We first examined the sensitivity of the wage elasticity of Japanese married women to various economic and statistical assumptions by using a sample from the Japanese Panel Survey of Consumers. We then provided a new estimate of the female labor supply that simultaneously controls for the wage endogeneity, sample selection into labor force, and the
endogenous selection between different segments of the non-linear and often discontinuous budget constraint. We found that the OLS estimate of the wage elasticity substantially drops when job related characteristics are included in the model. The assumption of wage exogeneity is rejected. We found that the wife’s labor market experience is a valid excluded instrument while the wife’s education should be directly included in the hours worked equation in the 2SLS procedure. This finding validates most of the model specifications in the previous studies in Japan that utilize instrumental variable methods. The assumption of no-sample selection bias is rejected, though the correction of the bias in the Heckman two step procedure has little impact on the estimated wage elasticity.
Our new estimate of labor supply shows that there are notable differences in the labor supply behavior of women who choose different segments of the budget constraint. In particular, the wage elasticity of women who choose the budget segment I (annual income less than 1.03 million yen ceiling) is twice more negative (-1.28) than the women who choose the budget segment III (annual income greater than 1.41 million yen ceiling) (-0.60). In the case of the budget segment I, we cannot reject the null hypothesis that the wage elasticity is smaller than -1, suggesting that these women may be adjusting their hours of work so that their income does not exceeds the 1.03 million yen ceiling. The effect of the number of kids on the hours worked is also twice more negative for the budget segment I (-0.08) than for the budget segment III (-0.04). The effect of age on the hours worked is positive for the women in the budget segment III while it is negative for the women in the budget segment I. Younger women are more likely to choose the budget segment III while older women tend to choose the budget segment I. These results indicate that age-hours worked profile for Japanese married women has a hump-shape, a shape similar to the well-known M-shaped age-labor force participation profile for Japanese women. Our maximum likelihood estimation improved upon the previous literature in that it simultaneously controlled for
various sources of bias discussed in the literature, and that it provided a much more efficient estimate of the wage elasticity than the 2SLS procedure.
References
Abe, Yukiko, 2009. The Effects of the 1.03 Million Yen Ceiling in a Dynamic Labor Supply Model. Contemporary Economic Policy, April, Vol. 27, Issue 2, p147-163,
Abe, Yukiko and Fumio Otake. (1995). “Zeisei Shakai Hosho Seido to Part-Time Roudou-sha no Roudou Kyoukyuu Koudou (Tax and Social Security System and Part-time Workers’ Labor Supply Behavior), Kikan Shakai Hosho Kenkyu, Autumn, Vol. 31, No. 2, pp. 120-134 (in Japanese).
Akabayashi, Hideo, 2006. The Labor Supply of Married Women and Spousal Tax Deduc-tions in Japan. Review of Economics of the Household, Issue 4, pp. 349-378.
Bound, John; David A. Jaeger; and Regina Baker, 1995. Problems with instrumental variables estimation when the correlation between the instruments and the endogenous explanatory variable is weak. Journal of American Statistical Association, 90, pp.
443-450
Borjas, George J, 1980. The Relationship between Wages and Weekly Hours of Work: The Role of Division Bias. The Journal of Human Resources, Vol. 15, No. 3, Summer, pp. 409-423.
Debroux, Philippe, 2003. Human Resource Management in Japan: Changes and Uncer-tainties, Gower Publishing Co.
Hansen, Lars P, 1982. Large Sample Properties of the Generalized Method of Moments Estimators. Econometrica, 50, 1029-1054.
Hausman, Jerry A, 1980. The Effects of Wages, Taxes, and Fixed Costs on Women’s Labor Force Participation. Journal of Public Economics, 14, pp. 161-194.
Hayashi, Fumio, 2000. Econometrics. Princeton: Princeton University Press.
Heckman, Jamese, 1979. Sample Selection Bias as a Specification Error. Econometrica, 47, pp. 153-161.
Hill, Anne, 1989. Female Labor Supply in Japan: Implications of the Informal Sector for Labor Force Participation and Hours of Work. The Journal of Human Resources, Vol.
24, No. 1, Winter, pp. 143-161.
Ihori, Toshihiro, Ryuta Ray Kato, Masumi Kawade, and Shun-ichiro Bessho, 2006. Public Debt and Economic Growth in an Aging Japan. K. Kaizuka and A. O. Krueger eds Tackling Japan’s Fiscal Challenges, Palgrave.
Kato, Ryuta Ray, (2002) Government deficit, public investment, and public capital in the transition to an aging Japan. Journal of the Japanese and International Economies, Vol. 16, pp. 462-491.
Kuroda, Shoko and kaoru Yamamoto, 2008a. Ijitenkan no Roudou Kyoukyuu Dansei-chi (Frisch dansei-chi) no Keisoku. (The Estimation of the Intertemporal Labor Supply Elasticity (Frish-elasticity)). Mita Shougaku Kenkyuu, June, Vol. 51, No. 2, pp.
77-92 (in Japanese).
Kuroda, Shoko and kaoru Yamamoto, 2008b. Estimating Frish Labor Supply Elasticity in Japan. Journal of Japanese and International Economies, Vol. 22, pp. 566-585.
MaCurdy, Thomas E, 1981. An Empirical Model of Labor Supply in a Life-Cycle Setting.
The Journal of Political Economy, Vol. 89, No. 6, December, pp. 1059-1085
Moffit, Robert, 1986. The Economics of Piecewise-linear Budget Constraints. The Journal of Business and Economics Statistics, 4(3), pp. 317-328.
Mroz, Tomas A, 1987. The Sensitivity of an Empirical Model of Married Women’s Hours of Work to Economic and Statistical Assumptions. Econometrica, Vol. 55, No. 4, July, pp. 765-799.
National Institute of Population and Social Security Research, 2008. Population Projections for Japan: 2006 - 2055 December 2006. National Institute of Population and Social Security Research.
Nakamura, Jiro and Atsuko Ueda, 1999. On the Determinants of Career Interruption by Childbirth among Married Women in Japan. Journal of the Japanese and Interna-tional Economics, 13, pp. 74-89.
Oishi, Akiko, 2003. Yuu Haiguu Sha Josei no Roudou Kyoukyuu to Zeisei Shakai Hoshou Seido (Married women’s labor supply and the Tax-Social Security System). Kikan Shakai Hoshou Kenkyuu, Vol. 39, No. 3, Winter, pp. 286-305 (in Japanese).
Sasaki, Masaru, 2002. The Causal Effect of Family Structure on Labor Force Participation among Japanese Married Women. The Journal of Human Resources, Vol. 37, No. 2., Spring, pp. 429-440.
Stock, James H, and Morihiro Yogo, 2002. Testing for Weak Instruments in Linear IV Re-gression. National Bureau of Economic Research Technical Working Paper 284, Oc-tober. Downloadable at http://www.nber.org/papers/t0284 (Accessed July 30 2009).
Ueda, Atsuko, 2007. A Dynamic Decision Model of Marriage, Childbearing, and Labour Force Participation of Women in Japan. Japanese Economic Review, Vol. 58, Issue 4, December, pp. 443-465.
Wooldridge, Jefferey M, 2002. Econometric Analysis of Cross Section and Panel Data. The MIT Press Cambridge, Massachusetts
Table 1: National Income Tax Brackets in 2002
Taxable Income Range Marginal tax rate (%) in 1000 yen (y)
1≤y<3,300 10%
3,300≤y<9,000 20%
9,000≤y<18,000 30%
18,000 and more 37%
Tax schedule has been changed in 1995 and 1998. We took into account these changes when we computed the after tax wage rate.
Table 2: The Employee Tax Deduction Schedule in 2002
Gross Income Range Total Deduction
in 1000 yen (y) (Basic + Employer deduction)
1≤y<1,625 1,030
1,625≤y<1,800 0.4y+380 1,800≤y<3,600 0.3y+560 3,600≤y<6,600 0.2y+920 6,600≤y<10,000 0.1y+1,580
10,000 and more 0.5y+2,080
Tax deduction schedule was changed in 1995 and 2003. We took into account these changes when we computed the after tax wage rate.
Table 3: Summary of the Prior Literature
Oishi Abe & Otake Kuroda & Hill Akabayashi
(2003) (1995) Yamamoto (1989) (2006)
(2008a)
(1) Method OLS 2SLS IV + Heck- 3SLS Structural
man Selection Estimation
(2) Data Note(a) GSPT JPCS Note(b) GSPT
(3) Uncompen- -0.36*** -0.51*** 0.053 0.25 0.098***
sated wage elasticity to -0.24*** to 0.20 to 0.26** to 0.245***
(4) Explanatory Variables
(4-1) Basic Variables
•Log(wage) ° ° ° ° °
•Non-wife income ° ° ° °
•Wife’s age ° ° ° ° °
•Dummy kids≤age 6 ° ° °
(4-2) Tax Related
•Not eligible for SAS °
•Husbands’ °
social security is category II
•Income adjustment °
dummy
(4-3) Household Characteristics
•Living with parents ° °
(4-4) Household Assets
•Amount of saving °
•Amount of borrowing °
•Have debt °
•Own a house °
(4-5) Work Characteristics
•Industry dummies °
•# employees °
(4-6) Human Capital Charact.
•Wife’s education ° °
•Husband’s education °
•Labor market °
experience
(4-7) Other variables
•Large city dummies °
•Prefectural unemployment ° °
•Consumer Price Index ° °
(5) Excluded Instruments
•Reported hourly °
wage
•Labor market ° °
experience
•Above squared °
•Prefectural °
average income
•Industry dummies °
•# employees °
•Occupation dummies °
Notes: (a) Kokumin Seikatsu Kiso Chousa. (b) 1975 Survey of women in Tokyo Metropolitan Area. (c) *,
**, *** significant at 10%, 5% and 1%.
Table 4: Summary Statistics
Segment I Segment II Segment III All
(#Obs=1413) (#Obs=231) (#Obs=1555) (#Obs=3199)
Mean Mean Mean Mean
(Std.Dev) (Std.Dev) (Std.Dev) (Std.Dev)
Dependent variable
Hours worked 940.033 1561.375 2078.038 1538.072
(437.910) (523.879) (532.276) (736.148)
Explanatory variables Basic Variables
Wage (After tax) 761.413 645.814 1487.486 1106.002
(378.447) (346.085) (655.464) (647.415)
Non-wife income 437.788 379.929 382.903 406.931
(169.688) (128.590) (130.231) (151.321)
Age 35.452 35.017 34.350 34.885
(4.265) (4.159) (4.692) (4.501)
# Kids≤age 6 0.485 0.541 0.556 0.524
(0.721) (0.732) (0.759) (0.741)
Household Characteristics
†Living with parents 0.401 0.372 0.473 0.434
(0.490) (0.484) (0.499) (0.496)
Assets
Assets (in 10,000 yen) 483.971 376.269 686.442 574.613
(1,021.527) (808.890) (1,340.496) (1,180.478)
Monthly loan 8.576 8.079 8.721 8.611
payment (in 10,000 yen) (17.753) (13.802) (18.509) (17.870)
†Own a house 0.587 0.450 0.572 0.570
(0.492) (0.499) (0.495) (0.495)
Job Characteristics
†Manufacturing 0.143 0.160 0.176 0.160
(0.350) (0.368) (0.381) (0.367)
†Retail 0.348 0.355 0.139 0.247
(0.477) (0.480) (0.346) (0.431)
†Service 0.311 0.351 0.271 0.294
(0.463) (0.478) (0.444) (0.456)
# Employees 265.188 272.937 514.935 387.147
(368.191) (366.456) (434.167) (420.197)
†Full-time 0.042 0.238 0.826 0.437
(0.200) (0.427) (0.379) (0.496)
†Managerial 0 0 0.007 0.003
position - - (0.084) (0.059)
Human Capital and Other Characteristics
Wife’s labor market 11.506 12.750 14.345 12.976
experience (years) (4.281) (4.102) (4.827) (4.743)
Wife’s education 14.102 14.017 14.408 14.245
(years) (1.068) (1.374) (1.122) (1.130)
Mother-in-law’s 10.735 10.528 10.723 10.714
education(years) (1.746) (1.686) (1.565) (1.657)
(a) † indicates that the variable is a dummy variable.
Table 5: OLS Results for the Budget Segment III
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Basic variables
Log(After -0.342a -0.298a -0.344a -0.365a -0.481a -0.503a -0.489a tax wage) (0.018) (0.016) (0.018) (0.019) (0.021) (0.021) (0.021)
Non-wife income 0.045a 0.040a 0.039a 0.041a 0.041a 0.038a
(in 1 million yen) (0.008) (0.008) (0.008) (0.008) (0.008) (0.007)
Age 0.003b 0.007a 0.001 0.0001 0.003 -0.006a 0.003
(0.002) (0.002) (0.002) (0.002) (0.002) (0.002) (0.002)
# Kids≤age 6 -0.008 -0.010 -0.013 -0.018b -0.018b -0.016b -0.018b (0.009) (0.009) (0.009) (0.009) (0.008) (0.008) (0.008) Household characteristics
Living with 0.030b 0.030b 0.007 -0.002 0.007
parents (0.015) (0.015) (0.014) (0.014) (0.014)
Asset variables
Assets 0.002a 0.002a 0.001a 0.001b 0.001b
(in 1 million yen) (0.001) (0.001) (0.001) (0.001) (0.001)
Monthly Loan -0.014 -0.010 -0.004 -0.003 -0.001
(in 1 million yen) (0.028) (0.028) (0.026) (0.026) (0.026)
Own a house 0.021 0.023 0.009 0.008 0.009
(0.015) (0.015) (0.015) (0.014) (0.015) Job characteristics
Manufacturing -0.070a -0.066a -0.069a -0.059a
(0.016) (0.015) (0.014) (0.014)
Retail -0.088a -0.024 -0.010 -0.016
(0.022) (0.020) (0.020) (0.020)
Service -0.051a 0.008 0.017 0.008
(0.017) (0.017) (0.017) (0.017)
log(#employees) 0.033a 0.030a 0.028a
(0.004) (0.004) (0.005)
Full-time 0.224a 0.203a 0.218a
(0.019) (0.019) (0.019)
Managerial 0.123c 0.123c 0.111c
Position (0.065) (0.068) (0.063)
Human capital characteristics
Labor market 0.012a
experience (0.002)
Education 0.041a 0.033a
(years) (0.007) (0.006)
Constant 9.798a 9.516a 9.869a 10.080a 10.462a 10.205a 10.091a (0.128) (0.120) (0.130) (0.141) (0.134) (0.155) (0.153)
Year dummies Yes Yes Yes Yes Yes Yes Yes
R squared 0.259 0.225 0.272 0.286 0.403 0.430 0.417
#obs 1555 1555 1555 1555 1555 1555 1555
Note: (a) Inside the brackets are robust standard errors. (b) a Significant at 1%, b Signifi-cant at 5%, c SignifiSignifi-cant at 10%.
Table 6: OLS Results for the Budget Segment I
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Basic variables
Log(After tax -1.300a -1.303a -1.298a -1.295b -1.285a -1.243a -1.283a
wage) (0.042) (0.042) (0.042) (0.042) (0.043) (0.041) (0.043)
Non-wife income -0.007 -0.010 -0.009 -0.010 0.001 -0.009
(in 1 million yen) (0.007) (0.008) (0.008) (0.007) (0.008) (0.008)
Age -0.001 -0.002 -0.002 -0.003 -0.003 -0.014a -0.002
(0.004) (0.004) (0.004) (0.004) (0.004) (0.005) (0.004)
# Kids≤age 6 -0.076a -0.075a -0.078a -0.078a -0.086a -0.081a -0.086a (0.026) (0.025) (0.025) (0.025) (0.025) (0.025) (0.025) Household characteristics
Living with 0.011 0.004 -0.002 -0.016 -0.002
parents (0.032) (0.032) (0.033) (0.032) (0.033)
Asset variables
Asset 0.002b 0.002b 0.002a 0.002b 0.002b
(in 1 million yen) (0.001) (0.001) (0.001) (0.001) (0.001)
Monthly Loan 0.078 0.072 0.073 0.041 0.067
(in 1 million yen) (0.070) (0.069) (0.069) (0.069) (0.069)
Own a house 0.025 0.026 0.033 0.031 0.034
(0.033) (0.033) (0.033) (0.032) (0.033) Job characteristics
Manufacturing 0.045 0.074c 0.088b 0.072c
(0.041) (0.043) (0.042) (0.042)
Retail -0.096a -0.066c -0.010 -0.066c
(0.037) (0.039) (0.039) (0.039)
Service -0.026 0.007 0.065 0.009
(0.040) (0.044) (0.043) (0.044)
log(#employees) 0.013c 0.016b 0.013c
(0.007) (0.007) (0.007)
Full-time 0.175a 0.150a 0.176a
(0.054) (0.053) (0.054) Human capital characteristics
Labor market 0.030a
experience (0.004)
Wife’s -0.004 -0.012
education (0.014) (0.014)
Constant 15.194a 15.209a 15.200a 15.239a 15.092a 14.880a 15.235a (0.325) (0.326) (0.323) (0.326) (0.330) (0.344) (0.361)
R squared 0.437 0.437 0.438 0.443 0.447 0.471 0.447
Year dummies Yes Yes Yes Yes Yes Yes Yes
#obs 1413 1413 1413 1413 1413 1413 1413
Note: (a) Inside the brackets are robust standard errors. (b) a Significant at 1%, b Signifi-cant at 5%, c SignifiSignifi-cant at 10%.
Table 7: Specification Tests for the 2SLS Models
Model Log(wage) Excluded Instruments Instruments Hansen’s J C-stat Exogeneity Coefficient & Comments (Underlined Relevance (pval) (pval) test (Wage)
(StDev) are the tested instruments) F-stat(pval) (pval)
Results for the Budget Segment III(# obs=1555) OLS5 -0.481***
(0.021)
IV1 0.079 Basic 4.48 0.393 - 16.16
(0.42) (0.00) (0.94) - (0.00)
IV2 0.067 Basic+Exp 8.73 0.42 0.02 32.47
(0.123) (0.00) (0.98) (0.88) (0.00)
IV3 0.191 Basic+Exp+WifeEduc 8.87 3.69 3.34 53.15
(0.129) (0.00) (0.69) (0.068) (0.00)
IV4 0.009 Basic 3.65 0.60 - 10.70
(0.198) (WifeEduc in Hours (0.006) (0.90) - (0.00)
worked equation)
IV5 0.096 Basic+Exp 9.48 0.71 0.18 39.70
(0.123) (WifeEduc in Hours (0.00) (0.96) (0.67) (0.00)
worked equation)
IV6 0.075* Basic+Exp+(Work 44.65 9.79 8.82 145.58
(0.042) Charact.). (WifeEduc in (0.00) 0.46) (0.18) (0.00) Hours worked equation)
Results for the Budget Segment I (# obs=1413) OLS5 -1.285***
(0.043)
IV1 -5.007*** Basic 1.02 0.623 - 25.52
(2.001) (0.395) (0.891) - (0.00)
IV2 -3.950*** Basic+Exp 5.28 1.59 0.53 63.83
(0.620) (0.00) (0.81) (0.47) (0.00)
IV3 -3.562*** Basic+Exp+WifeEduc 4.72 5.38 3.396 52.93
(0.539) (0.00) (0.37) (0.065) (0.00)
IV4 -5.306*** Basic 0.76 0.77 - 22.82
(2.503) (WifeEduc in Hours (0.55) (0.86) - (0.00)
worked equation)
IV5 -4.008*** Basic+Exp 4.66 2.82 0.65 59.51
(0.66) (WifeEduc in Hours (0.00) (0.72) (0.42) (0.00)
worked equation)
IV6 -2.017*** Basic+Exp+(Work 10.66 56.56 47.52 21.22
(0.176) Charact.). (WifeEduc in (0.00) (0.00) (0.00) (0.00) Hours worked equation)
Basic={Mother’s education, its squared term, mother-in-law’s education and its squared term}
Work Charact. ={Industry Dummies, # of employees, full-time dummy, managerial dummy}
Notes: (a) Instruments relevance test is the F-test for the null hypothesis that the excluded instruments are not jointly significant in the first stage wage regression. (b) Hansen’s J tests the null hypothesis that the over-identifying restrictions are valid. (c) C-statistic tests the orthogonality of the excluded instruments underlined in the third column. (d) Exogeneity test is the test for the null hypothesis that the suspected endogenous variable, log(wage), is actually exogenous. (e) The entire results for our preferred model (IV5) are shown in Table 10. (f) Insider the brackets are robust standard errors. (g) *, **, *** significant at 10%, 5%, 1% levels.
Table 8: Heckman Selection Bias Correction Results
Heckman Selection Bias Correction 2SLS&Heckman Selection(a) Segment III(#obs=6124) Segment I(#obs=5982) Segment III Segment I
Hours Selection Hours Selection Hours Hours
equation equation equation equation equation equation
log(After tax -0.505a -1.237a -0.043 -1.958c
wage) (0.015) (0.040) (0.159) (1.179)
Non-wife income 0.044a -0.077a -0.006 0.043a 0.001 0.003 (in 1 million yen) (0.005) (0.020) (0.009) (0.012) (0.016) (0.017)
age -0.0003 -0.148a -0.014a -0.016b 0.000 -0.016a
(0.002) (0.009) (0.005) (0.007) (0.002) (0.006)
#Kids≤age 6 0.001 -0.458a 0.040 -0.487a -0.026b 0.037
(0.008) (0.032) (0.026) (0.029) (0.013) (0.032)
Living with -0.007 0.233a -0.044 0.146a 0.059b -0.071
parents (0.014) (0.064) (0.034) (0.052) (0.026) (0.056)
Assets 0.001a -0.003 0.004a -0.012a 0.0004 0.004a
(in 1 million yen) (0.0004) (0.002) (0.001) (0.002) (0.001) (0.001)
Monthly loan -0.003 0.050 0.040 -0.039 -0.007 0.030
(in 1 million yen) (0.030) (0.138) (0.081) (0.120) (0.029) (0.075)
Own a house -0.00001 0.138b -0.027 0.212a 0.004 -0.018
(0.014) (0.061) (0.035) (0.049) (0.017) (0.037)
Manufacturing -0.070a 0.082c -0.011 0.008
(0.016) (0.049) (0.027) (0.129)
Retail -0.011 -0.009 0.040 -0.046
(0.018) (0.040) (0.031) (0.074)
Service 0.019 0.062 0.035 0.126
(0.015) (0.043) (0.022) (0.118)
log(#emploees) 0.030a 0.016b 0.006 0.018b
(0.004) (0.007) (0.010) (0.009)
Full-time 0.204a 0.158b 0.063 0.016
(0.016) (0.068) (0.055) (0.253)
Managerial 0.131b 0.087
position (0.065) (0.084)
Wife’s 0.029a 0.246a -0.007 0.031c 0.018b 0.006
education (0.005) (0.023) (0.013) (0.017) (0.008) (0.028)
Mother’s 0.118 0.077
education (0.187) (0.170)
(Mother’s -0.005 -0.004
Education)2 (0.008) (0.008)
Mother-in-law’s -1.400a -0.910a
education (0.216) (0.172)
(Mother-in-law’s 0.062a 0.040a
education)2 (0.010) (0.008)
Labor market 0.299a 0.154a
experience (0.008) (0.006)
Constant 10.381a 5.222a 15.531a 2.870b 7.547a 19.983a
(0.132) (1.576) (0.348) (1.294) (0.983) (7.259)
Inverse Mill’s -0.083a -0.352a -0.032 -0.272b
Ratio (0.012) (0.039) (0.025) (0.137)
rho -0.381 -0.615
Hansen’s J (pval) (0.40) (0.02)
Wage exogenous (0.0001) (0.984)
test(pval)
Note: (a) The first stage wage regression results are shown in Table 10. P-values for the instrument relevance test are 0.0006 for the Segment III, and 0.815 for the segment I. (b) a Significant at 1%, b Significant at 5%, c Significant at 10%. (c) All equations include year
Table 9: Joint estimation incorporating unobserved heterogeneity term
Segment III Segment I Segment Labor force
Hour Wage Hours Wage selection participation
equation equation equation equation equation(a) equation(b)
Log(After tax -0.603a -1.287a
wage) (0.014) (0.055)
Non-wife income 0.033a 0.072a -0.006 0.014a -0.453a 0.032 (1 million yen) (0.004) (0.006) (0.012) (0.005) (0.073) (0.040)
Age 0.003 -0.013a -0.003 0.001 -20.290a -0.121a
(0.002) (0.003) (0.005) (0.003) (3.877) (0.030)
# Kids≤age 6 -0.043a 0.005 -0.079a 0.054a -1.118a -1.355a (0.008) (0.011) (0.022) (0.013) (0.148) (0.067)
Living with 0.027c -0.100a -0.009 -0.063a 0.459c 0.606a
parents (0.015) (0.019) (0.030) (0.022) (0.239) (0.164)
Assets 0.007 0.0005 0.002 0.0007 -0.002 -0.018a
(1 million yen) (0.005) (0.001) (0.002) (0.001) (0.022) (0.005)
Monthly loan 0.001 0.025 0.064 -0.017 0.355 0.103
(1 million yen) (0.004) (0.058) (0.115) (0.081) (0.527) (0.353)
Own a house 0.008 0.004 0.028 -0.020 0.006 0.526a
(0.014) (0.020) (0.034) (0.023) (0.226) (0.152) Manufacturing -0.031c -0.080a 0.074 -0.102a -0.351
(0.018) (0.020) (0.066) (0.032) (0.250)
Retail -0.044a -0.125a -0.053 -0.036 -1.195a
(0.017) (0.025) (0.046) (0.025) (0.239)
Service -0.015 -0.048b 0.021 0.096a -0.288
(0.014) (0.022) (0.047) (0.025) (0.239) log(#employees) 0.019a 0.035a 0.013c 0.003 0.146a (0.003) (0.005) (0.007) (0.005) (0.048)
Full-time 0.279a 0.323a 0.172 -0.197a 4.676a
(0.014) (0.020) (0.133) (0.038) (0.215)
Managerial 0.042 -0.012
position (0.118) (0.104)
Wife’s 0.460a 0.053a -0.155 -0.004 0.571a 0.349a
education (0.054) (0.009) (0.141) (0.009) (0.122) (0.083)
Mother’s 0.134b 0.001 0.653 0.047
education (0.059) (0.088) (0.879) (0.658)
(Mother’s -0.006b 0.0004 -0.033 -0.007
education)2 (0.003) (0.004) (0.039) (0.029)
Mother-in-law’s 0.034 0.093 -1.577c -3.671a
education (0.069) (0.076) (0.837) (0.761)
(Mother-in-law’s -0.001 -0.004 0.067c 0.159a
education)2 (0.003) (0.003) (0.036) (0.034)
Labor market 0.022a -0.027a 0.437a 0.617a
experience (0.003) (0.004) (0.039) (0.029)
Constant 10.621a 4.827a 15.320a 6.143a -0.932 14.177a
(0.125) (0.506) (0.445) (0.592) (6.303) (5.547) χi (ρj for -0.142a -0.160a 0.035 0.136a -2.841a -2.941a
j=1,..6) (0.008) (0.010) (0.023) (0.023) (0.243) (0.137)
Year dummies Yes Yes Yes Yes Yes Yes
Note: (a) The ‘dependent variable’ isBit=1 if the individual chooses the budget segment III, and Bit=0 if otherwise. (b) The ‘dependent variable’ is Iit=1 if the individual participate in the labor force, and Iit=0 if otherwise. (c) a Significant at 1%, b Significant at 5%, c Significant at 10%. (d) Log likelihood = -4825.54906. (e) The null hypothesis that ρj for j=1,...,6 are simultaneously equal to zero is rejected (log likelihood ratio test statistic is 2018.269).
Table 10: Data Appendix : Other results
2SLS Results (IV5 in Table 7) Fixed effect log 2SLS&Heckman hours worked selection equations
Segment III Segment I equations (See Table 8)
Hour eq. Wage eq. Hour eq. Wage eq. Seg. III Seg. I Seg. III Seg. I
log(After tax 0.096 -4.008a -0.648a -1.214a
wage) (0.123) (0.665) (0.034) (0.063)
Non-wife income -0.012 0.086a 0.028 0.010 0.015 0.006 0.087a 0.011 (1 million yen) (0.013) (0.009) (0.026) (0.008) (0.011) (0.012) (0.009) (0.008)
age 0.001 -0.009b -0.021b -0.003 0.018a 0.032a -0.004 -0.003
(0.002) (0.004) (0.010) (0.003) (0.003) (0.010) (0.006) (0.003)
# Kids≤age 6 -0.038a 0.037a 0.013 0.034a -0.036a -0.091b 0.050a 0.013 (0.013) (0.013) (0.058) (0.017) (0.011) (0.041) (0.016) (0.039) Living with 0.082a -0.136a -0.143b -0.048b 0.014 0.036 -0.140a -0.043b
parents (0.024) (0.024) (0.071) (0.020) (0.033) (0.088) (0.024) (0.021)
Assets 0.000 0.001b 0.003 0.000 (0.0004) 0.001 0.001b -0.00002
(1 million yen) (0.001) (0.001) (0.003) (0.001) (0.0004) (0.001) (0.001) (0.001)
Monthly Loan -0.008 0.015 0.005 -0.015 0.013 0.031 0.015 -0.016
(1 million yen) (0.033) (0.047) (0.129) (0.039) (0.028) (0.063) (0.047) (0.039)
Own a house 0.007 0.008 0.017 -0.003 -0.038 0.044 0.002 0.007
(0.019) (0.024) (0.067) (0.021) (0.023) (0.055) (0.025) (0.027) Manufacturing 0.008 -0.121a -0.203c -0.105a 0.012 0.153 -0.121a -0.104a
(0.025) (0.027) (0.109) (0.030) (0.067) (0.095) (0.027) (0.030) Retain 0.054c -0.114a -0.159c -0.053c 0.022 0.060 -0.114a -0.051c
(0.032) (0.032) (0.084) (0.028) (0.078) (0.095) (0.032) (0.028)
Service 0.037 -0.032 0.301a 0.085a 0.044 0.098 -0.032 0.087a
(0.024) (0.026) (0.121) (0.031) (0.050) (0.081) (0.026) (0.031) log(# Employees) -0.002 0.052a 0.025c 0.003 -0.013 0.037a 0.052a 0.004
(0.009) (0.007) (0.014) (0.005) (0.010) (0.014) (0.007) (0.005) Full-time 0.025 0.303a -0.385 -0.195a 0.070c 0.094 0.302a -0.196a
(0.048) (0.027) (0.271) (0.079) (0.038) (0.073) (0.027) (0.080)
Managerial 0.070 0.065 0.084a 0.070
position (0.093) (0.149) (0.020) (0.144)
Wife’s 0.015c 0.037a 0.044 0.017 0.028b 0.019c
education (0.008) (0.009) (0.036) (0.011) (0.012) (0.011)
Mother’s 0.215a 0.000 0.206a 0.005
education (0.065) (0.072) (0.066) (0.073)
(Mother’s -0.009a 0.000 -0.009a 0.000
education)2 (0.003) (0.003) (0.003) (0.003)
Mother-in-law’s 0.095 -0.011 0.127 -0.047
education (0.081) (0.070) (0.087) (0.093)
(Mother-in-law -0.004 0.000 -0.005 0.002
education)2 (0.004) (0.003) (0.004) (0.004)
Labor Market 0.019a -0.011a 0.010 -0.005
experience (0.003) (0.003) (0.009) (0.009)
Constant 6.650 4.182a 32.566a 6.451a 11.626a 13.262a 4.169a 6.493a (0.746) (0.558) (4.240) (0.510) (0.229) (0.555) (0.558) (0.513)
Year dummies Yes Yes Yes Yes Yes Yes Yes Yes
R squared(a) 0.389 0.106 0.551 0.391 0.389 0.107
(within) (within)
# obs 1555 1555 1413 1413 1555 1513 6124 5982
Note (a) R squared for the 2SLS hours worked equations are not reported since they are negative. (b) a Significant at 1%, b Significant at 5%, c Significant at 10%
Figure 1: Budget Constraint for a Typical Wife
Figure 2: Graphs of Annual Income, Wage and Hours Worked
051015Percent
0 500 1000
Wives’ Pre−tax Annual Income in 10,000 yen Figure A: Wive’s Pre−tax Annual Income
05101520Percent
0 1000 2000 3000 4000 5000
Wives’ After Tax Wage in 1 yen
Figure B: After Tax Wage
150020002500Average hours worked by age
25 30 35 40 45
Wife’s age
Figure C: Average hours worked by age