The epic tale of legal entity identifier implementation, the dazzle of artificial intelligence and, yes, even the gradual acceptance of new ideas about data management and governance, were among the provocative topics for these analyses
Changes are coming both for me and for Inside Reference Data, as our November print edition will be our last before this title and Inside Market Data are merged into a bigger monthly magazine (IMD and IRD will retain their present form for online news and other content). As it happens, I'm departing my role and Incisive Media for another opportunity.
Over the years, the perspective I took in these columns is that there are inevitably going to be many imperfections in financial data management operations, but at the same time there are exciting ways of thinking about how to correct issues or innovate and think differently about data concepts.
I've found, in the past five years editing Inside Reference Data, that covering data concerns as a journalist also reveals one's own imperfections, in understanding and portraying what matters to readers, but hopefully developed my ability to provoke thought about the management issues this part of the financial services industry has. So please indulge me one last time as I look back at the sweep of developments I've covered and tried to provide insights about, for a few parting thoughts.
Following The LEI's Evolution
Implementation of the legal entity identifier (LEI), put into motion after the 2008 financial crisis to avoid future inability to track holdings with counterparties should those firms fail, has had many twists and turns over the past five years. It still has a few years to go to reach the goal set by the body that now administers it, the Global Legal Entity Identifier Foundation (GLEIF).
GLEIF did not exist yet when I began this role, and in my very first online column as editor, "LEI and the Profit Motive," I suggested that a non-profit organization might not be the most effective way to implement the LEI. GLEIF has proven that wrong, but not completely, because it does lean on assigning responsibility to for-profit providers in specific individual markets. So perhaps that column was at least half right.
Along the way, the LEI picked up steam as European regulations such as the Markets in Financial Instruments Directive (MiFID) II, the European Market Infrastructure Regulation (EMIR) and Basel III required firms to better track their transactions and holdings. Those rules presage a complication now apparent in LEI implementation—reconciling differences being allowed by local operating systems, as discussed in a 2013 column, "It's Not Easy Being Extraterritorial." It could also be debated whether the entrenched interests of local for-profit service providers make such reconciliation and standardization more difficult.
Once LEI issuance actually began, the concerns shifted to the quality and accuracy of LEI data, and standardization is being seen as a key element to improve those traits. By early 2015, quality was just starting to appear on the radar of LEI concerns, as noted in "The LEI's Next Frontier." DTCC and Swift had by then formed the Global Markets Entity Identifier (GMEI) utility, which is the registrar of record in many jurisdictions, and their executives have been cognizant of the importance of LEI data quality.
By this year, the issue of LEIs that had already lapsed became more apparent, as noted in "Identification and Verification." GLEIF has now jumped on the problem of keeping LEIs current, launching a verification effort this July, as noted in the column. It could also be debated whether this was soon enough, but with only about a quarter of GLEIF's eventual goal of 2 million LEI registrations complete, the verification and weeding out of lapsed LEIs could have come along much later in the process. GLEIF has demonstrated aptitude and credibility as it heads into the latter phases of instituting LEIs. Although many more still need to be registered, the identifier is well on its way.
Final Provoking Thoughts
Often, my columns have reflected on ideas about data management and governance put forward by knowledgeable industry professionals—how to incentivize the changes they may advocate, what might help gain support for their ideas and whether a method or service can derive value by better managing reference data.
The most exciting takes, however, have been the opportunities to show how artificial intelligence (AI), technology advances, management theories—and even pop culture portrayals of such business and technology topics—can be relevant for improving on how data is managed. Such explorations can be instructive and useful for financial data management executives in the years ahead.
The way the brain processes signals is a lot like how data systems process information. Understanding that similarity can serve as a foundation for innovative ways to perceive information, including financial data, as explored in two columns from this year and 2014 highlighting these insights from popular science author David Eagleman.
To take the relationship between intelligence and data a step further, artificial intelligence has been finding its way into data management discussions for years now. In the movies "Her," "Ex Machina" and "Chappie," mentioned in these columns from 2015 and 2014, unexpected developments occur when an artificial intelligence acts on volumes of data that are exponentially greater than what one single human being can process or fathom.
In those pieces, I argued that the industry would be challenged by reducing the cost of artificial intelligence resources enough to make it viable for smaller firms and users, although there already were providers out there making AI more accessible. Before leaping headfirst into AI however, the data governance and management procedures one has in place need to be equally insightful, but also steady and well organized. As described in a June 2016 column inspired by the TV show "Silicon Valley," the first step before any plans or any innovation should be consulting with the data managers and technology staff who "walk the walk" with data users.
Finally, if any of these insights proves to be as on target five years from now as my first thoughts on legal entity identifiers five years ago—or if any of our readers think back on them as useful food for thought in the years ahead—I would be pleased to know that the approach, at least, was the right one.
Waters Wavelength Podcast Episode 75: An Update on the Julia Programming Language; AI & Alternative Data; Digital Currencies
Julia Computing's Viral Shah talks about the programming language he helped create and what's ahead for it. Then James and Anthony talk about the pairing of AI & alternative data, digital currencies, and Game of Thrones.Subscribe to Weekly Wrap emails