[Previous findings regarding the risk-shifting behavior of mid-year underperforming mutual fund managers are mixed. In this article, I show that this is due to a "sorting bias," which is caused by the sorting of first-half risk levels when establishing relative midyear performance. Even without risk-shifting behavior, mean reversion of these sorted risk levels results in the detection of tournament behavior. After correcting for this bias, I find evidence supporting the hypothesis that first-half underperforming managers increase portfolio risk during the second half of the year and that this tournament behavior is not dependent on first-half market conditions.]

We document that CRSP and Thomson contain many voluntarily reported mutual fund portfolios that are not in SEC filings while, additionally, CRSP and Thomson are missing many SEC mandated portfolios available in SEC filings. We document that the voluntary disclosures are likely driven by convenience rather than duplicity. Although mandated portfolios contain securities with more return momentum, we find use of SEC or Thomson data lead to similar empirical findings. CRSP, however, contains inaccurate position information prior to 2008. Our findings have important implications, such as highlighting a 35% increase in observed manager trading by combining data sources.

Previous findings regarding the risk-shifting behavior of mid-year underperforming mutual fund managers are mixed. In this article, I show that this is due to a “sorting bias,” which is caused by the sorting of first-half risk levels when establishing relative mid-year performance. Even without risk-shifting behavior, mean reversion of these sorted risk levels results in the detection of tournament behavior. After correcting for this bias, I find evidence supporting the hypothesis that first-half underperforming managers increase portfolio risk during the second half of the year and that this tournament behavior is not dependent on first-half market conditions.

We document that CRSP and Thomson contain many voluntarily reported mutual fund portfolios that are not in SEC filings while, additionally, CRSP and Thomson are missing many SEC mandated portfolios available in SEC filings. We document that the voluntary disclosures are likely driven by convenience rather than duplicity. Although mandated portfolios contain securities with more return momentum, we find use of SEC or Thomson data lead to similar empirical findings. CRSP, however, contains inaccurate position information prior to 2008. Our findings have important implications, such as highlighting a 35% increase in observed manager trading by combining data sources.