(Luận văn) capital structure and growth option, evidence from vietnams stock market

INSTITUTE OF SOCIAL STUDIES UNIVERSITY OF ECONOMICS ERASMUS UNIVERSITY ROTTERDAM HO CHI MINH CITY THE HAGUE VIETNAM THE NETHERLAND t to SCHOOL OF ECONOMICS ng hi ep w n lo ad y th VIETNAM - THE NETHERLANDS PROGRAMME FOR M.A ju yi IN DEVELOPMENT ECONOMICS pl n ua al n va fu ll CAPITAL STRUCTURE AND GROWTH OPTION: m oi EVIDENCE FROM VIETNAM’S STOCK MARKET at nh z z k jm ht vb by Nguyen Cong Thanh gm om l.c A thesis submitted in partial fulfilment of the requirements for the degree of an Lu n va Master of Art in Development Economics ey t re Ho Chi Minh city, October 2017 t to ng hi ep w n VIETNAM - THE NETHERLANDS PROGRAMME FOR M.A IN lo ad DEVELOPMENT ECONOMICS ju y th yi pl ua al CAPITAL STRUCTURE AND GROWTH OPTION: n EVIDENCE FROM VIETNAM’S STOCK MARKET n va ll fu oi m at nh by Nguyen Cong Thanh z A thesis submitted in partial fulfilment of the requirements for the degree of z k jm ht vb om l.c gm Master of Art in Development Economics n va Dr Nguyen Vu Hong Thai an Lu Academic Supervisor: ey t re Ho Chi Minh city, October 2017 DECLARATION I hereby declare that my dissertation entitled “Capital Structure and Growth t to Option: Evidence from an Emerging Market” is the result of my own work and includes ng hi nothing which is the outcome of work done in collaboration except as declared in the ep Preface and specified in the text w My dissertation is not substantially the same as any that I have submitted, or, is n lo being concurrently submitted for a degree or diploma or other qualification in any other ad University or similar institution except as declared in the Preface and specified in the ju y th text yi I further state that no substantial part of my dissertation has already been pl ua al submitted, or, is being concurrently submitted for any such degree, diploma or other n qualification at any other University or similar institution except as declared in the n va Preface and specified in the text ll fu oi m Date: October 10th, 2017 Signature nh at Full name: Nguyen Cong Thanh z z k jm ht vb om l.c gm an Lu n va ey t re ACKNOWLEDGEMENT A successful thesis cannot be fulfilled without the unrelenting support of my t to dedicated supervisor, Dr Nguyen Vu Hong Thai, who has devoted all of his time, ng hi efforts, and energy to guiding me during the course of doing the thesis His dedication ep is what makes me feel impressive when I have a chance to work with him, his friendliness is what makes me feel comfortable when I ask him some difficult questions, w n and his enthusiasm is what makes me feel energetic when I have difficulty in solving a lo ad problem in my research All of these leave me with the most unforgettable memory and ju y th experience My purpose of this acknowledgement is to express my inexpressible gratitude to my dear supervisor There is a good teacher, there is an excellent student yi pl I would like to send my special thanks to Prof Nguyen Trong Hoai, Dr Pham al n ua Khanh Nam, Dr Truong Dang Thuy for their valuable command, guidance and support n va during the program Furthermore, I am extremely grateful to give Dr Nguyen Phuc oi m everything they have done to support me ll fu Canh and Dr Tran Thi Tuan Anh all my best thanks from bottom of my heart for at nh Besides, my thanks are given to all of the lectures who have been my knowledge guiders and the staff who have been my service supporters throughout the master z z program at University of Economics and Erasmus University Rotterdam Without their vb jm ht help, never can I have an opportunity to proceed and complete my master thesis k The last but not least, I would like to thank Dr Tran Gia Tung, Pham Thi Anh Thu gm and Tran Le Khang who have always been a pillar for me to rely on during the hardships l.c of attempting to achieve the master thesis It is their unspoken sacrifice and untiring om work that bring me more spare time to be able to reach the final destination of my an Lu progress n va ey t re ABSTRACT The research identifies the impacts of growth opportunities on capital structure of t to Vietnamese companies All of the listed firms on Ho Chi Minh Stock Exchange (HOSE) ng hi and Hanoi Stock Exchange (HNX) from 2008 to 2015 are included in the research data ep The regression methods for panel data including Pooled Ordinary Least Squared (POLS), Fixed Effects Models (FEM), Random Effects Models (REM), and System w n Generalized Method of Moments (SGMM) are used for the whole sample and two lo ad subsamples for the years after and during the financial crisis The results indicate that ju y th growth opportunities have significantly positive effects on leverage measured by both short-term and long-term debt over total assets Interestingly, the sign of this yi pl relationship remains unchanged in the pre-crisis and post-crisis In addition, this finding ua al proves the existence of Pecking Order Theory in Vietnamese firms’ capital structure n during this period because Vietnam has more financial constraints in getting the bank va n loans during the crisis, its legal systems are strictly controlled by the government or by fu ll state-owned enterprises, its law codes are too weak to protect the rights of shareholders m oi and bondholders, and most of its firms are virtually comprised of small and medium at nh enterprises z Keywords: capital structure, growth option, emerging markets, pecking order z JEL classifications: G31, G34 k jm ht vb theory, trade-off theory om l.c gm an Lu n va ey t re TABLE OF CONTENTS CHAPTER 1: INTRODUCTION t to ng hi Problem Statement and Significance of research 1.2 The Research Objectives and Research Questions ep 1.1 1.3 Structure of Thesis Design w n lo CHAPTER 2: RELATED THEORIES AND LITTERATURE REVIEW ad Overview of the related theories 2.1.1 Theory of irrelevancy 2.1.2 Trade – off theory ju y th 2.1 yi pl ua al n 2.1.2.1 Static trade-off theory va n 2.1.2.2 Dynamic Trade-off Theory ll fu Pecking order theory 10 2.1.4 The market timing theory 13 2.1.5 Theory of agency costs 14 2.2 Literature Review 15 2.2.1 Some previous review on trade-off theory and pecking order theory 15 2.2.2 Some previous studies on growth options and firm leverage 17 2.2.3 Some related literature in Vietnam 26 oi m 2.1.3 at nh z z k jm ht vb om l.c gm an Lu CHAPTER 3: METHODOLOGY AND DATA 30 3.2 Research Data 32 n Research Methodology 30 va 3.1 ey t re CHAPTER 4: RESULTS AND DISCUSSIONS 35 CHAPTER 5: CONCLUSIONS 46 5.1 Remarkable conclusions 30 5.2 Limitation and further research 32 t to ng REFERENCES 47 hi ep Appendix Table Column result 53 Appendix Table Column result 53 w n lo Appendix Table Column result 54 ad Appendix Table Column result 54 y th ju Appendix Table Column result 55 yi pl Appendix Table Column result 55 al n ua Appendix Table Column result 56 va Appendix Table Column result 56 n fu ll Appendix Table Column result 57 m oi Appendix 10 Table Column result 57 nh at Appendix 11 Table Column result 58 z z Appendix 12 Table Column result 58 ht vb Appendix 13 Table Column result 59 jm k Appendix 14 Table Column result 59 gm om l.c Appendix 15 Table Column result 60 Appendix 16 Table Column result 60 an Lu Appendix 17 Table Column result 61 n va Appendix 18 Table Column result 61 Appendix 21 Table Column result 63 Appendix 22 Table Column result 63 ey Appendix 20 Table Column result 62 t re Appendix 19 Table Column result 62 Appendix 23 Table Column result 64 Appendix 24 Table Column result 64 t to Appendix 25 Table Column result 65 ng hi Appendix 26 Table Column result 65 ep Appendix 27 Table Column result 66 w n Appendix 27 Table Column result 66 lo ad Appendix 29 Table Column result 67 y th ju Appendix 30 Table Column result 67 yi pl Appendix 31 Table Column result 68 al n ua Appendix 32 Table Column result 68 n va Appendix 33 Table Column result 69 ll fu Appendix 34 Table Column result 69 m oi Appendix 35 Table Column result 70 nh at Appendix 36 Table Column result 70 z z Appendix 37 Table 10 Column result 71 vb jm ht Appendix 38 Table 10 Column result 71 k Appendix 39 Table 10 Column result 72 gm l.c Appendix 40 Table 10 Column result 72 om Appendix 41 Table 10 Column result 73 an Lu Appendix 42 Table 10 Column result 73 Appendix 47 Table 11 Column result 76 ey Appendix 46 Table 11 Column result 75 t re Appendix 45 Table 11 Column result 75 n Appendix 44 Table 11 Column result 74 va Appendix 43 Table 11 Column result 74 Appendix 48 Table 11 Column result 76 Appendix 49 Table 12 Column result 77 t to Appendix 50 Table 12 Column result 77 ng hi Appendix 52 Table 12 Column result 78 ep Appendix 53 Table 12 Column result 78 w n Appendix 54 Table 12 Column result 79 lo ad Appendix 55 Table 12 Column result 79 y th ju Appendix 56 Table 14 Column result 88 yi pl Appendix 57 Table 14 Column result 90 al n ua Appendix 58 Table 14 Column result 92 n va Appendix 59 Table 14 Column result 94 ll fu Appendix 60 Table 14 Column result 96 m oi Appendix 61 Table 14 Column result 98 nh at Appendix 62 Table 15 Column result 100 z z Appendix 63 Table 15 Column result 101 vb jm ht Appendix 64 Table 15 Column result 102 k Appendix 65 Table 15 Column result 103 gm l.c Appendix 66 Table 15 Column result 104 om Appendix 67 Table 15 Column result 106 an Lu Appendix 68 Table 16 Column result 108 Appendix 73 Table 16 Column result 116 ey Appendix 72 Table 16 Column result 115 t re Appendix 71 Table 16 Column result 114 n Appendix 70 Table 16 Column result 112 va Appendix 69 Table 16 Column result 110 LIST OF TABLES Table Control variables definition 31 t to Table Descriptive Statistics 33 ng hi Table Correlation Matrix 34 ep Table Estimated results using POLS with Total Debt to Total Asset represented for Leverage 35 w n Table Estimated results using POLS with Short-term Debt to Total Asset represented for Leverage 36 lo ad ju y th Table Estimated results using POLS with Long-term Debt to Total Asset represented for Leverage 37 yi pl Table Estimated results using FEM with Total Debt to Total Asset represented for Leverage 38 ua al n Table Estimated results using FEM with Short-term Debt to Total Asset represented for Leverage 39 n va ll fu Table Estimated results using FEM with Long-term Debt to Total Asset represented for Leverage 40 oi m at nh Table 10 Estimated results using REM with Total Debt to Total Asset represented for Leverage 41 z Table 11 Estimated results using REM with Short-term Debt to Total Asset represented for Leverage 42 z ht vb k jm Table 12 Estimated results using FEM with Long-term Debt to Total Asset represented for Leverage 43 gm om l.c Table 13 Parameters Test, Hausman Test and Breusch & Pagan Lagrange multiplier Test 44 an Lu Table 14 Estimated results using SGMM with Total Debt to Total Asset represented for Leverage 46 n ey t re Table 16 Estimated results using SGMM with Long-term Debt to Total Asset represented for Leverage 50 va Table 15 Estimated results using SGMM with Short-term Debt to Total Asset represented for Leverage 48 Appendix 64 Table 15 Column result Dynamic panel-data estimation, two-step system GMM t to ng hi Group variable: cross Time variable : years Number of instruments = 35 F(7, 231) = 47.90 Prob > F = 0.000 ep Number of obs Number of groups Obs per group: avg max w n stdta Coef stdta L1 = = = = = 1392 232 6.00 t P>|t| [95% Conf Interval] 4786399 0952647 5.02 0.000 2909411 6663386 0438157 0154698 -.0577287 1.233293 0372793 1.017146 -.8725102 0076926 0042209 0896032 4088802 0098958 1671671 1447436 5.70 3.67 -0.64 3.02 3.77 6.08 -6.03 0.000 0.000 0.520 0.003 0.000 0.000 0.000 0286592 0071535 -.2342726 427682 0177819 6877785 -1.157697 0589722 0237862 1188152 2.038905 0567768 1.346513 -.5873238 lo Std Err ad ju y th mb mb_crisis0 tang ndts size med_lev_industry _cons yi pl ua al n Instruments for first differences equation GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/3).mb L(2/7).L.stdta Instruments for levels equation Standard _cons GMM-type (missing=0, separate instruments for each period unless collapsed) D.mb DL.L.stdta n va ll fu oi m at nh in in in in first first first first differences: differences: differences: differences: z z z z = = = = -5.89 -0.93 -0.93 0.92 > > > > z z z z = = = = 0.000 0.352 0.352 0.357 jm Prob > chi2 = 0.139 k Prob > chi2 = 0.476 chi2 = chi2 = 0.366 0.512 ey t re 0.377 0.541 n chi2 = chi2 = va 0.704 0.207 an Lu chi2 = chi2 = om l.c Difference-in-Hansen tests of exogeneity of instrument subsets: GMM instruments for levels Hansen test excluding group: chi2(17) = 13.47 Prob > Difference (null H = exogenous): chi2(10) = 13.31 Prob > gmm(L.stdta, lag(2 7)) Hansen test excluding group: chi2(13) = 13.96 Prob > Difference (null H = exogenous): chi2(14) = 12.82 Prob > gmm(mb, lag(1 3)) Hansen test excluding group: chi2(7) = 7.63 Prob > Difference (null H = exogenous): chi2(20) = 19.15 Prob > gm overid restrictions: chi2(27) = 34.99 but not weakened by many instruments.) overid restrictions: chi2(27) = 26.78 weakened by many instruments.) Pr Pr Pr Pr ht AR(1) AR(2) AR(3) AR(4) vb Sargan test of (Not robust, Hansen test of (Robust, but for for for for z test test test test z Arellano-Bond Arellano-Bond Arellano-Bond Arellano-Bond 105 Appendix 65 Table 15 Column result Dynamic panel-data estimation, two-step system GMM t to ng hi Group variable: cross Time variable : years Number of instruments = 40 F(6, 231) = 33.06 Prob > F = 0.000 ep Number of obs Number of groups Obs per group: avg max w n stdta Coef stdta L1 = = = = = 1392 232 6.00 t P>|t| [95% Conf Interval] 4862784 0772898 6.29 0.000 3339954 6385615 -.2400855 -.1897228 7188523 0343531 9378368 -.6022447 0507047 0664028 3458213 0124732 1350507 1620869 -4.73 -2.86 2.08 2.75 6.94 -3.72 0.000 0.005 0.039 0.006 0.000 0.000 -.3399882 -.3205553 0374851 0097774 6717482 -.9216023 -.1401827 -.0588903 1.400219 0589288 1.203925 -.2828871 lo Std Err ad ju y th invepx_mb tang ndts size med_lev_industry _cons yi pl ua al n Instruments for first differences equation GMM-type (missing=0, separate instruments for each period unless collapsed) va n L.invepx_mb L2.L.stdta Instruments for levels equation Standard _cons GMM-type (missing=0, separate instruments for each period unless collapsed) ll fu oi m nh in in in in first first first first differences: differences: differences: differences: z z z z = = = = -6.20 -1.11 -0.88 1.08 z z z z = = = = 0.000 0.266 0.379 0.278 k Prob > chi2 = 0.000 gm Prob > chi2 = 0.164 0.110 0.547 chi2 = chi2 = 0.286 0.157 n chi2 = chi2 = va 0.062 0.592 an Lu chi2 = chi2 = om l.c ey t re Difference-in-Hansen tests of exogeneity of instrument subsets: GMM instruments for levels Hansen test excluding group: chi2(17) = 26.73 Prob > Difference (null H = exogenous): chi2(16) = 14.09 Prob > gmm(L.stdta, lag(2 2)) Hansen test excluding group: chi2(25) = 33.92 Prob > Difference (null H = exogenous): chi2(8) = 6.90 Prob > gmm(invepx_mb, lag(1 1)) Hansen test excluding group: chi2(22) = 25.23 Prob > Difference (null H = exogenous): chi2(11) = 15.59 Prob > 106 > > > > jm overid restrictions: chi2(33) = 100.85 but not weakened by many instruments.) overid restrictions: chi2(33) = 40.82 weakened by many instruments.) Pr Pr Pr Pr ht AR(1) AR(2) AR(3) AR(4) vb Sargan test of (Not robust, Hansen test of (Robust, but for for for for z test test test test z Arellano-Bond Arellano-Bond Arellano-Bond Arellano-Bond at D.invepx_mb DL.L.stdta Appendix 66 Table 15 Column result Dynamic panel-data estimation, two-step system GMM t to ng hi Group variable: cross Time variable : years Number of instruments = 41 F(7, 231) = 29.24 Prob > F = 0.000 ep Number of obs Number of groups Obs per group: avg max w n stdta Coef stdta L1 = = = = = 1392 232 6.00 t P>|t| [95% Conf Interval] 4001218 0731849 5.47 0.000 2559267 544317 -.2088195 -.0177889 -.2002975 9122292 0354028 1.006543 -.6298646 0452453 0097355 0707424 3795663 0124716 1438531 1623067 -4.62 -1.83 -2.83 2.40 2.84 7.00 -3.88 0.000 0.069 0.005 0.017 0.005 0.000 0.000 -.2979657 -.0369708 -.3396804 1643748 0108301 7231109 -.9496553 -.1196733 0013929 -.0609147 1.660084 0599755 1.289974 -.3100739 lo Std Err ad ju y th yi pl invepx_mb invepx_mb_crisis1 tang ndts size med_lev_industry _cons n ua al n va ll fu oi m at nh z z k jm ht vb om l.c gm an Lu n va ey t re 107 t to ng hi ep Instruments for first differences equation GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/3).tang L.invepx_mb_crisis1 L.invepx_mb L2.L.stdta Instruments for levels equation Standard _cons GMM-type (missing=0, separate instruments for each period unless collapsed) D.tang D.invepx_mb_crisis1 D.invepx_mb DL.L.stdta w n lo test test test test ad Arellano-Bond Arellano-Bond Arellano-Bond Arellano-Bond AR(1) AR(2) AR(3) AR(4) in in in in first first first first differences: differences: differences: differences: z z z z = = = = -6.35 -1.01 -1.01 0.84 ju y th overid restrictions: chi2(33) = 99.60 but not weakened by many instruments.) overid restrictions: chi2(33) = 37.08 weakened by many instruments.) yi pl Pr Pr Pr Pr > > > > z z z z = = = = 0.000 0.311 0.314 0.399 Prob > chi2 = 0.000 Prob > chi2 = 0.286 ua al Sargan test of (Not robust, Hansen test of (Robust, but for for for for n Difference-in-Hansen tests of exogeneity of instrument subsets: GMM instruments for levels Hansen test excluding group: chi2(16) = 20.37 Prob > Difference (null H = exogenous): chi2(17) = 16.71 Prob > gmm(L.stdta, lag(2 2)) Hansen test excluding group: chi2(25) = 29.26 Prob > Difference (null H = exogenous): chi2(8) = 7.82 Prob > gmm(invepx_mb, lag(1 1)) Hansen test excluding group: chi2(28) = 28.09 Prob > Difference (null H = exogenous): chi2(5) = 8.99 Prob > gmm(invepx_mb_crisis1, lag(1 1)) Hansen test excluding group: chi2(26) = 36.98 Prob > Difference (null H = exogenous): chi2(7) = 0.10 Prob > gmm(tang, lag(1 3)) Hansen test excluding group: chi2(13) = 12.62 Prob > Difference (null H = exogenous): chi2(20) = 24.46 Prob > n va ll fu oi m at nh z z 0.204 0.474 chi2 = chi2 = 0.253 0.452 chi2 = chi2 = 0.460 0.110 chi2 = chi2 = 0.075 1.000 chi2 = chi2 = 0.478 0.223 k jm ht vb chi2 = chi2 = om l.c gm an Lu n va ey t re 108 Appendix 67 Table 15 Column result Dynamic panel-data estimation, two-step system GMM t to ng hi Group variable: cross Time variable : years Number of instruments = 41 F(7, 231) = 29.24 Prob > F = 0.000 ep Number of obs Number of groups Obs per group: avg max w n stdta Coef stdta L1 = = = = = 1392 232 6.00 t P>|t| [95% Conf Interval] 4001218 0731849 5.47 0.000 2559267 544317 -.2266084 0177889 -.2002975 9122292 0354028 1.006543 -.6298646 0459438 0097355 0707424 3795663 0124716 1438531 1623067 -4.93 1.83 -2.83 2.40 2.84 7.00 -3.88 0.000 0.069 0.005 0.017 0.005 0.000 0.000 -.3171309 -.0013929 -.3396804 1643748 0108301 7231109 -.9496553 -.136086 0369708 -.0609147 1.660084 0599755 1.289974 -.3100739 lo Std Err ad ju y th yi pl invepx_mb invepx_mb_crisis0 tang ndts size med_lev_industry _cons n ua al n va ll fu oi m at nh z z k jm ht vb om l.c gm an Lu n va ey t re 109 t to ng hi ep Instruments for first differences equation GMM-type (missing=0, separate instruments for each period unless collapsed) L(1/3).tang L.invepx_mb_crisis1 L.invepx_mb L2.L.stdta Instruments for levels equation Standard _cons GMM-type (missing=0, separate instruments for each period unless collapsed) D.tang D.invepx_mb_crisis1 D.invepx_mb DL.L.stdta w n lo test test test test ad Arellano-Bond Arellano-Bond Arellano-Bond Arellano-Bond AR(1) AR(2) AR(3) AR(4) in in in in first first first first differences: differences: differences: differences: z z z z = = = = -6.35 -1.01 -1.01 0.84 ju y th overid restrictions: chi2(33) = 99.60 but not weakened by many instruments.) overid restrictions: chi2(33) = 37.08 weakened by many instruments.) yi pl Pr Pr Pr Pr > > > > z z z z = = = = 0.000 0.311 0.314 0.399 Prob > chi2 = 0.000 Prob > chi2 = 0.286 ua al Sargan test of (Not robust, Hansen test of (Robust, but for for for for n Difference-in-Hansen tests of exogeneity of instrument subsets: GMM instruments for levels Hansen test excluding group: chi2(16) = 20.37 Prob > Difference (null H = exogenous): chi2(17) = 16.71 Prob > gmm(L.stdta, lag(2 2)) Hansen test excluding group: chi2(25) = 29.26 Prob > Difference (null H = exogenous): chi2(8) = 7.82 Prob > gmm(invepx_mb, lag(1 1)) Hansen test excluding group: chi2(28) = 28.09 Prob > Difference (null H = exogenous): chi2(5) = 8.99 Prob > gmm(invepx_mb_crisis1, lag(1 1)) Hansen test excluding group: chi2(26) = 36.98 Prob > Difference (null H = exogenous): chi2(7) = 0.10 Prob > gmm(tang, lag(1 3)) Hansen test excluding group: chi2(13) = 12.62 Prob > Difference (null H = exogenous): chi2(20) = 24.46 Prob > n va ll fu oi m at nh z z 0.204 0.474 chi2 = chi2 = 0.253 0.452 chi2 = chi2 = 0.460 0.110 chi2 = chi2 = 0.075 1.000 chi2 = chi2 = 0.478 0.223 k jm ht vb chi2 = chi2 = om l.c gm an Lu n va ey t re 110 Appendix 68 Table 16 Column result Dynamic panel-data estimation, two-step system GMM t to ng hi Group variable: cross Time variable : years Number of instruments = 30 F(6, 231) = 50.44 Prob > F = 0.000 ep Number of obs Number of groups Obs per group: avg max w n ltdta Coef ltdta L1 .6490922 0619861 0100928 -.026854 -.0527756 0143191 0058665 -.1642965 003556 0670101 2432154 0048796 0202382 0585149 Std Err t = = = = = 1392 232 6.00 [95% Conf Interval] 10.47 0.000 5269618 7712226 2.84 -0.40 -0.22 2.93 0.29 -2.81 0.005 0.689 0.828 0.004 0.772 0.005 0030865 -.1588832 -.5319796 0047048 -.0340085 -.2795877 017099 1051751 4264284 0239334 0457415 -.0490054 lo P>|t| ad ju y th mb tang ndts size med_lev_industry _cons yi pl n ua al n va ll fu oi m at nh z z k jm ht vb om l.c gm an Lu n va ey t re 111 t to ng hi ep Instruments for first differences equation Standard D.med_lev_industry GMM-type (missing=0, separate instruments for each period unless collapsed) L3.mb L(1/2).mb L2.L.ltdta Instruments for levels equation Standard med_lev_industry _cons GMM-type (missing=0, separate instruments for each period unless collapsed) DL2.mb D.mb DL.L.ltdta w n lo ad test test test test AR(1) AR(2) AR(3) AR(4) ju in in in in first first first first differences: differences: differences: differences: z z z z = = = = -5.01 -1.91 0.69 0.51 yi overid restrictions: chi2(23) = 63.66 but not weakened by many instruments.) overid restrictions: chi2(23) = 24.06 weakened by many instruments.) pl Pr Pr Pr Pr > > > > z z z z = = = = 0.000 0.056 0.491 0.608 Prob > chi2 = 0.000 Prob > chi2 = 0.400 n ua al Sargan test of (Not robust, Hansen test of (Robust, but for for for for y th Arellano-Bond Arellano-Bond Arellano-Bond Arellano-Bond n va Difference-in-Hansen tests of exogeneity of instrument subsets: GMM instruments for levels Hansen test excluding group: chi2(12) = 17.03 Prob > Difference (null H = exogenous): chi2(11) = 7.03 Prob > gmm(L.ltdta, lag(2 2)) Hansen test excluding group: chi2(16) = 16.13 Prob > Difference (null H = exogenous): chi2(7) = 7.93 Prob > gmm(mb, lag(1 2)) Hansen test excluding group: chi2(9) = 17.00 Prob > Difference (null H = exogenous): chi2(14) = 7.06 Prob > gmm(mb, lag(3 3)) Hansen test excluding group: chi2(16) = 20.09 Prob > Difference (null H = exogenous): chi2(7) = 3.97 Prob > iv(med_lev_industry) Hansen test excluding group: chi2(22) = 23.97 Prob > Difference (null H = exogenous): chi2(1) = 0.09 Prob > ll fu oi m at nh z z 0.148 0.797 chi2 = chi2 = 0.444 0.339 chi2 = chi2 = 0.049 0.932 chi2 = chi2 = 0.216 0.783 ht vb chi2 = chi2 = k jm chi2 = chi2 = 0.349 0.767 om l.c gm an Lu n va ey t re 112 Appendix 69 Table 16 Column result Dynamic panel-data estimation, two-step system GMM t to ng hi Group variable: cross Time variable : years Number of instruments = 30 F(7, 231) = 36.54 Prob > F = 0.000 ep Number of obs Number of groups Obs per group: avg max w n ltdta Coef ltdta L1 = = = = = 1392 232 6.00 t P>|t| [95% Conf Interval] 5385319 0715368 7.53 0.000 3975839 6794799 -.0005827 0119255 0547116 -.2864284 0181696 0129483 -.214189 0051818 0022343 0806362 2819495 0052189 0253611 0621824 -0.11 5.34 0.68 -1.02 3.48 0.51 -3.44 0.911 0.000 0.498 0.311 0.001 0.610 0.001 -.0107924 0075232 -.1041649 -.8419497 0078868 -.0370204 -.3367062 0096269 0163278 2135881 2690929 0284523 062917 -.0916718 lo Std Err ad ju y th mb mb_crisis1 tang ndts size med_lev_industry _cons yi pl n ua al n va ll fu oi m at nh z z k jm ht vb om l.c gm an Lu n va ey t re 113 t to ng hi ep Instruments for first differences equation Standard D.med_lev_industry GMM-type (missing=0, separate instruments for each period unless collapsed) L3.mb L(1/2).mb L2.L.ltdta Instruments for levels equation Standard med_lev_industry _cons GMM-type (missing=0, separate instruments for each period unless collapsed) DL2.mb D.mb DL.L.ltdta w n lo ad test test test test AR(1) AR(2) AR(3) AR(4) ju in in in in first first first first differences: differences: differences: differences: z z z z = = = = -4.48 -1.53 0.45 0.78 yi overid restrictions: chi2(22) = 52.63 but not weakened by many instruments.) overid restrictions: chi2(22) = 18.32 weakened by many instruments.) pl Pr Pr Pr Pr > > > > z z z z = = = = 0.000 0.126 0.656 0.434 Prob > chi2 = 0.000 Prob > chi2 = 0.687 n ua al Sargan test of (Not robust, Hansen test of (Robust, but for for for for y th Arellano-Bond Arellano-Bond Arellano-Bond Arellano-Bond n va Difference-in-Hansen tests of exogeneity of instrument subsets: GMM instruments for levels Hansen test excluding group: chi2(11) = 14.50 Prob > Difference (null H = exogenous): chi2(11) = 3.82 Prob > gmm(L.ltdta, lag(2 2)) Hansen test excluding group: chi2(15) = 15.22 Prob > Difference (null H = exogenous): chi2(7) = 3.11 Prob > gmm(mb, lag(1 2)) Hansen test excluding group: chi2(8) = 14.37 Prob > Difference (null H = exogenous): chi2(14) = 3.96 Prob > gmm(mb, lag(3 3)) Hansen test excluding group: chi2(15) = 14.91 Prob > Difference (null H = exogenous): chi2(7) = 3.41 Prob > iv(med_lev_industry) Hansen test excluding group: chi2(21) = 18.28 Prob > Difference (null H = exogenous): chi2(1) = 0.04 Prob > ll fu oi m at nh z z 0.206 0.975 chi2 = chi2 = 0.436 0.875 chi2 = chi2 = 0.073 0.996 chi2 = chi2 = 0.458 0.845 ht vb chi2 = chi2 = k jm chi2 = chi2 = 0.631 0.842 om l.c gm an Lu n va ey t re 114 Appendix 70 Table 16 Column result Dynamic panel-data estimation, two-step system GMM t to ng hi Group variable: cross Time variable : years Number of instruments = 30 F(7, 231) = 36.54 Prob > F = 0.000 ep Number of obs Number of groups Obs per group: avg max w n ltdta Coef ltdta L1 = = = = = 1392 232 6.00 t P>|t| [95% Conf Interval] 5385319 0715368 7.53 0.000 3975839 6794799 0113428 -.0119255 0547116 -.2864284 0181696 0129483 -.214189 0041838 0022343 0806362 2819495 0052189 0253611 0621824 2.71 -5.34 0.68 -1.02 3.48 0.51 -3.44 0.007 0.000 0.498 0.311 0.001 0.610 0.001 0030994 -.0163278 -.1041649 -.8419497 0078868 -.0370204 -.3367062 0195861 -.0075232 2135881 2690929 0284523 062917 -.0916718 lo Std Err ad ju y th mb mb_crisis0 tang ndts size med_lev_industry _cons yi pl n ua al n va ll fu oi m at nh z z k jm ht vb om l.c gm an Lu n va ey t re 115 t to ng hi ep Instruments for first differences equation Standard D.med_lev_industry GMM-type (missing=0, separate instruments for each period unless collapsed) L3.mb L(1/2).mb L2.L.ltdta Instruments for levels equation Standard med_lev_industry _cons GMM-type (missing=0, separate instruments for each period unless collapsed) DL2.mb D.mb DL.L.ltdta w n lo ad test test test test AR(1) AR(2) AR(3) AR(4) ju in in in in first first first first differences: differences: differences: differences: z z z z = = = = -4.48 -1.53 0.45 0.78 yi overid restrictions: chi2(22) = 52.63 but not weakened by many instruments.) overid restrictions: chi2(22) = 18.32 weakened by many instruments.) pl Pr Pr Pr Pr > > > > z z z z = = = = 0.000 0.126 0.656 0.434 Prob > chi2 = 0.000 Prob > chi2 = 0.687 n ua al Sargan test of (Not robust, Hansen test of (Robust, but for for for for y th Arellano-Bond Arellano-Bond Arellano-Bond Arellano-Bond n va Difference-in-Hansen tests of exogeneity of instrument subsets: GMM instruments for levels Hansen test excluding group: chi2(11) = 14.50 Prob > Difference (null H = exogenous): chi2(11) = 3.82 Prob > gmm(L.ltdta, lag(2 2)) Hansen test excluding group: chi2(15) = 15.22 Prob > Difference (null H = exogenous): chi2(7) = 3.11 Prob > gmm(mb, lag(1 2)) Hansen test excluding group: chi2(8) = 14.37 Prob > Difference (null H = exogenous): chi2(14) = 3.96 Prob > gmm(mb, lag(3 3)) Hansen test excluding group: chi2(15) = 14.91 Prob > Difference (null H = exogenous): chi2(7) = 3.41 Prob > iv(med_lev_industry) Hansen test excluding group: chi2(21) = 18.28 Prob > Difference (null H = exogenous): chi2(1) = 0.04 Prob > ll fu oi m at nh z z 0.206 0.975 chi2 = chi2 = 0.436 0.875 chi2 = chi2 = 0.073 0.996 chi2 = chi2 = 0.458 0.845 ht vb chi2 = chi2 = k jm chi2 = chi2 = 0.631 0.842 om l.c gm an Lu n va ey t re 116 Appendix 71 Table 16 Column result Dynamic panel-data estimation, two-step system GMM t to ng hi Group variable: cross Time variable : years Number of instruments = 21 F(6, 231) = 26.34 Prob > F = 0.000 ep Number of obs Number of groups Obs per group: avg max w n ltdta Coef ltdta L1 = = = = = 1392 232 6.00 t P>|t| [95% Conf Interval] 6345365 0705779 8.99 0.000 4954778 7735952 -.0706623 0372152 1367878 0176321 0071242 -.1838426 0271635 0860895 3417998 0052844 0233178 0666755 -2.60 0.43 0.40 3.34 0.31 -2.76 0.010 0.666 0.689 0.001 0.760 0.006 -.1241823 -.1324058 -.5366559 0072202 -.0388185 -.3152125 -.0171423 2068363 8102314 0280439 053067 -.0524726 lo Std Err ad ju y th invepx_mb tang ndts size med_lev_industry _cons yi pl ua al n Instruments for first differences equation Standard D.med_lev_industry GMM-type (missing=0, separate instruments for each period unless collapsed) L.invepx_mb L2.L.ltdta Instruments for levels equation Standard med_lev_industry _cons GMM-type (missing=0, separate instruments for each period unless collapsed) D.invepx_mb DL.L.ltdta n va ll fu oi m at nh z z AR(1) AR(2) AR(3) AR(4) in in in in first first first first differences: differences: differences: differences: z z z z = = = = -4.76 -2.10 0.92 0.47 Pr Pr Pr Pr > > > > z z z z = = = = 0.000 0.036 0.359 0.640 k jm for for for for ht test test test test vb Arellano-Bond Arellano-Bond Arellano-Bond Arellano-Bond Prob > chi2 = Prob > chi2 = chi2 = chi2 = 0.375 0.222 chi2 = chi2 = 0.196 0.628 ey 0.828 0.071 t re chi2 = chi2 = n 0.043 0.682 va chi2 = chi2 = an Lu 117 0.241 om Difference-in-Hansen tests of exogeneity of instrument subsets: GMM instruments for levels Hansen test excluding group: chi2(4) = 9.85 Prob > Difference (null H = exogenous): chi2(10) = 7.45 Prob > gmm(L.ltdta, lag(2 2)) Hansen test excluding group: chi2(6) = 2.85 Prob > Difference (null H = exogenous): chi2(8) = 14.45 Prob > gmm(invepx_mb, lag(1 1)) Hansen test excluding group: chi2(3) = 3.11 Prob > Difference (null H = exogenous): chi2(11) = 14.19 Prob > iv(med_lev_industry) Hansen test excluding group: chi2(13) = 17.07 Prob > Difference (null H = exogenous): chi2(1) = 0.23 Prob > 0.008 l.c overid restrictions: chi2(14) = 29.77 but not weakened by many instruments.) overid restrictions: chi2(14) = 17.30 weakened by many instruments.) gm Sargan test of (Not robust, Hansen test of (Robust, but Appendix 72 Table 16 Column result Dynamic panel-data estimation, two-step system GMM t to ng hi Group variable: cross Time variable : years Number of instruments = 21 F(7, 231) = 23.13 Prob > F = 0.000 ep Number of obs Number of groups Obs per group: avg max w n ltdta Coef ltdta L1 = = = = = 1392 232 6.00 t P>|t| [95% Conf Interval] 5868907 0709622 8.27 0.000 4470749 7267066 -.0688831 0141723 0541329 0433571 0227303 000868 -.2478866 0268661 0050929 0820308 3160267 005498 0251321 0700461 -2.56 2.78 0.66 0.14 4.13 0.03 -3.54 0.011 0.006 0.510 0.891 0.000 0.972 0.000 -.1218169 0041378 -.1074912 -.5793061 0118977 -.0486494 -.3858975 -.0159492 0242067 2157571 6660204 0335629 0503855 -.1098758 lo Std Err ad ju y th invepx_mb invepx_mb_crisis1 tang ndts size med_lev_industry _cons yi pl ua al n Instruments for first differences equation Standard D.med_lev_industry GMM-type (missing=0, separate instruments for each period unless collapsed) L.invepx_mb L2.L.ltdta Instruments for levels equation Standard med_lev_industry _cons GMM-type (missing=0, separate instruments for each period unless collapsed) D.invepx_mb DL.L.ltdta n va ll fu oi m at nh z z in in in in first first first first differences: differences: differences: differences: z z z z = = = = -4.57 -1.93 0.84 0.55 z z z z = = = = Prob > chi2 = 0.000 0.054 0.398 0.583 0.015 l.c Prob > chi2 = 0.603 om chi2 = chi2 = 0.212 0.715 chi2 = chi2 = 0.540 0.641 ey 0.927 0.285 t re chi2 = chi2 = n 0.227 0.749 va chi2 = chi2 = an Lu Difference-in-Hansen tests of exogeneity of instrument subsets: GMM instruments for levels Hansen test excluding group: chi2(3) = 4.34 Prob > Difference (null H = exogenous): chi2(10) = 6.75 Prob > gmm(L.ltdta, lag(2 2)) Hansen test excluding group: chi2(5) = 1.37 Prob > Difference (null H = exogenous): chi2(8) = 9.72 Prob > gmm(invepx_mb, lag(1 1)) Hansen test excluding group: chi2(2) = 3.11 Prob > Difference (null H = exogenous): chi2(11) = 7.98 Prob > iv(med_lev_industry) Hansen test excluding group: chi2(12) = 10.87 Prob > Difference (null H = exogenous): chi2(1) = 0.22 Prob > 118 > > > > gm overid restrictions: chi2(13) = 26.49 but not weakened by many instruments.) overid restrictions: chi2(13) = 11.09 weakened by many instruments.) Pr Pr Pr Pr k AR(1) AR(2) AR(3) AR(4) jm Sargan test of (Not robust, Hansen test of (Robust, but for for for for ht test test test test vb Arellano-Bond Arellano-Bond Arellano-Bond Arellano-Bond Appendix 73 Table 16 Column result Dynamic panel-data estimation, two-step system GMM t to ng hi Group variable: cross Time variable : years Number of instruments = 21 F(7, 231) = 23.13 Prob > F = 0.000 ep Number of obs Number of groups Obs per group: avg max w n ltdta Coef ltdta L1 = = = = = 1392 232 6.00 t P>|t| [95% Conf Interval] 5868907 0709622 8.27 0.000 4470749 7267066 -.0547108 -.0141723 0541329 0433571 0227303 000868 -.2478866 0278744 0050929 0820308 3160267 005498 0251321 0700461 -1.96 -2.78 0.66 0.14 4.13 0.03 -3.54 0.051 0.006 0.510 0.891 0.000 0.972 0.000 -.1096313 -.0242067 -.1074912 -.5793061 0118977 -.0486494 -.3858975 0002097 -.0041378 2157571 6660204 0335629 0503855 -.1098758 lo Std Err ad ju y th invepx_mb invepx_mb_crisis0 tang ndts size med_lev_industry _cons yi pl ua al n Instruments for first differences equation Standard D.med_lev_industry GMM-type (missing=0, separate instruments for each period unless collapsed) L.invepx_mb L2.L.ltdta Instruments for levels equation Standard med_lev_industry _cons GMM-type (missing=0, separate instruments for each period unless collapsed) D.invepx_mb DL.L.ltdta n va ll fu oi m at nh z z in in in in first first first first differences: differences: differences: differences: z z z z = = = = -4.57 -1.93 0.84 0.55 z z z z = = = = Prob > chi2 = 0.000 0.054 0.398 0.583 0.015 l.c Prob > chi2 = 0.603 om chi2 = chi2 = 0.212 0.715 chi2 = chi2 = 0.540 0.641 ey 0.927 0.285 t re chi2 = chi2 = n 0.227 0.749 va chi2 = chi2 = an Lu Difference-in-Hansen tests of exogeneity of instrument subsets: GMM instruments for levels Hansen test excluding group: chi2(3) = 4.34 Prob > Difference (null H = exogenous): chi2(10) = 6.75 Prob > gmm(L.ltdta, lag(2 2)) Hansen test excluding group: chi2(5) = 1.37 Prob > Difference (null H = exogenous): chi2(8) = 9.72 Prob > gmm(invepx_mb, lag(1 1)) Hansen test excluding group: chi2(2) = 3.11 Prob > Difference (null H = exogenous): chi2(11) = 7.98 Prob > iv(med_lev_industry) Hansen test excluding group: chi2(12) = 10.87 Prob > Difference (null H = exogenous): chi2(1) = 0.22 Prob > 119 > > > > gm overid restrictions: chi2(13) = 26.49 but not weakened by many instruments.) overid restrictions: chi2(13) = 11.09 weakened by many instruments.) Pr Pr Pr Pr k AR(1) AR(2) AR(3) AR(4) jm Sargan test of (Not robust, Hansen test of (Robust, but for for for for ht test test test test vb Arellano-Bond Arellano-Bond Arellano-Bond Arellano-Bond