Dynamics of overqualification: evidence from the early career of graduates

Figures & data

Table 1. Summary statistics.

	Mean	SD	Min	Max
Dependent Variable
Overqualification (t = 1)	0.16	0.37	0	1
Overqualification (t = 5)	0.16	0.37	0	1
Overqualification (t = 10)	0.14	0.35	0	1
Time-invariant controls
Female	0.51	0.50	0	1
University grade${}^{\mathrm{a}}$	3.03	0.63	1.0	4.0
School grade${}^{\mathrm{a}}$	2.83	0.63	1.0	4.0
Study duration${}^{\mathrm{b}}$	11.12	2.55	6	20
University of applied sciences	0.28	0.45	0	1
Business Administration & Economics	0.15	0.36	0	1
Medicine, Law, & Teaching	0.24	0.43	0	1
STEM subjects	0.43	0.50	0	1
Social & Cultural science	0.18	0.38	0	1
Age at graduation	26.88	2.36	22	35
Parents high-skilled${}^{\mathrm{c}}$	0.51	0.50	0	1
Cohort: 1997	0.51	0.50	0	1
Time-variant controls${}^{\mathrm{d}}$
Married	0.44	0.50	0	1
Children	0.33	0.47	0	1
Months unemployed	1.93	3.77	0	53
Observations	6467

Note: t depicts the time of observation 1, 5, and 10 years after graduation; ${}^{\mathrm{a}}$ 1 is the lowest grade and 4 is the highest, i.e. the best, grade; ${}^{\mathrm{b}}$ Measured in semesters; ${}^{\mathrm{c}}$ At least one parent obtained a tertiary degree; ${}^{\mathrm{d}}$ Overall statistics are presented. Source: DZHW-Graduate Panel 1997, 2001

Table 2. Results, dynamic pooled probit.

	Dependent Variable: Overqualification
	(1)	(2)	(3)	(4)
Overqualification $(t-5)$	0.444*** (0.012)	0.424*** (0.013)	0.361*** (0.013)	0.342*** (0.013)
Female	0.010* (0.006)	0.014** (0.006)	0.020*** (0.006)	0.018*** (0.006)
University grade				−0.019*** (0.003)
School grade				−0.007** (0.003)
Study duration				0.009*** (0.003)
U. applied sciences			0.021*** (0.006)	0.020*** (0.007)
BusAdmin/Econ			ref.	ref.
Med/Law/Teach			−0.119*** (0.007)	−0.121*** (0.007)
STEM subjects			−0.056*** (0.007)	−0.058*** (0.007)
Soc/Cult science			−0.021*** (0.008)	−0.020** (0.008)
Age at grad. < 27		−0.029*** (0.006)	−0.023*** (0.006)	−0.010 (0.006)
Parents high-skilled		−0.019*** (0.006)	−0.009 (0.006)	−0.007 (0.006)
Married		−0.008 (0.007)	−0.007 (0.007)	−0.006 (0.007)
Children		0.013 (0.007)	0.015** (0.007)	0.015** (0.007)
Months unemployed		0.004*** (0.001)	0.005*** (0.001)	0.004*** (0.001)
Log. Lik.	−4383.6	−4327.8	−4186.3	−4147.3
Obs. Indiv.	6467	6467	6467	5987
Obs. Total	12934	12934	12934	11974

Note: Pooled Probit estimation; Average partial effects; Standard errors clustered at the individual level; Previous period indicated by $(t-5)$; Significant at 1% ***, significant at 5% **, significant at 10% *.

Table 3. Results, Wooldridge dynamic RE probit.

	Dependent Variable: Overqualification
	All	Female	Male
	(1)	(2)	(3)
Overqualification $(t-5)$	0.029** (0.013)	0.033* (0.018)	0.028 (0.021)
Overqualification $(t=1)$	0.224*** (0.023)	0.207*** (0.029)	0.248*** (0.037)
Female	0.015** (0.006)
University grade	−0.018*** (0.003)	−0.022*** (0.004)	−0.014*** (0.004)
School grade	−0.006** (0.003)	−0.012** (0.005)	−0.003 (0.004)
Study duration	0.008*** (0.003)	0.011** (0.004)	0.006 (0.004)
U. applied sciences	0.019*** (0.007)	0.023** (0.010)	0.016* (0.009)
BusAdmin/Econ	ref.	ref.	ref.
Med/Law/Teach	−0.095*** (0.006)	−0.114*** (0.010)	−0.076*** (0.008)
STEM subjects	−0.053*** (0.008)	−0.054*** (0.012)	−0.052*** (0.011)
Soc/Cult science	−0.017** (0.008)	−0.026** (0.012)	−0.006 (0.012)
Age at grad. < 27	−0.008 (0.006)	−0.002 (0.009)	−0.013 (0.008)
Parents high-skilled	−0.008 (0.006)	−0.009 (0.008)	−0.007 (0.008)
Married	−0.005 (0.009)	−0.008 (0.013)	−0.001 (0.013)
Children	0.009 (0.009)	0.018 (0.013)	0.000 (0.013)
Months unemployed	0.001 (0.001)	0.001 (0.002)	0.000 (0.002)
Mean: Married	−0.004 (0.012)	−0.001 (0.017)	−0.008 (0.017)
Mean: Children	0.010 (0.012)	0.015 (0.017)	0.008 (0.017)
Mean: Months unemployed	0.004** (0.001)	0.004* (0.002)	0.003 (0.002)
Occupation FE:	No	No	No
σ	1.222 (0.099)	1.353 (0.145)	1.078 (0.137)
ρ	0.599 (0.039)	0.647 (0.049)	0.537 (0.063)
Log. Lik.	−4073.9	−2039.8	−2019.9
Obs. Indiv.	6467	3263	3204
Obs. Total	12934	6526	6408

Note: Random-effects probit estimation; Average partial effects; Standard errors in parentheses; Previous period indicated by $(t-5)$; Initial condition indicated by $(t=1)$; ρ: estimate of the cross-period correlation of the composite error term $a_i+{ϵ}_{it}$; Occupation FE: 54 occupational dummies; Year dummies included; Significant at 1% ***, significant at 5% **, significant at 10% *.

Table 4. Results, Wooldridge dynamic RE probit, occupation fixed-effects.

	Dependent Variable: Overqualification
	All	Female	Male
	(1)	(2)	(3)
Overqualification $(t-5)$	0.036*** (0.014)	0.037** (0.18)	0.036 (0.023)
Overqualification $(t=1)$	0.164*** (0.018)	0.131*** (0.022)	0.217*** (0.034)
Female	0.019*** (0.006)
University grade	−0.018*** (0.003)	−0.022*** (0.004)	−0.013*** (0.004)
School grade	−0.004 (0.003)	−0.008* (0.005)	−0.002 (0.004)
Study duration	0.003 (0.003)	0.003 (0.004)	0.004 (0.004)
U. applied sciences	0.014** (0.007)	0.015 (0.010)	0.017* (0.010)
BusAdmin/Econ	ref.	ref.	ref.
Med/Law/Teach	−0.003 (0.014)	−0.011 (0.018)	0.020 (0.025)
STEM subjects	−0.014 (0.009)	0.001 (0.015)	−0.025* (0.013)
Soc/Cult science	0.027** (0.012)	0.016 (0.015)	0.051** (0.024)
Age at grad. < 27	−0.009 (0.006)	−0.005 (0.009)	−0.016* (0.009)
Parents high-skilled	−0.009 (0.006)	−0.010 (0.008)	−0.006 (0.008)
Married	−.007 (0.010)	−0.011 (0.014)	−0.002 (0.015)
Children	0.013 (0.010)	0.021 (0.014)	0.004 (0.015)
Months unemployed	0.001 (0.002)	0.001 (0.002)	0.001 (0.003)
Mean: Married	−0.001 (0.012)	0.004 (0.017)	−0.011 (0.019)
Mean: Children	0.013 (0.012)	0.017 (0.017)	0.015 (0.019)
Mean: Months unemployed	0.003* (0.002)	0.003 (0.002)	0.004 (0.003)
Occupation FE:	Yes	Yes	Yes
σ	1.052 (0.092)	1.134 (0.130)	0.967 (0.134)
ρ	0.525 (0.043)	0.562 (0.056)	0.483 (0.070)
Log. Lik.	−3738.7	−1814.3	−1881.3
Obs. Indiv.	6461	3257	3037
Obs. Total	12912	6509	6059

Note: Random-effects probit estimation; Average partial effects; Standard errors in parentheses; Previous period indicated by $(t-5)$; Initial condition indicated by $(t=1)$; ρ: estimate of the cross-period correlation of the composite error term $a_i+{ϵ}_{it}$; Occupation FE: 54 occupational dummies; Year dummies included; Significant at 1% ***, significant at 5% **, significant at 10% *.

Table 5. Results, Wooldridge dynamic RE probit, robustness.

	Dependent Variable: Overqualification
	(1)	(2)	(3)
Overqualification $(t-5)$	0.029** (0.013)	0.029** (0.013)	0.032** (0.014)
Overqualification $(t=1)$	0.224*** (0.023)	0.227*** (0.023)	0.310*** (0.029)
Female	0.015** (0.006)	0.019*** (0.006)	0.009* (0.005)
University grade	−0.018*** (0.003)	−0.019*** (0.003)
School grade	−0.006** (0.003)	−0.008** (0.003)
Age at grad. < 27	−0.008 (0.006)	−0.009 (0.006)	−0.024*** (0.005)
Parents high-skilled	−0.008 (0.006)	0.008 (0.006)	−0.018*** (0.005)
Married	−0.005 (0.009)	−0.005 (0.009)	−0.005 (0.009)
Children	0.009 (0.009)	0.009 (0.009)	0.008 (0.008)
Months unemployed	0.001 (0.001)		0.001 (0.001)
Married $(t=5)$	0.001 (0.009)
Children $(t=5)$	0.004 (0.009)
Months unemp $(t=5)$	0.002 (0.002)
Married $(t=10)$	−0.003 (0.009)
Children $(t=10)$	0.006 (0.009)
Months unemp $(t=10)$	0.002 (0.001)
Mean: Married		−0.008 (0.012)	−0.005 (0.011)
Mean: Children		0.009 (0.012)	0.008 (0.011)
Mean: Months unemployed			0.004*** (0.001)
Study characteristics:	Yes	Yes	No
σ	1.222 (0.099)	1.240 (0.099)	1.274 (0.102)
ρ	0.599 (0.039)	0.606 (0.038)	0.619 (0.037)
Log. Lik.	−4073.9	−4100.9	−4238.5
Obs. Indiv.	6467	6467	6467
Obs. Total	12934	12934	12934

Note: Random-effects probit estimation; Average partial effects; Standard errors in parentheses; Previous period indicated by $(t-5)$; Initial condition indicated by $(t=1)$; ρ: estimate of the cross-period correlation of the composite error term $a_i+{ϵ}_{it}$; Study characteristics: university type and subjects; Year dummies included; Significant at 1% ***, significant at 5% **, significant at 10% *.

Table 6. Results, Heckman dynamic RE probit.

	Dependent Variable: Overqualification
	All	Female	Male
	(1)	(2)	(3)
Main Panel Estimation
Overqualification $(t-5)$	0.354*** (0.088)	0.400*** (0.122)	0.315** (0.129)
Female	0.240*** (0.063)
University grade	−0.237*** (0.033)	−0.257*** (0.048)	−0.210*** (0.044)
School grade	−0.094*** (0.032)	−0.148*** (0.049)	−0.060 (0.042)
Study duration	0.121*** (0.030)	0.153** (0.046)	0.101** (0.039)
U. applied sciences	0.239*** (0.066)	0.237*** (0.099)	0.240*** (0.089)
BusAdmin/Econ	ref.	ref.	ref.
Med/Law/Teaching	−1.776*** (0.144)	−2.036*** (0.22)	−1.461*** (0.195)
STEM subjects	−0.734*** (0.085)	−0.792*** (0.139)	−0.684*** (0.109)
Soc/Cult science	−0.153* (0.085)	−0.252** (0.122)	−0.046 (0.128)
Age at grad. < 27	−0.118* (0.064)	−0.064 (0.093)	−0.172** (0.087)
Parents high-skilled	−0.130** (0.058)	−0.123 (0.085)	−0.127 (0.079)
Married	−0.059 (0.058)	−0.056 (0.083)	−0.059 (0.081)
Children	0.153* (0.059)	0.273*** (0.086)	0.050 (0.082)
Months unemployed	0.039*** (0.005)	0.042*** (0.007)	0.035*** (0.008)
Constant	−1.713*** (0.110)	−1.591*** (0.163)	−1.605*** (0.142)
Initial Conditions Estimation
Unemployment rate	0.107*** (0.016)	0.132*** (0.022)	0.087*** (0.024)
Female	0.242*** (0.055)
University grade	−0.176*** (0.027)	−0.136*** (0.036)	−0.229*** (0.042)
School grade	−0.086*** (0.028)	−0.077** (0.038)	−0.096** (0.043)
Study duration	0.132*** (0.026)	0.131*** (0.035)	0.136*** (0.040)
U. applied sciences	0.147** (0.058)	0.001 (0.08)	0.274*** (0.089)
BusAdmin/Econ	ref.	ref.	ref.
Med/Law/Teaching	−1.344*** (0.098)	−1.372*** (0.122)	−1.415*** (0.181)
STEM subjects	−0.704*** (0.069)	−0.783*** (0.103)	−0.690*** (0.100)
Soc/Cult science	−0.117 (0.079)	−0.210** (0.102)	−0.066 (0.135)
Age at grad. < 27	−0.068 (0.057)	−0.015 (0.075)	−0.123** (0.090)
Parents high-skilled	−0.157*** (0.051)	−0.114* (0.068)	−0.205** (0.081)
Married	0.100 (0.070)	0.094 (0.095)	0.127 (0.107)
Children	0.057 (0.089)	0.228* (0.122)	−0.145 (0.136)
Months unemployed	0.040*** (0.011)	0.042*** (0.014)	0.038** (0.018)
Constant	−1.348*** (0.098)	−1.101*** (0.132)	−1.388*** (0.145)
θ	0.563 (0.050)	0.457 (0.056)	0.724 (0.102)
ρ	0.622 (0.037)	0.660 (0.049)	0.580 (0.058)
Log. Lik.	−6549.9	−3343.4	−3185.8
Obs. Indiv.	6467	3263	3204
Obs. Total	19401	9789	9612

Note: Random-effects probit estimation; Estimation coefficients are displayed; Standard errors in parentheses; Stata command: redprob; ρ: estimate of the cross-period correlation of the composite error term ${\mu }_i+{ϵ}_{it}$; θ: statistic used to test whether the initial conditions are exogenous. The estimate of θ is significantly greater than zero, i.e. the hypothesis that the initial conditions are exogenous is rejected; Unemployment rate: occupation-specific unemployment rate at the time of labour market entry; Year dummies included; Significant at 1% ***, significant at 5% **, significant at 10% *.

Table 7. Results, predicted probabilities of overqualification.

	All	Female	Male
Wooldridge Estimator
P$(y_t=1|y_{t-5}=0)$	7.9	8.4	7.6
P$(y_t=1|y_{t-5}=1)$	10.8	11.7	10.4
Difference	2.9	3.3	2.8
Heckman Estimator
P$(y_t=1|y_{t-5}=0)$	5.8	6.2	5.4
P$(y_t=1|y_{t-5}=1)$	9.8	10.8	9.0
Difference	4.0	4.6	3.6

Note: $y_t$ depicts the overqualification status in time period t; Probabilities are predicted under the assumption that the individual heterogeneity takes its average value.

Table 8. Results, Wooldridge dynamic RE probit, realised matches.

	Dependent Variable: Overqualification RM
	All	Female	Male
	(1)	(2)	(3)
Overqualification RM $(t-5)$	0.032** (0.013)	0.032* (0.019)	0.031* (0.018)
Overqualification RM $(t=1)$	0.085*** (0.013)	0.097*** (0.020)	0.070*** (0.017)
Female	0.020*** (0.004)
University grade	−0.005*** (0.002)	−0.005* (0.003)	−0.005** (0.002)
School grade	−0.004* (0.002)	−0.009*** (0.003)	−0.000 (0.002)
Study duration	0.004** (0.002)	0.008** (0.003)	0.001 (0.002)
U. applied sciences	0.009** (0.004)	0.009 (0.007)	0.010* (0.005)
BusAdmin/Econ	ref.	ref.	ref.
Med/Law/Teach	−0.055*** (0.006)	−0.075*** (0.010)	−0.037*** (0.007)
STEM subjects	−0.043*** (0.006)	−0.039*** (0.008)	−0.042*** (0.009)
Soc/Cult science	−0.037*** (0.005)	−0.048*** (0.008)	−0.025*** (0.005)
Age at grad.< 27	−0.006 (0.004)	−0.008 (0.007)	−0.005 (0.005)
Parents high-skilled	0.003 (0.004)	0.002 (0.006)	0.003 (0.005)
Married	0.002 (0.007)	0.013 (0.011)	−0.007 (0.008)
Children	−0.007 (0.007)	−0.011 (0.011)	−0.003 (0.008)
Months unemployed	−0.000 (0.001)	−0.001 (0.002)	0.001 (0.002)
Mean: Married	−0.002 (0.009)	−0.008 (0.014)	0.003 (0.010)
Mean: Children	0.005 (0.009)	0.012 (0.014)	0.001 (0.010)
Mean: Months unemployed	0.002* (0.001)	0.003* (0.002)	0.001 (0.002)
σ	1.085 (0.106)	1.016 (0.142)	1.155 (0.160)
ρ	0.541 (0.049)	0.508 (0.070)	0.572 (0.68)
Log. Lik.	−3038.0	−1601.6	−1424.3
Obs. Indiv.	6222	3120	3102
Obs. Total	12444	6240	6204

Note: Random-effects probit estimation; Average partial effects; Standard errors in parentheses; Previous period indicated by $(t-5)$; Initial condition indicated by $(t=1)$; ρ: estimate of the cross-period correlation of the composite error term $a_i+{ϵ}_{it}$; Year dummies included; 3-digit occupations with less than 20 observations for calculating realised matches are excluded from the estimation; Significant at 1% ***, significant at 5% **, significant at 10% *.

Table 9. Results, Wooldridge dynamic RE probit, inverse probability weighting.

	Dependent Variable:
	Participation	Overqualification
	Wave 3	Weighted		Unweighted
	(1)	(2)	(3)	(4)	(5)
Overqualification $(t-5)$		0.030** (0.013)	0.037*** (0.013)	0.030** (0.014)	0.037*** (0.014)
Overqualification $(t=1)$		0.240*** (0.022)	0.175*** (0.018)	0.237*** (0.024)	0.172*** (0.019)
Female	0.011 (0.007)	0.017*** (0.006)	0.020*** −2006	0.016*** (0.006)	0.020*** (0.006)
University grade	0.003 (0.004)
School grade	0.008** (0.004)
Study duration	−0.002 (0.004)	0.011*** (0.003)	0.007*** (0.003)	0.011*** (0.003)	0.007*** (0.003)
U. applied sciences	−0.027*** (0.009)	0.020*** (0.006)	0.015** (0.006)	0.020*** (0.086)	0.015** (0.007)
BusAdmin/Econ	ref.	ref.	ref.	ref.	ref.
Med/Law/Teach	0.024** (0.011)	−0.094*** (0.005)	−0.001 (0.013)	−0.092*** (0.006)	0.001 (0.014)
STEM subjects	0.027*** (0.010)	0.053*** (0.007)	−0.013 (0.009)	0.053*** (0.008)	(0.013) (0.009)
Soc/Cult science	0.011 (0.011)	−0.018** (0.007)	0.028** (0.011)	(0.018** (0.008)	0.028** (0.012)
Age at grad. < 27	0.004 (0.008)	−0.012** (0.006)	−0.012** (0.006)	−0.011* (0.006)	−0.011* (0.006)
Parents high-skilled	−0.008 (0.007)	−0.010** (0.005)	−0.011** (0.005)	−0.010* (0.006)	−0.011** (0.006)
Time-variant controls:	No	Yes	Yes	Yes	Yes
Occupation FE:	No	No	Yes	No	Yes
σ		1.231	1.059	1.230	1.054
ρ		0.602	0.529	0.602	0.526
Log. Lik.	−3548.1	−4712.3	−4327.8	−4.101.7	−3762.7
Obs. Indiv.		6467	6461	6467	6461
Obs. Total	9523	12934	12912	12934	12912

Note: Probit (1) and random-effects probit (2,3) estimation; Average partial effects; Standard errors in parentheses; Weighted: weighted by inverse probability weights for participation in wave 3 predicted from specification (1); Previous period indicated by $(t-5)$; Initial condition indicated by $(t=1)$; ρ: estimate of the cross-period correlation of the composite error term $a_i+{ϵ}_{it}$; Time-variant controls: married, children, months unemployed; Occupation FE: 54 occupational dummies; Year dummies included; Significant at 1% ***, significant at 5% **, significant at 10% *.

Table 10. Results, Wooldridge dynamic RE probit, residual wages.

	Dependent Variable:
	Log Hourly Wage	Overqualification
	(1)	(2)	(3)
Overqualification $(t-5)$		0.023* (0.014)	0.027* (0.014)
Overqualification $(t=1)$		0.240*** (0.025)	0.176*** (0.021)
Residual wage		−0.002 (0.003)	−0.005* (0.003)
Female	−0.073*** (0.010)	0.016** (0.006)	0.020*** (0.007)
University grade	0.038*** (0.005)	−0.015*** (0.003)	−0.015*** (0.003)
School grade	0.006 (0.005)	−0.006* (0.003)	−0.005 (0.003)
Study duration	−0.007 (0.005)	0.007** (0.003)	0.003 (0.003)
U. applied sciences	−0.009 (0.010)	0.020*** (0.007)	0.015** (0.007)
BusAdmin/Econ	ref.	ref.	ref.
Med/Law/Teach	−0.733*** (0.015)	−0.092*** (0.006)	−0.001 (0.015)
STEM subjects	−0.058*** (0.011)	−0.051*** (0.008)	−0.016* (0.010)
Soc/Cult science	−0.284*** (0.015)	−0.013 (0.008)	0.023* (0.013)
Age	0.140*** (0.033)
Age sqared	−0.002*** (0.001)
Age at grad. < 27		−0.007 (0.006)	−0.009 (0.007)
Parents high-skilled	0.011 (0.009)	−0.008 (0.006)	−0.009 (0.006)
Married	0.018 (0.014)	−0.003 (0.010)	−0.005 (0.010)
Children	−0.032* (0.018)	0.008 (0.010)	0.012 (0.010)
Months unemployed	−0.008*** (0.002)	0.001 (0.001)	0.001 (0.002)
Mean controls:	No	Yes	Yes
Occupation FE:	No	No	Yes
σ		1.243	1.098
ρ		0.607	0.547
Log. Lik.	−1841.4	−3597.8	−3312.6
Obs. Indiv.	5720	5720	5715
Obs. Total	5720	11440	11423

Note: OLS (1) and random-effects probit (2,3) estimation; Average partial effects; Standard errors in parentheses; Residual wage: normalised residual predicted from specification (1); Previous period indicated by $(t-5)$; Initial condition indicated by $(t=1)$; ρ: estimate of the cross-period correlation of the composite error term $a_i+{ϵ}_{it}$; Mean controls: mean married, mean children, mean months unemployed; Occupation FE: 54 occupational dummies; Year dummies included; Significant at 1% ***, significant at 5% **, significant at 10% *.

Table 11. Results, wage regressions, current overqualification.

	Dependent Variable: Log Hourly Wage
	Wave t = 1		Wave t = 5		Wave t = 10
	(1)	(2)	(3)	(4)	(5)	(6)
Overqualification (in t)	−0.025* (0.014)	−0.036*** (0.013)	−0.045*** (0.013)	−0.075*** (0.013)	−0.136*** (0.014)	−0.145*** (0.015)
Female	−0.076*** (0.011)	−0.072*** (0.011)	−0.069*** (0.010)	−0.052*** (0.010)	−0.128*** (0.010)	−0.110*** (0.010)
University grade	0.037*** (0.005)	0.042*** (0.005)	0.024*** (0.005)	0.025*** (0.005)	0.024*** (0.005)	0.032*** (0.005)
School grade	0.001 (0.005)	−0.004 (0.005)	−0.002 (0.005)	−0.003 (0.005)	0.018*** (0.005)	0.005 (0.005)
Study duration	0.008 (0.005)	0.006 (0.005)	−0.007 (0.005)	−0.011** (0.005)	−0.008* (0.005)	−0.007 (0.005)
U. applied sciences	−0.012 (0.011)	−0.031*** (0.012)	−0.022** (0.010)	−0.026** (0.011)	−0.046*** (0.010)	−0.045*** (0.010)
BusAdmin/Econ	ref.	ref.	ref.	ref.	ref.	ref.
Med/Law/Teach	−0.749*** (0.017)	−0.621*** (0.026)	−0.244*** (0.014)	−0.068*** (0.021)	−0.193*** (0.015)	−0.073*** (0.021)
STEM subjects	−0.053*** (0.012)	0.001 (0.017)	−0.079*** (0.012)	−0.048*** (0.015)	−0.082*** (0.012)	−0.032** (0.015)
Soc/Cult science	−0.261*** (0.017)	−0.130*** (0.022)	−0.251*** (0.015)	−0.113*** (0.021)	−0.249*** (0.015)	−0.128*** (0.021)
Constant	2.917*** (0.014)	2.668*** (0.097)	3.173*** (0.013)	3.018*** (0.219)	3.337*** (0.015)	2.947*** (0.033)
Controls:	Yes	Yes	Yes	Yes	Yes	Yes
Occupation FE:	No	Yes	No	Yes	No	Yes
Log. Lik.	0.461	0.513	0.205	0.262	0.217	0.295
Obs. Total	4675	4675	4675	4675	4675	4675

Note: OLS estimation; Standard errors in parentheses; Period indicated by $\left(t\right)$; Controls: age, parents high-skilled, married, children, months unemployed; Occupation FE: 54 occupational dummies; Year dummies included; Significant at 1% ***, significant at 5% **, significant at 10% *.

Table 12. Results, wage regressions, current and past overqualification.

	Dependent Variable: Log Hourly Wage
	Wave t = 5		Wave t = 10
	(1)	(2)	(3)	(4)
Overqualification
Pattern t${}_1$/t${}_5$
0/0	ref.	ref.
1/0	−0.012 (0.017)	−0.026 (0.016)
0/1	−0.034** (0.017)	−0.072*** (0.016)
1/1	−0.062*** (0.018)	−0.089*** (0.018)
Overqualification
Pattern t${}_1$/t${}_5$/t${}_{10}$:
0/0/0			ref.	ref.
1/0/0			−0.041** (0.017)	−0.041** (0.017)
0/1/0			−0.018 (0.021)	−0.049** (0.028)
1/1/0			−0.019 (0.026)	−0.026 (0.025)
0/0/1			−0.104*** (0.026)	−0.117*** (0.026)
1/0/1			−0.092** (0.043)	−0.116*** (0.038)
0/1/1			−0.133*** (0.026)	−0.147*** (0.027)
1/1/1			−0.195*** (0.022)	−0.208*** (0.024)
Constant	3.176*** (0.014)	3.018*** (0.221)	3.347*** (0.015)	2.972*** (0.029)
Controls:	Yes	Yes	Yes	Yes
Occupation FE:	No	Yes	No	Yes
Adjusted R2	0.205	0.262	0.219	0.298
Obs. Total	4675	4675	4675	4675

Note: OLS estimation; Standard errors in parentheses; Period indicated by $\left(t\right)$; Controls: female, grades, study duration, university type, subject, age, parents high-skilled, married, children, months unemployed; Occupation FE: 54 occupational dummies; Year dummies included; Significant at 1% ***, significant at 5% **, significant at 10% *.