Investing using the three factor model
Three dimensions of risk profiling
Most advisers will give you a choice of risk profile, the names vary but generally you are pidgeonholed into classifications like “conservative”, “moderately conservative”, “balanced”, “growth” and “aggressive”. It varies, but most dealer groups have between three and six risk profiles, and once they’ve decided which risk profile suits you, you get a standardised portfolio to match.
We regard this as a drastic oversimplification. Filling in a little multi-choice questionnaire isn’t going to reveal to the adviser all that much about your risk needs and aversions, and there is no way that five different portfolios can really suit every single client. Many advisers appreciate this, but prefer a simplistic model because simple classifications make it easier to run a financial planning business, the adviser doesn’t need to do much work on a case by case basis and so really only needs to worry about a small number of standard client categories.
If you are not a standardised person, it is unlikely that you will be a perfect fit for any particular stereotype and therefore we reject the notion that we only need to calculate a few portfolios to suit all of my clients. There are actually three (not one) dimensions to a risk profile:
- How long before you need to spend this money?
- How much volatility can you tolerate along the way?
- Does it bother you if your portfolio underperforms major benchmarks from time to time?
The first dimension, time, is obviously not going to always be a single number. If you are saving up for a single goal then it will be, for example if you want to buy a house in three years you have three years to save up for it. But what if you have multiple goals? You want to buy a bigger house in 2006, buy a new car in 2007, go on a long and expensive holiday in 2010 and retire in 2020, then in that case the years you need money are 2006, 2007, 2010, 2020, 2021, 2022, 2023, 2024, 2025 etc. A pensioner doesn’t draw out one lump sum once and for all, a pensioner has short term goals (pension payments for this year, next year, and the year after that), medium term goals (year four, five, six and seven) and long term goals (all the pension payments after that, for the rest of your life).
We therefore reject the simplistic notion that a single portfolio design can suit anybody. A portfolio with risk designed for short term needs will fail in the long term, and a long term portfolio is too risky for the short term. Therefore, most clients belong simultaneously in several of the standard risk profiles for their various goals.
The second dimension, volatility, is addressed with the mixture of cash and bonds (defensive assets) to the mixture of shares and property (growth assets). This is the standard Modern Portfolio Theory approach, you decide how volatile a portfolio you are willing to tolerate, and design the highest performing portfolio you can with that level of volatility. This usually leads the investor to determining a ratio of defensive assets to growth assets, which is about as far as standard risk profiling and asset allocation techniques ever go.
But portfolio construction can and should go beyond just determining how much cash and bonds a portfolio should have, because there are many different growth asset classes as subsets of the obvious ones of domestic and international shares, and property trusts.
As explained in the article on market obtaining higher returns , “value” stocks tend to outperform growth stocks, and the evidence does point to there being at least some return premium from small company stocks, there is also higher return potential in emerging markets. Thus, for investors wanting the highest return it can be a good idea to tilt a portfolio away from a large company neutral index and increase exposure to smaller companies, value companies and some emerging markets.
The trouble with that is that most investors focus on the performance of the large company neutral indexes like the All Ordinaries index. A value and small stock portfolio ought to provide higher performance in the long term than the All Ordinaries, but it is precisely because it performs differently that some perceive another risk.
The risk in this case is not usually higher volatility, because a value and smaller company portfolio isn’t necessarily going to be significantly more volatile in an absolute sense than the widely followed indexes, but in the short term there will be plenty of times when the neutral index beats the value and small company indexes. An investor in this portfolio might wonder why they are bothering with all of these other semi-exotic asset classes when the plain old vanilla flavoured All Ordinaries index has done 5% better this year.
This is likely to happen quite frequently, it would be unrealistic to assume that any strategy, however sound, is always going to beat every other strategy over every short term time period. In fact, the best investors in the world, such as Warren Buffett appear to trail popular benchmarks around 40% of the time, it is in the other 60% of periods that these managers make the highest profits.
The risk in this case is called “tracking error”, and is a major concern for many fund managers that know how short term oriented many people are. If you are the type of person that likes to check your portfolio every week and see how it is performing against popular benchmarks, you will probably have a much lower tolerance for tracking error than someone that only checks on the portfolio once or twice a year. The latter type of person only sees long term results and would be happy if the strategy outperforms neutral benchmarks over time, but the former would have great reason to be worried 40% of the time.
Bear in mind that sometimes 40% of the time can be a long time. A strongly value tilted portfolio ought to produce superior results six years out of every ten, but remember this means you could be lagging behind the indexes for four years in every decade! Four years is a long time to stick with a strategy that doesn’t seem to be working, but that is precisely what you are going to have to do in order to get to the six years where you could reap the biggest rewards.
So we need to add a third dimension to risk profiling, tracking error tolerance. An aggressive long term investor might be more than happy to accept the volatility of the stock market, but won’t necessarily be too happy about following a strategy that lags behind the general market index for any length of time. On the other hand, another investor might be quite happy with the strategy after reading up on the works of Professors Fama and French, having noted how well the Fama/French value portfolios have performed over the long term this investor would tilt the portfolio heavily toward value and small companies, reducing their exposure to large cap growth and neutral indexes.
So some investors will need a fairly conventional portfolio, mostly exposed to large well known “market” benchmarks such as the Australian All Ordinaries Index, the MSCI ex Australia World Index, the ASX200 Property Securities index, and others. Other investors, happy with our long term plan will accept something entirely different. There is no difference between the mixture of defensive assets and growth assets for these investors, but the portfolios will nevertheless perform very differently.