The Case for Diversification

For readers who wish to learn more about Modern Portfolio Theory, there are numerous free resources that cover the topic in exhaustive detail, such as from the Wikipedia. I will spare you the fine details here. In short, the Modern Portfolio Theory tells us that diversification can help improve the risk-adjusted return of a portfolio. It also shows that the lower the level of correlation between the assets in a portfolio, the lower the portfolio’s risk will be.

Correlation
Correlation indicates the strength and direction of a linear relationship between two random variables. The value is always between -1 and 1, where 1 indicates perfect positive dependency and -1 indicates negative dependency; 0 indicates no relationship exists and the two random variables are said to be independent.

Here is a correlation matrix showing the relationships between some major equity, fixed income, and commodity indices:

Here are a few observations based on the correlation data from the past 3 years:

·         The Straits Times Index (STI) has had a correlation of 0.99 with the MSCI Singapore Index. Hence they move very similarly to one another and investors might not reap any benefits from owning both in their portfolio.

·         The STI has had a high correlation of around 0.7-0.8 with the offshore Chinese equity indices (MSCI China, Hang Seng China Enterprises Index) but a low correlation with the domestic Chinese equity indices (CSI 300 Index, FTSE China A50 Index) at around 0.2-0.3. This again highlights one of our regular reminders to investors: “don’t take index names at face value” (also see our “Picking the Right Chinese Dishes from the ETF Menu”)

·         The STI has a moderate level of correlation with the other Asian countries’ equity indices, such as India, Malaysia and Thailand, all at around 0.6. In turn, these countries’ equity indices have also had a moderate degree of correlation with each other. This implies that mixing these single country Asian equity indices with the Singaporean equity indices can help reduce portfolio risk. The relationship also holds for US stocks. The S&P 500 Index has had a correlation of 0.5 with the STI over the past 3 years.

·         STI has had a near-zero correlation to gold prices (-0.02) over the trailing three year period.

·         Meanwhile, the HSBC Asian USD Bond Index exhibited a negative correlation with the STI (-0.5) over this same span.

Building a Portfolio with ETFs
Using the data from the correlation matrix above, we can mix and match some ETFs which track these indices specifically or related indices to form a diversified portfolio.

Here, suppose 3 years ago I was an investor in Singapore who wished to:

·         Maintain most of my exposure to Singapore securities – my home market which I am most familiar with and have the most outlook on

·         Diversify my portfolio

Here we built 3 portfolios (Portfolio A, B and C), with increasing magnitude of diversification (Portfolio A is the least diversified, while Portfolio C is the most diversified):
 

Looking at the performance of these portfolios over the past 3 years, our example shows that Portfolio C performed the best, driven by the bull run in gold and fixed income in the past few years. In addition, portfolio risk was reduced as we included investments with low correlations, which is as stated earlier, central to Modern Portfolio Theory.

This is Just a Start

This is a simplified illustration of how diversification works for your portfolio. We have a few more tips when it comes to building a portfolio with ETFs:

·         Rule of thumb: (1) form your investment thesis; (2) choose your ETF; (3) execute

·         Past performance does not guarantee future results. Do your homework

·         Read Morningstar’s ETF research and articles; read economic and market research to form a well-rounded view.

Jackie Choy, CFA, is an ETF Strategist with Morningstar