X

Banking behind the scenes

Back-office IT processes can account for a huge chunk of a bank's operating costs. But a McKinsey study of bank operations in the U.S. and Europe finds that there is more than one solution to this perennial dilemma.

4 min read
Banks constantly battle to streamline their back-office "factories," with good cause--these facilities can be responsible for up to half of their operating costs, excluding interest.

Yet whether the factories process mortgages, securities or other products, banks tend to use the same approach across the board: Process reengineering and general cost-cutting are the favorites, but outsourcing is gaining ground.

Most banks also treat the information technology operations that support back-office processes as a single, separate function requiring a separate cost-cutting program. But world-class banks take a different tack. In 2001 and early 2002, we studied the relative efficiency of the back-office operations of 13 European banks as well as how 20 banks in Europe and North America handled such operations. Our two samples included various types of banks, with a wide revenue range. We found that the world-class banks manage their back-office and IT operations as a portfolio of individual factories, each demanding a unique solution depending on its characteristics. Using these findings, we developed a guide to help banks choose the right sourcing solutions.

Our first study, on the relative efficiency of the back-office operations of different banks, showed that unit processing costs for the same product varies a good deal. In mortgage processing, for example, a cluster of banks had unit costs in the range of $100 to $200 for each mortgage, while others had unit costs as low as $50. Such variations, closely linked to scale, were largely explained by how well the banks had designed their factories. Banks with low processing costs had intrinsically inexpensive operations because they had developed focused products, flexible IT platforms, and lean, highly automated processes. These banks also could support a massive increase in volume without raising fixed costs, thus sharply reducing unit-processing costs

Banks with low processing costs had intrinsically inexpensive operations because they had developed focused products, flexible IT platforms, and lean, highly automated processes.

The second study, on the way banks run their back offices, found that the wide range of options fall into five "smart-sourcing" categories:

• Internal upgrades: Most banks pursue the familiar path of business-process reengineering and IT rationalization for their factories. But few upgrades are successful enough on their own to yield a lasting competitive advantage.

• Outsourcing: A favorite way of streamlining standardized, small-scale noncore operations is to outsource them. The U.K.'s Abbey National, for instance, turned over its credit card processing to the U.S. specialist MBNA. Banks currently outsource roughly 10 percent of their factory operations. Our research suggested that most could outsource far more.

• External co-sourcing: In a new variation on outsourcing--external co-sourcing--several banks pool their operations. This is an attractive approach if no large-scale provider exists, but so far, given the problems of integrating different legacy IT platforms, a rare one. Three mortgage banks in Germany, however, have decided to combine their volumes in a jointly upgraded platform. We expect to see more such ventures.

• Internal co-sourcing (shared service centers): Typically, multinational banking groups require each bank in each country they cover to process its own product lineup. But by concentrating their operations for a particular product, such as loans, in a single factory serving many countries, they can capture substantial benefits.

ING, for example, is setting up such service centers for several of its products, including mortgage processing, and HSBC is consolidating its wholesale-banking factories. Since it's difficult to connect these central factories to a number of geographically dispersed banks within the same group, only a handful of players are attempting this strategy.

But as banks adopt more advanced middleware, we expect that it will become easier to execute. An increasingly popular variant is to move factories to cheaper locations offshore: Citigroup, HSBC and Standard Chartered, for instance, have used their knowledge of Southeast Asia to move activities such as IT development and credit card processing to that region.

• Insourcing: Best-practice factories take on outside business to reach more efficient volumes. The South African bank Nedcor, for example, has cut its credit card processing costs through an internal upgrade and now insources from a European credit card issuer. Mergers are another way of gaining scale in operations. For example, the Royal Bank of Scotland is now moving the processing factories of NatWest (which was twice its size in risk-weighted assets at the time of the acquisition) to its own more efficient platforms.

Best-practice banks decide which smart-sourcing option to use by examining the relative efficiency and scale of the back-office operation behind every product line. First, a bank must assess the strengths of each factory and the complexity and sales volumes of its products, as well as its operational processes, IT platform and IT architecture. The bank can then classify the factory by such measures as efficiency and the potential for scaling up. The bank can choose the smartest sourcing option for each product line, depending on its strategic importance and the back office's starting position.

Best-practice banks decide which smart-sourcing option to use by examining the relative efficiency and scale of the back-office operation behind every product line.

Small, inefficient operations are "nonperformers," best suited to outsourcing or to small-scale internal upgrades. Most banks will find that their core factory operations, such as processing current and savings accounts, are "laggards"--high-volume activities inextricably embedded in custom-made, undocumented legacy systems. Despite these operations' inefficiencies, internal improvement may be the only option, not least because their sheer size makes them so important strategically.

"Innovators" are highly efficient processors lacking the scale to exploit their full value. For them, managers should consider scale-enhancing options, such as insourcing or building shared service centers to concentrate internal volume. Last, the "shapers"--big, efficient processors--should pursue insourcing, either from less well-favored banks or through mergers.

For more insight, go to the McKinsey Quarterly Web site.

Copyright © 1992-2002 McKinsey & Company, Inc.