We study different channels through which well-known benchmark indexes impact asset allocations, capital flows, asset prices, and exchange rates across countries, using unique monthly micro-level data of benchmark compositions and mutual fund investments during 1996-2014. We exploit different events and the presence of countries in multiple benchmarks to study the impact of benchmarks. We find that movements in benchmarks appear to have important effects on equity and bond mutual fund

more »

... , including passive and active funds. The effects persist even after controlling for other relevant variables, such as time-varying industry-level factors, country-specific effects, and macroeconomic fundamentals. Exogenous, pre-announced changes in benchmarks impact asset allocations, capital flows, and abnormal returns in asset prices and exchange rates. These systemic effects occur not just when the benchmark changes are announced, but also later on when they become effective. By impacting country allocations, benchmarks explain apparently counterintuitive movements in capital flows and aggregate prices. JEL Classification Codes: F32, F36, G11, G15, G23 Effective date This table presents the results from an event study of all episodes of large benchmark changes. All returns are cumulative returns starting at the first day presented in the table. Raw returns are the net returns of the stock/debt market index or the exchange rate for the country. Excess returns are returns minus a global factor. Abnormal returns are residuals of a one factor CAPM model. The global factors used are the MSCI All Country World for equity, the J.P. Morgan GBI for local currency bonds, the J.P. Morgan EMBI for investment/non-investment grade, and the average currency return for the exchange rate. Panel A presents the returns for the pooled events from the announcement and effective date. Panel B shows results for the announcement date. Panel C depicts results for the effective date. Announcement date and effective date are denoted by TA and TE, respectively. Positive currency returns denote a depreciation. Standard errors are in parentheses.