Many investors believe that the return series of investing in emerging market is normally distributed, that is, the outliers are rare and insignificant to the long run performance. However, given the feature that emerging equity market is more volatile compared to developed markets, the outliers may appear frequently during long term investments. Using daily returns from emerging markets, Estrada (2009) quantifying the impact of outliers on the long-term performance of emerging equity markets and the results suggested that black swans do have a huge impact on long run performance. As for the concept of black swans, according to Tableb (2007), a black swan has three characteristics: 1) It is an outlier which lies outside the expected range of returns; 2) it has a massive impact; 3) it can be explainable after the fact and thus may be predictable.
The market risk is assumed to be divided into two kinds: the systematic risk and the unsystematic risk (BEJA, 1972). According to Gourieroux and Monfort (2013), for a portfolio, although the systematic risk (market risk) exists, the unsystematic risk can be partly diversified, which is called the advantage of diversification. It is assumed that regional indexes are similar to portfolios which contain several market indexes. Since there exist outliers which could have huge impact, can investors decrease the risk by investing in the regional index? Are the number of black swans shown in the regional indexes smaller than those appeared in the emerging market indexes? In order to find the answer, we selected one regional index: The MSCI EM EMEA1, and analyzed the daily returns of the emerging markets which are included in this regional index.
This article follows the methodology used by Estrada (2009) that quantifying the impact of black swans on the long-run performance of emerging markets using updated data for 10 emerging equity markets till 2017. It further extends that article by considering impact of diversification on long-term performance by comparing individual market index and the selected regional index.
The remainder of this article is organized as follows. Section 2 analyses the result and evidence from 10 emerging markets and 1 region. Section 3&4 discussed assessments and suggestions on investing in emerging markets. Section 5 suggests limitations, while the last part is the conclusion.
Table 12 displays the selected 10 emerging markets and 1 region, EM EMEA, the selected market index, the number of years and days of each market index as well as the regional index. This table also shows the start date of the index we selected,and all the data collected ends at 15/12/2017. P10, P20 and P100 are the proportion of 10, 20 and 100 days relative to the total number of days in each market. All the indexes are in local currency and explain capital gains but not dividends. It can be seen from the table that the full sample includes 238 years and 61,268 trading days for 10 emerging markets and 1 region. Moreover, it should be noted that 10, 20, and 100 days only takes a very small proportion of the whole sample period.
Table 2 exhibits the summarized statistics for the distribution of daily returns for the indexes within the selected periods. The table includes minimum (MAX) and maximum return (MIN), arithmetic mean (AM), standard deviation (SD),coefficients of skewness (Skw) and kurtosis (Krt), and coefficients of standardized skewness (SSkw) and standardized kurtosis (SKrt). We can see from the table that all the 10 markets and 1 integrated region have very large variations in daily returns.
For example, the highest MAX was in Russia at 27.4% with an average maximum return of 0.06% across 10 markets. In the meantime, the lowest MIN was also in Russia (-26.66%) with an average minimum return of -16.1%. All markets except Greece and Hungary have a significant degree of skewness and all the indexes have a significant degree of kurtosis3. Therefore, the daily returns of most of the 10 markets and MSCI EM EMEA are not normally distributed. Moreover, the SD in MSCI EM EMEA is smaller than the average SD across 10 markets (1.51% vs.1.88%), which indicates smaller variation in regional index than in a single market.
