Table of Contents
1. THE IMPORTANCE OF ANALYSING WOMEN’S LABOUR MARKET PARTICIPATION
2. THE INFLUENCE OF CULTURE ON WOMEN’S LABOUR MARKET ATTAINMENT
2.1.1 Preliminary theories including culture
2.1.2 Pfau-Effinger’s (1996, 2000, 2005) Gender Culture Models
2.2 HYPOTHESES, METHODS AND DATA
3.1 EMPLOYMENT STATUS
3.2 ATTITUDES TOWARDS GENDER ROLES
3.3 THE RELATION OF EMPLOYMENT STATUS AND ATTITUDES
APPENDIX 1: GENDER CULTURE MODELS IN WEST-EUROPEAN SOCIETIES
APPENDIX 2: SPSS SYNTAX AND IMPORTANT OUTPUT
APPENDIX 3: CONTRAST OF EVS EMPLOYMENT STATUS (EMPLOYED) AND ILO ACTIVITY RATE [IN %]
APPENDIX 4: EMPLOYMENT STATUS SPLITTED BY COHORTS
APPENDIX 5: ITEMS FACTOR 1 (IN EACH CASE THE MOST PRONOUNCED CATEGORY IS SHOWN (WITH “NEITHER BEING IGNORED)
APPENDIX 6: ITEMS FACTOR 2 (IN EACH CASE THE MOST PRONOUNCED CATEGORY IS SHOWN (WITH “NEITHER” BEING IGNORED)
APPENDIX 7: WOMEN’S ACTIVITY RATE BY AFFIRMATION OF EQUAL DIVISION OF WORK
1. The Importance of Analysing Women’s Labour Market Participation
Since women take part in paid labour the discussion about a gendered labour mar- ket seems to bring up more and more topics with numerous facets. This reflects the high proportion of societal interest in issues like inequality, child care, and welfare state just to mention a few. Women hold a growing number of all jobs in economy while in most nations they are still responsible for unpaid work at home. Still this picture coexists with important differences in gender relations between countries and in labour market positions and experience of women within countries (Fagan and Rubery 1999).
Aspects of women’s labour are shaped on a national level by two broad catego- ries of institution: the welfare state and institutions that more directly organise the labour market. Both regulate the social and economic conditions for women to en- ter the labour market. (Fagan and Rubery 1999)
Beside the named institutional determinates of women’s labour market partici- pation, there are cultural determinantes that also affect women’s attainment in the labour market directly and indirectly through norm that vary cross-nationally. This study seeks to analyse attitudes towards the family and gender roles to thereby grasp how and to what extend they are related to women’s labour market participa- tion.
2. The Influence of Culture on Women’s Labour Market Attainment
Theoretically there are several approaches that seek to explain women’s employ- ment patterns. Clearly, characteristics of male and female labour market participa- tion differ considerably. Although there has been a general change in women’s education and employment patterns to the extend that equality on the labour market is not only desired but to some extend already accomplished so that women could (theoretically) have the same premises to work than men, different types of female preferences exist depending on the central priority of family and employment (Ha- kim 2002)
Considering several feminist approaches on welfare state policies, welfares states shape women’s employment patterns to the respect that they give or withhold incentives to paid, unpaid work or a combination of both. Those regulating mecha- nisms refer to explicit and implicit norms of institutions that constantly adjust the expectations and behaviour of individuals to the general logit of societal guidelines (Mósesdóttir 1995). Similarly to the typology of Siaroff (1994) who connected “female work desirability” to the religion of a country, this indicates at least that cultural determinants like norms, values, and attitudes, shape the central priority of women, be it directly through role models or indirectly through institutions. At this point it has to be noted that although the focus in this article is on culture, to predict women’s labour market participation, neither culture nor institutional factors can explain the differences between countries on its own. Especially, when focussing institutional factors like policies, conclusions can conform correlations with women’s employment patterns, but the causal order is particularly ambiguous (Matysiak and Steinmetz 2006; Pfau-Effinger 2000; Abrahamson 1999)).
The following section gives a short overview of gendered and to a certain ex- tend “cultured” welfare typologies which are based on one aspect or another on the feminist critic on Esping-Andersen’s (1990) Three Worlds. After establishing the awareness for the culture sensibility of welfare state policies, an approach that is explicitly based on culture will be used as background for further analysis:. Pfau- Effinger (1996; 2000; 2005) presented several gender culture models to predict women’s labour market participation on the basis of gender arrangements. Those are used to retrace the way of three distinctive welfare states (Finland, Germany, and The Netherlands), reassess their current position (gender culture model) on the basis of an descriptive analysis and the importance of each factor for likelihood of women to participate on the labour market.
The presentation of preliminary typologies that used the culture component to a certain extend has to stay very brief and should sensitise for the meaning of the norms, values, and attitudes in the context of welfare states and women’s labour market attainment.
2.1.1 Preliminary theories including culture
As mentioned above, institutions and welfare state policies adapt to the norms and values of individuals. According to Mósesdóttir (1995) this constitutes in the social construction of women as workers, housewives, and / or mothers. This leads respectively to a certain kind of support of the welfare state that permits women to fulfil their role. The welfare state regulates the relation between the sexes and en- ables or hinders women’s entrance to the labour market. By using “the consolida- tion of the mode of regulation as well as the norms and institutional arrangements concerning paid and unpaid work” (Mósesdóttir 1995: 633) as dimensions, Móses- dóttir identifies three variations of regulation mechanisms that lead to the estab- lishment of gender relations (liberal (USA), ecclesiastical (Germany), and egalitar- ian (Sweden)) (Mósesdóttir 1995).
Similarly, Alan Siaraoff (1994) includes the culture component into his typol- ogy of welfare states by assigning special importance to the “work-welfare-choice” (Esping-Andersen 1990). This concept includes determinants to identify the link- ages between labour and social service as well as its amount and recipient. In re- spect to culture it is important to recognize that labour market participation of women can be statistically linked to religiosity. Furthermore female work desirabil- ity is connected with the recipient of social benefits. For example catholic nations that solely pay social benefits to fathers have the least incentives for women to work. Creating two indices, Siaroff establishes a quartering of the analysed OECD countries into Protestant Social Democratic welfare states (Denmark, Finland, Norway, and Sweden), Protestant Liberal welfare states (Australia, Canada, New Zealand, the United Kingdom, and the United States), advanced Christian Democ- ratic welfare states (Austria, Belgium, France, (West) Germany, Luxembourg, and the Netherlands), and late female mobilization welfare states (Greece, Ireland, Italy, Japan, Portugal, Spain, and Switzerland) (Siaroff 1994).
The mappings of the several countries by Mósesdóttir and Siaroff may (if Mósesdóttir did not only use ideal types for her classification) correlate to a large extend since they both use criteria concerning social citizenship basing on culture.
Social citizenship was also faced by many feminist theories that show that full integration of women into the labour market meaning that the full provision of child care is the way to full citizenship of women and therefore gender equality. The thought that gender equality depends on female labour market attainment though is denied because this would mean an unacceptable androcentrism and follows the strategy of “equality in difference” (Meulders und O’Dorchai 2003: 186). This would mean that traditional home and caring activities of women are being en- hanced and used as basis of civil rights and political standing. Social citizenship as well as the respective social praxis is defined and encouraged in different ways in different welfare states. Facing this feminist discussion often denies the role of cul- tural values and ideals which are required for an accurate explanation. Pfau- Effinger (1996) however includes these different cultural attitudes in her compara- tive analysis and classification of West European gender culture models.
2.1.2 Pfau-Effinger’s (1996, 2000, 2005) Gender Culture Models
Social integration and exclusion of women into the labour market was often discussed in the context of welfare state policy, social inequality, and social citizen- ship. According to Pfau-Effinger (2000) the way of conceptualizing social citizen- ship is in part a more one-sided normative than analytical approach. Furthermore it is often based on the implicit notion of coherence of welfare state policy and labour force behaviour of women, the cultural dimension is not integrated into the analysis systematically, and social change is often not conceptualised. Altogether the nature of welfare state policy and its definition of social citizenship of workers and carers should be incorporated into the respective social context and considered as theoreti- cal frame for cross-national analysis (Pfau-Effinger 2000). For this purpose Pfau- Effinger (1996) developed a theory to explain international differences of women’s labour market behaviour and to classify societies in the face of gender specific divi- sion of labour which systematically accounts for the importance of socio-cultural conditions. In this regard Tenbruck (1992) refers to the concept of representative culture. In this sense culture are ideas, meanings and values which are commonly seen as valid construction of the world and provide the necessary frame of coexis- tence for social action (Tenbruck 1992). In this sense labour market participation of women results from the interplay of cultural models and institutional constraints in the labour market, family, and welfare state (Pfau-Effinger 1996).
Pfau-Effinger’s (1996) analysis is based on the interrelated fundamental terms gender culture, gender order, and gender arrangements. The term g ender culture on the one hand refers to consistent values and guidelines which exist in modern societies in reference to gender relations and forms of gender specific division of labour. Those are fixed in the institutional system and therefore stable. Gender order on the other hand are actual detectable structures of the relation of sexes and the relations between different social institutions in respect to gender specific divi- sion of work. Gender arrangements surround the gender culture and gender order and result from social bargaining processes. These bargaining processes provide , according to Pfau-Effinger (1996), a central basis that particular cultural guidelines dominate socially and for the principles of arrangement of social institutions in re- spect to gender relations.
Discrepancies between cultural guidelines and real structures of gender rela- tions within the scope of the gender order may be caused by non-simultaneous change in the development of gender culture and order.Therefore the normative validity of gender culture guidelines may vary to a certain extend. Such asyncronic- ities and disruptions are the origin of change in dominating guidelines of gender relations (Pfau-Effinger 1996)
The profile of the respective gender arrangement can be described by the domi- nant gender culture model. Those base on (1) cultural ideals of the gender specific division of labour, the central sphere of work for man and women, social valuation of those spheres and the way dependency between man and women is constructed, and (2) the cultural construction of childhood, motherhood and fatherhood (Pfau- Effinger 2000) (see also appendix 1)
Based on those criteria Pfau-Effinger (1996; 2000; 2005) classifies more and more differentiated gender culture models. In her work on “Changing Welfare States and Labour Markets in the Context of European Gender Arrangement” (2000) she identifies six models for West-Europe: the (1) family economic gender model, the (2) male breadwinner/ female home carer model, the (3) male breadwin- ner/ female part-time carer model, the (4) dual breadwinner/ state carer model, the (5) dual breadwinner/ dual carer model und the (6) dual earner/ marketized female carer model. Pfau-Effinger admits that there also exist hybrids which may most often develop through on-going change (Pfau-Effinger 2000)
In contrast to earlier classifications, culture-gender-models allow for analysing change, meaning that gender arrangements in different countries modernize. By including discrepancies between cultural guidelines and social acting three ways of modernization can be identified (Pfau-Effinger 2000). Those especially influenced and influence women’s labour market participation. Firstly there is the development of Germany and the Netherlands. Starting point in those nations was respectively the traditional male bread-winner/ female carer model. In (West-) Germany it mod- ernized towards a male bread-winner/ female part-time carer model as cultural guidelines and institutional factors (labour market policy, welfare state policy) changed, often also through the influence of the women’s movement. In the Nether- lands the development went even further to a dual bread-winner/ dual carer model. Secondly, there is a variation of the just described way of modernization which is performed by countries like Great Britain (and the USA). There, intervention by the welfare state is seen not that much as solution for labour market exclusion than in other countries. Starting point of the development was a male bread-winner/ female carer model that developed towards a male bread-winner/ female part-time carer model. Furthermore because of the idea of full integration of both sexes into the labour market, there was an additional impulse towards a male bread-winner/ fe- male marketized carer model. This causes many contradictions on the cultural level, in particular in respect to the social construction of childhood and motherhood. Therefore the extend to which the combination of paid work and childbearing by women has changed differs considerably in-between social classes. The third way of modernisation started with a family economy model and ended in a dual bread- winner/ state carer model like it is in Finland. The welfare state was of great impor- tance for this development, particularly since it covers the child-rearing and there- fore enables the full and equal integration of both sexes into the labour market (Pfau-Effinger 2000) In her later work “Welfare state policies and the development of care arrangements” (2005) Pfau-Effinger combines in respect to the focus on care arrangements the first and second development because the main difference is the amount of informal child care in the family. This differentiates the “social- democratic” welfare regime of the Nordic states from the remaining states of West Europe (Pfau-Effinger 2005).
The following section seeks to reassess the development of Germany, Finland and the Netherlands. These countries were chosen because Germany and Finland differ considerably in there women’s labour market participation level and attitudes toward social citizenship of women and the social construction of childcare. Their assignment in different typologies is mostly clear and opposed so that these two may describe the extremes of welfare states (leaving the liberals states with the least welfare state intervention behind) accurately. The Netherlands somehow find itself right in the middle of these two countries. An important point though to put the Netherlands in the focus of examination is that this welfare state shows “the greatest progress in restructuring family policy in the direction of individualized, egalitarian dual breadwinner models with partner shared childcare” (Pfau- Effinger 2005: 337).
2.2 Hypotheses, Methods and Data
Pfau-Effinger (1996; 2000; 2005) predicts a certain constellation of women’s la- bour market participation. The employment status of men and women in Germany, Finland and the Netherlands will be used to reassess the current point of develop- ment in the respective countries. The first hypotheses therefore is:
Actual labour market participation of women will differ cross-nationally according to Pfau-Effinger’s gender-culture-models in the respective coun- tries. (Hypothesis 1)
In particular this means that, since the predominant gender-cultural model in Ger- many is according to Pfau-Effinger a male bread-winner/female part-time carer model the labour market participation of men and women will differ considerably and women will in particular work mainly part-time.. According to Appendix 1 and the respective argumentation the pictures in Finland labour market participation of men and women will be more equal. Both sexes will be fully integrated into the labour market. In the Netherlands the picture will according to the dual breadwin- ner/ dual carer model be a relative equal participation of men and women in paid labour as well as housework. Since both parents take care of their children and the main sphere of rearing children is the private household the working hours of men and women will be less than full-time paid work in a traditional male bread-winner model. This “spare” time is then usable for child care. (see appendix 1) Furthermore according to Pfau-Effinger (1996, 2000, 2005) gender-culture-models are based on social bargaining processes which lead to dominating gender arrange- ments. These should be reflected in predominating norms, social values and atti- tudes in the respective society. Therefore:
Since cultural guidelines toward the central spheres of men’s and women’s labour and the relation between them as well as the central sphere of child- rearing develops differently in different cultural backgrounds, attitudes in regard to gender roles concerning paid and unpaid labour and the social construction of childhood and family will differ in-between the countries and in correspondence to Pfau-Effinger’s gender-culture models. (Hy- pothesis 2)
Obviously the named tendencies towards individuality affect the actions and be- haviours of persons.
Therefore such tendencies will also affect the labour market participation of respondents. Since the distribution of the employment status (full-time employed/ self-employed, part-time employed, household) differs according to the regime, the occurrence of attitudes will differ accordingly. Further- more, since men’s primary sphere (the labour system) is stable to a large extend, this hypothesis is tested on women’s activity rate and their distribu- tion between the different employment states. (hypothesis 3)
The named hypothesis will be tested on the integrated dataset of the European and World Values Studies. It consists of four waves between 1981 and 2004 and con- tains several countries including Germany, Finland and the Netherlands which are of special interest to this study. To investigate the latest position of the countries the latest available data was used. This is for Finland data surveyed in 2000 and for Germany and the Netherlands in 1999.
To test (see appendix 2 for syntax and respective output) hypothesis 1 and 2 a descriptive analyses of respective variables is used. If labour market participation of women is highest in Finland and lowest in Germany hypothesis 1 will be consid- ered true. To gain a clearly arranged picture a factor analysis is used. Using the elbow criteria and for reasons regarding content a two-factorial solution was used. Furthermore, to be able to compare the cultural guidelines of Pfau-Effinger’s cul- ture-gender-models to the distribution of attitudes in the dataset, the items were assigned to the guidelines as regards content. If variables which are considered to represent attitudes toward gender roles and the social construction of childhood and family are most traditional in Germany and least traditional in Finland, hypothesis 2 will be considered true.
The influences on women’s labour market participation (hypothesis 3) are esti- mated by Cross-tabulation of factors and woman’s labour market participation. If highest women’s labour market participation coexists with the highest affirmation of equal gender roles in the labour market and the household (measured by factors), hypothesis 3 will be considered true. To test the settings more accurate for the Netherlands, where a dual earner/ dual carer model is expected, part-time employ- ment will also be considered in these circumstances. The countries are tested sepa- rately since influences of coefficients possibly differ cross-nationally (see appendix 2).
3.1 Employment status
Before interpreting the employment status of men and women in the different coun- tries, it is important to realise that the employment status does not necessarily cor- respond to the percentage of economically active population which shall character- ise the employed labour force in this study. The following activity rate is used to compare the status indicated by respondents to the amount of economically active population determined by the International Labour Organisation (ILO). This com- parison will give information about if the two indicators measure the same. The activity rate used indicates the proportion of the whole number of the population (working-age) that work for gain. (Encyclopedia Britannica Online). Appendix 3 contrasts the two indicators using the same method. Obviously the percentage of respondents indicating they are employed (whether full-time, part-time or self- employed) corresponds to the activity rate issued by the ILO (see also appendix 2; output 4). More interesting in the context of this study would however be the per- centage of people that would like to work, although they may not be employed. Since this study is based on attitudes which may not reflect the real circumstances but desired ones, attitudes towards the desired employment status would be more suitable in this context when relating it to attitudes towards gender models. This may be particularly true for women, since institutional barriers may on the one hand distract them from performing their desired labour or on the other hand force them to work although they might desire to be housewife. This points towards the ques- tion if there is a true “work-welfare choice” (Siaroff 1994: 95) in the respective country. To go into this matter would admittedly go beyond the scope of this study but should be kept in the back of one’s mind for the understanding of the following analysis.
The employment status indicated by respondents in the EVS distinguishes be- tween several states. In the context of this study it is suitable to concentrate on full- time (here merged with self-employed) and part-time employment in contrast to the status as housewife or –men. In contrast to the initiatively considerations at the be- ginning of this section, employment status now is stated as the percentage of all respondents in the respective country.
As may be expected, respondent’s employment status differs considerable in- between countries. In all countries the difference between male and female full- and part-time employment, as well the status as housewife/ -men is high. For Finland we find a high percentage of full-time employed men and women. Furthermore the relations of full-time, part-time employment and housework between men and women in Finland is by far not assimilable to the relations in Germany or even the Netherlands.
Abbildung in dieser Leseprobe nicht enthalten
Women in Finland do by about 6% more housework and by about 4% work more part-time then men. The per- centage of full-time employed women in Finland is closest to men’s. In contrast full-time labour market participation of women in Germany does not reach the one of men by 20 percent points and women do more than 10% more housework or work part-time, which men pretty much do not do at all. One might argue that the percentage of women’s labour market par- ticipation in the Netherlands is to a large regard the same than in Germany, but you also have to consider women’s part-time employment. When full- and part-time employment is summed-up, female labour market participation in the Netherlands exceeds the percentage of working German and Finish women. In fact the break- down of the status of Netherlands’ women show the most equal one: The percent- ages of full-time working women, part-time working women and housewives is roughly the same (see appendix 2; output 1).
The contingency coefficient indicates a variation from indifference (K’=0,452) (see appendix 2; output 1). To be sure about the correlation, it was also separately tested for men and women as well as for each country. For the Netherlands respon- dent’s sex is most influential (0,487) on the respective employment status. The in- fluence is less in Germany (0,333) and least in Finland (0,218) (see appendix 2; output 2). When looking at respondent’s nationality results show stronger influence of nationality for women (0,342) than for men (0,265) (see appendix 2; output 3). To sum this up the influence of sex on employment status exceeds the influence of nationality only in the Netherlands for women. For men, where the influence of nationality is less than for women, the influence of sex in Germany also outreaches nationality. Only in Finland the sex component has less influence than nationality for men and women.
To a large extend this suits the main findings of Pfau-Effinger and therefore hypothesis 1, although a few points do not fully fit to the according breadwinner models. For Germany a female part-time carer or worker is not found without res- ervations: The percentage of women working part-time is less than the half of women working full-time and women doing housework are even more than women working part-time. A similar picture appears for the Netherlands: Men’s labour market participation by far outreaches women’s participation rate. This point of view does not justify naming the participation rates equal. There is however, like before, the possibility of summing-up part-and full-time labour for women, which draws nearer equality. Also men in the Netherlands do work more part-time and do a little more housework than men in Finland or Germany do. Lastly, Finland’s women are by far most integrated into the labour market when speaking of full-time employment, but speaking of full and equal integration of both sexes is exagger- ated.
 For detailed explanation see appendix 1 and the further explanation of modernisation and hypothesis 1.
 Working age means for Finland the population from 15 to 74 years and for Germany and the Netherlands the population aged 15+.