Skip to main content

Table 3 Estimates according to workers’ age (threshold = 40 yearsa), three educational categories

From: Does education raise productivity and wages equally? The moderating role of age and gender

 

GMM-SYS

LP

Value added per hour worked (ln)

Wage cost per hour worked (ln)

Value added-wage cost gapc

Value added per hour worked (ln)

(1)

(2)

(3)

(4)

Lagged dependent variable (ln)

0.659***

(0.058)

0.465***

(0.138)

0.614***

(0.047)

0.792***

(0.043)

Shares of workersa

 Young low-educated (YE12)

Reference

Reference

Reference

Reference

 Older low-educated (OE12)

−0.028

(0.055)

−0.010

(0.039)

−0.034

(0.046)

−0.026

(0.032)

 Young middle-educated (YE34)

0.046

(0.044)

0.004

(0.022)

0.040

(0.043)

−0.017

(0.022)

 Older middle-educated (OE34)

0.039

(0.047)

0.059*

(0.034)

−0.033

(0.043)

0.035

(0.022)

 Young high-educated (YE567)

0.158***

(0.058)

0.096**

(0.048)

0.093**

(0.045)

0.120***

(0.033)

 Older high-educated (OE567)

0.080

(0.069)

0.235***

(0.080)

−0.029

(0.056)

0.106***

(0.036)

Hansen over-identification test, p value

0.451

0.238

0.799

 

Arellano-Bond test for AR(2), p value

0.132

0.306

0.224

 

Number of observations

6714

6714

6714

6691

Number of firms

1844

1844

1844

1844

Chi-squared statistic for equality of regression coefficients, H0

 OE12 = YE34

1.94

0.17

2.58

0.13

 OE12 = OE34

2.14

5.14**

0.00

5.27**

 OE12 = YE567

8.62***

4.21**

6.38**

20.39***

 OE12 = OE567

2.36

7.73***

0.01

11.25***

 YE34 = OE34

0.01

2.22

1.43

2.96*

 YE34 = YE567

4.68**

3.48*

1.94

17.51***

 YE34 = OE567

0.22

7.58***

1.26

8.87***

 OE34 = YE567

3.01*

0.64

4.34**

9.34***

 OE34 = OE567

0.36

5.33**

0.01

4.62**

 YE567 = OE567

0.91

3.88**

3.11*

0.19

Interpretationb

 a) Among young workers

YE12 < YE567

YE34 < YE567

but

YE12 = YE34

YE12 < YE567

YE34 < YE567

but

YE12 = YE34

YE12 < YE567

but

YE12 = YE34

YE34 = YE567

YE12 < YE567

YE34 < YE567

but

YE12 = YE34

YE567 significantly more productive than YE34 and YE12

YE567 significantly more costly than YE34 and YE12

YE567 significantly more profitable than YE12

YE567 significantly more productive than YE34 and YE12

a) b) Among older workers

OE12 = OE34

OE12 = OE567

OE34 = OE567

OE12 < OE34

OE34 < OE567

OE12 < OE567

OE12 = OE34

OE34 = OE567

OE12 = OE567

OE12 < OE34

OE34 < OE567

OE12 < OE567

Education has no significant impact on productivity (relationship only significant at the 12% probability level)

Education increases wage costs significantly

Education has no significant impact on profits

Education increases productivity significantly

  1. Notes: Standard errors, that are robust to heteroskedasticity and autocorrelation, are reported between parentheses. Regressions also control for: % of workers with 10 years of tenure or more; % workers younger than 30 and older than 49 years, respectively; % women; % part-time workers; % blue-collar workers; % workers with fixed term employment contract; % apprentices; % temporary agency workers; ln of firm size; ln of capital stock per worker; level of collective wage bargaining; region where the firm is located (2 dummies); industries (8 dummies), and years dummies (11). AR (2) refers to second-order autocorrelation in first-differenced errors. Models (1) and (3) include first and second lags of explanatory variables (except time dummies) as instruments. Model (2) uses first and third lags of explanatory variables (except time dummies) as instruments
  2. ***p < 0.01, **p < 0.05, *p < 0.1
  3. a Young (older) workers are defined as being less than (at least) 40 years old. Low-educated workers (E12) have at most a degree of lower secondary school. Middle-educated workers (E34) have at most a degree from upper (general, technical or professional) secondary school. High-educated workers (E567) have a tertiary educational attainment (i.e. at least a Bachelor’s or equivalent degree)
  4. ‘<’ (‘=’) indicates if regression coefficients are (not) statistically different at the 10% level
  5. Value added-wage cost gap = ln(value added per hour) − ln(wage cost per hour)