The foreign capital flows and the behavior of stock prices at BM&FBovespa.
| Autor | Sanvicente, Antonio Zoratto |
| Cargo | Report |
Introduction
The knowledge of the relationship established between foreign capital flows and listed stock returns is unquestionably considered important for individual and institutional investors. Some articles recently published in the Brazilian financial press may be mentioned as a concrete evidence of the relevance and of the concern in that relationship. Anaya (2010), in his paper, conjectures prospects for stock price increases in the local market and also emphasizes the positions taken by a few market professionals. Among them, the author quotes: "For Emy Shavo, a variable income security strategist with J. P. Morgan, the prospects are for the return of a foreign capital inflow to the BM&FBovespa, a determining factor in a more significant appreciation of assets in general" (Anaya, 2010, p. 16).
Subsequently, Valenti, Torres and Bellotto (2012), in a discussion of Brazilian stock market prospects, state that "the foreign investor is still the main vector of trends at the local exchange" (Valenti, Torres, & Bellotto, 2012, p. 12). Thus, a more detailed explanation is offered: "When he [the foreign investor] enters the market, Ibovespa rises, ... and when he leaves, the market is left in trouble".
The same authors display a chart that would confirm the strong positive correlation between the level of Ibovespa and the net cumulative flow (the cumulative difference between values of purchases and sales) of foreign capital at BM&FBovespa.
The foreign capital flow reached its minimum worth in December 2008, i.e.,-- R$13,397 million, after having reached R$11,991 million, in May 2008(1), a period in which Ibovespa dropped almost 50% of its points (from 72,592 to 37,550). The chart also shows that, starting in December 2008, the sign of the net cumulative flow became positive, while the level of Ibovespa rose to 69,304. And, to stress this point of view once more,
... when the large international investors buy or sell stocks in Brazil, the effects are shared by everyone [since they affect market prices], for better or worse [up or down]. And local asset managers are left with the following understanding: 'against the foreign capital flow there is no argument' (Valenti et al., 2012, p. 16).
In the discussion of the interaction between foreign capital flows and local market returns, which occur in the local financial press, (as illustrated above), it is clear that the main concern is with the short-run interaction. It is also noticeable that there is a widespread belief that capital flows cause changes in market prices; and this belief is consistent with what is known in the theoretical literature, that is to say: foreign investors would tend to be more sophisticated and informed than other investors. However, as pointed out in the review of the theories, the relationship may also be recognized from price changes to capital flows, depending on the factor described as the trend chasing explanation.
Seen in these terms, the main objective of this paper is to concentrate on the nature, i.e., the direction, and on the significance of the short-run interaction, with the use of appropriate analytical tools, as well as with the investigation of daily data. The period analyzed in this research is fixed between Jan. 03, 2005 and July 31, 2012. Special attention is given to the possibility of a significant change in interaction as a result of the global financial crisis, which began in the second half of 2008. Candidate control variables are also used, since both, capital flows and market prices, which are proxied by Ibovespa, may be determined by and/or associated with various macroeconomic and market factors, and such as changes in exchange, interest rates, returns on international stock markets and stock market volatility, etc.
The specific analytical tools chosen for this research take into account two major explanations for the existence of significant interaction. On one hand, there is the argument, as expressed in the local press material references, which were mentioned above, that capital flows cause returns. This is called the information contribution argument. On the other hand, it is also possible to consider that flows follow returns; therefore, foreign investment could be recognized as trend chaser. And, there is still the possibility that mutual feedback arises between returns with flows.
This paper is organized according to the following structure: second section discusses the relevant international and national theoretical literature. Third section explains the methodological approach to the analysis of the data, which, in turn, are described in fourth section, in terms of their definitions and sources. Fifth section presents and discusses the results; and, final section concludes and suggests possibilities for additional studies.
Review of Literature
The problem examined in this present paper is part of what is called the price- volume relationship literature.
In his often cited paper, Hasbrouck (1991) suggests that the interaction of trades and quote revisions could be modeled as a vector autoregressive (VAR) system, because his tests using market specialist-based data from the New York Stock Exchange (NYSE) indicated that a trade's price impact would arrive with a lag, the price impact being a positive but concave function of the trade size. However, he was discussing bid-ask spread revisions, and not the effect of trading volume on actual trade prices. Since he found that large trades tended to cause the widening of spreads, some authors attribute the idea of a positive relationship between contemporaneous trading volume and price changes to Hasbrouck (1991), such as, Sarwar (2005).
Using multiple linear regression analysis, Long (2007) presents evidences for the options market that duplicate existing statistics for a price-volume relationship in equity markets. Using 1983-1985 daily data for the Chicago Board Options Exchange (CBOE) market, he shows that a positive relationship exists between the absolute value of call price changes and the options trading volume; but that, as in the case of results for equity markets, demonstrates that positive call price changes are normally associated with significantly higher volume levels; otherwise, they are not connected to negative price changes. The discussions concentrate on the role of new information, as well as on the information arrival process. It is posited that the general result can be explained by a process of adjustment towards a new price equilibrium, after the arrival of new pieces of information that cause a change in the demand for security: when new, positive (negative) types of information cause an increase (decrease) in demand for a particular security, there is price increase (decrease) results; with this, the attendant volume of trading which was required tends to create a new equilibrium for prices.
McGowan and Muhammad (2011) examine the relationship between volume and price changes for spot and futures in Kuala Lumpur stock market index, using data since the inception of Malaysian futures market, in December 1995, until December 2003. They use cointegration analysis and find that trading volume and price changes present a significant long run relationship. They point out that most of the earlier empirical work focus on the contemporaneous relationship between trading volume and stock price changes, while more recent studies deal with the dynamic relationship, that is, the possible causality between price changes and trading volume.
Moosa and Al-Loughani (1995) highlight two so-called empirical regularities which are directly relevant to the objective of the present paper: (a) it takes volume to make prices move, and (b) volume is relatively heavy in bull markets and light in bear markets. Firstly, as pointed out already, there is ample evidence in the literature for a positive contemporaneous correlation between trading volume and price changes. Secondly, there is, for example, the evidence provided by Long (2007). Moosa and Al-Loughani (1995) test the existence of such a relationship using data from Malaysia, Philippines, Singapore, and Thailand; and they find contemporaneous and lagged association between trading volume and price changes, as well as between price changes and trading volume, i.e., the feedback story.
Llorente, Michaely, Saar and Wang (2002) develop a model for the dynamic trading volume-return relation, with focus on individual stocks traded in US markets, instead of aggregates represented by stock market index. Their model emphasizes the influence of information asymmetry, which is proxied by bid-ask spread, company market capitalization and analyst coverage. They argue and test, with the expected results, that stocks which present low asymmetry (low bid-ask spreads, high market capitalization and extensive analyst coverage) will be the stocks that will be used in allocation or hedging decisions. In this case, large trading volume is not an informative signal, and, since volume does not lead to expected future stock payoff revisions, negatively autocorrelated returns will result. The opposite will happen with large trading volume signals for high informationally asymmetric stocks. In this case, large trading volume will be followed by positively autocorrelated returns, because the relevant body of information would be only partially incorporated into prices. This leads Llorente et al. (2002) to test the impact of trading volume on the return autocorrelations.
Sarwar (2005) examines the role of the trading volume of S&P 500 options in the prediction of future volatility of the underlying index. Although that study covers the possible association between trading in one market (options) and prices in another (the underlying asset), it deals with the dynamic relationship between trading volume and price, that is, the eventual leads and lags in that relationship. It uses daily data from Jan. 04, 1996 to...
Para continuar a ler
PEÇA SUA AVALIAÇÃOCOPYRIGHT GALE, Cengage Learning. All rights reserved.
