating ones. (See People v. McLeod (1989) 210 Cal. App.3d 585, 590-591 [258 Cal. Rptr. 496], review den. [many prior convictions and factors in aggravation, none in mitigation]; People v. Scott (1988) 200 Cal. App.3d 1090, 1095-1096 [246 Cal. Rptr. 406] [numerous priors, etc.; one possible mitigating factor apparently disbelieved by trial court]; People v. Porter (1987) 194 Cal. App.3d 34, 39 [239 Cal. Rptr. 269] [nine priors, eight aggravating factors, none mitigating]; People v. Preyer, supra, 164 Cal. App.3d 568, 577 [five aggravating, none mitigating]; People v. Swanson (1981) 123 Cal. App.3d 1024, 1034 [176 Cal. Rptr. 915] [six aggravating, none mitigating]; People v. Hartsfield (1981) 117 Cal. App.3d 504, 509-510 [172 Cal. Rptr. 794] [two aggravating, none mitigating]; see also People v. Mobley *1645 (1983) 139 Cal. App.3d 320, 324-325 [188 Cal. Rptr. 583] [no unusual circumstances negating statutory ineligibility for probation].) Other cases have noted the great unlikelihood of a more favorable result given the trial court's apparent intent to sentence severely. (E.g., People v. Smith (1984) 155 Cal. App.3d 539, 546 [202 Cal. Rptr. 259] [court stated intent to sentence to maximum term allowed by plea bargain]; People v. Bravot, supra, 183 Cal. App.3d 93, 98; but see People v. Reeder (1984) 152 Cal. App.3d 900, 912, 916-917 [200 Cal. Rptr. 479] [remanded despite stated intent to sentence severely and ample basis for such a sentence].)[4]

It therefore appears to me that while an impressive preponderance of case law supports application of some harmless error rule, the standard actually applied is not the "probable more favorable result" test of Watson. Instead, reversal is mandated if there is a "reasonable possibility that a statement of reasons would have altered the trial judge's conclusion or revealed reversible error...." (People v. May, supra, 221 Cal. App.3d 836, 839.) That is, the error is reversible if the record indicates that the sentencing judge could reasonably have imposed a less severe sentence.

Here, I concur in the result because I detect no reasonable possibility that the sentencing judge would have ordered defendant's sentence on the theft conviction to run concurrently with his sentence on the burglary conviction. It is apparent that the primary factor supporting consecutive sentences was that the two offenses were separate and independent. (See Cal. Rules of Court, rules 425(a)(1), 435(a)(3).) That is, the crimes were separated in time by more than 10 months and there is no apparent connection between them other than the identity of the perpetrator. The court stated its reasons for imposing an aggravated sentence on the principal term; even if we assume for purposes of analysis that the court relied on some of these same reasons to impose consecutive sentences (see rule 425(b)), it did not make dual use of the primary factor favoring consecutive sentences, i.e., that the offenses were separate and independent. The record is replete with aggravating circumstances and defendant identifies no mitigating circumstances which might reasonably favor a more lenient result. Accordingly, I believe that under the rule stated in People v. May, supra, 221 Cal. App.3d 836, 839, the error is not reversible.

*1646 III.

Although I feel obliged by the weight of precedent to adhere to the analysis just outlined, I doubt the wisdom of applying any rule of harmless error to cases of this type. The considerations which usually justify the application of such a rule do not support it here. They would be better served by making error of this type reversible per se.

The harmless error rule should extend no further than its rationale, i.e., that it is a waste of time and resources to reverse a decision where it does not appear "reasonably probable" that reversal will produce a more favorable outcome. But this assumes that fewer resources will be consumed in the appellate process than would be consumed as a result of remand. Typically this is true. Here, it is not. Instead, the time and resources consumed in the appellate search for harmless error exceed those which would typically be consumed in an automatic remand and resentencing. Reversal per se would also serve important societal and individual interests. Therefore, these cases warrant an exception to the general rule.

This case illustrates the point. Even though we have affirmed the judgment, the issue presented would have been resolved more efficiently by remand and resentencing without consideration of the effect of the error. To present the issue to us, the parties had to review and analyze the record and precedents and prepare written statements of their respective analyses. To properly assess those arguments, we had to conduct our own review of the record and undertake a speculative analysis by which we attempt to determine what the trial court "must have" or "might have" had in its mind, and the likelihood that a different result "could have" been reached. This exercise almost certainly consumed more resources than would have resulted from per se reversal upon identification of the error. Reversal would not have required a retrial, but at most the preparation of a new probation report, transportation of the defendant to court, a review by judge and counsel of their files, and a hearing, of probably less than 10 minutes, at which counsel would offer their recommendations and the court would impose sentence. The error was not likely to be repeated; further appellate review, at least on this ground, would have been obviated. A per se rule requiring remand would also provide a strong incentive for trial judges to remember to state reasons at the original sentencing.

At the appellate stage there is usually little question whether the trial court erred; the issue on appeal is limited to the likelihood the error affected the outcome. The parties and the appellate court spend far more time puzzling over this issue than the parties and the trial would have spent correcting it. This simple fact renders inapplicable the usual calculus *1647 supporting application of the doctrine of harmless error. And while there may be individual cases where the balance of judicial economies is different, there are at least as many more where we search for harmlessness in vain and must remand for resentencing after all. It seems plain that, viewing the judicial system as a whole, the doctrine of harmless error in this context does not eliminate, but creates, wasted effort and delay.

As noted, the Supreme Court has not addressed the precise question with which we are concerned. Little guidance is found in capital cases involving the failure to state reasons for denying automatic motions to modify the penalty. In three such cases, the court remanded for reconsideration without considering whether the error was harmless. (People v. Sheldon (1989) 48 Cal.3d 935, 962-963 [258 Cal. Rptr. 242, 771 P.2d 1330]; People v. Brown (1988) 45 Cal.3d 1247, 1263-1264 [248 Cal. Rptr. 817, 756 P.2d 204]; People v. Rodriguez (1986) 42 Cal.3d 730, 792-795 [230 Cal. Rptr. 667, 726 P.2d 113].) In a fourth case the court affirmed despite the error, finding "no reasonable possibility" that the trial court's decision had been affected. (People v. Heishman, supra, 45 Cal.3d 147, 200-201.) However, the meaning and viability of that holding are open to question. In Sheldon the court expressly refused to consider a harmless error argument, distinguishing Heishman on the ground that the trial judge there had died; otherwise, the court said, that case too would have been remanded. (People v. Sheldon, supra, 48 Cal.3d at p. 962.) The court then stated that if the original trial judge was unavailable the motion could be heard by another judge. (Id., at p. 963; see People v. Brown, supra, 45 Cal.3d at p. 1264, fn. 7.) If this was true in Sheldon it would seem to have been equally true in Heishman. It is therefore unclear how, or whether, these cases can be harmonized.

In any event, these cases offer a poor analogy to noncapital cases because they involve a wholly different balance of interests and efficiencies. The court has suggested in these cases that reversal is required partly because the severity of the penalty warranted an "abundance of caution." (People v. Sheldon, supra, 48 Cal.3d at p. 962, quoting People v. Heishman, supra, 45 Cal.3d at p. 200.) "Caution" is of course warranted in noncapital sentencing as well, but it is not the primary basis of my belief that per se reversal is warranted. Rather, automatic reversal offers benefits in noncapital cases which are not available in death penalty cases. In the latter, before the court reaches any issue of harmless error it has typically reviewed the massive record of a lengthy trial with a separate penalty phase. Having considered in detail the evidence supporting the penalty, it can effect little or no savings of appellate resources by remanding. In contrast, the error in noncapital cases is easily detected and is frequently the only penalty related point urged on appeal. If reviewing courts were not required to consider the asserted harmlessness of the error, they would not have to review the penalty record at all *1648 except to determine whether the court indeed failed to state reasons for the sentence.

In these respects the instant case is more comparable to civil nonjury cases where the trial court erroneously fails to explain its decision. (See Code Civ. Proc., § 632.) Under the former rule requiring findings of fact and conclusions of law, failure to make a finding on a material fact was reversible per se. (9 Witkin, Cal. Procedure (3d ed. 1985) Appeal, § 365, p. 367; see San Jose etc. Title Ins. Co. v. Elliott (1952) 108 Cal. App.2d 793, 801 [240 P.2d 41].) The same rule has been carried over to the current requirement of a "statement of decision": where the statement is required, failure to provide it is reversible per se. (Miramar Hotel Corp. v. Frank B. Hall & Co. (1985) 163 Cal. App.3d 1126, 1129-1130 [210 Cal. Rptr. 114]; see In re Marriage of S. (1985) 171 Cal. App.3d 738, 746 [217 Cal. Rptr. 561] [affected order "void"]; Lavine v. Hospital of the Good Samaritan (1985) 169 Cal. App.3d 1019, 1029 [215 Cal. Rptr. 708] [reversing order for sanctions under Code of Civ. Proc., § 128.5].) The modern application of this rule rests largely on considerations of judicial economy, as explained in Miramar Hotel Corp. v. Frank B. Hall & Co., supra, 163 Cal. App.3d 1126, 1130: "Were we ... to condone a total or even a material failure by trial courts to observe the prescribed procedure for revealing the basis for their respective decisions, we would be thrusting a quite substantial burden upon the litigants and also upon the appellate courts.... [W]here ... an inadequate statement or no statement whatsoever has been provided, then each appeal is inevitably based upon what is tantamount to a claim that the judgment is not supported by substantial evidence. This in turn requires both the litigants and the appellate court to conduct an examination of the entire record in order to properly review the trial court decision."

Beyond the immediate burden resulting from harmless error review in such cases is the possibility that an appellate willingness to save judgments from the effects of such error actually increases its frequency. Similar concerns prompted Justice Spencer's concurrence in Miramar Hotel Corp., supra. Noting that noncompliance with the applicable statute seemed to be on the increase, she wrote, "The far-reaching and burdensome effects of that practice mandate that it end immediately. Since I perceive no means of effecting that result other than per se reversal, I join with the majority." (163 Cal. App.3d at pp. 1130-1131.)

The interests served by the requirement of a statement of reasons under the DSL seem even more compelling than those served by the requirement in civil cases. The private interest at stake is personal liberty, and the requirement directly serves a public interest — proportional and determinate sentences — whereas a statement of decision in civil cases concerns the *1649 public only insofar as it aids efficient utilization of appellate resources. Furthermore, the efficiencies realized through per se reversal are likely to be greater: remand is more likely to obviate a further (colorable) appeal in the sentencing context than in a typical civil case.

In short, I see no virtue in a rule which requires us to parse the record when the sentence can be corrected by the trial court far more readily and easily than it can be salvaged by us. For the courts of appeal to accept such a burden out of a misplaced sense of obligation works a disservice to every concrete policy involved. If I were writing on a clean slate, I would hold error of this kind reversible per se.

Appellant's petition for review by the Supreme Court was denied May 22, 1991. Mosk, J., Broussard, J., and Kennard, J., were of the opinion that the petition should be granted.

NOTES

[1] All statutory citations are to Penal Code unless otherwise specified.

[1] Unless otherwise noted, all statutory citations are to the Penal Code.

[2] One would also expect the statement to assist the Board of Prison Terms in carrying out its mandate to review all sentences for disparity. (See § 1170, subd. (f).) When a sentence seems statistically disparate, the board "reviews the case record and the probation report, looking for ... nonquantifiable factors which might justify an especially severe sentence." (People v. Martin, supra, 42 Cal.3d at p. 445.) Presumably, a statement of reasons identifying such factors may obviate a finding of disparity and the resulting procedure to consider resentencing.

[3] A few courts have remanded while refusing to consider whether the error was harmless. (E.g., People v. Reiley, supra, 192 Cal. App.3d 1487, 1491 [court would not "`speculat[e] on the probability the court will impose the same sentences on remand'"].) Most of these cases came out of the Fifth District. (See People v. Hernandez (1988) 204 Cal. App.3d 639, 655 [251 Cal. Rptr. 393], review den.; People v. Brandt, supra, 191 Cal. App.3d 143, 147; People v. Manning (1982) 133 Cal. App.3d 159, 170 [183 Cal. Rptr. 727]; People v. Bejarano (1981) 114 Cal. App.3d 693, 704-705 [173 Cal. Rptr. 71]; People v. Edwards (1981) 117 Cal. App.3d 436, 449-450 [172 Cal. Rptr. 652]; contra People v. Romero (1985) 167 Cal. App.3d 1148, 1152 [213 Cal. Rptr. 774], review den.) That court has now explicitly adopted a harmless error rule. (People v. May, supra, 221 Cal. App.3d 836, 839.)

[4] Cases relying on this rationale to affirm may be criticized for presupposing that the trial court picks a sentence and then constructs a rationalization to support it. Such a supposition is antithetical to the DSL, which contemplates a sentence arrived at through the systematic consideration of relevant facts and criteria in light of the available sentence choices. The court is not supposed to "reason backward" to justify a preconceived result. (People v. Swanson (1983) 140 Cal. App.3d 571, 574 [189 Cal. Rptr. 547].)