Answer to Question #204185 in Microeconomics for Eskinider Degefu

Question

Log (earnings) = 4.016 + 0.092 . educi  + 0.079 .expei +  0.002 . experi 2 

 ( 0.222)  ( 0.008) ( 0.025) (0.001)

log ( earnings )I  = β1 + β 2 ⋅ educI  + β 3 ⋅ experI  − β 4 ⋅ experI2 + β5 ⋅ abilityI  + εI . 

 In this regression, what do you expect the sign of β5 (the coefficient on ability) to be? 

 a). What do you think the sign of the correlation between ability and years of education? 

 b). If we estimate the regression function with ability included, do you think that the 

estimated value of β 2 will be greater or less than what it was in the regression without ability? Explain.

 

Answer 1

(a)

The coefficient of $\beta 5$ should be positive since higher-ability people will be able to succeed in more productive jobs to advance better and be paid more since this will be discovered by employers.

(b) The correlation between ability and years of education is likely to be positive. This is because higher-ability people will find schooling easier and pursue more of it.

(c) Given the positive influence of ability on waves as well as the positive correlation between education and ability, we are likely over-estimating the return to education if we do not control for ability. Therefore we expect the estimated value of $\beta2$ to go down if we control the ability.