3.1.1 Introduction

This sub-section deals with the solution adopted by South African constitutional property law. In 1981, the Alienation of Land Act was enacted.^[9] The presumption of this Act was that vulnerable individuals would normally not be able to pay the full purchase price for land at once. Therefore, the Alienation of Land Act aims to strengthen the position of the purchaser who pays the purchase price in installments. One of the provisions that strengthens the position of the purchaser is Section 22 of the Alienation of Land Act. This provision dictates that a purchaser is entitled to receive the title to the immovable property if the transferor has gone bankrupt prior to delivery, on the condition that the purchaser has purchased the land in terms of a ‘contract’ and that the purchaser has made arrangements for the payment of all costs in connection with the transfer of land and several other costs. A ‘contract’ in terms of the Alienation of Land Act is an agreement pertaining to the sale of land in more than two installments over a period exceeding one year.

A few years ago, the scope of this protection was widened by the Constitutional Court in Sarrahwitz v Maritz N.O. and Another. In this case, Ms Sarrahwitz, an unemployed woman, bought a house from Mr. Posthumus and paid the purchase price in full, by borrowing this amount from her former employer. However, as the years progressed, no transfer of the immovable property took place. On 14 April 2006, the estate of Mr Posthumus went bankrupt and Mr Maritz was appointed as the trustee. At common law, both the immovable and the purchase price formed part of the insolvent estate. After several unsuccessful attempts, Ms Sarrahwitz made an application to the Constitutional Court. The Constitutional Court held that the distinction between purchasers that paid the purchase price in installments and ones who immediately paid the full price was not in accordance with Sections 9 (equality) and 26 (right to housing) of the Constitution. The Court therefore applied the remedy of severance and read in, at the end of the definition of a ‘contract’ in section 1 of the Alienation of Land Act, the words “including residential property paid for in full within one year of the contract, by a vulnerable purchaser”. The Court further defined a ‘vulnerable purchaser’ as “a purchaser who runs the risk of being rendered homeless by a seller’s insolvency”. Therefore, under these circumstances, the transferee may acquire the ownership of the immovable property, despite the fixation of assets of the insolvent estate.

3.1.2 Sub-questions

The instrument applies only to the transfer of immovable property (question 1). The instrument applies automatically (question 2): as long as the purchaser qualifies as ‘vulnerable’ and the purchase price is paid at once in full, the purchaser is protected without any additional form of registration being required. Although the court does not explicitly limit the duration of the instrument, the judgment does provide some sort of limitation by stating that the order only applies to “a seller’s insolvent estate that has not been finalised” (under 78, point 5). However, as long as the purchaser pays the purchase price at once within one year after concluding the contract, the proprietary interest that the judgment bestows upon the purchaser is, in principle, unlimited in time (question 3). Finally, the purchaser is merely protected against the bankruptcy of the seller, and – unlike a priority notice – not against an attachment or a second grant by the seller (question 4).

3.1.3 Concluding remarks

The most important feature of this judgment for the rest of this paper, is that the judgment illustrates that policy arguments such as preventing homelessness can outweigh the interest of the creditors in bankruptcy (embodied by the establishment of the concursus creditorum) and hence provide an exception to the ‘first come’-rule, despite the fact that these are core principles of property/insolvency law.^[10] In the context of bankruptcy, the idea of the concursus creditorum and its first-come rule seem so impenetrable that there is hardly any point in challenging it. Instead of deciding in favour of the creditors or the buyer on the basis of the merits of their respective claims, the fixation principle seems to completely block out the possibility of discussing its merits. This all or nothing approach is hard to justify, especially given that the question whether an interest is proprietary or not, often depends on timing. To illustrate this, consider the English cases of Re Goldcorp Exchange Ltd and Re London Wine Company (Shippers) Ltd, in which the buyers were erroneously under the impression that ownership had already passed to them, with disastrous consequences when faced with the bankruptcy of the seller. For this reason, Worthington has compared the process of winding up a bankrupt estate as a game of musical chairs, because the relative positions of the parties at the time the music stops (the fixation principle kicks in) is crucial to their happiness.^[11] Worthington, therefore, pleads for a different approach that shifts the focus from doctrinal arguments to policy arguments. Sarrahwitz illustrates that policy arguments may in fact provide a deviation from the fixation principle. The judgment also proves that policy arguments may justify the existence of vacuum instruments, even though parties themselves have not explicitly chosen for such an instrument. This makes sense: from a policy perspective, the position of ms Sarrahwitz or mr Martiz is irrelevant. It is homelessness as a problem for society as a whole which must be tackled with a constitutional remedy.^[12]

3.2.1 Introduction

For Germany, my analysis will focus on the priority notice. The priority notice is, by origin, a German creation (Vormerkung), which was adopted by many other jurisdictions in the past decades. The Netherlands, for instance, adopted the priority notice in 2003. In order to illustrate the different functions of the priority notice, the German priority notice – or, more precisely, the way it is utilised in current legal practice – is juxtaposed with the Dutch priority notice.

3.2.2 Sub-questions

The first difference is the scope of application of both notices. In the Netherlands, the priority notice can only be used to anticipate the transfer of registered property (immovable property, ships and aircraft) itself (s. 7:3 Dutch Civil code). It is not possible to anticipate the creation of a security interest (hypothec). Whether and to what extent the creation, transfer and waiver of other limited rights can be anticipated upon, is unclear.^[13] The German priority notice, on the other hand, can be used to anticipate any real/proprietary change in rights in immovable property (Dingliche Rechtsänderung) (§ 883 BGB) (question 1). This wider scope of application boasts a huge benefit: since the buyer of land often requires external financing to cough up the purchase price, creating a security interest – and being able to anticipate it – is just as vital to the transaction as the transfer of the asset itself. Both instruments also differ in terms of duration: the Dutch Vormerkung lasts six months, whereas its German counterpart is, in principle, not limited in time (question 3).^[14] In terms of effect, both instruments are the same: they both provide the grantee a proprietary interest that can be invoked against other grantees and creditors/a bankruptcy trustee (question 4).

The biggest difference between the two instruments in this regard is the frequency with which the priority notice is used. This closely relates to the second question (type of application). In the Netherlands, the priority notice is not used very often. Instead, the performance interest of the buyer is protected by the client account of the civil law notary. The client account is a bank account held by the civil law notary that does not form part of the recoverable property of the notary itself. The money in the account still belongs to the parties involved. In the context of a transfer of real estate, the buyer will transfer the money to the client account. As soon as the civil law notary has established that the buyer has actually become the owner of the immovable property, the notary releases the money to the seller. If the buyer fails to acquire ownership – for instance, because the seller went bankrupt in the meantime – the notary releases the money to the buyer instead.^[15]

In Germany, on the other hand, the priority notice is used very often. It even forms part of the template of sale-purchase agreements of land that can be downloaded by legal practitioners.^[16] What must be noted is that in Germany, the contract by which land is transferred must be recorded by a civil law notary anyway (§ 311 b BGB), regardless of whether or not parties use a priority notice. In the Netherlands, the only requirement of the contract is that it is ‘in writing’, which means that the additional fee of the notary comes into play only when parties choose for a priority notice. This brings about that the ‘extra’ cost of registering the sale purchase agreement is substantially higher in the Netherlands. The German Vormerkung does not apply completely automatically like the Vendor Purchaser Constructive Trust (see next subsection), but because it is in practice used in almost all real estate transactions, the German Vormerkung closely resembles automatic application. Because the priority notice protects the performance interest of both parties,^[17] there is no need for a client account. This is reflected in German practice: the client account of the German civil law notary is hardly ever used in standard real-estate transactions, and the purchase price is transferred directly from the buyer to the seller. So why then does Dutch law have a priority notice in the first place? The difference between Germany and the Netherlands in this regard, perfectly illustrates the different functions that a vacuum instrument might have. Section 2 has listed three problems of a simultaneous transfer that can be overcome by vacuum instruments. For the sake of clarity, the problems are summarized here once more. Problem a is the necessity to check at the last possible second if the transferor still has the power to dispose, problem b is the possibility of an event that occurs with retroactive effect and problem c is the possibility that the preparations of the transferee have been in vain if the exchange does not occur. In Germany, the function of the priority notice is to solve all three problems.^[18] By contrast, in the Netherlands, the client account of the notary overcomes almost all difficulties relating to problem a and b by creating what is, de facto, a vacuum instrument in favour of the seller (see footnote 16). The priority notice provides an additional form of protection for the buyer by providing a remedy for problem c. It must be noted that the Dutch Vormerkung does have the potential to overcome problems a and b, just like its German counterpart.

3.2.3 Concluding remarks

In order to secure the performance interest of both parties, it has become the customary practice in both countries to use respectively the client account of the civil law notary or the priority notice, even though there is no rule that makes either mandatory. Also more practical interests, such as the interest of civil law notaries to secure the performance interest of both parties in a cost-effective way, can justify ‘imposing’ vacuum instruments on parties to the transfer. In fact, this result can also be considered as an interest of society as a whole, because it will result in lower transaction costs (section 4 elaborates on this).

3.3.1 Introduction

So far, my analysis has focused on immovable property. However, as mentioned in the introduction, nowadays, other types of assets are just as important in the context of commercial transactions. Examples are company shares, intellectual property rights, permits, claims and contracts.^[19] The problem with these types of assets, however, is that there is no register that keeps track of their legal status. For this reason, a priority notice cannot exist in the context of these types of assets. This section will demonstrate, however, that it is possible to protect the performance interest of the buyer without awarding the buyer a registered and hence publicly visible right. For this purpose, this section analyses Anglo-American law, more specifically the Vendor Purchaser Constructive Trust (hereafter: VPCT).^[20]

3.3.2 sub-questions

The basis of the VPCT-doctrine was laid down in Lysaght v Edwards, where it was held “that the moment you have a valid contract for sale the vendor becomes in equity a trustee for the purchaser of the estate sold, and the beneficial ownership passes to the purchaser (...)”.^[21] In the interest of readability, this section will not explore the theoretical background and properties of this doctrine.^[22] The core of its functionality is that, as soon as there is a) a valid sale/purchase agreement, b) a proper consideration and c) an asset that qualifies as ‘unique’, the purchaser of an asset acquires an equitable interest in the asset.^[23]‘Unique’ means that the asset cannot be easily acquired on the current market. This equitable interest has a proprietary nature and can therefore be enforced against third parties. The most notable difference between this vacuum instrument and the previously analysed ones, is that this instrument applies to every type of asset, except for movable property (question 1). The doctrine does not apply in the context of movable property, because the transfer of movables is exclusively regulated by the Sale of Goods Act.^[24] Furthermore, the doctrine applies, in principle, automatically (question 2). In the context of immovable property, however, the Land Registration Act of 2002 requires registration of the equitable interest in public land records in order to achieve third-party effect.^[25] In the context of other types of ‘unique’ assets (e.g., shares and intellectual property rights), the doctrine applies as explained, without requiring registration. This raises the question as to how the purpose of the publicity requirement – being able to quickly assess whether or not someone has the power to dispose of an asset, in order to conduct transactions efficiently – are met.

The lack of publicity with other types of assets than immovable property is overcome by the bona fide purchaser for value doctrine. If buyer A (the first buyer) has acquired an equitable interest in the property and, before delivery took place, a second buyer (B) buys the same asset, A cannot invoke A’s equitable interest against B, on the condition that B is a bona fide purchaser and furthermore provides valuable consideration.^[26] One may notice, however, that the bona fide purchaser doctrine cannot be invoked by creditors or bankruptcy trustee’s. The VPCT therefore brings about a similar result as the South African Sarrahwitz-judgment: the (first) buyer acquires a proprietary interest that can be invoked against creditors, but most often not against purchasers (question 4). The doctrine is, in principle, not limited in its duration (question 3). The ‘uniqueness’ criterion deserves closer attention, because it sheds light on a different type of policy goal that can be achieved with vacuum instruments.

3.3.3 ‘Unique’

On first sight, the criterion that an asset must be ‘unique’, may sound arbitrary. It has been submitted, for example, that – in terms of consequences for the buyer, if the seller goes bankrupt prior to delivery – the difference between shares in a public company (which are not considered ‘unique’) and private shares (which are ‘unique’) cannot be justified.^[27] However, it is submitted that it does in fact have a rational justification. As mentioned above, the VPCT provides purchasers a proprietary form of protection, mainly against creditors of the seller. It is in this particular context – a conflict between a purchaser and creditors – where the rationale of the ‘uniqueness’-criterion shines.

Assets are worth more to a buyer than to a creditor, because purchasers will in general cough up market value, whereas creditors are in general merely able to generate the forced sale-value out of an asset as collateral for their claim.^[28],[29] As a matter of fact, it quite often occurs that the value that the buyer places on an asset is higher than its market value.^[30] The market value of an asset often exceeds the value in a forced sale, but this statement is more accurate if – and to the extent that – an asset is ‘unique’. This can be easily explained. If a generic asset (e.g. oil) is sold by creditors as collateral for their claim, everyone in the world interested in buying oil is a potential buyer, meaning that it would be unlikely that the price of that oil (forced sale value) is much lower compared to an ‘open-market’ transaction. However, if a specific asset (such as immovable property) is sold, the statement that everyone in the world is interested in buying that immovable property is false. The specific properties of that particular piece of real estate (such as its location, its vicinity, size and esthetics) bring about that, if immovable property in Groningen (an area in the Netherlands) is sold, not even everyone who lives in Groningen would be interested in buying that piece of property, let alone everyone in the world.^[31] The more unique an asset is, the more it will be worth extra to a buyer that has already chosen that specific property for its unique traits.

It is important to notice that also creditors benefit from a vacuum instrument in this context: after the buyer has paid the purchase price, the creditors can take recourse against a market value price, whereas the prior tempore outcome (see section 2) outcome would be that the creditors have to make do with dividing the forced sale value amongst themselves. We are starting to explore an additional, yet till now hidden potential of vacuum instruments: they can be used to maximize value of assets, which is naturally also a policy goal of private law. Vacuum instruments that provide a grantee protection against a different grantee cannot be justified on the basis of this benefit, however, they still boast the benefit of providing a solution for problems a and b (see section 2) and hence lower transaction costs.

3.3.4 Concluding remarks

This section does not purport to give a final answer to the question as to whether the argument of reducing transaction costs outweighs the interest of maximizing value. Instead, this subsection merely intends to point out the different shapes and sizes of vacuum instruments, and the variety of policy arguments underlying the desirability of these instruments. In order to underline the differences amongst instruments once more, subsection 3.4 provides a chart that summarises the findings of this section.

4.2.1 Scotland: a tale of two judgments

The central issue of this paper – protecting the performance interest of grantees – has received an extraordinary amount of attention in Scotland. The debate was sparked by Sharp v Thomson, in which the buyers of immovable property had not yet completed delivery before a floating charge, that covered all of the ‘property and undertaking’ of the seller (s. 462(1) of the Companies Act 1985), crystallised.^[35] It was held by the House of Lords that the buyers were triumphant in this conflict. It was, however, unclear exactly why they were triumphant. There were two explanations: explanation A entailed that the buyers had acquired some sort of proprietary interest in the immovable property prior to completion of the delivery. This reasoning seems hard to unite with a traditio transfer system such as Scots law, but can be explained by the tendency of the House of Lords to take common law doctrines, such as the VPCT, into consideration, even when deciding on Scots law. Explanation B centered on a narrow interpretation of ‘property and undertaking’: this would include only what the company uses in its day-to-day business dealings, and not literally ‘all its property’ in a legal sense. The main arguments against explanation A were the lack of clarity in regard to exactly when buyers receive their proprietary interest, and the lack of publicity of the buyers’ interest. However, if explanation A were correct, it would entail that buyers of immovable property were also protected from, e.g., insolvency of the seller, if they fail to complete delivery prior to the sellers’ bankruptcy. Several years later, exactly these facts (interest of buyer conflicts with insolvency of the seller) occurred (Burnett’s Trustee v Grainger).^[36] The House of Lords held that the buyers were in this instance not protected, from which it followed that explanation B of the Sharp v Thomson judgment had been the correct one. In several reports following these judgments, Scottish academics submitted that because of Sharp’s narrow interpretation, there was a lack of transactional security for grantees involved in a transfer of land, and that “there is now a clear need to improve the position of grantees who act in good faith and with reasonable diligence”.^[37]

The considerations of the Scottish law commission are especially relevant for this paper, because they provide a clear insight into the central problem of this paper – the problems that a traditio transfer system gives rise to in the context of mutual contracts and the performance interest of both parties – from the perspective of academics that, just like the author of this paper, have a civil law oriented background whilst being familiarised with the common law solution to this problem (the VPCT). This section will therefore use the general structure of the first report as a backbone for discussing the question as to whether, and if so to what extent, vacuum instruments deserve a stronger effect and/or a bigger scope of application.

4.2.2 Protecting the performance interest: a tale of two approaches

The core of the two possible approaches has been outlined in the initial discussion paper following Sharp.^[38] According to the researchers, the two approaches are: a) “to provide for the registration of a contingent interest in the land to the effect of establishing a priority against all subsequent rights” (also known as the priority notice or Vormerkung) and b) “to continue with a single act of registration but to confer priority against diligence and insolvency for a period before registration”.^[39] These two approaches can be conceptualised as follows: approach A strengthens the interest of the grantee when this interest competes with creditors and other grantees, whereas option B does not strengthen the interest of the grantee, but weakens the position of just creditors when the interest of creditors conflicts with the interest of the grantee.^[40] Option B therefore does not operate in the conflict between two grantees, and seems to assume that the ‘normal’ rules of property law suffice in this context. As mentioned in the introduction of section 4, many countries have adopted a priority notice (option A). Why does the Scots law commission even consider an alternative?

4.2.3 Benefits of approach B

The first reason that the commission mentions, is that a double act of registration “would involve more trouble and expense than would seem justified by the risk sought to be guarded against”. At first sight, this argument seems convincing. After all, 99 % of the transactions take place as planned, without an actual double grant by or insolvency of the grantor. If priority notices became standard practice, the commission rightfully notes that “almost all would turn out to have been unnecessary”. However, as mentioned above, the mere possibility of a double grant/insolvency problem gives rise to transaction costs. Because Germany (also a civil law jurisdiction) uses the priority notice to prevent exactly these transaction costs from occurring, the idea of a priority notice could, in general, still provide a viable solution to reduce transaction costs, even though the commission is of the opinion that this is probably not the case for Scotland.

A second, more convincing argument against the priority notice – and, therefore, in favour of approach B – is that the priority notice not only operates in a conflict between a grantee and creditors, but also in a conflict of grantees against other grantees. Operation in this context, however, cannot be justified on the basis of the argument developed in section 3.3.3, namely that an asset is worth more to a grantee than to a creditor. There is no indication that the grantee that registers a priority notice (first), is able to generate more value than a grantee that registers a priority notice later (or does not use a priority notice at all). This is exactly what makes alternative B attractive: it is a vacuum instrument that only operates when its operation can easily be justified by the difference between market value and forced sale value. In this regard, the commission notes that protection against double grants is offered by other instruments, such as the ‘rule against offside goals’.^[41] This rule entails that the title of an acquirer can be set aside by reduction, if the acquirer knew about an earlier grant in favour of someone else. The commission then poses the question whether “the relatively low incidence of risk justifies the complexities of a system of [priority] notices, or that the few acquirers who turn out to be unlucky should be left to try their fortunes in the courts”. No answer is provided, although the phrasing of the question implies that priority notices are considered too complicated by the commission. Another benefit of method B is that method A requires a register which keeps track of the legal status of an asset: method B can therefore also be applied in the context of different types of assets, such as movable property.

4.2.4 Downside of approach B

However, the biggest downside of only providing protection against creditors/bankruptcy (approach B) is that, because it does not operate in the grantee vs grantee context, a double grant may give rise to the situation where one party performs and the other does not. An example to illustrate this argument: S sells immovable property to P, and intends to transfer/register on December 1 at 3 PM. Assume that none of the approaches applies. In order to ensure that either both parties perform or both of them do not – a simultaneous exchange – the legal practitioner has to check whether S still has the power to dispose of the property on December 1 at 2.59.59 PM. If we add approach B, the effect would be that even if bankruptcy took place before December 1 at 3 PM, the registration of the transfer would be backdated to an earlier date, to November 24 for example. This means that the insolvency would no longer affect the transfer. If insolvency would be the only reason why owners lost the power to dispose, the legal practitioner would no longer need to perform several checks – one of which at 2.59.59 PM exactly – and can instead perform only one check when it suits the practitioner best, at some point between November 24 and December 1, which would reduce transaction costs. This is the desirable result from a vacuum instrument (see section 2). Furthermore, because P knows for sure that he will acquire the property from November 24 onwards, P may opt to pay the purchase price directly to the seller, like in German real estate practice. However, sellers may also lose their power to dispose because of a double grant. This chance may be considered theoretical, but as mentioned, ‘theoretical’ risks do sometimes materialise, and hence, private law needs a solution for this issue. If this occurs – S registers a second grant in favour of X, at 2.50 PM – the legal practitioner needs to check whether S still has the power to dispose in favour of P at some point in the 10 minutes before 3 PM. If the legal practitioner fails to do so, P may pay the purchase price to S, only to discover that X has become the owner of the property and that the seller has left the country with both the purchase price from X and P in their pocket. As mentioned, P may invoke the ‘rule against offside goals’ or a doctrine to a similar effect against X, but even if P invokes this rule successfully, X is the one that has neither the purchase price nor the property. It is needless to say that X’s claim for breach of contract against S cannot be satisfied if S is untraceable, or has gone bankrupt in the meantime. Because of this particular risk that is not covered with approach B, the benefits of vacuum instruments cannot be fully seized: a last minute check for power to dispose is still necessary.^[42] However, the small risk of a double grant – and the dangers it poses to the performance interest – can also be remedied by an insurance-oriented remedy, similar to Anglo-American conveyancing practice.

4.2.5 The golden mean

An intermediary solution might be the following. It is submitted that, for the reasons listed above, approach A – giving the purchaser a proprietary interest from the moment of sale onwards, made public by a priority notice – is the most desirable one in the context of assets that have a register to keep track of their legal status anyway, such as immovable property. In order to reduce transaction costs further, I argue that the current form of the priority notice could be simplified. Many jurisdictions provide an extensive set of rules of what must and must not be registered, in order for a priority notice to be effective.^[43] However, in order to fulfill its function, the Scots law commission is of the opinion that the scheme for priority notices could be simplified considerably without impairing its overall effectiveness. I would like to take this argument one step further and argue that it suffices to merely register the fact that a contractual obligation to transfer the asset itself, or that there is an obligation to vest/transfer/cancel a limited real right in relation to the property. Current schemes are probably often complicated because the priority notice does confer some sort of proprietary interest and is therefore – in terms of requirements for eligibility – treated like a real right. However, the German term for the origin of this doctrine – ‘Vormerkung’ – roughly translates to ‘ear tag’ or ‘caveat’. It should merely be considered as a signal to the parties or, in practice, the legal practitioner involved with the transfer, that there is something going on, and that further investigation/inquiry is necessary before proceeding with the transfer itself. The arguments in favour of the priority notice also apply to other assets of which the legal status is tracked by a register such as company shares, but it is not self-evident that all of these registers have a similar quality as the public land records. Hence, they are in legal doctrine deemed less trustworthy as an indication for proprietary interests.

For other types of assets – assets with low quality registers or assets that lack such a register at all – approach B would be the most suitable. As mentioned above, approach A is in general preferable, but approach A cannot be applied if there is no register keeping track of legal status. The benefits of approach B are still seized: the performance interest of buyers is protected at the cost of creditors, which is justified because of the difference between market value and forced-sale value. Buyers are not protected against double grants, but ‘default’ private law provides a degree of protection by instruments such as ‘the rule against offside goals’ or the bona fide purchaser doctrine. If this protection is considered insufficient, an insurance system can be set up to compensate the unlucky few that encounter a fraudulent and/or bankrupt seller. On one hand, one could argue that the risk that one party performs whilst the other does not, is impermissible, and that a simultaneous transfer is more desirable. On the other hand, assets that lack a register are in general of less value, which means that some incidents could be considered acceptable if transaction costs in general are lowered by adopting approach B.