The Retroactive Effect of Conditional
Obligations in Tax Law

Author: Marie-Pierre Allard

Table of Contents



Foreword

Canadian bijuralism poses several problems of interpretation for courts, lawyers and legal scholars. In our federal system, property and civil rights fall within the jurisdiction of the provincial legislatures, so provincial private law often must play a major role in the interpretation of federal statutes. For example, in Quebec, the Civil Code is called upon to complete the Income Tax Act when that Act imposes tax consequences on certain relationships governed by private law. This phenomenon sometimes clashes with the objective of applying the Act uniformly throughout Canada.

One of the most frequently encountered problem in this regard is determining the moment at which a disposition of property has taken place within the meaning of the Income Tax Act specifically when a conditional obligation is involved. The Civil Code of Québec provides for suspensive and resolutory conditions with retroactive effects that have no equivalent under the common law. Given this, should the retroactivity be recognized for tax purposes in Quebec, or should the notion of disposition be applied uniformly throughout Canada? This paper will consider the issue and propose some solutions.

The task turned out to be much more demanding, lengthy and complex than expected. I would like to thank Professor Marie Jacques of the Master’s in Taxation program at the Université de Sherbrooke, who agreed to act as essay director, for her moral support and the time she devoted to this paper. My thanks as well to Me Sandra Hassan, Legal Counsel at the Civil Code Section of the Department of Justice of Canada, who did not keep track of the many hours she spent to bring this project to fruition. I would like to thank all the people at the Civil Code Section and Revenu Québec who read the first version and provided their invaluable comments. Finally, I would like to thank Me Diane Bruneau, M. Fisc., who agreed to read and comment on the first version of this document, thereby contributing to many improvements.

Although all of these people have provided outstanding assistance, all opinions expressed herein are mine and I bear the sole responsibility for them.

Introduction

Bijuralism is one of the exceptional characteristics of Canadian law. In its Canadian incarnation, bijuralism is the coexistence of two systems of private law: civil law and common law. This juxtaposition of two constantly interacting legal systems is without a doubt a source of extraordinary richness for both systems, though there is a risk of hybridization through judicial interpretation – a risk that one of our greatest legal minds has warned about.[1]

Bijuralism is deeply and inextricably rooted in the history and legal tradition of the Canadian federation. In 1774, the Quebec Act [2] reintroduced the civil law in Quebec, reversing a ban imposed by the Royal Proclamation of 1763,[3] which ordered the use of English law in both criminal and civil matters. In the Constitution Act, 1867[4], which confers upon the provinces the exclusive authority to pass laws in relation to property and civil rights, confirmed the coexistence of two private law systems.[5]

The courts have always held that tax law is accessory to private law, as it merely specifies the tax consequences of contractual relationships between parties, which are governed by private law:

In my opinion fiscal law is an accessory system, which applies only to the effects produced by contracts. Once the nature of the contracts is determined by the civil law, the Income Tax Act comes into effect, but only then, to place fiscal consequences on those contracts. Without a contract, without a law and an obligation, there can be no fiscal levy. Application of the Income Tax Act is subject to a civil determination, whether such a determination be according to civil or common law.[6]

Thus, when a federal statute uses a private law word or phrase without defining it or giving it a specific meaning, one must refer to private provincial law in order to interpret it. In such a case, we say that provincial private law stands in a relationship of complementarity with federal law. But the Parliament of Canada may also choose to exercise its ancillary or incidental powers under section 91 of the Constitution Act, 1867 and establish its own private law rules, for purposes of federal law. Federal law is thus dissociated from the province’s private law.[7]

Moreover, since Canada is officially bilingual, bijuralism requires that Parliament speak to four distinct legal audiences, namely the Francophones and Anglophones of each of Canada’s two legal systems.[8] For several years, Parliament has been involved in the process of reviewing federal legislation in order to harmonize it with the civil law and common law systems, with a view to ensuring that “the civil law and common law are adequately reflected in both language versions.”[9] The Income Tax Act[10] is one of the statutes reviewed in this process.

This paper comes within the scope of the harmonization process and proposes some solutions to one of the problems that arise in applying tax law: the retroactive effect of civil law conditional obligations. In civil law, suspensive and resolutory conditions have effects that are retroactive to the date the contract was formed, unlike conditions precedent and conditions subsequent at common law,which have no retroactive effect. For tax law purposes, the moment of disposition of property is pivotal in determining, inter alia, the moment at which a capital gain is taxed, capital cost allowance is recaptured, or control of a corporation has changed. This raises the following question: where the issue is determining the moment in time when a disposition has occurred, does federal tax law recognize the retroactive effect of conditional obligations?

In this paper, we will attempt to determine whether the Act and the administrative policy of the Canada Customs and Revenue Agency[11] adequately recognize the unique features of civil law when it comes to the effect of conditional obligations. If not, we shall assess how and to what degree one might go about harmonizing the Income Tax Act with Quebec civil law on this issue, while giving effect to Parliament’s intent. In this regard, we shall be mindful of the balance between two objectives that are often in conflict: applying the Act uniformly throughout Canada, and respecting the private law rules of both legal systems although they are occasionally incompatible.

To this end, we will consider, in Chapter 1, the civil law rules regarding conditional obligations and their effects, and other civil law concepts that could involve issues of retroactivity. In Chapter 2, we will consider the common law concepts that correspond to civil law conditional obligations in order to identify both the similarities and differences between the two legal systems in this regard. In Chapter 3, we will analyze the retroactive effect of conditional obligations in tax law by considering the legislation, case law and scholarly writing on the issue as well as the Agency’s administrative position. The fourth and final chapter is a critical examination of the current situation, aimed at predicting likely developments in the law and proposing some reforms.

1. Conditional obligations under Civil Law

1.1 Suspensive conditions

Conditional obligations are governed by articles 1497 to 1507 of the Civil Code of Québec,[12] which defines such obligations as follows:

1497. An obligation is conditional where it is made to depend upon a future and uncertain event, either by suspending it until the event occurs or is certain not to occur, or by making its extinction dependent on whether or not the event occurs.[13]

Thus, a condition must be future and uncertain, and what is more, it cannot be purely potestative, which means it cannot be solely dependent on the debtor’s will.[14] A condition must also be an external event that is not essential to the formation of the contract. For example, a buyer who agrees to pay the price if the seller agrees to deliver the item is contracting a pure and simple obligation.[15]

When an obligation is subject to a suspensive condition, the creation of the obligation will depend on the occurrence of an event or on the certainty that the event will not occur; thus, the condition delays the creation of a relationship between the parties.[16] As long as the condition has not occurred, the very existence of the obligation is in abeyance.[17] The obligation is not only inexigible, as with a term; in fact it does not exist, as it has not yet come into being. If the obligation in question is payment, the debt has not legally arisen and a person who has paid in error can claim the money back.[18] Thus, the seller has no right to the price until the condition is fulfilled.

Ownership of property sold subject to a suspensive condition is not transferred immediately. The seller retains ownership and all the incidents thereof.[19] Occasionally, possession of the property may be transferred when the contract is formed, for example in a trial sale[20] (which is presumed to be subject to a suspensive condition) but this does not have the effect of transferring ownership.

When the condition is fulfilled, it has a retroactive effect to the date of conclusion of the contract, both between the parties and against third parties:

1506.    The fulfillment of a condition has a retroactive effect, between the parties and with respect to third persons, to the day on which the debtor obligated himself conditionally.

This provision reiterates the principle that a condition that has occurred has a retroactive effect, as stated in the first sentence of articles 1085 and 1088 C.C.L.C.[21] Thus, it is not a departure from the previous law. In this regard, the Civil Code of Lower Canada was practically identical to article 1179 of the Code Napoléon.

Thus, with a suspensive condition, the occurrence of the event causes the agreement to have become pure and simple from the beginning:

From what point does the obligation exist as a pure and simple obligation, however? The answer might seem clear: until such time as the suspensive condition occurs, the obligation is a conditional one; but from the moment the suspensive condition occurs, a pure and simple obligation is substituted therefore. And yet, the result is different under French law: art. 1179 C. civ. states that the condition is retroactive: it provides a condition fulfilled takes effect retroactively to the date the obligation was contracted. Everything happens as though the obligation had been pure and simple from the date the contract was formed; it is deemed never to have been merely a potential obligation.[22](Emphasis added.)

Thus, when the contract provides for the transfer of the right of ownership, that right is deemed to have passed to the buyer on the date the contract was signed.[23] The effects and consequences of retroactivity will be analyzed below.[24]

Where the condition is not fulfilled within the allotted time or when it becomes certain that it will not be fulfilled, the contract is, for all intents and purposes, considered never to have been formed.[25] Professors Pineau, Burman and Gaudet have written: [translation] “the slate is wiped clean: the erstwhile buyer never was a buyer and the erstwhile potential seller is considered never to have been a seller.”[26]

1.2 Resolutory conditions

A resolutory condition suspends the extinction of an obligation, not its existence.[27] Thus, an obligation subject to a resolutory condition comes into existence immediately, as soon as the contract is formed.[28] This means that as long as a resolutory condition remains unfulfilled, the obligation is treated exactly like a pure and simple obligation. It takes full effect, so that in a sale, ownership is transferred, and the price is payable, immediately.

Here is how Mignault has summarized the immediate effect of obligations subject to a resolutory condition:

[translation] The moment the contract of sale is formed it comes into full effect, just like a pure and simple sale. Both parties are bound to perform their obligations: the vendor must deliver the thing sold, and the purchaser must pay the price; ownership is transferred from the outset.[29]

When the condition is fulfilled, the contract is cancelled retroactively in accordance with article 1506 C.C.Q. Mignault goes on to write:

However, if the condition is fulfilled, all the effects of the sale are revoked retroactively: they are terminated not only as to the future, but also as to the past; they are considered never to have arisen.[30]

Several scholars have written that everything thereafter occurs as though the obligation had never existed: “the buyer never became the owner of the building (not even conditionally) and the seller has never ceased to be the full owner of the building.”[31]

The outcome of the converse situation is obvious: if it becomes certain that the condition will not occur, the sale is pure and simple from the outset and the contract is retroactively consolidated.[32] Thus, the buyer is deemed to have been the owner from the date the contract was signed.[33]

One should note the “mirror” effect of suspensive and resolutory conditions: all conditional obligations are suspensive for one party and resolutory for the other. Mignault illustrates this reciprocity by using a concrete example:

I have sold you my house under this condition: that a certain ship arrives. If the condition is realized, it has a double effect; thus you will be deemed to have been, and I will be deemed to have ceased being, owner, from the date of the contract (art. 1085). Thus, we were both owners: you were an owner under a suspensive condition, and I was an owner under a resolutory condition.

I sell you my house, but on the condition that the sale will be resolved if a certain ship arrives. The sale, like a pure and simple sale, produces all of its effects hic et nunc: you are the owner from this very moment onward; but if the condition is fulfilled, you are deemed never to have been the owner, and I am deemed never to have ceased being the owner (art. 1088): the very same event that deprives you of a right has the effect of granting me that right. We were therefore both owners: you were an owner subject to a resolutory condition and I was an owner subject to a suspensive condition.[34]

Essentially then, before the condition is fulfilled in a sale subject to a resolutory condition, the seller is the owner under a suspensive condition and the buyer is the owner under a resolutory condition. In the reverse situation, i.e. a sale subject to a suspensive condition, the seller before the occurrence of the condition is an owner under a resolutory condition, and the buyer is an owner under a suspensive condition during that time.[35]

1.3 What it means for a condition to be retroactive

1.3.1    General effects of retroactivity

The main effect of the fulfilment of a suspensive condition is that the parties must perform their obligations as though the obligations had existed from the date the contract was signed. Conversely, the main effect of the occurrence of a resolutory condition is to restore the prestations under the contract as though it had never existed.[36]

What occurs when a suspensive condition is not fulfilled? If possession has not been transferred pendente conditione,[37] the contract has simply never existed and the parties owe nothing to each other.[38] But if possession has been transferred, the buyer subject to a suspensive condition must return the property to the seller; in this regard, a sale under a suspensive condition is subject to the same restitutionary process as a sale under resolutory condition.

Articles 1699 to 1707 C.C.Q. now govern the restitution of prestations, whether such restitution is needed because a resolutory condition has been fulfilled, a suspensive condition has failed, a creditor has demanded resolution,[39] or the contract has become null[40] because it does not meet the necessary conditions of its formation.[41]

Article 1700 C.C.Q. specifies that the restitution of prestations be made in kind. The general rule is that the buyer must give the property back to the seller, and the seller must refund such part of the purchase price as was received. As we shall soon see, this will only be problematic if part of the thing is lost.

Retroactivity also has effects on third parties. The fulfilment of a condition has the effect of annulling the rights granted by a seller under a suspensive condition, or by the buyer under a resolutory condition, since their ownership is defeated retroactively and thus those rights would have been granted in the “property of another.” On the other hand, all rights granted by the purchaser under a suspensive condition or by the seller under a resolutory condition during the period of uncertainty, are confirmed by the fulfilment of the condition.[42]

This rule was applicable both under French law and under the Civil Code of Lower Canada. But new article 1707 of the Civil Code of Québec seems to call this principle into question:

1707.    Acts of alienation by onerous title performed by a person who is bound to make restitution, if made in favour of a third person in good faith, may be set up against the person to whom restitution is owed. Acts of alienation by gratuitous title may not be set up, subject to the rules on prescription.

Any other acts performed in favour of a third person in good faith may be set up against the person to whom restitution is owed.

Does this article make a fundamental break with prior law and are thus conditional obligations no longer retroactive as against third parties?  We believe the answer is no.

One should begin by noting that the acts of alienation contemplated in the first paragraph of article 1707 include only acts that transfer ownership. All other acts, including grants of real rights (such as a hypothec or prior claim) in the thing, are covered by article 1707 paragraph (2).[43]

It must be understood that article 1707 C.C.Q., found in Chapter IX of the Civil Code entitled “Restitution of Prestations”, applies to all contracts retroactively annulled. Reasons for such annulment extend beyond the failure or fulfilment of a resolutory or suspensive condition; they also include the failure of a condition of formation of a contract. It is very difficult, if not impossible, for a third party to know the reasons why a contract may be, or may have been, annulled. However, where conditional obligations affecting immovables were involved, that land registration formalities under the former Code were considered sufficient to protect third party rights, as those formalities ensured that third parties knew how precarious their  rights were.[44]

The same principle remains applicable under the new Civil Code. In fact, where immovables are involved, the creditor’s right of restitution is published in the land register. All purchasers are deemed to know of the registered rights,[45] which means that a third party cannot invoke article 1707 C.C.Q. because he cannot claim to be acting in good faith.[46] Moreover, where rights published in the register of movable personal and real rights are involved, the same deemed knowledge rule applies to third parties, although that particular deeming is rebuttable.[47] Thus, in most cases where the debtor of the obligation of restitution grants real rights to third parties, the rights will be retroactively annulled, having been granted by a person who never was the owner, and the maxim nemo dat quod non habet applies.[48]

In addition, article 1506 C.C.Q., which specifically provides that a condition fulfilled has a retroactive effect as against third parties, articulates a specific rule on conditional obligations, as opposed to article 1707 C.C.Q. which contains a general rule. Thus, article 1506 should normally take precedence. This opinion is shared by certain scholars who believe that article 1707 C.C.Q. does not apply to conditional obligations.[49] Mr. Justice Baudouin and Professor Jobin still write, despite article 1707 C.C.Q., that the fulfilment of a suspensive condition retroactively annuls the rights granted by a seller pendente conditione:

[TRANSLATION] In contracts that transfer ownership, the right of ownership is deemed to have passed to the creditor on the date the contract was signed. Consequently, any act done by the debtor regarding the thing, prior to the fulfilment of the condition, is defeated. Thus, the disposition of the material thing that the contract, or hypothecs or other securities, or servitudes granted by the debtor, would in principle not have any effect in relation to the creditor. On the contrary, all acts entered into by the creditor during the same period are retroactively validated, since they were done when the creditor is deemed to be the thing’s owner.[50]

They have the same view regarding resolutory conditions:

As for contracts that transfer ownership, the buyer is deemed never to have been the owner. Thus, in theory, any right that he has granted to third parties retroactively fails.[51]

In fact, article 2682 C.C.Q. appears to confirm that article 1707 C.C.Q. does not apply to conditional obligations, at least in relation to hypothecs:

2682. A person whose right in a property is conditional or open to an attack in nullity may only grant a hypothec subject to the same condition of nullity.

Acts of administration pendente conditione, were considered under the old doctrine (i.e. scholarly writing) to be opposable to the person who owed restitution: in this regard, article 1707 C.C.Q. merely codifies the doctrine on the question. This will be discussed in greater detail below.[52]

Essentially then, it is reasonable to conclude that article 1707 C.C.Q. does not negate or undermine the principle that the retroactivity of a condition can be set up against third parties ? a principle that is expressly articulated in article 1506 C.C.Q. It would make no sense for the legislator to have stated the principle and then deprived it of any effect.

Finally, as explained in the following translation from Faribault, one of retroactivity’s predominant effects is the cristallisation of a legal situation despite any subsequent amendments to a statute: 

If a new law comes into effect between the date of the agreement and the date on which the condition is fulfilled, the obligation will remain subject to the old law, as though it had been pure and simple from the outset.[53]

As we shall soon discuss, this rule is particularly important in tax law:[54] the retroactive effect of conditions in tax matters can give the parties a reasonably secure idea, in advance, of the tax consequences of the transaction they have in mind. This is because they will be protected from any potential amendments to the Income Tax Act subsequent to the signature of their contract.

1.3.2    Limitations to the retroactive effect of conditions

1.3.2.1    Risks

Under the Civil Code of Lower Canada, the principle, for contracts that transfer ownership, was that the owner bore the risks.[55]  Thus, the maxim res perit domino[56] applied. However, articles 1087 and 1088 C.C.L.C. established an exception to this rule in relation to conditional obligations. Article 1087 read as follows:

1087.    When the obligation has been contracted under a suspensive condition, the debtor is bound to deliver the thing which is the object of it, upon the fulfilment of the condition.

If without the fault of the debtor, the thing have altogether perished or can no longer be delivered, no obligation exists.

If the thing be deteriorated without the fault of the debtor, the creditor must receive it, in the state in which it is, without diminution of price.

If the thing be deteriorated by the fault of the debtor, the creditor may either exact the thing in the state in which it is, or demand the dissolution of the contract, with damages in either case. (Emphasis added.)

If the thing be deteriorated by the fault of the debtor, the creditor may either exact the thing in the state in which it is, or demand the dissolution of the contract, with damages in either case. (Emphasis added.)

The second paragraph of this provision reversed the doctrine of general risk and ran counter to the principle that conditions are retroactive. Indeed, it places the burden of the risk of loss on the debtor of the obligation to deliver: if the thing perishes completely, the seller is no longer required to deliver, but the buyer is no longer required to pay.[57] Thus, the seller bears the loss. The normal rules of retroactivity would have led to the contrary result. Faribault notes this contradiction:

Naturally, this provision in our article 1087 contains an exception to the rule stated in article 1085 to the effect that conditions are retroactive.

Pursuant to this rule, the creditor must bear the risks, for after the condition has been fulfilled the sale is deemed to have been pure and simple from the moment the contract was formed, and the buyer owned the thing from that point onward. By applying the maxim res perit domino, the buyer, i.e. the creditor of the obligation to deliver, would logically have to bear the loss. That is not the result provided for by article 1087, which is quite similar to article 1182 of the Code Napoléon.[58]

When the Civil Code was reformed, the special risk allocation regime applicable to conditional obligations was purposely set aside. From then on, it was decided that the general rules pertaining to the transfer ownership further to the conclusion of a contract, i.e. the rules found in articles 1456, 1693 and 1694 C.C.Q., would apply to conditional obligations.[59] Those provisions are as follows:

1456. The allocation of fruits and revenues and the assumption of risks incident to property forming the object of the contract of a real right transferred by contract are principally governed by the Book on Property.

The debtor of the obligation to deliver the property continues, however, to bear the risks attached to the property until it is delivered.

1693. A debtor is released where he cannot perform an obligation by reason of a superior force and before he is in default, or where, although he was in default, the creditor could not, in any cases, benefit by the performance of the obligation by reason of that superior force, unless, in either case, the debtor has expressly assumed the risk of superior force.

The burden of proof of superior force is on the debtor.

1694. A debtor released by impossibility of performance may not exact performance of the correlative obligation of the creditor; if the performance has already been rendered, restitution is owed.

Where the debtor has performed part of his obligation, the creditor remains bound to perform his own obligation to the extent of his enrichment.

Although res perit domino continues to be the general rule as far as risks are concerned,[60] it has been set aside in favour of res perit debitori[61] where the contract transfers ownership. Here is how Professor Pineau explains the new provisions:

Granted, article 950 of the new Civil Code states that the owner of the property bears the risks of loss, but article 1456, paragraph 2 specifies that the debtor of the obligation to deliver the property (in which a real right is transferred by contract) continues to bear the risks attached to the property until it is delivered. Thus, the buyer who has become owner no longer assumes the risks. Instead it is the seller, who is the debtor of the obligation to deliver, that assumes them. Thus, the rule in this context is res perit debitori.

So the change is significant. The new rule of law is inspired in part by Article 69 of the United Nations Convention on the International Sale of Goods (Vienna, 1980) which became part of Quebec law by virtue of S.Q. 1991, c. 68. The rule takes into account the fact that [translation]“a person in possession of the property is in a better position to take the appropriate measures to protect it.” Thus, possession, not ownership, becomes the pivotal factor.[62](Emphasis added.)

Thus, the exceptional regime that applied to risks in the area of conditional obligations became the general rule regarding contracts that pass ownership: risks are no longer tied to ownership, but rather, to the obligation to deliver the property. In other words, risks follow possession. Thus, in a conditional sale, the allocation of risks to the debtor of the obligation to deliver is no longer an exception to the retroactivity rule: it merely follows the general rule for contracts that transfer ownership.

Legal scholars Pineau, Burman and Gaudet provide the following general explanation of the risks associated with conditional obligations:

In such contracts, the risks are borne by the debtor of the obligation to deliver, who had possession of the property when it was lost due to a superior force. Thus, if the thing perishes pendente conditione, i.e. if the condition is fulfilled, the debtor of the obligation to deliver assumes the risks. That person is normally the seller in a sale subject to a suspensive condition and the buyer in a sale subject to a resolutory condition. For normally, pendente conditione, the person in possession of the thing sold conditionally and debtor of the obligation to deliver is the seller in the former case, and the buyer in the latter.[63](Emphasis added.)

We shall now consider how this rule applies to obligations involving a resolutory or suspensive condition.

1.3.2.1.1   Resolutory conditions

Where a resolutory condition is involved, the risks, in accordance with the rule res perit domino,[64] are transferred to the buyer from the moment the contract is formed since the buyer obtains the right of ownership immediately. Since he generally takes possession of the thing at the moment the contract is formed, article 1456 C.C.Q. will not counteract the effect of article 950 C.C.Q.

If the resolutory condition never occurs, the buyer will of course have to assume the risks of loss. Conversely, if the resolutory condition occurs before the thing is lost, everything happens as though there had never been a contract and the seller will have to bear the loss of the property.

The problem arises if the resolutory condition occurs after the loss. As a result of retroactivity, the buyer is considered never to have been the owner. In addition, the fulfilment of the condition gives rise to an obligation of restitution under articles 1699 et seq. of the Civil Code.[65] Article 1701 provides as follows:

1701. In the case of total loss or alienation of property subject of restitution, the person liable to make the restitution is bound to return the value of the property, considered when it was received, or at the time of its loss or alienation, whichever value is the lowest, or, if the person is in bad faith or if the restitution is due to his fault, whichever value is the highest.

If the property has perished by superior force, however, the debtor is exempt from making restitution, but he shall then assign to the creditor, as the case may be, the indemnity he has received for the loss of the property, or, if he has not already received it, the right to the indemnity. If the debtor is in bad faith or if the restitution is due to his fault, he is not exempt from making restitution unless the property would also have perished if it had been in the hands of the creditor.

The first paragraph does not raise the question of risks: the debtor of the obligation of restitution, i.e. the buyer subject to a resolutory condition, is not released from the obligation, but rather, is required to make restitution by giving equivalent property. The seller is therefore also required to make restitution of the prestations received.[66]

The second paragraph, however, releases the buyer from the obligation of restitution if the loss is due to a superior force. It must therefore be determined whether the seller is also released from the obligation to refund the purchase price. Since article 1701 C.C.Q. does not speak to this point, one must refer to the general rules governing risks. The buyer under a resolutory condition, being the debtor of the obligation of restitution, might be considered here as the debtor of the obligation to deliver within the meaning of article 1456 C.C.Q.[67] Consequently, res perit debitori would apply and the debtor of the obligation to deliver (or of restitution) would have to bear the loss. Thus, the buyer who cannot return property that is lost by reason of a force majeure cannot recover the purchase price and will have to pay it if he has not already done so.

1.3.2.1.2   Suspensive conditions

Let us begin with the case in which the seller retains possession of the thing pendente conditione. When the suspensive condition is fulfilled after the fortuitous loss of the thing, the seller, being the debtor of the obligation to deliver, will be released of the obligation to deliver, but the buyer will not have to pay the price, as stated in 1456 C.C.Q. Thus, it is the seller who bears the risks of loss.

This result is logical, for as we have seen, the buyer subject to a resolutory condition is also the debtor of the property under a suspensive condition;[68] this stems, as we have discussed above, from the reciprocal nature of the suspensive and resolutory conditions.[69] It is therefore normal for the seller under a suspensive condition to be expected to bear the risks just as the buyer under a resolutory condition does.

If the condition fails, however, the problem of risks does not arise as the seller has remained at all times the owner and possessor of the property.

The problem is different in cases where the buyer has taken possession of the thing pendente conditione. If the thing is lost before the condition occurs, article 1456, para. 2 C.C.Q. cannot apply as the property has already been delivered. Instead, the general rule in article 950 C.C.Q. will apply and the owner will assume the risks. So the question will be whether one must take account of the retroactive effect of the condition. If so, the buyer should bear the loss, being the one deemed to have been the owner since the contract was formed. If not, the buyer will not have to pay the price and the seller will bear the loss. 

According to Mignault, the seller bears the loss. In his opinion, the contract cannot be formed when the condition arises unless all the elements necessary to the formation of the contract are present at that time. Since one cannot deliver something that no longer exists, the seller’s obligation no longer has an object. In addition, the buyer’s obligation to pay the price no longer has a cause since the cause was the seller’s obligation to deliver, which has evaporated. Consequently, the contract cannot be created since there is no cause or object.[70]

Nonetheless, we believe the buyer should assume the risks of loss if in possession of the property. For one thing, to do so is to be consistent with the modern principle that possession is the decisive factor as far as risks are concerned.[71] This result is more compatible with the treatment of resolutory conditions, since a buyer under a suspensive condition who has taken possession of the property is the potential debtor of an obligation of restitution, just like the seller under a resolutory condition. Article 1456 C.C.Q. should therefore apply to impose the risks on the buyer.

It can therefore be concluded that in conditional sales under the Civil Code of Québec, the transfer of risks is no longer tied to ownership of the property but depends on possession. In all cases, subject to the uncertainty mentioned above with regard to suspensive conditions where possession is taken pendente conditione, the risk of loss is transferred to the buyer with possession of the thing, whether or not the condition is fulfilled.

Thus, since risks are now tied to possession, the retroactive nature of the condition no longer has an impact on risks, because the retroactivity does not affect the possession of the thing pendente conditione. As for cases involving a suspensive condition and a transfer of possession, even if our opinion is not shared, retroactivity does not apply because the contract simply could never have come into existence.

1.3.2.2    Fruits

The general rule is that fruits and revenues (generally referred to as “fruits”) belong to the owner.[72] Given the retroactive nature of conditions, whoever is deemed to be the owner on the date the contract was signed should be entitled to the fruits. That would be the seller where the suspensive condition is not fulfilled or the resolutory condition occurs, and the buyer in the reverse situation. Consider, for example, a sale contract subject to a resolutory condition. If the condition is fulfilled, the seller is deemed always to have been the owner, and the buyer who had possession of the thing pendente conditione and collected the fruits is not entitled to them because he was never the owner.

This was not the solution under the old law, however. It had been well settled, by scholars under the Civil Code of Lower Canada[73] and French law,[74]that the person who collected the fruits could keep them, and did not owe anything to the real owner. The authors invoked a number of justifications for this position. For Baudry-Lacantinerie, it was essentially a matter of fairness:

It is incontestable that if the object of the contract does not bear fruit, the seller under a suspensive condition, or the buyer under a resolutory condition, are entitled to use it during the interim period. And once the condition is fulfilled, they are not required to pay rent to the other party for any benefit derived from the thing. If the thing is fruit-bearing, why would they have to restore the fruit? What would be the reason for the difference? ... Should one not conclude that there is no distinction between fruit-bearing and non-fruit-bearing things?  Some object, citing art. 547 [Translator’s Note: Art. 547 of the Code Napoléon.] The provision states that the fruits “belong to the owner through accession”. Thus, the property must belong to the contracting party who, by operation of retroactivity, was the owner when they were collected. Applying the principle in art. 547 would lead to a shocking result. Is this not proof that it should not be applied?  Is it not highly material that the very beneficiary of the alleged restitution chose to give possession to the seller under a suspensive condition, or the buyer subject to a resolutory condition?  Would not the likely intent of the contracting parties support my view?[75] (Emphasis added.) (Translated by author)

Faribault appears to state that the true reason for fruits being exempt from restitution is as follows:

For his part, Demolombe believes that the true reason underlying this doctrine is that retroactivity operates in jure and not in facto. This was the reason accepted by Mignault.

Demolombe says: “Since it is an ineffaceable fact that the debtor had both the possession and the enjoyment of the property, the consequences of this fact must also be ineffaceable. Thus, it would appear to me that one of the most logical consequences would be that fruits are collected and acquired.

To my mind, this reasoning reflects the true state of the law.”[76](Translated by author)

However, Faribault articulates an additional argument, also advanced by Baudry-Lacantinerie, and supported by Demolombe. Mignault, for his part, explains his reasoning as follows:

The person who alienates property does not have to restore the fruits collected pendente conditione from a thing alienated subject to a condition ...: retroactivity attached to a fulfilled condition only applies to de jure matters. It was meant to benefit the buyer, who would otherwise be subject to any alienations, servitudes or hypothecs granted by the alienator pendente conditione. It was also meant to benefit his heirs. It does not affect de facto matters. That the fruits have been collected cannot be denied. It is a fait accompli, which the fulfilment of the condition should not be allowed to reverse.[77]

Retroactivity cannot alter reality retroactively and will therefore not affect events, or “ineffaceable” facts, that actually happened pendente conditione. We will return to this question further on.

As we have seen, the possessor of the thing pendente conditione was entitled, under the old law, to keep the fruits he collected. This is codified in article 1704 of the Civil Code of Québec, which states that the “fruits and revenues of the property being restored belong to the person who is bound to make restitution”, except if he is in bad faith or if the restitution is due to his fault. Thus, if the condition occurs, a buyer subject to a resolutory condition who must restore the property to the seller is nonetheless entitled to keep the fruits, produced by the property, that he collected pendente conditione. Similarly, a buyer under a suspensive condition can keep the fruits even if he must restore the property if the condition is not fulfilled.

A seller under suspensive condition can keep the fruits even if the fulfilment of the condition retroactively defeats his ownership of the property. In our opinion, the same solution applies in such a case, even though this situation is not one of restitution stricto sensu and therefore is not directly mentioned in article 1704 C.C.Q. Clearly, the legislator wanted to retain the old rule: the person who collected the fruits was entitled to keep them, independent of the retroactive nature of the condition.

1.3.2.3    Acts of administration

Under the old law, acts of administration that the possessor performed pendente conditione could be set up against a party who retroactively became the owner after the condition took effect.[78] This position appears to have been adopted for practical reasons:

It would be very difficult, if not impossible, to administer the thing that is the object of the contract if the solution were otherwise. No one would want to deal with the interim possessor if leases he signed failed due to the occurrence of the condition. Would the legislator have accepted such an economically poor solution?”[79] (Translated by author)

The law would presume that the intention of the parties was to give the debtor a “mandate” to administer the property on behalf of the real owner:

 …The cases have long admitted that acts of administration, such as leases, performed by the owner subject to a resolutory condition remain valid despite the retroactivity. We do not deny that after the resolution, that owner is deemed never to have had any right in the thing. But he nonetheless would in some way have acted as mandatary (agent) for the owner subject to a suspensive condition for the purpose of administering the thing until the fulfilment of the condition.[80]  (Translated by author)

Faribault appears to support the idea of this tacit mandate:

The law is presumed to have excluded the parties from the rule of retroactivity on the basis that it was the parties’ intent to do so. The creditor, by leaving this specific thing within the hands of his debtor until the occurrence of the condition, necessarily expected him to administer it and collect its fruits.

Moreover, it is in the interests of the parties, and of society in general, that the object of the contract be administered pendente conditione.

It follows that acts of administration performed by the debtor while he was in possession are not covered by the rule that conditions are retroactive.

The creditor is presumed to have given the debtor a tacit mandate to administer the thing since he is the only one able to do so while he has it in his possession.[81] (Translated by author)

However, Faribault, in support of his position, also makes the same argument regarding fruits:

Under our law, it seems to me that acts of administration done by the debtor while he had possession, that were actually performed and that produced results, cannot be defeated by the fulfilment of the condition. Therefore, the condition cannot be retroactive in this area.[82] (Translated by author.)

As we discussed[83] article 1707 of the new Civil Code states that “any other acts performed in favour of a third person in good faith may be set up against the person to whom restitution is owed.” This article, if and when it applies, only codifies the former law on acts of administration. Only true acts of administration will be discussed here, such as conclusion of a lease, and not “other acts” possibly falling within article 1707, paragraph 2, which some would say include consent to real rights such as hypothecs.[84]

1.3.3    Interpreting retroactivity restrictively

The origins of retroactive conditions can be traced back to Roman law. However, modern authors believe that classic Roman law did not know of the principle of retroactivity. Rather, they believe that the principle results from a historical error of interpretation by old French authors, particularly Pothier, who apparently believed that retroactivity was the general rule that explained the transfer of conditional obligations to heirs. It is believed that this principle was subsequently codified in article 1179 of the Napoleonic Code,[85] upon which article 1085 C.C.L.C., and now article 1506 C.C.Q., are based.

Several authors have criticized the existence of the principle of retroactivity.[86]  For one thing, they argue that the wording of the law itself contradicts many consequences that logically stem from retroactivity, such as the allocation of risks. Moreover, they note that consequences usually attributed to retroactivity can very well be explained using other principles. For example, the creditor’s right to implement conservatory measures can be explained by the existence of a "contingent" right pendente conditione.[87] As well, real rights granted to third parties by the seller subject to a suspensive condition or the buyer subject to a resolutory condition can be cancelled using the nemo plus juris ad alium transferre potest quam ipse habet [88] rule, without recourse to retroactivity.[89] Furthermore, many civil law countries, specifically Germany, Switzerland and Japan, have not codified the rule of the retroactive condition in their Civil Codes.[90]

For these reasons, Baudry-Lacantinerie concluded that retroactivity is a “fiction”:

It is not legally necessary for a condition that is fulfilled to have a retroactive effect.

Furthermore, from the preceding we can see that from a practical standpoint, this fiction unnecessarily complicates the application of the condition . ...

Whenever it is expressly recognized in the Civil Code, civilian scholars cannot be expected to disregard the retroactive nature of a fulfilled condition. However, because this retroactivity is a fiction, it would be wise, at the very least, to apply a narrow interpretation in those cases where the application of that fiction is questionable.[91]

Faribault pushes this idea a little further:

The retroactive nature of the fulfilled condition must be given a narrow interpretation because it is a legal fiction. If there is a doubt, it must be considered not to exist because its fictitious nature cannot prevent a fact from having occurred in the past.

This fiction cannot erase the fact that the conditional debtor had possession pendente conditione where the object of the obligation is a specific thing. If this possession escapes the condition’s retroactivity, anything that flows naturally and logically also escapes it, such as acts of administration performed by the debtor, and the fruits he collected during this possession.[92] (Translated by author)

Mignault does not use the word “fiction” in reference to retroactivity. But he nonetheless appears to argue that retroactivity applies solely to de jure, not de facto, matters. For the sake of convenience, an excerpt found above in this paper is reproduced again:

The person who alienates property is not required to restore the fruits collected pendente conditione from a thing alienated subject to a condition ...: retroactivity attached to a fulfilled condition applies only to de jure matters. It was meant to benefit the buyer, who would otherwise be subject to any alienations, servitudes or hypothecs granted by the seller pendente conditione. It was also meant to benefit his heirs. It does not affect de facto matters. The acquisition of fruits is a fait accompli which the fulfilment of a condition should not be allowed to reverse.[93]

Although both Faribault’s and Mignault’s statements may appear to be sweepingly broad, we believe their scope is much narrower than it seems at first glance. In fact, most French scholars[94] seem to reject Demolombe’s contention that retroactivity does not apply to de facto matters. Moreover, even Baudry-Lacantinerie, who maintained that retroactivity is a legal fiction, did not adopt the idea that it would only apply to that which is de jure. As we have seen,[95] he relied on other grounds for his submission that fruits and administrative acts are not affected by the retroactivity, and the idea of “legal fiction” was only raised as a subsidiary argument in both instances.[96]

Professor Leloutre also criticized this position:

[TRANSLATION] It is said that the owner subject to a resolutory condition should keep the fruits because retroactivity operates in jure, not in facto. It can no doubt result in the owner being considered, from one day to the next, never to have been the owner. But this cannot defeat the fact that he collected the fruits of the thing or that he enjoyed it. It cannot prevent the acquisition of fruits from being permanent.

It is surprising that such reasoning was accepted. The retroactive nature of the condition cannot prevent the owner subject to a resolutory condition from having enjoyed the thing. And it does not mean that because he was not the owner, he enjoyed it without being entitled to do so. None of this will prevent him from having to make restitution, however.[97]

In addition, as pointed out earlier,[98] Faribault himself relied on other reasons for his position regarding fruits and administrative acts. As for Mignault, his discussion only applies within the context of the restitution of the fruits.

The scholarly writing does not enable us to come to a clear conclusion on this question. However, in our opinion, there does not appear to be a general principle stating that the retroactive nature of a condition applies only to de jure matters and not to de facto matters. Rather, it appears that some scholars supported this idea in an attempt to justify a policy not to apply retroactivity in certain very specific situations, such as the collection of fruits and acts of administration.

We believe that even if such a principle does exist, its application should be limited to what might be described as “real and incontestable facts”, or ones that, to quote Demolombe, are “ineffaceable” ¾ that is to say, tangible, irreversible events that have actually occurred. Possession and enjoyment of the thing, collection of the fruits, and performance of acts of administration, are the indelible facts in question. We have already expressed the view that neither possession of the thing pendente conditione, nor any of its direct consequences (i.e. the risks) can be affected by retroactivity.[99]

The significance of applying retroactivity to de jure and de facto matters will be re-examined further on in the section about disposition under tax law.[100]

In any event, the Quebec legislature deliberately chose to preserve the general rule regarding the retroactive nature of conditions, as set out in article 1506 C.C.Q., despite the scholars’ nebulous position and the limits to the rules pertaining to retroactivity. The legislator’s intent seems clear: the principle should apply to conditional obligations in Quebec civil law, subject to the specific exceptions in the code regarding fruits, risks and acts of administration.

1.4 Other Civil Law concepts that raise questions of retroactivity 

We have just considered the civil law rules that govern conditional obligations. But other civil law concepts sometimes have retroactive consequences as well. Although we will not discuss them at length, we will provide an overview so that the tax treatment of those concepts can be considered at the same time as we discuss the tax treatment of conditional obligations.

1.4.1    Resolution of contracts for non-performance of an obligation

A creditor whose debtor has failed to perform his obligation may demand execution in kind or ask that the contract be resolved.[101] Resolution has a retroactive effect as the contract is considered never to have existed.[102]

In situations involving sales, article 1740 C.C.Q. specifically provides that the seller of a movable has a right of resolution if the price is not paid. Article 1742 C.C.Q. allows the seller of an immovable to do the same, provided, however, that the contract contains a resolutory clause. If it does contain such a clause, the seller will also be required to abide by the formal requirements set out in the Civil Code, including the sending of a 60-day notice[103] and compliance with notice requirements of the Book entitled “Prior Claims and Hypothecs.”

Resolution for non-performance has the same effects as the fulfilment of a resolutory condition:

[TRANSLATION] Resolution annihilates the contract retroactively: it is deemed for all purposes never to have been formed, and if the contract was subject to a resolutory condition, the situation is in all respects as though the conditional event had occurred. This retroactive elimination places the parties back where they would have been if they had never contracted. Thus, if certain obligations have already been performed, the prestations must be restored in accordance with arts. 1699-1706 C.C.Q. (art. 1606 C.C.Q.).[104]

This means that our previous discussion about restitution applies here. However, article 1743 C.C.Q. states that the seller of an immovable takes the property back free of any charges that the buyer may have placed on it after the seller registered his rights. The rule makes perfect sense since the seller must register (publish) his right of resolution in order for it to be opposable to third parties:[105] as we have seen, third parties cannot be considered in good faith in such a case, as they are deemed to have had notice of the resolutory clause.[106]

It is important, however, not to confuse resolution for failure to perform and resolutory condition: the condition must be a future and uncertain event that is external to the legal relationship between the parties, which resolution is a sanction for the wrongful inexecution (non-performance) of a principal obligation of a contract. Professors Pineau, Burman and Gaudet explain the distinction as follows:

One must however make sure to distinguish between an express resolutory clause and the resolutory condition considered earlier in association with conditional obligations. Clauses are designed to punish wrongful inexecution of an obligation, not to subject the contract to the occurrence of a future and uncertain event. Admittedly, voluntary performance may be uncertain, but one must recall that payment itself is legally certain since its performance can be enforced in court and is therefore not up to the debtor’s whim.[107]

In addition, a clause that provides for the resolution of the contract if the debtor fails to perform cannot be considered a resolutory condition in civil law because, as the Mazeaud brothers explain, the condition cannot be an essential element to contract formation:

The event must be outside the legal relationship. The legal relationship must be able to exist independently of the condition, and the condition is a modality of it: thus, an essential element of the contract cannot be a condition of it. A sale subject to the condition that the price will be paid is not a conditional sale. It is a pure and simple sale. Payment of the price is an element of the sale – an intrinsic condition. By contrast, the sale of an immovable subject to the condition that the buyer marries is a conditional sale: the contract is conceivable without the condition, and the condition is merely a modality of it. Unfortunately, the cases still do not use rigorous terminology in this field.[108](Emphasis added.)

The distinction between resolutory conditions and resolution for non-performance may appear to be of little interest in this discussion, for the same retroactive effects are produced in both cases. But the importance of the distinction will become clear when we deal with ss. 79 and 79.1 I.T.A.

In conclusion on this subject, one should also take care not to confuse the resolution of a contract for non-performance on the one hand, with a clause that reserves ownership until complete payment has been made, on the other. Such a clause is actually an instalment sale, discussed hereinafter.

1.4.2    Instalment sales

An instalment sale is defined in the Civil Code as “a term sale by which the seller reserves ownership of the property until full payment of the sale price.”[109] The Minister of Justice wrote the following commentary on the provision:

[TRANSLATION] The first paragraph states the universally accepted definition of "instalment sale”: a sale in which the seller reserves ownership of the property until the price is paid in full. The provision [also] states that instalment sales are term sales, and the term relates to the transfer of ownership and the payment of the price. This was done to avoid this type of sale from being confused with a conditional sale.[110] (Empshasis added.)

This provision codifies the principles articulated by the Supreme Court in Venne v. Québec (Commission de la protection du territoire agricole)[111] where the Court addressed the question whether a sale, in which ownership is reserved until full payment of the price, was subject to a suspensive condition. If it had been, the transfer of ownership would have been retroactive to the date of the contract. The Court held that it was not a conditional contract. Rather, it was a term sale, and the transfer of was therefore not retroactive to the date the obligation was incurred. The Court cited with approval the judgment of McCarthy J.A. of the court below:

In my view, and with respect for the contrary opinion, there is no question of a conditional obligation here; accordingly, the retroactivity mentioned in art. 1085 C.C. does not apply. The “condition” referred to in arts. 1079 et seq. of the Civil Code is “an event future and uncertain” on which the existence of an obligation depends. The payment of the price by Venne does not fall in this category: Venne was obligated to pay the price, just as the Winzen company was obligated to convey the immoveable property, within a certain time. The obligations on either side were obligations with a term (arts. 1089 et seq. C.C.), not conditional obligations. They existed once the “Contract for Deed” had been signed, even though their performance was in abeyance. The same is true for the rights corresponding to the obligations.

In any synallagmatic contract performance of its obligations by one of the parties depends on performance by the other, but that does not make the obligations conditional within the meaning of the Civil Code.[112]

Thus, instalment sales create obligations with a term, which exist as soon as the contract is formed, but the performance of which is delayed until the future and uncertain event, which constitutes the term, occurs.[113] The term involves both the payment of the price and the transfer of ownership. Thus, from the very beginning, the parties make irrevocable commitments. One party agrees to pay the price, and the other agrees to transfer ownership of the property. But the transfer will only take place when the term occurs, i.e. at the moment the buyer has paid the full purchase price. Unlike sales subject to a suspensive condition, the transfer has no retroactive effect: the seller continues to own the property until the price is paid, and the transfer occurs at that time.

Obviously, if the buyer fails to pay, the seller is entitled to back the property.[114] The situation will then be similar to one in which a suspensive condition has not been fulfilled, and there is no retroactivity involved, since ownership was never legally transferred to the buyer at all.

Suspensive conditions and instalment sales both suspend the transfer of ownership until the occurrence of a certain event. But what sets them apart is that only suspensive conditions have a retroactive effect. A distinction should also be drawn between an instalment sale and the resolution of a sale for non-payment of the purchase price: in the former case, the sale is pure and simple, and ownership is transferred immediately. But the resolution of an instalment sale contract annuls it retroactively, just as the fulfilment of a resolutory condition would do.

Lastly, it should be emphasized that article 1746 C.C.Q. transfers the risks of loss to the buyer, even if he does not obtain any ownership right. The provision is consistent with the trend, discussed earlier, tying the risk to possession rather than to ownership:

It should be recalled that reservation of ownership is designed strictly as a mechanism to guarantee that the price is paid. In all other respects, the buyer is allowed to enjoy the property as would an owner. Thus, it would seem fair that he should assume any loss. It should however be noted that there has been a change in the basis for the rule: in instalment sales under the Civil Code of Québec, the risks are no longer tied to ownership; rather, they are tied to possession (in a broad sense).[115]

The provincial Minister of Justice explained the provision in similar terms:

[TRANSLATION] The provision places the risks of loss on the buyer except where the agreement states otherwise or where it is a consumer contract. This rule runs counter to the general principle, expressed in arts. 950 and 1456, and restated in section 133 of the Consumer Protection Act (R.S.Q., c. P-40.1). Since the buyer is in possession of the property and the need for protection is different from situations governed by the Consumer Protection Act, this special rule seemed better suited to the context and is in fact in keeping with the underlying principle of article 1456, para. 2, which, in another context, makes risks dependent on possession.[116](Emphasis added.)

1.4.3    Nullity when the conditions of contract formation are not met

Articles 1416 et seq. C.C.Q. provide that a contract is annulled if it does not meet the conditions of contract formation required by law, such as consent, cause and object. The nullity is absolute where the condition is one of public concern, in which case any interested person may request it and the courts must raise it on their own motion. On the other hand, if the condition protects individual interests, the nullity is relative: only the party in whose favour it has been stipulated may request that the agreement be annulled.

Whether the nullity be relative or absolute, its effects are the same:[117] Article 1422 C.C.Q. provides that the contract is deemed never to have existed and that the parties have an obligation of restitution, which is subject to the general rules of articles 1699-1707 C.C.Q. analyzed above.

Thus, the new Civil Code abolished the old distinction between absolute and relative nullity: a contract that was absolutely null was void ab initio, i.e. it had never existed and the court had no choice but to take notice of this, whereas only a judge could deem that a relatively null contract had never existed, and the nullity took effect only at such time as the judge made that determination, which meant that the contract could be effective so long as the injured party did not request the nullity.[118]

Today both kinds of nullity have the same effects as a resolutory condition.[119] They annihilate the contract retroactively, as though the contract had never existed. Thus, all of the comments made earlier about resolutory conditions apply here, except the specific reservations we made regarding article 1707 C.C.Q. This is because the provision applies fully, to protect third parties in good faith, who have no way, after all, of knowing why a contract is null.

1.4.4    Sales with a right of redemption

Articles 1750-1756 C.C.Q govern sales with a right of redemption. Such sales are defined as sales “under a resolutory condition by which the seller transfers ownership of property to the buyer while reserving the right to redeem it.”[120]

Although the condition is purely potestative because it depends solely on the seller’s will,[121] it is essentially a sale under a resolutory condition:

Although the legislator has used the words “rachat” and “racheter”, the arrangement constitutes a single contract under resolutory condition, not two contracts, the first of which is a purchase and the second of which is a repurchase. This distinction has important effects on third parties.[122]

Thus, the effects of resolutory conditions do apply. Before the right of redemption is exercised, the buyer is the true owner of the thing. He may collect the fruits from it, and he bears the risks of loss.[123]

The seller must comply with the provisions of the Civil Code regarding the exercise of his right of redemption.[124] He may take back the property, and since the sale is a true sale under resolutory condition, there is indeed a retroactive effect and he is deemed never to have given up ownership.[125] Thus, the seller takes back the property free of any charges that the buyer may have laid upon it, provided his right was published in accordance with the rules respecting the publication of rights.[126]

In sum, sales with a right of redemption are sales under a resolutory condition and, aside from the special rules that govern such sales, they have the same effects, notably in terms of retroactivity.

1.4.5    Trial sales

A trial sale is a sale subject to the condition that the buyer indicates his satisfaction with the property and decides to purchase it. Under article 1744 C.C.Q., the sale is presumed to be under a suspensive condition. Thus, the effects and matters discussed above apply to trial sales.

Trial sales are the most common example of a sale under suspensive condition with immediate transfer of possession. This is because the property must be delivered to the buyer to allow for trial of the thing sold.[127] Are the risks of loss therefore transferred to the buyer pendente conditione? For the reasons discussed earlier,[128] we believe they are, since the risk of loss is tied to possession of the property.

1.4.6    Promises of sale

Promises of sale that do not involve retroactivity issues nonetheless raise questions regarding on the moment ownership is transferred.

A promise of sale is a bilateral pre-contract by which both promisors agree to enter into a definitive contract of sale at a later date.[129]  The parties have an obligation to perform: their obligation is to sign a contract that corresponds to the provisions of the promise. If one party refuses to perform, the other may institute an action in passation of title.[130] Thus, a promise of sale is not an actual sale:

[TRANSLATION] The promise itself does not generate any effects of a sale. Among other things, this means that it does not transfer ownership of the property, and does not give the promisor-buyer any publishable real right.[131](Emphasis added.)

It should be emphasized, however, that this rule is merely suppletive. It applies only when it is impossible to determine whether the parties intended simply to pre-contract, or whether they intended to effect a sale and transfer ownership right away. Thus, one must interpret the contract to ascertain the true intent of the parties.[132]

By way of exception to the rule mentionned above, article 1710 C.C.Q. states that a promise of sale with delivery and possession is equivalent to a sale. Professor Jobin explains this rule as follows:

[TRANSLATION]

Under the Civil Code, a promise is a sale, and takes places immediately, when it is completed by the delivery of the property to the promising buyer and that buyer has actual possession thereof. The parties may sign the instrument of sale at some point in the future. [...]

The law presumes that the parties, under the circumstances, are agreeing that the sale takes full effect immediately, which means that it is indeed a sale, not a promise. Thus, the promisor-vendor, in delivering the property, and the promisor-buyer, in taking possession, have actually started to perform their obligations pursuant to the sale.[133]

Thus, in promises of sale where the buyer takes possession of the property, the law presumes that the parties intended to pass ownership immediately. Naturally, this is a simple presumption; it will only be made if the parties have not stipulated otherwise.[134]

Essentially then, a promise of sale will not generally transfer ownership, unless otherwise provied in the contract. But the law will presume a transfer takes place where possession of the thing whose sale is promised is transferred to the promising buyer. In such cases, ownership is deemed to have passed unless the parties provide otherwise.

1.4.7    Retroactivity by contract

May the parties stipulate, in their contract of sale, that ownership was transferred prior to the signing of the contract?

In both the Civil Code of Lower Canada and the Civil Code of Québec, the Quebec legislature adopted a consensualist approach, thereby following in the footsteps of modern French law.[135] This approach starts from the principle that the contract is perfected, and ownership passes to the buyer, as soon as the parties’ minds have met.[136] Baudouin and Jobin write:

[TRANSLATION] When the two minds have met on the essential elements, the contract is born, except, of course if the law makes its creation dependent on some additional act, such as where certain formalities must be completed. The mere fact that the parties agree to subsequently memorialize their agreement in writing does not postpone the formation of the contract to the date of that writing. The situation is different if the parties actually intended that the contract only come into being at the time of signing.[137](Emphasis added.)

A distinction must be made between the contract itself and the writing that memorializes it: the contract is validly formed from the moment the parties have agreed on its essential elements, such as the object and the price. Later on, when the contract is subsequently drafted, the parties are entitled to state that the contract was effective from the time it was truly formed.[138] However, as both these scholars have pointed out, the parties may only have wanted their contract to come into being, and ownership to be transferred, on the date their written contract was signed. This will bring the parties’ relevant intention into issue.

In any event, it is doubtful that the parties could stipulate an effective date that is earlier than the date on which the parties had their meeting of the minds. Ownership cannot, after all, be transferred prior to that time, since the agreement between the parties is the legal cause, and an effect cannot come prior to a cause. The parties cannot make their contract retroactive to a date that it did not exist – at least not in any way that would affect someone outside their contract.

1.4.8    Retroactivity under the Civil Code

There are many other instances of retroactivity under the Civil Code,[139] only a few of which will be discussed here; later on we will compare how they are treated in tax law as compared to conditional obligations.

Under the old matrimonial regime of community of property, the spouses co-owned the community property.[140] The Supreme Court held that where the community property passes to the surviving spouse under the marriage contract, ownership in all the community property devolves to the latter upon the death of the spouse retroactively to the date of marriage.[141] This is a declaratory, rather than a translative, effect of the dissolution of the community property regime.[142]

The declaratory effect of the succession’s partition is provided for in elsewhere in the Civil Code:[143] the liquidator has possession until the property is partitioned, after which each heir is deemed to have been the owner of the property included in his share from the date of death of the de cujus.

These examples demonstrate that it is not exceptional for events under the Civil Code to have a retroactive effect. However, contracts are not the only source of retroactivity, which may also arise solely out of the Act. We will see further on how the Income Tax Act applies to this retroactivity.


2. Conditions under the Common Law

2.1 Preliminary concepts

Before we consider conditions at common law, it is essential to review certain of this system’s basic rules pertaining to property.

First of all, it must be understood that common law ownership, unlike civil law ownership, is regarded as a bundle of divisible rights:

[TRANSLATION] The common law definition of property is different from the Civil Code definition. But that is not all. The entire approach to developing rules governing ownership, and even the examination of those rules, is different. For example, the common law does not even speak of “ownership” when real property is involved. The feudal system never recognized that an individual could own land outright, and many basic rules of the common law system are vestiges of that period, so there is no theory of real property “ownership”. Rather, the common law has devoted its attention to considering the rules that govern the range of rights, privileges and powers included within ownership. Thus, the common law conceives of ownership as a bundle of divisible rights. At least in theory, no one can own a parcel of land; rather, one “holds the land of the Crown” i.e. one owns an “interest” in the property, not the property itself.[144](Emphasis added.)

These rights, privileges and powers of ownership can be divided and distributed to several different people:

Ownership consists of innumerable rights over property, for example the rights of exclusive enjoyment, of destruction, alteration and alienation, and of maintaining and recovering possession of the property from all other persons. Those rights are conceived not as separately existing, but as merged in one general right of ownership....

Ownership is nevertheless divisible to some extent. For example, one or more of the collection of rights constituting ownership may be detached. Thus prima facie an owner is entitled to possession or to recover possession of his goods against the entire world, a right that a dispossessed owner may exercise by peaceable retaking. He may, however, voluntarily or involuntarily part with possession, for example by the pledging, lending, hiring out, bailment, theft or loss of his goods, in any of which cases he is left with a right of ownership without possession, accompanied or not accompanied, as the case may be, by the right to possess.[145]

In addition, the rights that make up ownership can be split between a legal owner and a beneficial, or equitable, owner. The legal owner holds title to the property “at law” whereas the beneficial owner has an interest in the property under the rules of Equity:

The separation of the enjoyment of property and its administration, though not unique to the common law, is solved by the fragmentation of title into a legal and equitable title. ... Both the holder of the legal title and that of the equitable title are regarded as owners of the land.[146](Emphasis added.)

Black’s Law Dictionary defines the term beneficial owner as follows:

Beneficial owner. Term applied most commonly to cestui que trust that enjoys ownership of the trust or estate in equity, but not legal title, which remains in trustee or personal representative. Equitable as contrasted with legal owner.

One who does not have title to property but has rights in the property which are the normal incident of owning the property. The persons for whom a trustee holds title to property are beneficial owners of the property, and the trustee has a fiduciary responsibility to them. (Emphasis added.) [147]

The split between beneficial and legal title lies at the basis of the common law trust. The trust is the easiest context in which to understand the difference between beneficial and legal owners:

The separation of ownership is critical to the concept of a trust. Once the settlor transfers the property to the trustees, the settlor has divested him – or herself of the ownership of the property. The trustees become the legal owners of the property while other persons, the beneficiaries, have the equitable or beneficial ownership (that is, the right to use and enjoy the property). The trustees hold title and manage the property for the benefit of the beneficiaries and are not entitled to enjoy or use the property.[148](Emphasis added.)

Thus, the common law’s conception of ownership is extremely different from the Civil Code’s, which does not recognize the distinction between legal and beneficial ownershipAs Rinfret J. has noted, civil law ownership is indivisible:

[TRANSLATION] The legal system of the Province of Quebec does not know of the common law concept according to which one person has beneficial ownership while another has legal ownership. In Quebec, both are always held by one person. Ownership is indivisible. The usufruct, the substitution, the trust, the pledge, the hypothec, the privilege grant on a thing various rights … but never transfer ownership[149].

This explains why the trust, as understood at common law, does not exist in civil law. After all, it is based on the division of ownership between the trustee, who holds legal title, and the beneficiary, who holds equitable title:

The division of legal and beneficial ownership is foreign to civil law and irreconcilable with the fundamental principle of unity of title. In civil law jurisdictions permitting the creation of trusts (such as Quebec), trusts often consist in the segregation of property into a separate patrimony for the carrying out of a particular purpose. In other civil law countries, trusts are either entirely ignored or assimilated to agency relationships and governed by the rules applicable to such relationships.[150]

Since the common law and civil law conceptions of ownership are so different, it is difficult to apply the Income Tax Act uniformly while taking the specificities of provincial law into account. As Addy J. so wisely put it:

The law of real property is one of the areas where common law and civil law principle’s are most likely to be at variance or at least to flow from different fundamental premises. At common law, the nature of the relationship existing between a vendor and purchaser of real estate under given circumstances is governed to a large extent by the distinctions between legal and equitable ownerships, estates and remedies and by the principles applicable to various categories of trusts and trustees. None of these concepts even exists in civil law. To seek by way of common law jurisprudence to reach a solution to the present issue would be to venture out on a perilous journey over rocky and tortuous roads, fraught with pitfalls, which would lead to a mere cul-de-sac, if one were fortunate. [151](Emphasis added.)

2.2 Conditions precedent

The condition precedent is the common law equivalent to the civil law suspensive condition. Rather than explore all the intricacies of this concept, our intention is to discuss the ways in which the two conditions resemble each other and the ways in which they differ. 

We must first define the term condition. It appears to have the same meaning as it does in civil law: 

Condition. A future and uncertain event upon the happening of which is made to depend the existence of an obligation, or that which subordinates the existence of liability under a contract to a certain future event. Provision making effect of legal instrument contingent upon an uncertain event.[152]

A condition precedent is defined as follows:

A “condition precedent” is one that is to be performed before the agreement becomes effective, and which calls for the happening of some event or the performance of some act after the terms of the contract have been arrested on, before the contract shall be binding on the parties.[153]

Thus, the effect of a condition precedent is to delay the creation an obligation until it is fulfilled. As long as the condition has not occurred the obligation has no legal existence. A contract that is subject to the fulfilment of a condition precedent does not become a binding agreement until such time as the condition has been met or waived ....[154]

One must, however, note the distinction between true conditions precedent, which suspend the creation of the obligation, from conditions that are inherent to the contract and merely suspend the performance of the obligation. As Professor Fridman explains, this distinction was first made in Turney v. Zhilka.[155]

A radical change in the approach to conditions precedent was effected by the Supreme Court of Canada in Turney v. Zhilka. The court differentiated what was called “a true condition precedent – an external condition upon which the existence of the obligation depends” from an ordinary or internal condition ...If a condition is a true condition precedent, there is no contract until it is satisfied. If a condition is the other sort of condition, then, in the event of its non-fulfilment, there may still be a binding contract between the parties, depending on the way in which the innocent party, guiltless of any breach, reacts to a breach of the condition. It follows from Turney v. Zhilka, therefore, that a distinction now exists between a condition relating to the existence of any contractual obligation and a condition that is precedent to performance of a contractual obligation by the other party, not the one subject to fulfilment of the condition precedent. [156](Emphasis added.)

It appears that true conditions precedentare similar to civil law suspensive conditions, at least pendente conditione, since they suspend the creation of the obligation. Likewise conditions that are not true conditions precedent are more like obligations with a term as they merely suspend the performance of the obligation.

Moreover, when a condition precedent is fulfilled, the obligation arises immediately but without retroactive effect. This stands in contrast with suspensive conditions[157] and it is the fundamental difference between conditions precedent and civil law suspensive conditions. Thus, the transfer of ownership is not retroactive: the buyer becomes the owner only when the condition occurs and is not deemed or presumed to have acquired it any earlier.

But what happens if the condition fails?  The contract is considered never to have existed. And since a non-existent contract cannot give rise to any obligation at all, beneficial ownership was never transferred:

In the event the condition is not met or waived, then the agreement is void ab initio; it has never come into existence. Beneficial ownership of the subject matter of the contract cannot pass until the condition precedent has been satisfied or waived.[158](Emphasis added.)

Indeed, unlike a voidable contract (i.e. one that can be annulled at the request of the injured party) a contract that is void at common law can have no legal effect. A voidable contract is valid until set aside, however, and it has certain effects, including the transfer of beneficial ownership.[159]

Thus the effects of a condition precedent are similar to those of the civil law suspensive condition, both pendente conditione and where the condition fails. However, when the condition occurs, conditions precedent, have no retroactive effect, unlike suspensive conditions. We will discuss the tax impact of these differences later on. 

2.3 Conditions subsequent

A condition subsequent is very similar to a civil law resolutory condition:

A condition subsequent is one annexed to an estate already vested, by the performance of which such estate is kept and continued, and by the failure or non-performance of which it is defeated; or it is a condition referring to a future event, upon the happening of which the obligation becomes no longer binding upon the other party, if he chooses to avail himself of the condition.[160]

Just like civil law resolutory conditions, conditions subsequent do not suspend the creation or coming into force of the obligation. In a contract for the sale of goods, ownership will pass, immediately despite the existence of a condition subsequent. The condition extinguishes the obligation when the uncertain event occurs:

A condition subsequent is an agreement between the parties that the contract is immediately binding, but that if certain facts are ascertained to exist or upon the happening of a certain event, either the contract ceases to bind or one party is to have the option of cancelling the contract.[161]

Despite its resemblance to a resolutory condition, one will note an important difference: the parties cannot waive a resolutory condition and treat the contract as valid notwithstanding that it the condition occurred. They may enter into a new contract but it will only come into force when that contact is formed. At common law, however, a condition subsequent may give the parties an option to rescind.

In fact, the essential difference between a resolutory condition and a condition subsequent is that the latter has no retroactive effect: it sets aside the contract in terms of the future, but has no impact on the past effect of the contract. Thus, beneficial ownership is transferred twice – once when the contract is formed, and again when the condition is fulfilled:

In instances where a contract is subject to a condition subsequent, beneficial ownership passes from one contracting party to the other subject to the revesting of the property in the original owner upon the happening of certain prescribed events. If those events do not occur, the contract remains in force. If they do occur, there is a second transfer of beneficial ownership to the original owner.[162]

In short, before the condition occurs (or if it never does) a condition subsequent essentially has the same effects as a resolutory condition. However, when a condition subsequent occurs, it has no retroactive effect, unlike a resolutory condition.

2.4 Other Common Law concepts that raise questions of retroactivity

2.4.1    Retroactivity provided for by contract

Can the parties to a contract validly stipulate that their agreement came into force at a date that precedes the actual signing date? The answer in common law jurisdictions is similar to the answer in Quebec. The earlier date will be acceptable if it reflects the time at which the parties came to a definitive agreement, whether oral, or memorialized in a written letter of intent, on all the essential elements of the contract.[163] The agreed-upon date cannot, however, be earlier than the moment at which the agreement between the parties became “binding and legally enforceable.” If the parties were continuing to negotiate any essential elements of the contract (such as the sale price) on the date they indicated, if they did not intend to be bound before the   contract was duly signed, or if the initial agreement was subject to a condition precedent, the stipulated coming into force date will be binding as between the parties, but not on third parties.

The real question in this area, then, is not so much whether the coming into force date is valid as between the parties, but especially whether it is binding on the tax authorities, i.e. whether it is considered the transaction date for tax purposes. We will consider this question in greater detail when we turn to the applicable tax law.

2.4.2    Retroactivity under provincial law

Certain provincial statutes allow courts may make retroactive orders. Most common law jurisdictions in Canada have statutes that ensure dependents a fair share of a decedent’s estate. These statutes generally contain such allowances. The Saskatchewan Dependant’s Relief Act[164] for example provides that a person who was a dependent of the deceased, notably the spouse, may apply to the court to obtain a greater share of the estate than the amount provided her under the will or by the rules of intestacy. The judge’s order on such an application is generally retroactive under the legislation. Thus, the property ordered granted to the applicant are deemed to have devolved by will or by law, on the date the de cujus died. 

This retroactivity may have major tax implications, especially in relation to the question of when the property vests indefeasibly in the heirs or legatees. We will soon see whether federal tax law gives effect to the provincially recognized retroactivity of such orders.

3. The retroactive effect of conditional obligations in Tax Law

3.1 The provisions of the Income Tax Act

We have seen that the retroactive effects of civil law conditional obligations extend to a number of matters, including the moment ownership is transferred pursuant to a sale. The I.T.A. imposes certain tax consequences on such transactions, but does not make these consequences flow directly from the transfers. Rather, it ties them to the concept of “disposition.” Consequently, it is essential to analyze this concept.

The Act provides that taxable capital gains for the year from the disposition of property[165] must be included in the computation of income for tax purposes. Disposition is a central concept in this computation, because capital gains are taxable in the tax year of the disposition. 

The Act also provides that recapture of capital cost allowance must be included in a taxpayer’s income for the tax year in which the last item of property in a class is disposed of.[166] Likewise, in order to claim  capital cost allowance on property in a class, the property must have been “acquired” prior to the end of the year.[167]  As we shall see, the concept of acquisition is the mirror of disposition: when a taxpayer disposes of property, someone else acquires it at the same time. This is why the concept of disposition is very important, both for a vendor and a purchaser, where a CCA is being claimed.

The I.T.A. defines the term “disposition.” The new definition, which came into force in December 23 1998, is contained in subsection 248(1) which provides inter alia as follows:

248(1) “disposition” ¾ “disposition” of any property, except as expressly otherwise provided, includes

(a) any transaction or event entitling a taxpayer to proceeds of disposition of the property...

but does not include

(e) any tranfer of the property as a consequence which there is no change in the beneficial ownership of the property, except where the transfer is... (Emphasis added .)

This definition is almost identical to the definition in the old section 54, which read as follows:

“disposition” of any property, except as expressly otherwise provided, includes

(a) any transaction or event entitling a taxpayer to proceeds of disposition of property…

but, for greater certainty, does not include

(e) any transfer of property by virtue of which there is a change in the legal ownership of the property without any change in the beneficial ownership thereof … (Emphasis added.)

This definition used to apply to Division C  (“Taxable capital gains and deductible capital losses”), which contained sections 38 to 55 I.T.A. A different definition of the term “disposition” applied to capital cost allowance:

13.(21) The following definitions apply to this section “disposition of property” includes any transaction or event entitling a taxpayer to proceeds of disposition of property

This definition, in relevant part, is the same as the one contained in section 54;[168] the concept of “disposition” is therefore the same, whether it is used for taxable capital gains or capital cost allowance recapture.

It should also be noted that subsection 248(1), section 54 and subsection 13(21) all refer to “proceeds of disposition” which are defined in section 54:

54. In this subdivision,

“proceeds of disposition” of property includes,

(a) the sale price of a property that has been sold,

Although subsection 13(21) does not contain any definition of “proceeds of disposition,” the case law has always held that the definition in section 54 applies where CCA is involved.[169]

Upon analysis of these definitions, one may realize that “disposition” includes, but is not limited to, any event entitling a person to the proceeds of disposition of property, i.e. the sale price of a thing that has been sold, except if “beneficial ownership” of the property was not transferred by that event. This will puzzle civil lawyers, particularly those who read the French version of the Act, where they will encounter the phrase “propriété effective.”  What does that mean, and in any case, what is “beneficial ownership?”  There is no equivalent in civil law.[170]

A paradoxical conclusion stems from this definition at first blush. First of all, by stating a specific definition of the concept of disposition for the purposes of the I.T.A., Parliament seems to have intended to dissociate the I.T.A. from provincial private law and create a special tax rule. Secondly, within that definition, Parliament uses a private law concept without defining it, which brings us back to the principle that the relevant private law must be consulted, and that is the civil law in Quebec and the common law elsewhere in Canada. This process works very well with the common law provinces, whose law defines “beneficial ownership” and thereby completes the I.T.A. definition. But matters are more complicated in Quebec. Must Quebec courts and lawyers actually apply the common law concept of “beneficial ownership”, or should they try to define it by analogy or equivalence? Moreover, the fact that the new definition in subsection 248(1) no longer refers to “legal ownership” (unlike old section 54) does not appear to have changed the substance of the provision, since the phrase “without any change in the beneficial ownership” could only be a reference to a transfer of legal ownership. In fact, since the definition continues to refer to “beneficial ownership”, the same problem of interpretation persists where civil law is involved.

Parliament did try to define “beneficial ownership” for the purposes of applying it to  Quebec in subsection 248(3) I.T.A. The provision formerly read:

248. (3) References to property beneficially owned and to beneficial owner of property.¾ In its application in relation to the Province of Quebec, a reference in this Act to any property that is or was beneficially owned by any person shall be read as including a reference to property in relation to which any person has or had the full ownership ,whether or not the property is or was subject to a servitude, or has or had a right as a usufructuaary, a lessee in an emphyteutic lease, an institute in a substitution or a beneficiary in a trust, and a reference in this Act to the beneficial owner of any property shall be read as including a refernce to a person who has or had accordingly as the context requires, such ownership as a right in relation to that property. (Emphasis added.)

Since 1991, the provision has been replaced by the current paragraph 248(3)(f):

248. (3) Rules applicable in relation to the Province of Quebec. For the purposes of the application of this Act in relation to the Province of Quebec,…

(f) property in relation to which any person has, at any time,

  •  (i) the right of ownership,
  •  (ii) a right as a lessee in an emphyteutic lease, or
  •  (iii) a right as a beneficiary in a trust

shall notwithstanding that such property is subject to a servitude, be deemed to be beneficially owned by the person at that time. (Emphasis added.)

This amendment does not appear to have made new law: new paragraph 248(3)(f) is simply worded more clearly and provides that beneficial ownership includes certain specified rights. Usufructuaries and institutes of substitution are no longer included in the new definition because they are now considered deemed trusts under paragraphs 248(3) (a) through (e). However, both versions clearly state that civil law owners have beneficial ownership.

Let us now consider how the courts and scholars have interpreted these definitions, before moving on to the CCRA’s administration position.

3.2 The cases and scholarly writing

3.2.1    The complementarity of provincial private law

Before we consider the case law on the definition of disposition and the retroactive nature of conditional obligations, it is important to set out the basic principles that underlie the complementarity of private law.

As we saw in the Introduction, tax law is accessory to private law, and follows private law in establishing the tax consequences of private law transactions, which transactions are, of course, governed by provincial law.[171]

In this regard, apart from the now famous excerpt from Lagueux & Frères, cited in the introduction, one of the best-known quotes is from the judgment of Boisvert J. in Perron v. M.N.R.:

If income tax is a creation of the Act which imposes it, that Act must apply within the framework of the civil laws governing legal relationships between individuals. The tax is grafted, as it were, on the legal tree which covers with its shadow the rights and obligations arising from the contracts.[172]

The Supreme Court confirmed this principle in R. v. Dominion Engineering Co.[173] That case dealt with an instalment sale, and the Act, at the time, stated that where monthly payments are due on a sale price, the tax on each payments was due at such times as “each of such instalments falls due and becomes payable”. The purchaser declared bankruptcy before paying the price, but the Department of Revenue[174] claimed that the vendor should pay taxes on instalments prior to bankruptcy, even if they had not been paid to him, because they have become due and payable.

The Court held for the taxpayer, and Rand J. held as follows:

Although the section declares the “transaction” to be a constructive sale and delivery, the fundamental support of the tax is an executory contract leading to the transfer of title and possession. That contract is conceived as a potential sale to which in turn is related a potential total tax: “the tax shall be payable”. Pro tanto portions of the tax are related to instalments of price and, when the latter become payable as parts of a whole, the right to the tax takes on the same character : but throughout, the tax depends for its efficacy upon the maturing contract. For the total tax there is only an inchoate liability created by the making of the agreement : and to sustain the right to the tax, the instalment become payable must remain an obligation of an executory contract.

The legal liability at any time for any portion of the tax in no degree restricts the parties in good faith from modifying the contract as they see fit, and a fortiori it does not prevent a modification by operation of law. If, in the legal result, the actual transaction ceases to be one of sale, then the necessary support for the tax disappears.[175] (Emphasis added.)

Thus, the Supreme Court recognized that the “necessary support” for taxation is the existence of a valid contractual relationship under applicable private law. The Court recently reiterated that tax law must be founded on the true legal relationships established by the taxpayers.[176]  Yet these legal relationships can only be established under private law.

Hence, when the Act uses a private law term without defining it, one must resort to the private law of the province involved in order to interpret it.[177] However, this principle, which holds that provincial private law is complementary, is sometimes cast aside based on the principle that federal laws should be applied uniformly. As Brisson and Morel explain:

In opposition to the commonly held view that the complementarity of provincial private law legislation is accepted failing any provision to the contrary, it is sometimes suggested that federal legislation should be applied in the same way everywhere, in the interests of uniformity. ... For the same reason it has sometimes been considered appropriate to interpret the Income Tax Act as overriding the civil law, using a common law rationale, to avoid giving the Act a broader scope within Quebec than it would have in some other province.[178] (Emphasis added.)

Having explained this point of view, the authors go on to criticize it:

According to this view, the usual logic applied to the distribution of powers in private law is to some degree reversed. The private law of the provinces is no longer regarded as the fundamental law of federal legislation, since the latter is self-sufficient, generating its own “common law.”  Whether in the name of the uniform application of federal statutes and their source of inspiration (often the common law) or because a particular statute  constitutes a “complete code”, the result is always the same: the federal privte law legislation need no longer be shaped to fit the particular features of the law of any particular province. On the other hand, it has rightly been observed, with regard to the private law, “if all aspects of the law should be exactly the same across the country why have a federal system?"  .

However, the influence of this opinion, which advocates the autonomy of federal statutes, should not be exaggerated. For every decision that argues in favour of the dissociation of federal law and civil law, there are many others affirming their principled complementarity, excluding thereby the application of the common law.[179](Emphasis added.)

Brisson and Morel conclude that the complementarity principle should be applied despite the resulting disparities in applying federal legislation:

If so, one can accept the direct and substantial complementarity of federal legislation with the civil law, since it is consistent with the federal scheme of Canadian federalism in matters of private law. Although, it should be noted that a similar approach must, in the same spirit, be followed when a federal private law statute is applied in a common law province. This means, of course, that incomplete federal legislation will not be applied uniformly throughout the country. If this is unsatisfactory, it is sufficient, albeit necessary, that the federal statute supply its own definition, as is so often done in the field of taxation. Otherwise, the complementarity of legal systems, and the ensuing diversity, will continue to prevail.[180](Emphasis added.)

In a recent decision, Canada (Attorney General) v. St-Hilaire[181], the Federal Court of Appeal clearly applied the complementarity principle to the relationship between civil law and federal law. The facts of the case are highly unusual. After stabbing her husband to death, the respondent claimed surviving spouse benefits under the Public Service Pension Act.[182] Naturally, the Attorney General opposed this. Since the statute in question had no provision on the subject, the Attorney General relied on the common law principle that no one may profit from his or her crime. The respondent submitted that one had to turn to the Civil Code of Québec in the case at bar in order to complete matters to which the federal statute did not speak. However, she claimed that she was not caught by article 620 C.C.Q., since in her submission, that article simply provides that persons who are convicted of murder are not fit to inherit, and she had been convicted of manslaughter.

Décary J.A. gave a comprehensive analysis of the complementarity of civil law in order to determine which rules the federal law should apply to the statute in question. Citing an article by Professor Brisson,[183] he concluded that there was no federal common law stricto sensu: unless the federal statute itself states otherwise, the private law of the provinces simply must be made applicable by default:

[Translation] In Quebec, federal private law consists of private law as set forth in an Act of the Parliament of Canada, and civil law if it is necessary to resort to an external source to apply a federal statute. The Parliament of Canada may enact private laws intended as complete codes, in which case there will be no need to refer to the civil law, since the civil law is an external source. Parliament may also enact private laws that expressly or impliedly call upon civil law for their application.[184](Emphasis added.)

On the uniform application of federal laws, Décary J.A. commented :

[Translation]The very Constitution of Canada itself provides that federal statutes may have different effects depending on whether they are being applied in Quebec or in the other provinces.  By ensuring the continuity of civil law in Quebec, and by s. 94, which encourages efforts to uniformize property and civil law statutes in the other provinces, the Constitution Act, 1867 enshines the principle of federalism, which holds that a federal statute that refers to an external source of private law will not necessarily be applied uniformly throughout thecountry. To associate federal legislation with the common law on a systematic basis would be to disregard the Constitution.[185](Emphasis added.)

In response to the Attorney General’s argument that the statute in issue was one of public law and that the common law should complete it, Décary J.A held:

[Translation] The relevant part of the Act simply designates the beneficiary of the plan of which a public servant was a member. In my view, the Act does not appear to differ in naturefrom the Public Service Staff Relations Act (R.S.C. 1985, c. P-35). And yet, in Ménard v. Canada, this Court applied the civil law theory of unjust enrichment, instead of common law estoppel, in ordering Her Majesty to pay overtime to an employee of the Correctional Service of Canada.  And this Court routinely applies civil law in Quebec cases involving the Income Tax Act, which is said to be a public law statute.[186] (Emphasis added.)

Décary J.A. then cited the following article by Prof. Morel in this regard:

First, there are a number of situations in which the civil law is required to assume what might be called a passive role.  Such situations include every instance where, in furtherance of its own purposes, a federal statute assigns certain effects to juridical acts or facts governed by the Civil Code.  Examples abound. One need only think of legislation concerning bankruptcy, bills of exchange or bank security, which in order to have effect, depends on the existence of contracts such as loans, sales, and movable or immovable hypothecs.  Divorce and the extracontractual liability of the Crown are equally good examples.  The Income Tax Act, which determines the tax consequences of sales, assignments of claims, gifts, or legacies, illustrates how certain public law statutes also require that recourse be had to the Civil Code to identify the precise nature of the juridical act in question.  This is an example of how the Civil Code governs a private law relationship that comes into indirect contact with federal law, which in turn intervenes to determine the consequences of such relationships as far as the federal legal order is concerned.[187](Emphasis added.)

Décary J.A. concluded as follows:

[Transaltion] In my opinion, the question whether or not resort should be had to private law (civil law in Quebec) should not depend on whether the federal statute in question is public or private in nature. Rather, the question is whether the federal law, in a given matter, is being applied to situations involving relationships that it has not defined, and that can only be defined in relation to the persons affected. In a sense, then, we have come full circle and are back to Section VIII of the Quebec Act: where the people affected are litigants, their civil rights are in issue and Parliament has not defined those rights, provincial private law will fill the vacuum. In short, civil law applies, in Quebec, to any federal legislation that does not exclude it.[188] (Emphasis added.)

Since the Public Service Pension Act refers to “succession”, a private law concept, without defining it, Décary J.A. held that the statute must be interpreted in accordance with civil law. He then considered the civil law on point and found that the respondent could not be considered unfit to inherit under art. 620 C.C.Q. The other two Justices concurred entirely in the opinion of Décary J.A. as to the application of civil law, but differed in their interpretation of the Civil Code and held that the respondent was unfit to inherit.

We are dwelling on the complementarity principle because it is the premise of our legal reasoning. Basing ourselves on the wording of the Act, we will attempt hereafter to determine whether it has established a definition of the concept of “disposition” that is exclusive to tax law matters, dissociated from provincial private law. If not, we believe civil law rules must apply, since they are complementary. If the rules cause negative tax consequences, it is up to Parliament to amend the Act to remedy the problems.

As mentioned, although Parliament has defined the word “disposition”, the definition itself contains a reference to “beneficial ownership”, a private law term that is understood by only one of Canada’s legal systems. Moreover, as we shall see below, the definition of “disposition” is not exhaustive, and one must therefore refer to provincial private law to complete it.

3.2.2    Disposition as a concept

3.2.2.1    In common law provinces

The first time the concept of disposition was judicially interpreted in the context of the Income Tax Act was in Victory Hotels Ltd. v. M.N.R..[189]

In December 1954, the parties signed an agreement to sell a hotel owned by the appellant. The agreement stipulated that the buyer would take possession on January 3, 1955, and that the sale would be cancelled if the buyer was unable obtain a liquor license or if the hotel was destroyed by fire prior to the taking of possession. The buyer made a deposit of $25,000 on the purchase price, which was to be held in trust by the real estate agent until the conditions were satisfied. If the sale was cancelled because one or more of the conditions was not fulfilled, the deposit was to be returned to the buyer.

Noël J.A. held that the “disposition” occurred on January 3, 1955. In His Lordship’s opinion, the parties had clearly intended the sale to take effect on January 3, 1955. In addition, the seller reserved possession, use and control of the hotel up to January 3 and kept the revenues generated by the business up to that date. Furthermore, the interest on the mortgage was calculated from that date.

The judge first established that the term “disposed of” in section 20 I.T.A. must be given a broad interpretation:

Indeed, in the context of s. 20 of the Income Tax Act it is not unreasonable to give the words “disposed of” their widest meaning which would be “to part with “,”to pass over the control of the thing to someone else” so that the person disposing no longer has the use of the property.[190]

However, Noël J.A. noted that Parliament might have narrowed the broad interpretation of “disposition.” In his opinion, this is the case where there is a sale of property:

We have seen that s. 20(5)(b) of the Income Tax Act states that “disposition of property” includes any transaction or event entitling a taxpayer to “proceeds of disposition of property” and 20(5)(a) states that “proceeds of disposition” of property include (i) “the sale price of property that has been sold.” These sections do not define but merely include as a disposition of property a transaction (a sale for instance) entitling a taxpayer to proceeds of disposition of property, i.e. to the sale price of the property sold. It would indeed appear that the meaning of “disposition of property” has been somewhat restricted by the Act when a disposal of property takes place by means of a sale; in such a case there is a disposal of property as soon as a taxpayer is entitled to the sale price of the property sold.[191](Emphasis added.)

Thus, Noël J.A. held that the relevant sections do not define the word “disposition”,[192] but merely include within the term such transactions as entitle a person to proceeds of disposition. Nonetheless, Noël J.A. concluded that the meaning of “disposition” was narrowed in relation to cases where the disposition resulted from a sale. The disposition, in this instance, would occur as soon as the seller was entitled to the sale price. In the case at bar, the seller was not entitled to the sale price until the conditions were fulfilled and the buyer had taken possession. This occurred on January 3rd, 1955.

The Exchequer Court actually established the “test” for disposition in MNR v. Wardean Drilling Ltd.,[193] however. The decision, which was followed by the subsequent case law, requires closer analysis.

Wardean Drilling, an oil-drilling company, wanted to buy a drill. According to the sale contract signed in December 1963, the drill was to be delivered in February 1964 because specific modifications had to be made. The contract also provided that title would pass upon delivery. In addition, Wardean Drilling purchased ancillary equipment from another supplier. That contract was signed in December 1963, but the equipment was only built and delivered in 1964. The point in issue in dispute was whether Wardean Drilling could claim a deduction for depreciation of the drill and the ancillary equipment for taxation year 1963, i.e. whether it had “acquired” them during that year.

Cattanach J. of the Exchequer Court held that the acquisition had only occurred in 1964. He began by explaining that the decisive point was not the moment at which the parties signed an enforceable sales contract, but rather the following:

As I have indicated above, it is my opinion that a buyer has acquired assets of a class in Schedule B when title has passed, assuming that the assets exist at that time, or when the buyer has all the incidents of title, such as possession, use and risk, although legal title may remain in the seller as security for the purchase price as is the commercial practice under conditional sales agreements.[194]

Thus, property is “acquired” when ownership is transferred, if the property exists at that moment, or when the normal incidents of title,[195] such as possession, use and risk, are transferred if ownership is reserved by the seller as security for the sale price.

After establishing this test, the Court went on to examine the private law in Alberta, in order to determine the moment when ownership is transferred. Under the Alberta Sale of Goods Act [196] however, the terms of the contract determine when ownership is transferred. If the parties fail to do so, the Act provides a set of rules.

Having reviewed the drill sale contract, Cattanach J. concluded that the parties intended title to pass on delivery. Since delivery took place in 1964, the property was “acquired” at that moment. The contract for the ancillary equipment was silent on this issue and the judge referred to Alberta law, which provided that ownership was transferred when the property was finished and ready for delivery: 1964, in this instance.

The ratio decidendi of this decision was that under provincial law, ownership was transferred in 1964. The Court’s decision was not based on the moment when the normal incidents of title (such as use, possession and risk) were transferred because this was not a case in which the seller had retained title as security. The second part of the “test”, therefore, was an obiter dictum.

In spite of appearances, Wardean Drilling supports the theory that provincial private law is complementary. The judge examined the moment at which ownership was transferred under the provincial private law involved. As we saw, at common law ownership may be held simultaneously by a legal owner and a beneficial owner.[197] We agree with Noël J.A.’s dissenting opinion in Construction Bérou Inc. v. The Queen:[198]

In setting forth that rule Cattanach J. did not have it in mind to overturn applicable private law, since the judgment he rendered was to the exact opposite effect. As Chief Judge Couture explained in the case at bar, Cattanach J. in stating that rule:

is merely confirming the distinction [at common law] between the owner  who holds legal title and the beneficial owner of the property, that is, the one to whom ownership belongs subsequent to a transaction, but who will receive title to the property at a later date.[199]

In the footnote, Noël J.A. attempts to find an explanation for the second part of the test:

See Black’s Law Dictionary, which defines a “beneficial owner” as being inter alia: "One who does not have title to property but has rights in the property which are the normal incident’s of owning the property”. It may safely be assumed that the second part of the rule of acquisition stated by Cattanach J. derives from the Supreme Court of Alberta judgment in Hendrickson v. Mid-City Motors, [1951] 3 D.L.R. 276, in which it was held that a conditional sale agreement gave rise to a sale under Rule I of s. 21 of the Alberta Sale of Goods Act, despite the fact that ownership title remained with the seller. At that time the Court said the following, at 284:

I conceive ‘title’ and ‘property’ to be two entirely different things. One person may hold bare title to property while the whole beneficial ownership rests in some other person. A reservation of title does not necessarily imply that no property shall pass to the buyer ...

...In my opinion, the whole effect of the agreement ...is to transfer to the buyer the ‘property’ in the goods in question, while reserving to the seller a seller’s lien and the right to defer the conveyance of legal ‘title’ to the property until payment in full.

The Supreme Court of Alberta came to this conclusion in accordance with the intent of the parties as disclosed by the wording of the contract. See also Douglas S. Ewens, "When is a ‘disposition’", Report of Proceedings of the Twenty-Sixth Tax Conference, November 11-13, 1974, at p. 538. As regards application of the rule that a change in the beneficial ownership of property results in a disposition for tax purposes, see Grey v. Inland Revenue Commission, [1960] A.C. 1, at 12-14.[200]

Thus, while Cattanach J. affirmed in the second part of the “test” that the buyer “acquired” the property even though legal ownership remained with the seller as security, this affirmation is nothing more than the application of the provincial private law in question, that is to say, the common law.

Furthermore, in contrast with others who have summarized the Wardean Drilling test, we believe it is untrue to state that the moment of disposition is the first to arise between the transfer of title and the normal incidents of ownership. In our opinion, the second part of the test as stated by Cattanach J. only applies where the seller reserves title as security for payment of the sale price. Therefore, a transfer of beneficial ownership is only equivalent to a disposition when the seller reserves legal ownership as security. Nonetheless, certain decisions have interpreted Wardean Drilling to mean that disposition occurs when the normal incidents of ownership, such as possession, use and risk, are transferred.

The Federal Court applied the Wardean Drilling test several years later in R. v. Henuset Bros. Ltd. (No. 1).[201] This case involved a “conditional sale agreement” in which ownership was reserved until the price was paid in full. Once the contract was signed, the buyer enjoyed all the normal incidents of ownership, such as possession, use and risk. The judge concluded thus:

The clause in the conditional sales agreements obliging the buyer to insure the tractors against such risks as the seller specified is evidence that the risk had passed to the buyer. Its failure to insure does not alter the legal effect of this obligation. On the completion of the sale the buyer had the right to use the tractors and could have taken delivery of the tractors at Peoria if it had had any use for them in that vicinity. It follows that all the incidents of ownership other than the legal title reserved in the seller by the conditional sales agreements, such as possession, risk and the right to use the tractors were acquired by the buyer on December 30, 1971. In my opinion the reservation of the legal title to the tractors in the seller as security did not affect the issue any more than the taking of security on the tractors in the form of a chattel mortgage would have done.[202](Emphasis added.)

This is clearly a direct application of the test established in Wardean Drilling. This time, however, it is applied to a situation where the normal incidents of ownership were transferred prior to title. Title was reserved by the seller as security for the balance of the sale price, in much the same way as a mortgage or hypothec.

Courts faced with tax disputes arising in common law provinces have therefore adopted a broad interpretation of the term “disposition” in addition to the test established in Wardean Drilling. We will now examine whether the courts have applied these rules to cases arising in Quebec.

3.2.2.2    In Quebec

The Supreme Court interpreted the meaning of “disposed of” under the I.T.A. in The Queen v. Compagnie Immobilière BCN Ltée.,[203] a case that involved an emphyteutic lease. First of all, it rejected the argument that the meaning of “disposed of” was restricted by the French term “aliénés”. The Court held that they both should have the same meaning, whether the translation is “disposes” or “aliénés”.

The Supreme Court also confirmed the broad interpretation of disposition:

In the context of s. 20(5), the definitions of “disposition of property” and “proceeds of disposition” cannot be said to be exhaustive; these expressions must bear both their normal meaning and their statutory meaning; it would be wrong to restrict the former because of the latter.[204]

The definition is not exhaustive. It does not narrow the ordinary meaning of the word "disposition” but adds concepts that would not normally be included in the ordinary meaning.

The Supreme Court then examined the meaning of the verb “disposer” in civil law:

The substantive definitions of “disposition of property” and “proceeds of disposition” in s. 20(5)(b) and (c) are a clear indication that the words “disposed of” should be given their broadest possible meaning. 

In French, the verb “disposer” would convey the same idea as “to dispose of”. In discussing the jus abutendi which is one of the three main attributes of the right of ownership, Mignault (Droit civil canadien, vol. 2, at p. 477) wrote:

Jus abutendi, or the right to dispose of, is the right to use the thing for a final purpose that cannot be repeated, at least by the same person. Disposing of one’s property is to transform it, to use it, to destroy it, and lastly, to alienate it, that is, to transmit it to another.” (TRANSLATION)

The same view is expressed by Mazeaud (Leçons de droit civil, t. 2, v. 2, #1332 and 1333):

[UNOFFICIAL TRANSLATION; EXCERPTS NOT TRANSLATED IN ACTUAL SUPREME COURT OF CANADA JUDGMENT]

1332. —  ...Ownership is a total right because it is absolute: the owner has complete power over the property. This set of powers can be broken into three incidents: jus utendi or the right to use the property, jus fruendi or the right to collect income, and jus abutendi or the right to dispose of the property: to retain it, to give it, to sell it, to destroy it, to abandon it....

1333. —  The prerogatives of jus abutendi. —  The right to alienate includes, aside from the right to abandon the thing and to destroy it, two significant prerogatives: the right to alienate the thing gratuitously or for value, and the right to keep it in one’s patrimony.[205](Emphasis added.)

Thus, the Supreme Court inferred that the meaning of “disposition” in the Act is the civil law meaning, i.e. to abandon, destroy, give or sell a thing. The meaning is unquestionably broader than “to alienate” (though the appellant would have wanted it to limit it that way) but certainly does extend far enough to encompass, within the definition of “disposition”, transfers of the normal incidents of ownership such as use, possession and risk.

Indeed, the excerpts cited by the Supreme Court draw a parallel between the term “disposition” and the concept of abusus, the right to dispose of the thing, that is to say, to divest oneself of ownership in the thing. Note that all the examples given by Mignault and Mazeaud refer to acts by which the owner divests himself of ownership. This is why the transfer of possession or the right of use is never mentioned. These rights are linked to the usus, not the abusus. I believe that the Supreme Court, far from having approved the application in civil law of the test in Wardean Drilling, implied that the concept of disposition in civil law is linked to abusus, that is to say, the right to abandon ownership in the thing.

The “normal incidents of ownership” test established in Wardean Drilling was introduced into civil law by the Federal Court decision in Olympia & York.[206] The taxpayer, who owned a rental income complex, entered into a bilateral promise of sale. The contract stated that the possession of the building, the income and the risk of loss would be transferred to the buyer as of the date of signature, i.e. August 31, 1969. But the document stipulated that this was not the equivalent of a sale and that ownership was reserved by the seller and not transferred to the buyer until the actual sale, which was to take place once a specific portion of the purchase price was paid.

Addy J. first discussed whether or not this was a “sale.” He referred to civil law given that the facts occurred in Quebec. The judge commented as follows regarding the complementarity of civil law:

It is evident that the rights of the parties to the contract and all matters governing various agreements and legal relations arising from the actions of the parties to those agreements must be determined in accordance with the law of the Province of Quebec.

The rights of the parties arise out of the agreement filed as Exhibit 1 and full consideration must be given to its terms. Since there is no special definition of the word “sale” or any special meaning to be attached to it in the Income Tax Act, one must consider that word in the light of the law of the Province of Quebec as applied to the relationship created by the agreement Exhibit I.[207]

The judge then analyzed the civil law rules that applied and concluded that there had not been a sale. Under the Civil Code a promise of sale with a transfer of possession is equivalent to a sale, but this rule is suppletive. The parties are free to stipulate otherwise[208] and did so in the case at bar, where they indicated their intention that ownership was not to be transferred immediately.

The judge, having concluded thus, continued his line of reasoning and examined whether there had been a “disposition” within the meaning of paragraph 20(5)(b) I.T.A., which corresponds to the present definition of “disposition of property” in subsection 13(21) I.T.A. He started by affirming that the definition of disposition in the Act is not exhaustive and must be given a broad interpretation. He then referred to the interpretation of “acquired” in Wardean Drilling, because in the judge’s view, these terms are perfect antonyms and therefore contain substantially the same elements:

The word “acquired” used in section 20(5)(e) is obviously the direct opposite of “disposed” (or disposition) as used in the same section and must contain substantially the same elements viewed from the side of the person acquiring the asset as opposed to the person disposing of it.[209]

He concluded that, based on the rule established in Wardean Drilling, there had been a disposition in the case at bar:

In the case at Bar, the Plaintiff had, after executing the agreement and upon delivering possession of the property to First General in September 1969, completely divested itself of all of the duties, responsibilities and charges of ownership and also all of the profits, benefits and incidents of ownership, except the legal title. It was absolutely and irrevocably obliged to execute and deliver a clear deed to the buyer upon receipt of the balance of the purchase price which was payable to it. Any additional rights to which it was entitled under the agreement were solely and exclusively for the protection of that balance of purchase price and are rights which would normally be granted to a mortgagee to protect his security.

Having regard to what the Supreme Court of Canada said in Her Majesty The Queen v. Compagnie Immobilière BCN Limitée, supra, as to how the concepts of “disposition of property” and “proceeds of disposition” must be interpreted and having regard also to the statement of Cattanach, J. in The Minister of National Revenue v. Wardean Drilling Limited, supra, (with which I fully agree) I find that there was in the circumstances of the present case, in September 1969, a “disposition” of Place Cremazie Complex by the Plaintiff within the meaning of section 20 of the former Act (section 13 of the new Act).[210](Emphasis added.)

This is a very significant ruling because it is the first time the court applied the definition of “disposition” in Wardean Drilling to Quebec. It is strange and difficult to understand why Addy J., who referred to civil law in order to determine whether a “sale” had occurred, failed to do likewise when interpreting “disposition” under the I.T.A., or when applying a common law precedent in civil law.

Nonetheless, it is apparent that Addy J. understood the test in Wardean Drilling as we described it above. He applied the second part of the test and concluded that the seller only retained title as security for the payment of the sale price.

This decision was followed in Robert Bédard Auto Ltée v. MNR.[211] The Tax Court of Canada had to decide whether a lease with an option to purchase within 5 years of the date of signing, at a predetermined price including any rent paid, constituted a “disposition” under the Act. Tremblay J. held that under civil law the transaction was a lease, not a sale. It was the judge’s opinion, however, that a disposition had occurred for tax purposes because the buyer had the possession and use of the building and had assumed the risks.

This case was also followed by a majority of the Federal Court of Appeal in Construction Bérou.[212] The facts of this case can be summed up as follows. The taxpayer had signed contracts for the lease of trucks. The lease agreements contained an option to purchase the trucks at a price significantly lower than market value when the options were exercised. The contract stipulated that the lessee, Construction Bérou, assumed all risk of loss for the trucks once the contract was signed and had to continue paying the rent in the event the leased object was lost. In addition, the lessee assumed all charges and expenses associated with the leased property, such as insurance, repairs, maintenance, taxes and defending the lessor in legal proceedings. The taxpayer claimed capital cost allowance, deductions for interest on the “sale price” as well as an investment tax credit, for the year in which the contract was signed.

Based on Wardean Drilling and Olympia & York, Couture J. of the Tax Court of Canada,[213] concluded that Construction Bérou had "acquired” the trucks within the meaning of the Act,[214] even though it wasn’t yet the “owner”.

The Federal Court (Trial Division)[215] overturned the Tax Court of Canada. Tremblay-Lamer J. agreed with the Department that the trucks had not been "acquired” by Construction Bérou within the meaning of the Act. The judge based his decision primarily on the fact that the lessee didn’t have to exercise the option to buy and, consequently, could not be considered to have “acquired” the property, even if the broad definition in Wardean Drilling was applied.

The majority of the Federal Court of Appeal agreed with the taxpayer that the property had been “acquired” within the meaning of the Act, granted the appeal and allowed the deductions in question. Noël J.A., in a dissenting opinion, obviously thought otherwise. In the judge’s opinion, the trucks had not been “acquired” under civil law. Since the majority refers to the opinion of Noël J.A., it will be examined first..

Noël J.A. described the issue as follows:

It is true that when Parliament frames its statutes by reference to private law concepts without defining them or otherwise attaching to them any particular meaning, it in effect adopts the laws of the provinces. The question is one of intent: it must be determined in light of the provisions at issue whether Parliament, in assigning fiscal consequences to property “acquired”, was referring to the concept of ownership as it exists under the laws of the provinces or, as the appellant contends, to a separate concept peculiar to the Act.[216]

Ownership was clearly not transferred in civil law. The question was whether Parliament had overlooked the civil law in favour of a rule specific to tax law. In other words, Noël J.A. discussed complementarity or dissociation.

The judge then conducted an exhaustive review of all the relevant case law, as well as Interpretation BulletinIT-233R.[217] He concludes as follows:

It appears from this review of the case law that to the extent that Olympia & York authorized the courts to ignore the effects of Quebec law in cases arising in Quebec, it has not been followed. Only Judge Tremblay relied on this decision in Bédard Auto Ltée as a basis for concluding that a sale had occurred for tax purposes, even though the contrary result arose under the civil law. He subsequently adhered to the civil law in Goulet. In my view, the trial judge and the Tax Court judges before her were right to ignore Olympia & York, since the Act does not cast aside private law. The word “acquired” contained in each of the provisions in issue must be understood in its ordinary sense that is as referring to the acquisition of ownership of property and in the absence of some indication to the contrary, ownership of property cannot be acquired otherwise than in accordance with the applicable private law.

Additionally, this review of the cases also indicates that the rule of acquisition stated by Cattanach J. in Wardean Drilling has been scrupulously followed in cases from common law provinces. As indicated by the passages I have cited, this rule is that property may be acquired from the time ownership is transferred to the buyer or when a buyer has possession or use and assumes the risks inherent in the property in question, even though legal title remains with the seller to guarantee payment of the selling price.

Bastin D.J. applied the second part of this rule in Henuset and Judge Bowman did likewise in Gartry. Reed J. applied this test in Borstad and Strayer J. in Kirch. "I note that none of those decisions suggests that the rule of acquisition proposed by Cattanach J. departed from the applicable private law." As Chief Judge Couture noted in the case at bar, "in Wardean Drilling Cattanach J. was simply referring to the dismemberment of the ownership right at common law and the rule that there is a disposition of property when there is a change in the beneficial ownership, even though the legal ownership remains unchanged.”[218](Emphasis added.)

Noël J.A. cast aside the principle of complementarity and concluded that the concept of acquisition under the Income Tax Act was not dissociated from civil law. This concept must therefore be interpreted in conformity with civil law. He also held that acquisition in civil law is directly tied to ownership and that cases that followed Wardean Drilling are not applicable to Quebec.

Noël J.A. then went on to examine the provisions of subsection 248(3) I.T.A. In his opinion, they had been enacted in order to introduce concepts equivalent to beneficial ownership for application to Quebec. In any event, he considered this provision irrelevant to the resolution of the dispute before the court.

Lastly, Noël J.A. emphasized the importance of the parties’ intent in civil law. They are free to provide the terms and conditions they desire, specifically with regards to the transfer of ownership. Therefore, a stipulation that the lessor retains ownership of the property until the option is exercised is valid. Neither the Department nor the parties themselves can ignore the terms of a contract freely negotiated. To this effect, Interpretation Bulletin IT-233R has little basis in law:

In short, I come to the conclusion that Bulletin IT-233R, to the extent that it attempts to anticipate the future and lay down a rule that property leased under a leaseback contract is sold on signature of the contract if the price of exercising the option is “substantially less” than the “probable” value of the leased property, or if at the time the contract was signed “no reasonable person would fail to exercise the said option”, is devoid of any legal basis.

In legal terms, there is nothing to prevent the parties to a contract from validly stipulating that ownership of the leased property will remain with the lessor even if the cost of exercising the option as compared with the “probable” value of the leased item may appear to be “substantially less” at the time the contract is signed.[219]

We believe Noël J.A. correctly interpreted the law and agree with his conclusions, notably his interpretation of acquisition under civil law. We will return to this question later in this paper.

Létourneau J.A. appeared to adopt a position diametrically opposed to that of Noël J.A. His main concern, it seems, is that taxpayers in Quebec and the common law provinces receive equal treatment. He relied primarily on the former wording of subsection 248(3) I.T.A.:[220]

However, I agree with him that subsection 248(3) of the Income Tax Act (“the Act”) evidences a valuable effort by Parliament to treat beneficial ownership in property in the same way as various forms of ownership recognized in the civil law of Quebec so as to obviously offer to the taxpayers in Quebec the same benefits that this concept affords the taxpayers in the common law provinces. This was not an easy task to perform at the time because the concepts of ownership were different in the two legal systems, and the divisions of the ownership right, more limited in civil law than in common law, were not conceptually necessarily identical to those of the common law. Yet, the attempt by Parliament to harmonize the two systems with a view to providing fair and equal treatment to all Canadian taxpayers cannot be doubted. Hence, the necessity for a judicial interpretation which allows for the implementation of this legislative intent.

In addition, subsection 248(3) of the Act, in my view, provides a legislative basis for the application in Quebec of Interpretation Bulletin IT-233R. That subsection confirms and supports the conclusion which I have reached in light of the Act regarding the concept of acquisition of property for purposes of a capital cost allowance.[221](Emphasis added.)

Thus, Létourneau J.A. was of the view that Wardean Drilling and Olympia & York stated Quebec law, under the Act, as far as disposition and acquisition are concerned:  an acquisition or disposition occurs when the normal incidents of ownership, such as possession, use and risk, are transferred. In his view, Parliament’s intent in enacting section 54 and subsection 248(3) was to apply the same criteria to Quebec taxpayers as to taxpayers in other provinces. He added the following in support of this interpretation:

For practical purposes this interpretation has the merit of recognizing, for tax legislation that applies throughout Canada, a business practice that has no boundaries and of avoiding the danger of becoming too embroiled in unnecessary, sectoral and above all sterile and inequitable legalism at a time when the trend in the civil law is to approximate more closely to the common law. In addition, it is significant that Parliament, which annually amends the Act inter alia to alter legislative provisions when they are so interpreted that they do not meet the objectives sought, has not thought it appropriate to overturn this thirty-year-old interpretation. Further, this interpretation is consistent with the legislative intent stated in subsection 248(3) of the Act, which, as I have already mentioned, is intended to treat beneficial ownership of property in the same way as various forms of ownership recognized in the civil law of Quebec.[222](Emphasis added.)

With respect, we believe Létourneau J.A.’s interpretation runs entirely counter to the rule that provincial private law is complementary, and the fact that tax law should follow the legal relationship between the parties and not drive it. As we have seen, if Parliament wants a single legal concept to apply across the country, it can and should expressly amend the Act.[223]

However, the legislative intent behind subsection 248(3) (now paragraph 248(3)(f) I.T.A.) may have been to equate certain civil law dismemberments of ownership, such as the usufruct, with beneficial ownership. But this subsection in no manner recognizes that the normal incidents of ownership such as use, possession and risk, are equivalent to beneficial ownership in civil law. This provision does equate various civil law concepts to “beneficial ownership.” However, according to the interpretation principle ejusdem generis,equivalency cannot be extended to a concept that is not itself a civil law institution. Therefore, one could not argue that subsection 248(3) expresses Parliament’s intent to apply the Wardean Drilling test in Quebec. Michael Templeton is of the same view:

The majority of the Court of Appeal appears to interpret the provision as making the broad statement that whenever the Civil Code recognizes a property interest that is similar to a property interest that is recognized at common law as beneficial ownership, for the purposes of the Act, the Civil Code property interest shall be treated as beneficial ownership. However, the language used in subsection 248(3) does not appear that broad. The provision states that certain Civil Code property interests (those specifically listed in the subsection) are to be treated as beneficial ownership for the purposes of the Act; however, it is only those property interests that are specifically listed that are to be given this treatment.[224](Emphasis added.)

It must also be pointed out that it was the taxpayer in the case at bar who argued that the Act should be applied uniformly across the country, while the Department of Revenue relied on the civil law; the oposite situation is seen more frequently. In addition, the Minister in this instance adopted a position contrary to the one published by the Department in Interpretation Bulletin IT-233R and relied upon by the taxpayer. Létourneau J.A. appears to have been influenced by the apparent unfairness toward the taxpayer:

It is a matter for surprise that in the appeal at bar the respondent is using the special nature of Quebec civil law as a reason for denying the appellant a deduction which is granted to taxpayers and businessmen operating under the common law system....

In my view, it would be inappropriate for this Court to ignore or repudiate the content of the said Bulletin 17 years later, as the respondent is asking us to do in the case at bar, since the Bulletin clearly and correctly reflects the state of the legislation and case law applicable in tax matters at the time to property acquired by a taxpayer through a leasing transaction.[225]

Desjardins J.A. analysed the wording of section 54 and subsection 248(3). He referred to the Carter Commission, which recommended that, for purposes of taxing capital gains, the word “disposition” be given a broad meaning. Parliament subsequently adopted the definition of disposition in section 54, and as there had to be only one meaning of “disposition” applicable throughout Canada, subsection 248(3) of the Act was adopted for its application in a civil law environment.[226] Desjardins J.A. then held:

The federal Parliament accordingly devised, for tax purposes and, all of Canada, a common concept covering the ideas of “disposition” (“disposition de biens”) and “beneficial ownership” (“propriété effective”), both in civil and common law;...[227]

It is entirely possible that Parliament intended to dissociate itself from provincial private law and establish a specific definition that is uniformly applicable whenever the Act is involved. As we have already seen, the problem is that Parliament refers to common law concepts (beneficial ownership and legal ownership) in the definition of “disposition”, but fails to define them.

Furthermore, subsection 248(3) does not exhaustively define "beneficial ownership” for purposes of applying the Act in Quebec. It only provides examples of Quebec civil law concepts that are deemed equivalent to beneficial ownership. This means that one must refer to provincial private law in order to complete the definition. How, then, can one maintain that Parliament defined a common concept of “beneficial ownership” for all of Canada? Moreover, the test involving normal incidents of ownership, such as possession, use and risk, are nowhere to be found in subsection 248(3). It is the case law, not the neutral definition found in the Act, that has led to the importation of this criterion into civil law. The confusion is clearly illustrated in this excerpt from Desjardins J.A.’s decision:

However, in 1982 when the leasing contracts were signed s. 248(3) of the Act was already in place and provided that for tax purposes certain contracts were capable of transferring the “beneficial ownership”. For the sake of comparison, and to clarify my analysis, I would add that although a usufructuary is not the owner of property at civil law he can in fact have “beneficial ownership” of the property within the meaning of s. 248(3) of the Act, since that subsection creates its own ideas of “beneficial ownership”.

In the case at bar, despite clause 20 of the contracts governing the parties’ rights at civil law, tax law by s. 248(3) of the Act recognized that the appellant had acquired beneficial ownership of the dump trucks since it met the three requirements, possession, use and risk, recognized by the courts.[228](Emphasis added.)

Desjardins J.A. himself was forced to admit that these criteria have not been enacted by subsection 248(3) I.T.A., which supposedly “creates its own ideas of ‘beneficial ownership’”, but by the “case law”, i.e., cases on tax law that interpreted the concept of acquisition in a common law context.

For reasons already discussed, we cannot agree with the majority’s conclusion in this case. We do not feel that the reasoning therein, which is essentially based on considerations of fairness and the objective of a uniform application of the Act, should be followed. We believe Noël J.A. has once again correctly interpreted the law.

Because the concept of “disposition”, and its reciprocal concept “acquisition”, was interpreted in a common law context, it was established that a disposition occurs when ownership, or the normal incidents of ownership (such as use, possession and risk) are transferred, if the seller has retained ownership as security.

The civil law cases are divided as to the application of these tests. There is a trend toward adopting the common law criteria in the name of a uniform application of the concept of “disposition” throughout Canada. An opposing trend, based on the complementarity of civil law, favours an interpretation consistent with civil law institutions where the concept of disposition is to be applied in Quebec. As stated, earlier, we favour this second trend.

3.2.3    The retroactivity of conditions in civil law

We will now consider how the cases deal with suspensive conditions, and the way the cases apply retroactivity in determining when a “disposition” takes place.

There is a trend in favour of recognizing the retroactivity of conditional obligations. Note that according to case law, there can be no tax consequences if a contract no longer exists, because the necessary basis for the tax no longer is present.[229]  The following question arises: does the extinction of a contract affect only future tax consequences?

In Dominion Engineering, Rand J., of the Supreme Court of Canada, wrote as follows per curiam:

If, in the legal result, the actual transaction ceases to be one of sale, then the necessary support for the tax disappears. That result, at least where the termination of the contract does not effect a total rescission, will not affect the right to taxes on any portion of the price paid to the seller nor does it touch those that have been collected or reduced to judgment by the Crown.[230](Emphasis added.)

Thus, Rand J., in obiter dictum, held that the extinction of the contract does not affect taxes already collected, at least where the contract has not totally been rescinded. Was he implying that in contrast, i.e. where the contract has been totally rescinded or terminated, tax paid in the past should be refunded because the essential reason for the tax is no longer present? Further on, Rand J. held:

...where the obligation of such an executory contract is by operation of law destroyed, then unpaid taxes related to its terms, themselves suffer a corresponding effect. If that were not so, sellers with unsold property on their hands would be liable for taxes in respect of purchase price not only unpaid but the legal right to which had been annulled....[231]

Isn’t the situation of a seller subject to a resolutory or suspensive condition who repossessed property following the resolution of the sale, the very situation that Rand J. considers absurd?

In a context similar to Dominion Engineering, the Federal Court of Appeal, in Price (Nfld.) Pulp & Paper Ltd. v. The Queen,[232] wrote in obiter:

...[I]t may also be that even tax paid on accrued instalments may become refundable if a total rescission of the agreement of sale occurs...[233]

The Federal Court of Appeal, in Perini Estate v. The Queen,[234] a case involving contingent liability[235] in common law, held that a fulfilled condition had a retroactive effect much like conditional obligations in civil law.

The taxpayer in that case had sold all the shares in a company. The sale contract provided that the sale price was to include a certain sum payable immediately plus additional sums calculated as a percentage of future profits (earn-out). The contract also stipulated that interest, computed from the date of the sale, would be payable on these additional sums.

The Minister assessed this income as interest income. The taxpayer argued that it was not interest as defined under the Act since the principal used to calculate the interest, i.e., the additional sums indicated in the contract, did not exist until extracted from the financial statements. An essential element was therefore missing in order for the interest to be “interest” within the definition of the Act, that is, a principal amount on which interest could accumulate.[236] The Minister, for his part, argued that the additional amounts, once determined, had a retroactive effect with the result that the principal was deemed to have always existed and the interest could have accumulated on the principal as of the date the contract was signed.

Writing for the Court, Le Dain J. began by explaining that the obligation to pay the additional sums was a contingent liability at common law. He then cited the Federal Court trial judge:

The learned Trial Judge concluded that the fulfilment of the condition had a retroactive effect. This conclusion is contained in the following passage from his reasons:

Once it was ascertained that the profits had been made and could be calculated and the seller was still alive his obligation for the payments in each of the years 1969, 1970 and 1971 became due, and the condition having been fulfilled it had a retroactive effect to the date of the contract. Interest ran from that day on the payments due in accordance with the terms of the contract.

He based this conclusion on the assumption that the common law as to the effect of the occurrence of a contingency did not differ in principle from the rule in article 1085 of the Quebec Civil Code that “the fulfilment of the condition has a retroactive effect from the day on which the obligation has been contracted. ”[237](Emphasis added.)

The Federal Court of Appeal agreed with the trial judge that the contingent liabilityhad a retroactive effect, just like suspensive conditions in civil law. Their opinion was based on a decision of the Privy Council[238] that drew an analogy between contingent liabilities in English law and conditional obligations in Scottish law. The Court of Appeal concluded that the Canadian common law contingent liability resembled the conditional obligation in Quebec civil law, and therefore had the same retroactive effect.

Thus, the Federal Court of Appeal established that both civil law conditional obligations, and common law contingent obligations, have a retroactive effect. However, the Court then stated that the common law does not clearly exclude retroactivity and that the parties are therefore free to do so. This is what occurred in the case at bar where the contract indicated that “interest” would be payable on the “principal” calculated on the future profits.

In any event, there does not appear to be any doubt as to the retroactivity of conditional obligations in civil law, even if the Federal Court of Appeal did not go so far as to state that contingent liabilities necessarily have a retroactive effect. In addition, this retroactivity applies in tax law, at least with respect to interest.

Moreover, in Construction Bérou, Tremblay-Lamer J. of the Federal Court also stated, in obiter, that the retroactive effect of a suspensive condition applied to the concept of acquisition in the I.T.A.:

The cases hold that the taxpayer may have “acquired” property where the transaction is considered to be a conditional sale of a suspensive nature....

Since the transaction is characterized as a sale on a suspensive condition, the effect, once the condition is performed, is to transfer the ownership as of the day it is signed.[239]

As we have seen, some cases are favourable to the application of retroactive conditions in tax law. But there are other cases, in which it is held that tax law does not recognize the retroactive effect of resolutory conditions. It must be pointed out from the outset, however, that these decisions were rendered under particular circumstances, involving obvious bad faith or the deductibility of employer contributions.

The Federal Court’s decision in Alepin v. The Queen[240]has been cited in support of the proposition that retroactivity cannot be set up against third parties, including Revenue Canada (now the CCRA).[241] In Alepin, three brothers had sold their land to Jar Investments Ltd. The balance of the sale price was secured by a resolutory clause that applied in the event of the buyer’s default. As we have seen, such a clause has the same retroactive effects as a resolutory condition.[242]

A dispute arose between the Alepin brothers and Revenue Canada regarding the nature of a significant payment made by Jar Investments Ltd. on the balance of the sale price. The taxpayers argued that it was capital and thus that the payment, received in 1970, was totally tax-exempt. The Minister argued that part of the payment was on account of accumulated interest, and was therefore taxable as interest income.

In 1975, the Alepin brothers exercised their rights under the resolutory clause and took back possession of the land by way of “giving in payment” (dation en paiement). This event took place several years after the payment had been received and the transaction had been assessed by the Department of Revenue. If the sale was resolved, the resolutory clause specifically provided that all payments received up to then would be treated as damages for breach of contract. For this reason, the taxpayers submitted that the payment received in 1970 was on account of capital.

Marceau J. rejected that argument:

So far as the alternative argument based on the retroactivity of the giving in payment in 1975 is concerned, I do not see how it applies. A contract subject to a condition of dissolution has full legal effect from the time it is concluded, and until the event provided for occurs, it is completely applicable. When the payment was received in 1970, a part of it covered the interest due under the contract and was taxable forthwith, and this legal situation could not be subsequently modified or extinguished by the effect of a subsequent dissolution of the contract itself. The occurrence of the condition of dissolution to which the contract is subject may well extinguish the obligations resulting from the contract, but it can only affect third parties who have acquired rights in the meantime based on the contract to the extent that such rights were themselves made subject to a condition. Furthermore, in the case at bar the dissolution did not take place independently of the parties: it required a free, deliberate act by one of them, and was in fact brought about by a freely concluded contract of giving in payment:  is it reasonable to suppose that an ex post facto dissolution in 1975 could modify the destination of payments made in 1970, and extinguish their consequences for the federal treasury? (Cf. Malkin v. Minister of National Revenue, [1942] Ex. C.R. 113.), [1942] C.T.C. 135, 2 D.T.C. 587).[243](Emphasis added.)

The judge appears categorical: the retroactive effect of the cancellation can have no tax consequences. This statement needs to be tempered. First, the resolution of the contract at stake was clearly intended to avoid the tax consequences of the transaction following the assessment and ensuing dispute. Marceau J. unquestionably took this obvious bad faith of the parties into consideration. Moreover, the judge relied on Malkin, where the parties, subsequent to a first contract, signed a second one that modified the nature of the amounts received. The case did not involve a resolutory condition that was part of the contract from the outset; and what is more, the case originated in a common law province: the retroactivity was not based on the Civil Code.

Secondly, it should be recalled that this case involved a resolutory clause, not a resolutory condition. As we have seen, a condition is an extrinsic event, independent of the will of the parties. The Court’s position that the resolution required a voluntary and free act of one of the parties would not apply to a true resolutory condition.

The Tax Court of Canada cited this case in Larose,[244] where the taxpayer sold land but the Superior Court subsequently annulled the sale at the taxpayer’s request. The taxpayer in Alepin argued that the tax consequences were retroactively cancelled by the court-ordered resolution.

The judge rejected this argument based mainly on the decision of Marceau J. in Alepin:

The judgment nullifying the sale rendered by Mercure J. in April 1990 certainly cannot affect the rights that the Respondent acquired as a result of the sale of the buildings in 1979. Relatively large sums were payable to the Respondent as a result of this transaction. In addition, the evidence established that the main object of that legal proceeding was to enable the Appellant to be discharged of his debts to the Minister of National Revenue. This Court is of the opinion that such a situation is directly dealt with in the case law as reflected in such judgments as Malkin (para. 4.02(1)), Alepin (para. 4.02(2)) and Adam (para. 4.02(3)). In these cases, the courts have held that any attempt to retroactively change the nature of certain payments in order to benefit from a more advantageous tax treatment could not affect the Minister of National Revenue. This is why such a principle applies a fortiori when a taxpayer attempts to annul a transaction retroactively in order to eliminate the tax consequences to which it gives rise.[245](Emphasis added.)

One should note that the court in this case also wanted to punish the taxpayer for his bad faith, and refused to endorse the “retroactive tax planning” from which the taxpayer hoped to benefit.[246] In addition, the retroactive nature of a resolutory condition in civil law was not an issue in this case.

Some decisions have also refused to recognize the retroactive effect of suspensive conditions in tax matters. For example, in Fédération des Caisses populaires Desjardins de Montréal et de l’Ouest-du-Québec v. The Queen,[247] the Tax Court of Canada considered the question of whether the employer could deduct contributions for vacations earned but not yet taken by employees. In calculating its income, the employer deducted an amount for holidays accumulated by its employees during the year but not yet taken. This deduction was allowed and accepted by the Department of Revenue. However, the employer also claimed contributions to various social programs (retirement pension, employment insurance, Quebec pension plan, group insurance, etc.) calculated on these vacation pays earned but not yet taken.

The Minister disallowed the deduction for these employer contributions under paragraph 18(1)(e) I.T.A. claiming they were not provided for by law. The issue was whether or not these deductions constituted future reserves, in other words, was the employer required to make these deductions at that moment even though they were not due until the employee left on vacation.

The Tax Court of Canada held that the obligation to make contributions under the various statutes arose when the employees received their vacation pay. This obligation was considered to be subject to a suspensive condition and was not an obligation with a term. Consequently, the judge ruled that, the contributions were not deductible until the fulfilment of the condition because the obligation did not exist until that moment. The judge unfortunately failed to consider the retroactive effect of the condition in civil law.

In any event, this decision was reversed on appeal.[248] The majority of Justices were of the opinion that the obligation was with a term and was not an obligation subject to a suspensive condition. The dissenting Justice did not discuss suspensive conditions.

To sum up, we believe there are two opposing trends in the case law regarding the recognition, in tax law, of the retroactive effect of conditional obligations. One favours retroactivity while the other maintains that it should be ignored. However, there has not been a direct ruling on whether the fulfillment of a condition has a retroactive effect regarding the moment of disposition of property within the meaning of the Act.

In our opinion, it is very difficult to predict where the courts will address this question. Nonetheless, we believe that the current case law does not necessarily support a rejection of retroactivity. On the contrary, we are inclined to believe that the Federal Court of Appeal is in favour of recognizing retroactivity, as suggested by Perini Estate.

As for the scholars, some support retroactivity in tax matters,[249] while others have criticized such a stance.[250]

Professor Diane Bruneau has written one of the only articles to specifically analyze the application of retroactive conditional obligations in tax law. Her starting point is that a disposition occurs when the seller is entitled to the sale price, as confirmed in Victory Hotels.[251] She then states that the right to the proceeds of disposition arises at the moment ownership is transferred. At that point, she writes, we are back at square one and must determine whether the transfer of ownership can be retroactive (where the condition is suspensive) or be revoked (where it is resolutory.)

Turning to the central issue of the condition’s retroactivity, Professor Bruneau writes:

[TRANSLATION] Even in civil law, the temporary situation that existed prior to the occurrence of the condition cannot be erased by retroactive ownership. Therefore, a person who retroactively acquires income-producing property is not entitled to restitution. Similarly, a buyer subject to a suspensive condition does not assume the risk if the property is completely destroyed. The parties may nonetheless modify these effects by mutual agreement. Faribault addresses other situations not specifically provided for by law:

The retroactive nature of a fulfilled condition must be given a narrow interpretation because it is a legal fiction. If there is a doubt, such an effect must be denied because its fictitious nature cannot prevent a fact from having occurred in the past.

This fiction cannot erase the fact that the conditional debtor, where the object of the obligation is a concrete thing, had possession pendente conditione. If this possession is unaffected by the condition’s retroactivity, anything that flows naturally and logically from it should also be unaffected. This includes but is not limited to any acts of administration performed by the debtor, and any fruits he collected while in such possession.

For his part, this is what Mignault says:

...retroactivity attached to a fulfilled condition only applies to de jure matters. It was meant to benefit the buyer, who would otherwise be subject to any alienations, servitudes or hypothecs granted by the seller pendente conditione. It was also meant to benefit his heirs. It does not affect de facto matters. The collection of fruits is a fait accompli that the fulfilment of the condition cannot defeat.

The position of these scholars is that the buyer’s title is protected by the retroactive nature of the sale, but that reality cannot be altered. The tax cases to date accept that there can a disposition can be the result of certain de facto events. It would therefore be possible to argue that even where there has been a de jure sale, the actual disposition only occurs at the moment the condition is fulfilled. [252] (Emphasis added.)

Bruneau refers to Olympia & York in support of her statement that “[t]he tax cases to date have accepted that a disposition can be the result of certain factual events.” She notes that [TRANSLATION] “the incidents of ownership were judged sufficient” in that case for the Court "to hold that a disposition had occurred, even if title had not been transferred and a sale had not yet taken place.”[253]

Our understanding of Me Bruneau’s contention is as follows:  in tax law a disposition can occur de facto where there is a transfer of possession, use and risk; and retroactivity cannot affect a de facto disposition because retroactivity, in civil law, applies only to that which is de jure, not de facto. In her article, Professor Bruneau also points out numerous practical obstacles to the application of retroactivity in tax law: its ambivalent effects; the absence of provisions permitting past declarations to be varied; prescription; fairness; and the impact and scope of statutory amendments. She concludes that, given these difficulties and the factual nature of dispositions, conditional obligations should not be given a retroactive effect for tax law purposes.

With due respect, we disagree with the premises underlying this reasoning. For the one thing, we have already stated that the civil law “principle” whereby retroactivity only applies to de jure matters does not appear to be well settled.[254] At the very least, we believe this principle should only apply to “concrete and incontestable” or “ineffaceable” facts, that is to say the specific cases provided for in the Civil Code, namely fruits and acts of administration. Under the new Code, risk is no longer linked to ownership, but rather to possession. Thus, the fact that retroactivity does not apply to risks does not actually stem from the principle invoked.

Me Bruneau’s statement that disposition arises de facto is based on Olympia & York.[255] As  stated earlier,[256] we feel that this decision erroneously introduced the “normal incidents of ownership” test, articulated in Wardean Drilling, into civil law. We believe the concept of “disposition” should be interpreted according to civil law when it is to be applied in Quebec.

In addition, we are unable to consider that a disposition is an “incontestable” fact. We believe that dispositions are tied to the transfer of ownership. Thus, the transfer of ownership itself cannot be a “fact” that is immune from the retroactive effect of a condition.

Our opposition to Professor Bruneau’s position stems from the fact that we have different starting points. As we previously explained, we proceed from the principle that private law is complementary, and that if Parliament wishes to cast aside the private law meaning of a term, it must expressly define the term in issue. Me Bruneau implicitly seems to proceed from the postulate that the concept of “disposition”, as defined by case law, which applies in tax law is dissociated from provincial private law.

3.2.4    Other retroactive situations in civil law

We saw in the first chapter that certain civil law concepts can have retroactive effects. We will now examine the tax consequences of these events based on the current state of the law.

3.2.4.1    Resolution of contracts for non-performance of an obligation

As discussed above, the resolution of a contract produces the same effects as a resolutory condition. However, this situation is unique in that a creditor usually requests that the contract be resolved when the debtor is in default, generally in default of payment. Therefore, sections 79 and 79.1 of the Act apply to the return of property following the debtor’s failure to pay.

In essence, there are then two successive dispositions of the property under these sections. They also establish rules to calculate the proceeds of disposition from the property that go to the debtor and the cost to the creditor.

Under what circumstances will these sections apply? Subsection 79(3) establishes the rules that apply to a debtor where “a particular property is surrendered at any time by a person…to a creditor of the debtor.” Subsection 79(2) states that where a person (the creditor) acquires the “beneficial ownership” in property after the other person (the debtor) has failed to pay an outstanding debt, the creditor acquires the property by surrender. Likewise, a creditor who seizes property, that is to say a creditor who acquires the “beneficial ownership” in the property after the debtor defaulted on payment, is covered by the rules in subsections 79.1(2) and (6).

This calls for two comments. First, as we have seen, the Supreme Court held that the obligation to pay the sale price is not a condition as understood by the civil law. Therefore, these sections will practically never be applied to true conditional obligations.[257] They will only apply where the contract is resolved for non-performance, or in an instalment sale. Second, whether or not the creditor acquires “beneficial ownership” following the debtor’s default where the contract is resolved for non-performance depends on whether retroactivity under the Civil Code applies to tax law in general. Because of the resolution’s retroactive effect, ownership was never transferred, and therefore, the creditor cannot “reacquire” property that was never given away.[258]

But the definition of “disposition” gives rise to the following problem, however: it refers to “beneficial ownership”, a concept that only has meaning in a common law context. Can a Quebec creditor, deemed by the Civil Code never to have disposed of ownership, give up beneficial ownership and then recover it following the debtor’s failure to pay? Me Pierre Archambault phrased the question as follows:

[TRANSLATION] Lastly, in order for section 79 of the Act to apply, there must be an “acquisition” or “new acquisition” of the “beneficial ownership” or ownership of the property. Under the Civil Code of Québec, no such acquisition or new acquisition of property exists. Rather each party returns that which he received so that they find themselves in the same position as if the contract had never existed (article 1088 of the Civil Code). It is therefore appropriate to question whether section 79 of the Act applies in these circumstances.[259]

3.2.4.2    Instalment sales

As we have seen, an instalment sale is a term sale, not a conditional sale. Ownership is only transferred in civil law once the term occurs (i.e. once the sale price is fully paid) and there are no retroactive effects.

But the question still arises as to when disposition occurs in tax law. There is a disposition (if the test in Wardean Drilling is applied) when the contract is signed even though ownership is reserved as security for the payment of the sale price because possession, use and risk of loss were transferred to the buyer at that moment.

Moreover, there is also the issue of whether sections 79 and 79.1 apply, for if the buyer defaults on payment, the seller will take back the property that he never stopped owning. Nevertheless, these sections will apply if it is held that beneficial ownership was parted with.

3.2.4.3    Nullity when the conditions of contract formation are not met

Nullity has the same effects in civil law as a resolutory condition, and therefore it produces the same tax consequences. Hence, our comments regarding conditional obligations[260] also apply to annulled contracts.

3.2.4.4    Sales with a right of redemption

A sale with a right of redemption is essentially a sale subject to a resolutory condition, so it should have the same tax consequences. The use of the word “redemption” is no reason for the courts to hold that there are two dispositions, and to deny the sale its retroactive effect. Contrary to what the term might suggest, the “redemption” is not a second sale, nor does it refer to a second sale. It is simply a retroactive cancellation of the first sale in the same manner as if a resolutory condition had occurred. The legislator unfortunately created this confusion when it opted for the term “redemption” instead of “right to repurchase.”

In any event, a sale with a right of redemption is often contemplated by subsection 248(1) “disposition” (j) I.T.A., which states that the term disposition does not include:

any transfer of the property for the purpose only of securing a debt or a loan, or any transfer by a creditor for the purpose only of returning property that had been used as security for a debt or a loan.

A sale with a right to repurchase is most often intended to secure the seller’s debt to the buyer.[261]

3.2.4.5    Trial sales

Similarly, a trial sale is simply a type of sale subject to a suspensive condition where possession and use of the thing is transferred immediately. It should, therefore, have the same effects in tax law as a sale subject to a suspensive condition, depending on whether or not the retroactive effect of conditional obligations is recognized.

3.2.4.6    Promises of sale

As we have seen, a promise of sale with delivery and possession is equivalent to a sale under the Civil Code, unless the parties provide otherwise. Thus, the civil law allows one to hold that a disposition under such circumstances takes place upon delivery.

This is what occurred in Dubois v. The Queen.[262] The issue in that case was the deductibility of interest attributable to a period during which the buyer had possession of the property, and was responsible for all the charges, but before a notarial contract of sale had been signed. The Minister submitted that there had been no disposition prior to the closing of the sale.

The Court held that there was a promise of sale with a transfer of possession, and that the parties had intended to transfer ownership immediately even though the notarial contract was only to be signed later. Thus, at the moment possession was taken, a sale had occurred under the civil law, for this is when the parties agreed that ownership would be transferred. As a result, the interest that had run from that date was deductible, since it pertained to property “acquired” at that date.

The date of “acquisition” (within the meaning of the Act) coincided with the transfer of ownership (as the civil law understands it.)  Although neither retroactivity nor suspensive conditions were involved in this case,[263] the court nonetheless stated that the moment property is acquired must be determined with reference to the moment the civil law considers ownership to have been transferred.

The result is therefore the same as in common law when the ordinary attributes of ownership are transferred to the buyer after the deed is actually signed. In Kozan v. MRN[264] for example, the parties agreed to make the sale effective in November 1979; the possession, use and risks were transferred at that date but the signature of the deed was postponed until certain formalities were completed. The deed was finally signed in 1980 but the court held that the disposition had taken place in 1979 because of the parties’ conduct: they treated their transaction as completed from the moment the ordinary incidents of ownership were transferred.

The tax consequence was the same in both these decisions, while respecting the private law of their province of origin.

Conversely, the Court held in both Robert Bédard Auto Ltée v. MRN[265] and Olympia & York,[266] that there had not been a sale within the meaning of the Civil Code.  The parties had expressed their intention to delay the transfer of ownership until the act of sale was signed, even though the promise of sale was accompanied by delivery and transfer of possession of the thing sold. It was nonetheless held according to the criteria in the common law case law that a “disposition” had taken place within the meaning of the I.T.A., which is the trend in the case law that applied to test in Wardean Drilling.[267]

3.2.4.7    Contractual retroactivity

We have seen that the parties can stipulate that the contract is to take effect at a date earlier than the signing date, as long as the parties had reached a definitive agreement on this earlier date regarding the essential elements of the contract. The question then becomes whether this earlier date then becomes the transaction date for tax purposes. The relevant authorities in both systems will be reviewed because the issue is the same in both civil law and common law, so the answer should also be the same.

As explained above, for tax purposes, ownership can validly be transferred before the moment of signature of the deed of sale, if the parties so intend. In Dubois and Kozan, the Court held that there had been a disposition within the meaning of the I.T.A. as soon as there was a promise of sale with a transfer of possession, even though the deed of sale was not signed until a later date.[268]

In Reilly Estate v. The Queen,[269] the Federal Court (Trial Division) held that a disposition, for tax purposes, had occurred as soon as the parties entered into a binding agreement.

The taxpayer in this case had agreed, in a letter of agreement signed in 1972, to sell certain lots of land. The final contract was only signed in 1973. Muldoon J. determined the moment of disposition based on a binding agreement test: he examined the letter of agreement and held that it was a valid contract binding the parties according to the rules of common law because they had agreed on all the essential elements of the contract. The final contract reproduced the terms in the letter of agreement, except for a few details. The judge held that the sale had taken place as soon as the letter of agreement was signed, and that it was therefore at this moment that a disposition had occurred, since this is what gave the seller the right to the proceeds of disposition, as decided in Victory Hotels.[270]

The Federal Court Trial Division recognized a stipulated retroactive effect of a contract in Miller v. The Queen.[271] The taxpayer, a teacher, was party to a collective agreement that expired in December 1979. Arbitration ensued and the new agreement, to take effect in 1981, provided for a retroactive increase to January 1, 1980, plus interest payable on the increase.

The Minister claimed the additionnal amount was not interest as there was no principal at the time the interest accrued. However, the taxpayer relied on Perini Estate and submitted that it constituted interest, even if the amount of the principal could not be determined in advance.

Reed J. agreed with the taxpayer because he was of the view that this case was similar to Perini Estate and that the retroactive effect stipulated by the parties in their agreement had to be applied:

Equally, I cannot find in the Perini and Huston cases a requirement that in order to constitute an interest payment the formula for such payment must be decided upon prior to the commencement of the time period to which the interest relates. It is open to the parties to govern their relationship by retroactive agreements: Trollope & Colls et al. v. Atomic Power Constructions, Ltd., [1962] 3 All E.R. 1035. And it is open to them, when they do so, to provide for interest to be payable on the outstanding sum left due over the relevant period of time. In my view the taxpayer’s situation in this case is similar to that of the taxpayer in Perini....

In my view the $62.51 was genuinely a payment of interest. The parties agreed that their relationship would be governed on the basis of the retroactive agreement. This involved the retention of monies owing to the Plaintiff for which compensation was ultimately paid. The compensation paid was described by the parties and the arbitration board as interest. It was calculated on an accrual basis by reference to a normal rate of interest then current or with respect to the employer’s cost of borrowing. I can see no reason why this does not fall within the meaning of the word “interest” as it is used in section 110.1 of the Income Tax Act.[272](Emphasis added.)

As for the doctrine, most of the scholars[273] are of the opinion tax law can apply an earlier effective date than that on which the contract was signed, provided that the parties had a verbal but legally binding agreement at this earlier date; an offer must have been accepted and all the essential terms of the contract must have been definitively determined in order for there to have been such an agreement.

3.2.4.8    Retroactivity under the Civil Code

We have seen that the Civil Code contains a number of situations involving retroactivity, including dissolutions of marriages and successions. Is this retroactivity recognized in tax law?

In M.N.R. v. Faure, the Supreme Court answered this question in the affirmative.[274]   The issue in that case was whether the property of the deceased spouse had “passed” to the surviving spouse at the time of death, within the meaning of the Estate Tax Act.[275] The spouses were married under community of property according to the rules of the Belgian Civil Code, which the parties agreed, would be treated like the rules contained in the Civil Code of Lower Canada. Their prenuptial agreement contained a right of survivorship clause.

The Supreme Court held that the property had not “passed” to the surviving spouse upon the spouse’s death because in civil law she was deemed to have acquired the community property retroactively to the date of marriage. Writing for the Court, De Grandpré J. held:

Whatever the nature of the community may be, on its dissolution by the death of the husband, giving rise to application of the above-mentioned stipulation in the marriage covenants, the widow became owner of all the property, retroactively to the date of the marriage. In Sura v. Minister of National Revenue, [1962] S.C.R. 65, speaking of the share of the community property going to the spouse in a case in which the exclusive right of the survivor was not at issue, Taschereau J., as he then was, stated (at p. 71):

[TRANSLATION] ...If the wife was not co-owner of the community property, she would have to pay succession duties on dissolution of the community, because there would then be a passing of property from her husband. However, this is not the case here, because there was no passing, but partition, in which she took the share coming to her, which had belonged to her since the marriage. What she received did not come from the estate of her husband.

In support of his views, Taschereau J. cited as authorities several authors, including Mignault, who stated, in volume six of his Droit Civil, p, 337, that in the event of renunciation the interest is retroactively terminated, the other spouse being [Translation] “deemed to have always been the sole owner of the property which made up the community.[276](Emphasis added.)

Thus, the Supreme Court expressly recognized that retroactivity under a provincial statute – the Civil Code in the case at bar – applied to tax law. The Court held that the Civil Code determined when ownership was transferred for tax purposes; and since the Code stated that the surviving spouse’s ownership was retroactive, tax law had to follow suit.

The parallel with retroactive conditional obligations is obvious: If the Supreme Court recognized that civil law’s retroactivity applies to tax law in determining when ownership of property “passed” for the purposes of the Estate Tax Act, how could the Court disregard a condition’s retroactive effect when determining the time of disposition for purposes of the Income Tax Act?

This same reasoning was applied in Furfaro-Siconolfi v. The Queen[277]where the issue revolved around the moment at which a $30,000 gift under the marriage contract had been “transferred” to the spouse within the meaning of s. 160 I.T.A. Although the prenuptial agreement preceded the tax debt, the amount had been paid to the spouse three years later. The taxpayer argued that s. 160 could not apply because the amount had been “transferred” when the marriage contract was signed.

Pinard J. began by pointing out that the I.T.A. does not define the term “transfer.” He referred to the different ordinary and legal definitions of this word and held that the Act contemplated a simple transfer of ownership, where the beneficiary did not necessarily have to possess the property. He then considered Quebec civil law to determine the moment when ownership of the amount in question had been transferred:

It is by the operation of articles 777, 782, 787, 788, 795, 817, 819, 821, 822 and 1085 of the Civil Code of Lower Canada that the gift of $30,000 stipulated in the marriage contract here had the effect of transferring ownership of the money to the plaintiff when the contract was signed on September 2, 1977, a contract in fact followed by a marriage of the parties.[278]

However, in Riverin v. The Queen, [279] the Tax Court of Canada refused to recognize the retroactive effect of a giving in payment for the purpose of determining when there was a “transfer” within the meaning of s. 160 I.T.A. This case involved a loan granted by the taxpayer to a third party and secured by an immoveable hypothec. The debtor became insolvent and voluntarily gave the building in payment after receiving a 60-day notice. The Minister of Revenue, assessed the taxpayer for the debtor’s tax liability claiming a “transfer” of the immovable had taken place  within the meaning of section 160 I.T.A.

The taxpayer argued that the giving in payment was not a “transfer” within the meaning of s. 160 because the notice served under article 1040a C.C.L.C. forced the payment. This argument was rejected.[280] However, Archambault J. also examined the question as to when the transfer took place. The giving in payment under the Civil Code of Lower Canada had a retroactive effect to the date the loan was signed.[281] Archambault J. ruled that the transfer had occurred in February 1989 by virtue of a “legal fiction” under the C.C.L.C. Nonetheless, he held that retroactivity only applied to de jure , not de facto, matters and relied on Professor Diane Bruneau’s article on this issue.[282] In his opinion, the building was only transferred within the meaning of section 160 I.T.A. in 1990: 

I agree with Professor Bruneau’s analysis. In 1990, the immoveable passed from Mr. Demers’ patrimony to Mr. Riverin’s. This is a fact. In law, the effects of this transfer are deemed to be retroactive to the date of the agreement that created the giving in payment. For the purposes of subsection 160(1) of the Act, the transfer of the immoveable took place in May 1990.[283]

We are just as unable to agree with the Tax Court of Canada’s decision in this case as with Me Bruneau on this question, for the reasons previously stated.[284]

3.2.5    Conditions precedent and conditions subsequent

We now turn to the case law dealing with the question of when a “disposition” occurs where a condition precedent or a condition subsequent is involved

First of all, the case law has clearly established that where there is a true condition precedent within the meaning in Turney v. Zhilka, beneficial ownership cannot be transferred to the buyer until the condition occurs.[285]  As a result, there is no “disposition” prior to this time within the meaning of the Act.[286]  Most of the scholarly writing is in keeping with this point of view.[287]

It is important note that the test in Wardean Drilling is not applied to true conditions precedents at common law: it would be impossible to argue that a disposition occurred where the use, possession and risk were transferred if there was a condition precedent that had not yet materialized. As already noted, the second part of the test contemplated conditional sales agreements, where the seller reserves the title to secure payment of the sale price. This concept is more like a civil law instalment sale than a conditional obligation. We do not believe that the second part of the test in Wardean Drilling would apply to either a true condition precedent in common law or a suspensive condition in civil law.

The issue of the effect of conditions precedent on the moment when income is earned has arisen frequently. Paragraph 12(1)(b) I.T.A. states that the following sums are to be included when calculating business income for the year:

any amount receivable by the taxpayer in respect of property sold or services rendered in the course of a business in the year, notwithstanding that the amount or any part thereof is not due until a subsequent year (....)

les sommes à recevoir par le contribuable au titre de la vente de biens ou de la fourniture de services au cours de l’année, dans le cours des activités d’une entreprise, même si les sommes, en tout ou en partie, ne sont dues qu’au cours d’une année postérieure (...)(Emphasis added.)

Former paragraph 85B(1)(b) was essentially the same. The Exchequer Court established in M.N.R. v. John Colford Contracting Co.[288] that the creditor must have a clear, though not necessarily immediate, right to receive the amount in order for it to be “receivable”:

In the absence of a statutory definition to the contrary, I think it is not enough that the so-called recipient have a precarious right to receive the amount in question, but he must have a clearly legal, though not necessarily immediate, right to receive it.[289](Emphasis added.)

The judge then held that the law of the province where the contract was signed or performed must be examined in order to determine whether the amounts in question are receivable, that is to say, if the taxpayer had a clear right to the amounts. Thus, if the contract contains a condition precedent in common law, the taxpayer is only entitled to the amount once the condition occurs. As a result, the taxpayer need only include the amount in his income[290] when the condition precedent is fulfilled. The doctrine states likewise.[291]

The wording of the law supports the analogy, though not perfect, between the time when income is included and that of the disposition: disposition occurs in a sale once there is an “event entitling a taxpayer to proceeds of disposition”, the proceeds of disposition being the sale price. The sum in question must be included in income according to 12(1)(b) I.T.A. when the taxpayer has a clear, although not necessarily immediate, right to receive it. Me Brian J. Arnold wrote on this subject:

In summary, revenue from the sale of property is generally considered to be realized for income tax purposes when the vendor becomes legally entitled to receive and retain payment for the property. ...

Therefore, the time of recognition of business or property income under section 9 for an accrual basis taxpayer is virtually the same as the time of disposition of property for purposes of recapture of capital cost allowance and capital gains under sections 13 and 54 respectively.[292]

The case law and the doctrine[293] are unanimous as to a sale subject to a condition subsequent:  a disposition occurs as soon as the contract is signed, and the disposition is unaffected by the condition subsequent. There is a second disposition once the condition is fulfilled. A common law condition subsequent has no retroactive effect, so retroactivity is not an issue.

3.2.6    Retroactivity elsewhere in common law

3.2.6.1    Contractual retroactivity

Tax law treats both common and civil law contractual retroactivity (i.e. where the parties stipulate that the contract takes effect at a time other than the signing date) in the same way, same so this subject has already been discussed. We refer the reader to the corresponding section in this paper on other civil law concepts.[294]

3.2.6.2    Retroactivity under provincial legislation

We have seen that in certain cases, the courts are authorized by provincial statute to issue orders having a retroactive effect.[295] The tax consequences of such orders will now be examined. Hillis v. The Queen[296] is very representative of the uncertainty and conflict reflected in the cases between the various approaches to this subject.

Mr. Hillis died intestate on February 21, 1977, leaving behind a spouse and two sons. His spouse was entitled to $10,000 plus one-third of the remainder of the estate under the Saskatchewan Intestate Succession Act[297] and his two sons were entitled to the rest.

For unknown reasons, no concrete steps were taken toward settling the estate until 1979. The widow of the de cujus filed a motion under the Dependant’s Relief Act [298] in the Court of Queen’s Bench of Saskatchewan on November 29, 1979. Dependants of the deceased were entitled under this law to ask that the Court allocate a fair and equitable share of the estate for their support. The Court of Queen’s Bench issued an order on December 14, 1979, vesting all the deceased’s property in Mrs. Hillis.

Section 14 of the Dependant’s Relief Act was crucial in this case and reads as follows:

14. (1) Where an order is made under this Act, then for all purposes, including the purposes of enactments relating to succession duties, the will shall have effect, and shall be deemed to have had effect from the testator’s death, as if it had been executed, with such variations as are specified in the order, for the purpose of giving effect to the provision for maintenance made by the order.[299]

Thus, an order of the Court made under subsection 14(1) of the Dependant’s Relief Act is retroactive to the date of death.

It was necessary in the case at bar to determine whether Mrs. Hillis was entitled to benefit from subsection 70(6) I.T.A, as it then read, which provided for a rollover to the surviving spouse in the following circumstances:

...if the property can, within 15 months after the death of the taxpayer or such longer period as is reasonable in the circumstances, be established to have become vested indefeasibly in the spouse or trust, as the case may be, not later than 15 months after the death of the taxpayer, the following rules apply:...(Emphasis added.)

In order to be entitled to a rollover, the property had to be vested indefeasibly in Mrs. Hillis within 15 months after the death of her spouse. Furthermore, this event had to occur no later than 15 months of her spouse’s death or within a reasonable period in the circumstances. Only the first condition is of interest for this paper.

In the Federal Court of Appeal, Clément, Heald and Pratte JJ.A. wrote separate opinions leading to totally different conclusions. Only that part of their opinions on the share of the estate granted to Mrs. Hillis by order dated December 14, 1979 will be examined[300]:  was the order retroactive to the date of death for the purposes of subsection 70(6) I.T.A.?

Clément J. examined subsections 4(2) and 14(1) of the Dependant’s Relief Act and held as follows:

S. 4(2) of this statute deems the existence of a will with provisions such as those made by the Intestate Succession Act. But this fiction does not have force until an application for relief is made by or on behalf of a dependant, and then only for the purposes of the statute which cannot include federal income tax. And by s. 14(1) it is not until an order is made that the fictional will, modified as to the court may seem proper in the circumstances having regard to the purposes and directives of the statute, is to have effect. Then, the deemed will is deemed to have effect from the date of death of the intestate. These provisions are stated to be for all purposes, but obviously that can mean only for all provincial purposes. They cannot be taken to intrude fictions for provincial purposes into the interpretation and operation of the Act. The latter takes its operation in the realities of the circumstances subject only to such directives as it may itself prescribe.[301](Emphasis added.)

Clément J. categorically stated that the Income Tax Act only considers the reality of the circumstances, subject solely to the dispositions it may itself prescribe and that it is not subject to legal fictions born of provincial law. The principle of complementarity of law is clearly disregarded.

Pratte J. held that the property only vested when the order was issued because the devolution did not exist prior to the order’s pronouncement, even where it was admitted that the order was retroactive to the date of death. In other words, Mrs. Hillis was only entitled to the estate once the order was pronounced. Pratte J. explained his reasoning as follows:

...when did the estate become indefeasibly vested in Mrs. Hillis? In my view, ... when the order was pronounced since the effects of ... the Court order, in spite of [its] retroactivity, did not exist as long as ... the Court order was not pronounced. It is only when ... the Court order was pronounced that Mrs. Hillis became entitled to the whole of her husband’s estate with retroactive effect to the date of his death. If, therefore, ... the Court order had, as contended by the appellants, the effect of vesting the estate in Mrs. Hillis, that effect did not take place within 15 months after the death of Mr. Hillis.[302]

Heald J. immediately established that he intended to apply the principle that private law was complementary:

I agree with appellant’s counsel that the wording of subsection 70(6) of the Income Tax Act contemplates the disposition of property other than by will, as well as by will, since it deals with the transfer or distribution of property after the death of a taxpayer and “...as a consequence thereof...”. This wording, in my view, makes it clear, that Parliament contemplated that the law of the provinces in respect of the disposition of property on or after death, being matters relating to property and civil rights, would apply so as to control the application of subsection 70(6) in accordance with the law of the particular province concerned.[303](Emphasis added.)

The judge applied the provisions of the Dependant’s Relief Act, the provincial law in this instance. Heald J. held that under subsections 4(2) and 14(1) of the law, the order’s retroactive effect meant that the estate vested indefeasibly as of the date of death. Heald J. focussed primarily on the purpose of subsection 70(6) I.T.A. and held that Parliament had not intended to cast aside provincial law in this regard.

Although the three judges were divided as to the appeal, two judges out of three held that the retroactivity provided for in the provincial act did not apply to the I.T.A., albeit for very different reasons. Clément J. favoured the trend in the case law supporting a uniform application of the Act across the country regardless of provincial private law. At the other end of the spectrum, Heald J. held that provincial private law should be applied because it had not been cast aside by Parliament. Pratte J. took the so-called “ factual” stance: retroactivity cannot alter facts that occurred prior to the event giving rise to the retroactivity, in particular the “devolution” of property.[304]

3.2.7    Conclusion

This completes our analysis of the concept of disposition and retroactivity under the Civil Code and the common law as interpreted by the cases and scholars. We have tried to identify the major trends in the cases, although the cases have sometimes been contradictory.

We have seen that the cases have interpreted the concept of disposition broadly: although it is not exhaustive and does not limit the ordinary meaning of the word, the definition in the Act extends the concept of disposition to include concepts that would not normally be included. However, the definition in the Act states that a disposition only occurs in a sale where there is an event entitling the seller to the proceeds of disposition, which is, of course, the sale price.

Because the definition of disposition in the Act is not supposed to be exhaustive, it should be completed and interpreted be reference to provincial private law.

We have expressed our disagreement with both the decision in Olympia & York and the majority opinion in Construction Bérou on this subject. With all due respect, we believe that Wardean Drilling  test should not have been applied to a “disposition” or an “acquisition” in Quebec. For one thing, Wardean Drilling is a common law decision whereas the civil law is the private law required to complete the I.T.A. in Quebec. Moreover, this decision is based on the division of ownership in common law between the legal owner and the beneficial owner, a division that is not recognized in civil law.

However, the test developed in Wardean Drilling does not mean that there is a disposition or an acquisition at the moment beneficial ownership is transferred:  a disposition only occurs where the normal incidents of ownership are transferred to the buyer if the seller reserves ownership as security for the payment of the sale price, for example, in a conditional sales agreement or an instalment sale. We also believe that this test, were it to be adopted in Quebec, should only apply to true suspensive conditions because title is not reserved as security, but simply until the condition is fulfilled. Likewise, the seller does not reserve title as security in a resolutory condition: to the contrary, title is transferred to the buyer and will only be retroactively returned to the seller as though it had never been transferred if the condition occurs.

The test in Wardean Drilling is based on a broad interpretation of “disposition” encompassing something other than that which is explicitly provided in the definition under the Act. It must not be confused with the restricted definition found at subsection 248(1) “disposition” (e) I.T.A. which states that a disposition does not occur within the meaning of the Act where the legal ownership is transferred without the beneficial ownership; nowhere is the reverse mentioned (i.e. that there can be a disposition where only the beneficial ownership is transferred.) This last statement is not supported by the wording of the statute but by the decisions from the common law provinces.

Let us now return to the concept of disposition in the Act. We insist and repeat that the definition of disposition in the Act is not exhaustive and must be completed by private law. The complementary private law in Quebec is the civil law. As was suggested by the Supreme Court in Compagnie Immobilière BCN [305] and stated by Noël J. in Construction Bérou,[306] we believe the concept of “disposition” refers to civil law’s disposition of the ownership of property. This means that in the case of conditional obligations, one must resort to the Civil Code to determine when the seller disposed of his right of ownership in the thing.

We have seen that retroactivity provided for by provincial private law – be it retroactivity under the Civil Code, the common law or provincial statute – can be treated in three different ways for purposes of the I.T.A. The first approach rejects the application of retroactivity on behalf of a uniform application of the Act across the country. Clément J. expressed this somewhat marginal approach in Hillis.

The second approach also rejects the application of retroactivity, but for a different reason: retroactivity under the provincial law cannot alter the past nor affect third-party rights acquired during this time, in particular those of the Agency. This is the opinion of Pratte J. in Hillis: property cannot have “vested” before the order was issued because it “vested” by the order despite the order’s retroactive effect. The judgments in Alepin,[307] Larose[308] and Riverin[309] reflect this trend; it is unnecessary to repeat the comments already made on this subject.

The third trend in the case law would apply retroactivity for the purposes of the I.T.A. Where provincial law provides that retroactivity will apply to a specific event or private law concept, tax law must take this retroactivity into account when ascribing consequences to this private law concept. The Supreme Court recognized the retroactive effect of provincial law in Faure,[310] as did the Federal Court of Appeal in Perini Estate.[311] The Federal Court Trial Division decided likewise in Furfaro-Siconolfi [312] as did Heald J. in Hillis.

We feel this position accurately reflects the law. We can not agree with the second approach, i.e. that retroactivity is a legal fiction and that the I.T.A. only applies to the reality of the facts. In our view, retroactive legal consequences imposed by provincial statute are no less “real” than other legal concepts such as ownership. We agree with Me Joel Nitikman when he states:

In short, it is agreed that the Act focuses on “reality”, but it is submitted that that “reality” is found in provincial/common law. A provincial statute or a rule of common law which imposes retroactive legal consequences on persons is no more, but no less, real, than a statute or a rule which imposes those consequences prospectively. The Act should recognize both equally.[313]

The same author responds as follows to the argument that retroactivity is not enforceable against the Agency:

If a retroactive agreement is binding as between the contracting parties in a provincial court then it is binding for tax purposes, because the tax system is an accessory to the provincial law system. The fact that the Minister was not a party to the amending agreement is completely irrelevant; he was not a party to the original agreement but there is no doubt he is bound by it as far as the tax consequences arising from it are concerned. [314](Emphasis added.)

3.3 Canada Customs and Revenue Agency’s Administrative position

3.3.1    The concept of disposition and conditional obligations

Let us review the Agency’s position on disposition in general and on conditional obligations in particular. The different positions taken by the Agency will be explained in chronological order so that we can see how its position has evolved.

The first administrative policy issued on the concepts of disposition and conditional obligations was published in Interpretation Bulletin IT-170R.[315]

The Bulletin begins by restating, at paragraph 2, the principle from Victory Hotels that: “the date of disposition of capital property sold occurs at the time that the vendor is “entitled to...the sale price.” ... the date of disposition is given a somewhat restricted meaning when a disposition of capital property involves a sale.”

Paragraph 4 of the Bulletin contains the following comments about paragraph (e) of the definition of “disposition” in 248(1) I.T.A., which states that there is no disposition if the beneficial ownership is not transferred:

Subparagraph 54(c)(v) makes it clear ... that the Act is interested only in dispositions that involve a change in beneficial ownership (unless the contrary is expressly stated). This is also the Department's view in respect of dispositions of depreciable property described in paragraph 13(21)(c) and the sale of trading assets under paragraph 12(1)(b). A transaction that can be described as a "sale" is therefore disregarded for purposes of this bulletin if there is no concurrent change in beneficial ownership. Such transactions will usually involve a "purchaser" who can be described as an agent, nominee, trustee or prête-nom corporation of a "vendor" who basically retains the right to deal with the property as though it were his own.... (Emphasis added.)

One can see that the Minister of Revenue is saying that the concept of disposition is equivalent to the transfer of beneficial ownership.

The Minister commented on the acquisition of depreciable property in another Interpretation Bulletin – IT-50R, which has since been replaced by IT-285R2:

17. Generally, a taxpayer will be considered to have acquired a depreciable property at the earlier of:

  • (a) the date on which title to it is obtained, and
  • (b) the date on which the taxpayer has all the incidents of ownership such as possession, use, and risk, even though legal title remains in the vendor as security for the purchase price (as is commercial practice under a conditional sale agreement).

In order that the cost of an asset may fall within a specified class, the purchaser must have a current ownership right in the asset itself and not merely rights under a contract, of which the asset is the subject, to acquire it in the future.

18. In determining whether or not depreciable property is acquired by a taxpayer, the legal relationship between the vendor and the purchaser of the property should be reviewed. For example, where chattels are being acquired, the relevant sale of goods legislation would be applicable. Each of the provinces (other than Quebec) has a Sale of Goods Act pertaining to sales of chattels laying down substantially the same rules for the ownership rights to assets bought and sold. The basic rule is that property in respect of specific assets passes, and is therefore acquired by the purchaser, at the time when the parties to the contract intend it to pass as evidenced by the terms of the contract, the conduct of the parties and any other circumstances.

19. If, however, the intention of the parties is not evidenced as discussed above, the following rules apply to determine when property is to pass:

  • (a) if there is an unconditional contract for the sale of a specific asset in a deliverable state, property will pass to the purchaser when the contract is made, and it is immaterial whether the time of payment or delivery or both are postponed;
  • (b) if there is a contract for the sale of a specific asset and:
    • (i) the seller is bound to do something to the asset to put it into a deliverable state, or
    • (ii) the asset is in a deliverable state, but the seller must weigh, measure, test or do some other act or thing to ascertain the price, then property does not pass until the seller has satisfied those conditions and the purchaser has notice thereof.[316] (Emphasis added.)

Paragraph 17 is worded very similarly to the test in Wardean Drilling. Paragraph 18 states that the legal relationship created by the applicable provincial private law between the parties must be examined in order to determine if property has been acquired. The Bulletin mentions that specific statutes govern the sale of moveable property in all the provinces except for Quebec and that the moment of acquisition is determined according to these statutes. The rules that exist in the other provinces are explained at the end of paragraph 18 and in paragraph 19. The Bulletin gives the impression that in Quebec, one must refer to the Civil Code rules.

The way in which the Agency treats suspensive and resolutory conditions is explained in Interpretation Bulletin IT-170R, which states:

5. ...it is the Department's view that the sale price of any property sold is brought into account for income tax purposes when the vendor has an absolute but not necessarily immediate right to be paid. As long as a "condition precedent" [“condition suspensive” in the French version] remains unsatisfied, a vendor does not have an absolute right to be paid. However, the fact that an event subsequent to the completion of a sale restores the ownership of the property involved to the vendor or adjusts the sale price does not alter the fact that the vendor was at a particular time entitled to the sale price and therefore disposed of the property for tax purposes at that time. Similarly, the fact that a contract of sale is subject to ratification is of no consequence in determining a date of disposition unless it is made a condition precedent of the agreement.

6. A "condition precedent" [“condition suspensive” in the French version]is an event (beyond the direct control of the vendor) that suspends completion of the contract until the condition is met or waived and that could cancel the contract "ab initio " if it is not met or waived. (Emphasis added.)

The Agency takes the position that in the case of a suspensive condition there is a disposition once the condition is fulfilled. There is an immediate disposition in the case of a resolutory condition that is not retroactively cancelled if the condition occurs. It is obvious from the English version of the Bulletin that when the Bulletin refers to a suspensive condition, it is contemplating a common law condition precedent.

In addition, the Bulletin does not mention the retroactive effect of a fulfilled suspensive condition. However, it seems to imply that if the suspensive condition was not fulfilled, there will be no tax consequences because the contract will be cancelled ab initio.

Mention must also be made of paragraph 19 of Bulletin IT-170R, which states the following with respect to resolutory conditions:

19. Many agreements contemplate the reacquisition by the vendor of property that has been sold upon the happening of a specified event, the failure of a specified event to occur or a specified default of the purchaser. Where a reacquisition of beneficial ownership occurs by reason of the purchaser's failure to pay all or any part of an amount owing, section 79 provides rules to determine the tax consequences for both vendor and purchaser. Although the Act provides no specific rules where reacquisition occurs in situations to which section 79 does not apply, it is clear that such an occurrence does not retroactively nullify the effects of the original disposition for income tax purposes even if the agreement restores the vendor and purchaser to their relative positions before the sale took place. (Emphasis added.)

At the time, therefore, the retroactive effect of resolutory condition was not recognized by the Agency.

At the 1981 conference of the Association québécoise de planification fiscale et successorale,[317] Revenue Canada representatives were asked the following question:

[Unofficial Translation] Question 17: TIME OF DISPOSITION OF PROPERTY

The time of disposition of a property by a taxpayer is an important factor in calculating capital gain and recapture of capital cost allowance. For the purchaser, whether a CCA can be claimed in respect of the property depends on the time of disposition. The word “disposition” is not defined in the Income Tax Act. However, certain specific situations are included in this expression by subsection 54(c) I.T.A. Given the relative shortage of details from Parliament, there may be circumstances in which the time of disposition is subject to interpretation. This is the case where there has been an assignment of the incidents of the right of ownership (possession, use, risk) of a property, but the transfer of ownership is suspended pending the fufilment of a condition. When can one say that the disposition of the property occurs in such a case?

The Department’s position

It is the Department’s view that, for tax purposes, and thus for the purposes of section 54 of the Act, the sale price of a property must only be taken into consideration when the vendor has acquired the absolute right to be paid. So long as a suspensive condition [“condition suspensive” in the original French version] has not been met, the seller does not have the absolute right to be paid even if the purchaser has taken possession of the property.[318](Emphasis added.)

Thus, the Department was taking an unequivocal position:  where a suspensive condition exists in civil law, the disposition does not occur before the condition is fulfilled. However, the Department did not express an opinion on the issue of the retroactivity once a condition has been fulfilled, though the answer given implicitly suggests that it will not be given effect.

At the 1981 conference of the Canadian Tax Foundation[319], Revenue Canada expressed its intention to apply the Olympia & York decision:

[Unofficial Translation]

Q. 54  Disposition


(2)  Has the Department of National Revenue accepted that a disposition may take place when the incidents of ownership (possession, use and risk) have been transferred, even though legal title may not have been transferred? In this regard, will the Department apply Olympia & York Developments Ltd. v. The Queen?…

(2)  Generally, if all the incidents of ownership, i.e. possession, use, and risk, are surrendered and the taxpayer becomes entitled to proceeds of disposition, it is the Department’s view that a disposition has taken place whether or not legal title has been transferred. The Olympia & York decision supports us in this regard.[320]

At a 1983 round table held at the AQPFS annual conference, Revenue Canada was asked to answer the following question:

[Unofficial Translation] QUESTION 3: DISPOSITION WITH A SUSPENSIVE OR RESOLUTORY CONDITION

Where a sale is made subject to a resolutory condition, the seller delivers title to, and possession of, the property sold to the buyer. However, if the condition provided for in the contract of sale is fulfilled, the Civil Code of the Province of Quebec provides that the sale is deemed never to have occurred.

A sale can also be made subject to a suspensive condition. In such a case, ownership remains in the hands of the seller until such time as the condition is fulfilled and the buyer may or may not be given possession. However, where this condition is fulfilled, the buyer is deemed to have been owner of the thing sold since the date the contract of sale is signed, not the date the condition was fulfilled. For income tax purposes, could you confirm that no disposition is deemed to have occurred further to a sale with a resolutory condition if this condition is fulfilled? Similarly, could you confirm, for income tax purposes, that a disposition, further to a sale with a suspensive condition, is considered to have occurred when the contract of sale is signed, and not when the condition is fulfilled?

POSITION OF REVENUE CANADA TAXATION

The tax consequences of the fulfillment of a condition attached to a suspensive or resolutory clause will depend on the specific facts and circumstances of each situation.

First, it is important to stress that suspensive or resolutory conditions only have an effect on the contract of sale, and therefore do not apply to the events occurring subsequent to the date of the contract of sale. Consequently, the fulfilment of these conditions will not, for purposes of the I.T.A., cancel the transactions or acts carried out by the parties to the contract between the date of signature and the date of fulfillment of the suspensive or resolutory condition.

As for the effect of the suspensive or resolutory conditions applying to the sale itself, the Department generally [applies] the following procedure, based on the circumstances.

In the case of a conditional sale where possession or enjoyment of the property is transferred before the condition is fulfilled (condition “precedent” under customary law) the Department’s position is that no sale has occurred so long as the condition has not been fulfilled.

In the case of a suspensive condition, i.e. one that must be fulfilled to confirm the sale, where there is transfer of possession and enjoyment, so that the purchaser obtains the “beneficial ownership” as soon as the contract is signed, the Department is of the opinion that the sale took place at that time whether or not the condition is fulfilled.

However, where the seller reacquires property due to the fulfillment of a resolutory condition the effect of which is to cancel a sale when it is fulfilled, or due to a suspensive condition that is not fulfilled, the Department’s position is that there has been a disposition to the seller under paragraph 54(c)(i). Although legally, the right of ownership may be cancelled retroactively, the seller’s reacquisition of the “beneficial ownership” only occurs at that time. Consequently, there will be a disposition for purposes of the I.T.A. and section 79 may apply.[321]

The wording of the answer is very ambiguous: the Department talks about “condition precedent”, “customary law”, “suspensive conditions” and “beneficial ownership”, without specifying whether it is referring to civil law or common law concepts. Since the question was posed at a conference on Quebec tax law, the answer should apply to civil law.

Our understanding of the administrative position is as follows. Where there is a suspensive condition in civil law and the purchaser is taking immediate possession of the property, thereby obtaining “beneficial ownership,” the Department’s position is that there has been a “sale”, whether or not the condition is subsequently fulfilled. It is difficult to assess whether the Department is referring solely to the transfer of possession, or whether it requires that all the normal incidents of the right of ownership be transferred. Since this position seems to be based on Wardean Drilling, it is likely that possession, use and risk, at the very least, need to have be transferred.

In addition, with respect to conditions precedent, we have seen that at common law, beneficial ownership cannot be transferred to the purchaser as long as an unfulfilled true condition precedent exists.[322]  Thus, in tax law, there cannot be a disposition before a condition precedent is fulfilled.  The Bulletin confirms this implicitly, without clearly articulating it.

Conversely, in a sale subject to a suspensive condition without immediate transfer of possession, the “sale” takes place only once the suspensive condition has been fulfilled. It should be noted that the retroactive effect of the suspensive condition in civil law is not always mentioned.

The Department of Revenue has therefore changed its position with respect to suspensive conditions: contrary to what was stated in 1981, it will take account of the transfer of possession (and possibly other ordinary attributes of ownership) to determine whether a sale subject to a suspensive condition gives rise to an immediate disposition.  It appears that this change of position follows the decision of the Federal Court in Olympia & York.

It is also interesting to note that the Department uses the term “sale” rather than “disposition”: in its view, whether or not there has been a “sale” will depend on the circumstances.  Thus, it seems that the tax authorities themselves are using the terms “sale” and “disposition” as synonyms, which supports our position that disposition is tied to transfer of ownership.[323]

In addition, one can clearly infer from the response that if the suspensive condition is not fulfilled, the Department, where possession has been transferred to the purchaser, deems the return of the property to the seller to be a second disposition for tax purposes, since there is a second transfer of beneficial ownership. This is also the case if a resolutory condition is fulfilled.

 In 1987, during the ACEF conference, the Department was asked to comment on the following question:

Q. 70 Transfer of Property: Timing of Income

A taxpayer purchases the assets of a business from an arm's-length vendor and the completion of the transaction is conditional upon receipt of approval from Investment Canada. The purchaser and vendor agree that the transaction will be regarded as effective as at a previous date, and the profits from the operation of the business by the purchaser between that effective date and the date of completion will be regarded as profits of the purchaser.

Does the department consider the reporting of such income by the purchaser appropriate for tax purposes, notwithstanding that the assets are not transferred until the date of completion?

Department’s Response

As stated in paragraphs 5 and 6 of Interpretation Bulletin IT-170R, where the transfer of property is subject to a true condition precedent [“condition suspensive” in the French version], the disposition will not occur until the condition precedent [“condition suspensive” in the French version]  is satisfied. Accordingly, for the purposes of the Act, the transfer of the business will not occur until approval of Investment Canada is received. Any agreement between the taxpayer and the vendor purporting to give retroactive effect to the transfer is not effective for tax purposes. Any income from the operation of the business prior to the transfer will be income of the vendor. [324] (Emphasis added.)

At first glance, the Department seems to be returning to its previous position, as expressed in 1981 at the AQPFS Conference, which is that civil law suspensive conditions do not give rise to a disposition prior to their fulfillment, independently of the transfer of the possession of the property.  However, we believe the Department’s response pertained only to conditions precedent of common law.[325]   There are two reasons for this: first, this response was given during a conference of the Canadian Tax Foundation, a Canada-wide association.  Moreover, the English version of the response uses the term “true condition precedent.”[326]  Yet, as we have seen, a condition precedent prevents disposition so long as it is not fulfilled, given the impossibility of transferring beneficial ownership.  Secondly, the question mentions that “the assets are not transferred until the date of completion” – a reference to the closing date.  Thus, this question did not apply to civil law suspensive conditions where possession is transferred immediately, nor did it apply to the question whether suspensive conditions are retroactive.  Rather, the gist of the question is whether the Department should recognize the effective date of the transaction agreed upon by the parties.  We shall return to this aspect later.

The following question was asked during a round table discussion at the 1989 conference of the Association de planification fiscale et financière conference:[327]

“1.29. —SALE SUBJECT TO A SUSPENSIVE CONDITION

According to civil law, a taxpayer who sells a building subject to a suspensive condition (for example, by retaining ownership until the complete payment of the price) remains the owner of the property.  In the event of the non-fulfillment of the condition, the obligations of the parties to each other are cancelled.  From a taxation point of view, Revenue Canada will treat a sale subject to a suspensive condition like a disposition.

Consider the case where a Quebec taxpayer makes such a sale in exchange for the balance of a sale price, payable over five years.  Following a default in payment in the first year, the taxpayer retakes possession of the asset.  The taxpayer had claimed a reserve under subparagraph 40(1)(a)(iii) of the Income Tax Act (the Act) for the taxation year of the sale.

In circumstances such as these, is it the policy of Revenue Canada to apply section 79 of the Act, even if, in civil law, the Quebec taxpayer has not acquired or reacquired the beneficial ownership or the ownership of the building following a default in payment?

Response of Revenue Canada

The Department deems such a sale subject to a suspensive condition to be a disposition for the purposes of the Act.

If the property is returned following a default in payment, the Department considers that there has been a second disposition for the purposes of the Act, and that section 79 of this Act may be applicable.

We refer you in this respect to the responses given at the 1981 CTF round table discussions (Q. 54(2)) and the 1983 A.Q.P.F.S. conference (Q.3).” [328]  (Emphasis added.)

Thus, the Department reiterated the position it adopted during the 1983 AQPFS Conference: there is a disposition when possession is transferred even if the suspensive condition is not fulfilled. A second disposition occurs when the property is returned to the seller if the condition fails.

This time, the question asked during the APFF congress clearly contemplated Québec civil law. In addition, it implied that there had been a transfer of possession to the buyer pendente conditione. It should be noted, however, that this question concerned an instalment sale, not a true suspensive condition, and Venne had not yet been decided at the time.

The Department confirmed its position in a 1989 technical interpretation on resolutory conditions within the context of a sale with a right to repurchase: there is a second disposition when the option to purchase is exercised, even if the sale is retroactively cancelled in civil law.[329]

Then in 1991, the Department, in response to a question asked during a CTF round table congress, once again took the position adopted in 1987 regarding the contractual effective date of a transaction. The question and response read as follows:

« Q.41 Répartition du revenu lorsqu’une condition suspensive existe

En réponse à la question 70 de la table ronde de 1987, le Ministère a déclaré que, lorsqu’un bien est vendu, les revenus qui proviennent du bien entre la date de signature de l’offre de vente et d’achat et la date de transfert de la propriété appartiennent au vendeur s’il existe une condition suspensive. Si le vendeur et l’acheteur concluent une entente exécutoire par laquelle le vendeur est constitué agent de l’acheteur pour la période concernée, le revenu appartiendra-t-il à l’acheteur? La réponse serait-elle la même s’il n’existait aucune condition suspensive?

Position du Ministère

La position du Ministère demeure la même que celle donnée en 1987. Puisque la disposition n’a pas lieu, aux fins de l’impôt, tant que la condition suspensive n’est pas remplie, tout revenu que le bien génère avant le transfert n’appartiendra pas à l’acheteur, malgré l’existence d’une entente par laquelle le vendeur est constitué agent de l’acheteur pour la période concernée. Aucune convention visant à donner au transfert un effet rétroactif n’est valable aux fins de l’impôt. La date de disposition d’un bien vendu est la date où la propriété effective du bien doit passer à l’acheteur et le moment où le vendeur a un droit absolu, quoique pas nécessairement immédiat, de se faire payer.

Pourvu que le vendeur ait droit à son paiement et que la propriété effective ait été transférée, tout revenu que le bien produit entre la date de signature de l’offre de vente et d’achat et la date de transfert de la propriété doivent être reconnus par l’acheteur.  »[330](Emphasis added.)

The Department’s response suggests that where there is a suspensive condition, a transfer of beneficial ownership alone will not lead to a disposition: the seller must also have an absolute right to the sale price, i.e., the suspensive condition must be fulfilled.

If so, this would contradict the 1983 position, reiterated in 1989, which pertains to civil law suspensive conditions where possession is immediately transferred to the buyer. However, we believe that the Department’s response only contemplated common law conditions precedent, as it did in the 1987 CTF conference. The English version supports this view:

...As the disposition will not occur for tax purposes until the condition precedent is satisfied, any income arising before the transfer will not belong to the purchaser regardless of an agreement appointing the vendor as the purchaser’s agent for this period....The date of disposition of property sold is the date on which beneficial ownership is intended to pass to the purchaser and the time at which the vendor has an absolute but not necessarily immediate right to be paid.

Provided that the vendor is entitled to payment and beneficial ownership has been transferred, any income earned in the period between the effective date and the closing date must be recognized by the purchaser.[331](Emphasis added.)

Furthermore, it seems the two conditions must occur before a disposition can take place because the beneficial ownership cannot be transferred in common law until the condition precedent is fulfilled. The Department simply wanted to remind us that in order for there to be a disposition, it is not enough that the condition precedent be fulfilled. The beneficial ownership must also be transferred either at the same time or subsequently, as implied in paragraph (e) in the definition of “disposition” in subsection 248(1) I.T.A.

Practitioners would nonetheless find it helpful if the Agency could clarify its position as to whether it contemplates civil law suspensive conditions or conditions precedent, and whether these two concepts are to be treated differently in tax law.

Recently, both Revenue Canada and the Department of Finance were asked the following question during a 1998 APFF congress. The reply is interesting:

Question 4.8- Effect of resolutory and suspensive clauses

Property transfers are subject to the principles of civil law and tax law. It has been established that the principles of tax law are subordinate to those of civil law.

Articles 1507 and 1750 of the Quebec Civil Code set out the rules governing contracts containing suspensive conditions and resolutory conditions. A suspensive condition can be defined as a condition which suspends "the effects of the contract" while a resolutory condition can be defined as a condition which suspends the "cancellation of the contract". This resolutory condition, when satisfied, cancels the sale retroactively.

From a tax standpoint, the sale produces its full effects as soon as it is concluded and the vendor is immediately entitled to the sale price. Hence, disposition occurs at that moment. In the event that the transaction is not completed by reason of a resolutory condition, Revenu Québec recognizes the retroactivity from a tax standpoint and does not apply sections 484 to 484.13 of the Quebec Taxation Act.

According to paragraphs 5 and 17 of Interpretation Bulletin IT-170R, Revenue Canada does not recognize the retroactivity of the cancellation from a tax standpoint and applies sections 79 and 79.1 I.T.A. depending on the case.

Does the Department of Finance of Canada recognize this position?  Is it willing to review itin light of the rules of the Quebec Civil Code?

Does Revenue Canada still maintain this position?  Is it willing to review it in view of the Quebec Civil Code which governs transactions effected in Quebec?

Revenue Canada’s reply

There are two legal principles, which are in conflict in this question. As you point out, tax law applies to the effects produced by civil law. However, the Department must, in computing the taxes payable for a taxation year, operate on the basis of the facts as they exist at the end of a taxation year.

In our opinion, recognition of the retroactive effect of the cancellation of a sale is not compatible with the Act read as a whole. The Act is not designed to allow the application of new facts that occur during a taxation year to a prior taxation year. To this end, it does not allow reassessments in respect of statute-barred taxation years in order to apply retroactivity.

Moreover, in Clément Alepin (79 DTC 5259) and Michel Larose (92 DTC 2045), the courts refused to apply, for the purposes of the Act, the retroactivity provided for in civil law. In these two cases, the honourable justices stressed that the rights of the Department could not be affected following the cancellation of sale contracts.

Finance Canada’s reply

The Department of Finance agrees that the tax legislation must take the relevant provinciallaw into account. However, certain basic principles of tax law, such as those applicable to retroactivity, may not be entirely compatible with certain effects of provincial law. This is also the case of partnerships, which, regardless of their attributes, rights and obligations under provincial law, are generally not recognized in tax law.

We wish to examine the analysis of Revenue Canada, Revenu Québec and Justice Canada on this question in greater detail before concluding that Revenue Canada's position is notappropriate in the circumstances. However, we share Revenue Canada's concerns about certain practical aspects, such as the restrictions imposed in the case of statute-barred years.[332] (Emphasis added.)

We can see that the two departments are primarily concerned that the retroactive effect of conditional obligations will make it difficult to amend tax declarations filed in previous years in order to take into account a retroactive effect caused by an event in a subsequent year. This preoccupation applies mainly to time-barred years.

We have already expressed our doubts regarding the general application of Alepin and Larose.[333]

To conclude, the current position of the Agency on conditional obligations can be summed up as follows:

3.3.2    Contractual retroactivity

Interpretation Bulletin IT-170R adopted a more liberal position on contractual retroactivity. The date of entitlement agreed upon by the parties is used for tax law purposes:

7. Formal agreements of purchase and sale are frequently explicit as to the date of exchange and, unless circumstances indicate that a specified date was changed or was not the true intent of both parties, the date so specified is presumed to be the date of entitlement. ....[334]

However, the position expressed in the Bulletin was contradicted by the position adopted during the 1987 CTF congress[335] and reiterated during the 1991 CTF congress.[336] It states that any business income earned prior to the closing date will be attributed to the seller unless the seller had an absolute right to the sale price, i.e., there is no condition precedent and the beneficial ownership has been transferred.

The Department indicated in 1994 that in certain situations, a disposition could occur at a date prior to the closing date if all the parties to the contract agreed and if it did not result in a significant tax benefit:

You have described a hypothetical situation wherein:

  • the purchase and sale agreement stipulates both an effective and a closing date;
  • the terms of the agreement are such that the beneficial ownership and assumption of liabilities relating to these properties pass to the purchaser on the effective date, except for a few minor liabilities, which pass on the closing date; and
  • there are no conditions precedent to be met under the purchase and sale agreement and all that is required prior to closing is the usual due diligence and completion of appropriate documentation....

In this situation, the transfer is not legally effective until the closing date, and the vendor is legally liable to report the income between the effective date and the closing date. However, there have been instances where the Department has administratively accepted that the transfer occurred on the effective date where:

  • both parties to the transaction agree that the effective date should be used;
  • no significant tax benefit arises from the use of this date.[337](Emphasis added.)

This technical interpretation seems to impose additional conditions on, rather than increasing, the number of cases in which the Department will recognize that the contract took effect at a date agreed upon by the parties. In 1987, the Minister admitted during a CTF congress that the income could be attributed to the buyer as soon as the seller had an absolute right to the sale price and where there had been a transfer of the beneficial ownership. Thus, the Minister not only reiterated these two conditions, but he added two new ones, i.e. that the parties to the transaction agree to use the effective date (which appears self-evident in the case of a contract), and that this did not create a tax benefit. This last requirement does not appear to have any basis in law, in our opinion.

In fact, considering the cases reviewed above, we do not believe that the Agency’s position on the application of a contractual effective date to tax matters is well founded. The subsequent agreement only serves to confirm that the parties agreed on the essential elements of the contract on the date in question and that it was valid as of that moment.

3.3.3    Paragraph 248(3)(f) I.T.A.

As seen previously, Parliament enacted subsection 248(3) I.T.A.[338] in order to equate certain civil law concepts with common law’s beneficial ownership. The majority in Construction Bérou interpreted this subsection in a way that includes the normal incidents of ownership, such as possession, use and risk, in the concept of beneficial ownership for the purposes of applying the subsection to Quebec.[339]

However, the Agency maintains that the Court of Appeal’s interpretation in Construction Bérou applies to the former version of subsection 248(3) I.T.A. (which was amended as of 1991). The Agency expresses its opinion as follows:

[UNOFFICIAL TRANSLATION] We believe that, although the wording of paragraph 248(3)(f) is essentially similar to the Act at that time, it is now different, and cannot be interpreted to mean that the incidents of ownership, such as possession, use and risk, can give rise to a right of ownership. Rather, the current version of paragraph 248(3)(f) states that a person must first fully own property before that person is deemed to hold the “beneficial ownership” for the purposes of the Act....

The Agency is therefore of the view that the decision of the Federal Court of Appeal in Construction Bérou Inc. can be limited to similar cases that arose in tax years prior to 1991. This is in compliance with the Court’s interpretation of subsection 248(3) as it was previously worded.[340](Emphasis added.)

Therefore, the Agency’s interpretation is that paragraph 248(3)(f) could not support a transfer of the beneficial ownership in Quebec without the legal ownership; the beneficial ownership would be limited to the concepts listed in the subsection, i.e. ownership, the rights of a lessee under an emphyteutic lease and the rights of a beneficiary of a trust.

For the reason expressed previously, we agree with this interpretation of paragraph 248(3)(f) I.T.A., but do not believe that the situation changed following the 1991 statutory amendments. We believe the Agency did not agree with the decision of the majority of the Federal Court of Appeal in Construction Bérou and is trying to avoid its application by introducing subtle distinctions.

3.4 Conditional obligations in Quebec Tax Law

Although this paper is about the harmonization of the Income Tax Act with the Civil Code of Québec, it could be interesting, for the purpose of comparison, to analyze the way in which conditional obligations are treated under Quebec statute law and the Ministère du Revenu du Quebec .

The relevant provisions of the Taxation Act[341] and Regulation respecting the Taxation Act[342]are as follows:

248. (1) Disposition of property. — For the purposes of this Title, the disposition of property includes, except as expressly otherwise provided:

a) any transaction or event entitling to proceeds of disposition of property;...

2) Restriction. — A disposition of property does not include however:...

d) any other transaction provided in the regulations.”

248R1. For the purposes of section 248 of the Act, any transfer of a property governed by a common law jurisdiction which does not entail a change in the beneficial ownership thereof is not a disposition of property.

Similarly, any transfer of a property governed by civil law which does not entail a change in the right of the person who has the full ownership thereof, although such property be subject to a servitude, or in the right of the usufructuary, the emphyteutic lessee, an institute in a substitution or a beneficiary in a trust, is not a disposition of property.”

251. Proceeds of disposition of property.

The proceeds of disposition of property include, for the purposes of this Title, the same elements as the proceeds of disposition of property referred to in paragraph f of section 93...”

93. ... f) “proceeds of disposition”. — “proceeds of disposition” of property includes:

i. the sale price of property disposed of;

Besides a few minor differences, the provisions of the Taxation Act are very similar to those found in the I.T.A. One of the differences is the use of the term “aliénation”, found only in the French version, rather than “disposition”. Yet, just like the definition of “disposition” in the I.T.A., the definition of “aliénation in the T.A. is not exhaustive. We should therefore refer to its everyday meaning to define the term:

L’“aliénation” n’étant pas définie par la loi, on doit y appliquer le sens général de ce mot donné par Le Petit Robert comme étant “transmission qu’une personne fait d’une propriété ou d’un droit, à titre gratuit ou onéreux”. [343]

Given the definition found in the Taxation Act, it is clear that a sale constitutes a disposition:

[TRANSLATION] It is surprising to find that a detailed section like 248 T. A., whose role is to define the disposition of property for the purposes of Title IV of the Act, does not allude to the concept of sale. This section provides, however, that the disposition of property includes a transaction or event that triggers entitlement to a sale price of the property disposed of. This is the obvious conclusion when subsection 248(1)(a), section 251 and paragraph 93(f)(i) T.A. are read together.

For the purposes of both the Income Tax Act and the Taxation Act, it is best to refer to the Civil Code of Québec to determine the exact nature of a sale and at what moment the right to the sale price arises.[344](Emphasis added.)

Furthermore, we can see that Section 248R1 of the Regulation contains a specific rule for property governed by common law, and another one for property governed by the Civil Code. In fact, there are references to “beneficial ownership of the property”[345] for property governed by common law; however, for property governed by civil law, there are references to the concepts of full ownership, usufruct, emphyteutic lessee, institutes in a substitution, and beneficiaries of a trust. These concepts are essentially the same as those found in paragraph 248(3)(f) I.T.A.

Sections 79 and 79.1 of the I.T.A. also have an equivalent under the Taxation Act: sections 484 et seq. Section 484.1 of the Taxation Act, which corresponds to subsection 79(2) I.T.A., reads as follows:

484.1 Surrender of property — For the purposes of this subdivision, a property is surrendered at any time by a person to another person where the beneficial ownership of the property [“la propriété à titre bénéficiaire”] is acquired or reacquired at that time from the person by the other person and the acquisition or reacquisition of the property was in consequence of the person's failure to pay all or part of one or more specified amounts of a debt owed by the person to the other person immediately before that time. (Emphasis added.)

This section was added in 1996,[346] and generally applies to property acquired or reacquired after February 21, 1994. Before being amended, section 484 provided the following:

484. Where a creditor has acquired or reacquired, at any time in a taxation year, the possession as proprietor or the full ownership of property [“la possession à titre de propriétaire ou la propriété absolue”] in consequence of the debtor’s total or partial failure to pay the creditor’s claim, the following rules apply: ... (Emphasis added.)

Does replacing the expression “possession as proprietor or the full ownership of property” [“la possession à titre de propriétaire ou la propriété aboslue”] with “beneficial ownership of the property” [“la propriété à titre bénéficiaire”] change the meaning of this provision? The term “beneficial ownership of the property” [“la propriété à titre bénéficiaire”] is not defined in the Taxation Act. We saw above, however, that these terms have replaced the expression “beneficial ownership” in Section 248R1 of the Regulation respecting the Taxation Act.

We obtained confirmation from an authorized representative of the Ministère du Revenu du Québec that the administrative policy regarding the application of sections 484 et seq. remained unchanged after the expression “beneficial ownership of the property”/ [“propriété à titre bénéficiaire”] was introduced in 1994  and this leads us to believe that the new expression reflects the same reality as the prior expression, “possession as proprietor”/ [“possession à titre de propriétaire ”].

Interpretation Bulletin IMP. 484-2/R1 gives the position of the Ministère du Revenu du Québec on the effects of the dissolution of a contract. Here is an excerpt [from the official English translation]:

  • 4. Where dissolution (“résolution”) is involved, section 484 of the Taxation Act (the “Act”) is not applicable since the creditor has not acquired or reacquired the possession as proprietor or the full ownership of property. Indeed, his demand for dissolution (“résolution”) leads to annulment of the contract with respect to both the past and the future, such that he is deemed never to have ceased to be the owner of the property

  • […]
  • 6. Moreover, the seller may be reimbursed for the tax paid in respect of the capital gain or the income from a business, as the case may be, or in respect of the recaptured depreciation relating to the sale. As opposed to this, if the sale allowed him rather to deduct a terminal loss, he must pay the tax in respect of this deduction for the taxation year of the sale.

    Where the seller has financed, in whole or in part, the payment of the sale price, he cannot be reimbursed for the tax paid in respect of the interest paid by the purchaser though. Indeed, dissolution of the sales leads to resiliation of the loan contract; i.e., it is annualled for the future only.

  • 7. In order to obtain the reimbursement of the tax paid on income, which, owing to the dissolution (“résolution”), is deemed never to have been earned, the seller must file a modified fiscal return. The current version of bulletin IMP 1010-2 provides for the possibility of a reassessment by the Ministère where the prescription is acquired. This bulletin should be consulted in such cases.

  • 8. The Ministère du Revenu, though, considers that the income earned from property acquired by the purchasee that the deduction does not exceed the amount prescribed in the Regulation respecting the Taxation Act.

  • 9. The capital cost or the adjusted cost base of the property subject to dissolution is equal to what that [sic] capital cost or the adjusted base would be immediately before the sale as if the disposition had never taken place.[347](Emphasis added.)

Thus, the Ministère du Revenu du Québec recognizes the retroactive effect of resolution, whether it stems from a resolutory condition or from another cause of resolution contemplated in the Civil Code. Since the sale is deemed never to have taken place, the seller is considered never to have transferred ownership of the property. He may therefore recover the income tax paid on the capital gain, the business income or the recaptured depreciation. He may produce an amended return to claim the refund of the income tax paid, even if the tax year in question is time-barred.

Not all of the effects of retroactivity are produced, however: if the buyer had possession of the property pendente conditione, then it is he who must pay the tax on the income earned during that period, and he who may claim the depreciation. Consequently, the seller is not entitled to retroactively claim the depreciation for the period during which he is deemed to have been the owner while the buyer in fact had possession of the property.

This position is consistent with the civil law, which recognizes that "[t]he fruits and revenues of the property being restored" belong to the debtor of the obligation to make such restitution.[348] As for depreciation, even though it is harder to justify the fact that the buyer is entitled to the deduction, this solution seems to be based on considerations of fairness.  Indeed, since the buyer retains the income and pays the tax on this income, it is logical that he should be allowed to claim capital cost allowance.

Moreover, as for suspensive conditions, an authorized representative of the Ministère du Revenu du Québec confirmed that the administrative position applied to resolutory conditions applies, mutatis mutandis, to suspensive conditions.

One should also examine other Quebec statutes relating to the tax consequences of the retroactivity of conditional obligations. The Act respecting duties on transfers of immovables,[349] which gives the right to tax transfers of immovables, is an example. The definition of the term “transfer” found at section 1 of that Act refers, inter alia, to the transfer of the right of ownership on a property. A nearly identical definition is found in section 1 of the Land Transfer Duties Act.[350]The Ministère du Revenu du Québec has stated its position on the application of the latter Act in cases where the right of redemption was exercised further to the sale of an immoveable by a municipality for unpaid taxes. Here an excerpt [from the official English translation] of its Interpretation Bulletin DTT 1-2:

  • 2. Pursuant to section 532 of the Cities and Towns Act (R.S.Q., chapter C–19), the registration of an authentic copy of a deed before a notary establishing the reimbursement of the monies and the redemption of the immoveable restores to the transferee the right of ownership of the immoveable possessed by him at the time of sale.

  • 3. Therefore, following sale of an immoveable for non-payment of taxes, the land shall only belong to the purchaser when the time limit within which the owner may redeem the immoveable has expired. If, within this time limit, the owner recovers the immoveable through the exercise of his right of redemption, his right of ownership shall be restored in the state in which it was at the time of sale.

  • 4. The purchaser who becomes the owner of land following a municipal sale for unpaid taxes does so subject to its being redeemed. If the owner exercises his right to rescind the sale by redeeming, the redemption of the immoveable shall restore to the owner exactly the same ownership right he possessed at the time of sale.

  • 5. ...Hence, exercise of the right of redemption does not constitute a transfer within the meaning of the Land Transfer Duties Act since it does not lead to a transfer of rights between the transferor and the transferee.[351] (Emphasis added).

One may therefore conclude that in cases involving a resolutory condition, the fulfilment of a condition produces its retroactive effects in regard to transfer duties under both the Land Transfer Duties Act and the Act respecting duties on transfers of immovables. According to Me Marie-Pier Cajolet, not only would the deed of retrocession not be governed by these Acts, but the annulment of the original deed of sale would create an obligation for the municipality to refund the transfer duties previously collected.[352]

As for sales taxes, pursuant to the federal Excise Tax Act[353] and the Act respecting the Québec sales tax,[354] the event that gives rise to tax is a “taxable supply.”[355] The term “supply” is defined in both statutes as “the provision of property or a service in any manner, including sale.”[356] One might be tempted to believe that the retroactivity of conditional obligations has no impact on the GST or the QST, since retroactivity cannot nullify the supply of property.[357] And yet, the Ministère du Revenu du Québec appears to recognize the retroactivity of resolutory conditions where the application of the QST is involved:

However, if the contract as a whole constitutes a sale with a right of redemption, then the transactions conducted by B to A will not be construed as taxable supplies in view of the retroactive effect of this sale. B will then be considered as an SMB entitled to all of its ITRs.

Under article 1750 of the Civil Code of Québec, a sale with a right of redemption is a sale under a resolutory condition by which the seller transfers ownership of property to the buyer while reserving the right to redeem it.

Upon analysis of the contract, it is very clear that the transaction is not a sale with a right of redemption. There is no clause evidencing A's intention to reserve a right of ownership in a specific thing, to be exercised within a certain period of time.

Thus, we find that the matter in issue comprises two transactions, two sales within the meaning of article 1708 C.C.Q.: one by A to B and the other by B to A.

Since these sales are taxable supplies, subject to the goods and services tax and the Québec sales tax, they shall be taken into account in determining the status of B.[358] (Emphasis added).

The opinion is about a sale with a right of redemption, but since such a transaction is a sale under a resolutory condition,[359] one can reasonably infer that the opinion is likely to apply to any resolutory condition.

Finally, it should be noted that the retroactivity of the resolutory condition is not always recognized for purposes of computing paid-up capital, where the capital tax under Part IV of the Taxation Act is concerned. In one particular case, the Ministère du Revenu issued the following opinion,[360] based on section 1131 of the Taxation Act, which provides that the tax payable on capital in a taxation year is tied to the paid-up capital shown on the financial statements of the corporation for the year. However, generally accepted accounting principles did not, in this case, allow for the financial statements to be amended, and resubmitted to the shareholders, for the previous years affected by the retroactivity of the transaction in question.

4. The tax treatment of conditional obligations: A Critical review

Having discussed the current civil law and common law on conditional obligations, we are now in a position to review the Act, the case law and the CCRA’s administration position from a critical stance in order to determine whether tax law conflicts with civil law as far as the retroactive effect of conditional obligations is concerned.

Once this has been done, we will outline some possible solutions and indicate how they might be translated into legislative proposals.

4.1 Conflicts between Federal Law and Civil Law

4.1.1    Suspensive conditions

As we have seen, the Agency’s position on suspensive conditions distinguishes between cases in which possession and enjoyment of the property have been transferred to the buyer pendente conditione and cases where they have not.

4.1.1.1    Without transfer of possession

The solution regarding the period prior to the fulfilment is consistent with civil law: ownership cannot have been transferred because the obligation has not yet come into existence. And of course, nothing happens in civil law or tax law if the condition fails.

The situation obviously becomes more complicated where the condition is fulfilled. The Agency’s position in this situation is that the disposition occurs at the moment the condition is fulfilled. The transfer of ownership is supposed to have occurred when the contract was signed, due to the retroactive effect of the suspensive condition under the Civil Code. The tax authorities do not recognize the retroactive effect, hence the conflict with civil law.

The Agency’s position makes it harder for the taxpayer, especially as far as tax planning is involved. The transaction is subject to any amendments that might be introduced pendente conditione if the condition’s retroactivity is not recognized. Any tax planning based on the tax consequences becomes very risky. For example, consider a sale of shares in a small business under a suspensive condition: if the capital gains exemption were abolished, or if the company no longer qualified as a corporation operating a small business before the condition was fulfilled, the seller would no longer be entitled to these benefits and the tax consequences could be disastrous.

Nonetheless, we do not believe the Agency’s position to be well founded under the current law because it relies on erroneous judicial applications of the test in Wardean Drilling, in particular the decision in Olympia & York. Indeed, as mentioned, the current provisions of the Act do not specifically state that suspensive conditions are not retroactive. Since tax law is accessory to private law, and private law is to be used to complete tax law, we must turn to the civil law in order to interpret the word “disposition” as understood under the Act, because the definition is incomplete.

This is why the Supreme Court, in Compagnie Immobilière BCN,[361] implicitly recognized that the civil law concept of disposition refers to the moment ownership is transferred. Transfer of ownership, under the Civil Code, is deemed to take place when the contract is concluded, therefore disposition for tax purposes should occur at the same moment.

Transfer of ownership, under the Civil Code, is deemed to take place when the contract is concluded, therefore disposition for tax purposes should occur at the same moment.

Interestingly, disposition at common law is also dependent on the concept of ownership or title. The splitting of title between a beneficial owner and a legal owner is recognized at common law, and this is why common law may consider there to have been a disposition when beneficial title is transferred. We agree with Noël J’s statement in Construction Bérou that Wardean Drilling did not oust provincial private law in cases where the definition of “disposition” is to be applied to tax matters. On the contrary, the judge simply applied the applicable provincial private law, i.e., the common law.

Our position does not necessarily contradict Victory Hotels. In Victory Hotels, the Exchequer Court recognized that the definition of “disposition” in the I.T.A. is not altogether complete. It appears to have circumscribed the concept, in the case of a sale, to the point at which the seller is entitled to the sale price. This conclusion was based on the fact that the definition of “disposition of property” includes an event which entitles the taxpayer to the proceeds of disposition of property; and this was held to include the sale price of the property sold.

In cases involving a sale, these definitions do not mean that a disposition takes place solely when the seller is entitled to the sale price. This event is simply included as one of the events that come within the concept of disposition, i.e. there is a disposition, inter alia, when the seller is entitled to the sale price. Thus, there can be a disposition before the seller is entitled to the sale price, but not afterwards. In fact, the admittedly ambiguous excerpt from Victory Hotels arguably supports this:

These sections do not define but merely include as a disposition of property a transaction (a sale for instance) entitling a taxpayer to proceeds of disposition of property, i.e. to the sale price of the property sold. It would indeed appear that the meaning of “disposition of property” has been somewhat restricted by the Act when a disposal of property takes place by means of a sale; in such a case there is a disposal of property as soon as a taxpayer is entitled to the sale price of the property sold.[362](Emphasis added.)

As we have seen, the seller subject to a suspensive condition is only entitled to the sale price once the condition is fulfilled. As a result of retroactivity, the transfer of ownership occurs before the seller is entitled to the sale price; and thus, it not logically aberrant that a disposition occurs at the same time as ownership is transferred.

We have also seen that the Supreme Court[363] and other courts[364] have given effect, in taxation law, to the retroactive nature of provincial statutes. As previously stated, these decisions are well-founded in law because tax law must recognize the legal consequences imposed by provincial law on parties to a contract.[365]

In our opinion, the argument that the retroactive effect of the suspensive condition in civil law only applies to de jure, and not de facto, matters, does not prevent retroactivity from being applicable in tax cases.[366] We have concluded that disposition is a legal concept, and that, where a sale is involved, it contemplates the transfer of ownership. We cannot consider that the concept of disposition applies only to real, indisputable and ineffaceable facts that are immune to retroactivity, such as the collection of fruits, the enjoyment of the property or acts of administration. To the contrary, retroactivity should also apply to the concept of disposition since the right of ownership is subject to retroactivity, because these two concepts are related.

In addition, even if the Wardean Drilling test was applied in a civil law content, the disposition should be considered to have taken place when the contract is concluded. The judge in Wardean Drilling held that there is a disposition when ownership is transferred, or when the normal incidents of ownership are transferred and the seller reserves title as security for the sale price. The seller subject to a suspensive condition does not reserve ownership as security; this is different from an instalment sale. We must therefore go back to the first part of the test whereby disposition occurs upon the transfer of ownership, i.e., in civil law, when the contract is concluded given the retroactive effect.

One other objection might be raised against our position, however. This argument flows from the definition of “disposition” in paragraph 248(1)(e) I.T.A., which states that there is no disposition where the “beneficial ownership” is unaffected by the transfer of property. In other words, a disposition would only occur when the beneficial ownership is transferred, despite the transfer of ownership at an earlier date. Under this argument, even where, as a result of the retroactive effect of the condition, ownership was transferred when the contract was signed, there will be no disposition until the beneficial ownership is transferred, i.e. when the condition is fulfilled.

Nonetheless, this argument can be rebutted, because paragraph 248(3)(f) I.T.A. states that “beneficial ownership” specifically means ownership of property when applied to Quebec. Therefore, the “beneficial ownership” is transferred at the same time as ownership according to the civil law.

4.1.1.2    With a transfer of possession

The Agency’s position is that when possession of the property has been immediately transferred to the buyer in a sale subject to a suspensive condition, a disposition occurs within the meaning of the Act as of the time possession is transferred.

This position is based on Olympia & York, where the Federal Court (Trial Division) held that there had been a disposition for tax purposes even though, in civil law, there had been no sale.[367] This decision was itself based on the criteria established in Wardean Drilling. As previously mentioned, Olympia & York is not well founded in law because it imported a common law precedent into civil law. In addition, the case was based on the division of ownership between the legal owner and the beneficial owner, a concept foreign to civil law. Consequently, in our opinion the Agency’s position is not founded in law.

In civil law, the seller is not entitled to the sale price before the sale occurs, and the sale cannot occur until the suspensive condition is fulfilled. This rule is recognized by the Agency where the suspensive condition is not accompanied by a transfer of possession. But there is no reason to treat the situation differently when possession is transferred pendente conditione to the buyer. Indeed, as previously explained, disposition in tax law is related to the transfer of ownership as determined by the civil law, and the latter does not give the possessor of property any right of ownership, contrary to the common law.

Ironically, when the Agency’s position is applied to a suspensive condition that has been fulfilled, it produces the same result as the civil law: the disposition occurred when the contract was signed. This is merely a coincidence because the result arose for different reasons: in civil law, it is the condition’s retroactivity that causes ownership to be transferred when the contract was signed, while according to the Agency’s position, there is a disposition for tax purposes because at that moment the possession, use and risk are transferred.

The Agency takes the view that there is a second disposition, this time in favour of the seller, when a suspensive condition fails. For the reasons just discussed, we believe that this position is incorrect because ownership was never transferred in civil law. The seller is simply taking back the property, which he always owned. There is therefore a conflict between tax law and the civil law in such circumstances.

This conflict has major tax implications for the seller. He will be taxed on the capital gain realized when capital property, other than depreciable property, was  “disposed of”, even if he was never entitled to the sale price. The adjusted cost base of the property will be increased accordingly:  the adjusted cost base is “involuntarily crystallized” and in many cases the capital gain will not be exempt.

Therefore, the seller must pay a capital gains tax even though he still owns the property and does not have the money from the sale price to pay the tax. The Supreme Court refused to allow this situation when it ruled in Dominion Engineering.

There will be no capital gain for the buyer of such property because the deemed proceeds of disposition, or the sale price, is by its very nature the same as the cost of acquisition.

A seller will be taxed on the recapture of CCA as soon as possession of the depreciable property is transferred to the buyer. He will not be allowed to claim a refund of any tax paid when the property is returned and will get have to live with the corresponding increase in capital cost. Conversely, the buyer who claimed capital cost allowance while he was in possession of the property must also pay tax on recapture of CCA because he will be deemed to have “resold” the property for the original sale price.

What happens if there is a capital or terminal loss? The seller should logically be allowed to claim the loss and should not have to reimburse any tax saved when the property is returned. In this case, the seller is considered to have bought the repossessed property for the amount of the sale price, which becomes the new adjusted cost base. The seller has therefore “cashed” a capital loss or terminal loss without giving up ownership of the property. In such cases, the taxpayer benefits from the refusal to recognize retroactivity.

Other problems arise where control of a company is acquired further to the sale of its voting shares. The Agency’s position is that a disposition occurs, in a sale subject to a suspensive condition, as soon as the buyer is in possession of the shares.  If the number of shares disposed of is sufficient to enable the buyer to control the company, the provisions of the Act relating to takeovers will apply, and in particular, the provisions on deemed year-ends and the restrictions of loss carry-overs. There would be a second acquisition of control if the condition fails and the buyer takes back his shares. According to the Civil Code, the seller is deemed never to have sold his shares and would therefore not be entitled to the loss carry-over even for losses incurred prior to the transaction. Me Pierre Martel noted this problem and added that this interpretation would be contrary to the spirit of the anti-avoidance provisions that apply to takeovers.[368]

The Agency justifies its position by arguing that the Act read as a whole, does not allow it to reassess returns filed in previous years in order to correct the tax consequences of retroactive conditions.[369] However, a taxpayer would be entitled, under the new fairness provisions, to claim a refund for the time-barred years. The limitation provisions of the Act could be amended to allow the Agency to reassess and claim any amounts due in such instances. Moreover, the Ministère du Revenu du Québec already applies these administrative procedures apparently without any problems. In any case, this argument holds no water where a suspensive condition fails, and the contract was signed, in the same tax year. The Agency nonetheless takes the position that a taxable disposition has occurred in such a situation.

4.1.2    Resolutory conditions

If a resolutory condition is involved, the Agency’s position is that a disposition occurs when the contract is concluded. When the condition occurs and the contract is resolved, a second disposition occurs.

This position is correct in civil law, pendente conditione. As long as the condition does not occur, the civil law considers a sale subject to a resolutory condition to be a pure and simple sale: ownership is transferred to the buyer immediately and the seller is entitled to the sale price.

This interpretation is also in keeping with the tax cases we discussed on resolutory conditions and conditions subsequent. They should be treated identically for tax purposes because both the civil law and the common law treat these conditions similarly.

The situation will remain the same if the resolutory condition never occurs: there is a disposition in civil law and in common law as soon as the contract has been signed, and no conflict arises.

However, under Civil Code if a resolutory condition occurs, ownership is deemed never to have been transferred. Since the tax authorities do not recognize retroactivity, there is a conflict with civil law.

As discussed earlier, in our opinion the Agency’s position is not well founded in law in light of the tax cases. Disposition, in tax law, refers to the transfer of ownership as governed by the civil law, and retroactivity under the civil law applies in tax matters. A disposition for tax purposes would never have taken place if the resolutory condition occurs because ownership was never transferred. Likewise, we believe the seller’s right to receive the sale price is retroactively extinguished: the seller was never entitled to the sale price because the event on which this right was based is deemed never to have occurred. In fact, this is why the sale price is returned.

The Agency’s refusal to recognize resolutory conditions gives rise to the same problems discussed in relation to suspensive conditions with transfers of possession: the seller is taxed on a capital gain, the seller recaptures CCA and the takeover rules are applied to the buyer.

Likewise, the Agency feels that sections 79 and 79.1 I.T.A. could apply in such circumstances. As previously underlined, these sections only apply to property repossessed further to the buyer’s failure to pay and are therefore irrelevant in the case of a true resolutory condition. A transfer of the beneficial ownership is considered to have occurred when the sale is resolved for non-performance by the debtor even though the ownership has not been transferred under the civil law. Moreover, as we saw, under paragraph 248(3)(f) I.T.A., the term “beneficial ownership” contemplates ownership in civil law but does not include the simple transfer of possession, use and risk. We therefore believe the Agency’s position is erroneous.

4.2  Proposed solutions

Under current tax law, the concept of disposition is necessarily dependent on the civil law. There are two reasons:  first, the Income Tax Act does not define this concept. All it does is include concepts that would otherwise not have been included. Second, the definition in the Act itself refers to beneficial ownership – a private law concept – but does not provide any significant or substantive description. Since the Act has not cast aside private law in this area, conditions must be given retroactive effect in tax law.

Having said this, it may very well not be desirable for tax law to recognize the civil law’s retroactivity. If this is the intent, the Act should be expressly amended to exclude provincial private law on this issue and establish its specific rules. In other words, if Parliament, so desires, it should expressly dissociate itself from provincial private law on this matter.

We will now consider whether it would be advisable to recognize the retroactive effect of conditional obligations in tax law further to which, we will attempt to determine what the tax policy behing the concept of disposition ought to be. Finally, we will recommend legislative amendments we believe are necessary.

4.2.1    The consequences of retroactivity in tax law

What would be consequences if tax law recognized the retroactive effect of conditional obligations?

As discussed above, further to a sale subject to a suspensive condition, the capital gain should only be recognized once the condition is fulfilled, but it should be retroactive to the date the contract was signed. The taxpayer should declare the capital gain by filing an amended return for the tax year during which the contract was signed, if it was signed in a previous year. The same logic applies to business income or a recapture of CCA.

In the reverse situation, a taxpayer should be allowed to file an amended return in order to claim a capital loss or terminal loss.

The provisions of the Act on the assessment and reassessment periods should obviously be amended to allow the Agency to reassess a particular year, even if it falls outside the normal reassessment period.

The recognition of the retroactivity effect of suspensive conditions will cause certain problems however. First, once the condition is fulfilled, a seller who claimed CCA pendente conditione has done so on property he is deemed never to have occurred. Should the tax savings resulting from the CCA be reimbursed? In addition, can the buyer, who retroactively own the property on the date the contract was signed, claim CCA for the years during which the condition was suspended? Either of these solutions might be inequitable for one of the parties to the transaction. It does not seem fair that the buyer should be able to claim CCA for the period during which the seller had the possession, enjoyment and control of the property and assumed the risks and charges. But it would also be unfair not to allow the buyer to claim CCA for these years if he has possession of the property during that period.

A solution to this problem would be to adopt the position similar to that of the Quebec tax authorities. A party need not reimburse any CCA previously claimed and the other party cannot retroactively claim CCA. The parties, therefore, would be free to make a joint decision and to stipulate which of them would be entitled to claim CCA, such as is currently the practice with financial leasing agreements.[370] However, the problem would still exist where the parties failed to decide together: legally, the right to claim a capital cost allowance must be withdrawn from the seller and given to the buyer if the retroactive effects are fully recognized. This could lead to the inequities mentioned above and make administration more complex.

However, as previously mentioned, the taxpayer is sheltered, via the suspensive condition’s retroactivity, from any statutory amendments that might be introduced before the condition is fulfilled. While this undoubtedly places the taxpayer in an advantageous position, it goes without saying that it causes the government to lose some of its revenue. Furthermore, the retroactivity of a suspensive condition could open the door to numerous advantageous tax planning opportunities for the taxpayer, and Parliament would once again be forced to complicate the Act in order to stop avoidance transactions.

Applying retroactivity to a resolutory condition would also give rise to certain problems.

Further to a sale subject to a resolutory condition, a capital gain or loss, recapture of CCA or a terminal loss would be recognized for the tax year in which the contract was signed. The disposition would be deemed never to have occurred if the condition happened. The taxpayer would then be required to file an amended return seeking a refund of taxes paid on the capital gain or the CCA recapture. By the same token, any tax initially saved if a capital loss or terminal loss was declared should be paid. Naturally, the comments made earlier regarding the necessary legislative amendments to the rules on the periods of reassessment also apply in this case.

The question of who can claim CCA during the pendente conditione period must also be addressed with suspensive conditions: the buyer will have claimed CCA for the years during which he had possession of the property, assumed the risks and the charges and received the income. However, he is then in the position of having claimed CCA for property that he did not acquire if the condition occurs, and the seller, who never disposed of the property, could theoretically claim CCA for the years in question. However, in contrast with the suspensive condition, the buyer had possession pendente conditione , and a true right of ownership in the property. This right was retroactively extinguished: the fair solution would be for the buyer to claim CCA during this period. The Quebec tax authorities have opted for this position.

4.2.2    Parliament’s intent

During the APFF’s 1991 tax conference, the Department of Finance made the following comments on the decision of the Tax Court of Canada in Fortin & Moreau (Construction Bérou):

The Department of Finance feels that it is important to define the concepts of acquisition and ownership under the civil law in the same way as under the common law in order to avoid inconsistency between the taxation of transactions occurring in Quebec and the taxation of transactions occurring in a common law province.[371]

This clearly expresses the intent to apply a uniform concept of acquisition and disposition under the Act throughout the country.

Although we are of the opinion that tax law, as it currently stands, should recognize the retroactivity of conditional obligations, we are nonetheless aware of the importance of its uniform application. From a tax policy point of view, we understand that the Department of Finance would prefer that the Act applies in the same manner in Quebec as in the other provinces, thereby encouraging interprovincial trade and an equitable application of the Act.

However, this objective collides with both the common law and the civil law, two major private law systems in Canada; the concept of ownership and the consequences of conditional obligations are fundamentally different in both systems. A definition of disposition, specific to tax law, that does not refer to either system of private law or that takes their different concepts into consideration, is necessary if the Act is to be applied uniformly.

Moreover, we must recognize that Parliament originally intended to broadly define disposition. The Report of the Royal Commission on Taxation,[372] commonly called the “Carter Report”, published in 1966, favoured an all-encompassing tax res that would include any increase in economic power, including any increase in the market value of the property held, whether or not this increased value was materialized. For practical purposes however, the Commission recommended that the increased value should not be taxed until the property was “disposed of”. This concession was a compromise on the theoretically ideal res and it was recommended that a broad concept of disposition be adopted.[373]

Nonetheless, the government of the day did not think it wise to go as far as the Carter Commission was recommending.[374] Parliament ended up adopting back then the same definition of disposition that was in force since 1949 for the purposes of CCA recapture.[375]

As we saw, a significant number of civil and common law cases have attempted to apply a uniform concept of disposition for the purposes of the Act. These cases have equated disposition with the transfer of beneficial ownership. We must admit the objective is desirable even though the wording of the Act and the complementarity of private law with federal tax law makes it impossible for us to agree with the cases: the property’s increased value should be taxed when it is realized, that is to say, once the gain is definitely determined.

We believe that the tax policy underlying the taxation of capital gains or the recapture of CCA is that a gain is to be taxed when it is realized, that is to say, once the seller has a certain and absolute right to the sale price.

A basic principle of taxation in Canada, the United Kingdom and the United States[376] is that income must be realized or accrued before it can be taxed. One of the main reasons underlying this principle is that the property’s value stops fluctuating at this time and the gain or loss can thus be calculated. The seller also has the necessary cash to pay his income taxes at this time.[377]

4.2.3    Proposed amendments

We will attempt to suggest legislative amendments that strike a balance between the uniform application of the Act from coast to coast and the  respect of the two major private law systems in Canada. The reader will understand that we have not considered every possible combination and permutation, but that these proposals should serve as a helpful starting point.

One possible solution would be to define the concept of definition clearly and exhaustively, specifying that provincial private law is not to be considered the complementary source of law. The definition could refer to the time when the seller has an absolute, although not necessarily immediate, right to the sale price. Nonetheless, it would be extremely difficult to develop an exhaustive, neutral definition of the concept of disposition: the definition would not refer to any private law concept (e.g. ownership or beneficial ownership) in order to be complete.

We are of the opinion that it would be much better to define the concept of disposition by providing what is contemplated for Quebec purposes and, furthermore, what is meant for common law purposes. For purposes of Quebec, the definition could include the following elements:

When applied to the common law provinces, a “disposition”, would occur once the beneficial ownership is transferred, even if the vendor has reserved legal title.

According to this definition, in Quebec, a sale under suspensive condition, would be deemed a disposition once the condition was fulfilled, because the seller would not have an absolute right to the sale price until that time. A disposition would occur as soon as the contract was signed in the case of a resolutory condition, because the obligation arises immediately.

A definition excluding the condition’s retroactive effect would avoid the difficulties mentioned earlier and would make it simpler to apply the Act by eliminating the need for amended tax returns. Clearly, the inconveniences caused by a denial of retroactivity – inconveniences discussed earlier – would remain. That is the price to pay for a uniform federal statute: it would be the legislator’s intent. At the very least, a clearly drafted Act would allow Quebec taxpayers to know, in advance, that the retroactive effects of their transactions would not apply for income tax purposes.

Because there is a transfer of the beneficial ownership at common law when a condition precedent is fulfilled, there would only be a disposition at that moment, according to the proposed definition. Conditions subsequent do not prevent the beneficial ownership from being transferred, so the disposition would take place immediately.

Hence, civil law conditional obligations would be treated in the same way as common law conditions for tax purposes. The goal of applying the Act uniformly across Canada would be attained.

However, an instalment sale of civil law would give rise to a disposition under the proposed definition once the contract is signed. Indeed, in such a sale, possession, use, the right to collect the fruits and the risk of loss are transferred to the buyer and the seller, although subject to a term, has an absolute right to the sale price.

Last, this definition respects both civil law and common law concepts and terminology, and is in keeping with the principle that the federal Act should not conflict with the civil law.

Subsections 79(2) and 79.1(2) I.T.A. should be amended to remove any reference to “beneficial ownership” in relation to the resolution of the contract due to the debtor’s non-performance. Reference should be made to the same concepts we have included within the definition of disposition. The retroactive effect of a resolved contract should also be excluded.

Conclusion

We have discussed the law applicable to civil and common law conditional obligations and the tax consequences of those obligations. 

We have considered various civil and common law concepts pertaining to conditional obligations and retroactivity, devoting special attention to the similarities and differences. We then looked into the tax consequences of these different private law concepts. Our conclusion is that tax law essentially follows the rules of private law when it comes to determining when property is disposed of, although there are contradictory cases on the subject and the Income Tax Act is sometimes ambiguous on the point.

This conclusion is based on the principle of complementarity which holds that when the Act does not exhaustively define a private law term, the civil law must be applied in Quebec and the common law must be applied in the rest of Canada.

Finally, we concluded that if Parliament wishes to exclude provincial private law, and articulate its own rules, it must do so expressly. We proposed amendments to the legislation that would do so, amendments the purpose of which is to apply tax law uniformly throughout the country while remaining respectful of both the civil law and common law systems.

Bibliography

Statutes and Regulations

Constitutional enactments

Federal enactments

Quebec statutes

Non-Quebec statutes

Cases

Scholarly writing

Monographs, booklets and brochures

Journal articles, conference papers, etc.

Administrative positions

Federal

Quebec

Footnotes