In part two of his examination of risk on the buy side, Anthony Malakian looks at how the regulatory environment is changing the way risk is managed and reported on, and delves into the buy versus build trade-off for risk-related technology.
The lasting effects of the 2008 financial crisis are still plain to see. Many components of the Dodd–Frank Act have yet to be finalized and deadlines remain undecided. But one success—depending on how you look at it—is the implementation of Form PF, an offshoot of the sweeping regulatory overhaul, designed to allow the US Securities and Exchange Commission (SEC) to examine systemic risk among buy-side institutions such as hedge funds and private equity firms.
Starting late last year, large hedge funds—those with more than $5 billion under management—filed their first Form PF submissions. Most midsize and smaller funds either filed their first returns in the first quarter of this year, or will do so midway through 2013.
Form PF requires COOs—along with technology and risk officers—to answer a slew of questions pertaining to enterprise risk, excluding positions. While up to this point investors haven’t requested these forms, they do want to know that funds are submitting them. But that may change, and if it does, it might well alter the way firms go about collecting and distributing data across the enterprise.
One chief risk officer at a hedge fund with more than $5 billion under management says that right now the buy side is primarily concerned with populating the form and getting used to doing it on a regular, quarterly basis. But fast forward a few years from now and he says the filing will be something that funds are used to completing. More importantly, the SEC will have a better grasp of what it is looking at and will be able to make comparisons with previous filings and with similar funds. The risk officer says that in about two years, regulators will come back and ask why there are differences between current filings and those filed in 2015.
“I was having a conversation with another senior risk guy who had been around for a long time and he said that this was going to force firms to take a more detailed look at their numbers,” he says. “In the past, they may have generated the data and understood that it existed, but they may not have taken ownership of it. But now they are putting it in a hard document that’s going to the federal government and is accessible to investors and other people, or brought up in court, so it’s something that they need to be able to speak to and act on. They don’t need to change the way they trade, but much more that, the number they put on that form has to have much more belief behind it.”
Part one of this feature looked at risk management and modeling in the front office. This piece focuses on how risk metrics are captured and distributed in the middle and back offices, which starts with a high-end data warehouse.
It’s not just Form PF that is changing the way that firms aggregate and examine risk-related data. The Open Protocol Enabling Risk Aggregation (Opera) is an industry-led initiative aimed at standardizing reporting procedures for the collection, collation and conveying of risk and exposure information. In Europe, the Alternative Investment Fund Manager Directive (AIFMD), adopted by the European Parliament in November 2010, could lead to similar risk reporting requirements for European Union-domiciled funds.
With these sweeping new requirements, firms are finding out just how arduous the task of data aggregation can be. Retrieving data from internal systems, third-party vendor systems, and fund administrators can be daunting. The ability to aggregate that data from one point, with a single login, and then tailor it to the individual user—since different managers require different reports—is the key change in data warehousing and in developing end-to-end data management solutions, according to Leon Abudaram, CTO of Citigroup’s institutional clients group, which includes fund administrator duties. As a result, Citi has launched OpenAi, an interface used by middle-office personnel at funds of hedge funds and other buy-side firms to better aggregate performance and risk data.
“The chief risk officer will look at different data points than a chief operating officer,” Abudaram explains. “We have a tremendous amount of data; the key is to be able to figure out what’s relevant to that specific role and that specific person—that’s the value this system provides.”
More so than the disconnect between what CROs and COOs look at, Mark Connors, founder of Risk Dimensions, who left Stamford, Conn.-based hedge fund Diamondback Capital as its head of fixed-income and counterparty risk in January 2012, says there has always been a disconnect between a hedge fund’s risk group and its investment group in terms of reporting and daily workflow. The ideal scenario is that hedge funds’ risk systems can develop reports that analyze what happens if multiple factors go awry. But Connors says that only about 10 percent of firms are really trying to develop this, and that’s not because it’s a technology challenge. The real reason, he says, is mental.
“Behavioral, behavioral, behavioral—that’s it,” Connors says. “It’s a case of the manager saying: ‘You’re not hearing what I’m thinking.’ You’re used to seeing something one way. Risk departments have left a lot on the cutting room floor that’s not either offered to or picked up by the investment team, which has the potential to add a dimension the manager is not currently seeing. It has been this way forever.”
But before buy-side firms start contemplating analyzing multiple risk factors and what they might mean to the business, they need to address the ongoing challenge of ensuring good data quality. That is a prerequisite or else it’s back to the old maxim: bad data in, bad data out. That’s where the data warehouse comes in.
Marshall Saffer, COO of MIK Fund Solutions, says there are four dimensions necessary to provide the plumbing for a successful data warehouse: knowing everything about ones securities; knowing everything about the market data for those securities; knowing how that security is held, i.e. positional data; and a time series that provides the outlays of the previous three dimensions, day, over day, over day.
“Once you’ve got these dimensions, any type of reporting that you want to do on any data set—whether it is market data, risk, or Form PF—is academic,” he says. “The problem is that most people don’t solve those four dimensions.”
Furthermore, flexibility is a must, according to Joseph Amarante, founding partner at NorthPoint Solutions.
“Historically, data warehouses have been very static,” Amarante says. “Every time you wanted to make a change to them, it required a programming effort. The securities master needs to be flexible and not require constant programming. Our clients want to be able to add new attributes, link those attributes with minimal programming to different data sources, and have it automatically populate and flow through the whole system.”
Buy versus Build
The debate around whether to buy or build the required technology in-house ebbs and flows, regardless of whether it is a data warehouse, a risk system, or a trading platform. But of late there has been greater acceptance of third-party-provided solutions. Constrained budgets and IT staff numbers—not to mention diminishing expertise of those dedicated to IT internally—has made the proposition of buying a more palatable one. At a minimum, hedge funds are now more likely to buy the “skeleton” of a system and then build their specific requirements around it with internal developers.
One COO of a $5 billion hedge fund says his decision to enlist outsource provider Gravitas was an easy one post-2008: “We’re not a database firm and we’re not a risk firm—we’re an investment firm,” he says.
By switching to Gravitas, he says his firm was able to eliminate two full-time positions and reduce the workload on IT. Additionally, the fund cut expenses by 65 percent. The COO says his firm has a broad-ranging portfolio that at any time could have exposures across a wide swath of areas. While the fund normally doesn’t experience multiple exposures simultaneously, when there are opportunities in, say, distressed debt, portfolio managers want to be able to respond immediately and trust the data they are being fed.
“One of our goals on the risk side was to have capabilities in all of those areas, but at the same time, being the size that we were, it’s not like we had a 12-person staff on the risk side,” he explains. “So in late 2007 and early 2008, we bought a multi-billion-dollar bank debt portfolio. We hadn’t really done a lot of that before—it was really something new to what our infrastructure and risk team had experience in. Within a couple of weeks, we went from almost no exposure to a couple billion dollars of exposure.”
Jayesh Punater, founder and CEO of Gravitas, says the ability to provide data warehousing, a security master, reporting functionality, and pricing tools, all in one place is what the industry is now demanding.
“It’s very expensive to do everything in-house and a lot of funds can’t afford it,” Punater says. “The needs keep changing. Managers want control, but it’s expensive to create risk reports on a daily basis—before, requirements were weekly or monthly.”
Steven Harrison, CEO of Imagine Software, says he concurs with Punater and that hedge funds simply aren’t interested in being software developers anymore.
“In this environment, they put their money to use for their investors very quickly, rather than spending it on IT,” Harrison says. “By outsourcing, they’re saving investor money on overheads—they don’t have to build a datacenter and they don’t have to hire system administrators, and so on.”
Even with the cost savings that buying as opposed to building can provide, it’s inevitable that the buy side will always push to get as much out of their vendors as possible. One head trader at a hedge fund that manages about $150 million says vendors still need to better understand how hedge funds use these reports. The problem with the vendor community, he says, is that when there is a problem, they are too busy developing tools and systems without really understanding the strategy or how the output is incorporated.
“People run these reports and distribute them to investors,” he says. “But the stuff they really care about for internal purposes, trading purposes, are very specific to the hedge fund community that they live in.”
● Regulations and standards such as Form PF and Opera have led hedge funds to reexamine how to aggregate and distribute their vast data volumes. This leads to fund managers needing a better understanding of, and a true belief in, the information that is being provided in those documents.
● Key to these endeavors for reporting and analyzing risk is a robust securities master and data warehouse.
● With decreased budgets and shrinking internal staff, the idea of buying off-the-shelf solutions has become more popular. Funds now tend to buy the skeleton and develop their proprietary tools in-house.
Click Here to view as a PDF