2004 4 ( 100732) ( ) : 2000,,,20 : ;,, 2000 6 43 %( 11 %), 80 90,, : :,,, (), (),, (CUHIES2000), Heckman Vytlacil (1999, 2000, 2001),Carneiro, Heckman Vytlacil (2001) Carneiro (2002) (MTE), Bjorklund Moffitt (1987), : 2000 6 ( ), 4 (ATE) 43 %( 11 % ) (OLS) ( IV) ATE 29 %56 %( 7 %14 % ),- 22 % Carneiro, Heckman Vytlacil (2001) : 91
: : (OLS) (ATE), ( IV),80 90, :; (MTE),MTE ; ; (Mincer),, ln Y i = S i + X i + U i (1) i ( i = 1,2,, n),ln Y i, S i, X i,, U i,, (OLS) (1) : A i, U i Cov( A i, S i ) 0, E( U i S i ) 0 Griliches(1977),,,( IV), S i U i I i ;,,;, X i,,,,, (Carneiro Heckman (2002) Carneiro (2002) ) :, (Carneiro Heckman,2003) : (1),, ln Y i = i S i + X i + U i (2) i,x i, :S i = 1, S i = 0 ( ),,,,, ( Heckman, Lochner Todd, 2003) 92
2004 4 (ln Y 0 i,ln Y 1 i ) : ln Y 0 i ln Y 1 i, E( U 0 i X i ) = 0, E( U 1 i X i ) = 0 = 0 X i + U 0 i (S i = 0 ) (3a) = 1 X i + U 1 i (S i = 1 ) (3b), ln Y 0 i ln Y 1 i F(ln Y 0 i X i, S i = 0) F(ln Y 1 i X i, S i = 1), F(ln Y 0 i X i ) F(ln Y 1 i X i ), OLS IV,: ln Y i = S i ln Y 1 i + (1 - S i ) ln Y 0 i (4a) = [ ( 1-0 ) X i ] S i + 0 X i + [ U 0 i + ( U 1 i - U 0 i ) S i ] (4b) = [ ( 1-0 ) X i + ( U 1 i - U 0 i ) ] S i + 0 X i + U 0 i (4c) = i S i + 0 X i + U 0 i (4d) i = ( 1-0 ) X i + ( U 1 i - U 0 i ) (5) i 1 0 ( ( 1-0 ) X i ),U 1 i U 0 i ( ( U 1 i - U 0 i ) ), i, X, i : : g= E( i X i ) = E[ ( 1-0 ) X i ] (6) S 3 i = P i ( Z i ) - U si S i = 1 S 3 i 0 0 S 3 i < 0 S 3 i,, Z i ( Z i X i ) P i = P i ( Z i ) ( ), U si i, U si [ 0,1 ]( Heckman Vytlacil, 1999), i, P i ( Z i ) U si U si, (7) i = ln Y 1 i - ln Y 0 i i S i = 0 S i = 1 (3a) (3b) (5), i = i (3a) (3b) (6),(OLS) : plim( ^ OLS ) = E(ln Y i X i, S i = 1) - E(ln Y i X i, S i = 0) = E( 1 X i + U 1 i X i, S i = 1) - E( 0 X i + U 0 i X i, S i = 0) = ^ ( A TE) + [ E( U 1 i S i = 1) - E( U 0 i S i = 0) ] ( ), ATE X, : (8) 93
: : A TE = E( i X i ) = E( i X i ) = g (9) U si ( U 0 i, U 1 i ), S i U 0 i U 1 i, (8) OLS ATE (8) : plim( ^ OLS ) = E(ln Y i X i, S i = 1) - E(ln Y i X i, S i = 0) = E( i X i, S i = 1) ( TT) + [ E( U 0 i S i = 1) - E( U 0 i S i = 0) ] ( ) TT ( ) ( ),: TT = E( i X i, S i = 1) = E( i X i, S i = 1) = g+ E( U 1 i - U 0 i S i = 1) = A TE + E( U 1 i - U 0 i S i = 1) ] ( ) E( U 0 i S i = 1) - E( U 0 i S i = 0) ( ) (), E( U 1 i - U 0 i S i = 1) ( ) (8),IV S i U 0 i U 1 i - U 0 i I i g, : plim^ IV = Cov( I i,ln Y i ) Cov( I i, S i ) = g+ Cov( I i, U 0 i ) Cov( I i, S i ) = g+ Cov[ I i, ( U 1 i - U 0 i ) S i ] Cov( I i, S i ) + Cov[ I i, ( U 1 i - U 0 i ) S i ] Cov( I i, S i ) = g+ Cov[ I i, ( U 1 i - U 0 i ) S i = 1 ] P i Cov( I i, S i ) P i = Pr ( S i = 1) U 1 i U 0 i, U 1 i - U 0 i S i, (12), plim( ^ IV ) giv U 1 i - U 0 i = 0 ( ) U 1 i U 0 i U 1 i - U 0 i S i (, ), (12), IV g(heckman, 1997 ;Heckman Navarro2Lozano, 2003),,OLS IV,(Heckman Vytlacil (1999, 2000, 2001), Carneiro, Heckman Vytlacil (2001), Carneiro (2002) Navarro2Lozano (2002) ), ( LIV), ( MTE) MTE X i U si ln Y 1 i (ln Y 0 i ) ( WTP), : 94 MTE( X i = x, U si = u s ) = E( i X i = x, U si = u s ) = E( i X i = x, U si = u s ) (10) (11) (12)
2004 4 = ( 1-0 ) x + E( U 1 i - U 0 i U si = u s ) (13) MTE, LIV (Heckman,2001) : MTE( X i = x, U si = P i = p) = LIV ( X i = x, P i = p) = 5 E(ln Y i X i = x, P i = p) 5 p,x, Heckman Vytlacil (1999, 2000, 2001) Carneiro (2002) MTE : : ATE = 1 MTE( u s ) du s 0 TT = 1 MTE( u s ) h TT ( u s ) du s 0 TUT = 1 MTE( u s ) h TUT ( u s ) du s 0 h ATE ( u s ) = 1 h TT ( u s ) = 1 - F P ( u s ) E( P i ) h TUT ( u s ) = F P ( u s ) E(1 - P i ) ( ) ( ) () 1 f ( p) dp u s = E( P i ) u s f ( p) dp 0 = E(1 - P i ). ( TUT) ( ) ( ), : TUT = E( i X i, S i = 0) = E( i X i, S i = 0) (14) = g+ E( U 1 i - U 0 i ) S i = 0). (15) 2000 (CUHIES) 1992 2003,, 2000 6,6 : 2000, 6 7627,6280 6 4250, 7 :,6 2000, 587 (331,256 ), 273, 314 2613,,, 95
: : OLS IV 1 29 %56 %, 7125 %14 % OLS (CHOW,2001) (20 80 90 ) OLS : (, ), 1 OLS IV 8. 3189 0. 1493 8. 3040 0. 1552 4 e 0. 2929 0. 0630 0. 5609 0. 1695 0. 0380 0. 0194 0. 0196 0. 0202-0. 0016 0. 0010-0. 0007 0. 0010 0. 0117 0. 0020 0. 0098 0. 0023 0. 1537 0. 0602 0. 1439 0. 0607 0. 7543 0. 1255 0. 7908 0. 1267 0. 2693 0. 1085 0. 3142 0. 1092 0. 2278 0. 1181 0. 2759 0. 1192 0. 7246 0. 1241 0. 7775 0. 1256 0. 6241 0. 1297 0. 6739 0. 1314-0. 3679 0. 0855-0. 3873 0. 0868-0. 4786 0. 1288-0. 5890 0. 1298-0. 4649 0. 1179-0. 5304 0. 1179 IND - CON - 0. 2793 0. 0788-0. 3048 0. 0792 TRA - COM - 0. 4512 0. 1762-0. 4645 0. 1779 SPO - SOC - 0. 2880 0. 0900-0. 3106 0. 0905 FIN - INS - 0. 3220 0. 1050-0. 3327 0. 1061 :IND - CON ; TRA - COM ; SPO - SOC ;FIN - INS, 2, (Marginal effects) : Pr ( S = 1 X) = 5Pr ( S = 1 X) Marginal effects = 5 X exp ( X) = ( X) (16) 1 + exp ( X) = ( X) [1 - ( X) ], 2 1 Pr ( S = 1) 96
2004 4 2-4. 7370 0. 7305-1970 1. 5076 0. 7534 0. 3134 0. 1017 0. 0297 0. 0211 1971 3. 0771 0. 7138 0. 6396 0. 0605 0. 0342 0. 0126 1972 2. 6424 0. 7183 0. 5492 0. 0190 0. 0069 0. 0040 1973 2. 5395 0. 6809 0. 5279 1964 2. 0008 0. 7969 0. 4159 1974 2. 7740 0. 6753 0. 5766 1964 1. 7285 0. 9189 0. 3593 1975 2. 7931 0. 6763 0. 5806 1965 3. 3423 0. 8257 0. 6947 1976 2. 8634 0. 6669 0. 5952 1966 3. 1813 0. 8552 0. 6613 1977 2. 5890 0. 6672 0. 5381 1967 1. 8455 1. 1126 0. 3836 1978 2. 5572 0. 6656 0. 5315 1968 2. 9030 0. 8161 0. 6034 1979 1. 3631 0. 7636 0. 2833 1969 2. 2569 0. 7941 0. 4691 3 (, = 014) 0 1 0. 0360 0. 0225 0. 0141 0. 0278-0. 0013 0. 0011-0. 0009 0. 0013 0. 0188 0. 0038 0. 0077 0. 0038 0. 1365 0. 0723 0. 1913 0. 0777 0. 5712 0. 1961 0. 8853 0. 1590 0. 1901 0. 1263 0. 3929 0. 1049 0. 2612 0. 1364 0. 2296 0. 1081 0. 7122 0. 1695 0. 7971 0. 1301 0. 6930 0. 1551 0. 5461 0. 1744-0. 3368 0. 1188-0. 4471 0. 1093-0. 6060 0. 2065-0. 5868 0. 1771-0. 4205 0. 1511-0. 6256 0. 1677 IND - CON - 0. 2297 0. 0821-0. 3978 0. 0990 TRA - COM - 0. 3527 0. 1318-0. 5040 0. 1557 SPO - SOC - 0. 3702 0. 1282-0. 3040 0. 1202 FIN - INS - 0. 3345 0. 1560-0. 3543 0. 1331 :IND - CON ; TRA - COM ; SPO - SOC ;FIN - INS 97
: : 3 2 MTE 3 (3a) (3b) 2 MTE, u s MTE u s : u s ((7) ) u s 2 MTE, OLS IV, = 0. 4 2 1 P( S = 1) 4 ATE 43 % (11 %), OLS 29 %(7 %), ATE : IV > ATE > OLS,- 22 %TT, TUT, 51 % 36 % 8 %,, : = E( i - g X i, S i = 1) = E( i X i, S i = 1) - g = TT - A TE (17) 4 2,2000 6, 4 OLS 0. 2929 TUT 0. 3630 IV 1 0. 5609 2-0. 1407 ATE 0. 4336 3-0. 2220 TT 0. 5149 4 0. 0813 :11 ;21 = OLS - ATE ; 31 = OLS - TT ;41 = TT - ATE,, 3 MTE u s, 2, ( ) ( ), 98
2004 4, MTE Carneiro (2002) (OLS IV ) 2000, 6 3 :, = 0. 33,43 %( 11 %), OLS, OLS,,20 80 90 Bjorklund, A. and R. Moffitt,1987,The Estimation of Wage Gains and Welfare Gains in Self2Selection Models, Review of Economics and Statistics, 69 :42 49. Carneiro, P. Dissertation, University of Chicago., 2002,Heterogeneity and Selection in the Return to Schooling : Implications for Education Policy Evaluation, Ph. D. Carneiro, P. and J. Heckman,2002,The Evidence on Credit Constraints in Post2Secondary Schooling, Economic Journal, Volume 112, issue 482, pp :705 734.,2003, Human Capital Policy,in J. Heckman, A. Krueger, Inequality in America : What Role for Human Capital Policies, Cambridge, MA : MIT Press. Carneiro, P., J. Heckman and E. Vytlacil,2001,Estimating the Return to Education When It Varies Among Individuals, Working paper, University of Chicago. Chinaπs Urban Household Income and Expenditure Survey ( CUHIES),2000, National Bureau of Statistics, Urban Socio2Economic Survey Organization. Chow, G.,2001, Chinaπs Economic Transformation. Malden, MA : Blackwell Publishers. Griliches, Z.,1977,Estimating the Returns to Schooling : Some Econometric Problems,Econometrica, Volume 45, number 1, pp. 1 22. Heckman, J.,2001,Microdata, Heterogeneity and Econometric Policy Evaluation, Nobel Memorial Lecture in Economic Sciences, Journal of Political Economy, 109 (4), 673 748. Chicago. Heckman, J., L. Lochner and P. Todd., 2003, Fifty Years of Mincer Earnings Functions,Unpublished manuscript. University of Heckman, J. and S. Navarro2Lozano.,2003,Using Matching, Instrumental variables and Control Functions to Estimate Economic Choice Models forthcoming in Review of Economics and Statistics. October. Heckman, J. and E. Vytlacil,1999,Local Instrumental variable and Latent variable Models for Identifying and Bounding Treatment Effects, Proceedings of the National Academy of Sciences, 96 :4730 4734. 36.,2000, Local Instrumental variables, in C. Hsiao, K. Morimune, and J. Powells, ( eds. ), Nonlinear Statistical Modeling : Proceedings of the Thirteenth International Symposium in Economic Theory and Econometrics : Essays in Honor of Takeshi Amemiya, Cambridge : Cambridge University Press, 1 46. ( 116 ) 99
: : The Cropland Conversion : Contracting Out and Public Regulation Wang Xiaolong (CCER ;Peking University) Abstract :The Cropland Conversion Program is an important public policy for improving ecological environment. Focusing on the arrangement of structure of property rights, the Program tries to establish a new contractual relationship between government and farmers so as to reduce negative externalities by cultivating activities and encourage the afforesting activities that are conducive to the improvement of ecological environment. Within the framework of principal2agent model, this paper studies incentive2 incompatibility problems that may exist in the enforcement of the contract. Firstly, the paper makes an analysis on the social rationality of the contacting out. The analysis is followed by a model to show how market shocks may cause the behavior of farmers to be inconsistent as to the goal of the program. Then, taking The Cropland Conversion Program of Shannxi Province as a study case, the paper makes an empirical analysis on incentive2incompatibility problems facing the enforcement of specific public policies. Finally, the paper proposes a series of public regulation schemes in the hope that they can be used to improve the efficiency of the enforcement of the program. Key Words :Cropland Conversion ; Environment ; Incentive ; Contract JEL Classification :Q230 ( : ) ( : ) ( 99 ) Heckman,J. and E. Vytlacil,2001,Structural Equations, Treatment Effects and Econometric Policy Evaluation, Fisher Schultz Lecture World Congress of the Econometric Society. National Bureau of Statistics (NBS). 2001, China Labor Statistical Yearbook. Beijing : China Statistics Press. Heterogeneity,Selection Bias and the Return to Education : A Empirical Analysis Based on Chinese Micro2Data Li Xue2song James J. Heckman ( Insitute of Mathemetical and technical Economics,CASS) (Chicago University) Abstract :This paper uses newly available Chinese micro data to estimate the return to college education for late 20th century China when allowing for heterogeneous returns among individuals selecting into schooling based on these differences. We use modern micro2econometric methods to identify the parameters of interest. We demonstrate that heterogeneity among people in returns to schooling is substantial. People sort into schooling on the basis of the principle of comparative advantage, which we document to be an empirically important phenomenon in modern Chinese labor markets. Standard least squares or instrumental variable methods do not properly account for this sorting. Using new methods that do, we estimate the effect on earnings of sending a randomly selected person to college is a 43 % increase in lifetime earnings (nearly 11 % annually) in 2000 for young people in urban areas of six provinces of China. Our evidence, and simple least squares evidence, suggests that after 202plus years of economic reform with market orientation, the return to education has increased substantially in China, compared to the returns measured in the 1980πs and the early 1990πs. Key Words : Micro2Data Heterogenetity Selection Bias Return to Education of China JEL Classification :C340,I210,J310 ( : ) ( :) 116