Big U.S. custody banks mining Big Data need more engineers

By Tim McLaughlin

BOSTON, Dec 10 (Reuters) - The world's two largest custody banks, BNY Mellon Corp and State Street Corp, are loading up on engineers to crunch mountains of data into juicy chunks of information that they can use to win more customers and generate more fees.

The shift in hiring strategy comes as the cost of computing power, data storage and bandwidth plunges, giving the U.S. banks more opportunities to capitalize on information about customers that include the world's largest hedge, mutual and pension funds.

"At State Street this year, we've hired more engineers than MBAs," said John Klinck, the bank's head of global strategy and new ventures. "We're a data-intensive business, with $26 trillion of assets flowing through the pipes of the company every day. That's why the engineering side is critical, and it's hard to find."

State Street declined to give any specific hiring figures. But chief rival BNY Mellon said that this year it had hired 30 engineering graduates from college campuses, with plans to bring on 50 more in 2014.

The banks are not looking for people who can design roads and bridges, but graduates who are schooled in the latest techniques of data visualization, for example.

BNY and State Street safeguard $53 trillion in assets and perform mundane but vital tasks such as calculating mutual fund prices and keeping track of dividend and interest payments for investors.

The banks are sitting on so much data that even their top chief information officers concede that they do not know what to do with it all.

Still, custody banks are in prime position to generate more fees from their securities lending data by offering that information to a larger market in a real-time format, said Likhit Wagle, global banking leader for International Business Machines Corp.

BNY Mellon and State Street each play a key role in brokering transactions that allow hedge funds, for example, to borrow stocks from mutual funds and other institutional investors. That activity gives them access to massive amounts of information that may be valuable to a broader audience by giving investors ideas about how to play the market, Wagle said.

However, he said the banks must also figure out how to navigate privacy concerns and stiffer regulations.

REAL-TIME ADVANTAGE

David Lewis, a senior vice president at SunGard Financial Systems' Astec Analytics division, cited last week's credit rating downgrade of Australian airline Qantas Airways Ltd to "junk" status as an example of how investors can benefit from real-time securities lending information.

Short interest in Qantas stock was quietly building five months before the downgrade, and it more than doubled to 53 million shares from 24 million in November, Lewis said. Someone analyzing the data could have taken advantage as big, savvy investors began placing larger bets against a company with a lot of problems.

Qantas shares plunged 13 percent last week after Standard & Poor's issued the credit downgrade.

"You were five months behind if you moved on the downgrade," Lewis said. "If you're not looking at the whole picture, you're kind of investing with one arm behind your back."

But the full potential of data is still unclear.

BNY Mellon sent an email in the summer to about 50,000 of its employees, asking for ideas about how to harness the power of Big Data. It received more than 200 suggestions, said Chief Information Officer Suresh Kumar.

Kumar told Reuters that securities lending transactions were fertile territory for data mining, as some of the employees had proposed.

Custody banks generate a bulk of their securities lending fees from finding the hardest-to-get stocks sought by large short sellers such as hedge funds. Identifying those stocks in a more efficient way could bring more profit to banks, Kumar said.

"So many things take place in a financial institution that we don't capture," Kumar said. "That's the opportunity to say, 'If it's easy and economical to capture and we have the computing and intellectual power to analyze, then what insight can be gained for us and the client?'"