IV. Does Advertisings Influence on Content Matter? A. Do Investors Respond to Media mentions? Studies on the determinants of fows into U.S. mutual funds typically focus on the relation between measures of past performance and other characteristics and future inflows(see, for example, Ippolito [1992and Chevalier and Ellison [1997)). However, as Sirri and Tufano [1998 argue, it is costly for investors to gather and process information on the universe of available funds, suggesting that the media may play an important role in disseminating this information. Consistent with investors having lower search costs for mutual funds they have been exposed to through the media, Sirri and Tufano find that mutual funds receiving more media attention receive correspondingly higher inflows. Similarly, Jain and Wu 2000, Cronqvist 2004 and Gallagher, Kaniel, and Starks 2005 provide evidence that mutual fund flows respond to fund-level and family-level advertising, while Del Guercio and Tkac[2005 provide evidence that flows respond to changes in Morningstar ratings. Collectively, these papers suggest that investors rely on both advertising and the media when deciding which mutual funds to buy. Since media mentions are only valuable to mutual fund families o the extent that they influence investor behavior, in this section, we ask whether the media mentions we study have a significant impact on mutual fund inflows In Table V, we estimate the impact of media mentions on fund inflows using Fama-Mac Beth [1973 style regressions. Each month between January 1997 and December 2001, we estimate a separate cross-sectional regression. We then report the time-series means and standard errors associated with these cross-sectional estimates. Our dependent variable measures net inflows into fund i between months t and t+ll. The number of media mentions that fund i receives in month t-l is measured separately for positive and negative mentions in each publication. To test whether family-level advertising expenditures are systematically related to fund level inflows, we include advertising expenditures on print and non-print media over the prior twelve months normalized by the average dollars under management within the fund family over this period. In addition to the fund characteristics used to predict media mentions in Table Ill, we also control for the raw return of fund i between months t and t+1l. Finally, we include fixed effects for each investment objective within each monthly cross-sectional regresIV. Does Advertising’s Influence on Content Matter? A. Do Investors Respond to Media Mentions? Studies on the determinants of flows into U.S. mutual funds typically focus on the relation between measures of past performance and other characteristics and future inflows (see, for example, Ippolito [1992] and Chevalier and Ellison [1997]). However, as Sirri and Tufano [1998] argue, it is costly for investors to gather and process information on the universe of available funds, suggesting that the media may play an important role in disseminating this information. Consistent with investors having lower search costs for mutual funds they have been exposed to through the media, Sirri and Tufano find that mutual funds receiving more media attention receive correspondingly higher inflows. Similarly, Jain and Wu [2000], Cronqvist [2004], and Gallagher, Kaniel, and Starks [2005] provide evidence that mutual fund flows respond to fund-level and family-level advertising, while Del Guercio and Tkac [2005] provide evidence that flows respond to changes in Morningstar ratings. Collectively, these papers suggest that investors rely on both advertising and the media when deciding which mutual funds to buy. Since media mentions are only valuable to mutual fund families to the extent that they influence investor behavior, in this section, we ask whether the media mentions we study have a significant impact on mutual fund inflows. In Table V, we estimate the impact of media mentions on fund inflows using Fama-MacBeth [1973] style regressions. Each month between January 1997 and December 2001, we estimate a separate cross-sectional regression. We then report the time-series means and standard errors associated with these cross-sectional estimates. Our dependent variable measures net inflows into fund i between months t and t+11. The number of media mentions that fund i receives in month t−1 is measured separately for positive and negative mentions in each publication. To test whether family-level advertising expenditures are systematically related to fund￾level inflows, we include advertising expenditures on print and non-print media over the prior twelve months, normalized by the average dollars under management within the fund family over this period. In addition to the fund characteristics used to predict media mentions in Table III, we also control for the raw return of fund i between months t and t + 11. Finally, we include fixed effects for each investment objective within each monthly cross-sectional regression. 12