GENDER AND RACE IN THE LABOR MARKET

GENDER AND RACE IN THE LABOR MARKET

Sponsored by a Grant TÁMOP-4.1.2-08/2/A/KMR-2009-0041 Course Material Developed by Department of Economics,

Faculty of Social Sciences, Eötvös Loránd University Budapest (ELTE) Department of Economics, Eötvös Loránd University Budapest

Institute of Economics, Hungarian Academy of Sciences Balassi Kiadó, Budapest

2

Author: Anna Lovász Supervised by Anna Lovász

June 2011

Week 10

Roma in the labor market I: employment and wages

Literature for next week

• Roma II:

• Kertesi–Kézdi 2010 (HLM)

• Kertesi–Kézdi 2010 (BWP)

• Further recommended reading:

• Kertesi–Kézdi 2006 (BWP)

Empirical project

• Due next week: first draft of essay

• Printed at the beginning of class

• Meeting with each group

• General comments

3

Ratio of female workers

• Calculation:

• We don’t have the actual number of female workers, so we estimate from the sample of workers: number of women in the sample / number of workers in the sample within each firm.

• Need to use sampling weights!

• Worker vs. firm-level database

• Use worker-level data to calculate ratios

• Then switch to firm-level for the estimation

Dependent variable

• Sales revenue

• Value added (VA) = sales revenue – material costs

• Profit?

Controls

• Region, industry, ownership

• Age and education: ratios of other groups as controls

• Occupational composition of the workforce (also ratios)

• Endogeneity?

4

Results/interpretation

• Statistics: include statistics of the regression sample and variables (dependent and explanatory variables, calculated worker shares)

• Calculation of relative values:

• Production function (nonlinear):

• Linear approximation:

• Relation:

• Hypothesis tests

• Comparison to previous international results

Roma employment: Kertesi–Kézdi 2010 (BWP)

• There is rarely good data available on the labor market situation of the Roma

• Administrative data do not contain measures of ethnicity.

• Categorization of self is questionable (magyar or Roma, …)

• Goals:

• Evaluate the Roma/non-Roma employment differential between 1993–2007, based on the best data available so far.

( )

jt F F

jt jt

jt

jt Z u

L L L

M K

Y ^jt + ⋅ +











 + −

+ +

+ +

=α αln βln γln γln 1 ϕ 1 δ

ln ⁰

( ) ( )

L L L

L _F

F F

F 1 1

1

ln ≈ −



 



 + ϕ − ϕ

jt jt jt

F F jt jt

jt Z u

L L L

K

Y =α +α⋅ln +γ⋅ln +φ ^jt +δ⋅ +

ln ⁰

( ϕ

) ₌ φ

^/ γ ₊ ¹

5

• Decompose the difference: what part is due to observable demographic, education, regional differences.

• Also calculate the wage differential and decompose it.

• Emphasize the importance of schooling as the main determinant

K&K 2010: employment differentials

Employment by cohorts between 1945–64

6

7

8

9

10

Summary – K&K 2010

• Roma employment has decreased significantly since the transition.

• By 1994, there was a roughly 40% difference between Roma and non-Roma employment.

• Since then, it continued to grow slowly.

• Similar for men and women

• Macroeconomic changes did not affect the differentials.

• Wage differential: roughly 1/3

• Selection bias

• Decomposition:

• Education explains roughly 1/3 of the difference

• The role of education is increasing, even without consideration of the quality of schools.

• For women, the number of children is a significant factor.