oaxaca blinder decomposition: explained is greater than difference

[sally yin]

Dear all,

I am using oaxaca blinder decomposition to analysis labor productivity using several binary variables (i.e. name ends with YN), and here is my code:

xi: oaxaca lp numEmp regiesterdYN firmAge managerExp managerFemaleYN insuffWaterYN paySecurityYN expCrimeYN haveLoanYN, by(womenOwn) weight(0)

However, I am getting explained coefficient greater than difference coefficient. I am not sure why this is? Could explained be greater than difference? Any assistance is much appreciated! Thank you!

Blinder-Oaxaca decomposition Number of obs = 2,631
Model = linear
Group 1: womenOwn = 0 N of obs 1 = 1,843
Group 2: womenOwn = 1 N of obs 2 = 788

explained: (X1 - X2) * b2
unexplained: X1 * (b1 - b2)

---------------------------------------------------------------------------------
lp | Coefficient Std. err. z P>|z| [95% conf. interval]
----------------+----------------------------------------------------------------
overall |
group_1 | 7.886623 .0379214 207.97 0.000 7.812299 7.960948
group_2 | 7.788183 .0551115 141.32 0.000 7.680166 7.8962
difference | .0984401 .0668978 1.47 0.141 -.0326772 .2295573
explained | .1113455 .041434 2.69 0.007 .0301363 .1925547
unexplained | -.0129054 .0713335 -0.18 0.856 -.1527166 .1269057
----------------+----------------------------------------------------------------

Best,
Sally

[FernandoRios]

There is nothing unusual here
simply that the market structure already compensates Some
of the wage gap., which is fully explain and exacerbated by differences in characteristics
F