Table 7 Exogenous shocks since the establishment of the People’s Republic of China.

From: Long-term impacts of historical education policy on wages in China: insights on over-education

 

(1)

(2)

(3)

(4)

(5)

(6)

 

1952

1959

1969

1979

1989

1999

Dependent Variable:

Wage (logged)

Panel A: OLS

      

Jinshi density in Ming-Qing(logged)

0.197*** (0.037)

0.134*** (0.027)

0.084*** (0.024)

0.056*** (0.016)

0.054** (0.020)

0.127*** (0.028)

R2

0.705

0.582

0.479

0.563

0.561

0.731

Panel B: 2SLS

      

Jinshi density in Ming-Qing(logged)

0.158*** (0.049)

0.092* (0.050)

0.044 (0.044)

0.085** (0.039)

0.070* (0.040)

0.173*** (0.051)

Observations

22

20

20

26

213

199

KP-F statistics

9.717

9.009

9.009

11.626

115.374

129.608

Cluster

Province

Province

Province

Province

Province

Province

Controls

Yes

Yes

Yes

Yes

Yes

Yes

Provincial fixed effects

No

No

No

No

Yes

Yes

  1. Each column in each panel represents a separate cross-sectional regression. Panel A reports OLS estimates while Panel B clusters 2SLS estimates. Columns (1) to (4) provide provincial regression regarding historical Jinshi density effects on employee pay across 1949–1979, with 10 years as an interval due to limited data availability at the prefecture level. Due to a lack of data in 1949–1951, we regard the wage in 1953 as the alternative year. Columns (5) and (6) report prefecture-level regression regarding historical Jinshi density effects on employee pay in 1989 and 1999. These estimated Jinshi density effects can be interpreted as the percentage changes in wages per worker with a percent change in Jinshi density. Columns (1) to(4) do not include fixed effects due to limited observations, while Columns (5) and (6) include province fixed effects. Covariates include rainfall and air pollution, nightlight in 2004, population density in Ming-Qing, the urbanization rate in Ming-Qing, the distance to the coast, agricultural sustainability, and terrain ruggedness. The KP F-statistic is the Kleibergen-Paap Wald rk F-statistic for weak identification in the first stage (Kleibergen and Paap, 2006). ***denotes significant at 1% level, **denotes significant at 5% level, *denotes significant at 10% level.