Starting with the Orange County disaster in 1994, risk has assumed a higher visibility. Why would you want to do this? The main reason why the plan sponsors are getting more into this area is that they want to minimize exposure to risk that they're not being compensated for.

Plan sponsors are also continuing to look at performance attribution. They are beginning to look at different sources of risk and an attributing return to those sources of risk. So they understand that there are some risks in here for which we're not being compensated.

There are two basic risk measurement techniques.

The first is the information ratio which is very simple as all it does is look at the annualized alpha that a manager is producing and then compares it to the volatility of that alpha. It is the active risk divided by the volatility of an active risk. Essentially, the information ratio is a measure of the efficiency with which the manager uses risk. If a manager has a low information ratio it means that it is hard to tell whether the alpha is as a result of skill or whether it's just luck. On the other hand, a high information ratio points to the fact there is some likelihood that there is some skill being employed. For application purposes, you would compare the information ratio of different managers on an after fee basis. You would look for managers who have high ratios.

The second measure is Value at Risk. It establishes a minimum threshold for downside risk that a plan is willing to take with a certain probability. So a plan might set down a threshold of $20 per $100 invested with a 99% of volatility. If all that you are prepared to lose is $20 for every $100 you invested in a particular market or with a particular manager and this will be true 99% of the time. With VaR you can set the threshold for the plan as a whole--a maximum amount that you are prepared to lose. You could also do the same thing with each portfolio manager. Asani Sarkar, economist, Federal Reserve Bank of New York.

There are benefits to diversifying internationally, but what is the magnitude of this benefit? Do I get an additional 1% annually? What we argue is that the magnitude of the gains from international investment is very critically dependent upon whether the manager can short sell the foreign country stocks or not.

How difficult, or how costly is it for managers to shortsell? This varies from country to country. The laws are different. It also depends on what kind of an investor you are. In the United States, as well other countries, there are rules and laws which preclude certain types of institutional investors from using short positions. For example, this might be considered breach of fiduciary duty for some investors. Then there were Internal Revenue Service rules in the U.S. which basically prevented U.S. pension funds from short selling.

In the global context, shortselling is even more difficult because you need a derivative on the country index. With emerging markets, derivatives of country indices do not even exist.

If a U.S. investor is diversifying into G7 countries, Latin America and Asia, and if there are no constraints in the investment (they can short sell to whatever extent they want) the gain from diversification is 3.5% per year over this period. This number has a 99% competence attached to it. So there's only a 1% probability that it's going to be less than this. So, they are fairly substantial benefits if managers are allowed to short sell.

If you redo this exercise and you say that managers are not allowed to short sell at all, then that number comes down to 0.5%. So this is saying that if managers are not allowed to shortsell, there is virtually no benefit for the U.S. investor of diversifying internationally. Whatever little benefit you get, pretty much goes away in the transaction costs. What they suggest is that from a historical point of view, any benefit that active managers have obtained will be crucially dependent on the ability to assume additional risk such as using derivatives or shortselling country stocks.

Fred Francis, vice-president, global securities lending and finance, Royal Trust, Toronto.

Securities and lending is an essential ingredient of risk management. In effect, when you lend securities, you're trading credit upwards. It is an exchange of securities, in which you lend a specific security and receive primarily a fixed income security.

It also enhances securities market stability. Most people associate securities lending with shorting and that is correct. Most shorting does not take place as naked shorts. They primarily take place for different strategies, whether it is hedging strategies, arbitrage strategies or others. About 60% to 70% of all the securities lent represent activities related to arbitrage. Whenever there is arbitrage, the effect on the market is to bring back the convergence of pricing to a more rational, stable level, as opposed to affecting the emotional volatility in fluctuation in pricing.

In the U.S., Canada, and the U.K., where the markets are very mature and securities lending is practiced quite heavily, the markets are fairly stable. In markets such as the Far East, where the regulators don't necessarily support the activity, you have dramatic volatility.

Securities lending should be practiced whether the markets are volatile or not. When the markets are volatile and unstable, usually the returns on securities lending are much better than they would be under normal circumstances. Also, professional risk management processes that are put in place should neutralize any effect of any additional risks.

One of these risks is regulatory risk. Market instability comes for a number of reasons, one of which is sudden and unexpected regulatory changes that take place in different markets. business. The key is how to deal with such situations.

The second risk is volatility risk. In effect, volatile markets mean that you have very little ability to adjust your pricing models or margins on an ongoing basis. One of the benefits is that we found out that in declining markets, when volatility is on the decline, in effect, margins are increasing quite rapidly. So volatility is negative or needs to be managed primarily when the markets are going up, not when they're coming down.

Mark Seasholes, professor, Harvard Business School, Cambridge.

When foreign investors are net buyers today, are they net buyers tomorrow, in a week, in a month? More importantly, when foreigners are net buyers, what's happening to the prices?

We have found that international portfolio inflows are slightly positively correlated across countries, and are more strongly correlated within regions. But what we really want to know is how do the flows interact with prices? The question is, do the flows and prices move together?

There is some ability for international inflows to forecast returns. In emerging markets, inflows predict on average to positive future returns. The majority of price increases do not occur over a short period of time, such as a few days. Rather, prices seem to rise subsequent to inflows for a month or two. In developed markets, inflows do not forecast positive returns.

Doug McCalla, administrator, San Diego City Employers' Retirement System.

Typical approaches to rebalancing a portfolio have been based on calendar-based adjustments (monthly, quarterly, semi-annually) or proportion-of-portfolio-based (percentage) decision rules.

The hypothesis to be tested was whether a rebalancing decision rule based on equal probability trigger points around each allocation in the portfolio would be more efficient than using conventional processes such as calendar-based or percentage based.

So there is information about the allocations in the portfolio in terms of their relative volatility compared to one another and that measure essentially normalizes around each allocation. If you plot one standard deviation around each allocation, you have the same probability for that allocation hitting the trigger point no matter whether it's a very small allocation--the micro cap growth stocks which were extremely volatile--or whether it's a large allocation to a very stable fixed income exposure.

Rebalancing is just about the most cost effective thing that you can do to add value. The question is, does your group have the right risk tolerance to implement it and systematically follow the process? If you do so, you have the opportunity to reduce your funding costs to the plan sponsor and you can literally set the table for benefit enhancements.

Session Highlights

Global markets are volatile markets. If there was one common theme to the presentations at the conference, it was that. But that doesn't necessarily hold that volatility is necessarily bad, as markets that are at a current low point can bring opportunities. It just means that investors must understand the challenges of going global, and manage the resulting risk accordingly.

Investors are certainly talking about risk management. "I'm not sure the overall level of risk has changed that much but certainly it's much more in our consciousness," says Rajiv Silgardo, chief investment officer, Barclays Global Investors, Toronto. But talk doesn't necessarily lead to action. "In spite of the fact that we're thinking more about risk and that it's more visible today, we still don't have a comprehensive set of tools or measures that one can use to measure this risk," adds Silgardo.

Though many are struggling to build an integrated framework to managing risk throughout an investment portfolio, risk management techniques can be integrated into familiar portfolio building steps. Take the setting of asset mix policy. "Instead of just stopping at the optimal mix, when you get into risk allocation what you're doing is going a step further and you're actually now trying to allocate the risks that asset mix brings to your plan. You are looking for an optimal mix that's going to reduce the volatility of the plan surplus in the coming years. That's basically what risk allocation and risk budgeting is about."

This is neither a straight forward or easy process. Most of the frameworks for looking at risk work well during fairly normal market conditions, but fail at market extremes. Comments from conference participants questioned much of the understanding behind such approacheds--just when you need it the most, some risk management techniques fail to help.

Simple approaches to risk, such as merely measuring downside risk, don't always tell the whole story either. Jim Stothers, chief investment officer, Ontario Hydro, Toronto. summed up the situation by quoting a high-profile investor Warren Buffet. "The emphasis on price volatility as a proxy for risk reflects the preference of the academic community for being precisely wrong rather than approximately right."

But while complete risk management frameworks may be lacking, there are still tools available to help control risk. Securities lending can be an important part of risk management, according to Fred Francis, vice-president, global securities leading and finance, Royal Trust, Toronto. "It is a risk management tool, and it also allows all sorts of hedging transactions which, again, are risk management tools." Securities lending is basically an exchange of securities and functions much like a swap. A specific security is lent against primarily a fixed income security.

And there are other risk management tools that are effective, but never share in the limelight. "Portfolio rebalancing is the low man on the totem pole in terms of time and effort spent in the portfolio creation process," explains Doug McCalla, administrator, San Diego City Employees' Retirement System. Yet he explains that it is actually a valid risk control process that maintains the long-term strategic investment mix that has been set. In a sense, it is the most fundamental way of controlling risk--know what your position is at all times.

So with all of the problems associated with implementing global investment strategies, why bother? Asani Sarkar, economist, Federal Reserve Bank of New York, supplies the primary reason--earning excess returns. "Theoretically there is a lot of research in academia and other places which have shown that you should expect that there are gains to diversifying internationally," he says, but adds that few have quantified it. He does, and shows that the gains from diversifying can be as high as 3.5% annually. So there is potential in international markets, to accompany the real risks.