Temporal Aspects of Smart Contracts for Financial Derivatives

Abstract

Implementing smart contracts to automate the performance of high-value over-the-counter (OTC) financial derivatives is a formidable challenge. Due to the regulatory framework and the scale of financial risk if a contract were to go wrong, the performance of these contracts must be enforceable in law and there is an absolute requirement that the smart contract will be faithful to the intentions of the parties as expressed in the original legal documentation. Formal methods provide an attractive route for validation and assurance, and here we present early results from an investigation of the semantics of industry-standard legal documentation for OTC derivatives. We explain the need for a formal representation that combines temporal, deontic and operational aspects, and focus on the requirements for the temporal aspects as derived from the legal text. The relevance of this work extends beyond OTC derivatives and is applicable to understanding the temporal semantics of a wide range of legal documentation.

Smart Derivatives Contracts: the ISDA Master Agreement and the automation of payments and deliveries

“…Although many aspects of contract performance are currently (separately) automated, smart contract code would add the novel aspect of autonomous operation, where the code takes higher-level control over the activities performed, makes decisions based on observed events, and has far less need for human intervention. 10 The internal model envisages that all or some of the autonomous smart contract code (written in some formal representation) would be part of the contract and would have legal effect: 11 Certain clauses would be drafted in natural human language, as is the case today. But other clauses would effectively be set down on the page in some form of code, or other formal representation.…”

“…The final observation in the above quote is important -that the code need not be constrained to snippets of "formal representation" dispersed through the 10 We note in passing that it is not true that only provisions that are "formally specified" can be automated -simple aspects of derivatives contracts are already performed automatically (though perhaps not autonomously), and it is trivially true that simple provisions written in English can be performed automatically by computer. 11 An extreme version of the internal model might be for all of the contract to be represented formally, though many problems would have to be solved first (such as how to embody in a formal representation the degree of ambiguity and flexibility that is currently used in legal text).…”

“…Other aspects of automation occur in terms of streamlining the pre-trade workflow and in the processing of disputes, using a blockchain to provide an immutable history of versions of legal documentation. Through the use of Smart Contract Templates [8,9] the testing and debugging of code can be achieved much earlier in the code lifecycle, 3 and through the use of semantic analysis of the legal documentation [5,7,10] test cases can be constructed and used for code validation.…”

Literature Review: Smart Contract Semantics

“…For example, obligations may have deadlines, prohibitions may apply only for some specified time period, or some permission may be granted only after some obligation is fulfilled. Clack and Vanca [25] provide a thorough report on this subject, identifying many such examples within the ISDA Master Agreement. The model described above cannot express these links between deontic elements of the contract and associated discrete temporal instances, continuous temporal time spans, or conditions.…”

“…Clack and Vanca [25] identify some issues with Lee's framework. One criticism is that the system cannot handle all of the complex or nuanced temporal expressions that sometimes arise in the natural language of legal text.…”