Essays in financial economics

Abstract

In Chapter 1, we seek to understand the relation between liquidity and market imperfections from two dimensions: 1) Across liquidity measures, we compare the influence of imperfections on two commonly used measures, Kyle's lambda and price reversal; 2) Across imperfections, we study the interaction between two sources of market imperfection, information asymmetry and participation cost. We show that the two liquidity measures may be affected in opposite directions by the same imperfection, or may not capture liquidity changes at all; imperfection interactions can cause the market to appear "less illiquid" than single-imperfection benchmarks. Our model also suggests that imperfections and liquidity shocks may influence expected returns in opposite directions, which complicates the liquidity-asset price cross-sectional relation. In Chapter 2, joint with Andrew Lo, we perform an empirical comparison of systemic risk measures. In a recent survey paper, Bisias et al. (2012) provide a summary of 31 proposed measures for systemic risk in the financial system. In this paper we examine a subset of these measures to determine their time series properties before, during, and after the Financial Crisis of 2007-2009. By comparing their empirical properties over time, we hope to identify which measures were most informative for navigating through the 1998 and the 2007-2009 crises. By constructing rolling-window estimates of these measures using only prior data, we control for the most blatant forms of look-ahead bias to assess the value of these measures as "early-warning signals". Finally, we explore the possibility of combining these measures to produce even more informative indicators of systemic risk. In Chapter 3, joint with Andrew Lo and Silvia Sgherri, we construct two global systemic In Chapter 3, joint with Andrew Lo and Silvia Sgherri, we construct two global systemic risk indicators as well as a panel of regional indicators, using monthly hedge fund data. Results show that our geographic-focus global indicator provided contemporaneous characterization of financial distress; the hedge fund style-category global indicator generated early-warnings for the 2007 quant crisis and the 2011 European debt crisis, and typically led the geographic-focus indicator by 1~2 months. In addition, we use Granger causality network to visualize the interconnectedness of regional risks and track the transmission of crisis over time.