Asset-allocation managers must respond to changing conditions in financial markets if they are to best serve their clients, according to participants in a panel discussion sponsored by Morningstar Canada.
"Markets are dynamic, so (portfolio) optimization has to be dynamic as well," said Geoff Wilson, who leads the asset-allocation team at TD Asset Management Inc. and is a member of the TD Wealth asset-allocation committee.
Held on Oct. 21 in Toronto, the Morningstar Executive Forum panel also featured Michael Gates, director of the multi-asset strategies group at BlackRock Inc. in San Francisco, and Paul Kaplan, director of research for Morningstar Canada. The panel moderator was Michael Keaveney, director of the Morningstar Investment Management division of Morningstar Canada.
Wilson said TDAM's team spends a lot of its time on forward-looking analysis of what to expect from the markets. This helps the team form opinions on which "sub-asset classes" (within equities or fixed income) are likely to do well, and which investment style is likely to have better results in a particular market environment.
For example, he said the resources-heavy Canadian small-cap segment tends to thrive when natural resources are in high demand. At other times, he added, the team will reduce its Canadian small-cap exposure in favour of other types of assets.
Wilson said longer-dated government bonds, a major source of diversification, will be at the most risk if economic growth takes off and the U.S. Federal Reserve moves to raise interest rates. Conversely, he said that since there's little scope for further quantitative easing on the part of the Fed, another economic slowdown would pose a big challenge for investors.
It's important to have a view as to where you see the economy and the financial markets and what will change, said Gates, who is responsible for BlackRock portfolios for Canadian and U.S. mandates. The sharp increase in price-earnings multiples in the U.S. market is not going to continue, he said, "and that's going to change the trajectory of returns." Even so, his team at BlackRock has been overweighting equities in its portfolios for the past couple of years, and has continued to do so over the last six months.
Gates said bonds remain useful for mitigating risk in a 60-40 type of portfolio that combines equity and fixed income. "In terms of asset classes that are cheap, there aren't any right now," he added. "I don't think there are any slam-dunk asset classes at this point."
Decisions on risk are among the most important in determining the appropriate asset allocation, said Gates. This is not only a matter of deciding on the mix of asset classes.
In creating fund-of-funds portfolios, Gates said asset allocators also need to assess the costs versus the benefits of holding funds that are actively managed and therefore deviate from market benchmarks. Key questions to ask, he said, include: What risks are they taking? How correlated are these risks with other funds? What potential returns are they providing?
"Typically, active managers are much more expensive," said Gates. "So you've got to always hold their feet to the fire in terms of these measurements to make sure we're getting something for the cost."
Gates agreed with Wilson that when crafting optimized portfolios, asset allocators must be vigilant about changing market conditions. "Naïve use of historical data in optimization is dangerous."
Kaplan said portfolio optimization -- the quantitative process of combining asset classes to achieve the best risk-adjusted return -- is a useful tool to help understand risk-return trade-offs. But he cautioned that it is not enough to avoid some of the extremes in allocations that pure reliance on optimization techniques can produce.
He said extreme allocations can be mitigated by imposing constraints on the portfolio. Monte Carlo simulation is also a useful technique to determine what asset allocation is most likely to provide a smoother ride under various market conditions. "At the end of the day, there's still going to be a judgment call," he added.
Complicating the asset-allocation decision are potentially large future unknowns, said Kaplan, referring to market-shaking "tail-risk" events such as the global financial crisis of 2008. "The biggest risk is that something will happen that we haven't thought of."
Over the long run, said Kaplan, you can expect the growth in the equity market to mirror the growth in the economy. "So if you can come up with a reasonable number for how much the economy is going to grow, that could be kind of an anchor point for determining your equity expected return."
The types of risks that matter most will depend on the investor's objectives, risk tolerance, time horizon and age. For instance, Wilson noted that the sequence of returns -- in what order the investor is making or losing money -- is more important to retirees because of their shorter time horizon.
As for what risks investors generally care about, Wilson added: "A lot of times for retail investors, it comes down to drawdown." Or, as Gates put it: "The risk that investors care about is losses, so always keep that in mind."
Asset allocation must meet the needs of the client's liabilities that the portfolio is designed to fund, panelists noted. For example, an individual retiree will have future spending needs. "That's why we save," said Wilson. "We invest so we'll have more money down the road."
What constitutes a risk-free asset will depend on what future liabilities the asset allocation has been designed to address. "How one defines risk can be very context-sensitive," said Kaplan. For instance, if a client's liabilities are very much like the returns of a long-term bond, that bond might be the client's risk-free asset.
