Foreign portfolio investment flows, expectations, and Brazilian stock market returns.
| Autor | Meurer, Roberto |
-
Introduction
The importance of capital markets in financing economic growth and development in emerging markets is widely recognized in the literature (e.g., Bekaert et al., 2007; King and Levine, 1993). On the other hand, the use of foreign financing presents the risk of capital flowing out of the country during crises, jeopardizing economic growth (Obstfeld, 2009).
This paper analyzes the effect of foreign portfolio investment (FPI) flows on the Ibovespa, the main Brazilian stock market index. Brazil accounted for 3% of World GDP in 2003 and 2.5% in 2018 (International Monetary Fund, 2019). The market value of Brazilian public companies was 49% of GDP on average between 2003 and 2017. FPI relates to the real economy through effects on the balance of payments and exchange rate, among other channels.
Figure 1 shows that foreign investors represent a growing share of traded volumes in the Brazilian market. From 2003 to 2018, this share has never fallen below 20%, and stabilizes at around 50% from 2014 to 2018. Because of this high share, it is interesting to study the interaction of the behavior of foreign investors and the returns of the market, taking into account the local and international macroeconomic environment. The average flow for transactions in the country was 9.4 billion dollars between 2003 and 2018.
Despite the integration of international financial markets, dividing them into developed and emerging ones is still useful (Bekaert and Harvey, 2017) and makes use of pull and push factors relevant for international capital flows (Koepke, 2019; Fratzscher, 2012; Puy, 2016; Sarno et al., 2016, among others).
The main contribution of this paper is the explicit inclusion of the expectations of local macroeconomic variables as explanatory variables, which will allow for a better isolation of the effects of flows on market returns. Expectations about interest rates, inflation, and economic growth are used. Another contribution is the inclusion of commodity prices as an explanatory variable, given their weight in the Brazilian economy. Results show that FPI, expectations, and commodities help to explain returns in the Brazilian stock market. Specifically, Granger causality tests show that past changes in the monetary policy interest rate and expectations about the monetary policy interest rate help explain the Ibovespa returns. On the other hand, the stock market returns anticipate the behavior of traded volume in the stock market, commodity prices, real effective exchange rates, inflation, economic activity, inflation expectations, output expectations, country risk, and the VIX volatility index. Impulse-response analysis shows that a shock on stock market returns positively affects itself, which permanently impacts the index's level. On the other hand, exchange rate depreciation, increases in the interest rate, country risk, and changes in interest rate expectations have a permanent negative effect on stock market index returns, meaning a permanent change in the index's level. The impact of shocks on equity and fixed income FPI flows, changes in commodity prices, and S&P500 returns are positive but temporary, having no lasting effect on the index's level.
The paper is organized into three additional sections. Section 2 provides a brief literature review. Section 3 shows the data and results, and section 4 presents the concluding remarks.
-
Literature review
The literature on the influence of portfolio inflows and outflows on host countries' economies is growing. The main effects are on economic growth (e.g., Errunza, 2001; Durham, 2004; Damasceno, 2014), the foreign sector (Fry et al., 1999; Curcuru et al., 2008), and financial markets (Hau and Rey, 2008; Goncalves Jr. and Eid Jr., 2017).
One of the features of FPI flows that generates macroeconomic effects is an overreaction that leads to booms and "sudden stops" of these flows (Agosin and Huaita, 2012), which are also influenced by local macroeconomic factors (Durham, 2004). Global factors, such as risk perception, help to explain extreme capital flows (Forbes and Warnock, 2012). One of the main global variables in the literature that explains flows is global liquidity, usually proxied by an interest rate (Koepke, 2019). The author found that lower interest rates in developed economies push portfolio capital to emerging economies, a result also reported by Byrne and Fiess (2016). Domestic factors, such as institutions and macroeconomic fundamentals, explain the heterogeneity of global effects between countries (Fratzscher, 2012; Ghosh et al., 2014).
Information flows are a channel that relates FPI with the real side of the economy. Andrade and Chhaochharia (2010) show that FPI from the U.S. is explained by past direct investment. Direct investment makes it possible to gather information about the host economy, which is later used also to evaluate portfolio investments. Hence, the size of the host economy is important to make feasible the cost of obtaining information and to allow for better diversification. In the case of the stock market, the contemporaneous relation between flows and returns may be a consequence of information flows (Tille and van Wincoop, 2014; Sanvicente, 2014).
Foreign capital flows have an influence on stock market returns because they broaden firms' potential investor bases (Clark and Berko, 1997). Examples of the literature on the relation between FPI and host stock market returns are studies by Brennan and Cao (1997), Froot et al. (2001), Dahlquist and Robertsson (2004), Hau and Rey (2008), and Ulku and Ikizlerli (2012), among many others, which usually find a positive and lasting relationship. Because this effect is lasting, it is important from the point of view of the financial and real economy through its impact on firms' cost of capital and resource allocation.
There are different explanations for the relationship between FPI flows and stock market returns. The positive or negative feedback hypothesis states that flows and returns reinforce each other. The most well-known explanation is "trend-chasing," meaning that flows are directed to markets or assets with positive returns and leave ones with negative returns (Bohn and Tesar, 1996). Another explanation is "portfolio rebalancing": investors would buy or sell positions in the market to keep constant the share allocated in specific markets or stocks. This could explain how flows react to changes in stock prices and exchange rates, which is relevant for portfolios with assets in more than one country (Hau and Rey, 2008). As the origin of the rebalancing could be changes in asset prices in the origin country, international markets could be an important explanatory variable for portfolio capital flows. This possibility is analyzed in Tabak (2003) and Reis et al. (2010) for the Brazilian case.
The literature that analyzes the determinants of FPI for the Brazilian case is carefully reviewed by Goncalves Jr. and Eid Jr. (2016). Leal and Rego (1997) show that a regulation about foreign investments in the Brazilian financial markets, known as "Anexo IV", did not influence the returns and volatility of the Brazilian main market index, the Ibovespa. Meurer (2006) detects an indirect influence of foreign flows on Ibovespa returns through its effect on liquidity. Sanvicente (2014) tests the relationship between flows and daily returns and finds that flows influence returns; the study also uncovers the effects of country risk and the exchange rate. Employing daily data, Goncalves Jr. and Eid Jr. (2017) confirm the contemporaneous positive relationship between flows and stock returns, and find evidence of trend-chasing by foreign investors. Beyond this, they suggest that foreign flows may be destabilizing during crises. The influence of the exchange rate, country risk, and the U.S. stock market was also found by Ferreira and de Melo Zachis (2012) and Montes and Tiberto (2012).
The inclusion of macroeconomic variables as controls in the explanation of stock returns is usual in the literature. For the Brazilian case, examples are studies by Nunes et al. (2005), Silva et al. (2014), Araujo and Bastos (2008), Pimenta Junior and Hironobu Higuchi (2013), Goncalves Jr. and Eid Jr. (2011), and Soares et al. (2021), among others.
The size of the effect of foreign investors' purchases and sales on local stock prices may depend on the volume traded in the market, which has to be taken into account (Bekaert et al., 2007).
French and Li (2012) find a positive relationship between commodity prices and investment flows from the U.S. to Brazil. This is consistent with the importance of commodities for the Brazilian economy and its balance of payments. Likewise, Chicoli and Sousa (2016) found a positive relationship between Ibovespa returns and the commodity index.
Liberalization of the capital account in emerging markets has made these countries more integrated with international markets, making the behavior of global markets a part of the explanation of domestic markets' behavior (Reinhardt et al., 2013; French and Li, 2012). This structural change will also impact the effects of flows on the economy (Prasad et al., 2005). However, the Brazilian capital account opening happened in 1991 (Bekaert and Harvey, 2000), before the data span used in this paper.
This study uses gross flows, meaning that flows originated by foreign investors are separated from investments made by nationals abroad. Gross flows are considerably higher than net flows, which consider the difference between liabilities (foreign investors in the country) and assets (nationals investing abroad). Specifically, the flow used is the investment by foreigners, which is related to the country's liabilities. Beyond the higher volume, gross flows are also more volatile and exhibit asymmetric behavior, especially during crises (Broner et al., 2013). In addition, the information richness is higher for gross flows, as is also...
Para continuar a ler
PEÇA SUA AVALIAÇÃOCOPYRIGHT GALE, Cengage Learning. All rights reserved.
