Diversification with LATAM?

(Versión en Español)

One of the most important aspects of the modern theory of portfolios, work of the famous Harry Markowitz, is diversification. According to that theory, thanks to the diversification, you can reduce the risk (standard deviation, volatility, etc.) of a portfolio. The magic lies within the mathematics which tell us there is a positive relationship between the correlations of the assets and the volatility of those assets. (The correlation tells us the strength and direction of changes between two variables SEE MORE INFO).

For those of you interested, here is the math:

 --> This is the corr. coeficient.


In the development of the financial markets worldwide, I suppose, portfolio managers have searched and founded new markets to reduce the risk of their portfolios. Some of these markets could have been the Latin America markets. 

Let’s see what our friend, the portfolio manager, could have found constructing a correlation matrix between several indexes (I decided to use the S&P 500 and some others from LATAM) in 2006. Let’s suppose the manager decided to use data from 2000 to 2006: 

This matrix might look complicated to the non-expert eyes but trust me, it’s easy . The correlation between two assets is the value where the names of the assets cross (for example, the correlation between Argentina´s index and the S&P 500 is 0.254, the correlation between the Mexican and Brazilian stock markets is 0.482, etc.).

Some fund manager in the United States would have been interested in the Brazilian or Chilean markets because of their low correlation with the S&P 500 (assuming our friend, the manager, was invested in the S&P 500).

OPTIONAL: In fact during that period the daily average return of the S&P 500 was -0.003% with a standard deviation of 1.19%, meanwhile the average daily return of the Bovespa (Brazilian Index) was 0.06% with a standard deviation of 1.87% (according to Yahoo! Finance and my calculations). If the fund manager had invested just 47.02% on the S&P 500 and the rest on the Bovespa the average daily return would have been 0.03% with a volatility of 1.19%. Et voilà, our friend has achieved a greater return with the same risk!    

But, surprise! What has happened during the last years? If our friend decides to use the data of recent years, let’s say from 2006 to now, he would have some different numbers:

For the convenience of illustrating the difference I decided to put a “+” where the correlation has increased and a “-“ where the correlation has decreased: 

Our friend diversifying the risk using Latin America markets would be sad L. Remember, if the correlation increases the risk will increase.

One possible explanation for this increasing correlation is the integration of financial markets all over the world. In today´s financial markets, negative news from China have an impact on the S&P 500, as well as on the Brazilian stock market regardless of the commercial relationships between the countries.  Someone with just $100 USD can buy and sell stocks. The housing sector of the USA can have an impact on the share of an airport group in Mexico. If this tendency continues it will be harder and harder to diversify our assets :( . 

The other explanation is that the statistical differences between the correlations are not significant, the tendency changes… or that I don’t know how to use Excel XD