EGGE EC s Expert Group on Gender and Employment

Transcription

1 EGGE EC s Expert Group on Gender and Employment Title: Authors: Gender Equality and the European Employment Strategy: An Evaluation of the National Action Plans for Employment 2000 Final Version 2000 Jill Rubery, Damian Grimshaw, Mark Smith EWERC, Manchester School of Management, UMIST Copyright Disclaimer: This report was produced as part of the work of the European Commission s Expert Group on Gender and Employment (EGGE) and was funded by the European Commission. The European Commission has granted permission for this report to be published in this website. The opinions and views expressed in this report remain the responsibility of the authors and authors alone, and should not be taken to be representative of the views of the European Commission.

2 Gender Equality and the European Employment Strategy: An Evaluation of the National Action Plans for Employment 2000 Jill Rubery Damian Grimshaw Mark Smith European Work and Employment Research Centre (EWERC) Manchester School of Management UMIST Final Version 2000 Country fiches prepared by Phil Almond and Marilyn Carroll, EWERC EU experts group on gender and employment* Danièle Meulders, Belgium Tracey Warren, Denmark Friederike Maier, Germany Maria-Luisa Moltó, Spain Rachel Silvera, France Maria Karamessini, Greece Ursula Barry, Ireland Paola Villa, Italy Robert Plasman, Luxembourg Janneke Plantenga, The Netherlands Ulli Pastner, Austria Maria do Pilar González, Portugal Reija Hannele Lilja, Finland Lena Gonäs, Sweden Jill Rubery, UK * see page 80 for list of evaluation reports and authors

3 CONTENTS Page No. INTRODUCTION 1 1. GENDER AND EUROPEAN EMPLOYMENT I.1 GENDER AND THE EU EMPLOYMENT CONTEXT 3 I.2 EMPLOYMENT OPPORTUNITIES 5 I.3 UNEMPLOYMENT GAP 9 I.4 GENDER PAY GAP 12 I.5 SEX SEGREGATION 15 I.6 INFLUENCE OF EDUCATION AND PARENTHOOD ON EMPLOYMENT OPPORTUNITIES 18 I.7 SUMMARY 21 II EVALUATION OF THE NATIONAL ACTION PLANS II.1 GENDER MAINSTREAMING IN THE NATIONAL ACTION PLANS 23 II.2 GENDER MAINSTREAMING IN THE FIRST THREE PILLARS 40 II.3 GENDER EQUALITY POLICIES: THE FOURTH PILLAR 53 III. FUTURE PROSPECTS AND FUTURE PRIORITIES III.1 DEVELOPMENT OF EQUAL OPPORTUNTIES POLICY SINCE THE BEGINNING OF THE LUXEMBOURG PROCESS 65 III.2 CONCLUSIONS AND FUTURE PRIORITIES 67 III.3 BEST PRACTICES 74

4 INTRODUCTION Gender equality has been established as a central part of the European employment strategy since the start of the Luxembourg process. Strengthening equal opportunities between women and men was adopted as the fourth pillar of the employment guidelines at the Luxembourg summit in 1997, alongside the pillars of employability, adaptability and entrepreneurship. This decision set the scene for equal opportunities to be considered an integrated and complementary part of the European employment. Each Member State is under an obligation to draw up a National Action Plan following the agreed employment guidelines, under the four pillars. These National Action Plans are subject to assessment by the European Commission and since 1999 Member States have also been the recipient of recommendations based on these assessments, but endorsed by the Council of Ministers, as to how the National Action Plans could be strengthened in the future to meet the objectives of the European employment strategy. The inclusion of equal opportunities as a compulsory fourth pillar has thus provided a new platform for the development of a European-wide approach towards reducing gender inequalities. The inclusion of a new guideline in 1999 requiring Member States to adopt a gender mainstreaming policy throughout the National Action Plan provided a further major impetus to the integration of equal opportunities issues into the employment framework. Moreover, mainstreaming has not been used within the guidelines as a substitute for direct gender equality measures; indeed in the 1999 guidelines the second major change to the fourth pillar involved the introduction of the gender pay gap as a policy issue to be integrated into the employment strategy. The overall approach, therefore, follows the twin track recommended by the Council of Europe and other bodies that have explored the scope and role for mainstreaming. Mainstreaming provides the opportunities to build gender equality issues into all policy programmes and has the potential double benefit of ensuring that gender effects are taken into account in the initial design and of providing a basis for new and transformatory approaches to policy-making. However, the introduction of mainstreaming does not reduce the need for gender specific policies to deal with gender inequalities and deficits. The maintenance of a specific equal opportunities programme has the added benefit of providing a continuing impetus towards gender mainstreaming which may be lacking if interests in dedicated gender equality policies were to lapse. The objective of this report is to evaluate the extent to which this twin tack approach to gender equality in employment policy, as implied by the employment guidelines, has been translated into actual policy programmes and initiatives at the Member State level. The Luxembourg process has provided an unparalleled opportunity to put equal opportunities at the centre of European employment policy-making, but the extent to which this opportunity is turned into a reality depends on the reactions and responses in the Member States. The report consists of three parts. First we outline the background to European employment policy in 2000 from a gender perspective. This statistical analysis of the employment situation also adopts a twin track approach; we focus not only on gender equality issues, relating to the continuing patterns of gender differences and gender inequalities, but also introduce gender mainstreaming, by considering the interrelationships between the gender patterns of employment and the pattern of restructuring and growth within European labour markets. In the second 1

5 section we focus on an evaluation of the 2000 National Action Plans. First we consider the development of gender mainstreaming, both from a methodological and process perspective and from a policy perspective, followed by a detailed discussion of mainstreaming in the employment measures included in the first three pillars. The final part of this evaluation considers the attention paid to specific gender equality policies within the National Action Plans through the measures included under pillar four. The third section assesses progress to date under the Luxembourg process, both with respect to gender mainstreaming and gender equality, and considers prospects and priorities for the future. Examples of best practice are provided as a contribution to the possible spread of best practice across Member States and as examples of the various ways in which gender can be taken into account in a very wide range of different policy areas. 2

6 I. GENDER AND EUROPEAN EMPLOYMENT I.1 GENDER AND THE EU EMPLOYMENT CONTEXT The Presidency Conclusions from the Lisbon European Council (March 2000) stated that despite the apparently stable and healthy macroeconomic outlook across the EU, major weaknesses remain: More than 15 million Europeans are still out of work. The employment rate is too low and is characterised by insufficient participation in the labour market by women and older workers. Long-term structural unemployment and marked regional unemployment imbalances remain endemic in parts of the Union. The services sector is underdeveloped, particularly in the areas of telecommunications and the Internet. There is a widening skills gap, especially in IT where increasing numbers of jobs remain unfilled. (Lisbon, Presidency conclusions, March 2000) It is by now widely recognised that the key to raising the EU employment rate lies in enabling women s greater participation in the labour market. This is demonstrated in countries such as Denmark, Sweden and the UK where a high female employment rate is accompanied by a relatively high employment rate - closely matching the US pattern of employment participation. Part of the explanation lies in women s high share of net job creation in the expanding services sectors. In addition, a significant portion of new jobs are part-time, which are more likely to be filled by women. However, across the EU there are a number of obstacles to increasing the number of women working, reflecting weaknesses on the demand side of the economy as well as adverse incentive structures on the supply side. In many countries, there is sluggish growth in public and private sector services. Underdeveloped childcare services infrastructure hinders women s participation over the life-cycle. Persistent high levels of sex segregation, in a limited number of sectors and occupations, constrain women s employment opportunities. Also, slow progress in closing the gender pay gap diminishes the incentive to joining the workforce by concentrating opportunities for women s employment in low paid jobs. In this section, we assess the significance of gender equality to the overall EU employment programme, drawing on harmonised pay and employment data as well as the national reports. We explore progress across the EU and within individual member states towards closing the gender gaps across five dimensions of labour market policy: employment opportunities; unemployment; pay; employment segregation; and the influence of parenthood and education on employment. 3

7 Figure I.1a. Composition of male population of working age in the EU, 1999 inactive 22% unemployed 7% FT perm 48% family worker 1% self employed 13% PT temp 1% FT temp 6% PT perm 2% Source European Labour Force Survey 1999 Figure I.1b. Composition of female population of working age in the EU, 1999 FT perm 26% inactive 42% PT perm 13% unemployed 7% family worker 1% PT temp 2% self employed 5% FT temp 4% Source: European Labour Force Survey

8 I.2. EMPLOYMENT OPPORTUNITIES According to labour force survey data employment in the EU grew by 2.8 million during 1999, 1 continuing the trend of net job creation since As a result, the overall headcount employment rate increased by more than one percentage point - from 60.8% in 1998 to 61.9% in Adjusting for full-time equivalent jobs, 2 the EU employment rate rose by a smaller amount (56.3% to 57.2%) reflecting the persistent significance of part-time work as a motor of employment growth; more than one third (35.2% 3 ) of net headcount job growth was in part-time work. During 1999, the majority of employment growth across the EU came through an expansion of female employment. The number of women working grew by 1.7 million in 1999, compared to 1.0 million for men, reversing the pattern in 1998 when male employment increased slightly more than female employment. The disaggregated picture - by member state and by sex (Figure I.2) - demonstrates that there is both a gender-specific and a country-specific dimension to employment growth across the Union. Growth in women s employment in the former West Germany, Italy, the UK, the Netherlands and Belgium made a major contribution to overall EU employment growth. The case of West Germany is notable; in 1999 women in West Germany constituted just 8% of total employment in the EU (excluding Greece), yet accounted for 12% of net job growth (outstripping men s contribution by more than 300%). In member states where male employment growth registers a significant contribution (Spain and France), this is coupled with similar increases in female employment growth, suggesting that it is the expansion of all employment in the country that is significant. Between 1994 and 1999, six years of sustained job growth have contributed to a slight narrowing of the gender gap in employment rates (Figure I.3). Between 1998 and 1999, the EU female employment rate rose from 51.2% to 52.6%, compared to a smaller rise from 70.8% to 71.6% for men. Nevertheless, the pace of change is still slow. The gap in 1999 was still 19 percentage points. Across the 15 member states, there continues to be an approximate correlation between the size of gender gap in employment and the female employment rate. Sweden, Finland and Denmark combine narrow gaps with high female employment rates, whereas Greece, Spain, Italy, Luxembourg and Ireland all combine wide gaps (above the EU average) with low female employment rates (Figure I.4). The former East Germany stands out somewhat as an exception to the general pattern, with one of the narrowest gender employment gaps (9.3 percentage points) but a female employment rate of just 58%. Also notable is the case of Ireland, where the relatively wide gender gap masks a strong rise in the employment rate for both men and women (from 68.02% to 71.8% and 44.62% to 47.7%, respectively, between 1997 and 1999). 1 Data refer to E14, since ELFS 1999 data for Greece are unavailable. 2 Here, the FTE measure is defined as total number of full-time workers plus half the number of part-time workers (i.e. we assume part-time work is, on average, equivalent to half a full-time job). 3 Again, data exclude Greece. 5

9 Figure I.2. Contribution of male and female employment by country to E14 employment growth (headcount), Dw F 12.0% OTHER 31.2% ES F 11.9% Dw M 3.3% BE F 3.4% FR F 5.1% FR M 5.2% NL F 5.4% UK F 5.5% IT F 6.9% ES M 10.2% Note: 'Other' refers to contributions that fall below 3.0%; data for Greece are unavailable for 1999 and are therefore excluded here. Source: European Labour Force Survey 1998, Figure I.3. Employment rate in the EU 75% 70% 65% 60% 55% 50% 45% 40% male headcount male FTE female headcount female FTE Source European Labour Force Survey

10 Figure I.4. Gender gap in employment and the female (headcount) employment rate, ES IT GR 25 LU IR E15 BE Dw AT PO FR NL UK 10 De DK 5 FI SW 0 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% Female employment rate Note: Data for Greece are for Source European Labour Force Survey Headcount vs Full-time Equivalent (FTE) measures Comparison of member states performance on the basis of the headcount measure of employment rates needs to be supplemented by use of data adjusted for full-time equivalent jobs calculated here by taking a part-time job as equivalent to half a fulltime job. This has particularly striking implications for measures of the gender gap in employment, reflecting the very different compositions of full-time and part-time employment in different member states, the tendency for women to be overrepresented in part-time work and for men to work relatively long hours. For the EU as a whole, the FTE gender gap in employment in 1999 was 25.6 percentage points, compared to the headcount measure of 19.0 percentage points. The difference in gaps between the two measures has increased slightly over recent years, due to a larger reduction in the headcount gap relative to the FTE gap (Figure I.3). This reflects the growing proportion of women, and to a lesser extent of men, in part-time employment. During 1999, while full-time job growth among men and women was relatively similar across the EU (1.0 million and 0.9 million respectively), women also made substantial net gains in part-time work (a rise of 0.8 million jobs, compared to 0.2 million for men). 4 Job growth through part-time work was particularly important in a number of member states: in the unified Germany female part-time employment was the main driver of employment growth (a rise of 0.27 million out of a total of 0.55 million), and among men net creation of part-time work outstripped that of full-time 4 As with preceding analysis of employment change between 1998 and 1999, data for Greece is unavailable. 7

11 Figure I.5a. Employment gender gap AT BE Dw DK De ES FI FR GR IR IT LU NL PO SW UK E Source European Labour Force Survey work (a rise of 0.05 million and 0.04 million respectively); in Italy, where expansion in women s employment made a significant contribution to overall EU growth, part-time work was the main engine of growth; and in the Netherlands female part-time employment grew by 0.1 million, while female full-time employment remained unchanged. Overall, once we adjust for full-time equivalent jobs, differences in female employment rates between states are considerably diminished. Using the FTE measure, the spread extends from a low of 34% (Belgium) to a high of 60% (Denmark), whereas the more commonly used headcount measure displays a far wider spread, from 37% (Spain) to 72% (Denmark). In practice the top four performers based on the smallest gender gaps in employment on a headcount basis, Sweden, Finland, East Germany and Denmark, were also the top four countries on the FTE measure (Appendix Table 1). However, for a number of member states, use of the headcount measure as opposed to the FTE measure is critical in determining their relative performance regarding the female employment rate (Figure 5; Appendix table 1). In particular, the high ranking of the UK and the Netherlands on the headcount measure drops substantially (from fifth to eighth place, and from eleventh to bottom respectively); these changes in position clearly reflect the high proportion of part-time employment in both countries. Conversely, the ranking of Portugal improves from seventh to fifth, due to the relatively low rate of part-time employment. For both the UK and the Netherlands, the aboveaverage gender employment (FTE) gaps were identified as key structural challenges in the Commission recommendation for Council Recommendations (2000). 8

12 Figure I.5b. Full time equivalent gap AT BE Dw DK De ES FI FR GR IR IT LU NL PO SW UK E Source European Labour Force Survey I.3 UNEMPLOYMENT GAP Since 1997, there has been mixed progress in closing the gender gap in unemployment across the EU. At the 1997 Luxembourg summit, it was noted that the gap between male and female unemployment in the Union was close to three percentage points (9.8% and 12.5% respectively). During 1998, the gap widened to 3.1 points. However, in 1999 some progress was achieved, with a slight narrowing of the gap to less than three points (8.4% for men and 11.2% for women). Across member states, the gender unemployment gap in 1999 was by far the highest in three countries - Spain, Greece and Italy - where the gender gap (in percentage points) was 12.1, 9.6 and 7.4, respectively. As Figure I.6 demonstrates, women experience a higher risk of unemployment than men in 12 member states (considering East and West Germany separately). However, the general pattern in these countries is towards a narrowing of the gender gap; notably, a substantial improvement was achieved in Belgium, East Germany, Denmark and Portugal. Exceptions to the general pattern were Finland and Italy where the gap widened during In four member states (West Germany, Ireland, Sweden and the UK), women constitute the under-represented gender among the unemployed, and while the gap closed marginally in Ireland and Sweden, it widened in West Germany and the UK. It is worth noting that in Germany, women are slightly under-represented among the unemployed in the old Länder (an unemployment rate of 7.2 for men and 6.8 for women), but over-represented in the new Länder (15.0 and 18.8, respectively). 9

13 Figure I.6. Unemployment gap AT BE Dw De DK ES FI FR GR IR IT LU NL PO SW UK E Source: European Labour Force Survey No straightforward relationship can be found between trends in the employment and unemployment gender gaps. This is largely because patterns of women s employment are more likely to be associated with changes in inactivity status than is the case with men (Rubery et al. 1998). Member states can be grouped according to the pattern of trends in the two gaps during As we saw above, in twelve out of fourteen member states (no data for Greece), the employment gap narrowed. However, while the majority of countries made simultaneous progress on both gaps, with a narrowing of the employment gap and the unemployment gap, in Finland and Italy a narrowing of the employment gap occurred alongside a widening of the unemployment gap. This pattern of change suggests that the bulk of women s employment expansion is fuelled by a shift out of inactivity status rather than out of unemployment: in Finland, the proportion of women in inactivity dropped by almost four percentage points during 1999 (from 30.0% to 26.2%) compared to a smaller reduction in the unemployment rate (from 13.6% to 12.5%); and in Italy, the share of women in activity fell by more than one percentage point (55.5% to 54.4%), against a marginal reduction in the female unemployment rate (from 16.7% to 16.4%). Finally, in Austria and Spain - the two countries that witnessed a widening of the employment gap - there was apparent progress in reducing women s risk of unemployment relative to men. Notably, in the case of Spain, while the gender gap in unemployment rates narrowed, if we express unemployment as a share of the working age population, men experienced a greater reduction in the risk of unemployment than did women (from 10.6% to 8.4% for men and from 12.7% to just 11.2% for women). 10

14 Across the EU, systems of unemployment benefits in many member states tend to discourage women from registering as unemployed during periods of job search in favour of remaining outside the labour market (Rubery et al. 1998). Hence, whether the data demonstrate an over- or underrepresentation of women among the unemployed, it is likely that women s share of those willing and able to work is underestimated. National reports for selected member states illustrate this situation. In the UK, combining the ILO measure of the number of persons inactive but willing to work with the ILO measure of unemployed suggests a slightly larger potential labour supply of women than men (2.048 million and million respectively, 1999 data); this contrasts markedly with the UK claimant count measure with female unemployment of just 0.3 million and male unemployment of million. In addition, in many member states women are more likely than men to be ineligible for unemployment benefits due to over-representation in marginal employment - employees with weekly (or monthly) income that falls below the entitlement threshold for unemployment benefits. Forty six per cent of the unemployed in the EU had been out of work for 12 months or more in 1999 and for women the share was 47% compared to 45% for men. The share of unemployed women who had been unemployed for a year or more exceeded half in Belgium, Germany, Greece, Italy and Spain while in the Nordic countries and the UK the share was less than a quarter (figure I.7). A higher share of women than Figure I.7: Share of the unemployed without work for 12 months or more by gender, men women AT BE DK DU ES FR GR IRL IT LU NL PO SW SF UK Note: Date for Greece refer to 1998 Source: European Labour Force Survey:

15 men had been unemployed for more than twelve months in seven member states and there were particularly large gender gaps in Austria, Spain and Greece but in the UK, Ireland and Luxembourg the gender gap was reversed by a similar magnitude. Of course, data on long-term unemployment do not reveal whether the trends reflect entry into employment or inactivity. In certain contexts (such as Ireland, for example). I.4 GENDER PAY GAP Pay data for 1995 from the Structure of Earnings Survey show that women, on average, are paid 73% of men s gross hourly wages across the EU (Ireland excluded). The data exclude public sector earnings and are thus likely to under-estimate women s relative pay since the public sector provides an important source of high paid jobs for women in particular. The pay gap is largest in the UK and the Netherlands, where women s average pay is only two thirds of men s (66% and 69% respectively). By contrast, the gap is relatively narrow in East Germany, the Nordic countries, Luxembourg and Belgium (women s pay exceeds 80% of men s in each case, Figure I.8). It is worth noting that action on the pay gap was explicitly identified in the Council Recommendations for the UK. The structure of Earnings Survey does not include data for Ireland; however, national data reveal a gender pay gap of 71% (Barry 2000) As figure I.8 demonstrates, the average gender pay gap is narrower across the Union when data are restricted to full-timers only (75%). The pattern varies by member states, for while some register only a marginal difference (Greece, Italy, Portugal, Finland), the majority follow the trend at EU level with a difference between the two measures of women s average pay between one and three percentage points. For example, women s average pay compared to men in Belgium for full-time workers is 83.2%, but drops to 80.4% with inclusion of part-time pay data. However, one country stands out. In the UK, women s average pay compared to men drops by eight percentage points (from 74% to 66%), accompanied by a drop in the position of the UK in the EU ranking from twelfth place to bottom. The explanation is largely due to a high average wage differential between full-time and part-time employees, coupled with a high share of women in part-time employment; on average, female part-time workers in the UK earn only slightly more than half the hourly pay of male full-timers (52%) compared to an average ratio across the Union of close to two thirds (65%). By contrast with the UK, the situation of Sweden is notable since despite a similarly high share of women in part-time work (41% in Sweden and 43% in the UK), inclusion of part-time pay data reduces the gender pay ratio by just two percentage points (from 87.0% to 84.8%); the obvious reason is the relatively good pay prospects for women in part-time work in Sweden - in 1995 female part-timers earned, on average, 80% of the pay of male full-timers. In fact, ranking member states by the pay differential between women in part-time work and men in full-time work, and between women in part-time work and women in full-time work, the UK ranks bottom in both cases. 12

16 Figure I.8. Gender pay gap, 1995 (a) all workers 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% UK NL PO AU ES FR GR Dw IT BE LU FI DK SW De E14 (b) full-time workers 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% NL PO AU UK ES GR IT FR Dw FI DK BE LU SW De E14 Note: gross hourly earnings data, overtime included. No data for Ireland. See Appendix table 2 for details. Source: SES

17 Figure I.9 demonstrates the positive relationship between the size of the gender pay gap and the level of wage inequality (expressed as the inter-decile ratio). Member states with a high wage differential between the lowest and highest paid groups (Spain, France, the UK and, in particular, Portugal) are at the bottom of the ranking of countries by size of gender pay gap. Conversely, those states with the most compressed wage structures (Sweden, Finland and East Germany) have a relatively balanced gender pay differential (ranked in the top four by size of gender pay gap). A notable exception to the general pattern is Luxembourg, which combines high wage inequality with a relatively narrow gender pay gap. Recent policy developments in a number of member states have contributed to a narrowing of the gender pay gap by aiming to reduce overall wage inequality. In particular, the implementation of a minimum wage in the UK (April 1998) narrowed the pay differential between the lowest decile group and the median; similar expectations are held regarding the minimum wage recently introduced in Ireland (April 2000). Figure I.9. Wage inequality and the gender pay gap 3.5 PO UK D1/D9 2.0 ES NL FR GR IT BE LU DK Dw De 1.5 FI SW % 65% 70% 75% 80% 85% 90% 95% 100% Gender pay ratio Note:, The gender pay ratio is for all employees (full-time and part-time); Wage inequality is measured as the inter-decile ratio, D1/D9, calculated from industry pay data (48 classifications); Gross hourly pay data, overtime included; Austria is excluded due to lack of industry pay data. Source: SES,

18 Finally, given that the bulk of opportunities for women s employment growth are in the private services sector, it is of interest to note that, for the Union as a whole, the size of the gender pay gap in industry is only marginally narrower than that in private services (73.5% and 73.0% respectively). However, this conceals divergent patterns across member states. In Denmark, France, Italy and the Netherlands the gap in private services is at least five percentage points wider than that in industry, whereas in East Germany, Greece and Portugal there is far greater pay equality in the private services than in industry (Appendix Table 2). Among the latter group of countries, however, the main driver behind pay equity in services in East Germany and Greece is a levelling down of men s wages between industry and services, rather than a substantially higher pay for women in services. I.5 SEX SEGREGATION The task of addressing sex segregation in employment is considered as secondary to the need to raise the female employment rate. In the words of the 1999 Joint Employment Report: However, it should be recognised that a segregated labour market with high employment rates for women is preferable to one with less segregation but low employment rates for women (p.70). It is no surprise, therefore, that the three countries targeted by the Council Recommendations to improve sex segregation are among the best performers in terms of the female employment rate (Denmark, Finland and Sweden). A strategy that prioritises female employment rate over sex segregation would of course be undermined by evidence that there may be a causal relationship between high sex Figure I.10. Female employment rates and the level of segregation in the EU ES IT AU Du PO Note: Working age employment rates; for index of dissimilarity, 1991 data for AU, 1995 data for FI. Source: Bettio et al. (1998) for index of dissimilarity; European Labour Force Survey ( ) for employment rates. GR FR BE LU IR 0% 10% 20% 30% 40% 50% 60% 70% 80% Female employment rate NL FI UK SW DK 15

19 segregation and slow employment growth among women. For example where strong job growth is concentrated in male-dominated sectors, failure to improve equal opportunity of access to jobs would hinder efforts to increase women s employment. Also, as identified in a number of national reports (in particular Italy, Sweden and the UK), women s concentration in the public sector means that levels of female employment are highly vulnerable to policies of privatisation, often accompanied by downsizing, in the context of tightened public spending. The data for 1999 demonstrate a positive relationship between the degree of sex segregation and the female employment rate. Figure I.10 plots the female employment rate for 1999 against the index of segregation to show a strong positive correlation (r=0.84). The high employment rate countries of Finland, Denmark and Sweden are clustered at the top of the distribution with index of segregation scores of around 29.0 while the low employment rate countries of Italy and Greece have a low level of segregation at around The Netherlands has a relatively low level of segregation compared to its female employment rate while the Spanish employment rate is lower than countries with a similar level of segregation. Across the Union, women are still concentrated into a limited range of occupations. 5 On the basis of the 3-digit level classification of occupations, which includes 110 different categories, we find that one in three women (33%) work in just five occupations (Table I.1). The largest occupational group among women is the category of salespersons (8.1% of all women), followed by domestics (6.8%) and personal care workers (6.8%). This clustering of women s work is exacerbated by an even more concentrated pattern of employment among women in part-time work; 59% of all women in part-time work are employed in the top ten occupations as shown in table 1, compared to 44% of women in full-time work. Moreover, there is an overrepresentation of female part-timers in the top three occupations (female part-time shares of 49%, 66% and 47%, respectively, compared to an average for E13 of 34%). Table I.1. Top ten occupations for women's employment in the EU, 1999 Concentration of women's employment ISCO code All Full-time Part-time 522 Salespersons Domestics Personal care workers Secretaries Other office clerks Housekeeping and restaurant services Administrative associate professionals Numerical clerks Nursing and midwifery Secondary education teachers Total top ten concentration Note: Data exclude Greece and Ireland since occupational employment data are for 1999 Source: European Labour Force Survey Here, the data exclude Ireland and Greece since 1999 occupational employment data are unavailable. 16

20 Impact of restructuring It is commonly claimed that the potential for job growth in the EU is contingent upon improving the total employment share in the services sector. The key variable in this equation is women s employment, since a far greater share of women work in services than men (81% and 56% across the EU respectively). During 1999, net job growth in services was 1.8 million for women and just 0.9 million for men (3.5% increase in jobs for women and 2.0% for men) 6. Women accounted for 65% of the growth in services employment between 1998 and 1999 at the EU level. However, across member states women s role in job creation through the services sector varied considerably. In three member states the UK, France and Denmark women s share of job growth in services fell to just over a half (figure I.11), while in another three - Austria, Luxembourg and West Germany women s share was greater than 75%. Figure I.11: Female share of employment growth in services, % 80% 70% 60% 50% 40% 30% 20% 10% 0% AT BE Dw De DK ES FI FR IRL IT LU NL PO SW UK Note: Greece is missing due to lack of data for 1999 Source: European Labour Force Survey Sex segregation in employment may hinder efforts to close the gender gap in employment where women are not well represented among sectors of economic activity that most contribute to job growth. Considers the three sectors (at the 2-digit level, NACE) which contributed most to total net job growth at the EU level, one was female-dominated (health and social work) one was mixed (other business services) and one was male-dominated (computer and related activities) (Table 1.2). These three sectors combined accounted for 42.5% of total employment growth between 1997 and 1999; and female employment growth in these sectors accounted for 23% of total employment growth. Across member states, there are notable examples of both high and low female shares of employment among the fastest growing sectors. The relatively mixed sector other business services appears in the three sectors contributing most to employment growth in nine countries, and the female-dominated sectors of retail and health and social work appear in seven countries and six 6 Again, data exclude Greece 17

21 countries, respectively. By comparison, male-dominated sectors such as construction and computer and related activities appear among the top three in just four countries and two countries, respectively. In some countries, job growth individual femaledominated sectors outstripped total net job growth. For example, in Denmark, total employment in `health and social work grew by 28,000 compared to total net job growth of just 14,000 (a contribution of 215%). I.6 INFLUENCE OF EDUCATION AND PARENTHOOD ON EMPLOYMENT OPPORTUNITIES The gender gap in employment can be further examined by exploring different characteristics that shape the supply side of male and female labour. Here, we explore the role of education and parenthood in promoting, or hindering, equality of employment opportunity between men and women. Education For both men and women, higher levels of education are associated with higher rates of employment. Moreover, equality of employment opportunities between men and women tend to rise with higher levels of education. Across the EU (excluding Greece and Ireland), among those with a low level of education women s employment rate is only half that of men s (27% and 50% respectively). By contrast, women and men with a high level of education have a relatively similar rate of employment (73% and 79% respectively). The gender gap is thus substantially narrowed as a function of education (from 23 points, to 13, to 6, for low, medium and high levels of education respectively, Figure I.12). Figure I.12: Employment rate gap between women and men by education, low medium high 0-10 Source: European Labour Force Survey

22 Table I.2: Contribution to employment growth of top three NACE 2-digit sectors, Contribution to total growth Contribution to total growth Contribution to total growth Sector F F + M F share Sector F F + M F share Sector F F + M F share AU Retail 53% 78% 68% Construction 9% 63% 8% Computer & related 6% 34% 23% BE Other Business services 35% 58% 48% Metal manufacture 3% 58% 6% Retail 29% 57% 57% Dw Health & social work 41% 48% 76% Other Business services 23% 47% 53% Motor manufacture 5% 18% 18% De Health & social work 66% 82% 81% Other service activities 37% 74% 61% Other Business services 16% 56% 52% DK Health & social work 285% 215% 86% Retail 154% 208% 57% Food manufacture 38% 154% 38% ES Construction 3% 54% 4% Retail 21% 32% 56% Other Business services 12% 26% 50% FI Other Business services 22% 61% 49% Retail 39% 45% 65% Wholesale 12% 43% 33% FR Other Business services 9% 18% 48% Education 6% 12% 65% Health & social work 10% 11% 75% IR Construction 0% 21% 4% Other Business services 12% 18% 51% Retail 12% 18% 60% IT Public admin 14% 43% 34% Other Business services 17% 36% 44% Health & social work 16% 22% 57% LU Health & social work 22% 33% 73% Publishing 11% 22% 33% Extra territorial 11% 22% 44% NL Other Business services 23% 60% 43% Education 34% 38% 56% Construction 3% 30% 7% PO Other service activities 37% 39% 86% Financial intermediation 13% 29% 31% Education 20% 20% 75% SW Retail 100% 188% 62% Education 124% 165% 69% Health & social work 112% 112% 87% UK Education 39% 57% 70% Computer & related 11% 47% 25% Other Business services 21% 44% 44% E14 Other Business services 10% 20% 48% Health & social work 11% 14% 76% Computer & related 2.2% 8.4% Note top 3 sectors ranked by contribution of male and female employment to total employment growth F share = female share of sector in 1999 F = female contribution to total employment growth F + M = female and male contribution to total employment growth 1999 data for Greece is missing Source : European labour force survey

23 This pattern is generally consistent across all member states. Minor exceptions are evident in Finland, France and the UK where the gap among medium educated men and women is marginally wider than among those with a low level of education, suggesting that the employment gains to increasing education from low to medium are greater for men than for women. The major finding is the divergence in size of gaps by education across member states. The gender gaps in employment rate by level of education are relatively similar in Denmark, Finland, France, Sweden and the UK, while the spread is far wider in Austria, Belgium, Germany (East and West), Spain, Italy and the Netherlands. For example, in West Germany while low educated men are almost twice as likely to be in employment as low educated women (rates of 55% and 29%), there is relatively little difference among high educated men and women (77% and 72%). There are also differences across member states regarding the differential impact of education on women s employment rate. Denmark, Portugal and the UK stand out as member states where women with low education have a high employment rate (53%, 54% and 49% respectively) and the gains to increasing education are (partly as a consequence of the high starting point) relatively small. Conversely, women with low education in Belgium, Italy and Spain are far less likely to be in employment (rates of 19%, 20% and 21% respectively) and the gains to education are far higher. For example, the gap between women with low and high education in Denmark is 26 percentage points (rates of 53% and 79%), whereas in Belgium the gap is 57 points (rates of 19% and 76%). Parenthood The presence of children in the household has a double-pronged adverse impact on the gender gap in employment: women s employment rate is reduced; and men s employment rate is increased. Hence, the gender gap among men and women with young children is significantly wider (Figure I.13). Employment rates for men aged with children under seven were at least 90% in all member states 7 compared to just three member states for men with no children. The impact of a young child on the employment rates of women is particularly strong in the UK, Germany and Italy where rates are more than twenty points lower than for women with no children. In another six countries the employment rate falls by at least ten percentage points. These patterns open the gender gap in employment rates to over 45 percentage points in Ireland, Italy, Spain and Greece, while only in Austria and Belgium does the gap fall below 30 percentage points. 7 Data exclude Denmark, Finland and Portugal 20

24 Figure 1.13: Employment rates of men and women with a child aged 0-6yrs, AT BE Du ES FR GR IR IT LU NL SW UK men women Note: Greek data refers to 1998 Source: European Labour Force Survey 1999 I.7 SUMMARY Low levels of female employment in some member states remain a weakness of the economic outlook in the European Union. Although some countries display high level s of female employment, there remains considerable diversity on aggregate measures of employment and other measures of equality. During 1999 some countries made significant progress in reducing gender gaps in employment and unemployment but other gender differences in the labour market persist. The following points summarise the main areas of both continuity and change with respect to gender which therefore provide the context for the development of the gender dimension to the European employment strategy Women were the major contributors to net employment growth between 1998 and 1999 accounting for 1.8 million of the 2.8 new million new jobs. Strong employment growth for women has helped narrow the gender gap in employment rates slightly. Wide gender gaps tend to be associated with low female employment rates. Headcount employment rate measures still overestimate the amount of female employment in the EU compared to full-time equivalents. Measuring employment on a full-time equivalent basis reduces the relative performance of the Netherlands and the UK. Women face higher unemployment rates in 12 of the 15 member states. In the Southern member states of Spain, Italy and Greece women experience considerably higher unemployment rates than men. There has been a substantial improvement in the gender gap in Belgium, the former East Germany, Denmark and Portugal. However, in Finland and Italy the gender gap in unemployment rates widened. 21

25 The gender gap in hourly earnings persists across the EU with women s earnings only 73% of men's gross hourly wage. There are considerable differences between countries with women in the Netherlands and the UK experiencing the largest pay gap. The lower hourly pay received by many part-timers increases the gender pay gap. The introduction of minimum wages in the UK and Ireland has proved beneficial to low paid part-timers in particular. Sex segregation of employment remains high particularly in the Nordic member states. Across the whole EU women are concentrated in a limited number of occupational groups with one third working in just five occupations. Service sector employment growth remains the major area of job opportunities for women in the EU but women's concentration in the public sector means that they are vulnerable to policies designed to cut public spending. Education attainment boosts women's employment rate and closes the gender gap across the EU. The increases in the female employment rate associated with education are particularly high in Belgium, Spain and Italy while in the UK, Denmark and Portugal the returns are lower. The presence of children increases the gender gap in employment rates by raising the employment rate of fathers and reducing the employment rate of mothers. Young children increase the gender gap in employment rates to over 45 percentage points in Spain, Ireland, Italy and Greece. 22

26 II. EVALUATION OF THE NATIONAL ACTION PLANS II.1 GENDER MAINSTREAMING IN THE NATIONAL ACTION PLANS The requirement to mainstream gender into all areas of the National Action Plans can rightly be considered one of the most innovative aspects of the European Employment Strategy. To what extent has this requirement been implemented and how should we assess the impact of this measure on the ways in which European employment policy has evolved since the start of the Luxembourg process? The answers to this question must be given at an individual country level, for wide variations remain between member states in both the level of gender equality achieved and in the implementation of gender mainstreaming into their employment actions. Nevertheless, it is possible to conclude that in a majority of member states the incorporation of this requirement has had a tangible and positive impact on both the visibility of women in employment policy and on the development of a policy approach which has the prospects of improving gender equality. While there are clear differences between countries in the enthusiasm and effectiveness with which gender mainstreaming has been taken up, it is not easy to classify countries with respect to the implementation of gender mainstreaming. This arises from two main factors. First there is the problem of different starting points; in some cases countries with an already strong record of gender equality have not apparently made major changes to their policies in the light of the mainstreaming requirement and thus as far as the interpretation of the NAP is concerned they may be considered rather weak with respect to mainstreaming; at the other end of the spectrum some countries with major gender gaps have apparently embraced the gender mainstreaming objective with more enthusiasm. However, it may be premature to decide that these countries can be said to have adopted gender mainstreaming more than the former as we would be basing the classification more on words than deeds and on future commitments rather than past achievement. The second problem is that in all cases there are strong and weak elements to the gender mainstreaming process and it is difficult at this stage to judge the trade-off between these different elements, particularly when the weak elements relate to the implementation and evaluation of mainstreaming. Therefore we have tried to summarise the strengths and the weaknesses separately, without at this stage coming to an overall judgement (see table II.1). Evidence of gender mainstreaming needs to be derived not just from the NAP document but also from knowledge about the processes of policymaking in general within the member state concerned. If we take this broader perspective then it is clear that the three Nordic countries have a more developed commitment to mainstreaming, with, for example, mainstreaming principles influencing local government policy as well as national government policy. Moreover the mainstreaming initiative is supported both by a good care infrastructure and a high involvement of women in decision-making. Against this background it can nevertheless be said that in some respects the 23

27 Table II.1 Gender mainstreaming: assessment of strengths and weaknesses Country Strengths Weaknesses Belgium Stronger commitments to mainstreaming at regional level, especially Flemish area Limited use of targets or evaluation. Few references to gender in the text. than at federal government level. Increased diversity of measures in 2000 NAP and some measures to allow flexible reduction of working time. Denmark Mainstreaming has been integrated into Danish labour market and taxation policies which are based on the individual; equal opportunities are included in all the employment guidelines and all counties in Denmark have to report on equal The gender implications of policies, for example toward activating older workers, towards more flexible working time arrangements or towards lower taxes for low income jobs are not identified. opportunities. The commitment to mainstreaming in policy is supported by a strong care infrastructure and by the involvement of women in decision-making. Germany Strong in Federal Employment Service and in women and work programme Federal government has not fully adopted mainstreaming and has failed to implement part of manifesto commitments on gender; absence of gender issues in social partners Greece Spain France Ireland Italy Mainstreaming and specific equality measures have increased; quotas have been set for active labour market programmes and some positive action measures taken for women to set up businesses. The care infrastructure has been improved and there is a commitment to set up an inter-ministerial committee to promote mainstreaming. Goal of increasing share of working age population in labour market is consistent with gender mainstreaming, but not fully followed through into policies. Pillar 1 and III mainstreamed, to include both targets in pillar 1 and equality issues in working time changes in pillar III. Policies to involve local actors and social partners. Strong commitment to data analysis. New inter-ministerial committee committed to 8 fields of action reviewed on an annual basis. Commitment to mainstream gender equality across all six operational programmes of the National Development Plan. Gender impact assessment guidelines have been adopted by cabinet and are to apply to 105 out of 109 measures in the NDP. References to women under all pillars- declared intention to approve a plan for equal opportunities which should allow the transition from an experimental to a general approach. Wide ranging policies to raise gender awareness. Mainstreaming policies supported by new initiatives to strengthen equality legislation and to provide innovative leave arrangements. negotiations; too little information re the position of women in the East. Equal opportunities is conceived as helping women to conform to the male norm. There are no measures to promote active parenting among men. The quotas under pillar 1 are not observed in practice and there are no equal opportunities measures under pillar 3. Very weak with respect to mainstreaming; all measures addressed to women. Most of the equality budget devoted to leave arrangements without specific commitments to a care infrastructure. Pillar II not mainstreamed but gender specific policies still present. Inconsistencies between family policy- particularly parental child allowances- and policy of integrating women into the labour market not yet resolved. Over-emphasis on registered long term unemployed introduces gender bias. Insufficient attention is paid to women returners and the lack of a care infrastructure. Limited commitment to develop a care infrastructure and failure to include women in active labour market policies according to their share of unemployment. 24