European Foundation or the Iproveent o Living and Working Conditions The gender pay gap Background paper Click or contents Wyattville Road, Loughlinstown, Dublin 18, Ireland. - Tel: (+353 1) 204 31 00 - Fax: 282 42 09 / 282 64 56 eail: postaster@euroound.europa.eu - website: www.euroound.europa.eu
Contents Introduction Dierences in hourly wages Dierences in incoe distribution Sectoral gender segregation Gender tie gap Factors in deterining pay Actions by the social partners Conclusions 1 1 3 4 6 7 8 9
Introduction Although collective agreeents and iniu wage laws are in principle gender-neutral, woen still end up earning less than en. This paper will briely explore soe dierent acets o the gender pay gap issue, based on data collected by the European Foundation or the Iproveent o Living and Working Conditions. The data presented below coes ro two o the Foundation s projects: the European Industrial Relations Observatory (EIRO) report Pay developents 2005, 1 and the ourth European Working Conditions Survey, due to be published early in 2007. Since the data is derived ro two dierent sources, the extent o coverage varies, as do deinitions o pay. While not a survey on pay per se, the European Working Conditions Survey does provide aggregated data on gender disparities in the labour arket, oering soe interesting cross-country coparisons: soe snapshots ro it o the gender pay gap will be provided in this paper. First, the paper will look at dierences in hourly wages between woen and en. Next, it will present dierences in the distribution o net onthly incoe between woen and en. Sectoral gender segregation in the labour arket will then be looked at. Following that, pay dierentials resulting ro gender divisions o working tie and the gender tie gap will be exained. Finally, the paper will explore the relative iportance o individual characteristics and job-related variables as deterinants o the position-related pay gap in the labour arket in the EU27 (the current 25 Meber States plus Bulgaria and Roania). Dierences in hourly wages The update Pay developents 2005 includes data on gender pay dierentials, in the or o an unadjusted gender pay gap : the dierence between the gross hourly pay o en and woen across the entire econoy. (The dierences in calculation ethods and period o reporting are indicated in the report.) As Figure 1 indicates, quite a wide, although narrowing raw hourly pay gap still exists throughout Europe. Narrowing wage disparities are progressing only slowly in the EU15 and Norway: there was a all o 2.3 percentage points between 2001 and 2005 (ro 20.4% to 18.1%). A slight break in the trend is, however, visible in 2004 2005, with the upturn in the hourly pay gap in the EU15. In the new Meber States (NMS), hourly pay is soewhat ore equal; in these countries, the pay gap oscillated at around 17% over the period 2003 2005. 1 Available online at http://www.eiro.euroound.europa.eu/2006/06/update/tn0606101u.htl European Foundation or the Iproveent o Living and Working Conditions, 2006 1
Figure 1: Unadjusted hourly gender pay gap, 2001 2005 21% 20% 19% 18% 17% 16% 15% 2001 2002 2003 2004 2005 EU 15 + NO NMS All Source: Collated ro EIRO Pay developents updates, 2001 2005 Yearly upward and downward changes see to be a eature o gender pay statistics; urtherore, soe variations in the igures can be explained by changes in the source, or nature o the data used. Where individual country data is available that spans several years, it oten shows sall decreases in the gender pay gap as in Austria, Denark, France, Ireland and the UK. By contrast, slight increases in the pay gap can be seen in Belgiu, the Netherlands and Sweden. However, all such changes should be interpreted with care. According to Pay Developents 2005, the lowest hourly-wage pay gap is in Malta: the gap is 4% (a woan on average earning 96% o a an s salary). The widest gap, o 26%, is in Gerany. This irrors the situation reported in the EIRO update Pay Developents 2004, in which Malta and Gerany held the sae positions. Other countries consistently reporting relatively narrow gender pay gaps are Slovenia (6.9%), Denark (10.1%) and Ireland (13.3%), while those with coparatively wide gaps include Greece (25.2%), Cyprus (25%), Czech Republic (25.1%) and Slovakia (23.7%) see Figure 2. 2 European Foundation or the Iproveent o Living and Working Conditions, 2006
Figure 2: Unadjusted hourly gender pay gap 2005, by country 30% 25% 20% 15% 10% 5% 0% Malta Slovenia Denark Ireland France Bulgaria Hungary Lithuania Norway Latvia Roania Poland Sweden Belgiu UK Austria Italy Luxebourg Netherlands Spain Portugal Finland Estonia Slovakia Cyprus Czech Republic Greece Gerany Source: Pay developents 2005 (EIRO, 2006) Dierences in incoe distribution When looking at the data ro the ourth European Working Conditions Survey o 2005, yet another type o pay gap between the sexes becoes apparent naely, the unequal distribution o incoe between ale and eale workers. The easure o incoe being used here should irst be explained. Using data or the average net onthly incoe ro the ain job, an incoe scale o 10 deciles has been created. This is designed to atch the distribution o earnings as collected by the European Structure o Earnings Survey 2002. The incoe scale relects the respondent s relative position in the pay structure or each country. A pay gap, as well as a position gap, is thus indicated. For ease o analysis, the 10-point scale has been converted into a three-point scale denoting lower, ediu and upper pay levels (in ters o earnings received ro the ain paying job). Each respondent is thereore classiied according to whether they are in the lower, ediu or upper third o the population in their country o residence. Figure 3, on p.4, illustrates the pay gap in six clusters o countries representing speciic welare odels (according to the Esping-Andersen classiication): Scandinavian (DK, NL, FI, SV), liberal (UK, IE) continental (BE, DE, FR, LU, AT), Mediterranean (EL, ES, IT, CY, MT, PT), and central and eastern European (CZ, EE, LC, LT, HU, PL, SI, SK), as well as a cluster o the two acceding countries (BG, RO). European Foundation or the Iproveent o Living and Working Conditions, 2006 3
Figure 3: Dierences in incoe distribution between sexes, by country cluster EU: DK, NL, FI, SE EU: IE, UK EU: BE, DE, FR, LU, AT EU: EL, ES, IT, CY, MT, PT EU: CZ, EE, LV, LT, HU, PL, SI, SK ACC: BG, RO ale eale ale eale ale eale ale eale ale eale ale eale 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Lower er pay Middle Mediu pay Pay Upper pay Source: Fourth European Working Conditions Survey, 2005. Note: Figures are based on sel-reported net onthly incoe, only or eployees (EU25 and BG, RO) having worked at least an hour in the week prior to the survey interview. The dierences in incoe distribution are striking. In ost o the country clusters, with the exception o Bulgaria and Roania, alost 50% o woen are positioned in the lower pay bracket. At the other end o the spectru, only an average o 20% o woen (again, with the exception o Bulgaria and Roania) all into the upper pay category, copared to twice as any en in that incoe bracket (40%). Sectoral gender segregation According to the results o the ourth European Working Conditions Survey 2005, European labour arkets are highly segregated: only 26% o Europeans work in ixed occupations, where the workorce is at least 40% eale and at ost 60% ale. In addition, around hal the eale workorce have jobs in two sectors: 34% o woen work in education and health, while 17% work in the wholesale and retail trade. By coparison, hal o the jobs taken by en are in three sectors: anuacturing (22%), wholesale and retail trade (14%) and construction (13%). Figure 4 urther illustrates the situation in dierent sectors. It shows the distribution o low, ediu and high pay across genders and sectors and it gives the percentage o eale workers eployed in each o the sectors. As can be seen, even in those sectors with a large eale workorce, such as education and health and the retail and wholesale trade, ore en than woen occupy the higher pay bracket. 4 European Foundation or the Iproveent o Living and Working Conditions, 2006
Figure 4: Gender segregation, by sector 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Agriculture Manuacturing Electricity, gas and water Construction 36% 31% 20% 11% Retail Hotels and restaurants Transport and counication Financial interediation Real estate Public adinistration Education and health Other services 55% 49% 26% 45% 43% 41% 75% 61% Lower pay Mediu Pay Upper pay Source: Fourth European Working Conditions Survey (2005) European Foundation or the Iproveent o Living and Working Conditions, 2006 5
The gender tie gap The concentration o woen in the lower pay bracket can be explained in part by their overrepresentation in part-tie positions (see Figure 5), which as shown in Figure 6 on p.7 are ostly low paid. The gender tie gap is the dierence between the contributions to the labour arket in ters o working tie o en and woen. According to the ourth European Working Conditions Survey, there is a tie gap in the EU27 o 23%: 7% o en work part tie, copared to 30% o woen. Figure 5: Extent o part-tie work and gender tie gap, by country 70 60 50 40 30 20 10 0 CH NL UK NO BE LU SE IE AT DK EU DE IT FR ES TR MT PL EL RO FI CZ EE LT HU BG LV PT CY SI HR SK ale eale Source: Fourth European Working Conditions Survey (2005) The consequences or earnings o working part tie are illustrated in Figure 6. In all the country clusters, between 70% and 80% o part-tie workers are in the lower pay bracket. However, within the subset o part-tie workers, low pay is distributed airly evenly between en and woen; high pay is also quite evenly distributed aong a sall proportion o en and woen working part tie (roughly 10%). Ireland and the UK are an exception to this: aong ale part-tie workers, 21% are in the higher pay bracket, copared to 5% o their eale counterparts. 6 European Foundation or the Iproveent o Living and Working Conditions, 2006
Figure 6: Gender dierences in pay in EU27, by eployent status EU: DK, NL, FI, SE EU: IE, UK EU: BE, DE, FR, LU, AT EU: EL, ES, IT, CY, MT, PT EU: CZ, EE, LV, LT, HU, PL, SI, SK ACC: BG, RO Part-tie Full-tie Part-tie Full-tie Part-tie Full-tie Part-tie Full-tie Part-tie Full-tie Part-tie Fulltie 0% 20% 40% 60% 80% 100% Lower pay Mediu Pay Upper pay Source: Fourth European Working Conditions Survey (2005) Figure 6 urther illustrates the dierences in pay between en and woen in ull-tie jobs: overall, en tend to be overrepresented in the upper pay bracket, and there are roughly twice as any woen as en in the lower pay bracket. Factors in deterining pay The indings ro the ourth European Working Conditions Survey were exained to deterine the probability o a respondent being in the botto third or the top third o the incoe scale. A statistical analysis peritted the study o a range o actors in deterining a worker s position on the incoe scale (while siultaneously controlling or all the others). The results are highly signiicant: the chances o a part-tie worker alling into the botto incoe category are 6.25 ties greater than the odds o a ull-tie worker doing so: this indicates that part-tie work is one o the key deterinants o risk or being in the lower incoe bracket. Furtherore, the results also deonstrate that gender has a signiicant ipact on an individual s place in the incoe scale, even ater all other variables have been controlled or. This eans that a woan has a signiicantly higher chance o alling into the lower incoe category, and a signiicantly lower chance o alling into the higher incoe category than has a an with a siilar occupation, ull-tie status, tenure etc. (This data ust, however, be interpreted with care as they take relative pay as a dependant variable and are, thereore, very sensitive to the act that woen tend to get lower-paid jobs, even i those jobs have siilar characteristics to those taken by en.) European Foundation or the Iproveent o Living and Working Conditions, 2006 7
Actions by the social partners Although issues relating to the gender pay gap are ostly regarded as a atter o national legislation and politics, soe aspects o pay equality have been the subject o collective bargaining. There is debate as to whether collective agreeents redress or reinorce existing gender pay gaps: while collective bargaining is a good echanis or identiying and cobating discriination, it can also preserve eleents contributing to persistent pay dierences. In principle, the social partners agree that the gender pay gap exists, as was relected in the signing o the Fraeworks o Actions on Gender Equality, a cross-sectoral agreeent o the European social partners in May 2005. However, in their capacity as parties to the collective bargaining process, the social partners oten disagree on what is the best strategy to ollow in order to address the issue whether copulsory easures, collective regulations or individual arrangeents should be adopted. In general, eployer organisations consider legislation and regulation in the area o gender equality as counterproductive, and would preer tailor-ade copany-level solutions. Many eployers would also preer to include gender issues as a part o wider diversity policies. The trade unions, by contrast, tend to preer regulation to individual solutions. They also tend to take active easures in the ield o gender ainstreaing in general and regarding the gender pay gap in particular. Such easures include raising public awareness, conducting research, providing training courses on negotiation or equal pay or training using a odel agreeent on equal pay. They also ipleent procedures to cobat gender discriination. Apart ro ipleenting legislation allowing or equal pay or equal work (or work o equivalent value), the social partners have ipleented soe proactive easures: cobating the gender pay gap by giving unds to the eale doinated/and or low-paid sectors through central level negotiations or exaple, in the case o an equality-pay increent included in the national wage agreeent in Finland in 2006. (re-)evaluating jobs, or classiying the quality o jobs (based on objective criteria with a gender perspective) could lead to a avourable valuation o typically eale occupations or exaple, the job evaluation ethodology developed in Lithuania in 2004; prooting voluntary pay audits: in the UK, the Trades Union Congress (TUC) undertook a pilot project in 2003 in which hundreds o workplace equal-pay representatives were trained to address the gender pay gap and carry out pay reviews. An evaluation o the project suggested that it had played a signiicant part in pushing equal pay up the negotiating agenda and in propting eployers to agree to equal pay audits. concluding national intersectoral agreeents on gender equality and gender balance in workorce coposition, such as the one concluded in France in April 2004: it addresses such atters as narrowing the gender pay gap, preventing aternity leave ro adversely aecting woen s career developent, and tackling labour arket segregation. More inoration on the actions o social partners in the ield o gender equality will be published in early 2007, in the report Coparative study on gender and career. 8 European Foundation or the Iproveent o Living and Working Conditions, 2006
Conclusions Dierences in pay or en and or woen are prevalent throughout Europe. As shown in this paper, woen tend to have lower hourly wages than en. Moreover, woen tend to be segregated in the labour arket into certain sectors, and a vast ajority work in part-tie jobs, which are oten low paid. There are urther labour-related consequences or woen resulting ro the gender pay gap that accuulate over the course o a lietie. Due to the part-tie nature o their work, woen ay not be covered by the collective agreeents in which yearly wage increases are negotiated or their ull-tie counterparts. Furtherore, lower wages translate into lower social security and pension contributions, and perhaps less access to suppleentary pension schees. Although the social partners have taken soe action to urther greater gender equality, the issue could still be higher on the collective bargaining agenda. Perhaps a acilitation o the conclusion o a raework agreeent at the European level could hasten that process and lead to real action. In addition, it could be exained how best practice in prooting gender equality in the public sector could be transerred to the private sector in a eaningul way. Further research needs to be done to enable a better understanding o woen s early career choices, to deterine what attracts woen to certain occupations and sectors. In addition, it would be interesting to see which particular skills and personal characteristics are attractive to eployers who oer higher pay, and ro that work towards acilitating young woen in developing these skills. Finally, ore research exaining the dynaics o the gender pay gap over the lie course would help to deterine where the pay gap starts and how it ipacts on other policy areas, such as education and social and personal services. Kasia Jurczak, European Foundation or the Iproveent o Living and Working Conditions EF/06/101/EN European Foundation or the Iproveent o Living and Working Conditions, 2006 9