The scorecard system, described in an article in the November 2000 issue of Risk, is a heat-map system that measures 13 types of operational risks, such as fraud, modelling error, process and personnel. These are given different weights in each of ANZ’s 17 core business lines and generate a more accurate op risk capital charge than the previous method of charging a percentage of operational spend. According to ANZ chief executive John McFarlane, the change fundamentally altered the risk/return trade-off in a number of core business areas like credit cards and transaction services. And efforts to make risk management “a competitive advantage” have also boosted ANZ’s reputation with investors. ANZ’s stock price rose 11% for the year – 22% relative to the S&P 500 – making it one of the best-performing banking stocks in the world last year.
Lawrence’s petitioning as a member of the steering committee on regulatory capital at the Institute of International Finance and chairman of a 17-bank Basel taskforce has also brought a semblance of sense to the Committee’s proposals for op risk. Making frequent trips from Australia to the US and Europe, he demonstrated to the Committee how op risk can be measured in reality within a financial institution. The Committee responded by including the ANZ scorecard approach in its September paper on op risk IRB treatment. This eased bankers’ concerns about its over-reliance on op risk statistical modelling based on poor historical loss data.
The Australian native began developing the op risk framework within months of joining ANZ in 1999, after a near two-decade stint with a number of investment banks in the US. The idea was to change ANZ’s cost-cutting and revenue-chasing culture to one that also rewarded business line managers for sound risk management practices. Lawrence was supported by then-boss, ANZ chief financial officer Peter Marriot, as well as McFarlane – to whom he now reports. “We are reducing the cost of capital as a consequence of risk decisions… and this [op risk treatment] allows us to make superior business decisions. Mark has led the way,” says McFarlane.
The system has recently completed its first full financial year. All business managers now simulate their cost of capital savings/losses for planned new ventures via an intranet tool called ‘manage-my-capital’. This has created a new risk management culture at ANZ, with the bank significantly altering its economic capital provisions, says McFarlane, citing the credit card business as one area that looked like it had very high returns under the old op risk framework but more normal returns when analysed with the scorecard system. ANZ is now implementing a credit card fraud detection system called Falcon after it estimated an A$40 million op risk capital charge reduction – or a cost of capital saving of about A$5 million per year.
The bank also factors in op risk when analysing new purchases. “The global transaction services unit has been looking at acquiring a big piece of custody business. We wanted to understand the capital impact of acquiring these additional assets and calculated that. Most of the capital is op risk capital, and it had a big impact on the economic business case,” says Lawrence. A decision is pending on whether to proceed with the purchase.
Large projects – ANZ has about 50 or more – are also managed using the scorecard process: “We now bring a status report on all the major projects to the op risk executive committee every month, and bring a lot of transparency to the way those projects are running,” says Lawrence. “Are they on time? Are they under budget? Are the benefits consistent to the original business case? Projects that have trouble incur additional capital.
“You are penalised down the road if you ‘gild the lily’ in your business case – understating the cost, time, or overstating the benefits. In some cases we have replaced project managers,” Lawrence adds. “People now bite the bullet early. Risk management is about getting the right information to the right people at the right time.”
Lawrence admits that ANZ’s risk-weight estimates could be significantly out of kilter. “What we never claimed was spurious accuracy. We could be off by plus or minus 10%, even 20%. There is always more detail, but chasing it would be a ‘false Grail’, a waste of time. We needed to find a way to drive behaviour and we think we have highlighted the main risk areas and got people’s attention on those.”
So what’s next for Lawrence? “We are looking at a big initiative in active credit portfolio management. Not only securitising various parts of our portfolio, but actively shaping what comes in, in terms of limits, policies, asset classes, geographies.” But for once, instead of being the only pioneer, Lawrence will have the company of JP Morgan Chase and Bank of Montreal.
The week on Risk.net, March 10-16 2018Receive this by email