This paper uses sample from 15 countries from different regions around the world to examine how past performance of exchange-traded funds (ETFs) impact on their fund flow in the following period. Piecewise-linear regression following (Sirri & Tufano, 1998) which allow past performance to be non-linear are employed in the model to examine the different sensitivity of past performance ranking to flow. The evidence confirms that past bad-performing ETFs associate with outflow and past good-performing ETFs associate with inflow in the following period. Furthermore, I also find that investors are more sensitive to bad performers than good performers suggesting that there is asymmetry in flow-performance relationship in ETFs. This can be explained by the behavioral bias form investor that react more to bad news than good news. Apart from the generalize result from across countries, I also investigate the difference in flow-performance relationship in each country level. The evidence suggests that difference countries exhibit the different shape of flow-performance relationship depending on the level of investor sophistication. Investors in more developed countries imply more sophisticated and thus have less behavioral bias to chase past winner than those in less developed countries.
The flow-performance relationship of ETFs in international market
Post by MSF Chula at Friday, 8 January 2021 10:47 PM
Last updated at Friday, 8 January 2021 10:47 PM