Global Markets Need a Common Language

There is a quiet revolution brewing in the world of security identifiers. Over the past 18 months, many proposals have surfaced to create new identification systems for the millions of financial instruments traded worldwide each day. They have ranged from open source identifiers made available over the internet to randomized codes issued by government-sponsored entities, to proprietary, vendor-specific codes created by trading platform providers. While the intentions behind many of these proposals are to improve the infrastructure of the capital markets, we think others are merely veiled attempts at co-opting users into reliance on their proprietary systems.

Like most industry participants, the American Bankers Association (ABA) has watched with interest the unfolding debate over symbology in the financial markets. We have read all the press releases and white papers, and heard myriad panelists opine on the topic: Should symbology be proprietary or standardized? Should symbols be “open” or “closed?” Free or subject to licensing?

The ABA has some very strong views on this topic. As owners of the Cusip system, operated on our behalf by Cusip Global Services (CGS), we have been creating unique identifiers for financial instruments and their issuers since 1968. The Cusip system was developed at the dawn of the computing age to bring order to a chaotic settlement process, in which every securities processing shop on Wall Street had developed its own proprietary coding system for its punch card computers. In the years since, we’ve seen many symbology concepts come and go. After sifting through this latest round of proposals for new identification systems, we have found several concepts and ideas we think are detrimental to efficient markets.

Some proposals have also called for vendor-developed “open symbology,” a process through which identification codes are provided to all market participants, free of charge, by data vendors. Though the codes are provided for free, they do not appear to be open in the sense they can be used across multiple vendor platforms as a truly interoperable identification standard.

Since its inception, Cusip identifiers have been made available to any and all market participants. Fees for the assignment of a Cusip identifier are nominal.

A single license allows a firm to access Cusip identifiers and the associated data from any number of secondary vendors and distribution channels, or to get it directly from the source at CGS. As a result, Cusip and Cusip-based ISIN numbers are interoperable, allowing efficient communication between financial institutions around the globe regardless of individual platforms or data vendors of choice. Simply calling something “free” does not render it “open” in the sense of universal interoperability.

Why not issue random codes?
Other recent proposals have suggested the computerized assignment of random identification codes to all financial instruments. While this would require much less manpower to create each individual code, the elimination of specific focus and management of instrument identifiers may significantly increase the risk of failed trades and inaccurate reporting. CGS maintains a full-time staff of 30 data analysts to distill hundreds of pages of issuer offering documents down to nine digits that will become every publically traded security’s face to the world.

CGS also follows strict numbering constructs to build consistency and predictability into the “smart” codes. For example, the Cusip issuer number for Motorola is 620076 and any 9-character Cusip number starting with those digits must be a Motorola security. By contrast, another identification scheme may use a random alpha-numeric code for Motorola common stock and a completely different random code for a 5.375% bond from the same issuer.

Are there rules for constructing or maintaining the codes? How might symbols of varying lengths impact internal systems? Will a trading counterparty recognize the code? These are important questions when considering interoperability. If a code’s universal acceptance is in any way questionable, then the underlying symbology is of limited utility.

The ABA agrees that making standardized codes widely available is beneficial to the industry as a whole, which is why Cusip has done so in a fair and equitable manner for more than 40 years. We disagree, however, with the implicit notion that the entire undertaking can be reduced to generating and publishing random strings of characters.

It is possible to create identifiers at a reduced cost by removing proprietary intelligence from the process and simply issuing random codes, but that is hardly a step forward in the quest for transparency in financial markets. The only way to ensure the market-place continues to receive identifiers that speed-up trade processing and settlement, while giving market participants the confidence their data is accurate, is through focused reinvestment into the system.

Without this entrepreneurial zeal to continue to grow the business and improve the Cusip offering, it would be impossible to expand into new markets with the level of accuracy and due diligence the market-place requires.