: ; (2) ; (3) ( 15 ) (BE) (NI),,,, CSMAR,,,,,,,,,,,,, : ; Ohlson (1995) ; ;,,,, Q,, ( ) Q (MΠB),, Q (MΠB) Q,,, (2001) Q,,, Q 74

2008 3 3 :,,,,,,,,,,,,, :,,,,,,,, (MVE) ( EPS), (MΠB),,,, : (1) 3,, :610074, :yangd @swufe. edu. cn ;,, :40114021 @swufe. edu. cn ;, Baruch, :Jianming - Ye @ baruch. cuny. edu ( 70672112) ( NCET20620816), 211 151, 73

: ; (2) ; (3) ( 15 ) (BE) (NI),,,, CSMAR,,,,,,,,,,,,, : ; Ohlson (1995) ; ;,,,, Q,, ( ) Q (MΠB),, Q (MΠB) Q,,, (2001) Q,,, Q 74

2008 3 Q, Q (2002) Berger Ofek (1995) (2005), (2003), (2003), (2003), (2006) Q, : Q = ( 3 + 3 + )Π, Q,, Fama French (1992) (BΠM), MΠB (2006), (2004), (2004), (2005),, (2004) t 6 t ; (2004), (2004), (2005),, (2006), IPO, IPO ( ), (2006), (2005), ; (2004), (1999), (2003), (2003), (2004), (2002), (1998), (1999), (2001), (2005), (2005), (2005), (2006) (2004) ; (2005) Π ; (2001),, Altman (1968) Z (2002) (2005) Altman (1968) Z, X 4, Z,,,, 75

: ( ),,, 11 : = MVE,, MVE, : MVE0 = Price 3 Shares = Price 3 # TS + Price 3 # NTS # TS # NTS,,,, : : MVE0,, MV,, 21 : = ( ), MVE :,BPS, BVEΠ( # TS + # NTS) MVE1 = Price 3 # TS + BPS 3 # NTS (1), (1), : (1), : : MVE1 = a + b 3 BVE + c 3 NI P 3 # TS + BPS 3 # NTS = a + b 3 BVE + c 3 NI P 3 # TS = a + (b - %NTS = # NTSΠ( # NTS + # TS) %NTS) 3 BVE + c 3 NI,,,NI MVE c, NI NI : :,,,,, 31 : = 15 3, (15 ), MVE : (2), : 76 MVE2 = Price 3 # TS + EPS 3 15 3 # NTS (2)

2008 3 Price 3 # TS + %NTS 3 NI 3 15 = a + b 3 BVE + c 3 NI : Price 3 # TS = a + b 3 BVE + (c - %NTS 3 15) 3 NI,,BVE BVE MVE,, BVE : :,,,,, ( ),,, Price of NTS = f (z) 3 (Price of TS) f (z) z : True MVE = (Price of TS) 3 # TS + (Price of NTS) 3 # NTS = Price 3 ( # TS + # NTS 3 f (z) ) = Price 3 # S 3 (1 - ntr 3 (1 - f (z) ) ) = MVE0 3 (1 - ntr 3 (1 - f (z) ) ) (3),MVE0 = Price 3 Shares,Price,Shares,, :True MVE = a + b 3 BVE + c 3 NI, (3), : MVE 0 3 K = a + b 3 BVE + c 3 NI (4) K= 1 - ntr 3 (1 - f (z) ) f (z) = d, d (4), P 3 # TS f (z) 3 P 3 # NTS, D, # TS + D # NTS - D, ( # TS + D)Π( # NTS - D), : ( # TS + D)Π( # NTS - D) = P 3 # TSΠ[f 3 P 3 # NTS], : DΠ# TS = 1ΠK - 1 = MVE0 Π(True MVE) - 1 10 10 3 DΠ# TS, 10 3 DΠ# TS = 10 3 (1ΠK - 1), f = 015, ntr = 015, K= 1 - ntr 3 (1 - f) = 1-0125 = 0175, 10 (1ΠK- 1) = 3133 f = 015, ntr = 0175, K= 01625, 10 (1ΠK- 1) = 6,,, z,,,, 77

: (4) BVE, : MVE 0 ΠBVE 3 (1 - ntr 3 (1 - f) = aπbve + b + c 3 ROE +, 0 < f < = 1, N(0, 2 ) : - Likelihood = n 3 [log(s 2 ) - 2 log(1 - ntr 3 (1 - d) ) ] s 2 2 = (1Πn) ( [MVE0ΠBE] 3 (1 - ntr 3 (1 - d) ) - [aπbve + b + c 3 ROE]), : MVE 0 ΠBVE = [aπbve + b + c 3 ROE]Π(1 - ntr 3 (1 - d) ) + N (0, 2 ), ( ) 11, UEPS, : CAR = b 3 UEPSΠPrice,,, EPS,, f, # TS + f 3 # NTS, EPS : True EPS = NIΠ( ) = NI Π( # TS + # NTS 3 f (ntr) ) = NIΠ[ # S 3 (1 - ntr 3 (1 - f ) ] = EPS0ΠK # S = # TS + # NTS,EPS0 = NIΠ( # TS + # NTS), K = 1 - ntr 3 (1 - f) : CAR = b 3 UEPS0Π(Price 3 K) K= 1 - ntr 3 (1 - f) 1ΠK = 1Π(1 - ntr 3 (1 - f) ), : CAR = b 3 UEPS0ΠPrice + g 3 (ntr 3 UEPS0ΠPrice),g > 0, : :, 21, : True MVE = MVE0 3 ( 1 - ntr 3 (1 - f) ) MVE 0 ΠBVE 3 (1 - ntr 3 (1 - f (ntr) ) ), (ntr) MVE 0 ΠBVE K = 1 - ntr 3 (1 - f (ntr) ), ntr MVE 0 ΠBE, f < 1,, 78

2008 3 K= 1 - ntr 3 (1 - f) ntr,, ntr, : ;,, ( ) CSMAR2005, 1995, 1995 2005,,,,,, (ROE) - 2 013,,,,,, : 1A Mean stdev 10th Percentile Median 90th Percentile BVE 9651600 27691384 2011389 5581000 17411233 ntr 01660 01132 01492 01667 01836 MVE 0 41440 21984 11667 31666 81164 MVE 1 21114 11085 11178 11816 31357 MVE 2 11982 21306 01617 11950 31863 roe 01052 01173 01000 01074 01156 DIV - 01026 01115-01003 01000 01000 1B (Lower triangle is Spearman (rank) correlation) BVE ntr MVE0 MVE1 MVE2 roe DIV BVE 1100 0113-0112 - 0114-0103 0105 0101 ntr 0109 1100 0115-0127 - 0110 0107 0100 MVE 0-0145 0119 1100 0184 0139 0101-0104 MVE 1-0149 - 0126 0185 1100 0148-0101 - 0105 MVE 2-0124 - 0108 0167 0173 1100 0180-0107 roe 0103 0112 0128 0125 0177 1100-0105 DIV - 0101 0103 0101-0101 - 0102 0100 1100 :BVE,ntr,MVE 0 = Price 3 # TS + Price 3 # NTS, MVE 1 = Price 3 # TS + BPS 3 # NTS,MVE 2 = Price 3 # TS + EPS 3 15 3 # NTS,roe,DIV, 79

: ( ) Ohlson 11 (formal specification) Ohlson Ohlson, Ohlson Ohlson, X t X a t = X t - r 3 BV E t - 1, r = R - 1 > 0 Ohlson (1995) : X a t = X a t - 1 + V t + a t V t = V t - 1 + v t 0, < 1,V t,ohlson (1995) : MV E t = BV E t + 1 X a t + 2 V t (5) 1 = / ( R - ), 2 = R/ ( R - ) ( R - ), BV E t = BV E t - 1 + X t - DIV t, DIV t, (5) : MV E t = 0 + 1 BV E t + 2 X t + 3 DIV t + r V t (6) 0 = 0, 1 = 1 - r / ( R - ), 2 = R / ( R - ), 3 = 1-1, 4 = 2, : 0 = 0,0 < 1 < 1, 3 = - (1-1 ), < 1 1 > 0,, : w = 0195, R = 1104, 2 = R / ( R - ) = 1104 3 0195Π(1104-0195) = 10198 beta2,ohlson,bernard (1995) Feltham2Ohlson 68 % 80 %, Feltham2Ohlson,, (1999) 1993 1997 Feltham2Ohlson, (1999), Feltham2Ohlson,, (2002) Ohlson, ( ) 21 Ohlson,, : MV E t = 0 + 1 BV E t + 2 X t + 3 DIV t + 4 V t, MV E t t ; BV E t t ; X t t ; DIV t t ; V t t,, (Collins et al,1999 ;Darrough and Ye,2006) 2 Ohlson Vt = 0, Fama2 MacBeth, 11 80

2008 3,t 2A, MVE 0,BVE 1129, Ohlson (6) 1 0 1, t 2126, 5 % 1 1 X t (16118) t (6153) 10125 Ohlson MVE 1,, X t NI 5136 (t = 5176), MVE 0 1Π3,, MVE 2, EPS 15, BVE 0124 (t = 4115), MVE 0 1Π5, 2A ( ) MVE 0 t2value MVE 1 t2value MVE 2 t2value 707119 4147 270158 4196 266195 4175 BVE 1129 10107 0196 19104 0124 4115 X t 16118 6153 5136 5174 16123 17161 DIV 1131 2113 0168 1165 0163 1161 R 2 4315 % 3913 % 7114 % 2B ( ) MVE 0 t2value MVE 1 t2value MVE 2 t2value Constant 700197 319 261154 3154 177173 1149 BVE 2125 3183 1128 4161 1106 3135 X t - 2127-3151 - 0176-3113 10101 4129 DIV 5146 1106-4144 - 1161-9128 - 0182 R 2 5318 % 5311 % 7212 % :MVE0 = Price 3 # TS + Price 3 # NTS;MVE1 = Price 3 # TS + BPS 3 # NTS;MVE2 = Price 3 # TS + EPS 3 15 3 # NTS 2B MVE 2,,, MVE 2, 10101 (t = 4129), MVE 2 2 R 2 R 2 Ohlson,, R 2, MVE = 15 3 X t, R 2 = 100 % 2,,, 3, 31, ( ) = f (ntr) 3 ( ) Ohlson : MVE 0t 3 K = a + b 3 BVE + c 3 X t,mve 0t t MVE 0 = Price 3 Shares = Price 81

: 3 # TS + Price 3 # NTS, # TS # NTS,Price K = 1 - ntr 3 (1 - f) ntr,f BVE t t X t t 3 Ohlson, 1995 2005 11 1995 1996,, 0111 0126 1997, 014 015, 0145, 0105 1999 2005 3 (NTR) BVE X t DIV f R 2 R 2 (f = 1) 2 3 LIK 1995 134145 0112 2122 0111 3914 % 818 % 21519 3 1996 161164 0144 4105 0126 4214 % 2211 % 14213 3 1997 183142 1161 23155 0172 5715 % 5615 % 1512 3 1998 358141 0173 19134 0167 5212 % 5110 % 1813 3 1999 352127 0158 10110 0116 0145 5417 % 4914 % 7219 3 2000 885187 0172 18198-0114 0156 4910 % 4711 % 5410 3 2001 1091190 1116 12188 0125 0150 6013 % 5710 % 11117 3 2002 824172 0198 7159-0189 0143 5818 % 5513 % 15017 3 2003 631185 0185 8104 2146 0149 4910 % 4814 % 9719 3 2004 415110 0177 6117 2113 0138 4915 % 4615 % 18210 3 2005 220105 0142 6106 2137 0142 3616 % 3314 % 9213 3 478115 0176 10182 0190 0145 5010 % 4315 % 10418 104153 0113 2121 0156 0105 T2 4157 5185 4190 1161 8134 : 3 = 01001 3 2A (BVE) 0176, Ohlson 1 ( X t ) 10182, Ohlson 10198 3, :H 0 f = 1 vs. H a f 1 (B) (A),,, n B n B - n A 2, 2 3 (Log - likelihood (B) - Log - likelihood (A) ) 2 n - n B A n A (B) (A) (A) H 0 f = 1 Ohlson, (B) f Ohlson n B - n A = 1, H 0 f = 1, 2, 1, 2 3, 1, 10418,H 0 f = 1 a = 01001, f 82

2008 3 1995 1996 (f = 0111 0126),,,, 3, R 2, R 2 ( f = 1) f = 1,,Ohlson R 2 f Ohlson f, R 2,, f, Ohlson Core, Guay, and Van Buskirk (2003) R 2 35 %,Ohlson (ntr)?,,, 4 BVE X t DIV f R 2 R 2 (f = 1) 2 3 LIK 628 67317 0126 15168-2165 0156 6414 % 5512 % 1016 289 32815 1106 5107-11124 0121 5313 % 4215 % 1417 1369 38017 0188 10102-1141 0136 5114 % 4411 % 2411 5068 49218 0180 11107 1126 0150 5112 % 4419 % 5518 824 57510 0167 12101-2123 0158 5815 % 5311 % 1119,, : ;,, ;, ( ) f, 10 10 (1ΠK- 1), K= 1 - ntr 3 (1 - f) f 0145, K = 1-0155 3 ntr 5 5, ntr = 015, 10 5 3179 10 3,, ntr = 017, 10 6126,,,,,,, ntr NTR K 015 01725 3179 016 01670 4193 017 01615 6126 018 01560 7186 0185 01505 8178 016, 4193, 3,, 10 1193, 1913 %, 20 %, 10 3, ntr 0142, 231,ntr 83

: 0142 18, 7179 %, 92 %,,,, (2006), (2006), (2006), (2006), (2006),, ( ),, (ntr), ( ERC),,,,,, : :CAR = a + b 1 3 MΠB + b 2 3 UE + b 3 3 UE(neg) + :CAR = a + b 1 3 MΠB + b 2 3 UE + b 3 3 UE(neg) + b 4 3 NTR + b 5 NTR 3 UE + b 6 3 NTR 3 UE (neg) +,CAR MΠB UE,UE = UEPS 0 ΠPrice, UE < 0, UE(neg) UE, 0 NTR 6 6 (NTR) UE ( neg),, (Fama and French,1992), UE,UE(neg) 2 Coefficient T2ratio Coefficient T2ratio MΠB - 0110-12126 - 0110-11197 UE 4147 21177 1146 1152 UE(neg) - 2180-916 - 1116-0187 NTR - 0106-1167 NTR 3 UE 4174 3119 NTR 3 UE(neg) - 2143-1117 1 ( ERC) 4147,t 21177, ERC = 4147-2180 = 1167, NTR 3 UE, 4174 (t = 3119), UE ( 1146, t = 1152) UE NTR 3 UE,,,,, 84

2008 3 : (1), ; (2) Ohlson,, ; (3),,,,,,,,,,,2002 : :, 4,2001 :, 1,2002 :, 8,2006 :, 4,2002 :, 9,2003 :, 9,2005 : :, 7,2003 :, 4,2003 :, 11,2004 :, 4, 1999 : :,,2006 :, 11,2004 : :, 2,2003 :, 8,1999 :, 12,2001 :, 1,2005 :, 10,2005 :, 5,2001 :, 7,2003 :, 8,2004 :, 7,2006 :, 8,2005 : : A, 8,1998 :, 8,2004 :, 6 2005 :, 2,2002 : ZZ, 11,2006 : :, 4,2004 :, 2 85

:,2001 :, 11,2004 :, 6,2003 :, 2,2006 :, 1,2006 : IPO, 11,2005 :, 12,2003 :, 9,2006 :, 8,2006 :, 2,1999 :, 9,2003 :, 10,2006 :, 11,2006 :, 12,1999 :, 5,2001 :, 12 Berger,P. G. and E. Ofek,1995, Diversification s Effect on Firm Value, Journal of Financial Economics,Vol. 37, (1),39 65. Bernard,Vitor L,1995, The Feltham2Ohlson Framework : Implication for Empiricists, Contemporary Accounting Research,11(2) :733 747. Collins,D. W.,M. Pincus and H. Xie,1999, Equity Valuation and Negative Earnings : The Role of Book Value of Equity, The Accounting Review,Vol. 74 :29 61. Core,J. E.,Guay,W. R. and A. Van Buskirk,2003, Market Valuations in the New Economy : An Investigation of What Has Changed, Journal of Accounting and Economics,Vol. 34,43 67. Darrough,M. and J. Ye,2006, Valuation of Loss Firms in a Knowledge2Based Economy, Review of Accounting Studies,forthcoming. Fama,E. and French K.,1992, The Cross2section of Expected Stock Returns, Journal of Finance,Vol. 47,427 4651 Ohlson,J.,Earnings,1995, Book Values,and Dividends in Equity Valuation, Contemporary Accounting Research,Vol. 11 (2) :661 687. The Effects of Split Share Structure on the Empirical Capital Market Research in China and the Corrections Yang Dan, Wei Yunxin and Ye Jianming (Southwestern University of Finance and Economics ; Baruch College at City University of New York) Abstract :The coexistence of tradable and non2tradable shares affects the calculations of the market value of a firm, and influences virtually all the results in the empirical capital market research using data from China. The effect arises from the finding that each non2tradable share is worth less than one tradable share. The problem will continue to exist for any research utilizing historical data before the combination of two types of shares. We show that different assumptions about the price of non2tradable shares lead to significantly different empirical results, and a scientific correction is of crucial importance. In this paper, we propose a method which defines the price of a non2tradable share as a fraction of that of a tradable share. The fraction is then estimated through a model. We show that our method leads to estimates that are economically more meaningful than existing alternatives. As an extension of our method, we study the whether the tradable shares were compensated fairly during the reform of the split share structure. We find tradable shares are fairly compensated for the companies with lower fraction of non2tradable shares. For firms with high fraction of non2tradable shares, tradable shares were exploited by the non2tradable shareholders. Key Words : Coexistence of Tradable and Non2tradable Shares ; Equity Valuation ; Price2to2book Ratio ; Model Specification ; Reform of Split Share Structure ; Empirical Capital Market Research JEL Classification : G320, G340, G390 86 ( : ) ( : )