Saturday, January 28, 2012

Portfolio Allocation and Performance III

I thought to leave this graph for last. My performance over the last 12 months! We all freshly remember this summer’s tribulations focussing one day on U.S.  debt load and political paralysis, the next on Greece and the European debt crisis and before the day was over we fretted about a collapse in the growth of emerging markets, specifically a hard landing in China. While pundits announced sagely that the ‘end is near’ yet again, stock markets experienced a significant correction if not a mild bear market. In September, I decided that the risk of actually ending up in a severe bear market was too great and I removed the riskier assets, in particular oil & gas stocks from my portfolio. My cash level, already 15% jumped to 30%.

 I knew, there was a new risk. Not one of losing a fair bit of net worth (temporarily) in case of a market collapse but the opposite: losing in upside because nearly 30% of my portfolio would not take part in a possible market rally. Hence the portfolio’s under performance.

As of today, I am still struggling with my real emotional motivation about this move. Was it, what behavioral finance experts call “recency”, i.e. the effect of a recent occurrence in the investment environment (the 2008-2009 crash) that influenced my thinking more than the long term investment history justifies or… was it a truly calculated move because of my changing investment requirements (I am getting older and maybe I should consider a 'lower risk' profile).
So over the last decade or so, overall performance of my portfolio was tremendous. But there were some major asset allocation shifts and changes in specific market sector performance that affected this overall performance. Investment performance, as James O’Shaughnessy points out, is to be compared to benchmark performance. Both O’Shaughnessy and William Bernstein point out that performance of an investment class or type always reverts back to its long term median annual return (the benchmark). From 1998 until 2008, the oil & gas sector outperformed its benchmark by a large margin; after 2008  its performance has coincided with the overall market and lately it has underperformed. The sector is in the process to revert to its benchmark. As far as I can see, with the economy placing a cap on oil prices and with gas prices near a 10 year bottom, this process is likely to continue for some time. But, I would not bet the barn on that.

The risk of a market crash and double dip recessions are on the decline and we seem to be in a grinding slow moving bull market that is climbing an enormous wall of worry. Overall, stock markets have underperformed their benchmark (7% plus inflation), and based on the principal of reverting to the benchmark, there is a good chance we’ll do better in the future. Yet with investor’s sentiment turning rapidly bullish (though not extremely so), the market could do the opposite.
The other principal we have learned is about defining risk. The main risk about investing in the stock market lies in its volatility and the time it takes to recover from any temporary decreases in portfolio value. This is often expressed in terms of the standard deviation of the investment type from its benchmark. I hope to discuss this topic in more detail in future posts but for now, one could describe it as follows. O’Shaughnessy calls its: ‘Minimum expected (annual) return’ and it basically describes the average fall (bear market) of an investment type based on historical performance data.

Take the next example: a portfolio of low P/E stocks has an average annual return of 18.23% and a standard deviation of 18.45%. Then the value decline one should expect in a typical bear market equals the benchmark rate minus twice the standard deviation. In our example that would be 18.23 – 2x18.45 = -18.68%.  However, the worst historical performance measured between 1927 and 2011 in a single year was a fall of 52.6% (call it a ‘black swan’). That occurred in 2008-2009 (not surprisingly). The best performance for this Low P/E category was + 81.42% for the 1927-2011 period.

The risk of losing 52.6% in one year is the motivator for a diversified portfolio. Diversification is achieved by investing in different asset classes or investment types. Most of the research on this has been focussed on stocks and bonds. Very little is known about the risk profiles of real estate, but just looking South should remind you that risk is everywhere. 
My conclusions: an investor should aim to have a well-diversified investment portfolio and as shown earlier, shifts in the allocation of the various investment types (which should have widely different performance patterns – or have a ‘poor correlation‘) is key to surviving (but not necessarily avoiding) significant market down turns. The latter are usually highly emotional and irrational events. After a time of above average investment type performance a reversion to the benchmark should be anticipated, i.e. a period of underperformance and the asset allocation should be based on that expectation. At the same time, dramatic asset allocation changes, such as going to high cash levels, may result in severe underperformance. This may be justifiable in a period after experiencing excessive profits, but overall it is best to stick to your long term allocation targets with minor adjustments.

No comments:

Post a Comment