Sally Bould¹

Claire Gavray²

¹CEPS/INSTEAD, Luxembourg/University of Delaware

²Faculty of Psychology and Education, University of Liège

Abstract

Understanding the meaning of a woman’s economic activity requires an in depth understanding of the labour force concept and how it has been applied to women. This paper will examine in detail some of the issues and problems with the data available and its usage as well as the problem of cross national comparability. These issues will be explored by a close examination of the International Labour Organization’s (ILO) recommendations and definitions and yearly statistics (http://laborsta.ilo.org/). The first part of the paper will focus on the development of the labour force concept and a history of how women’s labour force participation has changed in the last 30 years using the examples of Bangladesh and Thailand. A different historical picture is found in the four European countries examined: Portugal, Spain, Belgium and France. The second part of the paper will examine factors which need to be taken into account in forming a broader understanding of women’s work in these European countries. Age, education, working hours, wages, and motherhood are examined. In view of this analysis the paper will critique the European Union goal of 60% of women working in EU countries by 2010.

Keywords women’s work, employment, labour force participation. EU employment goals

Resumo

O Trabalho das mulheres: Mensuração e Significado

Compreender o significado da actividade económica das mulheres requer a análise aprofundada do conceito de população activa e de como este tem sido aplicado às mulheres. Este artigo examina em pormenor algumas das problemáticas e limitações associadas ao uso dos dados estatísticos disponíveis e à comparabilidade entre países. Esta temática é explorada a partir de uma análise focada nas recomendações e definições da Organização Internacional do Trabalho (OIT), assim como das estatísticas anuais (http://laborsta.ilo.org/). A primeira parte do artigo centra-se no desenvolvimento do conceito de população activa e no modo como a participação laboral das mulheres se tem alterado no decurso dos últimos 30 anos, recorrendo aos casos do Bangladeche e da Tailândia. Uma perspectiva histórica diferente pode ser encontrada nos quatro países Europeus analisados: Portugal, Espanha, Bélgica e França. A segunda parte é dedicada aos factores que importa considerar de modo a que se alcance uma compreensão mais ampla do trabalho das mulheres naqueles países Europeus. Idade, educação, horas de trabalho, salários e maternidade são factores aqui relevados. O objectivo da União Europeia de alcançar uma taxa de emprego feminino de 60% nos países Europeus, até 2010, é também objecto de crítica.

Palavras-chave trabalho feminino, emprego, população activa, objectivos de emprego da UE

Résumé

La mesure de l’emploi des femmes: vers une compréhension des enjeux.

Comprendre l’activité économique des femmes demande de bien connaître le concept de force de travail ainsi que la façon dont il a été utilisé à propos des femmes. Cet article examinera en détail la disponibilité de données en cette matière, leur valeur et leur usage; il abordera les problèmes liés aux comparaisons internationales. Nous explorerons cette question au départ des recommandations, définitions et statistiques de l’Organisation Internationale du Travail (OIT) (http://www.ilo.org/global/lang--fr/index.htm). La première partie du document rendra compte de l’historique du concept de force de travail. Il retracera l’histoire de la participation des femmes à l’emploi et montrera, au départ des exemples du Bangladesh et de la Taïlande, l’ampleur des changements opérés sur les trente dernières années. C’est un tableau historique différent qui ressort pour les quatre pays concernés par cet article: le Portugal, l’Espagne, la Belgique et la France. La seconde partie du document examinera certains facteurs dont il est essentiel de tenir compte pour mieux comprendre l’emploi féminin dans ces pays européens: l’âge, le niveau de diplôme, les heures de travail prestées, le niveau du salaire et les maternités. Cette analyse débouchera sur un avis critique concernant l’objectif fixé par l’Union Européenne d’arriver à un taux d’emploi féminin de 60% d’ici 2010.

Mots-clés emploi, emploi des femmes, participation à la force de travail, objectifs européens en matière d’emploi

Introduction to the definitions...

Discussions of women working are now part of the landscape, not only of women’s studies, but of politics and policy today. In both the developed world of Western Europe and the developing world political discussions now emphasize the importance of women working in order to prevent poverty or to stimulate economic growth. This was not the case in the past where women’s work was invisible at least to economists and policy makers from the West (Rogers, 1980; Bould, 1984; Amott and Matthaei, 1996). The fact that women’s work is now visible is a sign of definite progress but before any other positive or negative conclusion can be reached it is first important to understand exactly how women’s economic activity is defined today both in the international and national contexts. This paper will examine closely the exact meaning, as well as the problematic of this definition. Future research needs to move beyond this measure of economic activity in order to develop a more comprehensive picture. This problem is reflected in the current European Union goal of increasing women’s employment rate to 60%.

The history of women’s reported economic activity will be reviewed for two developing countries, Bangladesh and Thailand and four developed western European countries: Portugal, Spain, Belgium and France. These countries were chosen to illustrate a diversity of cultures and history in the four developed countries and the two developing countries. The overall goal is not to provide the most recent data, nor to provide a broad overview of women’s labour force participation in many countries; these are available elsewhere. Rather this is a cautionary story of how the concepts of labour force participation and employment have been developed and measured with special attention to their limitations. Each country presents a unique history with different approaches to asking the relevant questions, reporting the results, and interpreting the trends in the context of the changing cultural expectations for women’s work force participation. Recent efforts of the European Union to harmonize these data are an important effort in enhancing cross national comparisons but even here the issues of women’s work force participation are very complex and often country specific.

History of the measurement of women’s economic activity

Up until the mid 20th century economic activity was measured by the concept of «gainful worker,» asking the person to identify his or her usual occupation. This measure was developed in the censuses of the West during the period of rapid industrialization when the modern breadwinner-homemaker ideal was emerging. Since this «ideal family type» defined the woman, especially the married woman or mother, as not working, the work she did was often not recognized or even denied. She would not report her economic activity as her usual occupation since she was defined by her family status. The status of a woman was derived from the status of her father or husband. Her economic activity would remain invisible, often even to her (Gubin, 2000; Dejours, 2003). Reinforcing this was that she was typically dependent on her husband/or her inheritance from her father for the bulk of her income. In addition, her husband’s status often depended upon his claim that he was the sole support of the family and that he had no need for his wife’s help in this task. The earliest forms of public assistance were provided to women whose husbands were killed in war so that they might not have to go to work (Gordon, 1994).

The husband’s role as sole support of the family, however, has had only a very brief history in the West (Bernard, 1981). Prior to the industrial revolution both the husband and wife were working, typically in agriculture. For the man it was important to have a wife who could work and contribute to the farm activity (e.g. Arensberg, 1937). But this picture of women’s essential work in agriculture was overlooked in the first measure of economic activity. The man reported his usual occupation as «farmer», but the woman would report herself as the farmer’s wife, even if she worked for long days in the farming business. It should be noted that all economic activity recorded in the data is self report, or the report of a close family member. This means that she is likely to report behavior which is consistent with the cultural expectations of her community. If a farmer’s wife does not report her usual occupation as farming, then her work will be invisible in this data. In some developing countries her current labour force participation may in fact be reported by another member of the household.

There is an excellent example of this in the southern states of the United States in the census of 1900. At this time more than 50% of the labour force was in agriculture. White women reported themselves as housewives, consistent with the cultural expectations of married women. In the African American community, however, women’s work was valued and recognized (Amott and Matthaei, 1996). Consequently African American women reported themselves as having a usual occupation in agriculture. But a close examination of these communities indicated that they were both poor and engaged primarily in agricultural work. In this situation it can be assumed that the white women also worked in the farms and the fields because poor families need the labour of all members. Every able bodied adult is economically active in poor families, but the labour of white women was made invisible by the cultural expectations (Brainerd and Miller, 1957; Lebergott, 1966).

The development of the labour force concept was the result of criticisms of the measure of «usual occupation» or «gainful worker». Instead of a vague question like «what is your occupation?», the questions have been standardized by the International Labour Organization (ILO) in order to more accurately distinguish those who are in the labour force from those who are not. Those who are in the labour force include those who report that they are working or employed in the reference week and those who report that they have been looking for work during a specific reference period. The labour force then, consists of the employed plus the unemployed. All others are defined as out of the labour force. A labour force participation rate is calculated by dividing the sum of the employed and the unemployed persons by the total population. Usually there is an age restriction whereby this rate is calculated only for all persons over age 15. A common reason for non participation for younger people is schooling and for older people, retirement. In addition there are those who are disabled and, of course, those who are «housewives». An important group of non participants are the discouraged workers who are discussed below. The term «employment rate» is also in widespread use and this rate is calculated by dividing the number of people who report that they are employed at least one hour during the reference week by the total population. In this case those persons who are unemployed are included only in the denominator, not the numerator. Usually the words «employed» and «working» are interchangeable and the employed also include the self-employed as well as those with a job but not at work during the reference week. Sometimes the word «working» is used to mean all of those in the labour force, both those employed and unemployed but this is not standard usage.

This new concept was developed not to make women’s work visible but to make men’s unemployment visible. The problem of measuring unemployment was first raised at the Second International Conference of Labour Statisticians in 1925 (www.ilo.org/global/). Nevertheless, the development of this concept opened the door to the possibility of measuring women’s work in agriculture by specifying activities which counted as «economic» and including a measure of an unpaid family member’s work. Women did not need to claim that they were farmers in order to have their economic activity recognized; they could simply be the farmer’s wife who helped her husband on the farm at least 15 hours a week. In this case the labour force concept set the stage for a more accurate reporting of the actual work activity of rural women, especially in agricultural work. It also improved the reporting of urban women who worked part time or earned money with a home-based business such as providing rooms or laundry service (cf. Amott and Matthaei, 1996). If there is no family business, however, her unpaid work is still invisible in national accounts. An urban woman would often raise chickens for her family’s consumption but this was not «work» because she did not receive money for her activity.

The labour force concept, however, introduced a new bias for making women’s economic activity invisible. In the new effort to define economic activity as including the unemployed as well as the employed, it was generally assumed that a woman, especially a married woman would not be looking for work. Thus the labour force concept made men’s unemployment more visible, but a woman’s unemployment would very likely remain invisible. And even where women were actually questioned as to whether or not they were looking for work, this measure does not take into account the woman (or man) who would like a job but who does not believe that she or he could find work; these individuals are classified as a «discouraged worker» (discussed below). In labour markets where women are discriminated against, their unemployment is even more likely to be invisible.

The ILO measure of «working» or did you earn a few Euros last week?

Employed or working, the major component of those who are defined as in the labour force, is a simple in or out measure. And in terms of the definition of employment it is minimalist. Doing something for pay only one hour in the week before the interview qualifies a person to be considered as working or employed. This definition has been criticized but so far no changes have been made in the data collection process. In comparisons between the percentage of women working and the percentage of men working it is important to understand that some women may be working for only a few hours last week. Therefore any comparisons between men’s and women’s labour force participation rates or men’s and women’s employment rates must be made with this understanding. In addition, caution is necessary in interpreting cross-national employment rates for women because of the wide possible variation in the number of hours worked by women in different countries. In the EU only the Netherlands has moved to an internal definition of «employment» requiring that the person be employed for at least 12 hours a week (Hakim, 2000: 8).

Employment is defined, then, as persons who worked for a wage or salary at least one hour during the week identified as the reference period plus those «with a job but not at work» or persons in self employment or with an enterprise or an unpaid family worker who worked a minimum number of hours in a family enterprise. Current guidelines have become more inclusive for the unpaid family worker so that she is considered «as self-employed irrespective of the number of hours worked during the reference period.» But many countries still retain a minimum hours of work, usually 15 as an unpaid family worker in the family business, in order for her to qualify as working.

Cultural expectations in developing countries: Bangladesh and Thailand

In Bangladesh, women’s economic activity was dismissed by the colonial power, in this case the United Kingdom. The colonial rulers believed in the homemaker breadwinner family ideal which prevented them from seeing or valuing the labour of Bangladeshi women. This continued after independence with development organizations trying to impose the homemaker breadwinner family upon the Third World. W.W. Rostow’s influential book, The Stages of Economic Growth (1961: 91) only referred to women in a parenthetical remark about their full time duties in child care. A women’s project in Bangladesh in the 1970s taught Bangladeshi women to make proper wedding cakes (Kromberg and Carr cited in Rogers, 1980: 92). Given this imposed Western cultural definition of women as homemakers and caregivers, it is not surprising that, even with the possibility of reporting herself as an «unpaid family worker» Bangladeshi women’s work remained invisible. In 1974 the reported labour force participation for women age 15 and over was only 3.4% (http://laborsta.ilo.org/).

In the last 3 decades women working in international development and women scholars have pushed hard to make women’s work visible (Boserup, 1970; United Nations 1999, United Nations 2005). One of the results of these efforts is that there has been a dramatic increase in the reporting of women’s economic activity. In 2000 the labour force participation rate in Bangladesh was 55.9% (http://laborsta.ilo.org). The increase of women’s paid employment in factory work with globalization has also lead to higher levels of reporting. Women who have a paid job in the formal economy are more likely to report themselves as working.

Before concluding that the increase in reported labour force participation by women is entirely positive, it is important to examine the situation for those under age 20. Each country sets a minimum age for economic activity to be counted. Of course it is true that child labour is still important in many developing countries, including Bangladesh, often a rural child will spend at least 15 hours a week in the family agricultural enterprise. The ILO, however, focuses on persons 15 and over as the potentially economically active population. But many countries still report labour force participation at younger ages. This is the case in Bangladesh where the workforce participation of those aged 10-14 is also reported (http://laborsta.ilo.org/). Here the picture for girls and boys is disturbing. Thirty-three percent of girls in these ages are reported as «working» and 38.9% of boys. Early marriage limits a girls education, but not her labour force activity (Ahmed and Bould, 2004).

It is important to remember that the situation for women may be very different in countries with different cultures and histories. While the reported rate of women’s participation was extremely low in 1974 (3.4%) in Bangladesh, the rate of women’s reported participation in nearby Thailand in 1970 was very high (73.4%) (http://laborsta.ilo.org). The high participation of Thai women in 1970 reflects the cultural recognition of women’s work in agriculture. In Thailand studies point to the fact that women’s work was valued in the traditional culture. Women are often seen as those who are better at managing the money so, in fact, the women may control the family business. This traditional recognition of the value of women’s work was never distorted by colonial rulers whose vision was limited to the breadwinner-homemaker family. Development programs in the 1980s included Thai women because their skills were recognized. And often the woman was the primary participant as long as the activity was compatible with childcare. (e.g. Rao, Anderson and Overholt, 1991).

In Thailand women’s labour force participation fell from 73.4% in 1970 to 64.9% in 2000. This fall in participation, however, cannot be interpreted negatively. A careful examination of labour force participation by age group indicates that the lower women’s labour force participation in 2000 is the result of fewer girls aged 15-19 working. This could be best explained by the increase in school enrollment for girls, which suggests that the fall in labour force participation reflects a positive and not a negative change.

There are many different cultural expectations and values attached to women’s labour in the developing world so researchers need to be cautious in making any broad conclusion. In cultures where daughters are under the control of their fathers, the unmarried daughter who works in a factory may receive limited respect and be able to retain only a small proportion of her wages (Salaff, 1981; Wolf, 1992). For married women, her own wages are likely to give her more respect and control over her life than her work in her husband’s or father-inlaw’s business. A married woman may be a slave labourer in her father-in-law’s business (Hsiung, 1996). Globalization, in providing jobs for women in the formal sector, has also resulted in more attention to women’s labour although the result can be that women gain wages in dangerous and inhuman working conditions (Rosen, 2002). In general there is no way to make any conclusions about the status or control over her life that working provides a woman by the use of the ILO labour force data. Cultures vary enormously and the limited self-report of labour force statistics only say what work is visible, but not its problematic issues like poor working conditions. It is also important to be aware of the different situations of women in the same country. For example, educated women in Bangladesh have opportunities for professional jobs; opportunities brought about in part because of the efforts to mainstream gender by the United Nations and other international organizations (Bould, 2008). Their situation is quite different from that of the poorly educated factory workers. The labour force statistics only provide an overview, an average. Understanding what working means to a group of women requires in depth interviewing.

The discouraged woman worker

In depth interviewing can also identify discouraged women workers. Discouraged workers, according to the ILO, are workers who would «like to work and are available to do so, but do not look actively for work for various reasons.» Of course men and women look for work primarily when they believe that they may be able to find work. Furthermore this measure relies on self-report in looking for work. Sustaining long term unemployment is difficult psychologically. Men, however, are more likely to continue to report that they are looking for work because that is considered their role; they usually have no other role. This is not true for older men, however, who with the experience of long term unemployment will report themselves «retired» rather than unemployed in order to save face (Bould, 1986). Married women still have the option of claiming the role of housewife when they believe that they are not likely to find work. So in situations of high unemployment the labour force shrinks especially because women simply give up looking for a job (cf. Uchitelle, 2008). A woman’s role as housewife gives her that option.

Western European Industrial Societies

While the advancement of women is part of the Western European image, it is important to remember that in much of Western Continental Europe until the second half of the 20th century a married woman’s wages were typically her husband’s property. She had no independent right to her earnings and was subject to her husband’s authority (Bould and Schmaus, 2008). Her earnings could give her influence as well as provide for her in the rare case of divorce. Nevertheless, if her earnings reflected her husband’s inability to support the family then her earnings brought shame upon him. Under the traditional breadwinner-homemaker family model a working wife represented the failure of the husband at breadwinning. It was the woman with the poorer husband, or the disabled husband or the widow who was most likely to work (Voldman and Schweitzer, 2002). If she did earn an income by taking in lodgers or laundry her husband would probably not see that as working, or as only working for pin money.

Under the labour force concept one reason that a woman’s work would not be reported is the lack of recognition of her work by the community or by her husband. Another reason could be the existence of an irregular economy. The irregular economy includes informal work which is paid «off the books» (Ferman and Ferman, 1973). The transactions are in cash and not reported to tax authorities; there is no record for tax collection. Those who receive the cash may feel entitled to keep it all without paying taxes. In the context of a mistrust of government, persons receiving such cash would not report it to any government authority, even to the census or to labour force surveys. More recently efforts to keep informal cash earnings hidden from government workers are happening in places like Bulgaria. While women’s reported labour force activity has dropped dramatically since 1989 women are most often engaged in the informal cash economy to earn at least some money (Ghodsee, 2000). The government worker who asks about earnings is not likely to be trusted. Overall the informal economy presents a challenge for accurate data collection, but the challenges are greatest where the transactions are in cash. There has been some attempt by the ILO to identify the informal sector as a statistical category but even this definition excludes the illegal economy which typically involves cash in the drug trade.

The Iberian Peninsula: Portugal and Spain

In 1970 the labour force participation rate for Spanish women 16 and over was 17.9% (see Table 1). The peak of participation was during the ages 20-24 at 39.6%, typically in the years before motherhood. There was also a high participation rate (36.6%) among women 15-19; these young women were probably no longer attending school. But the very low rates for women above the age of thirty probably reflect the strong expectation that in 1970 women with children should not «work» (Hakim, 2003) as Spain was then a very poor country with a high proportion of the labour force still in agriculture, poor women were no doubt working on the farms, but they or their husbands did not report their activities as working. Their economic activity was not recognized. No doubt the very conservative climate during the Franco years made much of women’s work invisible (cf. Casaca and Damião, 2008).

Although in l970 Portugal reported a higher labour force participation rate for women (24.6%) than Spain (17.9) (See Table 1), this rate still represents a substantial under-reporting of women’s work in agriculture. The agricultural sector was quite large and poor; experience indicates that poor women do work in agriculture but in traditional western cultures they are discouraged from claiming this activity as work. In Portugal traditional gender roles were set in law where husbands could restrict their wives access to professions, read their wives mail and refuse to allow her to leave the country (Ferreira, 1998). Traditional attitudes concerning mothers and married women working persisted in Spain until the late 1980’s (Hakim, 2003: 59). In both countries older women who would be working on family farms in 1970 apparently under reported their efforts; of women age 45-49 only 20% in Portugal and 14% in Spain reported working (http://laborsta.ilo.org). The higher rates of overall participation in Portugal reflect the higher rates of participation among teenage girls. Even at younger ages 10-14 more than 10% of girls and 13.6% of boys are reported as working in Portugal; this activity was probably in agriculture in 1970. Data for boys and girls under 15 years of age, however, is not included in the calculation of labour force participation rates. Portuguese girls (and boys) under the age of 20 in 1970 had the highest level of labour force participation rates of any of the three European countries with data on 10-14 year olds, (http://laborsta.ilo.org; this data is not available for Spain). More recent data on child labour in Portugal indicate that this is still an important as well as a complex problem (Lopes and Goulart, 2003; Goulart and Bedi, 2005); this recent analysis relies on special surveys as the official labour force surveys in Portugal no longer cover persons less than 15 years of age.

North Western Europe: France and Belgium

All four of the European countries, Spain, Portugal, France and Belgium experienced dramatic increases in the labour force participation of women 15 and over in the last three decades of the 20th century (see Table 1). More and more wage work in the formal economy became available to women, and when a woman gets a pay check she is more likely to report herself as employed. Furthermore, paid work in the formal sector is generally subject to taxation by withholding tax. Thus there is no reason to hide the earnings from government workers.

Portugal and Spain have had the most dramatic increases in reported labour force participation where it more than doubled between 1970 and 2000. In 2000 the highest rates of women’s labour force participation are in Portugal at 52.7%. Part of this increase in both countries probably represents the under-reporting of women’s unpaid family labour in 1970 when a relatively high proportion of the workforce was in agriculture but traditional expectations for women made this work invisible. In 2000 the highest rates of women’s labour force participation are in Portugal at 52.7% and the lowest rates are in Spain (Table 1). Portuguese rates climbed rapidly in the late 1970s. Factors in the rapid increase in rates in Portugal include the rapid growth of the service sector and new rights for women in the civil code (Ferreira, 1998). In addition there was the shortage of male workers due to emigration and the low wages for men (Casaca and Damião, 2008: 6). Liberalization of gender roles and the development of the service sector were also factors in the rapid increase in Spain, but this occurred about a decade later that the rapid increase in Portugal.

This dramatic increase in labour force activity in all four of the countries represents only the minimum of working one hour in the reference week and with the possibility that women’s unemployment is underrepresented. In Spain and Portugal women have been much more likely to be working full time in comparison with France and Belgium. All four countries experienced increases in school attendance and decreases in labour force participation for teenagers, 15-19, but this trend was most dramatic in France and Belgium; the participation rate for girls was 8.4%in Belgium and 5.9% in France in 2000 (www.laborsta.ilo.org).

The data in Table 1 refer to the standard ILO measure of labour force participation age 15 and over. This measure provides only a limited perspective of the long term trends because overall in these countries life expectancy has also increased rapidly and with no changes in the retirement ages, a higher proportion of those out of the labour force will be the elderly. Since all four countries have experienced a rapid aging of their populations in the last 3 decades their relative position is about the same although the later years are much lower because they include those over 65.

Recent developments

The 2007 data indicate that the greatest recent increase in women’s labour force activity since 2000 has been in Spain although the Spanish rate is still the lowest of the four countries (See Table 1). This is not surprising given recent rapid economic development in Spain. Portugal has experienced a modest increase to 56.3% (see Table 1). Increases in Belgium and France have also been modest to 51.5 in France and 46.5 in Belgium (for recent detailed comparisons between France, Belgium and Canada see Gavray, 2008). Portugal still stands out as the country with the highest level of women’s labour force participation of the 4 European countries. In fact, women’s work force participation in Portugal in 2007 is greater than that in Sweden. Portugal, however, still has a relatively high activity rate for teenage girls 15-19 of 14% (http://laborsta.ilo.org). It is also important to remember that these numbers include all of the elderly who are typically out of the labour force. Rates of employment and participation increase when only those under 65 are included as in Table 2.

The EU has set a target goal of a 60% employment rate for women aged 15-64. This measure is calculated by restricting the numerator and the denominator to women aged 15 to 64 only and it ignores unemployment; it does not represent labour force activity as a whole (Gadrey, 2001). This employment goal for women has already been met by Portugal and France (see Table 2). But one of the problems with this measure of achievement is that Portugal has met this goal with high rates of employment in the 15-19 age group. It is not in the interests of young women to spend their late teens at work rather than at school. There are also indications that a similar situation exists for teenagers in Spain, but the Spanish labour force data is collected only from the age of 16, without the 15 year olds there is no way to accurately compare Portuguese teenagers to Spanish teenagers.

In order to understand the meaning of these employment rates it is important to break them down by age. For the young, up to age 24 there is the importance of school attendance. For older women it is often a question of opportunities such as age discrimination and gender discrimination as well as possible retirement benefits. For example, the picture of the high rates of women’s employment in Portugal is due, in part, to the high rates among older women workers; 43.7% of women age 55-64 are in the labour force as compared to 35.2% in France and 22.1% in Belgium. While Portuguese women are a positive example of a longer work-life encouraged by the European Union, they no doubt reflect the Portuguese women’s lack of access to adequate retirement pensions. In the case of Belgium, part of the reason for the lower employment rates overall is that not only are girls out of the labour force (92%) but also older women workers are least likely to be working. This type of cross-national differences in young women and in older women make the comparisons of women aged 15-64 incomplete. One way to avoid this problem is to examine women workers only in the prime working ages of 25-54. Table 2 shows that in the prime working ages women’s employment is above the target of 60% in all four countries and above 70% in each of the countries except Spain.

Education and labour force participation

Higher education provides for job opportunities that are not available to those with limited schooling. Under the breadwinner-housewife model prevalent in the 1970s however, a married woman was not supposed to work if her husband could earn a good living. Since educated women were generally married to educated men, the ideology of the breadwinner-housewife model depressed her potential labour force participation in spite of her high educational level. On the other hand, the less educated woman married to a less educated man often had to work in order to have a minimum family income. With the possible exception of (West) Germany (Blossfeld, Drobnic and Rohwer, 2001) this picture has changed dramatically since the 1970s and the emergence of the dual earner family with children.

Now the relationship between education and labour force participation for women is similar to that of men in Western Europe. Since young women in Europe typically spend more time in school today, they are less available for labour force activity when they are young. Table 3 examines the labour force participation of women beginning at age 25-64 by educational attainment in 2002. For both women and men in each of the four countries the higher the education the more likely the women or man is a labour force participant. Those with more opportunities are more likely to be working or looking for work. In addition, persons with more education are also likely to benefit from better health. Their labour force participation is less likely to be limited by health impairments. This is especially true at older ages. Persons with health impairments are more likely to experience difficulties in the labour market, and give up the search for work and drop out of the labour force as a discouraged worker. In addition the jobs available to women with low education are often heavy cleaning and caring which require heavy physical labour. These jobs are hard to combine with family life (Gavray, 2007). The opportunity costs of working for mothers with low educational levels are likely to be high relative to the wages that they can earn.

Spain and Portugal: Is there a Southern model?

Portugal stands out as the country with the highest labour force participation rate for women at every level of education in comparison to the other three countries in Table 3. Women’s rates with upper secondary education and tertiary education in Portugal are almost as high as men’s. Portugal does not fit in the Southern model which was identified using the Mediterranean countries of Spain, Italy and Greece (Trifiletti, 1999; Wall, 2008). The overall lower labour force participation rate in Spain as compared to Belgium reflects the significantly lower rates among older Spanish women. This suggests that in the future, the southern model in Spain will gradually fade as older Spanish women reach retirement age and younger Spanish women sustain their higher labour force participation rates as they age (cf. Hakim, 2003).

In the case of education the situation in the South is quite different from that in the North. Women are much less likely to have completed secondary school and more likely to have dropped out in Spain and Portugal than France or Belgium. In both Spain and Portugal there is a low level of educational attainment for both women and men together with a high rate of labour force participation in the teenage years. In the case of Portugal, if the level of education was increased so that fewer Portuguese women were in the lower education category and more were in the upper, the overall labour force participation rate for ages 25-65 would increase although the rate for teenage girls would decrease.

Women with low education have low employment rates and low labour force participation rates (Flahaut, 2006; Gavray, 2008). If the goal is to increase the employment of women in the future, then more attention needs to be paid to the level of education that young women are receiving in Spain and Portugal. For Belgium and France, however, there is a high level of attendance at the advanced secondary level, 84.9% and 85.0% respectively (Table 4). The drop out rate for girls 18-25 before the beginning of secondary school is also low, just over 12% in 2006 (www.insee.fr/fr/themes). For these latter countries it will be necessary to target programs specifically at this small population of poorly educated young women to encourage further education and/or provide targeted job training. In addition it is necessary to enhance the quality of the jobs that are available to women with low education.

In Spain and Portugal, however, improving women’s employment prospects will first require broad reforms of the educational system to encourage young women (and young men) to stay longer in school. The drop out rate in Spain in 2006 for girls was 29.9% and for Portugal it is estimated to be 39.2% (www.insee.fr/fr/themes). Nearly 40% of Portugal’s young women age 20-24 have not attained the advanced secondary level (Table 4). This is an area where greater attention to the schooling of girls (and boys) will be important in the future. Now the school leaving age in Portugal is 15.

Unemployment and education

For both men and women more education results in less unemployment. But there is a greater impact for women (www.oecd.org/dataoecd/). Comparing unemployment levels across nations is always problematic because the unemployment numbers involve policy and politics; each state uses an internal definition for its own purposes. Nevertheless, Eurostat does harmonize the data according to the ILO standard definition, «unemployed persons are all persons 15 to 74 years of age who were not employed during the reference week, had actively sought work during the past four weeks and were ready to begin working immediately or within two weeks» (http://epp.eurostat.ec.europa.eu). A country may use a different definition for internal reporting, such as a reference period of actively seeking work for only the past week, not the past 4 weeks. This more restricted definition will result in fewer persons reported to be unemployed as the shorter the reference period, the less likely the individual will have looked for work. Such «lower» unemployment rates are often popular with governments.

In three of these countries in 2006 there is a very high rate of unemployment for women with low education. 15.4% for Belgium, 13.5% for France and 14.1 % for Spain; the Portuguese rate is lower at 8.8 but still higher than the rate for women with more education (www.oecd.org/dataoecd/). And this is true even though sustaining a job search is more difficult for those with low education. Highly educated women and men will usually have a broad network of friends and acquaintances to provide a constant opportunity to keep up the search for work (Granovetter, 1995). In contrast, a woman with low education is likely to use up her contacts, primarily family and friends, in a week or so and come earlier to the conclusion that there are no jobs available and become a «discouraged worker.»

Employment and motherhood

This examination of motherhood and economic activity will be limited to women who are employed. A key factor in understanding women’s employment patterns for developed countries has been the impact of motherhood on employment. Traditionally, women did not work after becoming mothers unless they had a home based business. But different countries have different cultural expectations for mothers and whether mothers work or not has been a key factor in explaining different employment and labour force participation rates between countries at a similar level of economic development (Gils and Kraaykamp, 2008). This is still true today where a significant part of the higher rate of employment in Portugal is because Portuguese mothers work. Nevertheless while mothers are most likely to work in Portugal, they report the most conservative attitudes of the four countries on whether mothers of preschoolers should stay home. In 2002 38% of mothers in Portugal indicated this conservative attitude in contrast with only 26% in Belgium (Charles and Cech, Forthcoming: Table 1) but mothers of two or more children in both countries had similar employment rates (Table 5).

In spite of conservative attitudes noted above in Portugal (38%) and in France (37%) the employment data indicate that for both countries women aged 25-54 who have one child are as likely to be working as women who have no children (Table 5). And for Portugal women with 2 or more children are almost as likely to be working as women with no children. In Spain, however, motherhood still has a strong impact on behaviour although the attitudes are similar to those in France as 35% of Spanish women report a conservative view on preschool children (See Table 5). The fact that attitudes are not congruent with behaviour indicate that cultural expectations may differ from attitudes. Women in Portugal may still be expected to work especially if they have a high education and good opportunities.

Overall the impact of motherhood is moderated by education. This effect showed up in 1991, over a decade and a half ago (Table 6). Mothers aged 20-39 who have a University degree are more likely to be working than similar mothers with only secondary education. In Spain where the overall rate for mothers is quite low at 46.8% in l991 the employment rate for those with university degrees is much higher at 83.9%. In Portugal education, not motherhood, is the key factor in employment rates while in Belgium both motherhood and education have a strong impact. The fact that motherhood has more of an impact in Belgium than in Portugal is curious because the availability of subsidised child care is greater in Belgium. Clearly there are other factors which influence the employment rates for mothers in addition to economic structures and institutional arrangements (Pettit and Hook, 2005).

The number of children is a key factor in increasing the probability of parttime jobs in Belgium and France (Gavray, 2007). In Spain there is a very limited impact and in Portugal none at all, but both of these countries have very low rates of part-time work, under 17% in 2000 (Pissarides et al., 2005: Table 2.2). Countries with very low birth rates such as Spain and Portugal women would be expected to have higher rates of labour force participation than in countries like Belgium and France, where mothers are much more likely to have 2 children (Shorto, 2008). While this is true for Portugal it is not the case for Spain which has low birth rates and low labour force participation rates.

Gender equality in wages

Everywhere the men make a higher wage on average. The hourly wage differences for those who work 15 hours or more are least among young women age 25 to 34 and greatest among women 45-54. In 1998 the gender wage ratio is lowest in Portugal where women age 25-34 make only 86.8% of what men make (Pissarides et al., 2005: Table 5.2). This measure furthermore does not take into account women who work less than 15 hours a week who are likely to have the lowest hourly wage. While the differences may be quite small in some countries, this reflects only the wages last week or some brief reference period. In each country women are more likely than men to have a temporary and/or part-time job rather than a permanent full-time job. Although education benefits both men and women there is still no equality in earnings, employment conditions or careers between very educated men and women. Inequalities in the private sphere require women to be realistic and strategic and this limits their choices and engagements (Marry, 2006).

To better understand the inequality between men and women in the case of earnings it is important to compare the lifetime earnings of women to the lifetime earnings for men. This is where the most significant difference lies (Hartmann and Whittaker, 1998). And life time earnings are more and more linked to the level of public and private pensions so that those with low lifetime earnings are at risk of poverty during retirement. A more accurate meaning of income and earnings is found in Titmuss’ (1962) concept of control of resources over time. It is now time to begin examining the lifetime picture for men and women and to better estimate the cost for women of their unpaid labour in the private sphere i.e. caregiving. Hourly wages, then, are a very poor indicator of labour market gender equality... The hours worked over the lifetime are different, differential careers result in different opportunities for promotion and pay increases, returns on investment in education differ. Measures of comparable worth all show that at the same skill level women are paid less. With all these factors to consider, a comparison of wage rates is not very meaningful (Silvera, 1998) except to highlight where women are still far behind, such as in Portugal.

Equality in hours of employment between men and women

In looking at labour force participation rates it is always important to remember that it is a minimalist definition of work for only one hour during the reference week. For the four countries in 2000, Portugal has the lowest incidence of women’s part-time work (12.6%) and Belgium has the highest at 34.4% (Pissarides et al., 2005: Table 2.2; see also Hakim, 1997). Recent data indicate a growth in part-time work among women (Casaca and Kovács, 2007: 114). There appears to be a north-south divide in that for Portugal and Spain the incidence of parttime work is significantly lower and more likely to be involuntary; more than one third of the Iberian women part-time workers indicated that they would like full time work in 1999 (Pissarides et al., 2005, Table 4.1).

In Belgium with better child care facilities the incidence of part time work is greater but it is less likely to be involuntary in contrast with Portugal. While having two or more children does result in an increase in part time work for mothers in Belgium and France, it has no effect on part time work in Portugal (where it remains low) and a limited effect in Spain in 2000 (Pissarides et al., 2005, Table 2.6). In addition to part time work, there is another type of work available in the market place, that of temporary work. Together with part time work these jobs are part of the flexible labour force. But in Spain and Belgium more than two thirds of these temporary jobs are reported as involuntary by the woman; in Portugal 41% of the temporary jobs are involuntary in 1999 (there is no data for France; Pissarides et al., 2005: 42).

Although the surveys on involuntary temporary work and part time work are not very reliable they provide a rough indicator of «flexploitation» (Casaca and Kovács, 2007). While economists (cf. Pissarides et al., 2005) encourage flexible work arrangements in order to increase women’s employment rates, much of this flexibility relates to what would be convenient for the employer not the type of work the woman would like. In Spain, for example, where there is a low employment rate for women, three quarters of women with temporary jobs would like a permanent job and more than one third of Spanish women working part time would like full time work. Introducing more of these flexible jobs in Spain is not likely to solve issues of women’s employment. In contrast, some national programs and cultures support part time work and regulate it so that it is not exploitive. In the Netherlands over 55% of women work part time and less than 8% report that it is involuntary. Thus there is the supply side for women, involving what a woman wants and what the cultural expectations are. On the demand side are the types of jobs available. Women in Spain and Portugal want full time permanent work and any move to flexible working conditions on the part of employers could result in «flexpoitation.» In Belgium newly created jobs have been part-time leaving women few options (Maruani, 1998). The offer of flexible work arrangements, however, is most problematic for women in low-paid jobs, and especially single parents. «When low-paid jobs are concerned, flexibility has most often a family-unfriendly character. In these cases, flexibility meant unsocial and atypical working hours and it actually acted as a hindrance to lone parents’ access to work» (Kroger et al. 2003: 40). Under the current push for liberalization and deregulation of labour markets (Pissarides et al., 2005) it is likely that women will lose. This restructuring of labour markets has resulted in the creation of «workfare» for women (Crespi, 2007).

The goal of an employment rate of 60% for women aged 15-64

The European Union’s goal of increasing women’s employment to 60% for women age 15 to 65 a misplaced one. Portugal had already achieved this goal by 2000 and was joined by France in 2007. But Portugal achieves this goal with high rates of employment for girls 15-19. The first priority for these girls should be access to education. Including this young group creates a false sense of accomplishment since the price they pay for working instead of going to school is a lifetime one. In Portugal educational opportunities are also restricted by child labour (http://www.eurofound.europa.eu/eiro/2002/07/word/pt0207105fpt.doc). The goal of 60% employment is also misplaced in countries like Belgium where this goal will soon be met but with a very high portion of part time work. Recent data indicate that 40.5% of women’s employment in Belgium is part time (Casaca and Kovács, 2007: 114). In addition there is the problem of temporary work, especially since much of it is appears to be involuntary.

In the discussions on providing access to employment for women, the situation of older women has not received sufficient attention. They remain the largest underutilized labour force in these four countries. The employment rates for women are already low in the ages 55-59, especially in Spain and Belgium at 35.2 and 32.5 respectively in 2004. Portugal and France have higher rates, 50% in these age group. But for all countries these rates drop precipitously at ages 60-64. In Belgium only 10% of these women are working. Similarly low rates are found in France (12.4%) and Spain (18.8%). Portuguese women have the highest rate at 36.1% (http://epp.eurostat.ec.europa.eu/extraction/, 2004). This is surprising because at ages 55-64 child care is no longer an issue. Are these women engaged in elder care and/or care of their grandchildren? In countries in Southern Europe the limited availability of assistance with these tasks can reduce the availability of women in these ages for employment (Casaca and Damião, 2008). In Spain, where succeeding cohorts will have had more labour market experience, will they also have to drop out for caregiving tasks?

It is likely that the next big hurdle in increasing women’s employment rates lies in how to provide better access to the labour markets for older women. While it is often the case that younger women benefit by lookism that same factor can work against older women. In addition older women may have more physical limitations in terms of the work that they are able to do. If employers are not sensitive to restructuring jobs for older women workers those workers may have to quit. An additional question is what about the older worker who has a temporary job? At older ages, both men and women find it difficult to get a new job and unemployment lasts longer. This is another area where more research by in depth interviews is necessary.

Women’s labour other than «economic»

Understanding women’s paid work is only part of the picture. The other part is women’s unpaid work. It might be important to develop a measure of leisure time which would no doubt indicate that women work in both paid and unpaid work more hours a week than men. If women’s care work and housework were included, for example, the picture of her working week would be very different (Hochschild, 1989). Overall the burden of women’s paid and unpaid work is greater than the man’s. While he is likely to work more hours in paid work, her unpaid work is much greater. Cultural and structural constraints limit this gendered choice (Aïach, Cèbe, Cresson and Philippe, 2001).

While the official statistics are a good place to begin cross national comparisons they do not provide the meaning of work for women or the value of a women’s work. And they do not take into account the total labour of women in society. Much of women’s labour is not counted; «persons who produce services for the consumption of their households which at present are not accounted for in national productions statistics and therefore are not considered as employed» (www.ilo.org/globa./What_we_do/Statistics, p 2). The ILO document does claim that «Their contribution to the well being of the populations is increasingly recognized» but many researchers doubt this claim. For example a woman who takes care of her own preschool children at home is not considered employed but if she takes care of her neighbors preschool children for pay then she is employed. So far there have been limited attempts to include women’s productive but unpaid work into national accounts. This aspect of women’s work for household consumption does present serious problems in data collection and interpretation.

Caution needs to be exercised in the interpretation of government data on  women’s work behavior. In the agricultural sector the reporting of an unpaid family worker can be problematic. A woman counted as «employed» may have only received pay for one hour of work in the reference week. Not only does her work behavior reflect cultural expectations but she may also report only behavior which is consistent with cultural expectations, especially in traditional societies.

The problem of cultural expectations is compounded in cross national research where cultural expectations for women differ. The movement of women into paid employment in the formal sector, however, increases the probability that they will report themselves as working but participation in paid employment has also been impacted by the cultural expectations for mothers since it takes place away from the household and children. The influence of cultural expectations can be very powerful as in the case of Portugal. The lack of adequate child care and the lower levels of education of Portuguese women would lead to the prediction of low labour force participation rates in the prime working ages (25-54). But the situation is quite the opposite.

The role of education and of the educational system is important in understanding women’s work behavior. In fact, the impact of motherhood has diminished and the impact of education has increased so that education has a greater impact than motherhood in Portugal. The importance of the education of young women requires a modification of the statistical target of 60% of women employed at ages 15-64 in current European labour policy. In particular the inclusion of young women (and young men) ages 15 to 20 as a target goal for working defeats the requirements of modern labour markets for a highly educated work force. And women with low education are particularly vulnerable.

At the other end of the age structure the level of employment of women aged 55-59 needs closer scrutiny. These low rates reflect generation effects, especially lower education levels as well as strong cultural expectations for mothers in this cohort. But part of this low rate of employment, is likely to be due to age discrimination. And women tend to be more affected by age discrimination than men. The issue of discrimination against older women brings to the forefront the demand side. What kind of jobs are being offered to women? Initially efforts need to be made in eliminating involuntary temporary work and involuntary part time work for women. In addition even the presumed «voluntary» character of women’s work decisions must be examined in light of the many constraints. Where are the real choices regarding the family-friendly environments?

While the official statistics are a good place to begin cross national comparisons they do not tell us the meaning of work for women or the value of a women’s work in society. Researchers who use the available labour force data need to be more explicit about the limitations especially in cross national research. There is a need for cross national in-depth interviews in order to understand the nature of women’s work and its meaning for the women themselves as well as problematic working conditions. Only this in-depth perspective can lead to specifying policies which can enable women to have a fuller work and family life and to reduce gender inequality.

Annexes

References

Ahmed, Sania S. and Bould, Sally. (2004), «One able daughter is worth 10 illiterate sons», Journal of Marriage and Family, 66, pp. 1332-1341.        [ Links ]

Aïach, Pierre; Cèbe, Claudine; Cresson, Geneviève; Philippe, Claudine (eds.) (2001), Femmes et Hommes dans le Champ de la Santé, Rennes, Editions ENSP.

Amott, Teresa and Matthaei, Julie (1996), Race, Gender and Work, Boston, South End Press.

Arensberg, Conrad. (1937), The Irish Countryman, New York, Macmillan.

Bernard, Jessie (1981), «The good-provider role: Its rise and fall,» The American Psychologist, 36, 1, pp. 1-12.

Blossfeld, Hans-Peter, Drobinc, Sonia and Rohwer, Gotz (2001), «Spouses’ employment careers in (West) Germany,» in P. Blossfeld and S. Drobnic (eds.), Careers of Couples in Contemporary Societies. Oxford, Oxford University Press, pp. 53-73.

Boserup, Ester (1970), Women’s Role in Economic Development. New York, St Martin’s Press.

Bould, Sally (1984), «Development and the family: Third world women and inequality,» International Journal of Sociology and Social Policy, December, pp. 38-51.

_____ (2008), «Critical resources for poor families: A critique of the United Nations Millennium Development Goals», in M. Ahmad (ed.) Poverty Reduction: Policies and Global Integration, Lahore, Pakistan, ISOSS Publications (forthcoming).

Bould, Sally and Schmaus, Gunther (2008), «The European Union roadmap (2006-2010): Equal economic independence for women and men? (in Italian), Sociologia e Politiche Sociali, 11, pp. 35-57.

Brainerd, Carol P. and Miller, Ann R. (1957), «Labor force estimates» in Everett S. Lee, Ann R. Miller, Carol P. Brainerd, Richard Esterlin, (eds.), Population Redistribution and Economic Growth: United States, 1870–1950. Vol. 1., Philadelphia, American Philosophical Society.

Casaca, Sara Falcão and Kovács, Ilona (2007), «Flexibilidad y desigualdad en el trabajo: tendencias y alternativas europeas,» Sociología del Trabajo, 61, Madrid, Siglo XXI, pp. 99-124.

Casaca, Sara Falcão and Damião, Sonia (2008), «Gender equality in the labour market and the role of the welfare state in Southern Europe,» Paper presented at the IV Symposium of the Network Gender and Well-being, Madrid, 25-27 June.

Charles, Maria and Cech, Erin (forthcoming), «Varieties of motherhood: Women’s beliefs about maternal employment in 32 Countries,» in Judith Treas and Sonja Drobnic, (eds.) Dividing the Domestic: Men, Women, and Housework in Cross-National Perspective.

Crespi, Isabella (2007), «Gender mainstreaming and reconciliation policies in Europe,» in Isabella Crespi (ed.) Gender Mainstreaming and Family Policy in Europe, Macerata, Italy, Edizioni SIMPLE, pp. 13-96.

Dejours, Christophe (2003), L’Évaluation du Travail à l’Épreuve du Réel. Critique des Fondements de l’Évaluation, Dijon, INRA Editions.

EUROFOUND [on line] Available at http://www.eurofound.europa.eu/eiro/2002/07/word/pt0207105fpt.doc (Accessed on 10/05/2008).

Ferman, Louis and Ferman, Patricia (1973), «The Structural underpinnings of the irregular economy,» Asia Pacific Journal of Human Resources, 8, pp. 1-17.

Ferreira, Virginia (1998), «Women in Portugal: Situation and Paradoxes,» [on line] Available at www.lxxl.pt/babel/biblioteca/women.html [Accessed 10/05/2008].

Flahaut, Erika (ed.) (2006), L’Insertion Professionnelle des Femmes: Entre Contraintes et Stratégies d’Adaptation, Rennes, Presses Universitaires de Rennes.

Gadrey, Nicole ( 2001), Travail et Genre, Approches Croisées, Paris, L’Harmattan, Collection Logiques sociales.

Gavray, Claire (2007), «Le travailleur n’est pas un électron libre en emploi», in Marie-Thérèse Casman, Caroline Simays, Riet Bulckens, Dimitri Mortelmans (eds.), Familles Plurielles, Bruxelles, Editions Luc Pire, pp. 109-116.

_____ (2008), «Genre, emploi et marché du travail: un tableau contrasté», in Annie Cornet, Jacqueline Laufer and Sophia Belghiti-Mahut (ed), GRH et Genre, les Défis de l’Égalité Hommes-Femmes, Paris, Editions Vuibert, pp. 5-24.

Ghodsee, Kristen (2000), «Women and economic transition», Center for Slavic and East European Studies Newsletter 17 (3), Berkeley, University of California.

Gils, Wouter van and Kraaykamp, Gerbert (2008), «The emergence of dual-earner couples: A longitudinal study of the Netherlands,» International Sociology, 23, (May), pp. 345-366.

Gordon, Linda (1994), Pitied But Not Entited: Single Mothers and the History of Welfare, New York, Free Press.

Goulart, Pedro and Bedi, Arjun S. (2005), «Child labour and educational success in Portugal», ISS Working Paper Series, n.º 412.

Granovetter, Mark S. (1995), Getting a Job, Chicago, University of Chicago Press, 2nd Ed.

Gubin, Eliane (2000), «Geneviève Fraisse, Les femmes et leur histoire; Michelle Perrot, Les Femmes ou les silences de l’Histoire», Clio, 12, [on line] Available at http://clio.revues.org/index199.html, mis en ligne le 20 mars 2003 [Accessed on 10/05/2008 ]

Hakim, Catherine (1997), «A Sociological perspective on part-time work,» in Blossfeld, H. and Hakim C., Between Equalization and Marginalization, Oxford, Oxford University Press, pp. 22-70.

Hakim, Catherine (2000), Work-lifestyle Choices in the 21st Century, Oxford, Oxford University Press.

_____ (2003), Models of the Family in Modern Societies, Andershot, England Ashgate Publishing Limited.

Hartmann, Heidi and Whittaker, Julie (1998), The Male/Female Wage Gap: Lifetime Earnings Losses, Washington D.C., Institute for Women’s Policy Research, [on line] Available at www.iwpr.org [Accessed 10/05/2008].

Heynes, Barbara (2005), «Emerging inequalities in Central and Eastern Europe», Annual Review of Sociology, 31, pp. 163-197.

Hochschild, Arlie. R. with Machung, Anne (1989), The Second Shift, New York, Viking.

ILO [on line] Available at www.ilo.org/globa./what_we_do/statistics, p. 2 (Accessed on 03/05/2008).

Koeva, Stefka and Bould, Sally (2007), «Women as workers and as carers under communism and after: The case of Bulgaria,» International Sociology, 17, 2, pp. 303-318.

Kroger Teppo, et al. (2003), SOCCARE Project Report 6, [on line] Available at www.uta.fi/laitokset/sospol/soccare/report6.pdf [Accessed 10/05/2008].

Lebergott, Stanley (1966), «Labor force and employment, 1800-1960,» Studies in Income and Wealth, 24, pp. 117-204.

Lopes, Margarida Chagas and Goulart, Pedro (2003), «Portuguese data on child work: What does it encompass?», Working Paper 04/2003/DE/CISEP, ISEG.

Marry, Catherine (2006), «L’éducation des femmes en Europe: une émancipation inachevée», in Christine Ockrent and Sandrine Treiner (eds.), Le Livre Noir de la Condition Féminine dans le Monde, Paris, Editions XO.

Maruani, Margaret (ed) (1998), Les Nouvelles Frontières de l’Inégalité, Hommes et Femmes sur le Marché du Travail, Paris, La Découverte-Mage.

Pettit, Becky and Hook, Jennifer (2005), «The structure of women’s employment in comparative perspective», Social Forces, 84 (2), pp. 553-578.

Pissarides, Christopher, Garabaldi, Pietro, Olivetti, Claudia et al. (2005), «Women in the labour force: How well is Europe doing?» in Tito Boeri, Daniela Del Boca, Christopher Pissarides (eds.), Women at Work: an Economic Perspective, Oxford, Oxford University Press, pp. 8-104.

Rogers, Barbara (1980), The Domestication of Women, New York, Tavistock.

Rosen, Ellen. I. (2002), Making Sweatshops, Berkeley, University of California Press.

Rostow, Walt W. (1961), The Stages of Economic Growth, London, The Cambridge University Press.

Salaff, Janet W. (1981), Working Daughters of Hong Kong, New York, Cambridge University Press.

Shorto, Russel (2008), «No babies?», New York Times, 29 June.

Silvera, Rachel (1998) «Les salaires, toutes choses inégales par ailleurs», in Maruani, Margaret (ed.), Les Nouvelles Frontières de l’Inégalité, Hommes et Femmes sur le Marché du Travail, Paris, La Découverte-Mage, pp. 127-138.

Titmuss Richard M. (1962), Income Distribution and Social Change, London, George Allen and Unwin.

Trifiletti Rossana (1999), «Southern European welfare regimes and the worsening position of women», Journal of European Social Policy, 9, 1, pp. 49-64.

Uchitelle, Louis (2008), «Women are now equal as victims of poor economy,» New York Times, 22 July.

United Nations Research Institute for Social Development UNRISD (2005), Gender Equality, Geneva, UNRISD.

Voldman, Danièle and Schweitzer, Sylvie (2002), Les Femmes ont Toujours Travaillé: Une Histoire du Travail des Femmes aux XIXe et XXe siècles, Paris, Editions Odile Jacob.

Wall, Karin (2008), «Leave Policy Models and the Articulation of Work and Family Life in Europe: a comparative perspective» (in Italian) Sociologia e Politiche Sociali, 11, 1, pp. 59-85.

Wolf, Diane L. (1992), Factory Daughters: Gender. Household Dynamics and Rural Industrialization in Java, Berkeley, University of California Press.

Sally Bould, PhD. is Senior Research Fellow, Centre for the Study of Population, Poverty and Socio-economic Policy (CEPS/INSTEAD) Differdange, Luxembourg and Professor Emerita of Sociology, University of Delaware, Newark. She has published numerous articles in the areas of work, family, poverty and ageing, with a focus on policy. She received a Fulbright Award in 2006 to study issues of women, aging and families in Europe. Her most recent publications include: «The European Union Roadmap (2006-2010): Equal economic independence for women and men? (in Italian with Gunther Schmaus) Sociologia e politiche sociali 11, 3, 2008; «Women as Workers and as Carers under Communism and After: The Case of Bulgaria» (with Setfka Koeva) International Review of Sociology 17 No. 2. 2007; «The Need for International Family Policy: Mothers as Carers and as Workers.» in L. Haas and S. K. Wisendale (eds.) Families and Social Policy: National and International Perspectives. New York: Haworth Press 2006.

E-mail: salbould@udel.edu

Claire Gavray is founder member of FERULG – Women, Education, Researches, University of Liège (Belgium) where she received her PhD in Sociology. She shares her activities of teaching and research between the Faculty of Psychology and the Sciences of Education, the Institute of Human and Social Sciences and the School of Criminology. Her main research concerns the evolution of the labour market, employment, well-being and life conditions by gender. Recent publications include: Gavray, C. (dir.) (2008). Femmes & Mobilités, Marcinelle, Cortext and Gavray, C. (2008), «Genre, emploi et marché du travail: un tableau contrasté », In Cornet, A., Laufer, J. et S. Belghiti-Mahut (Ed)., GRH et genre: les Défis de l’égalité hommesfemmes, Paris, Vuibert, 5-24.

E-mail: cgavray@ulg.ac.be

Artigo recebido em 31 de Maio de 2008 e aceite para publicação em 8 de Agosto de 2008.