COURT OF APPEAL FOR ONTARIO

DATE: 20230721 DOCKET: COA-22-CR-0184

Fairburn A.C.J.O., Zarnett and George JJ.A.

BETWEEN

His Majesty the King Respondent

and

Michael Smith Appellant

Counsel: Paul Socka, for the appellant Rachael Ward, for the respondent

Heard: June 30, 2023 by videoconference

On appeal from the sentence imposed by Justice Anthony F. Leitch of the Ontario Court of Justice on August 17, 2021.

Fairburn A.C.J.O.:

Overview

[1] The appellant pled guilty to firearm and drug related offences. He received a global seven-year sentence. This is an appeal from sentence that rests on a single issue: was there a calculation error as it relates to pre-sentence custody? The answer is yes.

Background Facts

[2] The police were called to a residence for a disturbance. They entered a residential location where they located the appellant. He had a satchel strapped to his body. Upon arrest, 28.6 grams of fentanyl were found inside. The police later executed a search warrant at the residence, where they located a loaded firearm in the bedroom in which the appellant had been staying. At that time, he was under multiple weapon prohibition orders.

Guilty Plea and Initial Submissions on Sentence

[3] On January 20, 2021, the appellant pled guilty to unauthorized possession of a loaded, prohibited firearm, possession of a firearm while prohibited (x3), and possession of fentanyl for the purpose of trafficking.

[4] The provincial Crown argued that a fit total sentence would be one of eight to nine years. The federal Crown asked for a six-year sentence, but acknowledged that it should run concurrent to the sentences imposed on the firearm counts.

[5] The appellant contended that a fit total sentence would be one of 7 years or 84 months. He said, though, that he should receive significant credit to reflect the harsh conditions in the custodial setting where he had been detained since his arrest 10 months earlier, conditions arising from COVID-19: R. v. Marshall, 2021 ONCA 344, at paras. 10-11. The appellant took the position that these collateral consequences should, effectively, result in a net sentence of 64 months left to serve.

[6] The appellant provided the court with two options for arriving at this net sentence. He suggested that the sentencing judge could either: (1) impose an 84-month sentence and then credit him at a 2:1 ratio for time served, which would result in a 20-month deduction from the total sentence, leaving 64 months to serve; or (2) impose a lesser 79-month sentence (which would already be reduced by 5 months to reflect the harsh conditions) and then credit him at a 1.5:1 ratio for time served, which would result in a 15-month deduction from the total sentence, again leaving 64 months to serve.

[7] Neither the provincial nor federal Crown made any submission as to the amount of credit for pre-sentence custody.

[8] Following submissions, the appellant made a lengthy allocution. In that statement, he made serious allegations about having been assaulted while detained in pre-sentence custody. The sentencing judge understandably expressed the view that these allegations could give rise to collateral circumstances that would further mitigate sentence. He granted the defence request to follow-up on the matter.

[9] Prior to adjourning, the sentencing judge expressed what he was thinking and what he expected when the matter returned:

All right. I can tell you … that I was, without revealing what my sentence would have been before this issue arose, I could tell you that I was going to treat the lockdown, sorry the pretrial custody he’s accumulated so far with the equivalent of a 2 for 1 analysis.

So I was not going to give him 2 for 1 credit because of my [indiscernible] in the area, but he would have gotten the equivalency of that through additional credit for time in lockdown. So just so it’s clear, I don’t need to have you prove what you said in your submissions concerning lockdowns, access to the yard, you know, the inability of people to visit and all that sort of thin[g]. My large concer[n] is the allegation of violence against your client, whether it be by other inmates or by correctional officers and whether or not he wishes to seek credit for those collateral consequences. [Emphasis added.]

[10] If the appellant had been sentenced when these comments were made on January 20, 2021, and had the sentencing judge used “the equivalent of a 2 for 1 analysis” to calculate credit on the 298 days of time served to that point, it would have been the equivalent of 596 days or approximately 20 months of credit for time served, which was 5 months longer than the standard Summers credit for time served: R. v. Summers, 2014 SCC 26, [2014] 1 S.C.R. 575, at para. 71.

The Matter Returns Almost Seven Months Later with An Agreement

[11] The matter finally returned before the sentencing judge on August 12, 2021, almost seven months after the adjournment. The delay was explained on the record as being, in part, due to a COVID-19 outbreak at the custodial institution where records to support the assault allegation were kept and the corresponding inability to access that institution during the outbreak.

[12] At the August 12, 2021 hearing, counsel for the appellant updated the court on the time spent in pre-sentence custody. As of August 12, 2021, the appellant had served 502 real days of pre-sentence custody, all of which had been during the COVID-19 pandemic. With a 2:1 credit applied, he would have served the equivalent of about 33 and a half months in pre-sentence custody.

[13] Defence counsel also informed the court about an agreement that had been reached between the parties. On top of whatever sentence the sentencing judge arrived upon as “fit and appropriate”, they asked the sentencing judge to deduct “a further” four months to reflect collateral circumstances.

[14] In particular, defence counsel said that this further reduction was to reflect a number of factors, including the fact that the appellant had saved significant court resources by abandoning his application to prove the alleged assault. Defence counsel also described the reason for the reduction as “primarily” being to reflect the time that the appellant spent in segregation in a dry cell after the alleged assault had occurred. The conditions in segregation included inappropriate clothing (the appellant was left dressed in what was described as a “baby doll dress”), with no access to water, a flushing toilet, a shower or a phone.

[15] The provincial Crown confirmed the agreement that had been reached for a four-month reduction from the otherwise appropriate disposition. While the Crown was clear that he was making no admission or acknowledgment of wrongdoing on the part of the correctional officers, he acknowledged that litigating the issue involving the alleged assault would have taken about four days of court time. The agreement reached ensured the better use of court resources. In addition, the Crown said that the deduction of four months was to reflect the time spent in segregation and that the appellant had spent “a significant period of time during the COVID-19 pandemic” in pre-trial custody, which the Crown characterized as a “collateral consequence, which would also justify a reduction in sentence.”

[16] On the actual day of sentencing, the federal Crown simply said she saw the agreement reached as reflective of the court resources saved in not proceeding to prove the assault allegations.

The Sentencing Reasons

[17] Sentence was imposed on August 17, 2021. As of that date, the appellant had spent 507 days in pre-sentence custody. By this point, the equivalent of 2:1 credit for time served would have been 1,014 days, which was 253 days longer than the standard Summers credit assigned at a ratio of 1.5:1.

[18] Hearkening back to the original sentencing submissions, the sentencing judge characterized the defence position as one of “seven years minus pre-trial custody”. He referred to this position as a responsible one. As a reminder, that defence position was predicated on pre-sentence custody reflecting the equivalent of a 2:1 ratio.

[19] The sentencing judge then said that it was his intention to impose a global sentence of seven years. In support of that conclusion, the sentencing judge held that the primary sentencing principles were denunciation and deterrence, but also noted the absence of aggravating factors and the presence of multiple mitigating factors, including his view that the appellant was “capable of rehabilitation.”

[20] Having settled on a seven-year total sentence as fit, the sentencing judge then looked to what credits he would apply to reduce that total. Ultimately, he deducted three credits.

[21] First, he dealt with the pre-sentence custody of 507 days, applying a 1.5:1 credit ratio for each day served. This resulted in 761 days of credit, which the sentencing judge truncated to 25 months of time served.

[22] Second, the sentencing judge reduced the sentence by another four months to reflect what he referred to as the “hard time” and “restrictions” experienced while the appellant was in custody for 507 days:

In examining the totality of the evidence on his time in custody for those 507 days, I find that a credit of a further four months custody is appropriate to reflect the hard time he has done and the restrictions that he has been under in custody.

[23] For ease of reference, I will refer to this four-month period of credit as the “Duncan credit”: R. v. Duncan, 2016 ONCA 754, at para. 6.

[24] Third, the sentencing judge reduced the sentence by another four months to reflect the agreement reached by counsel. The sentencing judge said he was prepared to accept that agreement for a few reasons:

One is to recognize the time saved in terms of court resources in litigating [the assault allegation], and also not only did Mr. Smith submit that he had been beaten while in custody, he also submitted a series of documents that showed he was triple-bunked and that he was placed under significant restrictions while in custody. It is for those reasons, without any formal proof of the more serious allegation of being beaten while in custody, that I am prepared to reduce his sentence by four months.

[25] For ease of reference, I will refer to this four-month period of credit as the “collateral consequences credit”.

[26] To sum up, the sentencing judge imposed a seven-year sentence on the possession of a firearm count. He then reduced the seven-year sentence by 25 months for time served (just 11 days shy of 1.5:1 credit), another 4 months for the Duncan credit, and another 4 months for the collateral consequences credit. This left a remnant of four years and three months to serve on the possession of a firearm count.

[27] The fentanyl count received a total 6-year sentence, also reduced by the 25 months of pre-sentence custody, leaving a net concurrent sentence of 3 years and 11 months. The possession of a firearm while prohibited counts each received concurrent sentences of one year.

Analysis

[28] The real issue on appeal is whether the ultimate sentence reflects the sentencing judge’s stated intention, as expressed on January 20, 2021, to grant the appellant “the equivalent of a 2 for 1” pre-sentence custody credit.

Parties’ Positions

[29] The appellant argues that he has been shortchanged by five months of credit. He maintains that the sentencing judge was clear on January 20, 2021, that while he would not grant 2:1 credit, he would grant the equivalency of that credit to reflect the harsh conditions experienced by the appellant while detained in pre-sentence custody. The appellant argues that if the sentencing judge had stayed true to that commitment, he would have credited him with five additional months of Duncan credit for the harsh conditions. The additional five months result from the fact that he had been in custody for 507 days by the time that he was sentenced, meaning that a 2:1 equivalency required an additional 9 months (not 4 months) credit over and above the standard Summers credit.

[30] The respondent argues that the sentencing judge’s comments were not nearly so definitive on January 20, 2021. The respondent says that the sentencing judge was merely speaking on a preliminary basis when he spoke in terms of the equivalency of 2:1 credit. He did not promise that credit and certainly did not promise it on a go-forward basis for the time between the January 20 appearance and the time of sentencing.

[31] The respondent also argues that the sentencing judge was required to consider any developments in the case between the original sentencing date and the sentencing. One such development was the collateral consequences credit jointly arrived upon by counsel. The respondent points out that one of the factors supporting that joint position for an additional four-month sentence reduction was the harsh conditions faced by the appellant in pre-sentence custody. Therefore, the respondent argues that although the sentencing judge said that he was specifically reducing the sentence by four months to account for harsh custodial conditions experienced by the appellant in pre-sentence custody (the Duncan credit), he also used some of the collateral consequences credit to account for those conditions. Therefore, in reality, the Duncan credit was longer than 4 months, although how much longer is unclear from the reasons.

Overview

[32] In my view, there was an inadvertent denial of procedural fairness in this sentencing proceeding. That denial arose from the appellant’s reliance upon comments made by the sentencing judge at the January 20, 2021 appearance, a reliance that later worked to his detriment when the matter returned to court for further sentencing submissions in August.

The Sentencing Judge Used a Pre-Marshall Framework to Deduct a Specific Period of Time to Reflect “Duncan Credit”

[33] In order to understand where the error lies in this case, it is first necessary to understand the methodology used by the sentencing judge to take into account the harsh custodial conditions arising from COVID-19.

[34] The sentencing judge’s reasons reflect what I will call a “Duncan framework”.

[35] In Duncan, this court recognized that in appropriate circumstances, “particularly harsh presentence incarceration conditions” could act to mitigate sentence beyond the statutory cap of a 1.5 day credit for every day served in pre-sentence custody: Duncan, at para. 6; Criminal Code, R.S.C. 1985, c. C-46, s. 719(3.1). In determining whether enhanced credit should be assigned, the court emphasized that we look to not only the conditions of the pre-sentence custody, but also the impact of those conditions on the accused: Duncan, at para. 6.

[36] The harsh conditions in custodial settings arising from COVID-19 have been described by this court as “collateral consequences”: see R. v. Morgan, 2020 ONCA 279, at paras. 8-9; R. v. Suter, 2018 SCC 34, [2018] 2 S.C.R. 496, at para. 48. As reinforced in Marshall, at para. 50, Duncan credit is directed at those collateral consequences which include “exceptionally punitive conditions” going beyond the “normal restrictions associated with pretrial custody.” The restrictions and health risks associated with COVID-19 were described in Marshall as a “good example” of the very kind of circumstance that could give rise to a Duncan credit: at para. 50.

[37] Marshall added an important nuance to the Duncan framework. While it does not amount to an error to deduct a specific number of days or months as Duncan credit, it is preferable if sentencing judges simply address any Duncan-type concerns as a type of mitigating factor when determining the fit global sentence. Therefore, the Duncan credit should not be approached as a “deduction from the otherwise appropriate sentence”, but as a factor to be taken into account when determining the fit sentence in all of the circumstances: Marshall, at para. 52.

[38] Although Marshall was decided between the January 20, 2021 adjournment and the sentencing date in this case, the sentencing judge’s reasons make plain that he approached the Duncan credit through a pre-Marshall lens, specifically assigning a period of credit for harsh custodial conditions, a period of credit which he later subtracted from what he deemed to be the otherwise fit global sentence of seven years in custody.

[39] Looking to the mitigating and aggravating factors, none of which had to do with harsh pre-sentence conditions, the sentencing judge said that it was his intention to render a seven-year sentence. Only after this pronouncement did the sentencing judge start to make deductions from that seven-year term, including specifically assigning a Duncan credit, over and above the regular 1.5:1 credit ratio, for what the sentencing judge described as follows: “a credit of a further four months custody … to reflect the hard time he has done and the restrictions that he has been under in custody.”

[40] Therefore, on their face, the reasons for sentence make clear that the seven-year total sentence did not take into account the harsh conditions served in pre-sentence custody. Rather, the sentencing judge’s reasons make clear that, in the absence of harsh conditions, a seven-year sentence would have been appropriate. Only then did he turn his mind to the harsh conditions and reduce the sentence by four months on account of those conditions.

[41] This was problematic because the appellant was told on January 20, 2021 (almost seven months earlier), that if he had been sentenced on that date, he would have received the “equivalent of a 2 for 1 analysis” to reflect the harsh conditions in pre-sentence custody. On that date, almost seven months before he was actually sentenced, the “equivalent of a 2 for 1 analysis” would have been five months of additional credit (as opposed to the four months of credit he received almost seven months later), over and above Summers credit.

[42] This brings us to the procedural fairness issue.

The Denial of Procedural Fairness

[43] Even if one were to accept the respondent’s position that the sentencing judge did not definitively say that he would give the equivalent of a 2:1 credit for time served to January 20, 2021, or on a go-forward basis, in light of the comments he made, one could be forgiven for thinking that was what the sentencing judge was going to do. One thing seems certain. Defence counsel thought that this was what the sentencing judge was going to do: provide enhanced credit with the “equivalent of a 2 for 1 analysis”. I say this because the record is silent with respect to any attempt by defence counsel to justify enhanced credit when the matter returned to court in August.

[44] Defence counsel’s reaction was understandable due to three specific aspects of the sentencing judge’s comments on January 20, 2021. First, and as already mentioned, the judge explicitly said he was going to treat the appellant’s pre-trial custody accumulated up to that point “with the equivalent of a 2 for 1 analysis” and give “the equivalency of that”. Second, the sentencing judge specifically told defence counsel that, upon her return, she did not need to “prove what [she had] said in [her] submissions concerning lockdowns, access to the yard, you know, the inability of people to visit and all that sort of thin[g].” Third, the judge indicated to defence counsel that his focus was, instead, on the alleged assault allegations, which he was potentially prepared to deal with as an additional collateral circumstance. As he said: “My large concer[n] is the allegation of violence against your client … and whether or not he wishes to seek credit for those collateral consequences.” These aspects of the sentencing judge’s comments, especially when taken together, could understandably have left defence counsel with the impression that the credit for the harsh conditions involved in the appellant’s pre-sentence custody was no longer at issue and the only remaining concern related to whether there would be a further credit due to the alleged assault allegations.

[45] Consistent with those directions, upon returning to court in August, defence counsel did not take steps to prove the specifics around the particularly harsh custodial circumstances the appellant had experienced to date. Rather, consistent with the comments made by the sentencing judge on January 20, counsel merely reiterated their request for a seven-year sentence, from which the equivalent of 2:1 credit for pre-sentence custody should be deducted to reflect both Summers and Duncan credit.

[46] While counsel did not attempt to prove the conditions of the appellant’s incarceration from January 20 to when the matter returned on August 12, there is no question that the pandemic continued to rage, and the conditions remained a good example of when Duncan credit may be applied. Indeed, when explaining why it had taken so long to access institutional records related to the alleged assault, counsel referred to the fact that in February and March of 2021, the institution where the appellant was detained experienced a significant COVID-19 outbreak and “for the better part of at least a month there were even greater numbers of extended institutional lockdowns while the detention centre did what they could to address that outbreak situation.”

[47] While the respondent argues that it was the sentencing judge’s intention to also assign a portion of the agreed upon four-month collateral consequences credit to Duncan credit, this does not solve the procedural fairness issue. Even if that had been the sentencing judge’s intention, and we took one month from the collateral consequences credit and slid it over to the Duncan credit, that would still only make for a total of five months of Duncan credit. As already mentioned, this would just get the appellant to the same point that he would have been in had he been sentenced on January 20, 2021. That would leave the following period of nearly seven months of pre-sentence custody unaccounted for from a Duncan perspective, in circumstances where the sentencing judge told counsel that they need not address the issue upon their return.

Conclusion

[48] Unfortunately, through no fault of anyone involved, the thread of this sentencing proceeding was lost. Undoubtedly, the appellant’s crimes are serious. However, he gave up his right to a trial and pled guilty in circumstances where the sentencing judge saw hope for rehabilitation.

[49] On January 20, 2021, the appellant was told that he would be receiving the equivalency of a 2:1 credit, at least up to that point. He was then told that upon his return, he need not address or prove the harsh custodial conditions he had experienced in pre-sentence custody. Indeed, he was told to direct his mind only to whether he should get even more credit for a possible further collateral circumstance involving an alleged assault.

[50] The appellant then spent another 209 days in harsh pre-sentence custody conditions. The matter took a significant time to return, in part because of those harsh conditions and a serious outbreak of COVID-19 in the institution where he was detained. He then gave up his right to litigate serious assault allegations that he made against the custodial authorities, an alleged assault on him that was followed by a period of segregation under harsh conditions. In exchange for giving up what would have taken at least four days of court time to litigate, the parties came to an agreement of four months of collateral consequences credit. While there is some dispute over how much, if any, of that collateral consequences credit should have been assigned to Duncan credit, there is no question that a good portion of the collateral consequences credit was to reflect things other than Duncan credit.

[51] The difficulty is that, having chosen to proceed under a pre-Marshall framework, using a more traditional Duncan formula to actually calculate enhanced credit for harsh conditions, the sentencing judge’s reasons leave us in the dark as to what was credited (if anything) over and above the four months clearly assigned to Duncan credit. Whatever was credited, however, it still does not account for what the appellant thought was going to happen when he returned to court in August and behaved, to his detriment, in accordance with the comments made by the sentencing judge in January.

[52] Accordingly, I would give effect to the sentencing judge’s comments on January 20, 2021 about providing the equivalency of 2:1 credit and then telling counsel that they need not address the issue further as he was focused on the issue of the alleged assault. This translates into nine months of enhanced credit. While in giving effect to the judge’s comments I treat this time as a quantified amount to be deducted from the otherwise fit sentence determined by the judge in this case, I reiterate this court’s comments in Marshall that the preferable approach going forward is to address any Duncan concerns as a factor in the course of determining the fit and proportionate sentence.

[53] For the purposes of this case, however, I will stay within the methodological framework used by the sentencing judge. Doing so, the Duncan credit will increase from four to nine months. This means that the 7-year sentence on the possession of a firearm count will be decreased by a Summers credit of 25 months, the agreed upon collateral circumstances credit of 4 months and now a Duncan credit of 9 months to leave a remnant of 3 years and 10 months to serve.

[54] The 6-year concurrent sentence on the fentanyl count will similarly be decreased by a Summers credit of 25 months, the agreed upon collateral circumstances credit of 4 months and now a Duncan credit of 9 months to leave a remnant of 2 years and 10 months.

[55] The one-year concurrent sentences on the possession of a firearm while prohibited counts remain untouched.

[56] All other aspects of the sentencing order remain the same.

Released: July 21, 2023 Fairburn A.C.J.O. I agree. B. Zarnett J.A. I agree. J. George J.A.