Market correlation….is there any….
By Karl Siegling
Having recently travelled in the USA and Asia presenting to investors, I was surprised how often I was questioned on the correlation between the Australian market and the US market, as well as Asian markets, in particular China.
Over time I have come to the view that highly technical and complex discussions and analysis on correlation are fraught with danger. I have always found it curious that the first thing many Australian investors check is the overnight performance of the Dow Jones Index. When we check this index we are actually determining how the top 30 stocks in the USA performed overnight. A better measure would be how the S&P 500 has performed overnight – at least this is a basket of the top 500 stocks listed on the US Stock Exchange rather than a basket of 30 stocks.
Nonetheless, this analysis is still rather limiting. How should the Australian market perform in relation to the top 30 or top 500 stocks listed in the US? How related are the companies listed on the Australian Stock exchange to those listed in the US? What percentage of the S&P 500 consists of resources stocks, what percentage are financial services? What percentage of the US index consists of technology stocks? This analysis quickly creates more questions than it does answers.
The correlation problem prompted me to find out what the performance of the US market (S&P 500 index) has been over the last ten years, relative to the Australian Market (All Ordinaries Accumulation Index). The chart appears below:
Interestingly the US markets significantly out-performed the Australian market during the ‘technology boom' of the late 1990's and 2000 and significantly underperformed the Australian market from the middle of 2000 to around early 2003 during the ‘technology bust'. Since 2003 the Australian market has outperformed the US market. The result is that an investment in the Australian market has grown around 250% in 10 years. A similar investment in the US markets has grown by around 65% in ten years.
The correlation discussion becomes even more complex when you adjust for currency movements between the AUD$ and US$. During this ten year period the exchange rate between the AUD$ and US$ has fluctuated between a low of US$0.485 per Australian dollar and the current exchange rate of US$0.865 per Australian dollar. This represents an 82% appreciation in the Australian dollar relative to the US dollar and results in only a 40% increase in the S&P 500 in Australian dollar terms. Clearly currency fluctuations affected correlation between the markets.
The ten year history of the Australian market relative to the US market should serve as a warning against making any emphatic statements on the correlation of the two markets. The problem of correlation over long periods of time is at best problematic and at worst meaningless. The reason for outlining this history is to illustrate a similar potential error in examining the Australian market relative to Asian indices, and in particular the Chinese market. Below is a chart of the All Ordinaries Index relative to the Shanghai Stock Exchange Index:
What can we say about the correlation between the Australian and Chinese market over the past five years? Well, at a very simple level the Chinese market significantly underperformed the Australian market for most of this period and then significantly outperformed the Australian market since late 2005. More importantly, what does this tell us about the likely correlation between the Australian market and the Chinese market in the future? Given our previous observations on the US market, the answer is probably very little. As unsatisfactory as this answer appears, any other answer is fraught with danger.
Article reproduced with kind permission from Cadence Capital: www.cadencecapital.com.au