* Toin s q *
999 997 46 Herfindahl 00 Matsusaka00 379 998-000 999 SIC Toin s q Weston970 CEO Gertner 994 CEO CEO smart money effectstein997 CEO winner-picking smarter money effecthuard Palia999 60
Khanna Tice 00 CEO CEO Jensen986 Stulz990 CEO stupid money effect Scharfstein Stein000 CEO more stupid money effect Lang Stulz 994 978990 Compustat Herfindahl Toin s qberger Ofek995 98699 Compustat 3%5%Comment Jarrell995Lang Stulz 978 Herfindahl 0. 4.3% Lins Servaes999 994996 Worldscope 0%5% Lamont997 986 50% CEO Shin Stulz998 98099 Compustat 6:: Raan 000 CEO CEO CEO CEO May995 979990 Compustat CRSP CEO CEO Dennis 997 98499 Compustat Value Line CEO CEO Aggarwal Samwick003 CEO CEO CEO CEO CEO 3
sample selection iases Campa Kedia00 Graham 00 Burch Nanda003 measurement error Villalonga004 BITS Compustat Compustat BITS Matsusaka00 CEO CEO Maksimovic Phillips00 Gomes Livdan004 4 SEC S-K 976 0% SIC Compustat Industry Segment Files 4 SIC 979 SIC 004 74 M: Claessens 000 Compustat Worldscope 8 5% 0% 5%Linss Servaes00 Worldscope 7 7% 4
M Herfindahl HI: HI = i = P i Pi Herfindahl M 3 EntropyEI EI = i = P i ln( / P i). MV/BVMV = BV = Toin s q: Chung Pruitt 994 Toin s q MV BD Toin s q = BV CL BD CL 3 EV MV MV EV = ln λ M i = = SALE i i λ MSALE i i SALE i i λi MSALEi i 3. 999 06 00000 cninfo.com.cn CSMAR CSMAR ST 79 96 Chung Pruitt994 Toin s q R 0.966 Berger Ofek995 998 0% 0% 0%% 00 9%0% 0 00 6% 6 5
. Toin s q Herfindahl 000 00 Toin s q 5% 00 00 0% Herfindahl 000 Herfindahl 0%5% Toin s q EV M HI EI MV / BV Toin s q EV M HI EI Toin s q EV M HI EI MV / BV Toin s q EV M HI EI Toin s q EV M HI EI MV / BV Toin s q EV M HI EI MV / BV EV M HI EI Herfindahl 0 Spearman Spearman Rank correlation coefficient 6
Herfindahl Herfindahl 5 35 00 Toin s q 0.3370.6540.0633 0.300.595 0.877 6.%9.56%3.0% 0.397 0.89.639 7.73% 5.3%8.87% Herfindahl M MV/BV Toin s q EV HI HI = 0.8 HI < 0.6 HI < 0. 8 0.4 HI < 0. 6 HI < 0. 4 MV/BV Toin s q EV EI EI = 0 0 < EI ln ln < EI ln 3 ln 3 < EI ln 4 ln 4 < EI MV/BV Toin s q EV Toin s q 3.54 Toin s q 7.48 7
. GLS Y = a D( ) D(3) 3 D(4) 4D(5) ε 3 Y Y = a ) HI( ) HI (3) 3HI(4) 4HI(5 ε 4 = a ) EI ( ) EI (3) 3EI (4) 4 EI(5 ε 5 Y Toin s q 948 =,, L, D(k) M k D( k) = D( k) = 0 k =,3,4, 5 HI (k ) Herfindahl 0.8 HI < HI ( ) = HI ( ) = 0 0.6 HI < HI ( 3) = HI ( 3) = 0 0.4 HI < HI ( 4) = HI ( 4) = 0 HI HI ( 5) = HI ( 5) = 0 EI(k) EI > ln( k ) EI( k) = EI( k) = 0 k =,3,4, 5 D() D() D(3) 3 D(k) k HI (k ) EI(k) k 8
3 D() EI() 5% HI () 5% 8 D(k) HI (k ) EI(k) k > 3 6 F - 3. i LNSIZE ii INTANG iii DIVID DIVID = DIVID = 0 iv LEVER Y = β 0 β β 5 DIVERSITY LEVER ξ β LNSIZE β 3 INTANG β 4 DIVID 6 Y t = β 0 β β 5 DIVERSITY LEVER t µ t t δ β ξ LNSIZE t t β 3 INTANG t β 4 DIVID t 7 Y Toin s q DIVERSITY DIVERSITY = 0 DIVERSITY GLS fixed effect 4 DIVERSITY Y Toin s q EV 0.090.35 0.30.6 5% Y MV/BV 00 0.0.4 0% LNSIZE LEVER DIVID INTANG roustness test 7 000-00 655 7 Toin s q 5 9 3 outlier 6 7 LNSIZE LEVER LNSIZE INTANG DIVERSITY DIVID LEVER DIVERSITY 9
â ˆ ˆ 3ˆ 4ˆ R ε = ) (5 (4) (3) ) ( 4 3 D D D D a Y MV/BV Toin s q EV ε = ) (5 (4) (3) ) ( 4 3 HI HI HI HI a Y MV/BV Toin s q EV ε = ) (5 (4) (3) ) ( 4 3 EI EI EI EI a Y MV/BV Toin s q EV
DIVERSITY LNSIZE INTANG DIVID LEVER R MV/BV Toin s q EV Toin s q - 0.090.6
CEO CEO Compustat ISF Worldscope Zingale 000 CEO / / / CASHFLOW = = CASHIN = CASHOUT = CSMAR Logit model
P, t ln P, t = β = β 0 0 ßX 3, t β CASHFLOW β VALUE, t 4, t β LNSIZE β MANAGER, t, t 8 P, t = Pr( DIVERSITY, t = X) t CASHFLOW, t CASHIN, t CASHOUT, t MANAGER, t VALUE, t MV / BV, t Toin' s q, t EV, t LNSIZE, t 5 CASHIN CASHOUT MANAGER MV / BV Toin' s q EV LNSIZE R CASHIN CASHOUT MANAGER MV / BV Toin' s q EV LNSIZE R MANAGER 00000 3
4 5 CASHIN CASHOUT / % 0.4% Toin s q EV 5%0% LNSIZE MANAGER M M CASHFLOW MANAGER VALUE LNSIZE censored sample OLS iased inconsistent Toit model > = 0 0,, 4, 3,, 0, t t t t t t LNSIZE MANAGER VALUE CASHFLOW M ζ φ φ φ φ φ 9 CASHIN CASHOUT MANAGER BV MV / q s Toin' EV LNSIZE CASHIN CASHOUT MANAGER BV MV / q s Toin' EV LNSIZE
6 Toit Logit Toin s q EV CASHIN CASHOUT 0% LNSIZE MANAGER Lang Stulz994 Toin s q 0.090.6 5
000 00 9 997 3 999 00 000 3 003 999 3 Aggarwal, R. and A. Samwick, 003, Why do managers diversify their firms? Agency reconsidered. Journal of Finance 58, 7-8. Berger, P. and E. Ofek, 995, Diversification s effect on firm value. Journal of Financial Economics 37, 39-65. Burch, T. and V. Nanda, 003, Divisional diversity and the conglomerate discount: The evidence from spin-offs. Journal of Financial Economics 70, 69-98. Campa, J. and S. Kedia, 00, Explaining the diversification discount. Journal of Finance 57, 73-76. Chung, K. and S. Pruitt, 994, A simple approximation of Toin s q. Financial Management 3, 70-74. Claessens, S., Dankov, S., Fan, J. and L. Lang, 000, The pattern and valuation effects of corporate diversification: A comparison of the United States, Japan, and Other East Asian Economies. World Bank and CEPR Working Paper Comment, R. and G. Jarrell, 995, Corporate focus and stock returns. Journal of Financial Economics 37, 67-87. Dennis, D. J., Denis, D. and A. Sarin, 997, Agency prolems, equity ownership and corporate diversification. Journal of Finance 5, 35-60. Gertner, R., D. Scharfstein and J. Stein, 994, Internal versus external capital markets. Quarterly Journal of Economics 09, -30. Gomes, J. and D. Livdan, 004, Optimal diversification: Reconciling theory and evidence. Journal of Finance 59, 507-535. Graham, J., Lemmon, M. and J. Wolf, 00, Does corporate diversification destroy value? Journal of Finance 57, 695-70. Huard, R. and D. Palia, 999, A reexamination of the conglomerate merger wave in the 960s: An internal capital markets view. Journal of Finance 54, 3-5. Jensen, M., 986, Agency costs of free cash flow, corporate finance, and takeovers. American Economic Review 76, 33-39. Khanna, N. and S. Tice, 00, The right side of internal capital markets. Journal of Finance 56, 489-58. Lamont, O., 997, Cash flow and investment: Evidence from internal capital markets. Journal of Finance 5, 83-09. Lang, L. and R. Stulz, 994, Toin s q, corporate diversification, and firm performance. 6
Joural of Political Economy 0, 48-9. Lins, K. and H. Servaes, 999, International evidence on the value of corporate diversification. Journal of Finance 54, 5-39. Lins, K. and H. Servaes, 00, Is corporate diversification eneficial in emerging market? Financial Management 3, 5-3. Maksimovic, V. and G. Phillips, 00, Do conglomerate firms allocate resources inefficiently across industries? Theory and evidence. Journal of Finance 57, 7-767. Matsusaka, J., 00, Corporate diversification, value maximization and or ganizational capailities. Journal of Business 74, 409-43. May, D., 995, Do managerial motives influence firm risk-reduction strategies? Journal of Finance 50, 9-308. Raan, R., Servaes, H. and L. Zingales, 000, The cost of diversity: The diversification discount and inefficient investment. Journal of Finance 55, 35-79. Scharfstein, D. and J. Stein, 000, The dark side of internal capital markets: Divisional rent- seeking and inefficient investments. Journal of Finance 55, 537-564. Shin, H. and R. Stulz, 998, Are internal capital markets efficient? Quarterly Journal of Economics 3, 53-553. Stein, J., 997, Internal capital markets and the competition for corporate resources. Journal of Finance 5, -33. Stulz, R., 990, Managerial discretion and optimal financing policies. Journal of Financial Economics 6, 3-7. Villalonga, B., 004, Diversification discount or premium? New evidence from the usiness information tracking series. Journal of Finance 59, 479-506. Weston, J., 970, The nature and significance of conglomerate firms. St. John s Law Review 44, 66-80. 7
Diversification and Firm Value: An Empirical Investigation of Diversification Premium Based on China s Stock-Market Listed Companies Su Dongwei (College of Economics, Jinan University, Guangzhou 5063) Astract: Diversification and firm value has een one of the most active research areas in corporate finance in recent years. By testing various empirical implications of internal capital market theories, this paper investigates diversification and firm value using data constructed from a sample of China s stock-market listed companies. The paper finds a statistically significant diversification premium on average, diversified firms have higher Toin s q, market-to-ook ratio and excess value measure than firms that are more focused. After controlling for factors that may affect firm value, such as size, intangiles, leverage and dividend policy, the paper finds that diversification premium ranges from 0. to 0.7. The paper also finds that high value firms are more likely to adopt diversification strategy and efficient internal capital markets may explain diversification premium for Chinese firms. Key Words: Diversification, Firm value, Diversification premium, Stock-market listed companies, Internal capital market JEL Classification: G30, G5, C0 8