The appeal is allowed.
The sentence of reparation is quashed.
Elias CJ, McGrath, William Young and Chambers JJ
Theft of gallantry medals
On 2 December 2007 two men burgled the National Army Museum at Waiouru and stole 96 gallantry medals. The theft of the medals caused much consternation throughout the country: among the medals stolen were nine Victoria Crosses, including the VC and Bar awarded to Charles Upham, the only VC and Bar ever awarded to a fighting solider.
On 17 December 2007 the Commissioner of Police decided to offer a reward of up to $300,000 “for material information or evidence, which leads to the identity and conviction of any person or persons responsible for the burglary of the Waiouru Army Museum, the subsequent receiving of any proceeds from that burglary, and/or the recovery of the stolen medals”. The Commissioner further advised that he would “determine the amount of the reward and [would], if necessary, apportion payment where there is more than one claimant”. The Commissioner has over the years often offered rewards for information leading to the identification of criminals responsible for particular crimes. On this occasion two people, Lord Ashcroft and Tom Sturgess, had offered to fund a reward if the Commissioner saw fit to offer one.
In January 2008 Christopher Comeskey, an Auckland lawyer, approached the police offering a deal under which the medals would be returned in exchange for the reward. Mr Comeskey did not disclose the names of his clients. The Commissioner and Mr Comeskey struck a deal. The medals were returned in February 2008. The police paid in excess of $200,000 under the agreement.
The police continued to hunt for the thieves. Eventually they arrested James Kapa, the appellant, and Ronald Van Wakeren and charged them with burglary. First Mr Van Wakeren and later Mr Kapa pleaded guilty. It turned out that these men had been Mr Comeskey’s clients. It is now common ground that $100,000 of the reward money had gone to Mr Van Wakeren and another $100,000 to Mr Kapa. Mr Van Wakeren repaid his share of the reward to the Commissioner, but Mr Kapa did not.
Mr Kapa came up for sentence in the District Court at Auckland on 26 August 2010. Judge Hubble sentenced him to imprisonment. He also imposed a sentence of reparation in the sum of $100,000. Mr Kapa appealed. The Court of Appeal reduced slightly the term of imprisonment. The Court of Appeal dismissed the appeal against the sentence of reparation.
Mr Kapa sought leave to appeal to this Court. This Court declined leave with respect to the proposed appeal against the sentence of imprisonment but granted leave on the question of whether the sentence of reparation complied with the requirements of s 32 of the Sentencing Act 2002.
The essential issue for our determination is whether Lord Ashcroft and Mr Sturgess are persons for whose benefit a sentence of reparation can be made. It appears this was simply assumed to be the case in the District Court: Mr Kapa’s then counsel (not Mr Edgar) made no submissions at all on the issue of reparation. The Court of Appeal appears to have been primarily concerned with the appeal against the sentence of imprisonment. All the Court said, in dismissing the appeal against the sentence of reparation, was:
Were Lord Ashcroft and Mr Sturgess persons for whose benefit a sentence of reparation could be made?
Section 32 read, at the relevant date, as follows:
Ms Markham, for the Crown, submitted the sentence of reparation was rightly imposed because Mr Kapa, through or by means of the offence of burglary, caused Lord Ashcroft and Mr Sturgess to suffer a loss of property (subs (1)(a)) or loss consequential on a loss of property (subs (1)(c)). She submitted the case could come within either paragraph, although she expressed a preference for para (a). Is this submission right?
The essential premise of the Crown’s argument was that any “person” could potentially be the recipient of a sentence of reparation. The person did not need to be a victim, save in the case of reparation in respect of emotional harm, where subs (2) limited the range of recipients to victims. The presence of subs (2), Ms Markham said, strengthened the Crown’s argument that non-victims could recover in respect of loss of or damage to property or loss or damage consequential on loss of or damage to property.
It is this essential premise we are unable to accept. For reasons we shall explain, only victims can be the recipients of a sentence of reparation. The key to understanding the purport of s 32 is its interrelationship with the definition of victim in s 4 of the Sentencing Act.
Section 32(1) permits reparation in three circumstances, which we consider in turn. First, in para (a) is “loss of or damage to property”. “A person who, through, or by means of, an offence committed by another person, suffers .... loss of, or damage to, property” is a victim in terms of para (a)(ii) of the definition of victim in s 4 of the Sentencing Act.
Paragraph (b) deals with emotional harm. A person who, through or by means of an offence, suffers only emotional harm is not a victim within the definition of victim in s 4. Because Parliament did not intend non-victims to be the recipients of a sentence of reparation, it provided in subs (2) that a person could recover for emotional harm only if that person could be shown to be a victim as defined. That is to say, it would need to be shown that the recipient met one or more of the following criteria:
Paragraph (c) of s 32(1) deals with consequential loss or damage. At first blush it might not be completely clear whether someone who suffers consequential damage must be a victim himself or herself. Could a person who was not himself or herself a victim recover for loss or damage consequential on a victim’s loss of or damage to property?
We are satisfied, for two reasons, that only victims can recover for consequential loss or damage.
First, it would seem unlikely that recovery of consequential loss was intended to be open-ended after reparation for loss of or damage to property and emotional harm had been carefully restricted to victims.
Secondly, the legislative history of s 32(1)(c) is consistent with a conclusion that consequential loss is recoverable only by victims. In order to understand this argument, we need to turn first to the sentence of reparation provided in the Criminal Justice Act 1985. Section 22(1) provided that, where the court was “satisfied that any act or omission that constituted the offence caused any loss of or damage to any property of another, the court [could] sentence the offender to make reparation”. Parliament restricted the broad nature of that provision by providing, in s 22(5), that the value of any loss or damage was not to include “any loss or damage of a consequential nature”.
The section generated much case law. One strand of it is relevant for current purposes. In R v Gill, Mr Gill had used a document to defraud and obtain money under false pretences. He had represented that a letter typed on the letterhead of a non-existent finance company, purporting to offer a $300,000 loan, was genuine. He had fraudulently prepared the letter to obtain commission on the loan for himself. The complainant discovered that the offer was spurious shortly before he needed the $300,000 loan to purchase a substantial commercial property. As he was unable to find alternative finance, he settled the contractual liability with the vendor of the property for $20,000. He also had to pay $2,296 in legal costs. The Court of Appeal held that Mr Gill was liable to repay the $3,000 commission he had collected from the complainant. But the Court held the $20,000 settlement amount and the $2,296 legal costs were not covered by s 22, as they were of a “consequential nature” arising from the special circumstances of the complainant.
It is clear from the tenor of the reasons that the Court reached its conclusion with some reluctance. But it held that, “so long as the exclusion [in s 22(5)] remains”, losses of the kind in question could not come within the ambit of s 22.
In 2001 the Government determined to reform sentencing law and introduced the Sentencing and Parole Reform Bill. The Bill set out to repeal most of the Criminal Justice Act. Clause 29 was to be the replacement of s 22. According to the Bill’s explanatory note, cl 29 was designed to change the existing law in two respects. First, it was proposed that a sentence of reparation could now be imposed not only for a victim’s property loss or emotional harm but also “for physical harm to the victim”. The second change was to provide for the recovery of consequential loss; the obvious intent was to remedy the problem the Court of Appeal had highlighted in Gill. The drafters’ initial attempt to provide for these changes was as follows:
It is clear from that drafting that consequential loss could be recovered only by a person who had suffered, through or by means of an offence of which the offender was convicted, emotional or physical harm or loss of or damage to property.
The Justice and Electoral Select Committee, to which the Bill was referred, reported to Parliament on 12 February 2002 on the submissions received. It realised the proposed extension to allow reparation for physical harm would result in double recovery, “because physical harm is [already] covered by current accident compensation legislation”. It was also concerned by the potential scope of reparation for consequential loss. To this end the Committee recommended a “new clause 29(2) [to place] appropriate limits upon consequential loss or damage”. The relevant part of the revised cl 29 read as follows:
It is also noteworthy that the new draft used the word “victim” for the first time in both sub-cl (1) and the new sub-cl (2). This appears to have been thought of as no more than a drafting change: there is no mention of this change in the commentary. Again, the redraft reinforced the original concept that consequential loss was recoverable only by victims.
Prior to the third reading of the Bill, the drafters recognised a tautology. The Select Committee’s substitution of “caused the victim to suffer” for “caused any other person to suffer” meant, given the definition of “victim”, that cl 29(1) now read as follows (if the definition were inserted):
A court may impose a sentence of reparation if an offender has, through or by means of an offence of which the offender is convicted, caused the person who, through, or by means of, an offence committed by another person, suffers physical injury or loss of, or damage to, property to suffer –
The drafters, by a Supplementary Order Paper, reverted in sub-cl 29(1) to “caused a person to suffer” rather than “caused the victim to suffer”. Subclause (1) now read as follows:
A court may impose a sentence of reparation if an offender has, through or by means of an offence of which the offender is convicted, caused a person to suffer –
The sole point of that drafting change seems to have been to remove the tautology. The change was not intended to widen the range of those who might be beneficiaries of a sentence of reparation. That range remained restricted to victims, as the Select Committee intended. Because, as we have explained, people suffering emotional harm were not necessarily victims, the drafters dealt with that problem by inserting a new sub-cl (1A), which became, on the Bill’s enactment, s 32(2), which we have discussed above.
The rights conferred on victims by s 32 are part of a suite of rights conferred on victims under various statutes this century. We refer to the Victims’ Rights Act 2002, the Parole Act 2002, the Prisoners’ and Victims’ Claims Act 2005 and the Criminal Procedure Act 2011. Each of those statutes uses the same or substantially the same definition of victim as the Sentencing Act does. Among the rights of victims are the following:
to make victim impact statements in the sentencing of the offender and to have those statements considered in the sentencing;
to participate in the process for making decisions about the offender’s release from prison or release to or from home detention;
to be given notice that the parole hearing of an offender is pending;
to attend a parole hearing and make submissions to the Parole Board;
to make submissions on and be notified about whether and when an offender will be released from detention and on what conditions;
to make a claim for damages or exemplary damages under the Prisoners’ and Victims’ Claims Act.
To recap at this point. We have established that s 32 is for the benefit of victims. Subsection (1)(a) allows reparation for direct loss of or damage to property. (We say “direct” as subs (1)(c) picks up a victim’s consequential losses.) Subsection (1)(b) allows reparation for emotional harm.
We mention in passing that this conclusion receives support from another provision of the Act. Section 38(1) provides:
Every sum payable under a sentence of reparation must be paid to the person who suffered the harm, loss, or damage, or, with that person’s consent, to that person’s insurer.
The fact the drafters saw s 38(1) as necessary indicates that they did not consider insurers to be entitled to recover under s 32(1), notwithstanding that insurers may, in one sense, be said to have been caused loss by the offending. Insurers are not victims and thus are not intended to be beneficiaries of a s 32(1) order. Their entitlement to recover for losses they have indemnified arises under the Sentencing Act, if at all, only under s 38(1) and is subject to the victim’s consent.
So what if a reward is offered and paid, as occurred here? This question can be looked at narrowly or liberally. One approach would be to say that the Commissioner paid the reward and it is irrelevant how he funded it. If that approach is adopted, it is obvious Mr Kapa cannot be ordered to pay reparation to the Commissioner. Neither the Commissioner nor the police generally have ever been treated as victims. Their costs have never been able to be recovered under the reparation regime. There were attempts under the Criminal Justice Act to recover, by way of reparation, costs incurred by the police in investigating suspected criminal activity, including costs of buying drugs and, in one case, the cost of employing Telecom to trace calls following a complaint of a telephone being used to annoy contrary to s 8(1)(a) of the Telecommunications Act 1987. These attempts failed. The point was made that all the police or prosecution can recover from a convicted defendant is such expenses as are allowed under the Schedule to the Costs in Criminal Cases Regulations 1987. That line of authority remains the position under the Sentencing Act. The cost to the Commissioner of paying out on a reward offer is not a loss which the Commissioner can recover from the offender under s 32. It is an investigative expense, no different from other costs, both internal and external, he incurs in the investigation of offending. If nothing is payable to the Commissioner, his funders obviously could not recover as his insurer or indemnifier.
A more liberal approach would be to ignore the Commissioner’s role and to concentrate on the payments made by the individual reward donors (as we shall call Lord Ashcroft and Mr Sturgess in the present case and others who might contribute, directly or indirectly, to the making of a reward in other cases). Assume in their favour they pay the reward direct to someone or that it is irrelevant that the reward was paid through the police. We are satisfied they do not, by their payments, make themselves victims. Indeed we think even the reward donors – and there might be scores of them – would be surprised to be considered as victims of the crime they were trying by their donations to help get resolved.
The act of contributing to a reward might take place weeks, months, even years after the offence to which it refers. The cost is not a direct loss of the offence so as to come within s 32(1)(a). The offender has not “caused” that loss, as s 32(1) requires. The offender has merely, by his or her offending, set up the occasion for the making of the reward to happen. So reward donors do not come within s 32(1)(a). At best, the loss they have suffered is consequential on the loss suffered by others. But, for reasons already given, only victims can recover consequential loss under s 32(1)(c).
Therefore, whether the matter is looked at narrowly or more broadly, neither the Commissioner when paying a reward nor reward donors can recover the cost of that reward by way of a sentence of reparation. The only situation in which the cost of a reward might be recoverable is where the victim has put up money for a reward – say, in the hope of recovering his or her stolen property. That cost might be a consequential loss of that victim. But that is not the present case: neither the museum nor its insurer contributed towards the reward.
We have not found it necessary to discuss all the cases counsel cited to us. We do mention, however, R v Ebdell as its facts are broadly similar to those in the present case. Leigh Ebdell stole from his employer, New Zealand Post, 1,500 items of mail, including packages. Many of the items were still in Mr Ebdell’s possession when he was caught, but they were in a frightful mess. Goods had been removed from their wrapping and were all over the place. The Crown sought reparation in favour of New Zealand Post, which incurred considerable cost in working out the intended recipients of the stolen items and then delivering the items. Judge Erber in the District Court, French J in the High Court and the Court of Appeal all considered New Zealand Post could recover its “costs of restitution” as consequential loss. New Zealand Post’s interest in the stolen goods meant it was a victim which had suffered a loss of property. The cost of restoring the property to the intended recipients of the mail was a consequential loss within s 32(1)(c). New Zealand Post’s status was accordingly different from a reward donor’s status. The result in Ebdell was correct but the case is distinguishable.
We hold that Lord Ashcroft and Mr Sturgess were not persons for whose benefit a sentence of reparation could be made. The Judge, in making a sentence of reparation in their favour, acted without jurisdiction.
We accordingly allow the appeal. We quash the sentence of reparation. Since Mr Kapa is legally aided, there will be no order as to costs.
Mr Kapa “earned”, as a result of his offending, $100,000. The fact he got the reward is irrelevant to the question of whether the reward donors suffered a loss – which is what the sentence of reparation is all about. (On the Crown’s argument, Mr Kapa would have had to pay the $100,000 even if the police had paid the reward to an innocent third party for information leading to the successful resolution of the crime.) Even though we have held the sentence of reparation is not available, there might be another method by which Mr Kapa could be stripped of his gain. That is under the Criminal Proceeds (Recovery) Act 2009. Since an application under that Act could still be brought, we do not comment on the chances of success, save to say that at least in principle Mr Kapa’s receipt of the $100,000 appears to be within the scope of the Act. That Act gives the police and the courts significant powers to trace the whereabouts of the $100,000 Mr Kapa admits to having received in early 2008.
This appeal concerns whether there was jurisdiction to order Mr Kapa to pay reparation of $100,000. That sum represents his share of a reward paid for the return of military medals taken from the Waiouru Museum in the course of a burglary.
The reward money was provided by Lord Ashcroft and Mr Sturgess but the offer of the reward was made by the Commissioner of Police and it was he who made the payment arrangements.
Mr Kapa and an associate, Mr Van Wakeren, were later identified as the burglars of the museum and convicted of burglary.
The issue in this appeal is whether Mr Kapa has “through or by means of” the burglary caused Lord Ashcroft and Mr Sturgess to suffer “loss of or damage to property” (direct loss) or to suffer “loss or damage consequential on any .... loss of, or damage to, property” (consequential loss) and thus if the imposition of a sentence of reparation was warranted.
Under s 32(1)(a) and (b) of the Sentencing Act 2002, a court may impose a sentence of reparation if an offender has, “through or by means of an offence of which he or she is convicted”, caused a person to suffer “(a) loss of or damage to property” or “(b) emotional harm”.
The wording “through or by means of an offence” is broad. The word “through” has been seen as requiring a more direct connection between the offence and the loss or damage, and the words “by means of” have been seen as capturing damage or loss closely associated with the offence, although not necessarily arising from the acts that constitute the definition of the offence. While not disagreeing that both concepts are covered, rather than seeing the phrase “through or by means of” as disjunctive, I see it as a composite phrase designed to capture loss or damage that naturally flows from or is closely associated with the offence in question.
It seems to me that, looked at this way, the loss of items stolen in the course of a burglary is clearly loss or damage to property “through or by means of” the burglary. I would also see the phrase as covering the cost of any repairs to property damaged or destroyed in the course of a burglary as well as the direct costs of being deprived of the property (such as the costs of a rental car as a temporary replacement for a stolen car).
It seems to me too that expenses incurred in the recovery of property, including the payment of a reward, also naturally flow from or are closely associated with the loss of the stolen items.
Indeed, in this case, receiving a reward was at least partly the reason for the burglary. Judge Hubble’s sentencing notes record Mr Kapa’s claim that he just borrowed the medals, always intending to give them back after “asking for a ransom”. The notes also record that Mr Kapa claimed that he was going to use the medals “to trade against the sentences” for other offending.
The Judge obviously did not consider the bargaining over sentence to be Mr Kapa’s sole motivation. He said that he took “all that with something of a grain of salt”. He could not overlook the fact that there were serious threats that the medals would be destroyed or disposed of overseas if Mr Kapa and his co-offender were not paid.
It would be odd if a reward payment was not considered to be a loss caused through or by means of an offence in circumstances where the receipt of a reward was one (and arguably the dominant) motivation for the offending in the first place.
Under s 32(1)(c), reparation may also be imposed where an offender has “through or by means of an offence of which [he or she] is convicted” caused a person to suffer “(c) loss or damage consequential on any emotional or physical harm or loss of, or damage to, property”.
The statutory wording
The most important point to note is that the loss must be consequential on the loss of or damage to property, rather than merely consequential on the offending itself. In my view, this means that the paragraph is concerned with economic loss that arises because of being deprived of property, including loss of profits arising as a result of not being able to use stolen property to generate income.
It is not stated explicitly that the person and the property referred to in s 32(1)(a) must be the same person and property as are referred to in s 32(1)(c). Indeed, the paragraphs in s 32(1) covering direct loss and consequential loss are separated by s 32(1)(b) covering emotional harm and are disjunctive, not cumulative.
It may, however, be unlikely that it was intended that a party who had not suffered direct loss would be able to claim for economic loss consequential on the loss of property suffered by another. There could be an unending spiral of people affected if that were the case.
The legislative history, as outlined in the majority judgment, may suggest that it was intended that reparation for consequential loss could only be awarded to a person who had also suffered direct loss. As noted by the majority, as originally introduced, the Bill referred to loss suffered by a “person”. This was changed to “victim” by the Select Committee and then changed back to a “person” by Supplementary Order Paper.
It is stated by the majority that the change from victim to person was to remove the tautology created. There was undoubtedly a tautology but it is not indicated in the legislative material that the change was made to remove the tautology. It may just have been a return to the wording of the Bill as originally introduced, which used the word “person”, and indeed to the wording used in the predecessor provisions. It may also have been that the change was made to deal with the issue of emotional harm.
That said, the wording in the Bill as introduced did make it clear that an order for loss “consequential on the offending” could be made in addition to an order for “harm, loss, or damage suffered directly through, or by means of the offence”, suggesting that consequential loss may only have been intended to be available to the person who suffered direct loss.
As noted above, I consider that a reward offered to recover property is, however, properly regarded as direct loss. This means that there was no need to rely on s 32(1)(c) to make a reparation order in favour of Lord Ashcroft and Mr Sturgess.
Two Court of Appeal cases
I am conscious that that there are two Court of Appeal decisions that take a broader view of consequential loss and, as a result, a narrower view of direct loss than I do. The first, R v Gill, arose under a different statutory regime. The second, Ebdell, arose under the current reparation provisions and bears some similarity to the current case.
In Gill, the relevant provision was s 22 of the Criminal Justice Act 1985. That section excluded consequential loss from its ambit. Section 22(5) provided that the value of the loss or damage “shall be limited to the cost of replacement or (as the case may require) the cost of repair, and shall not include any loss or damage of a consequential nature”.
Mr Gill had been convicted of using a document to defraud and of obtaining money under false pretences. He had represented that a letter typed on the letterhead of a non-existent finance company, purporting to offer a $300,000 loan, was genuine. He was paid $3,000 commission by the complainant. The complainant discovered that the offer was spurious shortly before he needed the loan to purchase a substantial commercial property. Unable to find alternative finance, he settled his contractual liability with the vendor of the property for $20,000. He also incurred $2,296 in legal costs.
It was held by the Court of Appeal that the reparation regime covered only the $3,000 commission paid to the fraudster. In the Court’s view, the other losses were losses of a consequential nature. The Court examined the notion of consequential loss in insurance cases but said that these cases bore “only obliquely” on the interpretation of s 22. The Court found the cases on damages more helpful whereby a distinction is made between normal losses (which every plaintiff in a like situation would suffer) and consequential loss in the sense of loss which is special to the circumstances of the particular plaintiff.
The Court considered that the payments to the vendor and the legal costs were due to the special circumstances of the complainant and his inability to arrange alternative finance. Not all persons defrauded in like manner would suffer similar loss.
If Gill was correctly decided, then the introduction of s 32(1)(c) to allow consequential loss does nothing for fraud victims like the one in Gill. This is because loss under the current provision, s 32(1)(c), has to be consequential on the loss of or damage to the property that was the subject of the direct loss. The payment to the vendor and the legal costs made by the victim in Gill were not consequential on the loss of the commission.
I do not consider Gill to have been correctly decided or, if it was correctly decided, then I consider that it was decided on the wrong basis. It seems to me that the Court took far too narrow a view in interpreting what loss arose “through or by means of” the offence. To my mind the settlement with the vendor and associated legal costs were clearly losses that naturally flowed from or were closely associated with the fraud and thus were suffered through or by means of the fraud.
It is not clear from the report of Gill whether the fraudster knew that the victim wanted the loan to purchase a commercial property but, even if he did not know the exact use to which the fictitious loan funds were to be put, he must have known that his victim had some use for the money in mind and that some loss was likely to arise should the money not be available. Further, the type of victim who would be deceived by the fraud was very likely to be a person who would have difficulty in arranging alternative finance, particularly at the last minute. The fact that a victim would want to use the fictitious loan was at the very heart of the fraud.
Further, a definition of consequential loss that distinguishes between loss that is personal to the victim and loss which would be common to all victims is not in my view apt in the criminal context where normally you take the victim as you find him or her. Nor is it apt in a reparation regime which, despite being a sentence, is essentially a compensation regime. It would not be within the spirit of such a regime to award compensation only on the basis of what other victims might suffer and not on the basis of what the particular victim did suffer.
It may be, however, that the result in Gill was required by the particular statutory provision the Court was dealing with. On the wording of s 22(5) of the Criminal Justice Act, direct loss to property was limited to the cost of repair or replacement, which would imply that anything that did not come within those words was consequential loss. Therefore the case may have been correctly decided, but for the wrong reasons. In any event, the wording of s 22(5) is much more apt to deal with physical property rather than mere financial loss such as was suffered by the victim in Gill.
The second case of Ebdell was decided under the current reparation provisions. Mr Ebdell had been convicted of two representative charges: one charge of theft in a special relationship and one charge of using a document. The theft was of over 1000 items of mail. The reparation order related to the cost incurred by New Zealand Post, Mr Ebdell’s employer, in investigating and returning the stolen items to their rightful owners.
French J accepted without any discussion that NZ Post’s loss did not come within s 32(1)(a). She considered that it was, however, consequential on the loss caused to the owners. In her view, it therefore fell within the intended scope of s 32(1)(c) which allows for consequential loss.
The High Court refused leave to appeal to the Court of Appeal. French J rejected the argument that her interpretation of s 32(1)(c) could lead to agencies claiming reparation for the cost of investigations. She stated that a distinction could be drawn between the cost of investigations (to which the provision would not apply) and the cost of restoring the goods to their rightful owners (to which it would).
Mr Ebdell then applied for special leave to appeal. This was refused by the Court of Appeal. The Court of Appeal held that the High Court was plainly right in holding that s 32(1)(c) permits reparation in favour of someone other than the owner of the property which has been lost or damaged. The Court said that the introductory words to s 32(1) do not limit the order of reparation to the owner of the property but instead use the more general expression “person”. If s 32(1)(c) had been intended only to cover property owners, it would have been easy enough to say so. Section 32(2) specifically limits awards of reparation for emotional harm to those who are persons against whom an offence is committed. The Court commented that the absence of a similar limitation in respect of s 32(1)(c) is striking.
The Court of Appeal did not consider whether the loss suffered by NZ Post was direct loss. Indeed, it seems to have assumed that it could not be because NZ Post was not the owner of the items of mail that were stolen.
Neither Gill nor Ebdell lead me to change my view that the cost incurred to recover stolen property (including the payment of a reward) is a direct loss. Indeed, I consider that NZ Post in Ebdell also suffered a direct loss because the money it expended was to recover property.
If, however, I am wrong in this view, then I would adopt the reasoning of the High Court and the Court of Appeal in Ebdell. I would also read the word “person” in s 32(1) as meaning what it says and not being tied to the person mentioned in s 32(1)(a). In my view, the word “person” applies to s 32(1)(a) and (c) disjunctively, as I earlier indicated. By this I mean that a person who qualifies under s 32(1)(c) to recover for consequential loss does not also have to be a person who, under s 32(1)(a), suffered direct loss or damage. This is clear from the use of the word “or” between the paragraphs.
Some limit would have to be placed on this to avoid an endless spiral of claimants, but the wording of the section already achieves this limit. The phrase, “through and by means of” qualifies both s 32(1)(a) and s 32(1)(c). In this way, consequential loss is restricted to those losses which flow naturally and are closely associated with the offending.
If the word “person” is not read widely, then it would appear to me that NZ Post would not come within s 32(1)(c). It is true that NZ Post was a deemed owner of the mail but this did not mean it had suffered direct loss. It was merely a deemed owner and not an actual owner. Thus the loss it suffered in recovering the mail was still loss consequential on the loss of someone else’s property (as the Court of Appeal in that case noted).
If the Ebdell reasoning is accepted (and the reward payment is not direct loss), then the reparation order to Lord Ashcroft and Mr Sturgess would come within s 32(1)(c).
I am bolstered in my view that the reward payment in this case is covered by the reparation provisions by the “liberal and non-technical” approach that has (I consider correctly) been taken to the interpretation s 32(1) of the Sentencing Act and to its predecessor provision.
Before turning to that case law, I first make a few points about the history of the reparation provisions.
History of the reparation provisions
The first point to make is that, as pointed out by the Law Commission, traditionally at common law, crimes were seen as “public wrongs against the community at large”. They resulted in punishment to express society’s disapproval, to deter future offending and provide for public protection. Criminal proceedings were not designed to be reparative or compensatory in nature. When an individual suffered injury, loss or damage from offending, then he or she had to institute civil proceedings in tort.
The compensation provisions in the Criminal Code Act 1893 as well as in the 1908 and 1961 Crimes Acts, were designed to provide some possibility of compensation in the course of criminal proceedings. These provisions were, however, little used as the courts ordered compensation only when the amount was not too large, the facts were clear and simple, there was no dispute as to liability or amount and the offender had the means to pay. Otherwise, the person suffering loss or damage was left to pursue a civil remedy.
In 1981, the Penal Policy Review Committee recommended the introduction of a reparation regime in respect of property losses “covering all direct or indirect loss or damage suffered by any person as a result of the conduct giving rise to the proceedings in which the offender was convicted”. It said that the courts should lean towards making reparation orders in every appropriate case.
As a result of the recommendations of the Penal Policy Review Committee, specific reparation provisions were introduced in the Criminal Justice Act. For the first time reparation was seen as being a sentence. The title of s 22 was “Court may sentence offender to make reparation”. This concept was carried through into the current reparation provisions. Section 32 of the Sentencing Act is entitled “Sentence of reparation”. This represents a shift in ideology. Previously (and explicitly in the 1908 Act) compensation orders were seen as different from sentences.
The sentence of reparation is, however, described by the Law Commission as still being a “loss-shifting” mechanism. It is “loss-shifting” because it is justified on the basis that it is less expensive to the victim than separate proceedings in tort, but its ability to compensate a victim is still dependent on the attribution of fault to the offender and his or her means to pay. As the Law Commission explained, reparation is not a system that is designed to spread loss; it simply transfers the loss from the victim to the offender. Additionally, the Commission commented that, while the criminal justice system now incorporates more explicitly a reparation element in terms of the relationship between the offender and the victim, this does not alter the relationship between the victim and the state.
The next point to make is that the phrases “through and by means of the offence” and “loss of or damage to property” are of some antiquity. They appeared in s 449(1) of the Crimes Act 1908, which provided that a Court could, “in addition to passing sentence on him [or her]” order a convicted accused person to pay the costs of prosecution and also order him or her to pay compensation “for any loss of property suffered by [any person aggrieved] through or by means of the crimes of the offender”. They also appeared in the compensation provision in s 403 of the Crimes Act 1961.
From 1987, those phrases reappeared in the reparation provision, s 22 of the Criminal Justice Act. Section 22, as first introduced in 1985, provided that, where a court “is satisfied that any act or omission that constituted the offence caused any loss of or damage to any property of another, the court may sentence the offender to reparation”. This formulation caused difficulties. For example, in one District Court case, the Judge ordered reparation only for the damage caused in gaining entry to premises through burglary and not for the theft of cash and stock that followed. I refrain from comment.
The final point is that in 1987, the first statute dealing with victims was introduced. That legislation also used the phrases “through or by means of a criminal offence” and “loss of or damage to property” in the definition of victim. There was thus congruence between the definition of victim in the Victims of Offences Act and those eligible for reparation under s 22 of the Criminal Justice Act but the wording of both provisions was the same as that used in the earlier Crimes Act compensation provisions.
When the Bill containing current reparation provisions was first introduced, “victim” was defined as having the same meaning as in s 2 of the Victims of Offences Act. That Act defined victim as:
a person who, through or by means of a criminal offence (whether or not any person is convicted of that offence), suffers physical or emotional harm, or loss of or damage to property; and, where an offence results in death, the term includes the members of the immediate family of the deceased.
Later, a majority of the Justice and Electoral Select Committee suggested amending the definition of “victim” by specifically including, among other things, every “complainant in relation to an offence”. The second part of the definition kept the old wording and included “every person who, through, or by means of, an offence committed by another person, suffers loss of or damage to, property”. The definition of victim as suggested by the Select Committee, unlike in the Victims of Offences Act, did not include a person who had only suffered emotional harm (unless they were a complainant).
Further changes were made by way of the Supplementary Order Paper. The changes replaced the “complainant in relation to an offence” by “a person against whom an offence is committed”. The second part of the definition remained the same. The definitions of “victim” in the Sentencing Act 2002 and in the Victims’ Rights Act 2002 are essentially the same.
In the Supplementary Order Paper, the provision which became s 32(2) was also inserted. This limits a sentence of reparation in respect of emotional harm or damage consequential on emotional harm to those I would call primary victims, being those “against whom an offence is committed”. Presumably this was done because of the potential magnitude of persons who may suffer consequential emotional loss. The inclusion of what became s 32(2), however, may support the argument that “person” in s 32 was not meant to be wider than “victims”.
It must be assumed that the continuation of the use of the term “through or by means of an offence” in the Sentencing Act (including in the definition of victim) was made in the knowledge of the wide and liberal interpretation that phrase had been given in the case law.
It is consistent with international rights instruments to have broad protection for victims. For example, the United Nations Declaration of Basic Principles for Victims of Crime and Abuse of Power 1985 defines “victims” as including people who have suffered “economic loss”.
Case law on reparation
With regard to s 22 of the Criminal Justice Act the Courts have held that reparation awards can be made in the following circumstances:
In Lovatt v Police the High Court upheld an order of reparation for $400 which represented the price paid by a second hand dealer to purchase a pottery head sold to him by an offender convicted of theft. Tipping J commented that the section should be given a “liberal interpretation consistent with the spirit of reparation being ordered in all proper circumstances”. The dealer’s loss derived from the fact that Mr Lovatt had purported to transfer title to the pottery head in return for the $400 when he in fact did not possess that title. This was because he had stolen it and it still belonged in law to the true owner, to whom the dealer was later required to return it.
In Wilmot v Police the issue was whether reparation for the full loss of a stolen car could be ordered against a party who was not the original thief but who had received it, knowing it to have been stolen. Tipping J held that it was sufficient that the conduct had “materially contributed” to the loss or damage. Tipping J rejected an argument that the loss was complete when the goods were stolen and that Mr Wilmot had not made any further contribution to the loss (albeit he had not helped to restore the loss). The Judge also rejected a subsidiary argument that it was unfair that Mr Wilmot was ordered to pay full reparation while the juvenile burglar made no contribution.
In Wilson v Police Gallen J upheld a reparation order made against the passenger in a vehicle which had been unlawfully taken. The passenger protested that the damage to the vehicle was caused not by the unlawful taking of it, but rather by subsequent irresponsible driving on the part of a co-offender. The argument continued that there was no causal connection between the offence and the property damage, which occurred later and in the course of a joyride. Gallen J said that, where it can be said a reasonable person could reasonably foresee the kind of damage which occurred as a result of the actions in which they participate, then reparation can be ordered.
Under the current provision, s 32 of the Sentencing Act, in R v Donaldson, a reparation order was upheld for damage to the Rolleston Rugby clubrooms, which occurred because of a fire started in the course of the burglary. This was despite the offenders being discharged during trial in relation to a count of arson because the Crown could not establish the mental element of arson. The fact that the fire was lit by one of the burglars and in the course of the burglary was, however, established to the criminal standard.
The Court of Appeal considered it significant that, when the Sentencing Act was passed in 2002, Parliament retained the formula “through or by means of an offence of which the offender is convicted”. The Court said (and I agree) that this must be seen as an endorsement of the liberal and non-technical interpretation of this phrase under the predecessor sections.
The Court discussed the nature of a reparation order, as compared to a fine, and the interrelationship of the two. A fine is punitive. It is a pecuniary penalty imposed by and payable to the state. By contrast, an order for reparation is compensatory in nature. Reparation is intended, wherever possible, to restore the victim’s position in relation to property loss or damage, emotional harm, or consequential losses.
The Court commented further that the statutory phrase “through or by means of an offence” is a wide expression and its outer limits are not immediately obvious. It may, therefore, prove helpful to have resort to the concepts of remoteness, materiality and intervening act (novus actus interveniens), at least in analysing more difficult factual situations. However, the Court endorsed the viewpoint that reparation is to be approached in a broad commonsense way, and resort to refined causation arguments is not to be encouraged. Again, I agree.
Arguments against reparation in this case
A number of arguments have been or could be made against reparation orders being made in this case. The main argument made on behalf of Mr Kapa was that the reward payment was voluntary and therefore could not be the subject of a reparation order.
The main reason for the majority decision is that reparation payments are limited to victims. A secondary argument relates to the position of insurers. There is also the issue that the reward was in fact offered by the Commissioner of Police. There may also be an issue of apportionment of reparation payments and as to the relevance of the fact that there might be another remedy available.
I discuss each of these points in turn.
Position of volunteers
Lord Ashcroft and Mr Sturgess, in putting up the reward, were acting purely altruistically. It was argued on behalf of Mr Kapa that the reparation regime does not cover volunteers.
Reliance was placed on a line of cases dealing with claims by the police for compensation (under s 403 of the Crimes Act) and for reparation (under s 22 of the Criminal Justice Act) for the purchase price of cannabis by undercover police officers. In all cases it was held that the compensation/reparation regime did not extend this far.
For example, in Cooper v Ward, Speight J held that the purchase price was money the police department voluntarily offered and that it had not been inflicted on an unwilling victim. Speight J considered that the entire context of ss 22 and 23 of the Criminal Justice Act and s 403 of the Crimes Act dealt with cases of involuntary loss suffered by an innocent party and not with the deliberate elective conduct of the police department in conducting its affairs.
I do not see these cases as controlling the situation here. A payment made (however, justifiably) to induce offending cannot be seen as a consequence of the offending – ie to have arisen through or by means of the offending. Another point of distinction may be that, unlike Lord Ashcroft and Mr Sturgess, the undercover police officer in Cooper did not suffer a loss in the sense that the officer received exactly what he understood he was paying for.
I concede that there is a possible argument that the word “loss” could import a connotation of an involuntary loss but the difficulty with such an interpretation is that it would require the drawing of fine distinctions. For example, in Ebdell such an interpretation might require an inquiry into whether NZ Post as bailee and employer had an obligation to try and trace the mail, rather than the payment being entirely voluntary, albeit motivated by a sense of moral obligation and for business reasons.
Further, although in this case, the reward payment was made by totally unrelated third parties for altruistic reasons, it could have been offered by, for example, a relative of someone whose valour had earned one or more of the medals or indeed by the museum’s insurers. It might in other cases too be a question of chance whether a relative offers a reward him or herself or whether the money is lent to the person whose property was stolen for the purpose of providing a reward for the return of the property.
I consider that such fine distinctions should play no part in a reparation regime. In line with the liberal and non technical interpretation under the case law to date, I would interpret the word “loss” as including money paid out voluntarily. Reparation would therefore be available if that payment arose as a natural consequence or was closely related to offending.
There seems no policy reason not to allow a reparation award to be made in favour of volunteers who have paid a reward to recover property and thus have (in my view) lost money through and by means of the offending.
Indeed, in a case like this where the money was paid to recover iconic items of such public importance and where the receiving of a reward was one of the motivations for the offending, one might have thought that the policy should be to ensure that those who put up rewards are entitled to reparation, and thus to encourage rather than discourage altruistic gestures of this sort.
I am conscious that the interpretation I favour means that Lord Ashcroft and Mr Sturgess would be victims as defined in s 4 of the Sentencing Act and s 4 of the Victims’ Rights Act 2002. This is, however, a necessary consequence of the definition of victim including not only primary victims (ie those against whom an offence is committed) but also persons who suffer loss of or damage to property through or by means of an offence.
The term “victim” must have been designed to cover a wider group than primary victims or there would be no need for the two limbs of the definition. This is reinforced by the fact that reparation for emotional harm is explicitly restricted to primary victims. A compensation regime, which is what the reparation regime remains, should be victim-focused and this supports not limiting compensation to the primary victim of the offence actually charged. A victim has no choice over how and what charges are laid.
It is significant too that the definition of victim used wording that was first employed in the context of the compensation and reparation regimes rather than the other way round. In addition, the definitions in the Sentencing Act and the Victims’ Rights Act 2002 were passed against the background of cases where the terms and phrases used in the definitions had been interpreted broadly in the context of the reparation regime under the Criminal Justice Act.
It is true that victims have a suite of rights under various statutes as outlined in the majority judgment. However, the rights set out at [27(b)]–[27(e)] of that judgment are only available for sexual assault or serious injury or where there is fear for safety. The rights set out at [27(b)]–[27(e)] of the majority judgment would thus not be available to Lord Ashcroft or Mr Sturgess in this case.
The right to make a victim impact statement would be available to reward givers but s 20 of the Victims’ Rights Act already gives the prosecutor the ability to treat a person as a victim if they have been “disadvantaged by the offence”. Losing a $100,000 reward payment would seem to me to be a clear disadvantage in terms of s 20.
The majority states that reward donors would be surprised to be considered victims. That might be so as most people would understand victim to mean primary victim. Reward donors might well, however, consider themselves as people who have lost money through or by means of the offending. That is in fact the inquiry to be made.
Up until now I have loosely referred to the reward as being paid by Lord Ashcroft and Mr Sturgess. Of course that is not strictly correct as the reward was both offered and paid by the Commissioner of Police. It was also offered not just for the recovery of the medals but also for information leading to conviction of the burglars.
The majority consider what they call both a narrow and a liberal approach to this issue. I do not accept that a narrow view is appropriate in this context. It seems to me that there are very good policy reasons why the police would want to be in control of any reward process, even if the money came from the victim or a third party. These relate to safety issues in any reward drop, the risk of private deals as to prosecution (no questions asked) and general public policy issues such as whether it might encourage offending by others. For this reason, I think the correct approach it to “look though” the police to the actual reward giver.
Taking what I consider to be the correct “look through” approach, I consider for the reasons set out above, that Lord Ashcroft and Mr Sturgess have suffered either direct or consequential loss through or by means of the burglary and are therefore entitled to reparation.
The question of whether the police would be able to claim reparation in relation to rewards funded by the Crown is not before the Court. For myself, I consider that allowing reparation for police investigation costs would not be within the scope of the regime. It would risk cutting across the Costs in Criminal Cases Act 1967. In addition, costs of police investigation are primarily incurred to bring offenders to justice on behalf of the community. That is the statutory function of the police. The police, as a public body with that statutory function, should not be able to recover the costs of that investigation at the expense of victims – ie by sharing the pool of assets available for victims.
The majority consider that the insurance provisions in s 38 of the Sentencing Act support their conclusion that the reparation regime does not cover rewards paid by third parties.
The explanatory note, when the Bill was introduced, said that “the ability to make payments to the victim’s insurer is new”. In the Select Committee report, it was also stated “that there is a new provision that a sum payable under a sentence of reparation can be paid to the victim’s insurer, provided the victim consents”.
The 1981 report of the Penal Policy Review Committee had recommended that insurance companies should be able to be reimbursed directly under any reparation regime if they have already paid out to the complainant or in other appropriate circumstances. They considered that insurance companies have a sufficient interest to warrant being given status as a party.
Despite the comments in the 1981 report, the Criminal Justice Act reparation provisions did not deal specifically with the position of insurers. It is therefore true that a statutory provision dealing specifically with insurance companies is new. But the ability to treat an insurer as a person who had suffered loss through or by means of the offence had been recognised by the Court of Appeal in R v O’Rourke.
Mr O’Rourke was convicted of receiving a stolen car valued at $9,500. The police found the vehicle in Mr O’Rourke’s garage in the process of being stripped down for parts. Mr O’Rourke was ordered to pay reparation of $8,350 to the insurers of the victim of the theft. This was the value of the car less the amount received by the insurer for the parts.
The Court held that the insurer had an interest in the vehicle, either as a result of its obligation to indemnify the complainant or through its rights of subrogation to the complainant’s interest in the vehicle. The Court considered that, either way, “within the meaning and intention of s 22(1), the insurer can be regarded as having suffered loss or damage to the car through or by means of the applicant’s offence of receiving”.
The Court rejected a submission that the provision was not intended to benefit insurers in this way. It said that these provisions furnish a simple and speedy means of compensating those who suffer from criminal activities. The language is clearly wide enough to cover insurers and the Court could see no sensible reason why they should be excluded. The Court commented that in many cases they are the real losers, because insurance against risks of this kind is very common.
It must be assumed that Parliament was aware of O’Rourke. I consider that clearer wording would be needed to overrule the principle it stands for. It seems to me that s 38 of the Sentencing Act was principally designed to deal with the situation where an insured had not been paid out at the time the order for reparation is made or where there is uninsured loss. If the insured had already been paid out in full then he or she cannot be said to have suffered a loss. In my view, s 38 ensures, by providing that a payment of reparation can only be made to a person’s insurer with that person’s consent, that uninsured loss is compensated for in priority to loss suffered by an insurer paying out on a claim.
I accept that there may be issues of apportionment of reparation payments if both primary victims and a wider class of persons are entitled to reparation. These arise, however, because the reparation regime covers a wider category than the primary victim.
There is, however, nothing in the reparation provisions that requires pro rata apportionment and judges could decide, where an offender’s means are limited, to order reparation that compensates a primary victim fully but only partially compensates other victims.
In this case, I consider it was reasonable for reparation to be ordered with regard to the reward as that had led to the medals being returned (and thus meant the avoidance of greater loss), in priority to a reparation order in favour of the museum and its insurers.
The majority raise the possibility that an application could be made under the Criminal Proceeds (Recovery) Act 2009 to strip Mr Kapa of the reward. I agree. However, the fact that another remedy is available does not exclude the reparation regime.
Indeed, in my view the reparation regime is more suitable. Lord Ashcroft and Mr Sturgess could not force the Crown to make an application under the Act. The proceeds of crime also do not go to Lord Ashcroft and Mr Sturgess but to the Crown, although no doubt the Crown would arrange in this case to reimburse them. Also the proceeds of crime legislation would not cover a reward paid that led to the return of property if it was paid to someone other than the thief.
I would have dismissed the appeal.
 An internet search reveals numerous examples of the police offering rewards for information over the past decade.
 We shall call this agreement “the medals agreement”. Many of its terms are subject to a District Court suppression order. For that reason, we do not disclose the total amount paid under it.
 In the agreed summary of facts upon which Mr Kapa was sentenced, Mr Kapa accepted that he and Mr Van Wakeren had used Mr Comeskey “to broker a deal to return the medals for a sum of the reward money”.
 R v Kapa DC Auckland CRI-2008-083-2487, 26 August 2010 at .
 Kapa v R  NZCA 504 at  and . The Court of Appeal reduced the total sentence for this and other offending to 13 years three months’ imprisonment and ordered Mr Kapa to serve a minimum period of imprisonment of six years six months.
 At .
 Kapa v R  NZSC 1.
 Paragraph (a) of the definition of victim.
 R v Gill (1992) 10 CRNZ 632 (CA).
 At 639.
 See the discussion at 638–639.
 At 639.
 Sentencing and Parole Reform Bill 2001 (148–1).
 Reparation for emotional harm had become available following the enactment of the Criminal Justice Amendment Act (No 3) 1987: see s 4.
 Sentencing and Parole Reform Bill 2001 (148–1) (explanatory note) at 11.
 Sentencing and Parole Reform Bill 2001 (148–1).
 Sentencing and Parole Reform Bill 2001 (148–2) (select committee report) at 17.
 At 17.
 Sentencing and Parole Reform Bill 2001 (148–2).
 Now called the Sentencing Bill. The Sentencing and Parole Reform Bill was split into the Sentencing Bill and the Parole Bill by way of Supplementary Order Paper 2002 (260) Sentencing and Parole Reform Bill (148–2).
 We have used only the second meaning of “victim”. The words in italics substitute that definition for the word “victim”. The subclause becomes even more complicated if the additional meanings of “victim” are also inserted.
 Supplementary Order Paper 2002 (262) Sentencing and Parole Reform Bill (148–2) at 5.
 The subclause was enacted in this form.
 Some of these rights are restricted to certain types of offending: see, for example, pt 3 of the Victims’ Rights Act. In order to access certain rights, the victim must give contact details to the police in accordance with s 31 of the Victims’ Rights Act.
 Victims’ Rights Act, ss 17–27.
 Sentencing Act, s 8(f).
 Victims’ Rights Act, s 47.
 Parole Act, s 43(2)(b).
 Parole Act, s 49(4).
 Parole Act, ss 50, 50A and 50B.
 In , Glazebrook J observes that, “[i]t must be assumed that Parliament was aware of O’Rourke.” While it usually is reasonable to assume that legislation is based on an understanding of the existing case law, we consider that such an assumption is not correct in the present case. This is because s 38 is not consistent with the legislative understanding that insurers are otherwise entitled to the benefit of reparation orders. If that had been the legislative understanding, the text of s 38 would be different. As well the explanatory note (referred to by Glazebrook J at ) makes it clear that the legislation was prepared on the assumption that but for what became s 38 reparation orders could not be made in favour of insurers.
 R v Johnstone CA14/85, 15 April 1986; R v Neave CA206/88, 9 December 1988; and Cooper v Ward  NZHC 241; (1998) 3 CRNZ 366 (HC).
 Ross v Police HC Hamilton AP98/96, 17 February 1997.
 See, for example, Ross v Police, above n 33, at 3.
 R v Ebdell  NZCA 536.
 To use French J’s term: Ebdell v Police HC Christchurch CRI-2009-409-4831, 11 September 2009 at .
 It was “the owner of [the] property .... at the time of the theft” for the purposes of pt 10 of the Crimes Act 1961: see s 218(1).
 There is no time limit on such an application.
 Sentencing Act 2002, s 32(1)(a).
 Sentencing Act 2002, s 32(1)(c).
 R v Donaldson CA227/06, 2 October 2006 at –.
 In the United States, appellate courts have upheld orders requiring offenders to pay restitution of reward payments to the victim of the offence, in circumstances where the recipient of the reward is not an offender: see People v Dillingham 881 P 2d 440 (Colo App 1994) at 442 and State v Hazlitt 713 P 2d 617 (Or App 1986) at 619 in which the Court authorised the inclusion of reward payment in a restitution order if the reward was paid as a result of the defendant’s criminal conduct.
 I note that the Crimes Act 1961, as originally enacted, contained a provision (s 262) that made it an offence (punishable by up to three years’ imprisonment) to demand or accept a reward in return for recovering anything obtained through a crime, unless efforts were made to bring the offender to trial. No explanation was given for why this provision was not continued into the current version of the Crimes Act.
 R v Kapa (sentencing notes) DC Auckland CRI-2008-083-002487, 26 August 2010 at . In a statement in April 2010, Mr Van Wakeren told police that “right at the start we agreed the medals would always be returned, either for negotiation of charges or monetary gain”. I understand that Mr Van Wakeren was prepared to give evidence for the Crown.
 At .
 At .
 At .
 The Bill as originally introduced allowed a reparation order to cover loss or damage that was “consequential on the offending”: see Sentencing and Parole Reform Bill 2001 (148–1), cl 29(2)(a). This was changed by the Select Committee to the current wording. The Select Committee did not explicitly explain the change in wording but it appears to have been made because the Committee, by majority, recommended removing the ability to award reparation for physical harm because of cover under the accident compensation legislation: see Sentencing and Parole Reform Bill 2001 (148–2) (select committee report) at 17.
 When deciding whether a sentence of reparation is appropriate, a court must, under s 32(3), take into account the availability of a civil remedy for the person who has suffered loss. The main purpose of this provision is to prevent double-recovery. The lack of a civil remedy may in fact provide good reason for ordering reparation.
 Supplementary Order Paper 2002 (262) Sentencing and Parole Reform Bill 2001 (148–2). These changes were agreed to on 18 April 2002: (18 April 2002) 599 NZPD 15682–15683.
 See  of the majority judgment.
 See  above.
 As discussed below at –.
 See above at  and .
 R v Gill (1992) 10 CRNZ 632 (CA).
 R v Ebdell  NZCA 536.
 This can be compared with the current regime which expressly permits recovery for consequential loss. See above at  and n 48.
 At 637.
 The original wording of the Bill as introduced would clearly have covered the victim in Gill, even on the Court of Appeal’s interpretation. See above n 48.
 See for example R v Roberts (1972) 56 Cr App R 95 (CA); and R v Blaue  1 WLR 1411 (CA). See also AP Simester and WJ Brookbanks Principles of Criminal Law (3rd ed, Brookers Ltd, Wellington, 2007) at [3.3.3(3)].
 See discussion of the legislative history of the reparation provision below at – and discussion of R v Donaldson, above n 41, and below at –.
 Set out above at .
 Ebdell v Police HC Christchurch CRI-2009-409-004831, 30 July 2009.
 Reparation was also ordered in favour of those individual victims of the theft who could be identified. Neither NZ Post nor the individual victims received full compensation.
 At .
 At .
 Ebdell v Police HC Christchurch CRI-2009-409-004831, 11 September 2009.
 At .
 R v Ebdell, above n 56.
 At .
 The Court had perhaps overlooked the deeming provision whereby NZ Post was deemed to be the owner of the property at the time of the theft for the purposes of pt 10 of the Crimes Act 1961: see s 218(1).
 See above at .
 Despite the legislative history, which in any event does not absolutely point to an interpretation that the word “person” was not to be interpreted as meaning what it says on its face. See discussion above at – and below at –.
 See  above.
 See discussion at  above.
 As direct or consequential loss.
 R v Donaldson, above n 41, at . As pointed out by the Court of Appeal in that case, this is similar to the manner in which the courts in England have advocated a broad and non-technical approach to reparation provisions. For example, in Bond v Chief Constable of Kent  WLR 40 (QB) at 459 McCullough J said that the court simply has to ask itself whether the loss or damage can fairly be said to have resulted from the offence, referring to R v Thomson Holidays Ltd  QB 592 (CA) at 599, where Lawton LJ said that Parliament never intended to introduce into the criminal law the concepts of causation, which apply to the assessment of damages under the law of contract and tort.
 Law Commission Compensating Crime Victims (NZLC R121, 2010) at [2.4] [Law Commission report].
 At [2.4].
 Criminal Code Act 1893, s 419 provided for the payment of costs for the restitution of property.
 Crimes Act 1908, s 449 provided for compensation for loss of property through offending.
 In addition to the current provisions relating to reparation, the Crimes Act 1961 used to have a s 403 (now repealed), which had provided for compensation for loss of property through offending.
 Penal Policy Review Committee Report of the Penal Policy Committee (presented to the Minister of Justice, 1981) at . See also R v Donaldson, above n 41, at  and R v O’Rourke  1 NZLR 155 (CA) at 158. In R v O’Rourke, the Court of Appeal rejected any approach that would tend to “relegate [the reparation provisions] to the sidelines, as an inferior alternative to civil remedies, or which confines their application to only clear and simple cases”.
 The Penal Policy Review Committee was formed in 1981 to consider penal policy. Its terms of reference, at , included examining means of dealing with offenders, the reduction of imprisonment, criteria for imprisonment, community-based sanctions, flexible sentences, the reintegration of prisoners into society, culturally-appropriate penal policies and the position of victims in the criminal justice system.
 At .
 Subject to the type of practical limitations referred to by the Court of Appeal in R v Rollo  2 NZLR 667 (CA). As both the Rollo case at 672 and the Report of the Penal Policy Committee, above n 83, at  recognised, it may not be appropriate to order a sentence of reparation where the offenders are impecunious and have no prospect of paying reparation in the foreseeable future. The Penal Policy Review Committee noted that ordering reparation in such cases would only serve to “raise false hopes and cause disappointment, accentuating the victim’s sense of loss”. Of course, s 35 of the Sentencing Act 2002 now requires the court to take into account the financial capacity of the offender when deciding whether to impose a sentence of reparation.
 Sections 22–25 were the new provisions relating to reparation. In addition to s 22, which allowed the court to order a sentence of reparation, s 23 dealt with the preparation of a reparation report, s 24 dealt with the conditions of the sentence (such as the total amount to be paid and how it was to be paid), and s 25 dealt with the enforcement of the sentence.
 See above n 81.
 Law Commission report, above n 78, at [2.11].
 At [2.11].
 Note that s 419(1) of the Criminal Code Act 1893 used the phrase “loss of property” rather than “loss of or damage to” property. See also s 886 of the Criminal Code of Canada 1892 (on which the New Zealand criminal code was based), which used the phrases “loss of property” and “through or by means of the offence”.
 Reparation for emotional harm in the Criminal Justice Act 1985 was introduced in 1987 by s 4(1) of the Criminal Justice Amendment Act (No 3) 1987.
 Criminal Justice Act 1985 (as originally enacted), s 22(1) (repealed).
 Police v Newham (1986) 3 DCR 204.
 Like the Court of Appeal did in R v Donaldson, above n 41, at . I note too that the explanatory note for the Criminal Justice Bill (No 2) in 1984 used the words “through or by means of the offence” to explain cl 20(1) of the Bill (which became s 22 of the Criminal Justice Act) even though cl 20(1) used the word “caused”: Criminal Justice Bill 1984 (70–1) (explanatory note) at v.
 The Victims of Offences Act 1987 was the first victims-focused legislation. The Victims of Offences Act was intended to make provision for the fair treatment of victims of offences. This Act intended to implement a number of recommendations made by the Ministerial Committee of Inquiry into Violence, which considered issues relating to victims of crime: Department of Justice Submission to the Committee of Inquiry into Violence (November 1986) at 144–148 and Violent Offences Bill (No 2) 1987 (126–1) (explanatory note) at 1.
 In the Sentencing and Parole Reform Bill 2001 (148–1).
 Sentencing and Parole Reform Bill 2001 (148–1), cl 4.
 Sentencing and Parole Reform Bill 2001 (148–2), cl 4(a)(i) in relation to the definition of “victim”.
 Sentencing and Parole Reform Bill 2001 (148–2), cl 4(a)(ii) in relation to the definition of “victim”.
 See above at  and n 50.
 Sentencing Bill 2001 (148–3A), cl 4.
 This Act replaced and expanded on a number of principles that had been contained in the Victims of Offences Act.
 The Sentencing Act definition is narrower in the sense that it does not include a person who suffers emotional harm as a result of an offence, unless that person is a person against whom the offence was committed. For certain very limited purposes, however, those who suffer only emotional harm and who are not primary victims are included as victims in the Victims’ Rights Act: see (b)(i) of the definition of victim under s 4 of the Victims’ Rights Act. The Sentencing and Parole Reform Bill 2001 and the Victims’ Rights Bill 2001 were being considered by the Justice and Electoral Select Committee very close in time.
 See  below.
 United Nations Declaration of Basic Principles of Justice for Victims of Crime and Abuse of Power A/Res/40/34 (1985), annex at .
 Lovatt v Police HC Christchurch AP156/9, 2 August 1991.
 At 2.
 At 2–3.
 Wilmot v Police HC Dunedin AP25/96, 15 July 1996.
 At 3. Contrast Collins v Police  NZHC 311; (1989) 4 CRNZ 219 (HC) where the High Court quashed a reparation order made against a party convicted of receiving stolen property. The order was for the value of a car that had been recovered by the police, but which had subsequently been stolen from the police yard. The Court held that reparation could only be ordered for loss or damage caused to the vehicle between the time it was stolen and the time it was recovered, dealing with the theft from the police yard as a novus actus interveniens. See below at  and n 122.
 At 4.
 At 5.
 Wilson v Police HC Napier AP60/94, 13 February 1995.
 At 2.
 At 3. Also, contrast this decision with Campbell v NZ Police HC Auckland AP157/97, 1 August 1997.
 R v Donaldson, above n 41, at  and .
 At .
 The Court noted the new use of the word “caused” in s 32 of the Sentencing Act as against the word “suffered” in s 22 of the Criminal Justice Act but did not consider that this change in wording dictated a change in approach. Again I agree. The words “through or by means of an offence” remain and provide the controlling concept.
 At .
 See above at –.
 R v Donaldson, above n 41, .
 At .
 R v Johnstone CA14/85, 15 April 1986.
 R v Neave CA206/88, 9 December 1988; and Cooper v Ward  NZHC 241; (1988) 3 CRNZ 366 (HC).
 At 368–369.
 I do not overlook the fact that NZ Post was a deemed owner but, as discussed above, this does not mean that it had suffered any direct loss from the loss of the mail as it did not in fact own it.
 A victim impact report from a security advisor for NZ Post stated that Mr Ebdell’s co-workers fell under suspicion and were abused by members of the public. NZ Post, too, was affected in terms of the costs incurred in its investigation, the damage to its commercial reputation and the loss of people’s faith in the integrity of the mail system.
 See above at .
 See above at .
 Sentencing Act 2002, s 32(2). See above at  and .
 This approach has been taken by the case law. See above at  and .
 See above at .
 See above at  and .
 See s 29 of the Victims’ Rights Act 2002, which limits the rights set out at ss 30–45 of the Act to situations where the offence is sexual assault or serious injury or where there is fear for safety. The rights set out at [27(c)]–[27(e)] of the majority judgment only apply to “victims” as defined in s 4 the Parole Act. To be considered a “victim” under that Act, a person must have asked to be given notice of the release on bail or parole of the accused or offender under s 31 of the Victims’ Rights Act. Section 31 is one of the sections that applies to a limited number of offences under s 29 of the Victims’ Rights Act.
 This judgment only covers rewards for the recovery of property. I make no comment on rewards provided for other purposes. I am, however, attracted to the distinction drawn by French J between investigative and recovery costs, despite boundary issues between the two concepts: see above at .
 See  of the majority judgment.
 Barr v Police  NZSC 109,  2 NZLR 1 at –.
 Policing Act 2008, s 9. Law enforcement is one of the functions of police.
 Therefore I consider that the cases relating to the recovery of the costs of crime investigation are correctly decided. See R v Johnstone, above n 124; R v Neave, above n 125; Cooper v Ward, above n 125; and Ross v Police HC Hamilton AP98/96, 17 February 1997. This approach also provides an alternative means of distinguishing the cases where police purchased cannabis to induce offending.
 Sentencing and Parole Reform Bill 2001 (148–1) (explanatory note) at 12.
 Sentencing and Parole Reform Bill 2001 (148–2) (select committee report) at 17.
 Penal Policy Review Committee, above n 84, at . The Committee considered that there should be similar provisions enabling secured creditors, hire purchase, or finance interests to be recognised and protected.
 R v O’Rourke, above n 83.
 This section of the Criminal Justice Act is set out at  above.
 At 158.
M A Edgar for Appellant.
A Markham and P D Marshall (instructed by Crown Law Office, Wellington) for Respondent.
all rights reserved