With the release of RiskMetrics by J.P. Morgan & Co. last October, the financial world was treated to what some experts called "a quantum leap in risk management."
It's very possible that when the history of finance in the 1990s is written, J.P. Morgan's sharing of its highly respected in-house risk management system will be cited as the most important financial development of the decade.
Why? Because after Oct. 10, 1994, no one in finance can admit ignorance about risk management. Just as the adaptation of commodities futures to money thrust their uses on a reluctant financial community - resulting in a plethora of derivatives - the ready availability of algorithms and data to manage investment risk and pressure from regulators are compelling all financial managers to put risk management on their "must do" lists.
Assuming a fee range of 4 to 12 basis points depending on how often investors use risk management services, the Employee Retirement Income Security Act market alone could potentially provide $2 billion annually in new fee-based revenues for risk management vendors.
This is good news for global custodians who care for both foreign and domestic assets and who are likely to be the new vendors. Risk management offers them a great opportunity to create, market, and sell a new service that all of their clients will eventually be buying. It is very likely that they will have few competitors because part of the problem in measuring investment risk quantitatively is getting current portfolio prices. Because of what they do, global custodians always have current price data for their clients' portfolios.
Are global custody clients now rushing to their custodians asking for value at risk measurements for their portfolios as a result of the October 10th announcement?
"Not yet," said Francis D. Antin, senior vice president in charge of market and product development at Boston-based State Street Bank and Trust Co. But has this announcement stimulated any thought of future product planning at State Street?
"We were already thinking about it," Mr. Antin said, "because we know what will be important to our clients."
The development of sophisticated risk measurement services will differentiate custodians, because not all of them will be willing to spend the money on a product that clients aren't using yet. Being first with a financial product often means getting the lion's share of business. Some of State Street's competitors - like Citicorp, Chase Manhattan Corp., Bankers Trust New York Corp., and Chemical Banking Corp. - already have proprietary risk management systems they use to measure their in-house trading risks, but none have yet said much about offering it to global custody clients. Morgan has said the bank now has no intention of offering risk management advice.
Exactly when the financial community will integrate risk management into its present activities is hard to say, but it is likely to be sooner rather than later if the Office of the Comptroller of the Currency has anything to do with it.
"We have a policy statement, Banking Circular 277, that tells banks to pursue value at risk," said Michael L. Brosnan, national banking examiner with the Comptroller's office.
Value at risk is the amount of money that is at risk on a securities position. If the position is leveraged, the value is higher than if it is not, but value at risk measurements can be applied to both speculators and investors.
Mr. Brosnan said the Comptroller's office knows that developing a consolidated risk measurement system requires significant resources. He said the agency believes these capabilities are necessary for banks that are both dealers and quasi-dealers.
"In practice, banks that are dealers or active position takers don't have a choice," he said. "We also just put out an advisory that said banks need to be able to assess the impact of changes in market factors on their earnings and capital." Mr. Brosnan also noted that the Securities and Exchange Commission is encouraging risk disclosure. As a result, value at risk numbers are beginning to appear on financial reports.
So far, risk management has been applied to proprietary trading, which is often leveraged. The recent failure due to unauthorized trading of Barings, Britain's oldest investment bank, will only serve to speed a general acceptance of risk management for proprietary trading.
Last September, Comptroller's office officials met with William Ferrell, founder of Ferrell Capital Management, to discuss risk management techniques. Begun in 1989 as a funds operation specializing in finding and supervising proprietary traders - mainly for commercial banks - the firm now offers a stand-alone risk management service that projects a gain or loss within defined time and confidence intervals for portfolios.
Last December, in what may have been a first, the European Bank for Reconstruction and Development in London retained Ferrell Capital Management to measure the risks of portfolios run by outside managers.
Why? "Since any bank's capital is always at risk, we see ourselves as risk managers who generate return rather than asset managers who look at risk," explained Andrew G.K. Donaldson, a treasury/investment officer at the bank.
The result is a "middle office" that can offer risk management through independent mark-to-markets, and accounting reconcilements. This is akin to performance measurement that global custodians already provide their institutional clients - for fees, the most coveted type of income in the banking business today.
Does it work? "For the moment, this is a very satisfactory solution," said Mr. Donaldson.
"Potentially, this is a very rewarding addition to the present portfolio of services that global custodians now offer," said Robert Gould Jr., head of the financial strategy practice for Coopers & Lybrand in New York. In addition to the ERISA market, there are potential customers in mutual funds, managed futures accounts, and trust departments. Anyone who manages money is a potential client.
"Risk management is much more than just a defensive tool," said Mr. Ferrell. "It can also be used to allocate capital and to appraise trading performances by quantifying the amount of market exposure a trader, methodology, or asset class used to create profits," he explained.
So the added bonus for risk management users is the ability to allow them to quantify reallocation of all of their assets to define a combination of the highest yielding/least risky mix available in domestic, international, or global markets. A money manager could approach a client saying quantitatively that he or she earned 13% on investors' capital while taking only 5% value at risk while a competitor earned 15% but took a 10% risk.
Since it would be prudent to be able to do this, it is easy to imagine that as this service becomes generally available, ERISA will ask pension plans sponsors to employ it. If and when that happens, the global custodians who can offer this service will have a distinct competitive advantage.
There is nothing new about quantitative risk management. Harry Markowitz began to describe how it worked in 1956. Nor is there anything new in the 100-page RiskMetrics manual, which makes all the algorithms available free to anyone who asks. Morgan also provides some 53,500 volatility calculations cross-referencing 325 markets in 15 countries free to anyone who can tune into the Internet or CompuServe.
But Morgan's system is limited to a comparison and compilation of volatility that describes only the probabilities of markets exceeding standard deviations. While this can efficiently cover 90% to 95% of all risks, there are some 10 to 15 times each year when a market's volatility can exceed almost any reasonable standard deviation barrier its users set.
"Key to successful risk management is using all four parts of the process," said Mr. Ferrell. His firm uses it daily to survey risks taken by 15 outside proprietary trading firms to which he allocates clients' capital. According to Mr. Ferrell, those parts are:
* Converting securities and derivatives to comparable equivalents and marking them to the market.
* Computing volatility and co-variance to determine value at risk by country, market, trader, or aggregate;
* Running stress tests on concentrations of risk affected by current market conditions.
* Taking action to reduce unnecessary risk exposure.
Clearly, considerable experience with markets would be needed to develop good stress tests or "what if" scenarios. How much risk will a portfolio have if there is a major earthquake in Japan? What if New York is hit with a 35-foot tidal wave? Dreaming up relevant questions is not easy.
State Street's Frank Antin is not concerned with this. "What is surfacing as today's requirement is our clients' need for more information," he explained.
"New investment vehicles are surfacing as we speak," he noted. "No matter what type of horror stories you may hear about derivatives, they are not going away." He sees the development of value at risk services not as something new, but merely an extension of performance and analytics in multiple markets that State Street and other global custodians already do.
"What clients really want is ever-more complex fiduciary reports that have up-to-the-date or up-to-the-minute information on the strategies and tactics their managers are pursuing," Mr. Antin explained. "They want to know if the strategies are in compliance with the framework their managers are supposed (to be) operating under."
Constructing risk management systems is not cheap. Even though all of the information is available in published sources, Mr. Ferrell spent three years and "several million dollars" developing his system. Investors who want one will either have to create their own system or outsource it. Given the start-up and maintenance expenses, the good news for custodians is that all but the very biggest investors will outsource it.
Another problem custodians face is liability. Suppose one of them estimates risk for a client's portfolio and the risk actually turns out to be much bigger?
"The very fine line we will all be walking is providing investment advice," Mr. Antin explained.
Isn't performance and analytics the same thing? He said it is a matter of how high the level of information goes. The higher the level, the closer one comes to giving investment advice - thus the bigger the risk and the more clients will have to pay. Mr. Antin does see this as additional fee income to the 4 to 8 basis points clients now pay global custodians. "The constant work involved in producing this kind of valuable information will let us build in and keep a margin," he said.
As for liability, "All of our clients are professionals," he noted. "They know how custodians have operated in the past and they know that the boundaries of where we have been providing information are changing all the time.
"The liability issue will resolve itself as the service evolves."
The pyramid of excellence in global custody is getting steeper and the bar for offering new products, particularly risk management, is being raised. "There will be varying degrees of custodians' appetites for developing this that depend on their target markets," said Mr. Antin.
Will State Street get over the bar and stay on top of the pyramid. "We are going to play in that game," Mr. Antin said confidently, "because risk measurement is truly adding value and will truly be rewarding."
Only time will tell, but the potential of this fee-based service argues that risk management will not only be the next hot custody product, but a highly profitable one as well.
Mr. MacRae is a freelance writer based in New Yor