Security is the process by which a creditor obtains an extra source for the repayment of his debt. In the financial context, it is an essential part of what may loosely be described as the enforcement of claims. This paper will seek to identify some of the main features of security in the light of current international trends. The subject is of interest to central bankers in their capacity as custodian of their countries’ financial systems. It is not proposed to enter into the debate as to whether security is necessary, desirable, or efficient.2 It is true that much lending is unsecured. Markets have sprung up to provide ready sources of unsecured credit, such as the commercial paper market. However, lenders like security, if they can get it, and this is the commercial reality of the situation.

Security is the process by which a creditor obtains an extra source for the repayment of his debt. In the financial context, it is an essential part of what may loosely be described as the enforcement of claims. This paper will seek to identify some of the main features of security in the light of current international trends. The subject is of interest to central bankers in their capacity as custodian of their countries’ financial systems. It is not proposed to enter into the debate as to whether security is necessary, desirable, or efficient.2 It is true that much lending is unsecured. Markets have sprung up to provide ready sources of unsecured credit, such as the commercial paper market. However, lenders like security, if they can get it, and this is the commercial reality of the situation.

On the other hand, the whim or desire of lenders is not in itself a sufficient justification for the great efforts being made in many countries to modernise their law of security. China’s law on the creation and use of security (the PRC Security Law) which came into effect on October 1, 1995 states as its goal

… to promote the availability of financing and the circulation of goods, to safeguard the fulfilment of creditors’ rights and to develop the socialist market economy.3

Leaving aside the word “socialist,” which now only applies in the case of a few countries, this serves as a readily acceptable economic rationale for the existence and proper functioning of the law of security.

The law of security makes little sense viewed in the abstract. At least four other areas of the law are necessary to breathe life into it.

  • The law of contract. This will define the secured obligation, and the events of default entitling the lender to look to the security to satisfy the debt.

  • The law of property. This will define the nature of property available as security. To revert to the example of China, the state is ultimately the owner of land, hence land itself (as opposed to the exclusive right to use it) is unavailable as security. Whether this makes much difference in practice may be debatable,4 but the point is that the law of property must be looked to to ascertain the rights encompassed within the collateral. More broadly, the law of property will define the nature of the charged rights, for example whether rights subsist in stocks and shares, or merely in the contractual obligation of the custodian to deliver stocks and shares, to foreshadow an example that is discussed below.

  • The law of civil procedure. It is self-evident that a well-drafted law of security is useless without efficient court procedures to implement it.

  • The law of bankruptcy. Bankruptcy is the process by which a debtor’s assets are liquidated for the benefit of the general body of creditors. A secured claim needs to survive that process intact, though rules differ as between legal systems. The law of bankruptcy may also impact on the validity of security, setting it aside in circumstances where the taking of the security amounts to an unfair preference.

One final preliminary point is this. Good lending practice dictates that loans be extended on the basis of the creditworthiness of the borrower, and the viability of the project expected to produce the cash flow to repay the loan. It is a mistake to allow lending decisions to be unduly influenced by the existence of collateral. Experience shows that even in the most advanced legal systems, the enforcement and realisation of security is often problematic. Disputes may arise as to ownership of assets, as to the amount of the secured debt, as to the validity of the acts of agents appointed to realise the security whether in or out of court, as to whether perfection requirements (sometimes technical) have been complied with, as to whether proper steps have been taken to achieve the best price, and so forth. In cross-border transactions, the room for impediment is even greater. This is certainly not an argument against the taking of security. It is, however, a reason not to place undue reliance on it.

Meaning of Security

The prime feature of security is that the property, the subject matter of the security, is available to the creditor to satisfy the secured debt, and is unavailable to other creditors in the event of the debtor’s bankruptcy. The key concept therefore is that property is, in a sense, earmarked for the claims of a particular creditor. However, this apparently simple concept raises in practice a number of issues such as the following.

  • Terminology. A wide number of terms are in common use even in the English language to denote security interests: charge, mortgage, pledge, hypothec, lien, security assignment, fiduciary transfer, etc. Sometimes these terms have a very specific meaning in national laws. For example, the essence of a pledge is that it is possessory, whereas a mortgage is nonpossessory. In international lending, it is wise not to place too much weight on the word used, but rather focus on the nature of the particular security created. Some systems seek to avoid the problem by a uniform term describing a more or less uniform security. For example, Article 9 of the United States’ Uniform Commercial Code (UCC) creates a single “security interest” in movables. The European Bank for Reconstruction and Development (EBRD) Model Law on Secured Transactions is based on the idea of a single right described as a “charge.” The approach of these Codes goes far beyond terminology, of course, and they have been trailblazers in the simplification and rationalisation of security law.5 It should, however, be noted that simplification of the concepts does not eliminate practical distinctions. There is a real difference in practice between (for example) security in the form of marketable shares, which can be disposed of unilaterally by the lender, and security over tangible movables, which may require court assistance for effective realisation.

  • Types of property. Most, if not all, legal systems draw a distinction between immovables (land) and movables (other forms of property). Subject to that, virtually every type of property that has an economic value is potentially available as security: land, goods, debt and equity securities, contractual claims, cash proceeds, inventory, receivables, and so forth. The diversity of available property, and the development of modern financing techniques, can be hard to accommodate within existing legal concepts. For example, under the German Civil Code (BGB) security over movables is taken by pledge involving the transfer of possession. In practice, however, German law does recognise nonpossessory security rights in the form of the fiduciary transfer of title to movables and rights capable of being transferred separately.6

  • Specificity. A key issue is the extent to which property needs to be identified before it can be charged. Many jurisdictions recognise concepts similar to the English “floating charge.” This type of charge is wholly nonspecific, covering matters such as inventory, receivables, and equipment. The charge “hovers,” leaving the chargor free to get on with his business, except in the case of default when it “crystallises” attaching to the available assets. The specificity issue turns up in practice again and again. For example, in the case of a charge over shares, is a general reference to type and number of shares sufficient? If so, how are competing rights to be dealt with in the case of a shortfall? A distinct but similar issue is the availability of security to secure fluctuating, future, and contingent claims. Again, such security is widely available internationally. But there are practical problems. In the case of bankruptcy, is it permissible to enforce against the security in respect of a contingent liability that has not yet occurred and, if so, how is the liability to be valued?

  • Security and quasi-security. Many financing techniques do not formally involve the taking of security at all. Sale and repurchase agreements (repos) and finance leases from an economic perspective may be little different from an unsecured term loan on the one hand and a loan secured on movables on the other. Setoff, or offset, is another example. A separate debt owed by creditor to debtor may provide functional security in the sense that one claim can be recovered by the simple expedient of refusing to repay the other. This in turn gives rise to issues of characterisation. These are not by any means purely academic. Whether a transaction security is characterised as security or not may affect perfection requirements, which may in turn affect validity. Here there is a difference of approach between, for example, UCC Article 9 which seems to look at the substance of the transaction, and English law, which will only look behind the form of the transaction in the case of a sham. Generally speaking, it has to be said that lenders prefer the latter approach since in a carefully structured transaction they will wish rights and obligations to be dealt with according to the characterisation that they have chosen, rather than a substitute characterisation imposed by a court.

There is, of course, a distinction between “securities” in the sense of collateral (which is the subject of the present discussion), and a “security” in the sense of debt, equity, or other securities (i.e., shares, government debt, etc.). The process of “securitisation” (involving the pooling and repackaging of loans into securities, which are then sold to investors) is also distinct and may be backed by complex collateral arrangements involving trusts (in jurisdictions recognising the trust) and collateral agents elsewhere.7 Securitisation will not be considered further here.

Contemporary Importance of Security Law

There has been an upsurge of interest in the law of security, equaled only by that in the law of bankruptcy. The reasons for this are partly economic. It has been pointed out that:

Until the debt crisis of the 1980s, a large share of public debt financing to developing countries was provided in the form of sovereign lending. Today, some medium- to long-term lending is still provided to, or guaranteed by, those countries or their governments or governmental agencies. However, in aggregate, most current debt financing does not benefit from sovereign or governmental guarantees. There are many reasons for this shift. Two in particular stand out: the budgets of these countries cannot sustain the level of borrowings needed, and with increased privatisation of assets that were once in the public sector, governments are not prepared to underwrite the risks of the (newly) privatised enterprises. Increasingly, therefore, providers of medium- and long-term debt financing to developing and transitional market economies are required to look to means other than sovereign backing to secure their lending. These lenders are especially looking to obtaining security for their lending over the assets of the cash flows of borrowing enterprises.8

So the first reason for the increased level of interest is that a modern security law is seen as attracting inward capital.

The second reason is to do with the transition of many countries from a socialist to a market economy. In most cases, the law relating to security required considerable modernisation. It should not be overlooked that these countries have their own legal traditions, which were by no means extinguished during the socialist period. To take as an example Romania, much of the law of contract is found in the Codul Civil of 1865, which is closely based on the French Civil Code, and the Codul Commercial of 1887, which is based on the Italian commercial code. The defects in the law of Romania identified at the beginning of the 1990s tended to reflect a failure to keep pace with developments in the postwar period. For example, it did not appear specifically to allow for modern French law rules under which the owner of a business (fonds de commerce) can create a charge (nantissement) over the business by an agreement that is registered at the local commercial court. Similarly, the law did not appear to take account of modern French practice in relation to charges over receivables, in particular pledges by way of subrogation conventionelle (contractual subrogation, in other words an agreement between pledgor and pledgee that the pledgee shall be substituted for the pledgor in his relation to the debtor). Reforming the law of security requires an appreciation of the wider legal culture in the country concerned, particularly as it relates to the law of contract, property, and bankruptcy for reasons already mentioned.

The third factor that has stimulated interest is the perception of the importance of law in promoting financial stability. This factor is likely to be of particular interest to central banks and their lawyers. In response to an initiative taken at the Lyon summit in June 1996, representatives of the countries in the Group of Ten and of emerging market economies jointly sought to develop a strategy for fostering financial stability in countries experiencing rapid economic growth and undergoing substantial changes in their financial systems. Their report highlights “ability to pledge and seize collateral” in the illustrative list of indicators of robust financial systems.9 The result of all this is to place an effective security law firmly on the list of indicia of a satisfactorily functioning economy.

Security Laws Around the World

Various classifications of securities laws may be made.10 One distinction is between common law and civil law traditions. Another focuses on functional distinctions, for example whether a system is “creditor-friendly,” and whether a universal charge is available, and on the treatment of security in bankruptcy. In fact, it is believed that the wide differences between national systems makes useful classification extremely difficult. That said, the essence of security (at least in an economic sense) is a constant. So, despite the wide divergence between legal systems, there is a growing interest in comparing how different legal systems deal with what are essentially the same issues. These are primarily what property is to be available as security, how security is to be taken, what perfection requirements are to be imposed, and how rights of secured creditors are to rank above those of other creditors.

The particular challenges faced by different legal systems may differ, but for present purposes three are particularly important and are shared in common and shall be discussed further.

  • Developing financing techniques. Often, these changes can be adequately accommodated within existing law and given effect by the courts (the German fiduciary transfer being an example), but sometimes specific legislative change becomes necessary. Examples are the revisions discussed below to Article 8 of the UCC, and the French Law of July 2, 1996, both of which are concerned to adapt the law to changing methods of holding securities. Some subjects remain relatively unexplored, such as the availability of intellectual property as security11 and the impact of technological developments like the Internet, and change may be anticipated here.

  • Cross-border security. This is important in international lending and will become increasingly important in Europe with the advent of the single currency. Issues, such as characterisation and perfection requirements, are generally resolved by reference to the lex situs, but this rule can be difficult to apply in cross-border transactions in which (for example) cash flow received in one jurisdiction may be paid into a secured account in another jurisdiction, or stock issued by companies of a number of different nationalities held as security.

  • Policy. There is a range of legitimate policy responses to issues arising in connection with security law such as the extent to which a jurisdiction should be “creditor-friendly,” the publicising of security, the circumstances in which security should be challengeable on the grounds of preference, and so forth. It needs to be kept in mind that there is no “right” answer to these issues. In writings and other analysis, policy issues are often considered on the grand scale, in relation to the overall objectives of an economic system. It should not be forgotten that they arise at the consumer level as well. In recent years, courts in England and Germany (and other jurisdictions as well) have had to grapple with the extent to which security granted by one member of a family to another, and debts secured on the family home, should be treated as subject to special rules.12 The courts in both these countries have placed restrictions on the recoverability of security in these cases for policy reasons. In essence, the courts are concerned to see that family relationships are not abused, and that limitations are placed on the ability of creditors to repossess the family home. The result is that the security is much harder to realise. Similarly, it is not surprising that economic difficulties in emerging economies create obstacles to the realisation of security. During the peso crisis of 1994, Mexican interest rates rose dramatically, hugely increasing the burden of debt repayment on domestic householders. In such circumstances, whatever the content of the law, practical considerations are bound to operate on courts, governments, and financial institutions, since no system will wish to see large numbers of people become homeless.

Nature of the Law of Security

The law of security is essentially about methods of taking and realising collateral. Generally this implies a proprietary right on the part of the creditor in the property constituting the collateral. To that extent, the law of security straddles the law of property and the law of obligations. An exception is the guarantee, which involves the personal promise of the guarantor without in itself creating any interests in his property. Lenders regard guarantees as “security,” but on a strictly conceptual basis they fall outside the law of security. In the following section, some of the principal features of the law of security are identified.

Types of Security Interest Available

This has been briefly touched on already. Most systems draw a distinction between possessory and nonpossessory security interests (pledge and mortgage). The distinction may affect realisation, in the sense that judicial approval may be required in the case of the latter but not the former. Beyond that broad division, the precise nature of security interests available under different legal systems becomes extremely technical. For obvious practical reasons, the nonpossessory interest has long been much more important than the possessory. However, strict legal concepts are often conveniently adapted to the circumstances. For example, in U.S. and English law, a bank issuing a letter of credit is considered as having a pledge over the shipping documents. To enable the goods to be sold on by the importer, the law considers the bank’s rights as pledgee continuing even after the release of the shipping documents under a “trust receipt.” This convenient fiction oils the wheels of commerce, though it is far from ideal from the perspective of the bank, because inevitably its security rights do not prevail against a purchaser for value without notice of its interest, and in the nature of the transaction the goods are likely to be sold on.

Types of Property Available

Again, this has been mentioned above. The tendency has been one of continual enlargement, though it may be noted that not all measures of law reform are intended to expand the frontiers of security. In the Netherlands, under the new Civil Code of 1992 it is not possible to create general security for future loans and in future assets, certainly not with priority as of the registration date. General floating charges are therefore not easily created either.13 Generally speaking, however, the trend is toward the floating charge largely because it is extremely attractive to both creditor and debtor as involving a “one-stop” security, whilst leaving the debtor free to carry on his business. In Japan, for example, a special kind of hypothec, which had developed in practice and was acknowledged in the prewar period, was incorporated into the Civil Code in 1971. This is called the floating hypothec or base hypothec, which secures unspecified obligations within a fixed limit. In the case of an ordinary hypothec, when the obligation is fulfilled, the hypothec is extinguished, while in the case of a floating hypothec it does not extinguish but covers future obligations whenever they arise.14


The security agreement is often required to be in writing, though sometimes the physical deposit (e.g., of title deeds or share certificates) is sufficient to create security provided that the deposit is made with the requisite intention. In addition to these formal requirements, there may be substantive requirements, such as the common law insistence on consideration (i.e., value given by the party taking the security), though this is not necessary where the transaction is recorded by deed. (Issues such as capacity are normally dealt with under the law of obligations.) Another important practical point in some jurisdictions (e.g., India) relates to stamping requirements under the Stamp Acts. Failure to comply where required may result in the creditor being unable to rely on the instrument in court.


Perfection denotes the process by which a security agreement becomes binding against third parties generally. It is associated with the concept of “publicity,” by which in theory a prospective creditor can find out what charges already exist on the prospective debtor’s property. In the case of land, title is typically subject to some form of public registration, and charges over the land are treated as ancillary to the registration process. However, in the case of movables, international patterns differ widely. In the rare case where physical possession is retained by the creditor, publicity is not an issue (there is no need, for example, for a filing under the UCC Article 9 rules). In the case of nonpossessory security over movables, Article 9 generally requires perfection by filing a notice of the interest in the state register, though the detailed rules are complex.

In England, registration is required for most but not all types of security created over the property of a company. There are provisions for the registration of chattel mortgages granted by individuals,15 but such mortgages are rare. This is in part because the legal requirements imposed last century on account of “false wealth” concerns proved to be commercially inconvenient. They provided a powerful spur for the creation of alternative financing techniques such as hire-purchase and leasing.

Germany, by contrast to the United Kingdom and the United States, has a more relaxed attitude to publicity in the case of security over movables. Under China’s Security Law of 1995, security over some types of property must be registered for the mortgage to be valid,16 but registration of all mortgages is advisable for priority purposes.17 The EBRD Model adopts a purist line, taking its stand:

… on the principle that charges are a matter for public knowledge. Since Roman law, the creation of secret rights and assets has been disfavoured. A person who gives assets as security but does not indicate this to his creditors creates an impression of “false wealth.” The Model achieves publicity mainly by relying on registration of charges at a separate registry.

Whilst laudable, it must be accepted that registration requirements in request of movables (particularly those with short life spans) can be burdensome and difficult to manage, even with the benefit of new technology. There is essentially a trade-off: ease of commerce, against the undesirability of false wealth, and in the end this is a policy choice.

In practice, perfection requirements rank high on the list of concerns of creditors intent on taking security. It is essential to comply with all applicable legal requirements so as to ensure the validity of the security. Sometimes, for various reasons, registration may be unacceptable, which may influence the form, or even the situs of the transaction, possibly relocating it offshore.


Realisation, or enforcement of the security, is the essence of security law. It is the means by which the creditor transforms a juridical advantage into hard cash. The main distinction is between self-help systems (the English system being a paradigm), and systems in which some form of judicial intervention is required, for example a judicial auction of the secured assets. Under the English system, on default a creditor with the benefit of a floating charge is entitled to appoint “receivers,” who take over the debtor’s business with a view to obtaining repayment.18 In practice, the self-help nature of the English system is qualified; for example, a creditor is prohibited from recovering possession of a dwelling house without a court order, and the court has wide powers to postpone possession on terms. Likewise, in countries frowning on self-help, judicial auction is not always required.

The EBRD Model adopts an ingenious compromise. It relies in the first instance on self-help, the person holding the charge being given broad but clearly defined rights to sell the charge property in whichever way he considers most appropriate. This is qualified by the right of any interested party to apply to the court for protection and to claim damages from the person enforcing the charge for any loss suffered as a result of wrongful or abusive enforcement. Another innovation is that the interests of persons entitled to the proceeds of sale are further protected by distribution being made through a proceeds depository (it is not known whether this proposal has actually been adopted anywhere). In this way, a balance is sought to be maintained between the interests of creditor and debtor.

Even where self-help applies, the creditors’ rights are by no means unfettered. Whilst the creditor may be entitled to choose the time when realisation takes place, he will be under a general duty to take reasonable steps to obtain a proper price. This is enforceable by the debtor in the courts by requiring the creditor to account.


Priority issues may arise in various different ways. For example, they may arise as between successive chargees, in which case the usual rule is that first in time prevails (subject to any perfection requirements having been satisfied). Priority issues also arise on bankruptcy. In “creditorfriendly” jurisdictions, security is taken out of the pot available for the general body of creditors and applied exclusively to the indebtedness of the secured creditor. Finally, there is the possibility of adverse claims. In this instance, the creditor may find the priority of his security prejudiced by parties asserting a superior claim to that of the debtor. Publicity is generally linked to priority.


A guarantee is distinguishable from other forms of security in that it involves no proprietary interest being vested in the creditor. An important practical distinction is between the guarantee and the letter of comfort. Letters of comfort are commonly taken where for some reason the support of, for example, a parent company is desirable. This sometimes happens in connection with capital adequacy concerns, where local supervisors seek assurances that a parent will satisfy the liabilities of a local subsidiary. The legal effect of a letter of comfort depends on its terms, but (as the name implies) comfort is distinct from security, and it should not be relied upon if an obligation binding in law is required.

One other type of guarantee, which is important in certain types of discount markets, is the “aval.” An aval is a guarantee endorsed on a bill of exchange (draft), typically by a bank or other creditworthy institution, which enables the paper to be discounted. Avals are specifically provided for in the Geneva Conventions on Negotiable Instruments (enacted into the laws of countries such as Germany, France, and Japan). English-type jurisdictions typically retain the 1882 Act, which does not specifically recognise the aval, but there is little doubt that the guarantee would be given effect to by the courts in favour of a holder of the instrument.

Enforcement of Claims

Ultimately, the law of security to be effective must be underwritten by an effective regime of enforcement through the courts. As has rightly been pointed out:

Efficient banking … requires a “credit culture”—an environment in which credit contracts are customarily honoured and enforced, in the context of a legal and judicial system that facilitates the enforcement of financial contracts, loan recovery, realisation of collateral, and bankruptcy. In some countries serious weaknesses in the judicial systems can negate improvements in corporate government and official oversight.19

Remedying deficiencies in enforcement procedures in courts is as important as it is difficult. Even in the case of countries with judicial systems generally regarded as efficient, it should be appreciated that the enforcement of collateral can be a laborious process. Protracted legal proceedings may be required to obtain vacant possession of property so that it can be sold, and even in the case of readily realisable security (such as stock), adverse claims, claims by trustees in bankruptcy, and others may impose practical constraints. This merely underlines the point made at the beginning of this paper. Good lending is always conditioned on the creditworthiness of the borrower and the viability of the project in respect of which the funds are being lent. Security is there as a safety net. If it needs to be enforced, the transaction has by definition gone wrong!

Some Recent Issues

Amongst the many issues currently of interest in the field of security law, I shall mention three. Each in its way illustrates the imprecision of boundaries in this field and the problems of applying apparently clear concepts to particular facts.

Reservation of Title

A seller of goods understandably wants to retain some rights in them if he has parted with them to the buyer prior to being paid the price, or the full price. One method of doing this is to stipulate that title (i.e., property) shall not pass to the buyer until the seller has been paid. This type of agreement carries different names, such as reservation of title and conditional sale, but the principle remains the same. In one sense, the seller treats the goods as security for the price, but different legal regimes adopt different approaches as to whether the transaction should be characterised as a security transaction or as a simple sale.

Under UCC Article 9, the seller’s right is indeed categorised as a security interest, though the consequences of this are limited by the fact that filing is not required in the case of consumer goods. In France, it seems that reservation of title is effective and enables the seller to reclaim the goods in the event of the buyer’s bankruptcy.20 A security interest is not created.

The position in England is closer to that of France than of the United States. Reservation of title does not in itself create a charge requiring registration for validity.21 On the other hand, a reservation of title clause that seeks to retain title in the seller until all obligations owed by the buyer are satisfied, or that seeks to give the seller rights in sale proceeds, or rights in items manufactured using other goods, may be characterised as security. The distinction can be hard to draw, but to this writer at least it seems valid in principle. It is the distinction between retaining property until the price of the goods is paid, which is consistent with a simple sale, and under the guise of a sale using the goods, the proceeds of sale or manufactured goods for some ancillary purpose. The former does not amount to security in the conventional sense, whereas the latter may. It is inherent in the analysis that the vendor must accept rights of a limited nature, and may be precluded from seeking to trace his product either into money, or into manufactured items, unless he is prepared to create security and satisfy any registration requirements.

Cash Deposits as Security

From a lender’s perspective, the most attractive security of all is that which can be readily applied to the debt. Cash held at the lender’s disposal is perhaps the best security available. In practice, there are a number of techniques in use for creating this type of security, such as blocked accounts, joint accounts, and escrow accounts. Sometimes these accounts contain one-off deposits, sometimes they are treated as the depository of a stream of payments. In principle, money in an account constitutes a debt owed by the bank to the account holder. That debt constitutes property, and should be available for security in the same way as other debts.

One situation that has caused controversy arises when the bank that holds the deposit (and is therefore debtor as regards the depositor) uses the money in the account as security for a debt owed by the depositor to itself. Should the bank’s rights be categorised as creating a security interest, or some other interest, and if so, what?

From a practical perspective, the issue is not normally specially important because the agreement between bank and depositor can be structured in such a way as to entitle the bank to withhold payment until the secured obligations are repaid, and give a right of set-off (or offset), so as to entitle the bank to satisfy the depositor’s obligations (created in the case of third-party liabilities by a guarantee) from the credit balance. In any event under many systems of law (such as U.S. and English law), banks have general set-off rights based on the principle of netting. However, a charge may be desirable in certain circumstances, for example, where set-off is limited under applicable insolvency laws.

In some countries, the position is straightforward. In Switzerland, for example, a clause is regularly contained in the Swiss bank’s general business terms and conditions providing for a general right of set-off and of pledge over all assets on deposit with the bank. There is no doubt under Swiss law about the possibility of creating a charge or lien (such as a right of pledge) in favour of a bank over a bank deposit.22

In English law, until recently, doubts were raised as to whether it was conceptually possible for a bank to take security over cash deposited with it, on the basis that in some sense the bank became its own debtor.23 However, these doubts have now been laid to rest. It has recently been clearly established that a bank can take a charge over cash deposited with it (and it is unlikely that such a charge is registrable).24

Securities Held by Custodians

Marketable stock makes good security since it is readily realisable at a value that can be objectively ascertained. There are various techniques for taking such security, ranging from a full transfer to the lender, to a nonpossessory charge. In the case of the latter, problems arise where the charge is taken over a fluctuating pool of stock and in English-type systems such charges are likely to be categorised as floating.25

However, with the advent of modern electronic clearing systems, paperless securities, and stock held through custodians, quite difficult legal issues can arise. Take the instance of an investor placing a portfolio of securities with a bank custodian. The securities may include registered and bearer securities, equity and debt securities, securities of domestic and foreign issuers, securities existing only in dematerialised form, and securities placed in a depository such as Euroclear or Cedel. None of the securities may be held in the investor’s own name. His contractual relationship is solely with the custodian. The securities may not be in the name of the custodian either. They may be held in the name of subcustodians, or in the name of the depository through which the securities can exclusively be transferred. The stocks “held” by all parties except for the ultimate holder may consist in practice of account entries in the books of intermediaries.

Now suppose the investor desires to use his stock as collateral. What is the legal nature of the interest being offered to the lender? And for perfection purposes (for example), where is that interest located?

These issues have been much discussed in the United States and England and elsewhere. Article 8 of the UCC has recently been completely revised with modern conditions in mind. It has been said that:

… under revised Article 8, Borrower would be described as an entitlement holder having security entitlements through Custodian. These security entitlements are the collateral in the secured transaction between Borrower and Lender. Once the collateral is described in that fashion, choice of law questions become rather simple. One need only specify the jurisdiction that is most appropriately associated with the package of rights that constitute the collateral. The 1994 revision does so by a series of mechanical rules that, in essence, may be regarded as specifying that the “location” of this collateral is the place where the securities account is maintained.26

Under English law, these issues fail to be resolved applying general principles. There is a consensus that the rights of the depositor against the custodian should be treated as proprietary (depending on the terms of the agreement between them).27 The nature of the proprietary rights will depend in turn on the nature of the custodian’s interest. They can be no more than “derivative” in the sense that investors with the custodian can have no greater rights than those vested in the custodian itself. However, in principle, the investors’ rights against the custodian should be treated as proprietary, should be available as security, and treated as located at the custodian’s office. In practice, however, security documentation taken over stock held by custodians often refers to the stock itself which may not reflect the true legal position.

In respect of securities held with Euroclear and Cedel, pledging arrangements are provided for by the laws of Belgium and Luxembourg respectively. In July 1996, France enacted a new type of security interest, taking into account the fact that much of the property of companies is in the form of a securities portfolio, the main feature of which is to be a changing collection of assets. The law creates a new kind of security device: the pledge of financial instruments accounts.28

Reform and Outlook

No one doubts the importance of the modernisation of the law of security. Very significant steps have been taken in this regard, worldwide. The law of China has already been mentioned. Other fundamental reforms include Russia’s Law on Pledge of 1992, reforms to the law of Hungary (influenced in part by the EBRD model law), and Poland’s Act on Registered Pledge and the Register of Pledges (which came into force on January 1, 1998). In addition to the EBRD, the World Bank has played a leading role in law reform efforts aimed at encouraging the grant of secured credit. International law reform agencies have also played their part, UNCITRAL in the field of receivables financing, and UNIDROIT in the field of factoring. Notwithstanding these efforts, harmonisation on an international scale seems a long way off.

It is noteworthy that this is the case, even in the European Union as it embarks on a single currency. The attractions of a harmonised law on security in terms of easing the provision of credit throughout the EU seem to be self-evident. But differences in property laws, registration techniques, and even on policy issues, make this a particularly difficult proposition. It is suggested that the ultimate goal for Europe in this field should be a Uniform Commercial Code along the lines of the U.S. model, including an article such as Article 9 standardising security interests in movable property. There is now a wealth of international experience to draw on.


I am grateful to Douglas Arner (Research Fellow at the Centre for Commercial Law Studies, University of London) for his research assistance.


E.g., J. Hudson, The Case against Secured Lending, 15 Int’l Rev. L. & Econ. 47 (1995).


See Art. 1 of the Security Law of the People’s Republic of China quoted in L. Barale, “China’s New Security Law: A Closer Look,” chapter 21 of Emerging Financial Markets and Secured Transactions, J. Norton and M. Andenas eds. (London: Kluwer, 1998) [hereinafter Emerging Financial Markets].


T. Benson, Taking Security in China: Approaching US Practices?, 12 BJIBFL at 247, 316, and 354.


Variants of Art. 9 have been adopted in Canada, and are under consideration in Australia and New Zealand.


K. Kreuzer, “The Model Law on Secured Transactions of the EBRD from a German Point of View,” chapter 9 of Emerging Financial Markets, supra note 3.


Reudelhuber and Vogt, The Issuance of Asset-backed Securities by Credit Institutions in Germany, 13 BJIBFL 98 (1998).


John Taylor, foreword to Emerging Financial Markets, supra note 3.


Report of the Working Party in Financial Stability in Emerging Market Economies (Basle: Bank for International Settlements, April 1997) Annex 1.


For a groundbreaking study, see P. Wood, Comparative Law of Security and Guarantees (London: Sweet & Maxwell, 1995).


See Kaufman, Simensky, and Bryer, International Law on Security Interests in Intellectual Property, 3 JIBL 120 (1991).


In Germany, see decisions of the Federal Supreme Court IX-ZR 93/93 24 February 1994, and IX-ZR 141/93 28 October 1993; in England, Barclays Bank plc v. O’Brien [1994] 1 AC 180.


J. Dalhuisen, “The Conditional Sale Is Alive and Well,” chapter 6 of Emerging Financial Markets, supra note 3.


H. Oda, Japanese Law (London: Butterworths, 1992) at 174.


Under the Bills of Sale Acts.


Art. 41, PRC Security Law.


Art. 43, PRC Security Law.


Floating charges can only be created by companies, not individuals.


David Folkerts-Landau and Carl-Johan Lindgren, Towards a Framework for Financial Stability (International Monetary Fund, January 1998) at 14.


Arts. 115 and 121, Bankruptcy Act of 1985.


Aluminium Industrie Vaassen B.V. v. Romalpa Aluminium Limited [1976] 1 WLR 676.


See M. Giovanoli, chapter 9 of European Banking Law, R. Cranston ed. (London: LLP, 1993) at 217.


Re Charge Card Services Limited [1987] 1 Ch 150.


Re Bank of Credit and Commerce International SA (No. 8) [1997] 3 WLR 909.


See, e.g., the decisions of the Singapore courts in Re Lin Securities pte [1988] 2 MLJ 137 and Dresdner Bank v. Ho Mun-Tuke Don [1993] SLR 114. This can affect registration requirements.


J. Rogers, Policy Perspectives on Revised UCC Article 8, 43 UCLA L. Rev. 1431, at 1460 (June 1996).


See generally, R. Goode, “The Nature and Transfer of Rights in Dematerialised and Immobilised Securities,” chapter 7 of The Future for the Global Securities Market, F. Oditah ed. (Oxford: Clarendon Press, 1996).


Gaillard and Baffreau, A New Regime of Financial Instruments Collateralisation in France [1997] 11 JIBL 453.