12 December 2008 Features

 

Been there, done that ... the role of IT in bank mergers

David Bannister & Dan Barnes

The past few months have seen bank merger and acquisition activity on an unprecedented scale. We spoke to some veterans of previous mergers for their perspective on what the IT and operations implications are.

When it was announced a few months ago that Lloyds TSB was to acquire Halifax Bank of Scotland, City wags were soon saying that IT management were hoping that Halifax IT was as un-integrated into Bank of Scotland as TSB was into Lloyds ... or it might have been the other way round.

Meanwhile, Nomura is trying to swallow parts of Lehman Brothers in Europe, Barclays Capital has other parts to deal with in the US, Bank of America has Merrill Lynch and Wells Fargo has Wachovia - and this is a far from exhaustive list.

The pitfalls of such large-scale integration projects are vast, and very public humiliation will follow any bank that screws it up.

But M&A among banks has been going on forever - in recent years the only thing that has been holding it back has been competition law. Many of the banking operations - and all of the big ones - that were operating at the beginning of the year are the result of mergers. JPMorgan Chase Chemical Cazenove Bank One, for instance.

So what lessons do those historical mergers have for those struggling to cope with the current crop?

David Weymouth, currently chief information officer at RSA, was CIO at Barclays at the time of its acquisition of Woolwich, and Nigel Walder, chief executive of operational risk specialist BCS, who was CIO at NatWest Global Markets when Royal Bank of Scotland took over, both have similar views of the experience.

"Get a good plan together and get the right people to execute it; don't agonise. Where these things get lost is when you end up in a big technology debate," says Weymouth. "Quite a lot of this happened at the time of the Lloyds TSB merger - there was massive religious warfare between the technology people. You just have to kill that and move quickly."

Walder agrees that this is an area fraught with pitfalls, and suggests that maintaining discipline is important. "Given the scale of what's happening, there is very limited experience in the marketplace for dealing with such massive transformations. It needs exceptionally formal programme and project management - the business as usual approach to change won't work in these large scale M&A deals," he says. "Evaluating and integrating technology should - as much as possible - be business led decisions. The default position is that technologists and business crossover managers will collapse into an analysis paralysis phase. They will overanalyse what is the best system, what needs to happen. They'll get out the theory books."

This is not a good thing, says Weymouth: "The theoretical attraction of ending up with some best-of-breed mix is just that - it's theory."

That said, there are obvious overlaps, and choices will have to be made.

"If we take Bank of America and Merrill Lynch - looking at it hypothetically - both will have an equity derivatives business," says Walder. "You only need one equity derivatives platform. Now organisational deployment must have highest priority. The way I'd like to see it happen is I'd like to know who will be the head of equity derivatives. Then I'd ask the head ‘what will be our equity derivatives platform?' Some won't need to run a selection analysis programme: they'll say ‘we're using system A' or ‘we're using system B'.  You need to know who is accountable or making that decision and on the front office side it should be the trading desk."

Weymouth says that this is sometimes overridden by management issues. "The first key thing is that you have to make sure that business and IT are lined up, but in general the IT community will put together a plan that says you will end up with one system and one set of infrastructures, and that will be largely from the acquirer. What sometimes occurs is intervention by senior management of the acquired business, who say ‘no, no: we like what we've got'. You've got to make clear-cut choices early in the process. You should test them, but by and large, you should stick to them."

There is also the question of brand. "You get some interesting questions around the brand, particularly with older systems and hard-coded Cobol - a lot of the association of the brand is in that hard code, so in a funny way a decision to retain a brand cascades into all sorts of systems areas," he says.

"From an operations point of view the brand and culture are key. To really maximise the cost reductions you have to be working toward a single culture - a single set of applications, a single set of processes, but also a single culture," says Walder. "If you're inconsistent from a brand perspective externally it's unlikely you'll be consistent internally."

Without commenting on specifics of the Barclays/Woolwich merger, Weymouth points out that as it was finished Barclays also "downplayed and repurposed" the Woolwich brand.

While externally RBS and NatWest have maintained separate identities, internally they are driven from the RBS systems, and this was generally regarded as one of the key factors in RBS's rapid rise in world rankings

"The one characteristic of the RBS/NatWest takeover that did come across was that ‘you pick one bit of the bank and you stick to it' approach. Generally, though not always, that is going to be the acquirer's systems. The reason that is important is that you can do some planning with some certainty," said Weymouth

And planning is another key to success: though in the case of some recent shotgun weddings, there has been precious little time for that, both Walder and Weymouth say that this is often the case in M&A situations.

"Senior management want to complete the integration as quickly as possible and you'll often read that the purpose for speed is to lock in the cost reductions. While that's true you also want to get out of this period of heightened operational risk as fast as possible," says Walder.

"The challenge, if it is a public acquisition, is that you just don't have the information. The critical thing is that very, very quickly after the deal is struck you have to get in there and kick the tyres properly. Particularly around the infrastructure," says Weymouth. "It is all about getting to the nitty-gritty and how quickly you can get to a detailed analysis of what you've got."

A further question is raised by the relative size of the two parties, and the hostility or otherwise of the acquisition.

"What we've seen in some situations such as Santander and RBS/NatWest is that, to shortcut the analysis paralysis, the banks' management say, as a general rule of thumb we're taking bank A's system," says Walder. "Some people will complain that their systems are actually superior and that may be true, however as an 80/20 rule it's much braver and much safer to get that mandate out so you can crack on with integration."

Weymouth says that it can work the other way around, when the smaller party's systems are small enough to be absorbed, though he too is more inclined to crack on with implementing the integration rather than risk losing control.

"The bit that people forget, I think, is that if you have a systems and infrastructure landscape that you own and manage, you've got to a point where you know how to make it work. If you start to make it more complex by bringing in different applications and different bits of infrastructure, as opposed to migrating, you end up with something you don't know and you have to go through all that process again," he says. "It's self evident in a way, but when I think of some of the smaller mergers we did at Barclays, where we thought we'd integrate someone else's system, it only really worked when we integrated stockbrokers. Generally it wasn't a good idea"

In the very large cases, IT efficiency has played a large part. Santander's acquisition of Abbey was largely predicated on cost savings generated by switching Abbey to the Spanish giant's Partenon system.

That wholesale replacement approach is also being adopted by HSBC, where the One HSBC strategy led by chief technology and services officer Ken Harvey is creating a global IT platform that can be rapidly deployed from a handful of regional databases. This will make the IT outcome of any future HSBC acquisition a given: the target will become part of One HSBC. Any sufficiently interesting technology will be absorbed and redeployed to other parts of the operation.

The RBS, Santander and HSBC top-down imposition of IT approach is often cited as an instance of how agility in IT can provide  leverage in an M&A situation, but Weymouth is sceptical: "I think it should be a major factor: I suspect in reality the thrill of the chase puts it on the back-burner," he says.

Outsourcing

"The extent to which people have outsourced makes things more complex, and you just need to understand the implications: are you going to cancel the contract and move to someone else, or can you live with it?" says Weymouth. "It's a dimension that is becoming more and more important. Who are your major partners, your service providers? Increasingly, people will have application development contracts or infrastructure management contracts and so part of your consideration - apart from the contractual issues - is who can you work with and how can you integrate that? More and more financial services businesses have an international dimension. Even if they are purely domestic, they'll still have offshore contracts. Again, that comes down to due diligence you can do."

Walder is also wary of the role of outsourcing "I would be very cautious around outsourcing. There are two philosophies: I've got a mess and I'm going to outsource it to let the service provider come up with a deal that's structured in a way to get me better service. The more logical is to fix the problem and then outsource it to a third party to see if they can do anything with the costs."

Human Factors

In all of this - particularly in the current climate - the personnel issue is very important, both in terms of the immediate worries people will have for their livelihoods, and for the choices that can be made.

"I don't think one should be shy about it, but you have to be realistic; it will, inevitably, create some tensions," says Weymouth. "Normally, what it is about is what are euphemistically called the Social Issues in any merger or takeover: who's going to be in the top jobs? If you have some of the people from the acquired business in the top jobs they are going to want to keep some of the things that are near and dear to them. No surprises there."

The people in the top jobs can also affect the overall morale of the firm and they have to beware of being seen to steamroller over people.

"Communication of values and strategies is essential, and people tend to put a low priority on the task. What hits the radar of a C-Level executive are the huge, broad, obvious issues that seem to be more immediate. One of my fears is that while the market is bad, senior management think they don't need to get peoples' buy-in because quite frankly where have they got to go?" says Walder. "While it's true that competition for resources is significantly less than in a good market, if you aren't communicating well with your team then frankly morale will be pretty low. During these periods of change within banks, you have to recognise that operational risk spikes dramatically and although people won't be looking to move elsewhere they won't perform at their best. You need to mitigate that."

In contrast with his views that taking a best-of-breed approach to systems and applications integration is more theoretical than practical, Weymouth says that on staffing it is essential to retain personnel from both sides, else there is a risk that crucial knowledge will be lost in the transition period. Building on that allows the merged group to build a team with strengths from both sides.

"That happened with the Woolwich: we ended up with a much richer, more talented team because of the people we brought in from the Woolwich side." BT