GLOBALISATION OR ISOLATION IN NEW DUTCH PROPERTY LAW? The New Civil Code of the Netherlands and the New Civil Codes of the Netherlands Antilles and Aruba Compared Sjef van Erp Readers are reminded that this work is protected by copyright. While they are free to use the ideas expressed in it, they may not copy, distribute or publish the work or part of it, in any form, printed, electronic or otherwise, except for reasonable quoting, clearly indicating the source. Readers are permitted to make copies, electronically or printed, for personal and classroom use. 1. The new Civil Codes of the Netherlands, the Netherlands Antilles and Aruba In this annual address, I would like to make a comparative property law analysis focusing on the relations within the Kingdom of the Netherlands. The Kingdom of the Netherlands is comprised of: the so-called ‘Kingdom in Europe’ (the Netherlands), the Netherlands Antilles and Aruba. The overall constitutional structure can be found in the Statuut voor het Koninkrijk der Nederlanden (Charter for the Kingdom of the Netherlands). In Article 42 of the Charter it is stated that the rules of constitutional law of the parts of the Kingdom are laid down in the Constitution (Grondwet) of the Netherlands and the Constitution (Staatsregeling) of the Netherlands Antilles and Aruba. Article 39 of the Charter contains the principle that within the Kingdom, among other legal areas, civil and commercial law should, as far as possible, be regulated in the same way. The governments of the three constituent parts are to consult one another during the legislative process if far-reaching changes in the law are proposed. This is the so-called ‘principle of concordance’. It is explicitly not required that the law within the Kingdom be uniform, as differences between the parts may be necessary because of differing local conditions. Because of this constitutional principle of concordance, it became imperative that after the enactment of the new Dutch Civil Code a new Civil Code for the Netherlands Antilles and Aruba be drafted as well, taking the Dutch Civil Code as the model law. The ‘new’ Dutch Civil Code has now been in force for more than ten years, and it is becoming clear in which areas the Code meets the expectations that existed when it was enacted and in which areas it does not function very well. It is, therefore, quite interesting - not to say intriguing - to see to what extent the Dutch model was followed in the area of property law. My approach will be to analyse the Dutch model from the perspective of policy choices and compare the choices made in the Netherlands and in the Netherlands Antilles and Aruba. I will take the concept of unitary ownership and personal property security interests as examples. 2. Policy choices underlying the concept of unitary ownership It is a well-known characteristic of civil law systems that ownership is considered not to be fragmented. ‘Ownership’, as the old Dutch Civil Code states in Book 5, Article 1(1) (NCC Article 1(1)), ‘is the most comprehensive right which a person can have in a thing’. Also ‘dismembered rights’ (limited real rights) exist, which are rights ‘derived from a more comprehensive right, the latter being encumbered with the dismembered right’ (Book 3, Article 8 NCC). The law in mandatory form fixes the number and content of real rights; this is the so-called numerus clausus of absolute rights. Of course, also in a civil law system several people can be the owners of one object at the same time. This is not, however, a case of fragmentation but a sharing of ownership: co-ownership. This means that all the owners have the same rights and are burdened by the same duties, but according to their share. Furthermore, no distinction between major (‘common law’) and minor (‘equitable’) interests is made. The civil law principle of unitary ownership has not only been reaffirmed as the basis of new Dutch property law, it was even decided that this principle should be maintained more strictly than under the old Civil Code. As a consequence, the introduction of the English and American concept of trust into Dutch law was explicitly not made possible and the transfer of ownership for security purposes, accepted by Dutch case law under the influence of German case law, became no longer allowed. The pivotal provision is Book 3, Article 84(3) NCC, the so-called fiducia ban, not permitting any transfer of ownership that is not ‘real’ (complete). Transfer of ownership for security purposes was used to evade the provisions on pledge, which under the old Civil Code could only be created in a possessory form. To replace this type of transfer, the non-possessory pledge was introduced. Originally, this was to be a publicly registered pledge, but after an intense debate the new non-possessory pledge was to be created secretly. Although such a pledge must still be registered, this has to be done in a non-public tax register merely to certify the date. Remarkably enough, the new Civil Codes of the Netherlands Antilles and Aruba do not contain this fiducia ban. The Netherlands Antilles and Aruba still accept a transfer of ownership for security purposes. It is even discussed whether it should be made possible to create a trust under the law of the Netherlands Antilles. It is also quite interesting to note that the Dutch non-possessory pledge has been adopted, though without the limited registration at the tax office. Several reasons are given for this deviation from the Dutch model. In the following sections, I will first examine briefly the background of the Dutch debate on unitary ownership, followed by the continuing criticism made about this choice and the reasons why the legislatures of the Netherlands Antilles and Aruba refused to follow the Dutch example. Finally, I will consider the developments in the light of globalisation. My conclusion will be that the fiducia ban in the Dutch Civil Code will prove to be a futile attempt to maintain the purity of a civil property law system in a world characterised by regional and global integration of markets and the worldwide legal integration which results from this economic process. 3. Policy choices underlying the law of personal property security interests Under the old Dutch Civil Code, it was accepted through case law that a transfer of ownership for security purposes as such was valid. Although such a juridical act was a transfer from the perspective of form, from the perspective of substance, however, it was the creation of a non-possessory security interest. If the property transferred were goods, the debtor/transferor did not lose control and if they were claims, the debtor of the claim was not informed. As a pledge, this was an invalid transaction given the publicity requirements that apply here. Pledging of goods could (and can) only be done by bringing the goods under the control of the creditor/pledgee or a third person agreed upon by the parties (Article 1198(1) of the old NCC, Book 3, Article 236(1) of the new NCC). In the case of claims, a deed had (and has) to be drawn up and notice of this deed must be given to the debtor of the claim (Article 1198 of the old NCC and Book 3, Article 236(2) in conjunction with Book 3, Article 94(1) of the new NCC). As a transfer or an assignment, however, the transaction was valid. Goods can be delivered constituto posessorio: the transferor is still in control of the goods, though no longer as possessor, merely as ‘holder’ (détenteur). Under the old Civil Code, giving notice to the debtor of the assigned claim was not a conditio sine qua non for assignment. Not informing the debtor, however, resulted in the protection of the debtor paying bona fide to the assignor and not the assignee. The ownership of the creditor/transferor for security purposes was not, however, complete. If the debtor did not pay his debts, the rules on pledge applied as far as possible to any forced sale of the goods or claims. The transferor was not able to invoke his ownership rights against certain privileged creditors. Parallel to this use of ownership for security purposes, old Dutch property law allowed the broad retention of title clauses. Clearly, the underlying policy was that ownership, whether transferred or retained, could be used for security purposes. This created an atmosphere in which a limited form of ownership fragmentation became accepted. The debtor/transferee for security purposes could be seen as ‘economic owner’, the creditor/transferor was the formal owner. In the case of retention of title, the seller/owner was the formal owner, the buyer/‘holder’ of the goods sold and delivered was the ‘economic owner’, especially if he was allowed to dispose of the goods in the ordinary course of business. This all changed when the new Civil Code entered into force, resulting in a fundamental policy change. In Book 3, Article 84(3) of the new NCC it is stated: A juridical act which is intended to transfer property for purposes of security or which does not have the purpose of bringing the property into the patrimony of the acquirer, after transfer, does not constitute valid title for transfer of that property. With regard to the assignment of claims, Book 3, Article 94 now provides: In cases other than those provided for in the preceding article [viz. rights payable to bearer or to order, the instrument of which is under the control of the transferee, JvE], rights to be exercised against one or more specifically determined persons are delivered by means of a deed intended for that purpose and notice thereof given by the alienator or acquirer to those persons. Furthermore, Book 3, Article 92 of the NCC severely curtails the possibilities to create retention of title. The underlying policy change is clear: ownership should no longer be used as security. There is only one exception: if an existing right of ownership is used as security, but even in such a case this is possible only to a limited degree. The policy change is in conformity with the ideas of Meijers, the auctor intellectualis of the main structure and basic underlying principles of the new Civil Code. In his De algemene begrippen van het burgerlijk recht (the general principles of civil law), Meijers wonders whether it makes sense in a civil law system to accept both pledge and ownership limited vis-à-vis third parties, such as ownership for security purposes. His answer is distinctly negative: in a general theory of law (or civil law) such a distinction is ‘foolishness’. He reiterated this approach during the parliamentary debates on the new Civil Code. If it is desirable that property be transferred for security purposes, a pledge should be created; if property is to be administered for the benefit of a person a bewind (administration) should be established. In his view, rules of mandatory law which have become obsolete should be replaced by new rules; if they are not, evasion of the law would undermine the authority of mandatory law. When one reads the parliamentary documents and debates, one gets the strong impression that, with respect to property law, the drafters of the new Civil Code wanted to return to principles of pure civil law. The policy choices that resulted from this thought process are the following: (1) ownership is the most complete absolute right and is, as such, unitary in nature; (2) a pre-formulated statutory list of other absolute rights, with pre-formulated mandatory contents (numerus clausus of absolute rights) must be strictly adhered to; (3) these other absolute rights (‘limited real rights’) are burdens resting on the right of ownership, not rights that are split off from the full right of ownership; and (4) freedom of contract does not apply to the creation of limited real rights, except in so far as the law explicitly allows the parties to give shape to these rights. It will be clear that these policies are interrelated. If, e.g., parties are free to create new limited real rights, numerus clausus has become simply a model and fragmentation of ownership can no longer be stopped. The above meant that a new statutory framework had to be created to allow the needs of legal practice to create non-possessory security interests and trust/beneficiary relationships to be met. This was done by the creation of the non-possessory pledge and by the introduction of special rules on administration, together with special rules on irrevocable powers of attorney. As to pledges, these rules can be found in Book 3, Articles 237 and 239 of the NCC. It may be helpful to cite the first paragraphs of these Articles, as they provide the general framework. Article 237(1) reads: The right of pledge on a moveable thing, on a right payable to bearer, or on the usufruct of such a thing or right, can also be established by an authentic deed or a registered deed under private writing, without the thing or the document to bearer being brought under the control of the pledgee or of a third person. Article 239(1) states: A right of pledge on a right which can be exercised against one or more specifically determined persons and which is not payable to bearer or order, or a right of pledge on the usufruct of such a right, can also be established by an authentic deed or a registered deed under private writing without notification thereof to those persons, provided that the right in question already exists at the time of the establishment of the right of pledge or will be directly acquired pursuant to a juridical relationship already existing at that time. In contrast to what Meijers originally proposed (the introduction of a so-called registered pledge), the non-possessory pledge is not registered in a public register, but with the tax authorities, merely to certify the date. In this respect, old and new Dutch property law still advocate secret personal property security interests and no policy change took place. Transfer of ownership for security purposes is not made public and retention of title or a non-possessory pledge is not publicly registered. Meijers’ attempt to introduce the principle of transparency (publicity) in this area of the law was (and, even more so, today still is) completely justified in the light of global developments; I need only refer to Article 9 of the U.S. Uniform Commercial Code and the 2002 Cape Town Convention on Mobile Equipment. Also, the influence of European law is now felt in this area, as it may very well be argued that secret security interests violate the European freedoms, especially the freedom of capital as it is laid down in Article 56 EC. Considering the developments in the Netherlands in hindsight, it is amazing to see that, on the one hand, an attempt was made to purify property law as it is laid down in the new Civil Code (cf. the fiducia ban), whereas on the other hand the - justified! - purification attempt with regard to the publication of personal property security interests failed. It is submitted that the only explanation for this apparent contradiction is that in spite of all the comparative legal research that was done to prepare the new Code, the final debates only focused on the domestic situation. The incoming tide of non-Dutch legal rules (either uniform law or law coming from a major economy such as the United States) resulting from regional and global economic integration was, at the end of the day, ignored. 4. Did the return to strict civil law in the new Dutch Civil Code work? Meijers developed the underlying principles of the new Dutch Civil Code in the 1940s and 1950s. In this period, market integration had only taken place within the structures of the various colonial empires. Each colonial power imposed its own Western system of law (French law, English law, Dutch law, etc.) upon its colonies. As a result, no need arose to accommodate civil law to common law or vice versa. Only as a result of the creation of the European Communities did legal integration within Europe become necessary, first of all between the various systems of civil law, next also between civil law and common law. In the following decades regional integration became a part of the global integration of markets and legal integration also became necessary at a global level. Given the strength of the American economy, globalisation not infrequently means Americanisation, also in the law. It may, therefore, be said that already at the time when the new Dutch Civil Code was enacted, several decades after the first drafts, the policy choices made in the area of property law were based on economic premises which no longer applied. Particularly the influence of Anglo-American property concepts is constantly increasing; I need only refer to derivative instruments (such as options, futures and swaps), which are standardised contracts whose value is based on the performance of underlying values (a financial asset, index or other investment). Such contractual values are considered to be assets that can be traded on an exchange. In many respects, particularly at the level of policies, principles and concepts, English and American real and personal property law differ from civil property law. It is, e.g., a well-known characteristic of Anglo-American property law that ownership is not a unitary concept; parties can fragment ownership and no numerus clausus in the civil law sense exists. Another example is the lease, a property interest under Anglo-American law, which is a contract under civil law. Under the growing pressure of economic integration, a certain cross-fertilisation is taking place between civil law and common law, which can be seen both at the level of private international law (the growing recognition and accommodation of foreign legal phenomena and acceptance of uniform private international law rules) and at the level of substantive law (growing harmonisation and unification). A famous (or infamous?) example of what is happening at the private international law level is the recognition of the common law trust by civil law countries. For civil law systems, it is a major problem who should be considered the ‘owner’ of trust property. To solve these problems, the Hague Trust Convention was drafted, which entered into force in the Netherlands on 1 February 1996. A trust/beneficiary regime violates the fiducia ban, as it is, from a civil law viewpoint, a transfer of ownership with limited purposes (it is not a true or ‘real’ transfer, as its purpose is division between entitlement to benefits and the management thereof). Consequently, the Act to implement the provisions of the Hague Trust Convention states in Article 4 that the legal consequences of the recognition of a trust according to Article 11 of the Convention is not to be overridden by provisions of Dutch law concerning the transfer of ownership, security interests or the protection of creditors in the case of insolvency. A further statutory exception to Book 3, Article 84(3) of the NCC was made as a consequence of the participation of the Netherlands in the European and Monetary Union. By special statute, the fiducia ban was declared non-applicable to repo transactions. In the Explanatory Memorandum it is stated that this statutory provision is necessary to avoid interpretation problems and that the special statutory provision should not be seen as an exception to Book 3, Article 84(3) of the NCC. Whether ‘interpretation’ or not, this Article does not apply to repo transactions. Also, the Bill on financial collateral arrangements contains a provision declaring the fiducia ban non-applicable to transactions made in performance of such arrangements. Finally, when the legislature wanted to give rules on so-called qualitative (third-party) accounts held by notaries and bailiffs, serious classification problems arose. The new Acts on notaries and bailiffs contain provisions which in essence provide that clients who deposit money with a notary or a bailiff as a neutral third party are the only persons entitled to the money; however, they do not have the power to dispose of the money. Furthermore, it is explicitly stated that a notary or a bailiff is not entitled to the money, which gives the clients protection if the notary or the bailiff becomes insolvent, but the notary and the bailiff do have the exclusive power to dispose. Is this fragmented ownership or full ownership of the client with a statutory irrevocable exclusive power of attorney given to the notary or the bailiff? In the parliamentary documents accompanying the Bill, this approach is defended. Already complicated questions have arisen as to what exactly the entitlement of the client is. These are, however, not the only problems, which so far have arisen in connection with the policy choices and the principle upon which the new property law is founded. As could, and probably should, have been expected, the fiducia ban also led to perplexing difficulties with regard to sale and lease-back arrangements. Are these ‘true’ or ‘real’ transfers or not? The Supreme Court gave a rather formalistic reply. If the arrangement is such that upon default of the debtor/transferor the creditor/transferee is allowed to terminate the lease and is then fully entitled to act as owner, the arrangement does not violate Book 3, Article 84(3) of the NCC. Being fully entitled to act as owner means that, in the case of a sale, the debtor is not entitled to any surplus. In other words, a positive conclusion that someone is a ‘true’ owner is based on a negative examination of the rights of the debtor. In practice, careful drafting may avoid problems here and will result in a perfectly legal arrangement. The punishment for the less careful drafter will be invalidity of the agreement. The same desire to stay strictly within the limits set by the policy choices behind the new Civil Code can be found in a recent case, in which the Supreme Court decided that a contractual clause limiting the transferability of a claim is binding upon third parties, because it affects the claim itself as the object that is to be transferred and does not affect, as such, the power to dispose. Therefore, no bona fide third-party protection rule can be applied in cases where a claim is being pledged in violation of such a clause and the pledgor did not in fact know (and could not have known) of the non-transferability clause. Let me pause here for a moment, to remark that even the present United States Supreme Court, a majority of which is undoubtedly inspired by a more conservative ideology, rejects this type of formalist reasoning and takes a more legal realist approach. As far as property law is concerned, I refer to the recent United States v. Sandra Craft case, in which the question was decided whether the possession of individual rights in a tenancy-by-the-entirety estate can be seen as ‘property’ belonging to one of the spouses that may fall under a lien in favour of the federal Internal Revenue Service. The majority of the Court answered this question positively, considering the substance of the right under discussion, not its form. Finally, a fairly recent development that should be mentioned here is the proposal to amend the above-cited provision on assignment of claims (viz. that notice to the claim debtor is a conditio sine qua non) and restore - in cases of securitisation and factoring - the rule under the old Dutch Civil Code that for an assignment to be valid the claim debtor need not be informed about the assignment. The new provision, as proposed, reads: These rights [viz. rights to be exercised against one or more specifically determined persons, JvE] can also be delivered by an authentic deed or a registered deed under private writing, which authentic or registered deed must be drawn up for that purpose, without notification thereof to the persons against whom those rights must be exercised, provided that these rights already exist at the time of the delivery or will be directly acquired pursuant to a juridical relationship already existing at that time. The delivery cannot be invoked against the persons against whom these rights must be exercised, except after notification of the delivery to those persons by the assignor or the assignee. To justify this radical change less than fifteen years after the enactment of the new Civil Code, the Explanatory Memorandum argues that the creation of a non-possessory pledge in cases of securitisation and factoring is not what the parties want. What they want is a complete assignment, without the claim debtor having knowledge of the assignment, which they cannot attain as this requires notification of the claim debtor. Therefore, the notification requirement must be abandoned. No mention is made of the impact of the fiducia ban: if the assignment were to be considered not ‘real’ or not ‘true’, it would still be invalid for that reason. Or should we read the Explanatory Memorandum as implying that also Book 3, Article 84(3) no longer applies in cases of securitisation and factoring? It is clear that the internal coherence of the system is at risk. It will be clear from the above that present Dutch property law, as laid down in the Civil Code, can be characterised as the result of an attempt to purify the civil law and re-establish statutory law as the prime source of legal rules for market participants. This attempt is also reflected in certain property law cases, in which rather formalistic reasoning is used to maintain the underlying principles of and the policies behind the new Civil Code. Of course it is understandable that the Supreme Court is extremely hesitant to go against such policies and principles only a few years after a new Civil Code entered into force and after a prolonged and at times heated political debate. However, that changes in Dutch property law are really necessary can be seen clearly when we look at the re-codification process in the Netherlands Antilles and Aruba. 5. No return to strict civil law in the Netherlands Antilles and Aruba As already remarked above, the major changes resulting from the re-codification in the Netherlands made it necessary to re-codify civil and commercial law in the Netherlands Antilles and Aruba. In that process, the first experiences in the Netherlands could be taken into account. Also, local conditions are different. The Netherlands Antilles and Aruba are Caribbean islands with a large offshore financial industry and contacts with the United States are intense. Financial institutions were afraid that the policy changes in the area of property law that characterised the new NCC might result in legal insecurity and might jeopardise their business prospects. This led to a dualist approach regarding the principles underlying and the policies behind the re-codified property law. First of all, ownership can still be used for security purposes. This applies both to transfer of ownership for security purposes and to retention of title, which can still be used broadly. In other words, Book 3, Article 84 of the NACC and the ACC does not contain a fiducia ban. As far as the assignment of claims is concerned, Book 3, Article 94(1) of the NACC and the ACC gives the same provision as the corresponding article of the NCC. However, Book 3, Article 94(3) of the NACC and the ACC provides the following exception: The assignment of a claim mentioned in paragraph (1) for the purpose of securing payment of a money claim does not require notification thereof, provided that the right in question already exists at the time of the assignment or will be directly acquired pursuant to a juridical relationship already existing at that time. In this way, the transfer (assignment) of ownership for security purposes is still possible in exactly the same way as under the old Civil Code. Assignment is complete after the deed of assignment has been signed. The debtor of the assigned claim who pays the assignor in good faith is protected against any claim for payment by the assignee. Secondly, non-possessory pledges also are allowed, but with even fewer requirements than under the new NCC. As stated above, Book 3, Articles 237 and 239 of the NCC require that the deed by which the non-possessory pledge is created is either in authentic (notarial) form or in the form of a registered deed under private writing. Such a formal requirement is absent in the NACC and the ACC. The reasons for this deviation from Dutch law are in part of a more principled nature and in part cost-inspired. It was feared that otherwise parties would always choose transfer of ownership for security purposes, as no formal validity requirement applies to such an arrangement, and that a system of registration would have to be set up particularly for this purpose. It is remarkable that, under the influence of regional and local market conditions (particularly the influence of Anglo-American law), the policy shift which took place in the Netherlands did not occur in the Netherlands Antilles and Aruba. Even more remarkable is that the proposal has been made to take over the amendments made to Book 3, Article 94 of the NCC. It is unclear, as I remarked above, whether this also means that such an assignment can also be made for security purposes. It should, finally, be added that in the Netherlands Antilles the introduction of trust arrangements is now under consideration. Such a development is, under the law of the Netherlands Antilles and Aruba, not blocked by a general and blind fiducia ban. 6. Concluding remarks The fundamental changes in property law, as proposed by Meijers and laid down in the new Netherlands Civil Code have resulted in an isolationist tendency in Dutch property law. The changes were characteristic of a period in which it was felt that a return to a purer civil law, laid down in a comprehensive code, was vital for the well-functioning of the Dutch civil law system. But since the 1940s and 1950s much has changed. European economic and legal integration has reached a stage that 50 years ago would have been difficult to imagine. The worldwide integration of markets has led to a growing influence of US law as a result of the enormous strength of the American economy. Anglo-American property principles and concepts are now entering the civil law world and need somehow be accommodated. This pressure was felt more intensely in the Netherlands Antilles and Aruba at the time of re-codification. The result is that the paradigmatic policy shift of Dutch property law did not take place in the Netherlands Antilles and Aruba. Here a dualist approach can be seen. The old policies and principles are still adhered to, as they enable the accommodation of Anglo-American concepts, such as the trust. The new policies and principles can also be found in the new Netherlands Antilles and Aruban Civil Codes, though to a very limited degree. Non-possessory pledges are now also possible, though not as the exclusive tool to create non-possessory security. The conclusion can only be that, at the end of the day, in the Netherlands Antilles and Aruba the old policies and principles prevailed. Given the problems that arise in Dutch property law as a consequence of strict adherence to the new policies and principles, a systematic rethinking of those choices could be very fruitful. Ad hoc problem solving, as seems to be taking place at present, may create more problems than will be solved. I am convinced that such a systematic rethinking would lead to a more flexible and transparent property law.