March 21, 2014
A merger of three investment managers had just been announced when BMO Global Asset Management made the decision to revamp its portfolio accounting procedures. What followed was a near three-year project that made the company the proud owner of an IBOR, and an example for the rest of the buy side. By Jake Thomases
As buy-side institutions consider the merits and costs of developing an investment book of record (IBOR), BMO Global Asset Management is relaxing on the porch with a pipe, an iced tea, and its feet on a shaggy dog. The debates are now quiet for BMO’s Chicago-based buy-side arm; the blueprints are put away; the implementation is largely over. While others mull philosophical topics like whether an IBOR should support the front office or the middle office, or whether IBOR represents a new concept or merely a new wrapper on a portfolio accounting system, BMO is already reaping the rewards of the system.
The scope of what constitutes an IBOR has been in constant evolution for decades. Barclays Global Investors has taken credit for inventing the term around 1999. The transition from user-developed tools to an enterprise IBOR barely got off the ground in the ensuing years, either because of a lack of need or a lack of understanding of the value of holistic, intraday position data. By beginning its adoption process in January of 2011, BMO could fairly be called an early adopter, deserving of both the acclaim and the scrutiny that accompanies such a title.
“The right third-party consultant will understand what it means to use the system with these other applications that you have hanging off of it: your portfolio analytics, your order management, and how you configure firewalls and deal with batch and file transfers.” —Todd Healy, BMO Global Asset Management
Yet that process was not triggered by a great faith in IBOR. The term never even touched the tongue of decision-makers during the initial months of 2011. Rather, the organization was in the process of merging with the investment management arms of BMO Harris Bank, its US affiliate, and M&I Bank, which BMO’s Toronto-based parent company agreed to buy in December 2010. It found itself with a heaping pile of portfolio management systems. The cost to maintain them all would have been high, to say nothing of the inefficiencies created by such a tangled web.
The collection needed to be pared. Todd Healy, the vice president in charge, considered his options. M&I utilized a trust system as its book of record and an operational data store to feed downstream systems, including the order management system (OMS). Harris was getting its start-of-day positions from reports, and through overlaying into the OMS. None of the legacy choices seemed sufficient. The $128 billion asset manager had plans to grow into new markets with new products and bigger assets; it needed a consolidated and intra-day overview of its positions on a grander scale than any of the available legacy systems would provide. To get truly consistent data it needed more than a synchronized solution—it needed a single solution.
Wouldn’t it be nice, Healy imagined, if we had a centralized book of record that we could feed down to the FactSets, to the Barclays, to all these other analytics systems we’re using? “We knew what we needed—we just didn’t have the right label to put on it,” he says.
An earlier organizational decision to avoid building any software internally meant Healy would have to find a third-party provider, and the money to hire one—the initial allotted budget was sufficient only to migrate customers onto an existing legacy system. The team had to push for additional investment into the ambitious project, which it received.
In January 2012, a year after beginning the brainstorming phase, an agreement with Copenhagen-based SimCorp was signed. Among the features that stood out about the flagship SimCorp Dimension investment management product were its scalability and breadth of securities coverage. SimCorp’s IBOR promised a single source of data for accurately determining positions, performance, and risk exposure. Without it, BMO would continue to work off partial or out-of-date data, or have its portfolio managers produce more accurate figures manually, something Healy had already seen too much of. It was an absurd use of their time.
“We had to make sure we had a really good understanding of who the users of the data were going to be and that we were meeting their specific needs for that data, both in terms of timeliness and the view they are going to get,” says Lance Ihinger, managing director at Cutter Associates, a Boston-based consultancy that was brought on to assist with some of the business decisions. “One of our top priorities was being able to get the cash balance to the people who needed it—in other words, right when these traders come in in the morning they’re going to need to know what their cash is—and being able to understand how the corporate actions cash affects their position.”
Originally, the users of the data were intended to be in the middle and back office. As the benefits to the trading desk became apparent, as additional modules were offered by SimCorp, the scope of the project expanded.
It was important to keep everybody on the same page as the vision changed. The single most important piece of advice Healy would offer to investment firms following BMO’s footsteps is that everyone in the organization agree on the purpose of the IBOR. A Waters breakfast briefing on IBOR in February revealed that there is still a lack of agreement on what functions it should encompass and which divisions should be the primary beneficiaries. No single organization can get industry participants to come to a consensus, but they can at least come to a consensus internally. With so many disparate IBOR opinions floating around, it would be easy to find the front, middle, and back office, as well as IT and the executive level, taking different tacks.
Mo’ Voices, Mo’ Problems
Unity took on special importance in this case because of all the parties involved. The M&I team was working out of Milwaukee, the Harris team, including project manager Healy, out of Chicago, and the BMO team out of Toronto. Most in-house IT was done in Toronto, although Sal Auditore, a consultant who acted as the technical go-between with SimCorp, recalls the IT staff coming in three groups, each handling a different part of the task. Most of the technologists were full time; most of the business people were not. There were some that spent as much as 80 percent of their time on the implementation, while others spent merely 10 percent.
Then there were the five dedicated team members from SimCorp, along with subject matter experts on security mapping, accounting, performance, and the like that slipped in and out. BMO also brought on Cutter Associates in May 2012 and Adeptyx, the Bloomington, Minnesota advisor where Auditore is a director, in late 2011.
“The vendor knows their product really, really well—in a test environment,” says Healy. “The right third-party consultant will understand what it means to use the system with these other applications that you have hanging off of it: your portfolio analytics, your order management, and how you configure firewalls and deal with batch and file transfers. By bringing them in, we avoided a lot of the pain we definitely would have felt if it had just been us and the vendor.”
At its peak, there were 25 to 30 people on the project, converging three lines of business onto one platform. It produced some confusion, and even a little friction, in the beginning.
“Sure, it was challenging,” says Gregory Nuckowski, leader of the SimCorp team. “There were a lot of moving pieces. There were a lot of stakeholders that needed to be aligned. But having clear direction from the top down was critical to our collective success.”
Healy made sure to keep in conversation with his portfolio managers and traders before and during the implementation. He wanted their input, and just as importantly, he wanted them to feel like their input was valued. Their recommendations would help shape the interface and mechanisms of the IBOR, because installing a robust system is only half the battle. If no one uses it, it becomes a robust piece of artwork, untouched above the fireplace. By encouraging the trading desk members to outline their desired specifications, Healy got them to buy into the project as something that would help them do their jobs better. He had no doubt that, without buy-in, they would greet the IBOR’s eventual debut with a sneer before turning back to their trusted spreadsheets.
“IBOR is pretty much a new phenomenon,” says Auditore, a veteran of SimCorp implementations, including at Wells Fargo. “It’s about trying to get your OMS, your client reporting, your performance, and your portfolio accounting on one system, or two at the most. Before that, it was all best-of-breed stuff. You’d end up siloing and have data going across four or five different systems. IBOR gives you all those modules on one platform. They’ve taken in post-trade, compliance, client reporting, performance, and accounting.”
Delayed But Engaged
Assigning all of the involved parties to their appropriate roles took a few months. There was also a small setback regarding a decision to have the SimCorp product run through a third-party application service provider (ASP) environment. The cost was deemed too prohibitive, forcing the Toronto IT team to abandon the ASP plan and host it internally. Configuring other aspects of the platform also proved occasionally trickier than anticipated.
These early delays led to a timeline readjustment. The original go-live target of July 2012, considered aggressive even at the time, was pushed back. Healy says he was always forthright with his bosses about the time commitment involved, so that small setbacks did not significantly disrupt expectations. No one thought it would be done in nine months. Still, he admits, there were times when the staff felt worn down from pushing as hard as it did for as long as it did.
The first major milestone, according to SimCorp’s Nuckowski, was the successful conversion of data off of the legacy platforms. Accounts from the three merged investment firms were onboarded in phases. M&I was first, in January 2013, one year after SimCorp commenced the implementation. Legacy BMO, slightly delayed because of additional OMS testing, came in May 2013. Harris, in September, was last. That marked the end of SimCorp’s official involvement.
From the start of the brainstorming phase, the project took a little less than two years; from the start of the actual implementation it was less than two. Auditore calls that typical for a portfolio accounting system. Since the IBOR should do more than a portfolio accounting system, BMO’s timetable should be considered a win. It helped that, although there were adjustments—for instance, Dimension wasn’t built to handle batch jobs on the 600,000 record files M&I was sending over to be processed—SimCorp did not have to do an excessive amount of customization to its core solution set.
The budget for all BMO projects is divided into a business bucket and an IT bucket. Healy says the business half was under budget, while the IT half was right near target. BMO would not release the final figure, except to call it a “very significant spend.”
Ahead of the Field
“SimCorp is such a massive system that can do so much, you find out what it can do and start adding more things and tweaking what you put in before to improve your organization,” says Auditore.
Sure enough, Healy’s team has been tinkering ever since September, when the vendor and the consultants and most of the internal people packed up their tools and went home. However, just because the IBOR is in live production doesn’t mean there isn’t work to be done, and will be for the foreseeable future. For one thing, while a high adoption rate by members of the trading desk is a good thing, it means their suggested adjustments will continue flowing in at a high rate. There were also modules, like for reporting, that were shelved for post-production because they weren’t critical to the basic rollout.
With the core system in place, BMO believes it has improved its efficiency, timeliness, and data quality. The IBOR illuminates not only current positions, but predicts future positions using anticipated corporate actions and reinvestment for short-dated instruments. The back office reports more accurately, the middle office derives more accurate exposure calculations, and the front office makes better trading decisions.
Yet BMO finds itself in a relatively small club. A SimCorp survey of 136 buy-side executives last month revealed that less than half are able to generate intra-day position snapshots. Sixty-three percent said their front, middle, and back offices do not work off a common set of real-time position data. Despite those recognized problems, 58 percent dismissed the possibility of an IBOR initiative in 2014. Many of those who have commenced work on an IBOR are in the early stages, where BMO was two years ago. BMO and Healy have become proselytizers—and guinea pigs—for the technology. An authoritative voice isn’t a bad position to hold. And if that position ever changes, it’ll surely be captured, intra-day, by their new IBOR.
- BMO Global Asset Management wasn’t a big believer in IBOR as a concept. Its original intent was to figure out a solution to the bevy of systems it inherited from a merger with the investment management arms of M&I Bank and BMO Harris Bank. IBOR was the idea that eventually sprang from that.
- Its problem-solving session began in January 2011. A year later, it contracted SimCorp to provide it with an IBOR. The first accounts were migrated to the SimCorp Dimension platform a year after that, in January 2013. The last were migrated in September 2013. The asset manager called it a “very significant spend.”
- The team of 25 to 30 people working on the project at any one time included people from all three banking entities, three IT teams, SimCorp, and at least two consultancies—not to mention the executives and traders that were consulted throughout. It took a few months to get everyone into their proper roles, but once the implementation direction was settled, it moved efficiently.