As I reviewed semi-annual performance reports for clients, I found myself thinking about this question. Our reports tell our clients the return on their portfolio for the year to date, the past three and five years. These are absolute returns, after all costs, which can be used to relate to individual needs and risks. If your financial plan shows a return of say 5% is required each year and your return is more, then things are looking positive, and a cushion is being built. If the return is less, then some changes may be required – save more, spend less, work longer.
Of course, we know that it is impossible to generate a consistent annual return when investing in a diversified portfolio containing a number of asset classes. When the markets declined in 2008, we recalculated our long term projections for clients, using the lower asset values, to determine what actions may be required. We have since compared actual to projected results at the end of each year, thus keeping the planning current and meaningful.
Absolute return should be the most relevant to a retail investor. However, our world is full of news and talk of relative returns. How did my portfolio perform compared to my neighbour’s? How did that fund manager perform relative to all funds in the class? How did that pension fund perform?
In order to look at relative returns, a “benchmark” must be established. And here is the first step in the imperfect world of relative returns. Many benchmarks can be supported as valid. For example, comparing the performance of a Canadian equity fund to the S&P/TSX Canadian Composite index is a commonly used benchmark. Or comparing to the peer group – i.e. all Canadian equity funds – is another. But there are many factors and strategies that can be built into the Canadian equity component of a portfolio (or any asset class, for that matter) – holdings in small companies versus large, value versus growth, concentration on a certain sector, to name a few. Should each of these be built into a benchmark?
Index suppliers such as Standard & Poor’s, Dow Jones, Russell, MSCI, FTSE, DEX and others build and maintain indices based on many criteria, often including the strategies referred to above. But if the “value added” benefit of a particular approach or manager relates to the use of a particular strategy (or dare I say stock picking ability!) should that be reflected in the benchmark? Or should the comparison be made to the broad market?
Similarly, should a benchmark be a weighted average of the components that are targeted to be in a portfolio – i.e. 40% fixed income, 20% Canadian Equity, 20% US Equity and 20% International equity – even if the manager chooses not to be in US equity for some period of time?
And don’t forget that it’s impossible to replicate an index without incurring costs. So those benchmarks come without any costs, making it difficult to compare “apples to apples”.
Performance reports from PWL report absolute returns for clients. (Securities administrators are contemplating making this a requirement for all securities and mutual fund dealers.) When we meet to review your portfolio and the performance report, we also review benchmark returns which compare your return to the broad market indices of each component of your portfolio.