David A. Rakowski

Fund Flow Volatility and Performance

This paper provides a detailed analysis of the impact of daily mutual fund flow volatility on fund performance. I document a significant negative relationship between the volatility of daily fund flows and cross-sectional differences in risk-adjusted performance. This relationship is driven by domestic equity funds, as well as small funds, well-performing funds, and funds that experience inflows over the sample period. My results are consistent with performance differences arising from the transaction costs of nondiscretionary trading driven by daily fund flows, but not with performance differences arising from the suboptimal cash holdings that arise from fund flows.

Geography and Local (Dis)Advantage: Evidence from Muni Bond Funds

We examine the relationship between the geographic location of mutual fund managers and fund performance using the unique setting of single-state municipal-bond mutual funds. We find that local managers underperform non-local muni-bond fund managers. Furthermore, we document that local muni-bond fund managers perform relatively better in states with more local funds, consistent with knowledge spillovers, business connections and networking effects associated with those areas. Locals also perform relatively better in states with higher levels of political integrity, consistent with less political pressure on local fund managers in these locations. Our results are robust to several sensitivity checks.

Geography and Local (Dis)advantage

We examine the relationship between the geographic location of mutual fund managers and fund performance using the unique setting of single-state municipal-bond mutual funds. We find that local managers underperform non-local muni-bond fund managers. Furthermore, we document that local muni-bond fund managers perform relatively better in states with more local funds, consistent with knowledge spillovers, business connections and networking effects associated with those areas. Locals also perform relatively better in states with higher levels of political integrity, consistent with less political pressure on local fund managers in these locations. Our results are robust to several sensitivity checks.

The Sources of Portfolio Returns: Underlying Stock Returns and the Excess Growth Rate

This paper decomposes portfolio returns into the underlying sources arising from the constituent stocks’ growth rates, as well as their variances and covariances. We employ this method to show that the difference between large and small stock portfolio returns is driven by a portfolio “excess growth rate” that is induced by the higher volatility of small stocks’ returns and not by the average growth rate of small stocks. Therefore, the “size effect” is not a small firm effect, but a small firm portfolio effect driven by the excess growth rate of the portfolios. In contrast, portfolios of high book-to-market stocks outperform due to higher average levels of growth by the constituent stocks and not due to their variance-covariance structure. Our results demonstrate the importance of considering the sources of portfolio performance as possibly distinct from the performance of the portfolio’s underlying stocks when designing and interpreting studies of portfolio performance, corporate events, or the cross-section of stock returns.

Is All That Twitters Gold? Social Media Attention and Stock Returns

Measures of information traditionally arise from curated, professional sources such as newspapers, analyst coverage, earnings announcements, and business news wires. We utilize Twitter activity to examine the impact of attention generated by individuals. Causal evidence shows that Twitter activity has a direct impact on trading volume in financial markets. Furthermore, increases in Twitter activity are associated with positive abnormal returns and, when occurring in conjunction with traditional information supply events, increase the diffusion of information to investors. Our results identify conditions under which attention generated by individuals drives price discovery and trading activity.

Cash Flow and Discount Rate Risk in the Investment Effect: A Downside Risk Approach

We examine whether sensitivities to cash flow and discount rate risk in down markets explain the investment effect, in which low-investment stocks earn higher expected returns than high-investment stocks. We show how productivity and financing constraints asymmetrically impact the systematic risk of low-investment and high-investment firms, conditional on market state. Our evidence is consistent with both productivity constraints and financing constraints as explanations for the investment effect, but, contrary to expectations, more when prices are rising than falling.

Explaining Demand for Financial Innovation: Evidence from Exchange-Traded Notes

Exchange-traded notes (ETNs) are a relatively new form of security design that appear similar to exchange-traded funds (ETFs), but with no underlying portfolio holdings. We identify those characteristics of ETNs that are distinct from ETFs, and we test which ETN characteristics are most associated with interest from investors. We find that some ETNs have return patterns that are not spanned by ETFs, while other ETNs employ different strategies to attract investors. We show that the market activity for ETNs is associated with the distinctiveness of ETN returns, improvements in tracking errors, access to leverage, and the extent of risk transformation that ETNs allow.

The Endogeneity of Trading Volume in Stock and Bond Returns: An Instrumental Variable Approach

This paper investigates the joint determination of two dimensions of a security: trading volume and return. In much of the existing literature, volume is modeled as being exogenously related to security returns. Our analysis evaluates the extent to which trading activity also depends on security returns, thus resulting in a reverse causality from returns to trading activity. Using instruments for security trading activity, we compare approaches using ordinary least squares (OLS), two-stage least squares (2SLS), and generalized method of moment (GMM) estimators. Using both stock and bond data, our results indicate that returns and trading volume are determined simultaneously. We show that estimates of the magnitude, direction, and significance of causality between volume and returns can be substantially changed once one corrects for the endogeneity of volume.