Social Change and Income Inequality Results of the German Mikrozensus 1962-2004

Peter Kriwy Institute of Social Sciences CAU Kiel

with the collaboration of Christiane Gross

Rational Choice Sociology: Theory and Empirical Applications Seminar at Venice International University, 3 Dez. 2007

Social Change and Income Inequality

Content

1. 2. 3. 4. 5. 6. 7. 8.

Introduction Theoretical Implications Hypothesis Data Methodological Challenges Descriptive Results Multivariate Results Summary

2

Social Change and Income Inequality

1. Introduction ¾ Educational expansion in Germany since the 1950s ¾ Education and work are rather closely connected to each other in Germany ¾ Winner of the educational expansion: -

girls/women non-native persons persons with lower socio-economic background

¾ Education as one of the main determinants of the risk of being unemployed and of the income level ¾ Do the returns of higher education change as well? ¾ Do these groups benefit to the same extent from their educational background regarding employment and income? ¾ How does income inequality in these groups and different sectors change over time? 3

Social Change and Income Inequality

2. Theoretical Implications

Theories to explain income inequality: ¾ human capital approach (Schultz 1960; Becker 1962, 1964; Mincer 1974) ¾ signalling and screening (Spence 1973) ¾ numerous social capital explanations ¾ social reproduction (Bourdieu 1983) Theories of social discrimination: ¾ taste for discrimination (Becker 1957) ¾ statistical discrimination (Phelps 1972) Mechanisms of social discrimination: ¾ glass ceiling ¾ hurdles ¾ threshold 4

Social Change and Income Inequality

3. Hypothesis (1/3)

¾ human capital approach education

productivity

higher income

¾ signaling and screening education

signal

higher income

Æ impact of education on income level 5

Social Change and Income Inequality

3. Hypothesis (2/3)

¾ social reproduction SES

education

higher income (SES)

¾ taste for discrimination sex, ethnic group, etc.

discrimination

lower income

¾ statistical discrimination minority status

estimation of productivity

lower income

Æ impact of personal characteristics on income level

6

Social Change and Income Inequality

3. Hypothesis (3/3)

Hypothesis regarding the correlation of educational expansion (EE) and employment position/income: ¾ "proletarianization" thesis (Schlaffke 1972) EE

limited # of high positions

unemployed academics

¾ absorption thesis (Teichler et al. 1976) EE

change of hierarchies in firms

status quo (unemployment)

¾ crowding out thesis (Fürstenberg 1978, Lutz 1979) EE

increasing # of academics

unemployed workers

7

Social Change and Income Inequality

4. Data ¾ Joint project: Social Change in Germany Christof Wolf (ZUMA) ¾ ¾ ¾ ¾ ¾ ¾ ¾

„Mikrozensus“ covers about 1% of the German population. It is conducted every year since 1957. Hourly wage available since 1973. Since 1991 data of “East Germany” are available as well. About 830.000 individuals are collected per year. Around 500.000 cases are available as scientific use files. The cumulated data file (1962-2004) contains 12 Mio. cases in trend design. 8

Social Change and Income Inequality

5. Methodological challenges ¾ Inflation rate Æ individual hourly wage is inflation discounted to levels of 2004 and adjusted to buying power of each year (Lengerer et al. 2007)

¾ Employment period especially starting one´s working life differs, depending on educational level (total returns versus age related returns) Æ age related returns because of trend design (multivariate models contain 30-60 years old persons)

¾ Only net income available (problematic due to the German "Ehegattensplitting") Æ interaction effect: marry*higher income in hh

9

Social Change and Income Inequality

5. Measurement ¾ Education in years: similar to equal-distance model from lowest: ‘Hauptschule’ without apprenticeship to highest: ‘Abitur’ including university degree ¾ Work experience: Age - (education + 7 years) [unaccounted for unemployment] ¾ Data is assigned as a household sample: Personal weight is used to create an “individual” sample

10

Social Change and Income Inequality

6. Descriptive Results Development of education (West Germany) according to gender 13,5 13 12,5 12

male female

11,5 11 10,5 10 76 82 89 91 93 95 96 97 98 99 00 01 02 03 04

11

Social Change and Income Inequality

6. Descriptive Results Development of education (West Germany) according to nationality 13,5 13 12,5 12

germans non-germans

11,5 11 10,5 10 76 82 89 91 93 95 96 97 98 99 00 01 02 03 04

12

Social Change and Income Inequality

6. Descriptive Results Development of individual net income 1600 1400 1200 1000 800 600 400 62

64

66

68

73

82

91

95

97

99

01

03

13

Social Change and Income Inequality

6. Descriptive Results Development of individual net hourly wage 12 11 10 9

west east

8 7 6 5 73 76 82 89 91 93 95 96 97 98 99 00 01 02 03 04

14

Social Change and Income Inequality

6. Descriptive Results Development of individual working hours (full time jobs) 45 44,5 44 43,5 43 42,5 42 41,5 41 40,5 40

west east

73 89 91 93 95 96 97 98 99 00 01 02 03 04 15

Social Change and Income Inequality

6. Descriptive Results Development of tertiary sector (service) vs. industry, crafts & agriculture 70 65 60 55

west east

50 45 40 35 62 63 64 65 66 67 68 69 73 76 82 89 91 93 95 96 97 98 99 00 01 02 03 04

16

Social Change and Income Inequality

6. Descriptive Results Development of unemployment 25 20 15

west east

10 5 0 62 63 64 65 66 67 68 69 73 76 82 89 91 93 95 96 97 98 99 00 01 02 03 04

17

Social Change and Income Inequality

6. Descriptive Results Development of unemployed academics and “workers”

16 14 12 10

unemployment academics workers

8 6 4 2 0 76 82 89 91 93 95 96 97 98 99 00 01 02 03 04

18

Social Change and Income Inequality

7. Multivariate Results Returns separated by gender (display of regression coefficient, dep. var: log. hourly wage, from 1991 including west dummy, Germans only) 0,1 0,095 0,09 0,085 0,08 0,075 0,07 0,065 0,06 0,055 0,05

male female

Covariates: Exp, exp2, marry, marryh, hhanz, tert 76 82 89 91 93 95 96 97 98 99 00 01 02 03 04

19

Social Change and Income Inequality

7. Multivariate Results Returns separated according to gender (display of regression coefficient, dep. var: log. hourly wage, from 1991 including west dummy, Non-Germans only) 0,075 0,07 0,065 0,06 0,055 0,05 0,045 0,04 0,035 0,03 0,025

Reference value of German

male female

Males: 0,08 Females: 0,07

76 82 89 91 93 95 96 97 98 99 00 01 02 03 04

20

Social Change and Income Inequality

-----------------------------------------------------------------Variable |

md04

wd04

mnd04

wnd04

-------------+---------------------------------------------------exp |

.01620*

.02197*

.00349

.00778

exp2 | -.00019*

-.00028*

.00005

-.00004

summej |

.07684*

.06488*

.05910*

.04474*

marry | -.22121*

-.28876*

-.29484*

-.31979*

marryh |

.43384*

.37029*

.42478*

.28057*

hhanz |

.03866*

.03944*

.05822*

.03101

tert | -.03298*

.04711*

-.14393*

.02244

west |

.34070*

.25300*

.30509*

.40139

.32

.16

.20

.11

R2

|

-----------------------------------------------------------------21 * p < 0.001

Social Change and Income Inequality

8. Summary ¾ Men still achieve higher levels of education than women. ¾ Non-Germans´ levels of education rises slightly after 1997 – the gap between Germans and Non-Germans expands. ¾ The tertiary sector expands extremely after 1973. ¾ The percentage of unemployed academics rises. ¾ The returns of education for men are higher than the returns for women. This gap seems to enlarge in the recent past. ¾ The returns of education for Non-Germans are lower than the returns for Germans.

22

Social Change and Income Inequality

Thank you for your attention!

Contact: Peter Kriwy Institute of Social Sciences, CAU Kiel [email protected] Tel: +49 (0) 431 / 880 - 4377 23