We combine Nobel Prize winning research with smart algorithms and expert analysis to find the optimal investment portfolio for you.
Diversification is key
The first step in our investment management process is selecting a set of asset classes that can be used as building blocks of our client portfolios. When doing this, we look for asset classes that:
1. Provide diversification benefits
2. Have at least 10-years of investment track record
3. Are covered by at least one low-cost, high-quality ETF
The following table shows the asset classes we use in our portfolio and the benefits that each provide.
|1||Canadian Stocks||Growth, long-term inflation protection, diversification|
|2||US Stocks||Growth, long-term inflation protection, diversification|
|3||International Stocks||Growth, long-term inflation protection, diversification|
|4||Emerging Stock Markets||Growth, long-term inflation protection, diversification|
|5||Canadian Bonds||Income, diversification|
|6||Emerging Market Bonds||Income, diversification|
|7||REITs||Inflation Protection, diversification, income|
We do not include commodities or natural resources in our asset class set. The reason is that the Canadian economy, and the Canadian stock market are highly correlated with commodity prices. As such, commodities as an asset class do not offer the same diversification benefits to a Canadian investor as a US or European investor.
The optimal asset allocation for each client
The proportions of each asset class in a client portfolio is a function of the risk-level for that portfolio. In order to determine the appropriate asset class proportions for each risk level, we follow a variation of a methodology developed by Nobel Prize winning economist Harry Markowitz called Mean-Variance Optimization (MVO).
MVO is a mathematical framework that is used by many institutional and sophisticated investors. By using MVO, we can find the optimal combination of assets that maximize returns for each level of risk. The inputs to MVO are expected returns for each asset class, standard deviation of returns, and correlation of returns of asset classes with each other
The standard deviation and correlation of asset class returns are calculated from the monthly return data for the last 10 years.
We have built internal models to calculate expected returns for each of the asset classes we include in our portfolios. For stocks, our models look at current dividend yields, growth in earnings per share, and long-term price to earnings ratios. For bonds, we look at factors such as current yields and interest rates; long-term interest rate forecasts, and the bond portfolio’s duration and convexity. For REITs, our models look at dividend yields, growth in dividends per share, and the yield spread between real return bonds and REITs.
After calculating the expected returns for each asset class we adjust them for taxes and fees. Our optimization has been run for both taxable and tax-sheltered accounts such as RRSPs and TFSAs. We have found that the tax status of the account makes a small difference in the recommended asset allocation. (The tax status of the account has a more important impact on the recommended funds for the account due to issues relating to withholding taxes.)
In order to account for the uncertainty in the expected returns of our asset classes, and to reduce the optimization’s sensitivity to input parameters, we simulate thousands of possible return scenarios within the constraints of asset class expected returns, standard deviation, and correlations.
For each scenario, we calculate the optimal portfolio. We then use an averaging technique to combine the results of each simulation to arrive at our recommended asset allocation. This process is often called Resampled Mean Variance Optimization in the financial industry.
The model portfolios that we arrive at using this method are less sensitive to changes in input assumptions, and are expected to perform well under different market environments.
Putting it all together
The following charts show the recommended asset allocation for taxable and tax-sheltered accounts at each risk level.