x̄ - > The study you referenced on Kenya's wage determination used advanced econometric techniques.

Econometric Analysis: Kenya's Wage Determination Study

The study you referenced on Kenya's wage determination used advanced econometric techniques. A study utilizing the World Bank's Skills Towards Employability and Productivity Survey (WBSTEPS) employed the following methods:

1. Mincer Earnings Regression

The Mincer equation models wage determination based on human capital theory. It estimates how education and work experience affect earnings. The basic form is:

\[ \ln(Wage) = \beta_0 + \beta_1(Education) + \beta_2(Experience) + \beta_3(Experience^2) + \epsilon \]

• Education: Number of years of schooling
• Experience: Work experience (squared term captures diminishing returns)

2. Heckman Selection Correction

The Heckman correction accounts for sample selection bias, especially when the sample is restricted to employed individuals, leading to non-random selection.

Steps:

1. Step 1 (Selection equation): Models the probability of being employed.
2. Step 2 (Outcome equation): Corrects wage estimation by considering the selection bias.

\[ Wage = \beta X + \lambda \theta + \epsilon \]

where \(\lambda\) is the inverse Mills ratio from the selection model.

3. Blinder-Oaxaca Decomposition

This method decomposes wage differences between groups (e.g., men and women) into explained and unexplained components:

• Explained: Due to observable characteristics (e.g., education, experience)
• Unexplained: Due to discrimination or differences in returns to these characteristics

\[ \Delta W = (X_m - X_f)\beta + X_f(\beta_m - \beta_f) \]

• \(X_m, X_f\): Average characteristics of men and women
• \(\beta_m, \beta_f\): Coefficients for men and women

4. Neumark Decomposition

A variant of the Blinder-Oaxaca, focusing on wage structure rather than group differences, often applied when suspecting discrimination.

Application to Kenya's Labor Market (Key Findings Recap)

• Women earn 84.5% to 86% of men’s wages.
• Wage gap mostly driven by returns to endowments, not differences in qualifications.
• Evidence of discrimination in returns to education and experience.

This work is licensed under a Creative Commons Attribution 4.0 International License.