Could data management — from managing securities-master files to cleansing reference data — be the next phase in outsourcing? Many industry executives say yes, although they believe it may take some time for vendors to gain the expertise to handle the specific needs of market-data departments on Wall Street.
"We've typically seen processes (outsourced) like human resources or payroll, which are horizontal and can be built on economies of scale," explains Tim Lind, a senior analyst with TowerGroup. "When we're talking about data management, companies must explore whether they can get the same kind of service in an outsource market. How do you get a deep understanding of capital markets in an outsourcing environment?"
Deciding which areas of data management fit best with an outsourcing model is a subject of debate among industry experts. However, the industry agrees that reference data, or descriptions of securities such as their terms and conditions, can potentially be outsourced. Firms are already doing so in the form of a securities-master database, which is comprised of all the descriptive fields of reference data.
Pricing data, both real-time and historical, is another area that is being considered for outsourcing.
A third area that remains open to debate is research data, such as counter-party information. Research data might include whether a company owns, or is owned by, other companies. This ensures that a firm isn't exposed to credit risk, and helps to prevent Enron-type debacles. "A hot area right now is creating links between issuing companies and the public-company securities that they guarantee," says TowerGroup's Lind. "This can give a firm a clear indication of where their exposure lies."
Areas that will prove to be more difficult to outsource include internal data - such as accounting and performance information, or proprietary special-purpose vehicles - according to industry sources. In addition, holdings data - specific securities owned by clients, dates of purchase and price paid - is nearly impossible to outsource, due to stringent client-privacy issues. Equally arduous is corporate-actions data, due to the variety of sources, says Lind.
Stephen Gouthro, senior vice president of Investment Data Management & Investment Applications at Putnam Investments, explains that data quality is a vital issue when examining outsourcing. "No matter how much you pay for this data, none of these providers are actually liable for the quality of the information. However, firms are liable, so we can't just hope that it's correct," he says. He adds that vendors' reluctance to cleanse multiple data sources has contributed to keeping firms from outsourcing this process.
Gouthro says that tolerance for error might determine what processes a firm is willing to outsource. "A trading area is hypersensitive to error versus the research area who might have a much higher tolerance for error," he explains.
As one might expect, the vendors that are poised for the best position in the data-management-outsourcing space are traditional data-management vendors. Market-data executives believe their providers can do more than just give them data, such as consolidate and cleanse their own, as well as other, vendor feeds.
"You can make a lot of headway with a vendor as it relates to the data that you buy from them, but when you're bringing in data from multiple vendors, it gets a little more complex," says Gouthro. "Are you going to ask Vendor A to help you manage the quality of Vendor B's data? Probably not. In fact, they'll be much more interested in talking to you about buying the data exclusively from them."
Many vendors provide software that can be used to consolidate and cleanse data, but don't actually offer the service of doing so. In addition, exceptions must be manually investigated, which software often cannot handle.
For these reasons, a financial-institution's market-data department employs data analysts to collect, integrate, cleanse, validate, monitor, and distribute data and exceptions into the appropriate data warehouses within the firm.
Tom Jordan, president and chief executive officer consulting firm Jordan & Jordan, says that, because of the importance of data integrity, data cleansing could be considered a core competency that is better kept in-house. "Data-content scrubbing is a very specialized business," he explains. "Everyone I've ever met that does data scrubbing believes that they do it better than everyone else."
Others, such as the American International Group, who use Cicada Corp. for enterprise-wide market- and credit-risk data management, disagree. Like many market-data vendors, Cicada provides software to consolidate and cleanse data. However, the vendor recently began offering a consolidation and cleansing service, by tapping into data analysts in Oklahoma City.
According to Hubert Holmes, executive vice president of Cicada, AIG supplies Cicada with external vendor feeds and proprietary AIG information, such as client data. The vendor then integrates, validates, analyzes and cleans the data looking for exceptions. Exceptions are investigated by live analysts and, once reconciled, the data is fed into AIG's database in a format that the firm can use efficiently, says Holmes.
He claims that the result is less costly than it would have been to do the process in-house because of savings on software purchasing and maintenance, hardware, data-center facilities, and data analysts. However, Holmes declined to discuss the specific savings that AIG realized. He says that Cicada's model has the ability to address outsourcing a firm's securities-master database, pricing data and research data.
The operation functions similarly to an offshore-outsourcing model, explains Holmes, as a result of the decreased cost for real estate and labor in the Midwest. Yet, the risk is minimal in comparison to a true offshore model, he says, because the analysts are familiar with the U.S. securities industry, an area where an offshore worker might not have competency.
In addition, there is a question as to whether an offshore outsourcer, with operational abilities in many areas, can have the expertise to handle the data-management processes.
Lind notes there are methods of providing offshore data analysts with an equal level of data aptitude. For example, he says, a firm can send its own analysts offshore to work with an outsourcing vendor to develop its business and train staff.
Girish Paranjpe, president of finance solutions at Wipro Technologies in India, says that his company is exploring business-process outsourcing in the data-management arena. While Wipro has yet to announce a client, Paranjpe notes that there is often high turnover within firms' data-management departments necessitating constant training. "The same effort that our potential customers spend to induct new staff could be deployed to train people elsewhere," he says.
Despite the perceived challenges, Cicada's Holmes says that internal-data management qualifies as one area where a vendor-analyst model of outsourcing might be employed. For example, he says, Cicada and Accenture could combine their technological and strategic expertise to facilitate content-checking for internal data. In this model, a firm could move its own in-house analysts to work at an outsourcing location run by Accenture.
One Boston-based investment manager, who declined to be identified, says he is exploring the idea of outsourcing data management. He's familiar with the Cicada/Accenture model, but notes that this model focuses on the securities-master database, which his firm has already established in-house. In addition, he says, the services are expensive. "They are probably the furthest ahead in this challenge, but they don't use an offshore model, so it's not necessarily cheaper."
TIME IS TICKING
Despite the quiet buzzes of interest, the industry has yet to see much business-process offshore outsourcing in the market-data sphere. Whispers of market-data executives can be heard exploring outsourcing arrangements, but rarely publicized. One reason for this might be fear of negative backlash from their own departments, who worry about losing their jobs to offshore workers, according to industry sources.
However, Jo McCann, a former manager of middle-office data at AIG, notes that outsourcing does not necessarily equal layoffs. Instead, she says, outsourcing can enhance the productivity of internal analysts by freeing up their time, or bring a new process to a firm that had not previously been a focus.
In addition to layoff fears, many firms are unwilling to be the first to take the plunge. The Boston-based investment manager says that vendors need to focus exclusively on data-management-operations outsourcing before firms will trust their services. In addition, he says, there are core services that he doesn't see being offered yet, such as handling every asset class adequately, as well as the need to identify missing data, not just erroneous data, when reconciling exceptions.
For these reasons and others, many say that the concept is progressing very slowly. However, some industry insiders believe that it is simply a matter of time before these processes are outsourced in order to focus on stronger core-competencies. "It is a maturation but clearly going to happen," says Jordan. "It's a labor intensive activity that is going to benefit from sharing industry resources."