Something curious is happening in the financial markets. The finer points of bond and loan documentation are usually the preserve of those bankers responsible for execution, in-house lawyers and the law firms that advise them. Rarely does the debate attract wider interest. But in the wake of recent sovereign bond defaults, there have been calls for changes in the underlying documentation to facilitate the restructuring process.
The reason for this is not hard to fathom. In the 1980s, when the modern approach to sovereign debt restructuring became established, most of the debt affected consisted of commercial bank loans. The structure of the syndicated loan with its agent bank and clearly identifiable syndicate members provided a framework by which creditors could, through a bank steering committee, establish a dialogue with the borrower.
In the 1990s, however the source of emerging-market funding switched from the loan market to bond markets. The international bond markets are - despite electronic registers held by the likes of Euroclear and Cedel - still characterized as bearer markets. Bond holders are not readily identifiable nor do they have or seek the same degree of intimacy with the issuer that a commercial bank has with its borrower.
This is reflected in the documentation. By contrast with syndicated loans, there is no ready mechanism to establish a dialogue between a sovereign issuer and its bond holders. Even if a transaction benefits from the inclusion of a trustee, the trustee is unlikely to feel comfortable acting as go-between although it may facilitate the establishment of a representative group of bond holders. Hence the calls for changes in the underlying documentation.
Three types of clause have been discussed. Majority action clauses would allow payment terms to be modified provided a super-majority of the bond holders agreed (this term already appears in English law bond documentation but US practice is to require unanimity). Sharing clauses would require any bond-holder litigant to hand over to other bond holders any amounts recovered on a ratable basis (ie, in proportion to their respective holdings). Collective representation clauses would provide the mechanism to allow the trustee or other designated party (namely, a group of bond holders acting as a bond-holder council) to conduct a dialogue with the issuer on behalf of all of them.
"A central premise of 1980s-style debt rescheduling was the need to achieve equal treatment of creditors," explains Lee C Buchheit, a veteran of the Latin American sovereign debt restructurings of the 1980s and a partner at Cleary Gottlieb Steen & Hamilton. "Sharing clauses were an innovation of the syndicated loan market in the 1970s. In most syndicated loans, some of the participating banks maintain continuing business relations with the borrower, and some do not. The latter group may worry that the borrower will be tempted to pay its regular bankers ahead of relative strangers. Sharing clauses provide that if any of them receives a payment that is disproportionate to what the other lenders in the syndicate have received, they will share that payment on a ratable basis with their fellow syndicate members."
Sharing clauses were carried over into restructuring agreements since these acted as little more than massive syndicated loan agreements that covered all of the commercial bank loans to a single sovereign debtor. Sharing clauses put off maverick creditors - those who buy the debt at deep discounts and then threaten a lawsuit in order to force their debt to be bought out. "Maverick creditors did not become a significant irritation again until countries began to exchange their restructured loans for Brady bonds in the early 1990s. The Brady technique had the effect of emptying the old restructuring agreements of all well-disposed creditors, leaving in some cases one or two litigious mavericks behind," says Buchheit.
Majority action clauses have a similar effect: they prevent the maverick from holding the majority to ransom. Originally, syndicated loan agreements provided that amendments to payment terms required unanimity: smaller creditors - say, regional banks - might otherwise feel that the larger money-centre banks with their greater assets and exposures might grant additional concessions to avoid a potential default.
This thinking was carried over into the first debt restructurings. "But by the mid-1980s," notes Buchheit, "there were fissures: big banks versus little banks; regional versus money-centre banks; banks in North America versus banks in Europe, Japan and the Gulf; banks with large loan-loss provisions versus those with none; and by then debt was trading into the hands of non-banks, which lay beyond the persuasive powers of bank advisory committees." The Brady initiative was deliberately structured as an exchange of new bonds for existing restructured debt, rather than as an amendment to the old restructuring agreements, in part to circumvent the unanimous amendment clauses.
Eurobonds - traditionally bearer and often written under English law - have tended to have a 75% majority provision for payment term changes. "The practical choices are either to specify a smaller percentage of total holders or a higher percentage of holders present and voting at a meeting," says Buchheit, "but market resistance to the inclusion of these clauses is likely to increase as the required percentage declines. It is not unusual to find a handful of large buyers owning more than half an issue, and small investors may worry that such clauses may encourage the large holders to ignore their view."
Finally, collective representation clauses make it easier to kick-start the process of negotiation by defining the channel of communication in advance, for instance through the appointment of bond-holder councils. "This is what happened in the 1930s following widespread defaults on sovereign bonds after the market crash of 1929," says Buchheit.
The principal objection to any of these clauses will be the perceived restriction on individual bond holders' rights to take action. Another stumbling block may be the unwillingness of underwriters to mention these documentation changes up front during the roadshow. "They have all the charm of a pre-nuptial agreement," points out Buchheit.
Even assuming general market acceptance of these modifications, it will take time for current issues to be retired and replaced with new-style documentation. So expect any sovereign bond restructurings in the near future to be bedevilled by precisely the problems that these clauses are designed to prevent.