Efficiency of sector diversification in the Brazilian stock market in times of financial crisis.

Author:Roquete, Raphael M.

    In the global financial system, it is natural that the appearance of crises causes great turmoil in the stock market. For example, examining the evolution of a particular index, such as the Bovespa, it is easy to identify clusters of volatility during periods of crisis. In these periods of high volatility, diversification emerges as an alternative to spread portfolio risk, since the various sectors of the economy tend to have different reactions to these shocks: some depend more on domestic demand, while others are more sensitive to international events. Markowitz (1952) was responsible for alerting investors to take into consideration, besides returns, the risk of the assets themselves (represented by standard deviation).

    As such, a study on the performance of sector diversification is particularly relevant, in the first place due to the increased importance of diversification based on correlations between assets as a tool in models of portfolio resource allocation and risk management; the second reason concerns the substantial growth in trading volume on the Bovespa in recent years, driven by the economic stability provided by the Real Plan. Investors should be able to create a portfolio effectively, minimizing the tendency of novices to enter into a state of despair with every downturn of the index and unload their positions without the slightest foundation. Thirdly, the influence of a crisis in a given nation has had an increasing effect on the economy of many countries, resulting in a significant correlation between the stock exchanges around the world.

    Thus, the research sought to investigate whether sector diversification can to a great extent reduce investor risk in times of financial crises. When the results pointed to an inefficiency in diversification, it was sought to investigate the possible causes of these results. We chose the Russian Crisis of 1998 and the Financial Crisis of 2008 because they have different backgrounds and especially because Brazil was in a completely different economic situation in these two periods. Additionally, both crises were identified as clusters of volatility in the evolution of the daily return of the Bovespa index.

    The study was restricted to shares in the Bovespa, to avoid distortions in the results due to liquidity problems arising from the shares traded on other Sao Paulo Stock Exchange indexes. The efficiency of diversification will be analyzed via a sector study and just from the point of view of a manager who keeps only domestic market shares in their portfolio. The innovation in the methodology used in this study is noteworthy given that it has not been used in any previous survey.

    The article first reviews the literature about sector diversification strategy and its advantages. It then looks at the causes and results of two moments of instability in the world financial market, Russian Crisis and the Crisis of 2008. The next section presents the methodology that evaluates the efficiency of diversification in the Brazilian stock market during financial crisis. The following section discusses the results of the empirical study. The article finishes with the conclusions, implications and some ideas for future research in this important area.


    2.1 Sector Diversification

    The 1950s was marked by the emergence of a new theory about the formation of investment portfolios, a key work being Markowitz' 1952 paper Portfolio Selection, in 1952 (Saito, Savoia and Fama, 2006). According to Markowitz (1952), the idea of building a portfolio based on returns alone would be mistaken. For the author, the rational investor should compose their portfolio considering only two measures: the expected return (average of past returns in a given period), which would be desirable, and risk, defined as undesirable.

    In the same article, the author points to diversification as the ideal way to minimize portfolio volatility, noting, however, that it is not sufficient to rely an diversification without an analysis of the correlation between the assets. In other words, it is use building a portfolio with a range of assets if they are all subject to the same risk. The risk of the portfolio would depend, then, on the proportion of each asset in the portfolio, the respective variance of each asset, and the covariance among all assets (Statman, 1987). Through total or partial elimination of unique risk (specific risk of a company and/or a sector), diversification would be able to reduce the risk of the portfolio.

    Sector diversification can be an important and useful strategy for any investor, because they may allocate their investments in (i) companies in the same sector, which, by definition, have similar characteristics, and (ii) firms in other sectors, with different characteristics, which, in theory, can become a good proxy to eliminate unique risk (Losekann, Velasquez and Vieira, 2009). Moreover, because the Bovespa, for example, is an index highly weighted in the commodities sector, when this sector does not fare well the index tends to fall in lockstep (Teves, 2009). Given this special feature of the Brazilian index, its behavior may not reflect the performance of other global indexes. For example, in the event of a world crisis, when the correlations between the world stock exchanges were increasing, there would not necessarily be a loss of sector diversification in the domestic market.

    2.2 Financial Crisis

    According to Kindleberger (2000), financial crises are situations in which there is a large devaluation of assets or the failure of important institutions for the financial system. They may be associated with several causes, such as runs the banks, market bubbles, stock market collapse, speculative attacks on the currency of a particular country, or default on external debt. The emergence of financial crisis adds to uncertainty about the future and is characterized by poor visibility vis-a-vis the global economic outlook. This aspect increases the fragility of the decisions of economic agents.

    In recent decades the process of globalization has intensified, thereby enabling the diffusion of information technology and greater integration of international markets which in turn facilitates the transfer of resources between countries. This increase in capital flows has resulted in a greater synchronization between the observed asset returns in emerging markets and those of developed countries. This fact can be demonstrated through the contagion of various crises originated in a particular country, such as, for example, the crises of Mexico, Russia, Brazil and Argentina. Objectively, investors could discern the effects through the loss of efficiency of international diversification, due to the increase in correlations among the assets of different countries (Fama and Pereira, 2003).

    It is interesting to observe the behavior of investors during these periods. Companies tend to unload their positions to offset the losses they have incurred in different markets. Speculators aware of this fact can leverage in order to profit, thereby amplifying a downward movement in the stock market. Material published by Valor Econdmico in 2008 argues that the behavior of foreign investors is crucial to explain the highs and lows in the Bovespa. At the time, Instituto Nacional de Investidores (INI) noted that 72% of the Bovespa index volume could be attributed to non-resident investors (Campos, 2008). Silva and Salles (2008) point out that in times of high uncertainty the so-called flight to quality occurs, which is defined as the flight of capital from risky assets, mainly in emerging countries, to low-risk assets such as U.S. Treasury bonds. Some authors believe that in times of crisis the markets also endure the herd effect: If one group of investors begins to sell their shares, other investors follow, resulting in a precipitous price drop. Aquiles Mosca, in an interview with the Conexao Dinheiro in 2010, touches upon this issue, reiterating that people will tend to follow a group. All try to avoid being the last to get out of the market when it begins to fall, causing a veritable stampede in times of crisis.

    Another important issue highlighted by Jacquier and Marcus (2001) and Pinto and Ramos (2004) is the fact that it is quite common to find portfolio management and risk models that are derived from the correlation matrix, based on historical data. According to the authors, in periods of turbulence these correlations tend to break down, making the manager's work that much more difficult. For Jacquier and Marcus (2001), systemic risk becomes much more present than unique risk in times of high volatility, which would cause an increase in correlations between assets. The authors also suggest a model to predict the correlation matrix using volatility as an independent variable; they suggest that most of the variation in the correlation matrix can be attributed to the change of volatility in time.

    2.2.1 The Russian Financial Crisis

    In 1998, Russia was contaminated by the East Asian Crisis, which raised the fears of thousands of investors in regard to emerging countries. To shore up its currency, Russia arranged for several IMF loans thereby increasing its debt. Following IMF "advice," the government allowed a free flow of capital to make the country more attractive to foreign investors and sustain the currency. The outcome of the resolution was not positive. Fears were such that the measure merely facilitated even further the withdrawal of money from Russia (Rabimov, 2004; Stiglitz, 2002).

    The Asian Crisis also contributed to the imbalance between oil supply and demand (a source of tax revenue for the Russian government), which culminated in a sharp drop in the price of the commodity in the first six months of 1998 (Brisac, 2008). Compounding the Russian situation at the time was an evident overvaluation of the ruble (Barbosa and Oliveira, 2006), mainly due to...

To continue reading