韓國開發硏究제 33 권제 3 호 ( 통권제 112 호 ) 재직자직업훈련관련공적재정의구조와성과 : 효과분석 이철인 ( 서울대학교사회과학대학경제학부교수 ) 유경준 ( 한국개발연구원선임연구위원 ) Training Incentives in the Korean Levy-Gran

韓國開發硏究제 33 권제 3 호 ( 통권제 112 호 ) 재직자직업훈련관련공적재정의구조와성과 : 효과분석 이철인 ( 서울대학교사회과학대학경제학부교수 ) 유경준 ( 한국개발연구원선임연구위원 ) Training Incentives in the Korean Levy-Grant System and the Performance: Evidences from the KLIPS Data Chul-In Lee (Professor, Department of Economics, Seoul National University) YOO, Gyeongjoon (Senior Research Fellow, Korea Development Institute) * 본논문은 2009 년도 KDI 연구사업인 직업훈련심층평가 의일환으로시작된연구를발전시킨것임을밝힌다. 한국노동연구원의노동패널자료협조에감사드린다. 익명의검토자두분의매우유익한 논평및제안들에많은도움을받았고이에대해깊이감사드린다. ** 이철인 : (e-mail) leeci@snu.ac.kr, (address) Seoul National University, Gwanak-ro 1, Gwanak-gu, Seoul, Korea 유경준 : (e-mail) yoogj@kdi.re.kr, (address) Korea Development Institute, 49 Hoegiro, Dongdaemun-gu, Seoul, Korea Key Word: (Public Training Schemes), (On-the-job Training), - (Levy-grant System) JEL Code: J01, J24, J08, H25 Received: 2011. 5. 12 Referee Process Started: 2011. 5. 16 Referee Reports Completed: 2011. 8. 31

ABSTRACT This paper examines how the levy-grant system for on-the-job training affects individual workers' training level and the subsequent wage growth. Some notable results include: (i) the workers at the firms facing high net benefits (i.e., grant minus levy) receive more firm training indeed, and (ii) training provision raises post-training earnings substantially. All these results are found to be robust to changes in firm size and estimation method.

재직자직업훈련관련공적재정의구조와성과 : 효과분석 89 Ⅰ. 서론. (schooling) (on-the-job training: OJT)..,,., (Becker[1964]),,,,., (Heckman[1999]). 1),. 2), /. 1) (self selection) Heckman(1979), Heckman(1999), Heckman and Smith(1998), Heckman and Vytlacil(2005), Heckman, Vytlacil, and Urzua(2006), Heckman and Vytlacil(2007a), Heckman and Vytlacil(2007b). 2) Becker(1964). Stevens (1994), Acemoglu and Pischke(1998).

90 韓國開發硏究 / 2011. Ⅲ., (positive). - (levy-grant)., ( levy-grant ),.,,, /. ( [1997]; [2000]; [2002]; [2002]; [2004] [2008] )., (i) 3), (ii).. (Heckman[1999]; [2008] ),. Ben-Porath(1967) Heckman(1976), Mincer(1974),.,., 3),.

재직자직업훈련관련공적재정의구조와성과 : 효과분석 91, 4) (variation),..,,,., levy-grant,. (2009) 2008 2007.,..,... Ⅱ. 분석의틀및모형 1. 직업훈련사업의재정조달구조를이용한효과식별 가. 수혜및분담률체계 5). ( ) 1 4) Brown and Medoff(1989),. 5) (2009).

92 韓國開發硏究 / 2011. Ⅲ. (i) I,, / /. (ii), (iii),, / /. <Table 1> <Table 4> 1, 1..., 1~100, 101~150, 151~999 1,000. 1~149, 150~500, 501~999 1,000, 1~149, 150~300, 301~999 1,000. 1,. (variation),. <Table 2>,.. <Table 3>,., (i), (ii),, (iii)., (variation). 6) 6).,,

재직자직업훈련관련공적재정의구조와성과 : 효과분석 93 <Table 1> Public Assistance System for Promoting On-the-Job Training: Cost and Maximum Grant Relative to Wages (1999. 1. 29 ~ 2002. 12. 29) (Unit: Percent) Industry Group Firm Size 1~100 101~149 150~999 Over 1,000 Cost 0.10 0.10 0.50 0.70 Max. Grant 0.18 0.12 0.60 0.84 Net Grant 0.08 0.02 0.10 0.14 Industry Group Firm Size 1~149 150~500 501~999 Over 1,000 Cost 0.10 0.30 0.50 0.70 Max. Grant 0.18 0.54 0.60 0.84 Net Grant 0.08 0.24 0.10 0.14 Industry Group Firm Size 1~149 150~300 301~999 Over 1,000 Cost 0.10 0.30 0.50 0.70 Max. Grant 0.18 0.54 0.60 0.84 Net Grant 0.08 0.24 0.10 0.14 Note: The Public Assistance System is part of Unemployment Insurance in Korea; Net Grant=Grant Cost in Rate (%). <Table 2> Public Assistance System for Promoting On-the-Job Training: Cost and Maximum Grant Relative to Wages (2002. 12. 30 ~ 2004. 9. 30) (Unit: Percent) Industry Group Firm Size 1~100 101~149 150~999 Over 1,000 Cost 0.10 0.10 0.50 0.70 Max. Grant 0.27 0.12 0.60 0.84 Net Grant 0.17 0.02 0.10 0.14 Industry Group Firm Size 1~149 150~500 501~999 Over 1,000 Cost 0.10 0.30 0.50 0.70 Max. Grant 0.27 0.81 0.60 0.84 Net Grant 0.17 0.51 0.10 0.14 Industry Group Firm Size 1~149 150~300 301~999 Over 1,000 Cost 0.10 0.30 0.50 0.70 Max. Grant 0.27 0.81 0.60 0.84 Net Grant 0.17 0.51 0.10 0.14..

94 韓國開發硏究 / 2011. Ⅲ <Table 3> Public Assistance System for Promoting On-the-Job Training: Cost and Maximum Grant Relative to Wages (2004. 10. 1 ~ 2005. 12. 31) Industry Group Firm Size 1~100 101~149 150~999 Over 1,000 Cost 0.10 0.10 0.50 0.70 Max. Grant 0.36 0.12 0.60 0.84 Net Grant 0.26 0.02 0.10 0.14 Industry Group Firm Size 1~149 150~500 501~999 Over 1,000 Cost 0.10 0.30 0.50 0.70 Max. Grant 0.36 1.08 0.60 0.84 Net Grant 0.26 0.68 0.10 0.14 Industry Group Firm Size 1~149 150~300 301~999 Over 1,000 Cost 0.10 0.30 0.50 0.70 Max. Grant 0.36 1.08 0.60 0.84 Net Grant 0.26 0.68 0.10 0.14 (Unit: Percent) <Table 4> Public Assistance System for Promoting On-the-Job Training: Cost and Maximum Grant Relative to Wages (After 2006. 1. 1) Industry Group Firm Size 1~100 101~149 150~999 Over 1,000 Cost 0.25 0.25 0.65 0.85 Max. Grant 0.60 0.25 0.65 0.85 Net Grant 0.35 0.00 0.00 0.00 Industry Group Firm Size 1~149 150~500 501~999 Over 1,000 Cost 0.25 0.45 0.65 0.85 Max. Grant 0.60 1.08 0.65 0.85 Net Grant 0.35 0.63 0.00 0.00 Industry Group Firm Size 1~149 150~300 301~999 Over 1,000 Cost 0.25 0.45 0.65 0.85 Max. Grant 0.60 1.08 0.65 0.85 Net Grant 0.35 0.63 0.00 0.00 (Unit: Percent)

재직자직업훈련관련공적재정의구조와성과 : 효과분석 95 나. 순지원율의경제적의미 7)..,.. (sunk cost),.. (2009), <Table 5>., ( ),.. <Table 5>, 151~500 30.2% 101~150 18.6% 1.5.,.., Oi(1999).,, 7)., ( ) ( ).. ( ) (, ). (omitted variables bias),, (bias).

96 韓國開發硏究 / 2011. Ⅲ <Table 5> The Ratio of Firms OJT Levy Burden to Total Labor Cost and The Ratio of Reimbursed Fund to Original Levy Burden (Unit: Percent) Industry Group Firm Size 1~100 101~150 151~500 501~999 Over 1,000 (1) Ratio of Levy to 0.199 0.367 0.698 0.646 0.650 Total Labor Cost (0.049) (0.165) (0.850) (0.464) (0.229) [N=3] [N=16] [N=53] [N=30] [N=36] (2) Ratio of Refund 7.7 20.9 19.2 37.7 37.3 to Actual Levy (13.3) (39.7) (26.7) (78.2) (29.0) [N=3] [N=16] [N=52] [N=27] [N=32] Industry Group Firm Size 1~100 101~150 151~500 501~999 Over 1,000 (1) Ratio of Levy to 0.483 0.889 0.656 0.623 0.805 Total Labor Cost (0.472) (2.322) (1.140) (0.483) (0.769) [N=13] [N=31] [N=153] [N=60] [N=51] (2) Ratio of Refund 8.1 18.6 30.2 24.7 38.2 to Actual Levy (11.6) (27.3) (38.2) (29.1) (37.5) [N=13] [N=31] [N=150] [N=58] [N=49] Note & Source: Human Capital Corporate Panel (HCCP), 2007; Cited from Kang and Yoo(2009). Standard errors in parentheses; N is the number of observations used for each estimation..,., (, 0%). 101~150 20.9%, 151~500 19.2%,., 101~150 151~500.,.

재직자직업훈련관련공적재정의구조와성과 : 효과분석 97 2. 계량분석모형 가. 훈련참여식추정모형 2) 고정효과통제로짓모형 Prob., (fixed effect logistic model)(chamberlin[1980]). 8) 1) 기본프로빗모형 Prob Prob (1) (2) Prob (3) (4), ( : 1 ) 1, 0 ; / (,,, ) ; (net) (policy). (2) (4). (fixed effect model) 8),. Chamberlain(1980), (conditional log likelihood estimation).

98 韓國開發硏究 / 2011. Ⅲ. 9), - (random).,.,., (variation) (within variation) (between variation),,... 나. 임금효과추정모형.. 1) 고정효과모형, Altonji and Spletzer(1991),.. (net grant) 9) 이때.

재직자직업훈련관련공적재정의구조와성과 : 효과분석 99.,. Lee(2008), (procyclical) (calibration).. ln, (5), (training) ( ),, ( ), ( :,, ),. 2) 일차차분임금성장률모형,., ln (6) ln : i t (real wage) ln ln ln g: ( =1), edu: a:, fs: firm size; 4 IDS: 3 DOJT: (on-the-job training) unemp: t

100 韓國開發硏究 / 2011. Ⅲ Ⅲ. 자료및분석결과.,,,,., (variation),.,. 1. 자료의구축. (t) (t-1). 10). (t+1).. 3 moving window. 3 window 10 7 10) p**4501(** ),,. 1, 0.

재직자직업훈련관련공적재정의구조와성과 : 효과분석 101., (fiscal benefits)., 150~399,,, 100, 450~499, 600~639, 640~ 649,. 11). 18 65. (missing) (999999).,,. -250% +250%. 2. 분석결과, ( 1 ). 17,538.. 11%. 12) <Table 6>. 11), 100~299, 150. 100~500 (n_grant) 0.63. (attenuation bias). 12) 15. P**4502, - -? 0 9, 1 8...

102 韓國開發硏究 / 2011. Ⅲ <Table 6> Descriptive Statistics-Sample 1 Variables* N Average Std. Err. Minimum Maximum age 17538 38.21 10.42 16 65 gender 17538 0.64 0.48 0 1 edu 17538 14.07 3.97 0 22 married 17534 0.70 0.46 0 1 firm_size (log) 5927 5.12 1.82 0.92 6.91 unemp. rate(%) 17538 4.08 0.89 3.3 6.3 manufact. 17538 0.28 0.45 0 1 mining 17538 0.18 0.38 0 1 others 17538 0.54 0.50 0 1 training=1 14602 0.11 0.32 0 1 n_grant 17538 0.14 0.08 0.08 0.68 ln(wage) 17538 4.96 0.59 1.20 8.05 dln(wage) 17538 0.11 0.36-2.44 2.46 dln(wage2) 14400 0.17 0.37-2.70 2.93 industry 17538 1.64 0.77 1 3 mon_wage(t-1) 17538 133.59 79.90 3 1000 mon_wage(t+1) 17538 160.70 103.34 3 3300 mon_wage(t+2) 14401 182.24 139.59 0 5300 year99 17538 0.12 0.33 0 1 year00 17538 0.13 0.34 0 1 year01 17538 0.14 0.34 0 1 year02 17538 0.14 0.35 0 1 year03 17538 0.15 0.35 0 1 year04 17538 0.15 0.36 0 1 year05 17538 0.17 0.37 0 1 Note: The sample is analyzed under a 3-year moving window ; Required is the information about changes in wages and whether or not sampled employees benefited from training over the 7-year period; variables with monetary units are measured in the 2005 price level. age = ; gender = ( =1; =0); edu = ; married = ( =1; =0); firm_size = (1-10 ) ; manufact. = ( 2); mining =,, / / ( 1); others = ( 3); training = ( =1; =0); n_grant = : net grant (grant-levy); dln(wage) = =log(w(t+1))-log(w(t-1)); dln(wage2) = =log(w(t+2))-log(w(t-1)); industry = (1-3); mon_wage(t-1) = 1 (monthly wage), ; mon_wage(t+1) = 1 ; mon_wage(t+2) = 2 ; year99 = 1999 1 0 ; year00 = 2000 1 0.

재직자직업훈련관련공적재정의구조와성과 : 효과분석 103 [Figure 1] Trends in the Ratio of OJT Recipients During the Sample Period 0.18 0.16 0.14 0.12 0.1 0.08 training 0.06 0.04 0.02 0 1999 2001 2002 2003 2004 2005 Note: Estimates from Korean Labor & Income Panel Study (KLIPS). n_grant,, 0 (0.14). 11%, 2. [Figure 1].,,.. 가. 훈련참여결정식 (participation equation) 의추정 1) 프로빗모형결과 <Table 7>. ( (1) ),

104 韓國開發硏究 / 2011. Ⅲ <Table 7> Estimates of Participation Equation Using a Probit Model Dependent Variable = training: training = 1 for participation Variables Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 n_grant 0.473*** 0.544*** 0.071 0.410* 0.192 0.517** 0.317 (0.153) (0.155) (0.208) (0.233) (0.235) (0.242) (0.244) age ---- -0.023*** -0.022*** -0.022*** -0.022*** -0.021*** -0.022*** (0.002) (0.003) (0.003) (0.003) (0.003) (0.003) gender ---- 0.169*** 0.209*** 0.211*** 0.239*** 0.233*** 0.263*** (0.030) (0.053) (0.053) (0.054) (0.054) (0.055) edu ---- 0.005-0.006-0.006-0.007-0.007-0.008 (0.004) (0.007) (0.007) (0.007) (0.007) (0.007) married ---- 0.329*** 0.321*** 0.320*** 0.315*** 0.323*** 0.317*** (0.036) (0.063) (0.063) (0.064) (0.063) (0.064) firm_size ---- ---- 0.259*** -1.736-1.851* -1.689-1.821* (0.016) (1.059) (1.074) (1.062) (1.077) firm_size2 ---- ---- ---- 0.881* 0.931** 0.863* 0.921** (0.460) (0.465) (0.461) (0.466) firm_size3 ---- ---- ---- -0.163** -0.170** -0.160** -0.169** (0.080) (0.081) (0.080) (0.081) firm_size4 ---- ---- ---- 0.010** 0.011** 0.010** 0.011** (0.005) (0.005) (0.005) (0.005) unemp. ---- ---- ---- ---- -0.260*** ---- -0.261*** (0.032) (0.032) manufact. ---- ---- ---- ---- ---- -0.069-0.088 (0.054) (0.055) mining ---- ---- ---- ---- ---- -0.135* -0.130* (0.072) (0.073) Constant -1.285*** -0.861*** -2.055*** -0.475 0.648-0.502 0.637 (0.027) (0.077) (0.171) (0.811) (0.835) (0.814) (0.838) #Obs 14602 14599 5191 5191 5191 5191 5191 Note: Standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1 ( (2) )., 1%p 9%p. 1%. 1.2%p ( (3) ),

재직자직업훈련관련공적재정의구조와성과 : 효과분석 105. (4),. 13),.,.,.,.,,, (Lee[2008] )... (1,2..,10) (imputation) n_grant.,, (net grant)..,, n_grant.,,. 13) firm_size, 4.

106 韓國開發硏究 / 2011. Ⅲ,.,,.. 2) 고정효과로짓모형결과 <Table 8> (fixed effect).,,. <Table 8> (logit model) 1, 2,.,. (variation). 1, 2 n_grant., 1%p 5.4%p 18%p..,. 나. 임금효과의추정 1) 임금성장률방정식의추정 <Table 9>, (training ).,

재직자직업훈련관련공적재정의구조와성과 : 효과분석 107 <Table 8> Estimates of Participation Equation Using a Fixed-Effect Logit Model Dependent Variable = training: training = 1 for participation Variables fixed effect model random effect model Model 1 Model2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8 n_grant 1.850*** 1.186*** 1.643*** 1.810*** 1.822*** 0.988* 0.988* 0.555 (0.446) (0.453) (0.370) (0.368) (0.366) (0.516) (0.516) (0.521) age ---- ---- ---- -0.034*** -0.051*** -0.044*** -0.044*** -0.046*** (0.004) (0.005) (0.007) (0.007) (0.007) gender ---- ---- ---- 0.456*** 0.375*** 0.458*** 0.458*** 0.529*** (0.094) (0.094) (0.126) (0.126) (0.129) edu ---- ---- ---- ---- 0.016-0.011-0.011-0.013 (0.012) (0.017) (0.017) (0.017) married ---- ---- ---- ---- 0.708*** 0.631*** 0.631*** 0.625*** (0.102) (0.146) (0.146) (0.148) firm_size ---- ---- ---- ---- ---- -4.461* -4.461* -4.708* (2.562) (2.562) (2.590) firm_size2 ---- ---- ---- ---- ---- 2.289** 2.289** 2.386** (1.105) (1.105) (1.117) firm_size3 ---- ---- ---- ---- ---- -0.425** -0.425** -0.439** (0.190) (0.190) (0.192) firm_size4 ---- ---- ---- ---- ---- 0.027** 0.027** 0.028** (0.011) (0.011) (0.011) unemp. ---- -0.475*** ---- ---- ---- ---- ---- -0.557*** (0.050) (0.071) Constant ---- ---- -3.353*** -2.367*** -2.370*** -0.775-0.775 1.670 (0.098) (0.183) (0.234) (1.972) (1.972) (2.015) #Obs 3385 3385 14602 14602 14599 5191 5191 5191 Note: Standard errors in parentheses. * means statistical significance at 10%, ** at 5%, and *** at 1%. net_grant=net Grant; firm_size=size of Firms; Industry Group (manu) represents manufacturing; Industry Group (mining) includes mining, construction, transportation/warehousing/ telecommunications; Industry Group I (others) includes all industries other than manufacturing, mining, construction and transportation/ warehousing/ telecommunications., 5.7% 1.9%.. <Table 9> dln(wage).,..

108 韓國開發硏究 / 2011. Ⅲ <Table 9> Effects of Training on Wages Dependent Variable: Wage Growth Rate = dln(wage) Variables Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 training 0.057*** 0.043*** 0.030** 0.031** 0.019 0.019 (0.009) (0.009) (0.014) (0.014) (0.014) (0.014) age ---- -0.009*** -0.005-0.005-0.005-0.005 (0.002) (0.004) (0.004) (0.004) (0.004) age2 ---- 0.000* -0.000-0.000-0.000-0.000 (0.000) (0.000) (0.000) (0.000) (0.000) gender ---- 0.007 0.008 0.007 0.011 0.012 (0.006) (0.010) (0.010) (0.010) (0.010) edu ---- -0.000 0.001 0.001 0.000 0.000 (0.001) (0.001) (0.001) (0.001) (0.001) married ---- 0.019** 0.001 0.002 0.003 0.003 (0.008) (0.013) (0.013) (0.013) (0.013) firm_size ---- ---- 0.006** 0.393*** 0.368** 0.372** (0.003) (0.151) (0.150) (0.150) firm_size2 ---- ---- ---- -0.168** -0.156** -0.158** (0.068) (0.068) (0.068) firm_size3 ---- ---- ---- 0.029** 0.027** 0.027** (0.012) (0.012) (0.012) firm_size4 ---- ---- ---- -0.002** -0.002** -0.002** (0.001) (0.001) (0.001) unemp. ---- ---- ---- ---- -0.035*** -0.035*** (0.005) (0.005) manufact. ---- ---- ---- ---- ---- 0.001 (0.011) mining ---- ---- ---- ---- ---- -0.006 (0.014) Constant 0.104*** 0.363*** 0.252*** -0.023 0.136 0.133 (0.003) (0.041) (0.069) (0.127) (0.128) (0.128) #Obs 14602 14599 5191 5191 5191 5191 R 2 0.003 0.024 0.025 0.026 0.036 0.036 Note: *** p<0.01, ** p<0.05, * p<0.1; First Difference OLS estimation. (Oi[1999]), firm_size, 4.,

재직자직업훈련관련공적재정의구조와성과 : 효과분석 109,. ( unemp )..,., ( : Altonji and Spletzer[1991]). 2) 고정효과모형을이용한임금효과추정 <Table 10> <Table 9>,., (fixed effect) (differencing).,. (within-variation)...,.. 3) 장기임금효과추정결과 2

110 韓國開發硏究 / 2011. Ⅲ <Table 10> Effects of Training on Wages Using a Fixed-Effect Model; Dependent Variable: log of Wages = ln(wages) Variable Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 training 0.057*** 0.017** 0.013 0.014 0.014 0.013 0.015* (0.009) (0.008) (0.014) (0.014) (0.014) (0.014) (0.008) age ---- 0.137*** 0.150*** 0.150*** 0.144*** 0.144*** 0.131*** (0.005) (0.011) (0.011) (0.011) (0.011) (0.006) age2 ---- -0.001*** -0.001*** -0.001*** -0.001*** -0.001*** -0.001*** (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) edu ---- 0.001-0.002-0.002-0.002-0.002 0.001 (0.002) (0.004) (0.004) (0.004) (0.004) (0.002) married ---- 0.047*** 0.055* 0.056* 0.056* 0.053* 0.046*** (0.016) (0.032) (0.032) (0.032) (0.032) (0.016) firm_size ---- ---- 0.007 0.302 0.300 0.311 ---- (0.005) (0.194) (0.194) (0.194) firm_size2 ---- ---- ---- -0.107-0.105-0.110 ---- (0.085) (0.085) (0.085) firm_size3 ---- ---- ---- 0.017 0.016 0.017 ---- (0.015) (0.015) (0.015) firm_size4 ---- ---- ---- -0.001-0.001-0.001 ---- (0.001) (0.001) (0.001) unemp. ---- ---- ---- ---- -0.019*** -0.019*** -0.020*** (0.007) (0.007) (0.003) manufact. ---- ---- ---- ---- ---- 0.048* 0.021 (0.029) (0.013) mining ---- ---- ---- ---- ---- -0.021 0.004 (0.036) (0.017) Constant 4.971*** 1.429*** 1.177*** 0.897*** 1.207*** 1.173*** 1.761*** (0.002) (0.109) (0.216) (0.259) (0.281) (0.281) (0.124) #Obs 14602 14599 5191 5191 5191 5191 14599 R 2 0.004 0.168 0.206 0.207 0.210 0.211 0.171 #Indivisuals 4904 4904 2648 2648 2648 2648 4904 Note: Standard errors in parentheses; fixed effect estimation. *** p<0.01, ** p<0.05, * p<0.1.,. <Table 11>., (specification), 1.

재직자직업훈련관련공적재정의구조와성과 : 효과분석 111 <Table 11> Effects of Training on Wages Dependent Variable = dln(wage2) Variables Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 training 0.070*** 0.050*** 0.007 0.007 0.006 0.006 (0.010) (0.010) (0.016) (0.016) (0.016) (0.016) age ---- -0.007*** -0.008*** -0.008*** -0.007*** -0.005 (0.000) (0.001) (0.001) (0.001) (0.004) age2 ---- ---- ---- ---- ---- -0.000 (0.000) gender ---- 0.015** 0.028** 0.028** 0.032*** 0.031** (0.007) (0.012) (0.012) (0.012) (0.012) edu ---- -0.000 0.001 0.001 0.001 0.001 (0.001) (0.001) (0.001) (0.001) (0.001) married ---- 0.016* 0.001 0.001 0.002-0.002 (0.008) (0.014) (0.014) (0.014) (0.016) firm_size ---- ---- 0.014*** 0.216 0.238 0.240 (0.003) (0.176) (0.176) (0.176) firm_size2 ---- ---- ---- -0.082-0.092-0.093 (0.079) (0.079) (0.079) firm_size3 ---- ---- ---- 0.013 0.015 0.015 (0.014) (0.014) (0.014) firm_size4 ---- ---- ---- -0.001-0.001-0.001 (0.001) (0.001) (0.001) manufact. ---- ---- ---- ---- 0.008 0.008 (0.012) (0.012) mining ---- ---- ---- ---- -0.030* -0.030* (0.016) (0.016) Constant 0.150*** 0.419*** 0.353*** 0.203 0.188 0.149 (0.004) (0.017) (0.034) (0.129) (0.129) (0.149) #Obs 12072 12069 4308 4308 4308 4308 R 2 0.004 0.041 0.052 0.053 0.054 0.054 Note: Standard errors in parentheses. * means it is statistically significant at 10%, ** at 5%, and *** at 1%. Standard errors in parentheses; First-Difference OLS estimation. *** p<0.01, ** p<0.05, * p<0.1. dln(wage)..

112 韓國開發硏究 / 2011. Ⅲ 3. 강건성점검 (Robustness Check) 1: 기업규모제약을가한표본이용 가. 기초통계량., 70 999 -.,,. 1,670. ( 2 ) <Table 12>. 14) 2 11%. 11.6%. <Table 12>, <Table 6>. 나. 훈련참여식의추정 <Table 13>,,. 14) (quasi-experimental approach)., (event).,.,, 150 149 151,, ( ).

재직자직업훈련관련공적재정의구조와성과 : 효과분석 113 <Table 12> Descriptive Statistics-Sample 2 Variables N Average Std. Err. Minimum Maximum age 1670 37.67 10.11 18 65 gender 1670 0.68 0.47 0 1 edu 1670 14.26 3.52 0 22 married 1670 0.71 0.45 0 1 firm_size (log) 1670 5.57 0.71 4.44 6.62 unemp. (%) 1670 4.09 0.89 3.3 6.3 manufact. 1670 0.38 0.49 0 1 mining 1670 0.17 0.38 0 1 others 1670 0.44 0.50 0 1 training=1 1463 0.11 0.32 0 1 n_grant 1670 0.21 0.18 0.08 0.68 ln(wage) 1670 5.07 0.55 2.83 7.07 dln(wage) 1670 0.12 0.32-1.84 1.86 dln(wage2) 1398 0.17 0.34-2.33 2.13 indus 1670 1.73 0.74 1 3 mon_wage(t-1) 1670 145.22 87.47 23 1000 mon_wage(t+1) 1670 176.26 108.91 15 1100 mon_wage(t+2) 1398 198.05 119.36 10 1150 year99 1670 0.13 0.33 0 1 year00 1670 0.12 0.33 0 1 year01 1670 0.13 0.33 0 1 year02 1670 0.16 0.33 0 1 year03 1670 0.16 0.37 0 1 year04 1670 0.15 0.36 0 1 year05 1670 0.19 0.39 0 1. 15) 15), 1%p 0.4%p 7.8%p,.

114 韓國開發硏究 / 2011. Ⅲ <Table 13> Participation Equation Estimation Dependent Variable = training: training = 1 for participation Variables Model 1 Model 2 Model 3 Model 4 Model 5 FE + n_grant 0.020 0.105 0.197 0.162 0.002 0.471 (0.232) (0.239) (0.243) (0.267) (0.269) (0.957) age ---- -0.038*** -0.037*** -0.037*** -0.037*** ---- (0.006) (0.006) (0.006) (0.006) gender ---- 0.109 0.106 0.104 0.134 ---- (0.100) (0.101) (0.101) (0.102) edu ---- -0.014-0.014-0.014-0.017 ---- (0.014) (0.014) (0.014) (0.014) married ---- 0.567*** 0.563*** 0.566*** 0.549*** ---- (0.125) (0.125) (0.125) (0.126) firm_size2 ---- ---- ---- 2.446 2.410 ---- (1.493) (1.511) firm_size3 ---- ---- ---- -0.589-0.580 ---- (0.361) (0.366) firm_size4 ---- ---- ---- 0.040 0.039 ---- (0.024) (0.025) firm_size ---- ---- 0.134** ---- ---- ---- (0.065) unemp ---- ---- ---- ---- -0.246*** ---- (0.064) Constant -1.216*** -0.156-0.956** -12.489* -11.253 ---- (0.067) (0.256) (0.466) (7.319) (7.405) #Obs 1463 1463 1463 1463 1463 195 Note: Standard errors in parentheses; Probit MLE estimation. *** p<0.01, ** p<0.05, * p<0.1 +: FE = Fixed Effect Logit Model 다. 임금효과의추정 (Table 14 ).,,

재직자직업훈련관련공적재정의구조와성과 : 효과분석 115 <Table 14> Effect of Training on Wages Variables Dependent Variable = dln(wage) Difference Model Fixed-Effect Model Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8 training 0.050 0.011 0.011 0.011 0.066** 0.049* 0.042 0.041 (0.032) (0.030) (0.030) (0.030) (0.026) (0.026) (0.026) (0.026) age ---- 0.053*** 0.054*** 0.054*** ---- -0.010-0.009-0.010 (0.005) (0.006) (0.007) (0.007) (0.007) (0.007) age2 ---- ---- ---- ---- ---- 0.000 0.000 0.000 (0.000) (0.000) (0.000) gender ---- 0.000 0.000 0.000 ---- 0.024 0.025 0.024 (0.000) (0.000) (0.000) (0.018) (0.018) (0.018) edu ---- 0.003 0.003 0.003 ---- 0.003 0.003 0.002 (0.008) (0.008) (0.008) (0.002) (0.002) (0.002) married ---- 0.089* 0.090* 0.090* ---- -0.012-0.014-0.013 (0.053) (0.053) (0.053) (0.025) (0.025) (0.025) firm_size ---- ---- 0.000 0.000 ---- ---- 0.000 0.000 (0.000) (0.000) (0.000) (0.000) firm_size2 ---- ---- -0.390-0.389 ---- ---- -0.321-0.326 (0.300) (0.300) (0.265) (0.265) firm_size3 ---- ---- 0.091 0.091 ---- ---- 0.079 0.080 (0.073) (0.073) (0.064) (0.064) firm_size4 ---- ---- -0.006-0.006 ---- ---- -0.005-0.005 (0.005) (0.005) (0.004) (0.004) unemp. ---- ---- ---- 0.002 ---- ---- -0.021** -0.022** (0.014) (0.009) (0.009) manufact. ---- ---- ---- ---- ---- ---- ---- -0.023 (0.018) mining ---- ---- ---- ---- ---- ---- ---- 0.006 (0.024) Const. 5.074*** 2.976*** 4.979*** 4.943*** 0.120*** 0.349*** 1.938 1.972 (0.007) (0.222) (1.470) (1.493) (0.009) (0.128) (1.300) (1.303) #Obs 1463 1463 1463 1463 1463 1463 1463 1463 R 2 0.005 0.185 0.191 0.191 0.004 0.032 0.038 0.040 #pid 951 951 951 951 ---- ---- ---- ---- Note: Standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1

116 韓國開發硏究 / 2011. Ⅲ 4. 강건성점검 2: 도구변수를이용한추정, (instrumental variable)., (endogeneity), (net grant). (net_grant). 16) (regression discontinuity model),., (+),,,.. Ⅳ. 결론및논의사항. 1~ 10. 3~4, (t-1), (t) (t+1)..,,.. 16), exact identification, over-identification test.

재직자직업훈련관련공적재정의구조와성과 : 효과분석 117, (stylized fact).,.,,.,.,.,,,.,,,,.,,.,, -.,.,., -., -

118 韓國開發硏究 / 2011. Ⅲ,.,. (regression discontinuity) 100%, (net_grant).,,.

재직자직업훈련관련공적재정의구조와성과 : 효과분석 119 참고문헌 Kang, Changhui and Gyeongjoon Yoo, Impact of the Korean Government s Policy for Employee Training Promotion on Firms' Investments in Training, Journal of Korean Economic Analysis, Vol. 15, No. 3, Dec. 2009, pp.209~254. (in Korean) Kim, Dong-Bae, Determinants of Employee Formal Training, Journal of Business Research, Vol. 7, No. 2, 2000, pp.119~134. (in Korean) Kim, Ahn-Kook, The Analysis of Productivity Effects of In-Firm Education and Training in Korea, KYONG JE HAK YON GU, Vol. 50, No. 3, 2002, pp.341~367. (in Korean) Kim, Ahn-Kook, The Determinants of Corporate Training in Korea Market Failure and State Intervention, Korean Journal of Labor Economics, Vol. 31, No. 1, 2008, pp.105~133. (in Korean) Ryu, Jang-Soo, The Size and Determinant Factors of Education Training Investment in Korean Manufacturing Firms, KYONG JE HAK YON GU, Vol. 45, No. 4, 1997, pp.227~249. (in Korean). Yoo, Gyeongjoon and Chulin Lee, The Estimation of the Effectiveness of Vocational Training for the Unemployed, Korean Journal of Labor Economics, Vol. 31, No. 1, 2008, pp.59~103. (in Korean).,,, 2002.,,,, 2009.,,,, 2004. Acemoglu, D. and J. S. Pischke, Why Do Firms Train? Theory and Evidence, The Quarterly Journal of Economics, Vol. 113, No. 1, Feb. 1998, pp.79~119. Altonji, J. and J. Spletzer, Worker Characteristics, Job Characteristics, and the Receipt of On-the-job Training, Industrial and Labor Relations Review, Vol. 45, 1991, pp.58~79. Becker, G. S., Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education, Columbia University Press (for NBER), New York, 1964. Ben-Porath, Y., The Production of Human Capital and the Life-Cycle of Earnings, Journal of Political Economy, Vol. 75, 1967, pp.352~365. Brown, C. and J. Medoff, The Employer Size-Wage Effect, Journal of Political Economy, Vol. 97, No. 5, 1989, pp.1027~1059.

120 韓國開發硏究 / 2011. Ⅲ Chamberlain, G., Analysis of Covariance with Qualitative Data, Review of Economic Studies 47, 1980, pp.225~238. Heckman, J., A Life-cycle Model of Earnings, Learning, and Consumption, Journal of Political Economy, Vol. 84, Part 2, 1976, pp.s11~s44., Sample Selection Bias as a Specification Error, Econometrica, Vol. 47, 1979, pp.153~161., Program Evaluation, The Economics and Econometrics of Active Labor Market Programs, Handbook of Labor Economics, Vol. 3, 1999, pp.1865~2097. Heckman J. and J. A. Smith, Evaluating the Welfare State, NBER Working Paper No. W6542, 1998. Heckman, J. and E. Vytlacil, Structural Equations, Treatment Effects, and Econometric Policy Evaluation, NBER Working Paper No. T0306, 2005., Econometric Evaluation of Social Programs, in J. Heckman and E. Leamer (ed.), Handbook of Econometrics, North-Holland, 2007a., Evaluating Marginal Policy Changes and the Average Effect of Treatment for Individuals at the Margin, Unpublished manuscript, Columbia University, 2007b. Heckman, J., E. Vytlacil, and S. Urzua, Understanding Instrumental Variables in Models with Essential Heterogeneity, Review of Economics and Statistics, Vol, 88, No. 3, 2006. Incer, J., Schooling, Experience and Earnings, Columbia University Press (for NBER), New York, 1974. Lee, C.-I., On-the-job Human Capital Investment and Intertemporal Substitution: New Evidence on Intertemporal Substitution Elasticity, Journal of Economic Dynamics and Control, Vol. 32, No. 10, 2008, pp.3350~3375. Mincer, J., Schooling, Experience and Earnings, Columbia University Press for National Bureau of Economic Research, New York, 1974. Oi, W. Y., Firm Size and Wages, O. Ashenfelter and D. Card (ed.), Handbook of Labor Economics, Vol. 3C, 1999. Stevens, M., A Theretical Model of On-the-job Training with Imperfect Competition, Oxford Economic Papers, Vol. 46, 1994, pp.537~562.