Inequalities in the European Union - European Trade Union Institute

0 downloads 0 Views 206KB Size Report
Jan 21, 2011 - Disposable income inequality increased (or stable) in almost all countries. Exceptions : F ..... N.B.: Lighter colours are n.s. at 95% level. Source: ...

Inequalities in the European Union Maurizio Franzini Sapienza University of Rome & Criss 21 January 2011 ETUI, Brussels

This presentation 1.

2.

3.

4.

High economic inequalities: the role of labour incomes Labour incomes and human capital: inequality “between” and “within” Why is “within” inequality so high (and overlooked)? Policy implications: not only human capital…

1. Market and disposable Income Inequality in EU from Mid 1980s to Mid 2000s [1]

Generalized increase in market income inequality, France and Netherlands being the only two exceptions

1. Market and disposable Income Inequality from Mid 1980s to Mid 2000s [2]

Disposable income inequality increased (or stable) in almost all countries. Exceptions : F, EL and E. In many countries inequality was already on the rise in mid-80’s. Everywhere (NL and F are exceptions) market income inequality increased no less than disposable income inequality.

1. Gross and disposable income, 2007 Gini index of household equivalent income. Source: elaborationss on EU-SILC 2007 data 0.500 0.450

0.430

0.400

0.377

0.391

0.382

0.382 0.365

0.356

0.350

0.329

0.322

0.300

0.311

0.274

0.321

0.334

0.301 0.266

0.344

0.337 0.322

0.317

0.298

0.296 0.281

0.269

0.333

0.324

0.321

0.279

0.284

0.272 0.255

0.252

0.245

0.250 0.200 0.150 0.100 0.050

Gross household equivalent income

Disposable household equivalent income

High inequality also on the basis of 2007 data.

Sp ai n

Po rtu ga l

Ita ly

G re ec e

U K

Ir el an d

Sw ed en

N or w ay

Fi nl an d

D en m ar k

N et he rla nd s

Lu xe m bo ur g

Fr an ce

G er m an y

B el gi um

A us tri a

0.000

2. Earning inequality and disposable income inequality around 2000s

0 .5 5

0,55

0 .5

0,5

0 .4 5

0,45

2004

Gini index of Earnings

Gini index of Earnings

and in a pool G in i in d e x oGini f eindex a r n inof g earning a n d d is p odisposable s a b le in cincomes o m e s in p o oof l ocountries f c o u n t r ie s 2010 S o u r c e :Source: e la b o relaborations a t io n s o n on L ISLISd data a ta 2 010

0 .4

0,4

0 .3 5

0,35

0 .3

0,3

2000

R2 = 0,4797

R2 = 0,5349

0,25

0 .2 5

0,2

0 .2

0,2

0 .2

0,25

0 .2 5

0,3

0 .3

0,35

0 .3 5

0,4

0 .4

0,45

0,5

0 .4 5

0 .5

0,55

0 .5 5

Gini index of Disposable Incomes

G in i in d e x o f D is p o s a b le In c o m e s EAR-DPI_2000

E A R - D P I_ 2 0 0 0

EAR-DPI_2004

E A R - D P I_ 2 0 0 4

Despite the influence of many other factors on disposable income inequality, earnings inequality exerts a major influence on it.

2. Inequality in labour incomes, 2007 Gini index of annual gross labour incomes in EU15 countries (plus NO). Source: elaborations on EU-SILC 2007 data 0.500 0.475 0.450 0.425 0.400 0.375 0.350 0.325 0.300 0.275 0.250 Austria Belgium Germany France

Luxemb Netherla Denmark Finland Norway Sweden ourg nds

Ireland

UK

Greece

Italy

Portugal

Spain

Average value

Employment income

0.392

0.315

0.421

0.362

0.392

0.438

0.345

0.403

0.403

0.379

0.458

0.393

0.380

0.365

0.435

0.363

0.390

Labour income

0.404

0.326

0.440

0.375

0.397

0.448

0.371

0.404

0.399

0.378

0.470

0.409

0.450

0.387

0.439

0.368

0.404

Employment income

Labour income

The Gini index on gross labour income is higher than 35%, Belgium being the only exception. Inequality in employment income only, is, in general, lower but still quite high.

2. Wage inequality and education, 2007 Hourly gross wage premia by educational attainment (reference: upper secondary) estimated trough a Mincerian equation on people aged 25-64. Source: elaborations on Eu-SILC 2007 data 80 62.5 60

52.2 44.8

40 28.4

24.4

27.8

31.0

30.6

41.5 36.1

34.7

33.4

31.5 27.1

22.9

20.2

18.3

20

0 -10.5

-13.3

-20

-16.2 -22.9

-25.9

-16.8

-15.9

-10.7 -17.2 -24.9

-26.7

-27.4

-24.6

-26.0

-22.8

-32.5

-40

At most lower secondary

Average

Spain

Portugal

Italy

Greece

UK

Ireland

Sweden

Norway

Finland

Denmark

Netherlands

Luxembourg

France

Germany

Austria

Belgium

-53.2

-60

Tertiary

Wage premia are in general high (and seem not related to the rigidity of the labour market). This seems to point to a major contribution of inequality between groups of workers with different education attainments to overall labour income inequality

3. Inequality within graduated workers, 2007 Gini index of earnings of tertiary graduated workers. Source: elaborations on Eu-SILC 2007 data 0.450

0.400

0.350

0.300

0.250

0.200

0.150 Austria

Belgium Germany

France

Luxembo Netherlan Denmark urg ds

Finland

Norway

Sweden

Ireland

UK

Greece

Italy

Portugal

Spain

Average

Annual wages

0.366

0.303

0.344

0.349

0.337

0.399

0.290

0.341

0.357

0.365

0.390

0.369

0.361

0.375

0.392

0.339

0.355

Hourly wages

0.322

0.232

0.287

0.282

0.300

0.342

0.207

0.276

0.261

0.295

0.328

0.349

0.335

0.331

0.346

0.307

0.300

Annual wages

Hourly wages

However, inequality in hourly and annual wages within the group of tertiary graduated workers (but also among other educationally homogeneous groups) is generally high. Especially with respect to annual wages.

3. Decomposing of employment income inequality, 2007 Theil index decomposition of hourly gross wages by workers' educational attainment. Source: elaborations on EU-SILC 2007 data. 100% 11.0

10.0

5.3

9.3

14.7

6.5

1.9

6.6

8.3

8.8

14.5

12.3

14.2

18.1

25.8 36.2

80%

60%

89.0 40%

90.0

94.7

90.7

85.3

93.5

91.7

93.4

98.1 85.5

91.2

87.7

85.8

81.9

74.2 63.8

20%

Within

Sp ai n

Po rtu ga l

Ita ly

G re ec e

U K

Ir el an d

Sw ed en

N or w ay

Fi nl an d

D en m ar k

N et he rla nd s

Lu xe m bo ur g

Fr an ce

G er m an y

B el gi um

A us tri a

0%

Between

By decomposing employment inequality it comes out that the share of it explained by inequality within educationally homogeneous groups of workers is overwhelming. This holds both for annual and hourly incomes (shown here).

3. Decomposition considering also occupations, 2007 Theil index decomposition of hourly gross wages by workers' educational attainment and occupation. Source: elaborations on EU-SILC 2007 data. 100% 15.8

14.1

18.0

12.3

14.1

12.1

15.8

5.1

10.9

14.9 22.7

16.7

38.4

80%

22.0

24.8 39.6

60%

40%

84.2

85.9

82.0

87.7

85.9

87.9

84.2

89.1

94.9 85.1 77.3

83.3

78.0

75.2

61.6

60.4

20%

Within

Sp ai n

Po rtu ga l

Ita ly

G re ec e

U K

Ir el an d

Sw ed en

N or w ay

Fi nl an d

D en m ar k

N et he rla nd s

Lu xe m bo ur g

Fr an ce

G er m an y

B el gi um

A us tri a

0%

Between

The importance of within-group inequality does not change much if we consider jointly workers’ education and occupation (three groups here: managers, white collars and blue collars).

How did “within” inequality change? Componente within del Theil dei redditi annui netti da lavoro (autonomo e dipendente) 1996: lordi in FI e FR; 2007: lordi in DE, DK, FI, NL, UK

100.0%

95.0%

90.0%

85.0%

80.0%

75.0%

U K

D E

FI

N LA N

D

IA

IA ST R

AL G TO R PO

AU

LO

A AG SP

EC R G

N

IA

A AL I IT

FR

N D A IR LA

BE L

G

IO

L N

D K

70.0%

1996

2007

12

3. Educated ….but in the lower decile, 2007 Share of workers with a tertiary education degree who are in the poorest decile and quintile of the distribution of the gross annual income from employment 1° decile Countries

1° quintile

25-29

30-34

35-54

25-29

30-34

35-54

Austria

8.1

6.2

4.3

20.1

15.1

7.5

Belgium

9.8

4.8

2.8

17.5

10.5

7.1

Germany

10.1

2.4

2.6

20.4

5.4

6.4

France

7.9

3.8

2.8

15.7

9.4

6.8

Luxembourg

10.0

5.3

2.2

14.2

7.2

4.0

Netherlands

6.3

3.3

1.8

13.7

6.2

5.4

Denmark

5.9

7.8

2.7

21.0

14.8

4.7

Finland

5.3

4.8

3.1

14.3

10.5

6.4

Norway

5.7

4.7

2.4

24.1

8.9

6.0

Sweden

8.4

4.4

2.9

25.5

15.1

7.6

Ireland

8.5

3.9

1.3

12.4

7.2

4.5

UK

1.7

1.6

4.3

6.1

7.1

10.1

Greece

8.7

7.4

3.2

19.8

14.2

5.4

Italy

14.6

8.4

1.9

28.3

14.9

4.9

Portugal

16.2

2.0

1.5

22.3

3.8

1.6

Spain

10.4

3.8

3.3

20.2

10.8

7.1

Average

8.6

4.7

2.7

18.5

10.1

6.0

Source: elaborations on EU-SILC 2007 data

Not few tertiary educated workers are in the lowest deciles of the distribution. However, variability across countries is quite marked.

3. Educated and “poor”: gender and contract type, 2007 Share of workers with a tertiary education degree aged 35-54 who are in the poorest quintile of the distribution of the gross annual income from employment, by gender and contractual arrangement Male

Female

Full-time

Part-time

Permanent

Temporary

Austria

4.1

11.7

5.5

17.5

5.8

30.5

Belgium

3.6

10.6

3.3

18.6

4.9

38.5

Germany

2.2

13.1

2.8

20.5

5.0

18.6

France

4.2

8.9

4.0

19.5

4.1

32.2

Luxembourg

1.6

7.5

3.0

10.0

2.4

29.3

Netherlands

3.1

8.7

3.9

7.9

3.1

15.9

Denmark

3.7

5.4

4.4

6.4

n.a.

n.a.

Finland

3.9

8.4

5.8

19.1

1.4

10.1

Norway

5.0

6.9

5.6

10.6

1.0

12.3

Sweden

5.2

9.4

6.6

13.0

3.7

20.5

Ireland

1.4

7.5

3.0

10.9

5.3

25.3

UK

5.5

14.3

3.6

37.5

1.8

32.8

Greece

2.7

8.1

3.7

42.0

8.3

30.9

Italy

2.4

7.2

3.3

21.1

1.5

14.9

Portugal

0.9

2.1

1.1

45.0

2.4

15.3

Spain

3.7

10.9

4.0

33.6

0.7

24.5

Average

3.3

8.8

4.0

20.8

3.4

23.4

Source: elaboration on EU-SILC 2007 data

Females and holders of temporary and part time contracts are exposed to a very high risk of falling in the lower decile, despite their education.

4. Persistence of inequality Gini index of gross annual labour incomes (net in FR, IT, GR, PT) in 2005-2007. Individuals active in the whole period and aged 26-54 in 2005. Source: elaborations on EU-SILC longitudinal data 0.500

0.451

0.450

0.437

0.400

0.390

0.352

0.350

0.435 0.422

0.340 0.329

0.366

0.366

0.340 0.328

0.333 0.317 0.305

0.312

0.300

0.385

0.368

0.366 0.354

0.298

0.316 0.295

0.290

0.279

0.274

0.266

0.274 0.261

0.251

0.250

Mean of annual Gini 2005-2006-2007

Sp ai n

Po rtu ga l

Ita ly

G re ec e

U K

Ir el an d

Sw ed en

N or w ay

Fi nl an d

D en m ar k

N et he rla nd s

Lu xe m bo ur g

Fr an ce

B el gi um

A us tri a

0.200

Gini of average income 2005-2007

Inequality persists over longer period (limited role of income varability,also among higher educated workers)

A FIRST SUMMING UP High labour income inequality is a major cause of disposable income inequality.  High labour income inequality seems to depend to a large extent on inequality within educationally homogenous group.  This is not to deny that inequality between educated and not-educated workers played a role.  However, especially in Europe, the importance of within-inequality has been overlooked.  It is necessary to go into this inequality and understand which are its causes. Here are some clues 

Possible causes of “within” inequality 



  

“Quality” of education…what is this, precisely? Soft skills otherwise learned…what are they? Social ties Type of contract Family background (which may encompass some of the above causes..)

4. Wages of temporary and permanent workers, 2007 Wage gap among permanent and temporary employees (OLS estimations; workers aged over 24). Source: elaborations on EU-SILC 2007 data 60.0

49.7

49.3

48.7

50.0

44.7

45.5 43.1

42.3

41.3

39.7

40.0

38.7 35.2

34.6 31.5

30.9 30.0 25.6 21.0

24.2

23.1

20.2

20.0

22.2

22.1

19.4

17.6

15.9

14.8

18.6 15.0

14.7 9.0

10.0

2.4

Annual gross wages

Sp ai n

Po rtu ga l

Ita ly

G re ec e

U K

Ir el an d

Sw ed en

N or w ay

Fi nl an d

N et he rl~ s

Lu xe m bo ~g

Fr an ce

G er m an y

B el gi um

A us tri a

0.0

Hourly gross wages

The wage gap between temporary and permanent workers, after controlling for several variables, is in general extremely high. Here is the first possible explanation of high within-group labour inequality.

4. Wages of temporary and permanent workers, by education Hourly gross wage gap among permanent and temporary employees (OLS estimations on workres aged over 24), computed through current gross monthly wages, by educational attainment. Source: elaborations on EU-SILC 2007 data 20.0 17.6

18.0 16.0

17.6

15.8

15.6 14.7

14.9

14.5

13.8

14.0

13.2 12.5

12.4

12.0

11.0

11.0

10.6 9.7

9.7

10.0

12.4

9.0

8.0

7.2

6.8 6.0 4.0 2.0

1.3

0.0 Austria

Ireland

UK All workers

Greece Tertiary graduated

Italy

Portugal

Spain

No tertiary graduated

In most countries (for which appropriate data were available) the wage gap within homogeneous groups is high, esp. tertiary educated.

4. Intergenerational inequality – The chances to attain a tertiary degree Percentage changes in the probability to attain a tertiary degree, by gender and parental occupations1 (compared to offspring of managers). Marginal effects estimated by an ordered probit model2.

Male

Germany

France

Spain

Italy

UK

Ireland

Denmark

Finland

White collar

-22.7%

-23.5%

-6.4%

-15.0%

-17.8%

-7.9%

-32.2%

-16.6%

Blue collar

-36.1%

-43.2%

-34.8%

-46.9%

-34.5%

-26.4%

-52.1%

-31.7%

White collar

-26.3%

-23.3%

-6.5%

-15.1%

-16.9%

-7.6%

-28.8%

-14.1%

Blue collar

-42.2%

-42.7%

-35.4%

-47.0%

-32.8%

-25.3%

-46.5%

-27.0%

Parental Occup.

Female

Parental Occup.

Highest occupation got by father and mother; 2 Representative individual: 40 years old, native, with one sibling, living with both parents during his youth and whose parents highest degree is upper secondary. Source: elaborations on EU-SILC 2005 data

1

Family background has several influences, some are well known. Focus on the not too well known: effects on the probability to work as temporary employees and to become a manager

4. Wage differentials and family background (2) Background premia. Gross (net in IT, ES) annual labour income gaps by parental occupations, controlling for offspring education and occupation. Individuals aged 35-49. N.B.: Lighter colours are n.s. at 95% level. Source: elaborations on EU-SILC 2005 data 16.0

14.8

14.0 11.7

12.0 9.9

10.0 8.2 8.0

6.3 6.0

4.9

5.1

4.0 2.8 2.0

0.8

0.9 0.2

0.0 -0.4

-0.6 -2.0 -2.3

-2.6

-3.1

-4.0 -6.0 Germany

France

Spain

Italy Parent white collar

UK

Ireland

Denmark

Finland

Parent manager

In UK and IE (at a lower significance level) as well as in Southern countries, family backgrounds have a further impact upon earned income of workers besides those working through the education and occupation channels.

A SECOND SUMMING UP o

“Within” inequality has among its causes:  The wage gap between temporary and permanent workers  Several influences of family background (at least in some countries): on the chances of getting a better education and a higher occupation, on the compensation - given both education and occupation

Conclusions and policy implications o

o

o

“Within” inequality is important and should be given more attention in the analysis of inequality in Europe Much remains to be done in order to know it better.  Our research is now focussing on its dynamic aspects It has far reaching implications also for policy making. In particular:  

 

on the notion of equality of opportunity on the relationship between education and inequality and, more generally, between inequality and growth on how “acceptable” inequality is on some crucial aspect of the European Employment

THANK YOU!

Suggest Documents