Algorithmic Contract Types Unified Standards

Algorithmic Contract Types Unified Standards, abbreviated to ACTUS, are a set of royalty-free, open standards for representing financial contracts. The standards combine (1) a concise data dictionary that defines the contractual terms present in financial contracts with (2) a simple but complete taxonomy of fundamental algorithmic contract type patterns that incorporate elements from that data dictionary that apply to a given contract type such that (3) the cash flow obligations that are established by the contract can be accurately projected, analyzed and acknowledged by all parties to the contract over the life of the contract.

Providing an open and royalty-free standard for the data elements and algorithms of financial contracts enables the consistent sharing of accurate financial data by organizations in the financial industry, whether to consolidate views of product lines within an enterprise, to manage obligations between institutions, or to facilitate the collection and consolidation of financial data by regulators. Adoption and uptake of ACTUS is viewed as a public good benefit to create a globally accepted set of definitions capable of representing the preponderance of financial contracts in the real economy. Such standards are regarded as important for transparency and efficiency in financial innovation, risk management, financial regulation, the tokenization of financial instruments, and the development of smart contracts for decentralized finance (DeFi) using blockchain.

History

The difficulty of defining and analyzing financial data were described by Willi Brammertz and his co-authors in a 2009 book, Unified Financial Analysis: The missing links of finance. The simplicity of the problem is described in an ECB paper, “Modelling metadata in central banks”. This cites the issue of how financial institutions have tried to overcome data silos by building enterprise-wide data warehouses. However, while these data warehouses physically integrate different sources of data, they do not conceptually unify them. For example, a single concept like notional value still might be captured in various ways in fields that might be labeled ‘nominal value,’ ‘current principal,’ ‘par value’ or ‘balance’. Standardization of data would improve internal bank operations, and offer the possibility of large-scale financial risk analytics by leveraging Big Data technology. Key to this is the idea of "contract types".

The concepts were expanded upon by Brammertz and Allan I. Mendelowitz in a 2018 paper in the Journal of Risk Finance. They describe the need for software that turns natural language contracts into algorithms – smart contracts – that automate financial processes. Financial contracts define exchanges of payments or cash-flows which follow certain patterns. They identify less than three dozen relevant patterns covering most existing contracts. Underlying these contracts there must be a data dictionary that standardizes contract terms. In addition, the smart contracts need access to information representing the state of the world and which affects contractual obligations. This information would include variables such as market risk and counterparty risk held in online databases that are outside the blockchain (sometimes called "oracles").

The idea of the standardized algorithmic representation of financial contracts predates digital currencies but is highly relevant for block-chains or distributed ledgers and the concept of smart contracts. Brammertz and Mendelowitz argue in their paper presented at the 2019 Cardano Summit in Miami that without such a standard, the chaos - already a reality in banks today - would potentiate on blockchains, since on block-chains with Turing complete languages, every contract could be written individually. They further argue that of the four conditions set by Szabo in his article published in 1996, block chains with a Turing complete programming languages fulfill only one, namely Observability. What is not met is Verifiability due to its unstandardized and difficult to read nature especially in the financial sector. The third condition Enforceability is therefore also not met since it is maximized by the combination of Observability and Verifiability. The ACTUS standard fulfills the Verifiability condition for financial contracts wherefore Enforceability is maximized. The fourth condition is Privity, which needs to be solved separately.

The authors argue that the adoption of a standard for smart contracts and financial data would reduce the cost of operations for financial firms, provide a computational infrastructure for regulators, reduce regulatory reporting costs, and improve market transparency. Also, it would enable the assessment of systemic risk by directly quantifying the interconnectedness of firms.

This led to the ACTUS proposal for a data standard alongside an algorithmic standard. Together, these can describe most financial instruments through 31 contract types or modular templates. The research foundation and users' association develop the structure to implement the ideas. The specifications are developed, maintained, and released on GitHub.

An ACTUS Financial Research Foundation and the ACTUS Users Association control the intellectual property and development approaches.

In October 2021, the ACTUS data standard was added as the second reference after ISO 20022 to a database run by the Office of Financial Research, an arm of the US Treasury. ACTUS is being used to help define five asset classes (equities, debt, options, warrants, and futures) in the OFR's financial instrument reference database (FIRD). A third reference, the Financial Information eXchange (FIX) messaging standard, was added a year later. Early 2023 ACTUS became a liaison A member of ISO TC68 / SC9.