I had the pleasure of joining a presentation from Mr Inmoo Lee (Ph.D.) a Vice President from Dimensional Fund Advisors in the United States. Inmoo along with two other colleagues from Dimensional has conducted research into the relationship between risk premiums and business cycles looking at whether there are systematic patterns and if so whether investors can effectively access better returns utilising these patterns. (The following is my own summary of the presentation and not that of the researchers.)
Before getting to the results let's first step back and define a couple of key concepts. For those who have followed our website and our underlying investment philosophy you will be aware that we subscribe to the academic research behind the three factor model as identified by Fama & French. Fama & French found that there are areas of investment markets where risk premiums exist over the long term. In particular, small and value companies, taken as a group, hold higher levels of risk for investors and therefore investors expect compensation for taking on that risk. i.e. higher risk leads to higher return in the long run. (Take a look at our Investment Philosophy and Our Research Based Approach pages for more details)
However, these risk premiums are not present all the time and over short run periods, small and or value companies may under-perform the index. Currently the Global Value fund that we use in portfolios has underperformed the Large Company fund year to date but has out-performed over 5 years.
Therefore, based on these presumptions which are supported by academic research, the question to ask is can you time your entry and exposure to the risk premiums so that you make the most of periods when the risk premiums are being realised and minimise the exposure when the risk premiums are not being realised and the index (and growth stocks) are performing better.
What Inmoo and his colleagues have done is to look at business cycles as a source of prediction. Let's stop here and break off for a brief economics lesson first. The business cycle basically tracks the periods of expansion (growth) and contraction (recession) in an economy. Throughout history economies move through this cycle moving from periods of expansion, reaching a peak and then contracting, reaching a trough from whence the economy starts to expand again.
Intuitively we should expect that the risk premiums (expected future returns) are greatest at the bottom of contractionary periods (troughs) and worst at the top of expansionary periods (peaks). The theory being that at the bottom of the market cycles, riskier investments such as small companies and out of favour companies (value) will be sold off the furthest. Therefore the expected future return is the greatest as these investments are at relatively low prices.
So theoretically, the best time to be buying into risk premiums is at the bottom of the business cycle and selling out at the top of the business cycle. Inmoo's research actually found that some relationship did indeed exist with small and value areas of the US market out-performing the market going forward from the bottom of the business cycle and under-performing after the cycle reached the peak.
Everyone should be getting excited now as there seems to be an opportunity to trade in and out of the premiums and thus achieve greater returns. Unfortunately there is a big BUT. A major problem is that it is very difficult to identify exactly when the business cycle reaches its peak and trough. The National Bureau of Economic Research in the US has the task of doing just that. They identified the last peak as being reached in March 2001. Unfortunately they were only able to make this determination in November of that same year. i.e. 8 months later. Similarly they identified that the last trough in the business cycle was November 2001 and this was determined in July 2003, some 20 months later.
If an organisation which has this task as its primary focus takes so long after the fact to make a determination, what hope to make the determination beforehand. This is the key problem.
Now some might say well what does it matter if you miss the peak or trough by a month or two you should still end up with a better outcome. Inmoo and his colleagues took this very consideration into account. They found that if you missed your timing either coming in or going out the benefit of the timing decision was statistically non-existent. Basically the conclusion was that you had to be lucky twice with your timing decisions. You would have to pick the precise month going in to the risk premium and then precisely time the month when to get out of that area of the market.
Yet again it seems the probability of greater performance is stacked against the "market timers". A much better approach is to determine your ideal portfolio exposure to risk factors i.e. the market, small companies and value companies, and hold this exposure through time.
If you would like more information about our approach to structuring investment portfolios please take a look at our Building Portfolios page.
Regards,
Scott Keefer