Effects associated with index composition changes: evidence from the Euronext Lisbon stock exchange

• Autor(a) principal: Duque, João
• Data de Publicação: 2004
• Outros Autores: Madeira, Gustavo
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10400.5/2258
• topic: Stock Index Revisions; Index Composition; Price and Volume Effects; Event Study; Abnormal Returns.

According to previous studies in many other markets, changes in a stock market index composition have shown abnormal returns are available at the date when index changes become effective. Stocks (added or deleted) tend to generate abnormal returns. But as market microstructure and stock index governing rules differ from country to country and from index to index, studies on previously unstudied markets are relevant. This study examines the stock reaction that occurs when shares are added to or deleted from the Euronext Lisbon stock index PSI-20, studied in terms of abnormal returns. And although the vast majority of papers in the literature are concerned with price effect and volume effect, this study also focuses on the trading volume and on the volatility effect. Although the rules governing the PSI-20 index are publicly available, the market seems to be surprised when announcements of composition change are released. stock price returns react positively for additions to and negatively for deletions from the index. All these observed effects are evident for the time period between the announcement day and the effective day of the change. when the effective day of the change arrives, the market still reacts to stock additions to the index, decreasing significantly the price that was being raised until then, in what may be a sign of a previous overreaction when the announcements were made. It also reacts in terms of trading volume. A positive abnormal trading volume is observed after additions to and deletions from the index, with some persistency after the event. The volatility effects do not seem to be statistically significant. All statistically significant findings on price returns seem to be temporary which lends itself to the empirical support for the price pressure hypothesis. However, the empirical evidence of a persistent increase in trading volume either for additions to or for deletions from the index becomes a puzzling and contradictory support for the liquidity hypothesis.