The year identifiers wanted some attention too

It’s been a year. If, on your bingo card, you somehow had Sam Bankman-Fried getting charged with money laundering, Russia invading Ukraine, Kanye West becoming a Nazi sympathizer, Elon Musk buying Twitter, and Taylor Swift (my girl) instigating a congressional inquiry into Ticketmaster, then congrats: You’re insane. But I bet what you didn’t see coming was all the weird—and arguably wonderful—drama unfolding in the world of reference data, particularly those pesky identifiers.

One might think the back office is a sleepy place. Keep your records straight, reconcile your transactions, pay someone some fees. Open a filing cabinet, take lunch, close a filing cabinet, go home. If you don’t look too closely, it’s tempting—even easy—to write off this little corner of the financial world, much like Taylor Swift’s detractors and haters. But what a mistake that would be! See: Reputation (2017).

In the midst of the 2021 winter holiday break, FactSet quietly issued a press release: The vendor would acquire Cusip Global Services (CGS) from S&P Global, its operator for more than 50 years, for $1.925 billion.

For the next three months, all was calm and quiet. Then, shortly after FactSet announced that its deal with S&P had closed, I got an email from someone I’d never heard of or spoken to, and it alerted me as simply and succinctly as FactSet’s December 27 headline: “Lawsuit imminent,” the email read, with only a link to our story about the announcement, which ran in early January. Those are my favorite kinds of emails (the cryptic kind), and the people who send them are my favorite people (the cryptic kind).

And so, a fire was lit—one that’s taken center stage certainly for reference data experts and enthusiasts, and potentially for anyone who is signing their name to a contract with a data licensing clause.

In the Southern District of New York—the powerhouse behind financial crimes prosecution—two firms filed a joint class action lawsuit against CGS, S&P Global, FactSet, and the American Bankers Association, the patent owner of the Cusip standard. Dinosaur Financial Group, a New York-based broker-dealer, and Swiss Life Investment Management, a European real estate-focused asset manager, had assembled a high-powered legal team to sue the quartet of companies, alleging violations of sections 1 and 2 of the Sherman Antitrust Act, enacted in 1890 to protect against monopolistic behavior; Section 4 of the Clayton Act, an anti-trust law enacted in 1914; and the US Copyright Act of 1976.

Like in Taylor Swift’s 16-year career, the hits kept coming.

Days after the first class-action was filed, Hildene Capital Management, a Connecticut-based asset manager, filed another one, also alleging violations of the Sherman Antitrust Act as well as various violations of the New York and Connecticut business practice laws.

As this was going on, I was developing a roster of reference data experts for insights. I asked the same questions over and over. What did they think? Did these companies stand a chance if the suits made it to trial? Would more financial firms join the class? How long has a suit like this been coming? While there was a general tone of cautious optimism (as Cusip fees are notoriously despised, but accepted, by trading firms), no one I spoke to was willing to make definitive bets on the outcome, or throw their hats in the class-action ring. The questions were endless.

There was some hemming and hawing about the lack of big names (Goldman Sachs, JP Morgan, and Pimco, for example, were noticeably absent) as plaintiffs. But several people directed my attention to the team of attorneys responsible for the first suit, as if to suggest there were still big guns with powerful ammunition on which to keep an eye.

First, there was Irell & Manella LLP, and the filing specifically named David Nimmer, a giant in the field of copyright law who has twice served as co-counsel representing clients before the US Supreme Court—his side won both times in unanimous decisions. Leiv Blad of Competition Law Partners PLLC is an expert in the field of antitrust law with a focus on financial services. And Robert Kaplan and Frederic Fox of Kaplan Fox & Kilsheimer LLP have been lead or co-counsels on several major class-action cases involving financial institutions.

But what did two competing class actions alleging similar wrongdoings and grievances mean for each other? We wondered that, too, for a while. In the end, the presiding judge over both cases, Judge Katherine Failla, ordered the cases and their respective legal teams to be consolidated.

In August, the three firms joined forces to file a joint suit. The trio sought a ruling on whether Cusip numbers—de facto serial numbers used to identify nearly all North American stocks and bonds—are copyrightable or not, challenging the ABA’s assertion of copyright over the Cusip numbering system, which it has held for more than 50 years.

The new suit also alleged CGS, S&P, FactSet, and the ABA violated sections 1 and 2 of the Sherman Antitrust Act by boycotting Cusip users who refused to enter into licensing agreements with S&P and FactSet, and by using the standard-setting process—Cusip is the national standard for securities identification under the American National Standards Institute’s Accredited Standards Committee X9—to squash would-be competitors to Cusip, such as the Financial Instrument Global Identifier (Figi), developed by Bloomberg in 2009.

However, the new suit set up a fallback, stating that if Cusip numbers are found to be copyrightable, then the foursome is at least in violation of the ABA’s commitment to X9 that it would offer licenses to Cusip numbers on fair, reasonable and non-discriminatory (FRAND) terms, including price, which the suit says was and is a condition of X9’s designation of Cusip as the national standard.

The case then saw a flurry of docket entries and conferences in November and December. Notably, lawyers for the defendants admitted that individual Cusip numbers aren’t subject to copyright under the current patent. What is copyrighted is CGS’s master database—similar to how a phone book can be copyrighted, but not the phone numbers it contains.

As Tim Baker, founder of fintech consultancy Blue Sky Thinking, pointed out in his handy trial synopsis, it “sets the stage for many industry participants to challenge the validity, necessity and even legality of the widely deployed CGS license, at least the part that requires CGS customers to restrict further access to content that includes the Cusip unless those customers also have a CGS license.”

And on December 21, the latest update arrived in the form of another amended joint complaint from the three customer firms on behalf of their class (representing all entities who pay license fees to use Cusip numbers in their businesses, including institutions, data vendors, fintech companies, and other potential competitors who could develop innovative products and services using the Cusip numbers).

The plaintiffs doubled down on their copyright challenges and included further allegations of wrongdoing. In one explosive example, lawyers for the plaintiffs claimed that the ABA explicitly told the Securities and Exchange Commission that individual Cusip numbers are copyrighted for the purpose of influencing the SEC’s rulemaking process. Big, if true.

We’ll keep you updated on what’s coming next.

Figi, LEIs, BTC, oh my!

But Cusip, for all its drama, is merely one standard identifier among many—several of which, frankly, also have their own drama. Let’s catch up with the aforementioned Figi, first.

A quasi-competitor to the Cusip standard, the Figi is a recognized US data standard authorized for financial instrument identification by the American National Standards Institute’s subcommittee X9, which is responsible for developing standards for the financial services industry (but it is not the national standard). Originally developed and introduced by Bloomberg in 2009, the free-to-use and open-source Figi was adopted by standards consortium the Object Management Group in 2014, which has put forth 11 standards accredited by the International Standards Organization (ISO), and it remains under the group’s governance.

While Figi is meant to be able to uniquely identify any security—even cryptocurrencies—its viability as a competitor to Cusip, in practice, is hindered by the fact that the Depository Trust & Clearing Corp. (DTCC), which processes most securities transactions in the US, allows only Cusip codes to be used in settlement.

Nevertheless, there were more than 92 billion requests from the Figi API in 2022 on OpenFIGI.com, up 20% over 2021, more than 1,400 new API accounts, and 1.4 million visits to the site. And in 2022, the identifier passed the milestone for having more than 1 billion unique Figi numbers identifying unique securities out in the world, 80 million of which were generated in this year alone, according to Steve Meizanis, Bloomberg’s global head of symbology and LEI services.

Speaking of the LEI—short for Legal Entity Identifier, and which identifies legal entities rather than securities—there’s been some progress there, too. With more than 2.2 million LEIs now in existence, and with 39 participating issuers, the Global Legal Entity Identifier Foundation (Gleif) is in the process of expanding the identifier’s scope beyond financial reporting to include know-your-customer processes, client onboarding, and supply chain relationships and recordkeeping.

And crypto continues to do what it does best: throw a wrench into everything.

We closed out the year watching Bankman-Fried’s FTX crash, burn, and implode—leading to a chaotic court hearing in the Bahamas and the extradition of the disgraced mogul back to the US. Ever the opportunists at WatersTechnology, my colleague Joanna Wright and I seized on the chance to capitalize on disaster. While everyone was pointing and laughing at Bankman-Fried, we asked those in the institutional crypto space (a crawl space, for all intents and purposes): “Hey, what about your reference data needs?!”

Well, they have some needs, and to no one’s surprise, little has yet been done to address them. But at least one organization, the Association of National Numbering Agencies (Anna), is trying. The organization is working on making the Isin, the selected identifier for transaction reporting under Mifid II, fit for identifying digital assets, much as it did for OTC derivatives. In 2019, the association said it was forming a technology task force on digital assets to examine how this might be achieved.

“All the providers use different symbology because there are no standards in crypto,” Ethan Feldman, co-founder and CTO of Talos Trading, told WatersTechnology in November. “So even if you’re trading something as simple as Bitcoin, you’re going to see different symbols across the providers. Some are using XBT, for instance, which is the currency code; others are using BTC, which is more canonical. This is a challenge for us every day.”

Cryptocurrencies like Bitcoin and Ethereum can be “forked” into new currencies when their developer community makes a change to the blockchain’s ruleset, or protocol. Bitcoin, for instance, was forked into Bitcoin Cash in 2017, meaning that Bitcoin kept running, but a new currency called Bitcoin Cash began to also operate. Bitcoin Cash was referred to by the BCH symbol. Then in 2018, BCH itself split in two, forking off into BCH and a new currency, Bitcoin Satoshi’s Vision (SV). In 2020, BCH split yet again into BCH ABC (BCHA) and Bitcoin Cash Node (BCN).

The tracking of all these symbols is an ongoing exercise in double- and triple-checking. To get around it, Talos creates its own standardization by using its own symbology, agnostic of what the crypto exchanges are naming the instruments. “If you’re using our API, you’re going to use BTC to interact with Bitcoin, whether it’s on Kraken or Coinbase. That’s a very vanilla case, though; it gets way more complicated as you get into more complicated assets,” Feldman said.

Three years on, it seems like Anna’s task is still very much a work in progress—a testament to the degree of difficulty involved in making and setting a standard, even for something fundamental, necessary, and seemingly low-stakes.

And that’s really a good way to summarize the entire world of reference data. It’s niche, perhaps it’s mundane, and one can reasonably draw the assumption that it’s boring—but it’s not. Sixteen years of defending Taylor Swift from her haters on the Internet have prepared me to make this case. It’s one of the most technical and esoteric topics I’ve covered in my four years at WatersTechnology—Oh my God, the acronyms!—but it’s also been the most fun. Everyone has an opinion, and while they won’t always put them on the record, I never get bored of them.