liquidity risk and asset pricing

LIQUIDITY RISK AND ASSET PRICING DISSERTATION Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University By Kuan-Hui Lee, M.A ***** The Ohio State University 2006 Dissertation Committee: Approved by G Andrew Karolyi, Adviser Kewei Hou Ren´ M Stulz e Ingrid M Werner Adviser Graduate Program in Business Administration ABSTRACT In this dissertation, I investigate the effect of liquidity risk on asset pricing In the first essay, I test the liquidity-adjusted capital asset pricing model (LCAPM) of Acharya and Pedersen (2005) for 1962-2004 in the US market using various liquidity proxies In a time-series test with a one-factor (market model), three-factor (excess market return, SMB, and HML) and four-factor model (excess market return, SMB, HML and MOM) as well as in a Fama-MacBeth regression, I find that test results vary according to the liquidity measures used, to the test methodology, to the test assets, and to the weighting scheme Tests based on the liquidity measure of Amihud (2002), Pastor and Stambaugh (2003) and zero-return proportion show some evidence that liquidity risks are priced, but in most cases, I could not find evidence that supports the LCAPM The second essay specifies and tests an equilibrium asset pricing model with liquidity risk at the global level The analysis encompasses 25,000 individual stocks from 48 developed and emerging countries around the world from 1988 to 2004 Though I cannot find evidence that the LCAPM holds in international financial markets, cross-sectional as well as time-series tests show that liquidity risks arising from the covariances of individual stocks’ return and liquidity with local and global market factors are priced Furthermore, I show that the US market is an important driving force of world-market liquidity risk I interpret our evidence as consistent with an intertemporal capital asset pricing model (Merton (1973)) in which ii stochastic shocks to global liquidity serve as a priced state variable The third essay investigates how and why liquidity is transmitted across stocks In a vector autoregressive framework, I uncover a dynamic interaction of liquidity across size portfolios in that past changes of liquidity of large stocks are positively correlated with current changes of liquidity of small stocks Furthermore, liquidity spillovers are not restricted among fundamentally-related stocks and are independent of the dynamics in return and volatility spillovers This finding implies that the process of liquidity generation is independent of information flows and that portfolio diversification strategies should consider different patterns in return, volatility and liquidity spillovers iii to my parents iv ACKNOWLEDGMENTS I don’t know how to thank enough for having such an exceptional doctoral committee I wish to thank Kewei Hou, G Andrew Karolyi, Ren´ Stulz, and Ingrid M e Werner for their continual support and encouragement I am most grateful to my advisor, G Andrew Karolyi, for his tremendous support, guidance and encouragement Throughout my doctoral work he encouraged me to develop my analytical thinking and research skills His assistance guided me not only to a better researcher, but also to a better instructor I wish to express my deep appreciation to Ren´ Stulz whose consultation, insightful e comments and encouragement were continual stimulation to my research I am deeply indebted to Ingrid M Werner whose help, guidance, patience and encouragement tremendously helped me in all my doctoral training process I would like to thank Kewei Hou for his insightful comments, efforts and guidance I also thank Karl Diether, Bing Han and Jean Helwege for their encouragement and friendship and thank Bong-Chan Kho for being an excellent role model even from my school life in Seoul many years ago I thank my brother Tae-Hwy Lee, who is a time-series econometrician, for guiding me in pursuing academic career I want to share my happiness and thankfulness with my friend, Jungwu Rhie, who passed away by tragic accident last year and may rest in peace in heaven v VITA November 9, 1970 Born - Taejon, Korea 1989-1996 B.B.A Business Administration, Seoul National University 1997-1999 MBA, International Finance, Seoul National University 1999-2001 M.S Statistics, University of North Carolina, Chapel Hill 2001-present Graduate Teaching and Research Associate, The Ohio State University FIELDS OF STUDY Major Field: Business Administration Studies in: International Finance Empirical Asset Pricing Market Microstructure vi TABLE OF CONTENTS Page Abstract ii Dedication iv Acknowledgments v Vita vi List of Tables x List of Figures xii Chapters: Introduction Testing the Liquidity-Adjusted Capital Asset Pricing Model using Different Measures of Liquidity 2.1 2.2 2.3 2.4 Introduction Liquidity-Adjusted Capital Asset Pricing Data and Iliquidity Measures Methodology 2.4.1 Fama-MacBeth Regression 2.4.2 Time-Series Tests 2.5 Empirical Results 2.5.1 Time-Series Tests 2.5.2 Fama-MacBeth Regression 2.6 Conclusion vii Model 11 16 17 19 21 21 23 28 The World Price of Liquidity Risk 30 3.1 3.2 3.3 3.4 Introduction Related literature Liquidity-adjusted capital asset pricing model Data and liquidity measure 3.4.1 Sample screening 3.4.2 Summary statistics 3.4.3 Is zero-return proportion a good proxy for illiquidity? Methodology 3.5.1 Innovations of return and illiquidity 3.5.2 Estimating predicted betas Empirical results 3.6.1 Local and global market betas 3.6.2 Fama-MacBeth test results for local liquidity risks 3.6.3 Fama-MacBeth test results for global liquidity risks 3.6.4 Fama-MacBeth test results for local and world market liquidity risks 3.6.5 Do US markets drive world market liquidity risks? 3.6.6 Time series tests 3.6.7 Subperiod analysis Robustness tests 3.7.1 Different innovation 3.7.2 Two-way sorts Conclusion 30 33 36 39 39 43 45 48 48 49 53 53 55 57 Liquidity Spillovers 78 3.5 3.6 3.7 3.8 59 63 66 70 72 73 73 76 4.1 Introduction 78 4.2 Literature Review 83 4.2.1 Theories 83 4.2.2 Empirical Studies 85 4.3 Hypotheses 88 4.4 Data and Liquidity Measure 90 4.5 Liquidity Spillovers 92 4.5.1 Vector Autoregression 95 4.5.2 VAR Estimation Results 97 4.6 Are Liquidity Spillovers Related to Correlated Fundamentals or Information Flow? 99 4.6.1 Intra-Industry Liquidity Spillovers 99 4.6.2 Intra- vs Inter-Industry Liquidity Spillovers 102 viii 4.6.3 Does Information Flow Contribute in Liquidity Spillovers? 4.6.4 Does Style Investing Contribute to Liquidity Spillovers? 4.7 Robustness Checks 4.7.1 Alternative Lags 4.7.2 Value-Weighted Average 4.8 Discussion 4.9 Conclusion 105 106 115 116 116 118 119 Conclusion 122 Bibliography 124 Appendices: A Tables 131 B Figures 174 ix LIST OF TABLES Table Page A.1 Summary Statistics 132 A.2 Betas by Size Group 133 A.3 Intercepts from the Market Model Regression 134 A.4 Intercepts from the Three Factor Model Regression 135 A.5 Intercepts from the Four Factor Model Regression 136 A.6 Fama-French Regression of Size Portfolios with Value-Weighted Market Illiquidity 137 A.7 Fama-French Regression of Size Portfolios with Equal-Weighted Market Illiquidity 139 A.8 Fama-French Regression of Illiquidity Portfolios with Value-Weighted Market Illiquidity 141 A.9 Fama-French Regression of Illiquidity Portfolios with Equal-Weighted Market Illiquidity 143 A.10 Summary Statistics (Liquidity and Return in World Financial Markets) 145 A.11 Correlation of Liquidity Measures by Size in US Market 147 A.12 Coefficients from the estimation of predicted betas 148 A.13 Mean of Portfolio formed based on Predicted Betas 149 x PS (three factor model) 2.00 12.0 10.0 1.50 6.0 1.00 4.0 0.50 t-value intercept (%) 8.0 2.0 0.0 0.00 low high 10-1 -0.50 -2.0 -4.0 liquidity net beta EW VW t-val (EW) t-val (VW) RV (three factor model) 1.60 10.0 1.40 8.0 1.20 intercept (%) 0.80 4.0 0.60 2.0 0.40 0.20 t-value 6.0 1.00 0.0 0.00 -0.20 low -0.40 high 10-1 -2.0 -4.0 liquidity net beta EW VW t-val (EW) t-val (VW) Figure B.4: Intercepts from Three Factor Model Regression (Liquidity Net Beta) Figure shows the coefficients (bar shaped) and t-values (bullet lines) from three-factor model regression specified in equation (2.11) where stocks are sorted into 10 equally-weighted (EW) or market value-weighted (VW) portfolios based on estimated liquidity net beta 10 − shows the intercept and t-value from buying the highest beta portfolio and selling the lowest beta portfolio PS denotes illiquidity measure is from Pastor and Stambaugh (2003), RV from Amihud (2002), TV for turnover, ZR for zero-return proportion and RO for Roll (1984)’s measure In computing betas, market illiquidity is formed as a market value-weighted average of individual stocks’ illiquidity and market return is from CRSP value-weighted index 184 (Figure B.4 continued) TV (three factor model) 0.80 8.0 0.60 6.0 4.0 0.20 2.0 0.00 low high 10-1 -0.20 t-value intercept (%) 0.40 0.0 -2.0 -0.40 -0.60 -4.0 liquidity net beta EW VW t-val (EW) t-val (VW) ZR (three factor model) 1.00 10.0 0.80 8.0 6.0 0.40 4.0 0.20 2.0 0.00 -0.20 low high 10-1 0.0 -2.0 -0.40 -0.60 -4.0 liquidity net beta EW VW t-val (EW) 185 t-val (VW) t-value intercept (%) 0.60 (Figure B.4 continued) RO (three factor model) 0.80 7.0 6.0 5.0 4.0 0.40 3.0 0.20 2.0 1.0 0.00 low high 10-1 0.0 -1.0 -0.20 -2.0 -0.40 -3.0 liquidity net beta EW VW t-val (EW) 186 t-val (VW) t-value intercept (%) 0.60 PS (three factor model) 1.00 7.0 6.0 0.80 5.0 4.0 3.0 0.40 2.0 0.20 1.0 t-value intercept (%) 0.60 0.0 0.00 low high 10-1 -0.20 -1.0 -2.0 -0.40 -3.0 net beta EW VW t-val (EW) t-val (VW) RV (three factor model) 1.60 10.0 1.40 8.0 1.20 6.0 0.80 0.60 4.0 0.40 2.0 0.20 0.00 -0.20 t-value intercept (%) 1.00 low -0.40 high 10-1 0.0 -2.0 net beta EW VW t-val (EW) t-val (VW) Figure B.5: Intercepts from Three Factor Model Regression (Net Beta) Figure shows the coefficients (bar shaped) and t-values (bullet lines) from threefactor model regression specified in equation (2.11) where stocks are sorted into 10 equally-weighted (EW) or market value-weighted (VW) portfolios based on estimated net beta 10−1 shows the intercept and t-value from buying the highest beta portfolio and selling the lowest beta portfolio PS denotes illiquidity measure is from Pastor and Stambaugh (2003), RV from Amihud (2002), TV for turnover, ZR for zero-return proportion and RO for Roll (1984)’s measure In computing betas, market illiquidity is formed as a market value-weighted average of individual stocks’ illiquidity and market return is from CRSP value-weighted index 187 (Figure B.5 continued) TV (three factor model) 6.0 0.50 5.0 0.40 4.0 0.30 3.0 0.20 2.0 0.10 1.0 0.00 -0.10 t-value 7.0 0.60 intercept (%) 0.70 0.0 low high 10-1 -0.20 -1.0 -2.0 -0.30 -3.0 net beta EW VW t-val (EW) t-val (VW) ZR (three factor model) 1.00 8.0 0.80 6.0 4.0 0.40 0.20 2.0 0.00 -0.20 low high 10-1 0.0 -2.0 -0.40 -0.60 -4.0 net beta EW VW t-val (EW) 188 t-val (VW) t-value intercept (%) 0.60 (Figure B.5 continued) RO (three factor model) 7.0 0.80 6.0 5.0 4.0 0.40 3.0 2.0 0.20 1.0 0.00 low high 10-1 0.0 -1.0 -0.20 -2.0 -0.40 -3.0 net beta EW VW t-val (EW) 189 t-val (VW) t-value intercept (%) 0.60 PS (four factor model) 1.60 10.0 1.40 8.0 1.20 6.0 0.80 0.60 4.0 0.40 2.0 0.20 0.00 -0.20 t-value intercept (%) 1.00 low high 10-1 -0.40 0.0 -2.0 liquidity net beta EW VW t-val (EW) t-val (VW) RV (four factor model) 1.40 12.0 1.20 10.0 8.0 0.80 6.0 0.60 4.0 t-value intercept (%) 1.00 0.40 2.0 0.20 0.0 0.00 -0.20 low high 10-1 -2.0 liquidity net beta EW VW t-val (EW) t-val (VW) Figure B.6: Intercept from four factor model regression (liquidity net beta) Figure shows the coefficients (bar shaped) and t-values (bullet lines) from four-factor model regression specified in equation (2.12) where stocks are sorted into 10 equallyweighted (EW) or market value-weighted (VW) portfolios based on estimated liquidity net beta 10 − shows the intercept and t-value from buying the highest beta portfolio and selling the lowest beta portfolio PS denotes illiquidity measure is from Pastor and Stambaugh (2003), RV from Amihud (2002), TV for turnover, ZR for zero-return proportion and RO for Roll (1984)’s measure In computing betas, market illiquidity is formed as a market value-weighted average of individual stocks’ illiquidity and market return is from CRSP value-weighted index 190 (Figure B.6 continued) TV (four factor model) 0.80 7.0 6.0 0.60 4.0 3.0 0.20 2.0 0.00 t-value intercept (%) 5.0 0.40 1.0 low high 10-1 0.0 -0.20 -1.0 -0.40 -2.0 liquidity net beta EW VW t-val (EW) t-val (VW) ZR (four factor model) 1.00 10.0 0.80 8.0 6.0 0.40 4.0 0.20 2.0 0.00 -0.20 low high 10-1 0.0 -2.0 -0.40 -0.60 -4.0 liquidity net beta EW VW t-val (EW) 191 t-val (VW) t-value intercept (%) 0.60 (Figure B.6 continued) RO (four factor model) 0.80 7.0 0.70 6.0 0.60 5.0 4.0 0.40 3.0 0.30 2.0 0.20 1.0 0.10 0.0 0.00 -0.10 low -0.20 high 10-1 -1.0 -2.0 liquidity net beta EW VW t-val (EW) 192 t-val (VW) t-value intercept (%) 0.50 PS (four factor model) 7.0 1.00 6.0 0.80 4.0 3.0 0.40 2.0 0.20 t-value intercept (%) 5.0 0.60 1.0 0.0 0.00 low high 10-1 -0.20 -1.0 -2.0 net beta EW VW t-val (EW) t-val (VW) RV (four factor model) 1.60 10.0 1.40 8.0 1.00 6.0 0.80 4.0 0.60 0.40 t-value intercept (%) 1.20 2.0 0.20 0.0 0.00 -0.20 low high 10-1 -2.0 net beta EW VW t-val (EW) t-val (VW) Figure B.7: Intercept from four factor model regression (net beta) Figure shows the coefficients (bar shaped) and t-values (bullet lines) from four-factor model regression specified in equation (2.12) where stocks are sorted into 10 equally-weighted (EW) or market value-weighted (VW) portfolios based on estimated net beta 10 − shows the intercept and t-value from buying the highest beta portfolio and selling the lowest beta portfolio PS denotes illiquidity measure is from Pastor and Stambaugh (2003), RV from Amihud (2002), TV for turnover, ZR for zero-return proportion and RO for Roll (1984)’s measure In computing betas, market illiquidity is formed as a market value-weighted average of individual stocks’ illiquidity and market return is from CRSP value-weighted index 193 (Figure B.7 continued) TV (four factor model) 6.0 0.50 5.0 0.40 4.0 0.30 3.0 0.20 2.0 0.10 1.0 0.00 -0.10 t-value 7.0 0.60 intercept (%) 0.70 0.0 low high 10-1 -0.20 -1.0 -2.0 net beta EW VW t-val (EW) t-val (VW) ZR (four factor model) 7.0 0.80 6.0 5.0 4.0 0.40 3.0 2.0 0.20 1.0 0.00 low high 10-1 0.0 -1.0 -0.20 -2.0 -0.40 -3.0 net beta EW VW t-val (EW) 194 t-val (VW) t-value intercept (%) 0.60 (Figure B.7 continued) RO (four factor model) 0.80 7.0 0.70 6.0 0.60 5.0 4.0 0.40 0.30 3.0 0.20 2.0 0.10 1.0 0.00 -0.10 low high 10-1 0.0 -1.0 -0.20 -0.30 -2.0 net beta EW VW t-val (EW) 195 t-val (VW) t-value intercept (%) 0.50 6% 80% 70% ZR1 5% ZR5 60% PQS1 PQS5 4% 40% 3% P Q S (% ) Z R (% ) 50% 30% 2% 20% 1% 10% 0% 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 0% Figure B.8: Correlation of Liquidity Measures in US Market From ISSM/TAQ data, daily proportional quoted spread (PQS; defined as quoted spread divided by quote mid-point) are calculated for AMEX and NYSE stocks for 1983-2003 and it is averaged every month to form monthly time-series Figure plots the zeroreturn proportion, ZR, and PQS in the US market over time The suffix denotes the variables are from the smallest quintile, while the suffix denotes the variables are from the largest quintile, where each quintile is formed based on individual stocks’ previous end-year market capitalization 196 Alphas from market model for portfolios based on global liquidity risks 3.50 5.00 3.00 4.00 2.50 3.00 2.00 1.50 1.00 1.00 0.50 t-value alpha (% ) 2.00 0.00 0.00 -0.50 (small) 10 (big) 10-1 -1.00 -2.00 -1.00 -1.50 -3.00 global liquidity net beta alpha (all) alpha (developed) alpha (emerging) t (all) t (developed) t (emerging) Panel A: Alphas from market model for portfolios based on global liquidity risk Alphas from market model for US portfolios based on US local and global liquidity risks 7.00 2.50 6.00 2.00 1.50 4.00 3.00 1.00 t-value alpha (% ) 5.00 2.00 0.50 1.00 0.00 0.00 (small) 10 (big) 10-1 US local / global liquidity risks alpha (US local) alpha (US world) t (local) t (world) Panel B: Alphas from market model for US portfolios based on US local and global liquidity risk Figure B.9: Intercepts from Market Model Regression Panel A and B plots intercepts and t-values from market model of panel A and C of Table A.18 Bar denote intercept while bullet lines are for t-values 197 0.30 0.25 Quoted Spread 0.20 0.15 0.10 0.05 QS1 QS5 20031124 20030529 20021126 20020531 20011129 20010529 20001127 20000531 19991201 19990604 19981203 19980608 19971205 19970610 19961210 19960613 19951214 19950619 19941219 19940622 19931222 19930625 19921228 19920630 19920102 0.00 Figure B.10: Daily Quoted Spread Eligible stocks are sorted into portfolios based on market capitalization at the end of previous year using NYSE ME breakpoint by K French Intraday quoted spreads are averaged each day to form daily measure of illiquidity QS5 is a daily quoted spread of largest stock portfolio, while QS1 is that of smallest stock portfolio 198 ... the effect of liquidity risk on asset pricing In the first essay, I test the liquidity- adjusted capital asset pricing model (LCAPM) of Acharya and Pedersen (2005) for 1962-2004 in Liquidity is... commonality in liquidity and the asset pricing implication of liquidity risk in three developed countries of Japan, UK and US He finds that global and industry-wide cross-country liquidity is more... market liquidity Acharya and Pedersen (2005) propose a theoretical asset pricing model that includes various liquidity risks: liquidity risk arising from the covariance of individual stock liquidity