This paper investigates two issues: whether there is heterogeneity for fund managers as investors and whether there is asymmetric volatility under short-sale constraints. If so, what are the driving factors in the Korean fund market? Fund return data from 2002 to 2008 are used to determine these factors. Specifically, for short-sale constraints, we test the hypothesis of difference of opinion developed by Chen, Hong, and Stein (2001) and Hong and Stein (2003). This hypothesis provides a unique opportunity to test directly the differences of opinion among fund managers that operate fund monies under short-sale constraints using asset-allocating strategies. The results of the GJR-GARCH model show an asymmetric volatility in returns and an increase in differences of opinion among fund managers, which extended to an increase in asymmetric volatility. Furthermore, the results of this study are consistent with the model of Hong and Stein (2003), which predicts that negative asymmetries are more likely to occur when there are large differences of opinion among fund managers. Therefore, our results imply that the overvaluation effect is more remarkable in funds for which a wider dispersion of the opinions of fund managers exists. These findings are consistent with Miller’s (1977) intuition and Hong and Stein’s (2003) model. In addition, our results also support the stochastic bubble hypothesis and are consistent with Blanchard and Watson (1982) and Wu (1997), even after controlling for fund characteristic variables.El trabajo estudia si existen heterogeneidades en la inversión y si existe volatilidad asimétrica bajo ventas cortas en el mercado de fondos coreano. Para ello se usan datos de retornos para el período 2002-2008. Para las restricciones de ventas cortas se testea la hipótesis de diferencias de opiniones de Chen, Hong, y Stein (2001) y Hong y Stein (2003). Los resultados del modelo GJR-GARCH muestra volatilidad asimétrica en retornos y un aumento en las diferencias de opiniones de los administradores de fondos. A su vez, los resultados son consistentes con el modelo de Hong y Stein (2003) y la intuición provista por Miller (1977), que predice que asimetrías negativas son más probables cuando existen diferencias de opiniones, habiendo también mayor sobrevaloración. Nuestros resultados también ofrecen soporte a la hipótesis de burbujas estocásticas de Blanchard y Watson (1982) y Wu (1997), aun después de controlar por las características de los fondos.