There is a lot of speculation on the extent to which capital markets and risk will need to adjust to “deglobalization”, foreshadowed by the pandemic crisis and punctuated by the Russia-Ukraine conflict, against a backdrop of rising investor expectations to be prepared for any number of unforeseen events. We speak with Tim Lind, managing director of DTCC Data Services, about the role of back office data for front office decisions in this complicated, volatile environment.
This content requires a Finadium subscription. Articles with an unlocked symbol can be accessed with free registration. Log in or create a free account by signing up here..
DTCC collects data as a result of its global position as the central securities depository for the US facilitating equities and fixed income transactions, as well as its trade reporting utilities for derivatives (GTR) and Institutional Trade Processing (ITP) services. That translates to collecting some 80% of the notional value of OTC derivatives under regulatory reporting regimes globally, like Dodd-Frank and EMIR, and between 60% and 70% of buy- and sell-side trade activity in global equities and fixed income.
“We see about half of the world’s cash-settled capital markets activity by value and volume on any given day,” said Lind. “The sheer magnitude of our clearing, settlement (and) trade reporting activities means that we have a natural source of empirical observations of what markets are doing.”
DTCC’s data products are branded “Kinetics”, a term that refers to the science of motion and its causes. In the capital markets context, this is intended to reinforce the point that DTCC’s collected data are real trade observations, or “evidence of the natural forces of supply and demand” across asset classes, he explained.
US equity markets, for example, are comprised of dozens of equity trading venues that add to the complexity of front office trading, but it all comes together at DTCC’s clearing subsidiary National Securities Clearing Corporation (NSCC). Not surprisingly, a bulk of the clients of this NSCC data are quantitative hedge funds that systematically trade equities, using data to tune trading models, which has only grown with the popularity of “nowcasting”.
“The breadth of equity trade observations that we have is really unparalleled, so we can keep all this data together and show you, at the individual security level, buying and selling momentum, how concentrated this activity (is),” said Lind. “Our clients are using it for valuation, to discover trading activity, in alpha generation, how it impacts signals that would (help) understand why prices are moving and make bets based on familiar patterns.”
Alt-data
Lind describes the alternative data space as “vibrant and innovative”. Shortly after the meme stock event in January 2021, alternative data suppliers began mining Reddit and social media feeds for hedge fund clients seeking to gain an edge. In general, hedge funds employ full time staff who scour the globe for alternative data sources, Lind noted.
From 2019 to 2021, the amount of retail participation market share went from single digit percentages to almost 30% of the consolidated volume of equity trading in the US in 2021. This was mostly fueled by “frictionless” trading that platforms like Robinhood have set up via the payment for order flow (PFOF) model.
“We thought we understood equity markets, and now we have a whole new dynamic that occurred in 18 months that we never really encountered before,” said Lind. “(Retail participants) don’t act like institutions, they are not rebalancing on an index, they are not necessarily investing in ESG or ETFs, they are fueled by other motivations (and) the institutional side needs to take that seriously.”
While equity markets are known for transparency and electronification, fixed income still has plenty of room to continue to improve, and Lind said that this asset class is experiencing a “super trend” towards understanding liquidity and valuations.
“Our notion of what fixed income pricing is, is based on models and isn’t always based on empirical evidence of the natural supply and demand of trade activity. Rather than models in people’s imagination, we are mining empirical evidence of market dynamics,” he said.
A new era
Finadium has worked with DTCC on its Treasury Kinetics product, which creates a broad source of benchmark information for collateral takers/providers and intermediaries derived from billions in daily repo settlements. The data is painting a picture of a major shift in short-term cash.
“Over the last 10 years, there’s been no returns on cash and people have been a little bit laissez-faire in terms of seeking the best return, especially with short-term cash. Now that benchmark rates are increasing by central banks around the world, we think we are going to usher in a new era of more active management of cash, and there’s real returns and yields to be had,” said Lind, adding that also there are penalties that could be paid by those that don’t manage their cash wisely.
Securities lending too is evolving as DTCC eyes its SFT clearing launch at a time when trading platforms aim to bring better visibility on activity, like hard-to-borrow securities pricing for example, while clearing on the backend creates a new source of information on collateral pricing.
In another upcoming product launch, the team will be beta testing ETFs to better understand pricing anomalies arising from basket creation/redemption processes. And Lind also noted that OTC derivatives markets data still has a lot of room for growth.
“The last 10 years, our industry has spent tens of billions of dollars to create reporting mechanisms for OTC derivatives, but the focus has been on reporting not analyzing, so I think we are going to enter a new era where the data is more appropriately mined and used to understand the dynamics of the derivatives market, but more importantly the underpinning risk factors of those derivatives — interest rates, equity values, credit, commodity and energy prices,” Lind said.
“Those derivatives collectively tell stories about the market and sentiment, and that data has not been appropriately utilized up to its potential,” he added.