Portfolio Diversification Getting Harder

Portfolio Diversification Getting Harder

If there is one thing we can all agree on, investing is more accessible than ever in our history. More countries have set up their stock exchanges, and their complete transformation to digitisation allows even foreign investors to easily buy and sell stocks from anywhere worldwide. With ever-greater access to world equity markets, one would think that the equities portfolio diversification should be more achievable than before. However, according to a recent study published just a couple of months ago, it's getting more difficult to achieve such a diversification.

If you want to spread your risk by diversifying across many different stocks, you need to analyse their correlation levels. For example, if two stocks are identical in terms of their price movements, they will have a correlation coefficient of 1. This means these two stocks are perfectly correlated and their price movements are identical. Even if you diversify across these two stocks, it will result in the same performance as buying just one of the stocks.

Analysed Global Indices

The study used the following global stock market indices going back over  the past several decades:

  • S&P 500 (USA)
  • FTSE 250 (UK)
  • DAX (Germany)
  • CAC 40 (France)
  • Nikkei (Japan)
  • HSI (Hong Kong)
  • SSE (China)
  • TSX (Canada)
  • BVP (Brazil)
  • RTS (Russia)
  • KOSPI (South Korea)
  • SNX (India)
  • AOR (Australia)
  • IPC (Mexico)

You can calculate the correlation coefficient for each pair in all permutations, creating a matrix of correlations.

Average Correlation Coefficient

Example - Correlations among Global Stock Indices: 2010s

By taking the average of all correlation coefficients in the matrix, we can arrive at a single number representing the Average Correlation Coefficient. This is a real number between -1 and 1 where:

  • -1 means the perfect negative correlation
  • 0 means not correlated at all
  • 1 means the perfect positive correlation

So basically, what you want is a number as close as possible to 0, which indicates that the indices are not correlated at all. In other words, if you spread your investment by purchasing all of the global indices, that would be a true diversification.

Average Correlation Coefficient Results

  • 1980s: 0.25
  • 1990s: 0.30
  • 2000s: 0.59
  • 2020s: 0.7

Conclusion

In the past 40 years, we have had nothing but an ever-increasing correlation between the global stock markets. The Average Correlation Coefficient of 0.7 is a high number, and at this point, you can pretty much say that diversification is not achievable across these global indices. This is due to the ever-increasing globalisation and economic interdependencies between the countries. If you are looking for diversifications in your portfolio, you will have to find some other strategies to diversify. Here are some examples:

  • Find investable industries that are not correlated
  • Find asset classes that are not correlated (bonds, commodities, etc)