2. (4 points) Jason plans to write a term paper regarding the male-female wage gap for one of his courses. As a starting point, he performs a simple Oaxaca decomposition using a public dataset that he acquired online. First, he starts with the simplest specification as follows: Specification 1: In hourly wage) =a + b (years of education) Using a statistical software, he gets the following Oaxaca decomposition results: oaxaca Inwage edyears [pw=resp_weight], by(female) weight() Blinder-Oaxaca decomposition Number of obs = 50,556 1: female - 2: female - 1 Robust Inwage Coefficient std. err. z P>z! [95% conf. interval] Differential Prediction_1 Prediction_2 Difference 3.205528 2.887424 .3181038 .0062104 .0055114 .0083032 516.16 523.90 38.31 0.000 0.000 0.000 3.193356 2.876622 .3018297 3.2177 2.898226 .3343778 Decomposition Explained Unexplained .0126001 .3055037 .0034021 0077183 3.70 39.58 0.000 0.000 .0059322 .2903761 .019268 .3206313 Next, he tries a different specification with additional controls: Specification 2: In(hourly wage) =a +b (years of education)+c (years of labor market experience) + d(years of job tenure) With the second specification, he gets the following decomposition results: 3 oaxaca Inwage edyears exp tenure (pw=resp_weight], by(female) weight(e) Blinder-Oaxaca decomposition Number of obs = 50,269 - 1: female - 2: female = 1 Robust Inwage coefficient std. err. . . Z P> 121 [95% conf. interval] Differential Prediction_1 Prediction_2 Difference 3. 207812 2.889002 .3188099 .0062003 0055231 .0083035 517.36 523.08 38. 39 0.000 0.000 0.000 3.195659 2.878177 . 3025353 3.219964 2.899827 .3350844 Decomposition Explained Unexplained .0532388 .2655711 .0040838 .0074259 13.04 35.76 0.000 0.000 .0452346 .2510167 .0612429 .2801256 Briefly describe three economic observations that can be made from these results