Wage Determinants: Inequality Mitigation

NB: We do not resell your papers. Upon ordering, we do an original paper exclusively for you.

NB: All your data is kept safe from the public.

Table of Contents

Introduction
Econometric Model
Data Description
Results Analysis
Conclusion

Introduction

Exploring the heterogeneity in wages in different countries and communities can be one of the most fundamental ways to spot the signs of economic inequality. This type of inequality in financial resources can also have numerous negative spill-overs on the social integration of citizens and communities, their access to education and healthcare, and overall levels of well-being. The current report is aimed at identifying the key patterns of wage inequality within the observed dataset and giving some recommendations as to how to mitigate them or eliminate them completely. Some of the factors hypothesized as wage determinants in the current study are work experience, education, tenure, race, marital status, and the level of urbanization.

Econometric Model

The three formulated econometric models were aimed to answer the six key questions of exploring the wage inequality in the considered dataset. The basic model containing only the variables of tenure, experience, and education answers the first question of the case study, as well as defines the joint significance of these three variables. The equation for the first model is as below.

lwage = β₀ + β₁* tenure + β₂* exper + β₃* educ + ε, where

β₀ reflects the constant of the considered multiple regression equation;
wage reflects the logarithmic transformation of the variable wage (defined as monthly earnings of the sample respondents);
tenure is the number of years that the individual has been working with the current employer;
exper reflects the number of years of work experience of the sample participants;
Educ measures the number of years of education of the individuals included in the sample;
ε reflects the stochastic error term of the regression.

The second model to be estimated includes an extended set of factors influencing the levels of wages. The specific equation to be used for this model is as below.

wage = β₀ + β₁ * tenure + β₂* exper + β₃* Educ + β₄* black + + β₅* urban + β₆* marital + ε, where

β₀, wage, tenure, exper, Educ, ε mean the same as in the first model;
black is the dummy variable of race, where 1 represents a black individual and 0 – non-black;
urban is the dummy variable of locality, where 1 represents a person living in an SMSA (standard metropolitan statistical area), and 0 – elsewhere;
married is the dummy variable of marital status, where 1 represents an individual that is married, and 0 – the one who is not married.

The third model also included the interaction terms between the variables of education and race and education and locality (black_educ is the interaction term of education and race, and urban_educ is the interaction term of education and locality). The outputs for all of the estimated multiple regression models will be discussed in the next section of the report.

Data Description

Within the further analysis, only the logarithmic transformation of the wage variable will be used. However, to have a better understanding of the original wage distribution, both values of wage and wage were summarized.

Table 1: Descriptive Statistics	No. of Observations	Min	Max	Average	St. Dev.
wage	935	4.75	8.03	6.78	0.42
wage	935	115.0	3078	957.95	404.36
tenure	935	0.0	22	7.23	5.08
exper	935	1.0	23	11.56	4.37
Educ	935	9.0	18	13.47	2.20
black	935	0	1	0.13	0.34
urban	935	0	1	0.72	0.45
married	935	0	1	0.89	0.31

It can be seen that the average value of wage within the sample is 957.95, with a range between 115 and 3078 and a standard deviation of 404.36. The average tenure and experience are 22 and 23 years, respectively, with the values of tenure having slightly higher variation. The variable of education varied between 9 and 18 years, with the mean at 13.47 years. The summary statistics on the three included dummy variables reflect that they are coded by the values of either 0 or 1. The information on the mean values shows that the dataset includes close to 89% of married people, 13% of people of the black race, and 72% of people living in urban localities.

Results Analysis

The first of the three estimated models included only the variables of tenure, experience, and education (Table 2). Each of the variables was found to be significant at 1% (p-values of 0.000) and positively associated with the level of wage. More specifically, as the tenure of an individual goes up by one year, the associated wage is expected to go up by 1.3%. For the cases of experience and education, one additional year is expected to increase the wages by 1.5% and 7.5%, respectively.

Table 2: Model One (Restricted) Dependent Variable – Log of Wage	Coefficient	Standard Error	t-stat	p-value
constant	5.497	0.111	49.73	0.000***
tenure	0.013	0.003	5.17	0.000***
exper	0.015	0.003	4.55	0.000***
Educ	0.075	0.007	11.50	0.000***
P-values: estimate significant at 1%.*

The R-squared of the first modeled regression is equal to 0.1551, which means that close to 15.51% of the variation in the dependent variable wage is explained by the dynamics in the chosen set of independent variables. As for the Wald test of joint significance, it showed that the variables of tenure, experience, and education are indeed jointly significant, with the test statistics of 56.97 and the associated p-value of 0.0000. In this test, rejecting the null hypothesis implies rejecting the possibility of zero beta coefficients next to tenure, experience, and education.

Within the second extended model shown in Table 3, the size, direction, and significance of the effects of tenure, experience, and education remain largely similar as in the previous estimation. The value of the R-squared went up to 0.2429, which means that this model explains close to 24.29% of the variation in the dependent variable.

Table 3: Model Two (Extended) Dependent Variable – Log of Wage	Coefficient	Standard Error	t-stat	p-value
constant	5.346	0.113	47.32	0.000***
tenure	0.012	0.003	4.94	0.000***
exper	0.014	0.003	4.42	0.000***
educ	0.066	0.006	10.58	0.000***
black	-0.219	0.037	-5.93	0.000***
urban	0.196	0.027	7.28	0.000***
married	0.195	0.039	4.97	0.000***
P-values: estimate significant at 1%.*

As for the values of the dummy variables, each of them was found to be significant at 1%. For the variable of race, a black individual is expected to have close to 19.7% lower wage, all else being equal (calculating the exp. of the regression coefficient and subtracting 1). For the variable urban, SMSA inhabitants are expected to have 21.7% higher wages, and being married is associated with an approximately 21.5% increase in wages.

Table 4 presents the third model with the added interaction terms. The R-squared of the regression went up from 0.2429 to 0.2454 as compared to the previous version (model two), which means that the interaction terms added very little, if any, explanatory power.

Table 4: Model Three (Interactions) Dependent Variable – Log of Wage	Coefficient	Standard Error	t-stat	p-value
constant	5.466	0.165	33.17	0.000***
tenure	0.012	0.003	4.93	0.000***
exper	0.014	0.003	4.32	0.000***
educ	0.058	0.011	5.27	0.000***
black	0.134	0.258	0.52	0.604
black_educ	-0.028	0.020	-1.38	0.169
urban	-0.003	0.167	-0.02	0.985
urban_educ	0.015	0.012	1.21	0.228
married	0.193	0.039	4.92	0.000***
P-values: estimate significant at 1%.*

The previous conclusion is also confirmed by the non-significant p-values of the interaction terms. Therefore, it seems that returns to education do not differ depending on the race and living or not living in an urban locality.

Conclusion

Basing on the performed calculations, it is now possible to give some recommendations. First of all, living in an urban locality seems to be associated with higher wages. Therefore, to stimulate the growth of wages in the future, the government might want to invest in increasing the levels of local urbanization. In addition, black individuals seem to be disadvantaged when it comes to the levels of wages. Judging by the output of the regression, this cohort of the population can try to mitigate the negative effect by investing in education, getting more working experience, or changing their family status to married.

However, the government should also interfere as the analysis shows that black people get lower wages in the considered area controlling for the levels of education, experience, and tenure, which is not fair. The good news is that the returns to education do not differ across races, and therefore black people can get the same benefits from their educational investments as white citizens. Overall, all groups of individuals could benefit in terms of wages from getting more education, tenure, and experience, living in an urban locality, and being married.

Click Here To Order Now!

Wage Determinants: Inequality Mitigation

Do you need this or any other assignment done for you from scratch?

We have qualified writers to help you.

We assure you a quality paper that is 100% free from plagiarism and AI.

You can choose either format of your choice ( Apa, Mla, Havard, Chicago, or any other)