A little-discussed problem has plagued the financial services industry for decades: a high percentage of trades break down — perhaps as many as 30% to 45%.
Each processing glitch means someone in a back office has to analyze the details of the trade, figure out what went wrong and repair it. Often, the problem is a data hiccup — a securities identifier code or a company identifier is not recognized by the receiving firm’s system, for instance, or a buy-side firm is using a Bloomberg product code while the supply-side firm’s using a code from IDC.
“The cost of repair is out of control,” said an executive who spoke off the record and is involved with a utility JPMorgan Chase, Goldman Sachs and Morgan Stanley are building to fix this problem. “For every million dollars a firm spends on data management, there is an average of $3 million in trade repair costs.”
Capital markets firms will spend about $1.2 billion next year on reconciliation technology and services, including external and internal investment, according to Aite Group.
The new bank-run utility, called SPReD (for Securities Product Reference Data), will use a combination of people and technology to build a clean, accurate securities reference data file and use it to fix trade errors. It will start with equities and certain types of derivatives.
“All the banks including us have all these different data streams being collected around the bank,” noted an executive at one of the participating banks. “We’re all receiving this data independently, bringing it into the company, and we have teams of people purchasing the data. For all this reference data, if we’re all using the same data anyway, [the question is] can we get it to a place where it’s scrubbed and clean, a golden copy.”
Most large banks are trying to do this in a more automated fashion, said Virginie O’Shea, senior analyst at Aite Group.
Securities reference data is data about the data in trades — the securities identifiers like CUSIP or ISIN numbers, the financial instrument data (name, interest rate, issue date, currency, etc.), the counterparty identifiers. Making sure all this data is accurate can sound trivial but is actually a huge challenge.
“People call reference data static data and it’s really not,” O’Shea pointed out. “Look at any of the standards they’ve tried to impose in various areas — the industry doesn’t tend to follow those standards.”
There are several standards for securities identifiers, including CUSIP (which stands for Committee on Uniform Security Identification Procedures), ISIN (International Securities Identification Number) and the Bloomberg Global Identifier, but because there are so many and because companies aren’t required to use standards at all, this doesn’t help.
“The industry pays lip service to standards and then doesn’t actually do it,” O’Shea said.
For the full story, please visit AmericanBanker.com