The chaos of 2008 saw a series of shotgun weddings between banks. IT teams were asked to integrate their technology with no time to prepare. Some of those integrations are still ongoing. By Jake Thomases
The sun officially rose at 6:37 a.m. on Sept. 16, 2008. Much of Wall Street would still have been asleep so early on any other Tuesday morning. But sleep was a luxury on this Tuesday.
The bottom of the financial world was falling out. Merrill Lynch had been sold to Bank of America the previous Sunday. Lehman Brothers had filed for bankruptcy protection less than 24 hours later and AIG, teetering on the brink of collapse, would need an $85 billion infusion from the Federal Reserve on Wednesday.
A day later, US Treasury secretary Henry Paulson and Federal Reserve chairman Ben Bernanke proposed the first edition of what would become the $700 billion Troubled Asset Relief Program (TARP) bailout fund. So on that Tuesday morning, with insolvency lurking behind every headboard, sleep did not come easily.
One of the sleepless was Bank of America enterprise change executive Eric Livingston, who received the news—“We’ve bought Merrill Lynch”—from his bosses, even though he’d already learned of the merger through the media. “'We need you to help put this thing together,” his bosses said.
As red-eyed financiers on both sides of the Atlantic chewed their nails to the quick, and as stockholders wondered if their fortunes would make it through the day, Livingston made his way to work to begin one of the biggest technology mergers in Wall Street history.
A New Frontier of M&A
“There was an immediate recognition that this transition was different from the transitions that we had done in the recent past,” Livingston recalls. “This was a merger of the nation’s largest retail bank, credit card company and mortgage lender with one of leading global wealth management and corporate and investment banking firms.”
The 2000s was a decade of voracious consumption by Bank of America CEO Ken Lewis. Between 2003 and early 2008, he acquired LaSalle Bank, US Trust Company, MBNA, FleetBoston, and Countrywide Financial, becoming a national powerhouse in the process. Livingston, however, didn’t see any of it firsthand. He had been brought over as enterprise change executive in July of 2008, two months before the economic meltdown. Now he was being asked to help engineer an integration with a massive investment bank whose platforms may or may not have been compatible with BofA’s, in a deal that was struck over a weekend, with a Kansas-sized spotlight on everything his team did.
He wasn’t alone. Banks everywhere were being absorbed overnight. JPMorgan Chase had acquired Bear Stearns in March and would go on to take over Washington Mutual. Barclays and Nomura divvied up Lehman in late September. Wells Fargo took on Wachovia in October. Handshake agreements became official in hours or days as the credit crisis spread. And so it was left to the IT teams to make those deals work. Quickly.
“I know that a number of players involved would have liked to have taken their time,” says Peter Duffy, CTO of IT analytics firm Sumerian. “But they felt that if they weren’t able to show to the board that they had a strategy in place, they would seem to be failing. Many of them would have preferred to take a step back, to take a scientific, analytical approach, do a proper comparison of the performance and the capacity and the cost profiles of the systems that they have, and based on that, make a decision to go forward armed with the appropriate evidence for one or another of the systems.”
Sumerian was hired by two such entities—one in the US and one in the UK—to assist with the integration process. Duffy says the technologists sprung to work immediately, even while knowing that it would be a multi-year process. By the time they called him, they’d already attempted to draw up their own technology roadmaps, with results that were mixed enough to require a consultant.
“Because of the pressure we were under, the economic crisis that was ongoing and the desire to integrate as quickly as possible, we consciously made decisions with the goal of getting the firms integrated as opposed to building out the strategic architecture,” says JPMorgan’s Peter Cherasia. “It wasn’t until the beginning of 2009 that we could take a breather and say, ‘OK, we’re stable, we’re operating well, we have good command and control, but we’re not happy with all of our metrics. So we’re now going to need to put a platform in place to drive down error rates, to reduce breaks, reconciliations, and so on.’ The platform at that time had far too many interconnects, it didn’t have the level of capacity and resiliency you like to see, and it was not flexible in its ability to add new products. In the beginning of 2009, we had 14 different derivatives trading platforms. If you had your strategic vision and infrastructure built out and time to do the planning work, you probably would have made different choices in terms of the integration. But, that all happens ex-ante.”
It feels like a long time since Cherasia’s business card read “Bear Stearns” in the upper left corner. JPMorgan’s head of markets strategies was one of five Bear executives—he was CIO there—to be co-opted onto JPMorgan’s investment bank management committee after the latter acquired the former in March of 2008. Cherasia was coerced by CEO Jamie Dimon and chief administrative officer Frank Bisignano, among others, out of an early retirement and into a job on the integration team. He knew as well as anyone what Bear brought to the table, and how his new employer could leverage it. (Read more about Cherasia’s transition from Bear Stearns to JPMorgan in the June 2010 issue of Waters.)
‘It Was Ugly’
At the beginning of any integration, the systems—risk systems, foreign exchange (FX) systems, equity systems and others—run in parallel as the transition teams study the ecology of both institutions. Once they know what they’re dealing with, they start the rationalization process to decide what to keep from Bank A, what to keep from Bank B, and what to fuse together.
It’s a process not limited to software—just as each institution has its own fixed-income desk, each has its own fixed-income team. Power struggles between personnel often leave one group victorious and the other to limp, defeated, out of the organization, say analysts. One insider expressed surprise at the partisanship shown by his IT people during 2008, when he expected everyone to pull together to avert a crisis.
In some cases, says Duffy, the winning team gets to keep its own platform—certainly a suboptimal way of picking technology. Others are more scientific.
“It was ugly,” says analyst Larry Tabb, CEO of the Tabb Group. “To a certain extent, it’s got to affect people. I’ve lived through a couple of these in my life. At Lehman, we bought EF Hutton and I think 90 percent of those guys were shot within the first couple of months. And even though Lehman was on the winning side of that, it was pretty hard. You have triage teams that go in there to figure out what’s going on and then you’ve got folks to come in and work on conversions. Generally, when you’re in that kind of battlefield mentality, the morale—it’s challenging. But overall, you’ve got so much work to do you can’t focus on what’s going on around you.”
The plans for target operating models that were cobbled together in 2008 weren’t as detailed as the traditional merger-and-acquisition (M&A) plans. But they still included the same steps. One was to create deadlines and quarterly goals—when, for example, should each conversion be completed? Another was to outline the right metrics to allow for oversight and crosschecking of the project. There was also a need to eliminate redundancies while maintaining scalability, as order management systems (OMSs) can overflow if the new trade volumes are not properly anticipated.
JPMorgan handled the process best, according to Tabb. Over the years, it had drawn up a formidable acquisition playbook thanks to purchases of Jardine Flemings, Chase, and Bank One. That highly prescriptive playbook was vital during the crisis, as it allowed the firm to work off a script, even if it had to ad-lib a few lines. The only hole in the playbook, which had successfully assimilated talent, cultures, and services, was that it had let some of the heritage technology continue to exist side-by-side, never doing a wholesale architecture overhaul.
That was the situation Cherasia stepped into as he and his team debated the merits of adopting Bear Stearns’ platform or choosing from among JPMorgan’s legacy systems.
“It would have been a lot easier if all the target platforms had been JPMorgan platforms,” Cherasia says. “It’s more complicated when you have to migrate some of JPMorgan’s products to the acquired platforms because, in certain cases, they are better. Before you begin to move the position risk, you would like to understand what the strategic architecture and operational platforms are going to be. But you don’t have that luxury. So you’re doing triage and trying to make quick decisions with less information than you’d like to have.”
The tech team wrote 400 programs over the course of two years in order to connect Bear’s prime services business to the existing cash equities settlement capability.
Don’t Blame the Machines
Meanwhile, over at Bank of America, Livingston and his transition team opted to convert BofA’s existing wealth management clients to Merrill Lynch’s platform, which was highly regarded in the industry. For the corporate and investment banking business, BofA chose an integrated model that pulled assets from both institutions. The methods required different approaches. Conversion involved testing the scale, the conversion routines and the data that sat within the legacy applications. Integration meant real-time data transmission testing and data architecting.
The process was a complicated one, but not because of the technology—market volatility, business priorities, and regulations all obstructed the transition, Livingston says, but the technology itself never presented problems.
“As an integrated leadership team, we were experiencing the pressure of the markets as well as the intense visibility that we were getting in the marketplace from the media and various reporting outlets,” he says. “In that instance, the pressure was generated from the outside. On the inside, the teamwork was exceptional, mainly based on the fact that we made these joint decisions at the very beginning.”
With the fate of the US economy potentially at stake, the glare from regulators could have burnt a hole through sheet metal. Paulson and Bernanke had arranged some of the acquisitions personally; they were not about to let the firms run off and handle their own business, given the precariousness of the situation in which the US financial services industry was in.
Risk systems were especially important. The Securities and Exchange Commission (SEC) and the UK’s Financial Services Authority (FSA) wanted to know that each newly merged entity could run its risk analytics overnight for all of its assets. According to Duffy, some were afraid to initiate efficiency cutbacks on the risk management front, for fear of reducing capacity too much and running afoul of regulators.
Insiders at the time said they did not anticipate the amount of regulatory requests that came in, or the amount of time they would have to spend updating and reporting their progress to the authorities.
Three years later, some of the banks are still putting the finishing touches on their integrations. Barclays, JPMorgan, Bank of America, and Wells Fargo hope they made the right decisions on personnel, platforms, and redundancies. Opinion about the final products depends on where an observer sets the bar.
Sumerian’s Duffy is surprised that a number of the projects are still ongoing. BofA, for example, only shut down its transition team in July and is still tweaking a couple of small operational items.
Duffy says he is still working with two other banks on their transitions, but declines to name them, citing non-disclosure agreements.
The merged and acquired firms could perhaps have achieved a top operating state sooner, he says, by acting with less haste at the start. “But, then, hindsight’s always 20/20,” he adds.
Then there are those who recall those sleepless nights in September of 2008, who remember the panicked phone calls, and who set the bar near their feet. Larry Tabb had so few expectations of success that he applauds the banks for simply making it to 2011.
“I can only look back on it in amazement that there weren’t more things that went wrong,” he says. “You’re talking about an almost complete meltdown of the financial system and you’re looking at four mergers of unprecedented scale that occurred literally over a weekend and were implemented over a couple of months. I’m not sure how any team prepares for that.”