By Robert Iati, Senior Director, Capital Markets, Dun & Bradstreet
We’ve been hearing about the importance of Big Data for what seems like decades, but investment banks often approach data projects with insufficient attention to reference data. Typically, this is due to a prominent focus on ultra-fast market data and a limited amount of discretionary funds resulting from the need to meet overwhelming demands for compliance with capital markets regulations such as Dodd-Frank and the Volcker Rule. Here are three reasons why that is a strategic mistake.
Data volumes continue to grow exponentially
After sifting through nearly unmanageable volumes of data, investment banks and asset managers face the challenge of discovering how to use all the data and, more importantly, how to find meaningful patterns in the data that deliver insights that others don’t have. With the focus of data on gaining greater insight into pricing (market data), many investment banks and asset managers retain the feeling that reference data is less important to their ability to generate trading profits. This is because customer data, counterparty data and indicative product data is perceived to be less dynamic. This, however, is a mistaken view.
In fact, reference data is far more complex today than it has been at any time in our history. Every entity has so many relationships – via acquisitions, affiliations, partnerships and similar dealings – that the spider web of connections for any single entity is nearly impossible to trace. Multiply the relationships by the number of distinct entities in the world and you quickly highlight the magnitude of the problem. The simple result is that we don’t know with whom we are dealing when transacting trillions of dollars of securities.
This influences the counterparties with which investment banks trade, the companies issuing the bonds they invest in, and the entire ecosystem of institutions that comprise the capital markets industry.
Competitive insight can yield giant returns
In the trading business, possibly the most competitive of all, it’s all about a firm’s ability to differentiate itself using data. We all know that the trading business has evolved (some say devolved, but I’ll refrain from that debate for now) into one where technology trumps human intelligence and quantitative process overrules fundamentals. So, in some ways, technology levels the playing field among the large firms, each of which invests hundreds of millions of dollars in trading technology every year simply to keep pace.
The well-worn fact is that alpha has become so hard to find that trading institutions go to the greatest efforts to identify an edge. However, technology too is often commoditised, making a genuine edge more difficult to gain. So, if trading firms have similar technology and use the same data as their competitors, the tide will rise but all firms will rise with it at the same rate. In trading – a zero sum game – that isn’t satisfactory. It is relative growth that matters, leading one firm to win more frequently than its competitors. The winners are those that have the best inputs into their technology. Pairing those inputs with the most refined algorithms and trading models effectively brings out the best in the data. Finding unique datasets, either raw or derived, is the winning formula.
Access to unique data enables well-informed investing
What am I talking about here? Reference data that offers insightful information about companies that form the core of our investments – equities or fixed income securities – is moving to the front of the line for investment models. It can often provide a way to identify a leading indicator for an investment position.
- It’s insight into the payment history for a private issuer of bonds that enables a trader to assess its price better than the other bidders.
- It’s the ability to create new models by using analytics that combine existing (and sometimes common) data with new data in different ways to uncover leading indicators to a trend in a sector or to identify a trigger for a named stock that enables the trader to benefit uniquely.
- It’s about getting an early view into the glitches in the supply chain of a manufacturing company that reduces its future earnings projections and triggers a ‘sell’ signal on its stock before the rest of the market can figure it out.
The point here is that we know the data used by most trading firms is quite similar. However, in spite of all the other machinations adopted by trading firms to make the most of their assets, all traders remain on the lookout for data that they can deploy uniquely to separate from the pack. That offers clear and measurable value.
Right now, the differentiating factor for many of these firms is in the models used by each to derive new data. That is the secret sauce they count on to achieve alpha returns. But improving the inputs by adding genuine, unique data to the equation can raise their game to its highest point.