Portal de Periódicos da CAPES

International Transmission Mechanism of Stock Market Volatilities

2008; Taylor & Francis; Volume: 9; Issue: 1 Linguagem: Inglês

10.1080/10978520802189302

1528-6932

Andrés Rivas, Rahul Verma, Antonio J. Rodríguez, Priti Verma,

Monetary Policy and Economic Impact

ABSTRACT We investigate the volatility spillover effects of European equity markets to the equity markets of Mexico, Brazil, and Chile. The results of the E-GARCH and VAR models suggest that the stock markets of Spain and Germany have stronger volatility spillover effects on Latin American markets than do Italy, the United Kingdom, and France. We find that these spillover effects of Spain and Germany have a greater impact on Mexico and Brazil than on Chile. Moreover, in all these cases negative innovations increase volatility more than do positive innovations. We tie our results to the relative degree of openness of Latin American countries and their level of international trade with the European economies. Mexico and Brazil are relatively more open economies than is Chile. Also, these two Latin American economies have higher levels of international trade with Spain and Germany than with other European economies. Our results are consistent with the notion that, as an economy becomes more open and integrated with the world economy, its financial sector becomes more susceptible to external shocks. RESUMEN. Investigamos los efectos secundarios que los mercados de capitales europeos provocan sobre los que operan en México, Brasil y Chile. Los resultados obtenidos con los modelos exponenciales E-GARCH y VAR, sugieren que los efectos indirectos de volatilidad causados por los mercados de valores de España y Alemania son mucho más potentes que los que Italia, el Reino Unido y Francia ejercen sobre los mercados latinoamericanos. Descubrimos que dichos efectos secundarios provenientes de España y Alemania ejercen un efecto más contundente sobre México y Brasil que sobre Chile. Además, en todos estos casos el aumento de volatilidad provocado por las innovaciones negativas es mayor que el resultante de las innovaciones positivas. Hemos vinculado nuestros resultados al grado relativo de apertura que existe en los países latinoamericanos y a su nivel de comercio internacional con las economías europeas. Las economías de México y Brasil son relativamente más abiertas que la chilena. Además, el nivel del comercio internacional de estas dos economías latinoamericanas con España y Alemania es mucho más alto que el existente con las demás economías europeas. Nuestros resultados coinciden con la noción de que, a medida que una economía adquiere una apertura más amplia y se integra más con la economía mundial, su sector financiero se torna más susceptible a los choques externos. RESUMO. Analisamos os efeitos do excesso de volatilidade do mercado de ações europeu em relação aos mercados de ações do Brasil e do Chile. Os resultados dos modelos E-GARCH e VAR indicam que os mercados de ações da Espanha e da Alemanha apresentam efeitos mais consistentes do excesso de volatilidades do que os da Itália, Reino Unido e França nos mercados da América Latina. Acreditamos que os efeitos destes excessos da Espanha e da Alemanha são de maior magnitude no México e no Brasil do que no Chile. Além disso, em todos estes casos, os efeitos das inovações negativas aumentam a volatilidade mais do que os efeitos das inovações positivas. Atrelamos os nossos resultados ao grau de relativa abertura dos países da América Latina e o seu nível de comércio internacional com as economias européias. A economia do México e a economia do Brasil são relativamente mais abertas do que a do Chile. Por outro lado, estas duas economias latino-americanas mantêm um volume maior de comércio internacional com a Espanha e a Alemanha do que com outras economias européias. Os nossos resultados coincidem com a noção de que, à medida que uma economia se torna mais aberta e mais integrada com a economia mundial, a sua área financeira torna-se mais suscetível aos choques externos. KEYWORDS: EGARCHemerging marketsLatin Americastock markets interdependence Notes Source: IMF, Direction of Trade Flows statistics, 2005. Source: UNCTAD. Brazilian stock market return (R_BR); Chilean stock market return (R_CH); Mexican stock market return (R_MX); Spain market return (R_SPA); Italian market return (R_ITL); German stock market return (R_GER); French stock market return (R_FR); U.K. stock market return (R_UK); and U.S. stock market return (R_US). All the variables are in the form of continuously compounded rate of change. ***, **, and * denote statistical significance at the 1%, 5%. and 10% levels, respectively. LB(n) for Ri,t is the Ljung-Box statistic identifying the presence of autocorrelation, while LB(n) for RFootnote 2 i,t is the statistic identifying the presence of heteroscedasticity. Jarque-Bera is the test for the null hypothesis of normality. Brazilian stock market return (R_BR); Chilean stock market return (R_CH); Mexican stock market return (R_MX); Spain market return (R_SPA); Italian market return (R_ITL); German stock market return (R_GER); French stock market return (R_FR); U.K. stock market return (R_UK); and U.S. stock market return (R_US). All the variables are in the form of continuously compounded rate of change. The variables are Brazilian stock market volatility (V_BR); Chilean stock market volatility (V_CH); Mexican stock market volatility (V_MX); Spain market volatility (V_SPA); Italian market volatility (V_ITL); German stock market volatility (V_GER); French stock market volatility (V_FR); U.K. stock market volatility (V_UK). The variables are Brazilian stock market volatility (V_BR); Chilean stock market volatility (V_CH); Mexican stock market volatility (V_MX); Spain market volatility (V_SPA); Italian market volatility (V_ITL); German stock market volatility (V_GER); French stock market volatility (V_FR); U.K. stock market volatility (V_UK). See Kaminsky (1999) for a detailed review of literature on this financial crisis in emerging markets. The vertical axis values on the IRF graphs are extremely small. This may be due to the following reasons: first, these numbers are represented in decimals and not in percentages; second, we have carried out our analysis by using daily stock index data and, consequently, the daily returns are extremely small in decimals; and third, we have generated IRFs from a VAR model that consists of the variance series obtained from volatility models. These data points in the variance series are much smaller compared with the actual daily return series. Additional informationNotes on contributorsAndrés Rivas Andres Rivas is Assistant Professor of Finance and Economics and Interim Coordinator of the Texas Center for Border Economic & Enterprise Development at Texas A&M International University. His research interests are international finance, banking, derivatives, and equity markets with special focus on emerging markets. Rahul Verma Rahul Verma is Assistant Professor of Finance and Coordinator of Finance Program at the University of Houston-Downtown. His research interests are international finance, behavioral finance, and investments. E-mail: vermar@uhd.edu Antonio Rodriguez Antonio Rodriguez is Professor and Associate Dean of the College of Business at Texas A&M International University. His research interests are international finance and investments and international banking. E-mail: Rodriguez@tamiu.edu Priti Verma Priti Verma is Assistant Professor at the College of Business at Texas A&M University, Kingsville. Her research interests are financial econometrics, volatility and risk management, and international finance. E-mail: priti.verma@tamuk.edu