Women's participation in the labour force

Johnston, Grant Working Paper Women's participation in the labour force New Zealand Treasury Working Paper, No. 05/06 Provided in Cooperation with: The Treasury, New Zealand Government Suggested Citation: Johnston, Grant (2005) : Women's participation in the labour force, New Zealand Treasury Working Paper, No. 05/06, New Zealand Government, The Treasury, Wellington This Version is available at: https://hdl.handle.net/10419/205571 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Women’s participation in the labour force Grant Johnston N EW Z EALAND T REASURY W ORKING P APER 05/06 J UNE 2005 Treasury:734607v2 NZ TREASURY WORKING PAPER 05/06 Women’s participation in the labour force MONTH / YEAR June 2005 AUTHOR Grant Johnston The Treasury PO Box 3724 Wellington New Zealand Email Telephone Fax [email protected] +64 4 471 5167 +64 4 472 4942 ACKNOWLEDGEMENTS Paul Callister, Sylvia Dixon, Tim Maloney and Bob Buckle provided very helpful comments on an earlier version of this paper. I am also indebted to my colleagues at the Treasury for their assistance and in particular to Angus White, who worked on this project during his summer internship. NZ TREASURY New Zealand Treasury PO Box 3724 Wellington 6008 NEW ZEALAND Email Telephone Website [email protected] 64-4-472 2733 www.treasury.govt.nz DISCLAIMER The views, opinions, findings, and conclusions or recommendations expressed in this Working Paper are strictly those of the author(s). They do not necessarily reflect the views of the New Zealand Treasury. The Treasury takes no responsibility for any errors or omissions in, or for the correctness of, the information contained in these working papers. The paper is presented not as policy, but with a view to inform and stimulate wider debate. WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE i Abstract Labour force participation is a topical issue in New Zealand. It is well known that the participation of New Zealand women aged 25-39 is low in comparison with women in other OECD countries. There has been considerable interest in policies which might raise women’s participation. This paper provides a base of information on women’s labour force participation in New Zealand and in other OECD countries. The low participation of younger New Zealand women seems to be driven largely by a combination of relatively low participation rates among mothers with young children and sole mothers, together with high fertility rates and high proportions of sole parent families. However, while New Zealand women tend to leave the labour force when they have children, they also tend to return strongly to the labour force when their children get older. Considered over all ages, New Zealand has a reasonably healthy female participation rate, and the total quantity of work done in New Zealand, relative to the size of the working-age population, is amongst the highest in the OECD. JEL CLASSIFICATION J10, J21 KEYWORDS women; participation; employment; labour supply; New Zealand WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE ii Table of Contents Abstract...............................................................................................................................i Table of Contents ..............................................................................................................ii List of Tables......................................................................................................................ii List of Figures....................................................................................................................ii 1 Introduction ..............................................................................................................1 2 The economics of participation ..............................................................................2 2.1 The basic model of labour supply...................................................................................2 2.2 Why raise participation? .................................................................................................3 3 Data and definitions.................................................................................................5 4 Women’s participation in New Zealand..................................................................7 4.1 Previous research...........................................................................................................7 4.2 Patterns of participation..................................................................................................8 4.3 Historical trends............................................................................................................11 4.4 Participation of mothers................................................................................................14 4.5 Projections of labour force participation .......................................................................20 5 International comparisons ....................................................................................23 5.1 Labour force participation in the OECD........................................................................23 5.2 Labour force participation by age.................................................................................25 5.3 Employment and participation of particular groups of women in the OECD................27 5.4 Accounting for paid parental leave...............................................................................31 6 Conclusion and discussion...................................................................................34 References .......................................................................................................................36 Appendix – Growth accounting......................................................................................39 List of Tables Table 1 – Women’s participation and non-participation in the labour force, by age group.................8 Table 2 – Full-time participation rates by age and sex, 1951-2001..................................................13 Table 3 – Participation of sole and partnered mothers by age of youngest child and highest qualification.............................................................................................................................17 Table 4 – Number of sole and partnered mothers not in the labour force by age of youngest child and highest qualification.................................................................................................18 Table 5 – Percent of women in each age group who are mothers, by age of youngest child..........20 Table 6 – Projections of the female labour force, 2001 (base) to 2031, under three scenarios.......21 Table 7 – Proportion of the total population who are in the labour force, 2001 (base) to 2031, under three scenarios.............................................................................................................22 Table 8 – Paid maternity leave provisions in Nordic countries and New Zealand............................32 List of Figures Figure 1 – Annual hours of work per person, men and women combined, for the population aged 15-64................................................................................................................................4 Figure 2 – Participation of men and women by age............................................................................9 Figure 3 – Employment status by age.................................................................................................9 WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE iii Figure 4 – Women’s participation......................................................................................................11 Figure 5 – Labour force participation, 1976-2001.............................................................................12 Figure 6 – Participation of successive cohorts of women.................................................................14 Figure 7 – Participation of mothers compared to women with no children and to men....................15 Figure 8 – Mothers’ labour force participation...................................................................................16 Figure 9 – Mothers’ participation by age of mother...........................................................................19 Figure 10 – Mothers’ participation by age of mother and age of youngest child..............................19 Figure 11 – Projected participation rates for women, 2001 (base) and 2011, under three scenarios.................................................................................................................................20 Figure 12 – Women’s participation rates from the HLFS, 1996-2004, for various age groups ........22 Figure 13 – Participation rates in the OECD for the population aged 15-64 ....................................23 Figure 14 – Part-time workers as a percentage of all female workers, 15-64..................................24 Figure 15 – Women’s labour force participation, 1976 and 2003.....................................................24 Figure 16 – Labour force participation in the OECD.........................................................................25 Figure 17 – Groupings of OECD countries by women’s participation profiles..................................26 Figure 18 – Employment of mothers in the OECD, 2001..................................................................28 Figure 19 – Employment of women aged 20-49 in the OECD with no children, 2001 .....................29 Figure 20 – Total fertility rates in the OECD, 2002...........................................................................29 Figure 21 – Employment of sole parents, various years (1996-2001)..............................................30 Figure 22 – Sole parent families as a percentage of all families with children, 1996.......................30 Figure 23 – Employment of women 25-54 with a university degree, 2002.......................................31 Figure 24 – Women’s participation in Nordic countries and in New Zealand, raw and adjusted for maximum time on paid parental leave................................................................32 WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE 1 Women’s participation in the labour force 1 Introduction This paper is part of a programme of work being undertaken at the Treasury on labour force participation in New Zealand. It uses Census data to show the patterns of participation and non-participation among different groups of New Zealand women, and compares these with the patterns found in other OECD countries. These comparisons are intended to inform policy discussions, but it should be noted from the outset that differences in participation rates do not, in themselves, constitute an argument for or against measures to increase the labour force participation of women in New Zealand. The focus of the paper is on women’s participation and non-participation in the labour force, rather than on the number of hours they work. In many places in the paper, however, labour force participants are divided into full-time workers, part-time workers and the unemployed, which gives some indication of the amount of time they work. A fuller consideration of the hours worked by New Zealand women, and by families as a whole, is given elsewhere, for example by Callister (2004, 2005a, 2005b). The basic economic model of labour supply and the rationale for investigating women’s labour force participation are outlined in Section 2. This provides a background and a framework for the study. Section 3 briefly discusses the sources of data and some key definitions. Sections 4 and 5 lie at the heart of the paper. Section 4 is concerned with the participation of New Zealand women, namely how participation rates differ between women and men; how women’s participation has changed over recent decades; how it may change in the future; and what personal and family characteristics are associated with participation and non-participation amongst women. In particular, groups of women are distinguished according to their age, qualifications, the presence and age of children, and whether women are sole or partnered mothers. Section 5 compares the participation rates of women across OECD countries, and identifies groups of countries that are similar in their participation profiles. This section also tries to identify the key differences between women’s participation in New Zealand and their participation in other countries. Section 6 summarises and discusses the findings of this paper. WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE 2 2 The economics of participation 2.1 The basic model of labour supply People choose how much time they will spend working (per week, say, or per year) by weighing up the competing attractions of work and non-work activities. By working for pay, people earn money with which they can buy goods and services. On the other hand, people also value what they can do outside of paid employment, including unpaid production (housework, looking after children, painting the house, voluntary work, etc), study and leisure time. According to the basic economic model of labour supply, people will spend more time in paid work if the goods they could buy with their extra income would be more valuable to them than the non-work activities they would have to forego (and vice versa). People do these sorts of calculations, for example, when they decide whether to work part-time rather than full-time, to get a second job, or to take a few months leave without pay to travel overseas. In this way, people decide how much work will maximise the satisfaction they get from both consuming goods and undertaking non-work activities.1 This description reveals the two key elements lying behind labour supply decisions. One is the net return that people get from different amounts of work, that is, what they will have in their pockets once they have earned wages; paid income tax, crèche fees, transport costs, etc; and have lost any entitlement to benefit income or subsidies. The other key element is the value that people place on non-work activities relative to the consumption of goods and services. This valuation is likely to be affected by factors such as family commitments, hobbies and interests, and how productive individuals are at performing household tasks. People often put a high value, for example, on spending time with their children, particularly during the first, formative years of their lives. Some people do not wish to work, since their satisfaction is maximised (given prevailing wage rates2, income taxes, etc) by working no hours at all. These people are not participating in the labour force. They are in a situation where the income to be gained from paid work is not sufficiently attractive to lure them into giving up any of their non-work activities. On the other hand, people whose satisfaction is maximised by working one or more hours a week are said to be participating in the labour force. People are participating in the labour force if they are actually employed, full-time or part-time, or if they are unemployed and actively seeking work. The “participation rate” for a particular group is the proportion of that group who are participating, that is, who are employed or unemployed. This simple model of labour supply has been extended in many different ways, to incorporate further aspects of labour supply decisions—for example to recognise that labour supply decisions are often made by families, rather than by individuals—or to take into account dynamic effects happening across the lifecycle. But even the simple model is 1 Labour supply theory is discussed in an accessible way in all introductory labour economics texts, eg Borjas (2000) and in much more detail in specialised texts such as Killingsworth (1983). 2 Wages are determined by the interaction of the demand for labour (from firms) and the supply of labour (from individuals). The demand for labour therefore also has an effect on participation rates. Notably, during an economic slowdown, when lower demand for labour pushes wage rates down, many people opt out of the labour force for a time—the so-called “discouraged worker” effect. Also firms may affect participation because they demand labour in lumpy amounts, at a given time of the day or week, which may not suit prospective workers. WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE 3 useful for understanding the patterns of participation we observe in New Zealand and for making broad predictions about the effects of policies. For example, we would expect to see that higher skilled people, who command higher wages, are more likely to be participating than lower skilled people. We would also expect to see that parents of children, especially young children, have relatively low participation rates as they have many things they would like to do, or feel they have to do, with their non-work time. In terms of policy changes, the model predicts, for example, that a childcare subsidy would increase the labour force participation of mothers of pre-schoolers, since the subsidy would have the effect of increasing the net return that these mothers get from working. 2.2 Why raise participation? Recent interest in raising labour force participation in New Zealand has been motivated, in the first instance, by a consideration of economic growth. Mathematically, an increase in participation leads to an increase in the number of hours of work done in the country, which in turn leads to an increase in GDP per capita (all else being equal). This is evident from a consideration of the growth accounting framework used by the Ministry of Economic Development and Treasury (2005), which is outlined in the Appendix. In its Economic Survey of New Zealand, the OECD (2003a) also made this point, stating that an increase in participation would help towards New Zealand’s goal of being in the top half of OECD countries for GDP per capita. The degree to which this might happen has been explored by Bryant, Jacobsen, Bell and Garrett (2004). From international comparisons, Bryant et al (2004) also identified that New Zealand stands out amongst OECD countries for having low participation rates for women aged 25 to 34 years. The marked dip in New Zealand women’s participation at this stage in the lifecycle does not appear to be a common feature in the OECD. In both the 25-29 year age group and the 30-34 year age group, participation rates for New Zealand women are ranked 22nd out of 24 OECD countries. This suggests, on the face of it, that younger women constitute a population group in which New Zealand could fruitfully encourage participation. The labour force participation of younger women, especially mothers, has therefore been a focus of recent policy discussions in New Zealand. The Prime Minister has said that the Government will look at a range of policies, including parental leave and childcare policies, to see how to boost both women’s participation in the workforce and good outcomes for children (Clark 2005). It is important to bear in mind, though, that the total quantity of work done in New Zealand, given the size of our population, is in fact very high by international standards. Figure 1 shows, for each country in the OECD, the number of hours worked each year per person of working age (15 to 64 years). Compared to people in other developed countries, New Zealanders spend a lot of time at work. The productivity of this work, however—the value of the goods and services produced per hour—is relatively low in New Zealand. Both the Ministry of Economic Development and Treasury (2005) and the OECD (2003a) consider that raising productivity, rather than increasing the quantity of work undertaken, is the key to a sustainable increase in New Zealand’s GDP per capita. WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE 10 b. men 0 10 20 30 40 50 60 70 80 90 100 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 Age group (years) Percent Employed full-time Employed part-time Unemployed Source: 2001 Census. For both men and women, the relatively low participation of young people is largely explained by the high proportion of students in this age group. Declining participation rates among older people are likely to be the result of a number of factors, for example worsening health and increasing prevalence of disability; lower consumption when children have left home and the mortgage is paid off; and increased levels of non-labour income from investments and pensions.11 Amongst 25-54 year olds (the so-called “prime ages” for labour force participation) it is clear that many women, and some men, are not participating because of childcare and other household responsibilities. The participation of mothers is discussed in detail in Section 4.4. Other non-participants might be notional job seekers, sick or disabled, fulltime students, or voluntary workers.12 At all ages, women’s participation in the labour force is stratified by qualifications and by ethnicity (Figure 4). As the economic model of labour supply predicts, more-educated women—who can typically earn higher wages—have greater participation rates than lesseducated women. At all ages, European women have greater participation rates than women in any other ethnic group. As discussed in Section 4.1, however, this may largely be because of differences in other measurable characteristics. Notably, women with no qualifications, and Maori women, show little sign of a dip in participation. At least partly, this may be because of higher fertility and younger parenthood in these groups (Statistics New Zealand 2004a, Dharmalingam, Pool, Sceats and Mackay 2004), so that these women do not establish themselves in the labour force before they begin to have children. 11 Jensen, Sathiyandra, Rochford, Jones, Krishnan and Mcleod (forthcoming) show that people with disabilities in New Zealand have lower participation rates. The prevalence of disability increases markedly with age (Maskill, Hodges, Burns and Carroll 2004). There is a significant international literature on the influence of health and disability status on labour force participation. The issues in this literature are discussed, for example, in Bound, Schoenbaum, Stinebrickner and Waidmann (1998) and Stern (1989). 12 People who are not in the labour force might still be looking for work, but not as actively as the official definition of unemployment requires. Some of these notional job seekers will be receiving the Unemployment Benefit and some will not. Likewise, some of the people who are sick or disabled might be receiving a benefit or ACC compensation, and some will not. Dixon (1999) categorises primeage men who are not in the labour force into these sorts of categories. WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE 11 Figure 4 – Women’s participation a. by highest qualification 0 10 20 30 40 50 60 70 80 90 100 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 Age group (years) Participation rate Post-school qual. School qual. No qual. b. by ethnicity 0 10 20 30 40 50 60 70 80 90 100 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 Age group (years) Participation rate European Maori Pacific Asian Source: 2001 Census. Note that ethnicity is defined here using the total responses definition: if a person reported more than one ethnic group they have been counted in each applicable group. 4.3 Historical trends 4.3.1 Trends from 1976 to 2001 Figure 5 shows a consistent series of Census-based labour force participation rates from 1976 to 2001.13 In most age groups, women’s participation has increased between each 13 Putting together a consistent historical series is complicated by issues of measurement—Census definitions, and therefore the data collected and presented, have changed over time. The wording of some questions has also changed between Censuses. Before 1986, full-time employment was defined as working 20 or more hours a week. Since 1986 full-time employment has been defined as working 30 hours or more a week. Also, before 1991, unemployment was defined simply as seeking work; from 1991 it was defined as actively seeking work, in line with the HLFS definition. The series shown here uses the latest definitions, and older Census data has been WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE 12 Census (Figure 5a). The dip in women’s participation, which was very pronounced in 1976, has also been shallowing out, so that the “m-shaped” participation profile has become much less distinguishable over time. The dip has also moved over to the right, probably reflecting the trend of increasing age of mothers. At the same time, participation rates for men, at least in the peak working ages, have been falling (Figure 5b). For older men, however, participation has increased markedly over the past decade, largely as a result of the change in eligibility for Superannuation from age 60 to age 65 (Hurnard 2003). Figure 5 – Labour force participation, 1976-2001 a. women 0 10 20 30 40 50 60 70 80 90 100 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 Age group (years) Participation rate 2001 1996 1991 1986 1981 1976 b. men 0 10 20 30 40 50 60 70 80 90 100 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 Age group (years) Participation rate 2001 1996 1991 1986 1981 1976 Source: Author’s calculations based on Census documents from 1976-2001. While the increases in women’s participation and decreases in men’s participation have to some extent cancelled each other out, the overall trend has been towards increasing adjusted to reflect this. It has not been possible to extend this series back beyond 1976. To adjust for differences in the definition of unemployment, HLFS unemployment backdates calculated by Chapple (1994) have been used. WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE 13 participation amongst the population as a whole. Except for 15-24 year olds, the total participation rate for men and women combined was higher for all age groups in 2001 than it was in 1976. Most of the increase in “prime-aged” women’s participation since 1976 has been in fulltime, as distinct from part-time, employment. In younger and older age groups, however, there have been large increases in part-time employment since 1976. 4.3.2 Longer-term trends Table 2 shows a longer-term series of labour force participation, going back to the 1951 Census. By necessity, this series uses the pre-1986 Census definition of full-time participation which counted people who were working for 20 or more hours a week. This series shows that the trend of increasing women’s participation and decreasing men’s participation extends back to the Second World War. The exception, for women, is in the 15-24 year age group, where participation has fallen markedly because of increased schooling and post-school education.14 Table 2 – Full-time participation rates by age and sex, 1951-2001 15-19 20-24 25-34 35-44 45-54 55-64 15-64 Women 1951 64.6 52.8 22.5 20.7 21.0 12.2 28.4 1961 64.1 49.8 20.9 25.4 28.7 17.8 31.8 1971 56.9 54.9 29.5 37.5 37.7 21.9 38.9 1981 49.5 63.7 41.1 51.9 48.6 22.3 45.8 1991 35.7 63.6 51.3 60.5 60.0 23.4 50.7 2001 32.1 59.0 59.7 61.7 68.0 41.9 56.4 Men 1951 71.5 95.9 97.8 97.6 95.6 75.4 91.3 1961 65.5 94.6 98.5 98.3 97.3 81.7 91.1 1971 57.1 91.0 98.2 98.7 97.2 81.6 88.5 1981 56.6 91.0 97.1 97.6 96.1 69.9 86.2 1991 40.1 78.0 86.7 89.2 87.6 52.4 75.8 2001 39.1 72.6 85.3 87.3 86.2 69.4 77.2 The labour force here consists of people employed for 20 or more hours a week, plus people who were ‘seeking work’ (1951-1981) or ‘actively seeking work’ (1991 and 2001). Source: Statistics New Zealand, Census documents 1951-2001. The 1951-2001 series can also be used to track the participation of different cohorts of women as this is measured in successive Censuses (Figure 6). At all ages, apart from the 15-24 age group, women in more recent cohorts have had higher participation rates than their predecessors. The dip in participation amongst 25-34 year olds has also become less marked with each successive cohort. 14 Longer-term series of women’s labour force participation, although not age-specific series, are also presented in Hyman (1978), Horsfield (1988), Davies and Jackson (1993) and Revell and Brosnan (1986). WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE 14 Figure 6 – Participation of successive cohorts of women 0 10 20 30 40 50 60 70 80 90 100 15-24 25-34 35-44 45-54 55-64 Age group (years) Participation rate born 1957-66 born 1947-56 born 1937-46 born 1927-36 born 1917-26 born 1907-16 Source: Census documents 1951-2001. The massive increase in women’s labour force participation since the Second World War has been a phenomenon in almost all developed economies (Jacobsen 1999). Some of the reasons given for this increase include: • demand for labour causing the real wages of women to rise, therefore making it more worthwhile for women to work, • changes in industrial composition (eg the rise of the clerical sector and the decline of agriculture and labouring), • a growing number of part-time jobs, and their increased availability in a range of different occupations, • technological advances making household work more efficient, and also producing goods (such as TVs) that cannot be produced at home, • an increase in the schooling and post-school education of women, • changes in family composition, due to factors such as an increase in divorces, delayed childbearing, smaller families, and an increase in the proportion of women not having children, • changing conditions of work, and • changing cultural and legal attitudes towards women at work. 4.4 Participation of mothers This section considers the participation of mothers with dependent children (referred to here as simply “mothers”). Dependent children can be aged up to 17 years. Throughout this section the analysis is restricted to women (and to men in some places) who are aged 20 to 54 years, since few mothers are outside this age group. Not surprisingly, mothers have lower participation rates than women with no children, and lower participation rates than men (Figure 7). Note that the category of “women with no WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE 15 children” is not a homogeneous group—it includes women who are yet to have children, those whose children are older than 17 or have left home, and those who have never had children. Figure 7 – Participation of mothers compared to women with no children and to men 0 10 20 30 40 50 60 70 80 90 100 Mothers Women with no children Men Percent Employed Full-time Employed Part-time Unemployed Source: 2001 Census. Analysis is restricted to people aged 20-54 years. Figure 8a shows that sole mothers have lower participation rates than partnered mothers. Sole mothers have lower qualifications, on average, than partnered mothers. They are also more likely to be receiving a social welfare benefit, and therefore to face the disincentives to work which are implicit in means-tested benefits. 15 Sole parents may also have a greater workload at home, since household tasks are not shared with a partner.16 Large families appear to have some effect, but in general the number of children in a family makes little difference to the mother’s participation (Figure 8b). On the other hand, the age of the youngest child in the family makes a marked difference to the participation of mothers (Figure 8c). In all, mothers with pre-school children (aged 0-4 years) have a participation rate of only 54%, but mothers with school-aged children (aged 5-17 years) have a participation rate of 79%. This latter rate is almost as high as that for women with no children (cf Figure 7). There are a number of possible explanations for the difference in participation between mothers of younger children and mothers of older children. One is that mothers consider it important to spend a good deal of time with very young children; another is that the school system effectively supplies free childcare for at least part of the day, making it easier and less expensive for mothers to work. 15 Benefits are highest at zero, or very few, hours of work and are progressively diminished as hours of work increase. These high effective marginal tax rates reduce work incentives for people on benefits. 16 Although, as Craig (2004) demonstrates, male partners can create more work for women, rather than reducing it. WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE 16 Figure 8 – Mothers’ labour force participation a. by sole and partnered status 0 10 20 30 40 50 60 70 80 90 100 Sole mothers Partnered mothers Percent Employed full-time Employed part-time Unemployed b. by number of children in the family 0 10 20 30 40 50 60 70 80 90 100 One Two Three Four or more Number of children Percent Employed full-time Employed part-time Unemployed c. by age of youngest child in the family 0 10 20 30 40 50 60 70 80 90 100 <1 1 2 3 4 5 6 7 8 9 1011121314151617 Age of youngest child Percent Employed full-time Employed part-time Unemployed Source: 2001 Census. Analysis is restricted to people aged 20-54 years. WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE 17 Consistent with the New Zealand studies discussed in Section 4.1, the 2001 Census shows that qualifications (Figure 4a), being a sole or partnered mother (Figure 8a), and age of youngest child (Figure 8c), make a substantial difference to women’s labour force participation. These factors are combined in Table 3, which shows that each of the three factors has an effect independent of the other two. So, for example, even when taking the age of children and mother’s qualifications into account, sole mothers still have lower participation rates than partnered mothers. Table 3 – Participation of sole and partnered mothers by age of youngest child and highest qualification No School qualification School Qualification Post-school Qualification Total Sole mothers with youngest child: 0-4 years 32 44 56 42 5-9 years 56 68 77 66 10-17 years 65 76 85 75 any age 48 60 73 60 Partnered mothers with youngest child: 0-4 years 47 56 64 57 5-9 years 72 80 86 80 10-17 years 79 85 91 86 any age 65 71 78 72 All mothers with youngest child: 0-4 years 41 54 62 54 5-9 years 66 77 84 76 10-17 years 75 83 89 83 any age 59 68 77 69 For comparison: Women with no children 74 84 90 83 Men 83 89 94 88 Source: 2001 Census. Analysis is restricted to people aged 20-54 years. Table 3 shows that participation rates vary widely for different groups of mothers. At one extreme, for example, sole parents with a pre-school child and no school qualification have a participation rate of only 32%. At the other, partnered mothers with a child aged 10-17 and with a post-school qualification have a participation rate of 91%. In terms of labour force participation, mothers are a diverse group. Having young children is perhaps the strongest single influence on labour force participation. Highly-qualified mothers with a child aged 0-4 years, for example, have a participation rate lower than unqualified mothers with a child aged 5-9 years. WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE 18 Table 4 follows the same format as Table 3 but each cell shows the number of mothers who are not participating in the labour force. Again this highlights the importance of the age of the youngest child. Nearly two-thirds of mothers not in the labour force have a youngest child under 5 (85,420 out of a total of 137,080). Table 4 – Number of sole and partnered mothers not in the labour force by age of youngest child and highest qualification No School qualification School Qualification Post-school Qualification Total Sole mothers with youngest child: 0-4 years 9,570 9,050 3,950 25,030 5-9 years 4,170 3,810 2,000 11,190 10-17 years 3,100 2,750 1,720 8,580 any age 16,850 15,610 7,670 44,790 Partnered mothers with youngest child: 0-4 years 10,940 27,770 18,550 60,390 5-9 years 4,130 7,690 4,290 17,360 10-17 years 3,940 5,980 3,370 14,540 any age 19,010 41,440 26,210 92,290 All mothers with youngest child: 0-4 years 20,520 36,830 22,500 85,420 5-9 years 8,300 11,510 6,290 28,550 10-17 years 7,040 8,720 5,090 23,120 any age 35,850 57,050 33,880 137,080 For comparison: Women with no children 20,610 25,770 15,550 71,610 Men 30,840 33,310 20,290 99,740 Source: 2001 Census. Analysis is restricted to people aged 20-54 years. Totals include people whose highest qualification was not stated or unidentifiable so may be greater than the sum of the individual columns. Table 3 and Table 4 also allow some “what if?” testing to take place. For example, if sole mothers were to have the same participation rates in each cell of Table 3 as partnered mothers, the overall participation of mothers would increase from 69% to 72%.17 Of the additional 13,700 labour force participants that would result, only a half would have school or post-school qualifications. We now turn to analysing mothers’ participation by the age of the mother. On the face of it, mothers’ participation increases with age, reaching a peak in the 45-49 year age group (Figure 9). Conversely, the participation of women with no children declines gradually with 17 The overall participation rate for all women aged 20-54 (not just mothers) would rise by 1.5 percentage points. WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE 19 age, although this may reflect a lower participation rate of women whose children have left home and who do not return to the labour force. Figure 9 – Mothers’ participation by age of mother 0 10 20 30 40 50 60 70 80 90 100 20-24 25-29 30-34 35-39 40-44 45-49 50-54 Age group (years) Participation rate Men Women no kids Mothers Source: 2001 Census. When the age of the youngest child is considered, however, mothers’ participation is in fact reasonably steady across age groups (Figure 10). What drives the steeply ascending participation profile for mothers in Figure 9 is simply that older women are less likely to have families with young children in them. Figure 10 – Mothers’ participation by age of mother and age of youngest child 0 10 20 30 40 50 60 70 80 90 100 20-24 25-29 30-34 35-39 40-44 45-49 50-54 Age group (years) Participation rate 10-17 years 5-9 years 0-4 years A ge of youngest child: Source: 2001 Census. Furthermore, the characteristic “dip” in women’s overall labour force participation, as illustrated in Figure 2, is explained largely by the age profile of mothers of young children. Women aged 30-34 years, for example, have relatively low participation rates because they are the most likely to have pre-school children (Table 5). WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE 26 Figure 17 – Groupings of OECD countries by women’s participation profiles Nordic countries and Canada 0 10 20 30 40 50 60 70 80 90 100 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 Age group (years) Participation rate Iceland Sweden Norway Denmark Finland Canada Western Europe 0 10 20 30 40 50 60 70 80 90 100 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 Age group (years) Participation rate Austria France Germany Eastern Europe 0 10 20 30 40 50 60 70 80 90 100 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 Age group (years) Participation rate Slovak Rep Czech Rep Poland Hungary Other European countries 0 10 20 30 40 50 60 70 80 90 100 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 Age group (years) Participation rate Portugal Netherlands Belgium Ireland Spain Greece Italy Countries with a dip 0 10 20 30 40 50 60 70 80 90 100 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 Age group (years) Participation rate Switzerland United Kingdom New Zealand Australia United States Asian countries 0 10 20 30 40 50 60 70 80 90 100 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 Age group (years) Participation rate Japan Korea Source: OECD Labour Market Statistics, 2003 data. WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE 27 The Nordic countries, together with Canada, have the highest women’s participation rates. Women’s participation profiles in these countries are similar to those of men, although slightly lower at each age. Austria, France and Germany also have an n-shaped profile, although participation rates are slightly lower on the whole, and considerably lower for younger and older women. The Nordic countries, France and Austria have the highest rates of public spending on childcare in the OECD, together with generous paid parental leave provisions (Jaumotte 2003). New Zealand, the United Kingdom, Australia and Switzerland all show a dip in participation around the peak childbearing ages. The United States has a hint of a dip, which has gradually flattened out over time (Jacobsen 1999). As in New Zealand, this dip is characteristic of women leaving the workforce when they have young children, and returning when their children are older. On the other hand, another group of European countries—many of them predominantly Catholic countries—have a participation profile which is consistent with a pattern of withdrawing from the labour force after marriage or childbirth and not returning, or returning intermittently, later in life. Eastern European countries tend to have a peak on the right-hand side (ie in the older age groups), which is striking but difficult to explain. The OECD (2002a) says that this is the result of increased barriers for women to participate in the labour market during the transition period from a planned economy, which appears to have disproportionately affected younger age groups. Finally, the Asian countries in the OECD, Japan and Korea, also display dips around the peak childbearing ages, although these dips are much bigger than in the Anglo countries, and women’s participation overall is much lower. What these groupings show is that participation profiles are not randomly distributed across OECD countries. Countries with similar profiles of labour force participation often occupy the same geographical region, or share similar national values, norms or histories. These parallels should not be overstated, and the match of profiles to countries is by no means perfect, but Figure 17 suggests that similarities in the labour force participation of women may be part of wider similarities between countries. 5.3 Employment and participation of particular groups of women in the OECD The data presented in Section 4.4 identified that qualifications, the presence and age of children, and being a sole or partnered mother, are important influences on women’s labour force participation. This section compares the employment or participation of different groups of women, defined by these characteristics, across the OECD. We consider, where we can, the employment or participation rate of different groups of women, and also their prevalence in the population. Both of these factors affect the overall participation of women in New Zealand compared to women in the OECD. For the most part, employment rates are reported here rather than participation rates, as this is how the data are presented in the various OECD publications. Employment rates differ from participation rates by not including people who are unemployed. Compared to other OECD countries, the employment of women with pre-school children in New Zealand is low (Figure 18a). Women with school-aged children, however, have reasonably strong employment rates compared to those in most other countries (Figure 18b). The difference in employment rates between New Zealand women with pre-school and with school-aged children (24 percentage points) is the third-highest in the OECD, behind only Japan and the Czech Republic. WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE 28 Figure 18 – Employment of mothers in the OECD, 2001 a. with youngest child aged under 6 0 10 20 30 40 50 60 70 80 90 100 Iceland Norway Sweden Denmark Portugal Finland Austria Netherlands Canada Belgium US France UK Ireland Germany New Zealand Greece Italy Poland Australia Spain Japan Czech Rep Employment rate b. with youngest child aged 6-14 0 10 20 30 40 50 60 70 80 90 100 Iceland Finland Sweden Norway Denmark Canada US New Zealand Czech Rep Netherlands UK Portugal Austria Ireland Germany Japan France Belgium Poland Greece Italy Spain Employment rate Source: Eurostat on-line data; individual countries’ statistical agencies; OECD (2002b) Indicator SS4.2. Note that the relative position of New Zealand in these two figures does not change if the defining age for a pre-school child is taken to be under 5 years in New Zealand rather than under 6 years. Furthermore, the employment of women with no children in New Zealand is up with the highest in the OECD (Figure 19). WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE 29 Figure 19 – Employment of women aged 20-49 in the OECD with no children, 2001 0 10 20 30 40 50 60 70 80 90 100 Iceland Canada Netherlands UK Norway Sweden Denmark New Zealand US Germany Austria Switzerland Czech Rep Finland Portugal France Australia Ireland Belgium Poland Spain Italy Greece Employment rate Source: Eurostat on-line data; individual countries’ statistical agencies; OECD (2002a) Table 2.4. Ideally, we would like to know what proportion of women in different OECD countries are mothers, and in particular what proportion are mothers of pre-school children. If New Zealand women are more likely to have children, especially young children, in their households compared to women in other countries, this would partially explain the low participation of New Zealand women of childbearing (and childrearing) age. Such information is difficult to find, however, so we use, as a proxy measure, total fertility rates across the OECD. This reveals that New Zealand has one of the highest fertility rates in the OECD (Figure 20). High fertility rates might explain some of the relatively low participation of younger New Zealand women; although it is also the case that Iceland has high fertility in conjunction with the highest participation rates in the OECD. Figure 20 – Total fertility rates in the OECD, 2002 0.00 0.50 1.00 1.50 2.00 2.50 United States Ireland Iceland New Zealand France Australia Norway Netherlands Denmark Finland Sweden United Kingdom Belgium Canada Portugal Austria Switzerland Japan Germany Italy Greece Spain Poland Czech Republic Total fertility rate Source: OECD (2005) Indicator GE3. The total fertility rate in a particular year is the average number of births a woman would have during her reproductive life if she were exposed to the fertility rates characteristic of various childbearing age groups in that year. Turning now to sole parenthood, New Zealand stands out for its low employment of sole parents (Figure 21). Differences between countries in the employment of sole parents are likely to be caused, in no small part, by the eligibility for, and generosity of, social welfare WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE 30 programmes for sole parents in the various countries. The OECD (2003a) suggests that the low employment of sole parents in New Zealand may be due to a lack of work requirements for sole parents, a relatively passive benefit system, relatively high benefit rates compared to the average wage, and high effective marginal tax rates. Figure 21 – Employment of sole parents, various years (1996-2001) 0 10 20 30 40 50 60 70 80 90 100 Sweden Switzerland Austria Portugal United States Italy Denmark Spain Hungary Greece Germany France Norway Netherlands Czech Republic Finland Belgium New Zealand Canada United Kingdom Australia Ireland Employment rate Source: OECD (2002b) Indicator SS3.2; OECD (2003b) Table 3.6; 2001 Census (New Zealand). New Zealand also has a high prevalence of sole parent families (Figure 22). In New Zealand, sole parent families make up over a quarter of all families with children: together with the United States this is the highest proportion of sole parent families in the OECD. Figure 22 – Sole parent families as a percentage of all families with children, 1996 0 5 10 15 20 25 30 New Zealand United States Iceland Canada Australia United Kingdom Norway Sweden Denmark Finland Belgium France Netherlands Germany Japan Austria Poland Ireland Czech Republic Switzerland Italy Greece Portugal Spain Percent of all families with children Source: Eurostat on-line data; individual countries’ statistical agencies; OECD (2002b) Indicator GE7.2. Given New Zealand’s position in Figure 18a and Figure 21 it is not surprising that New Zealand has a very low employment rate for sole parents who have a child aged under 6 (32%).19 This rate is the second-lowest in the OECD, ahead only of Australia (30%). By 19 Data for 17 countries is available from the OECD (2001). New Zealand data is from the 2001 Census. WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE 31 way of comparison, over 70% of Portuguese, Austrian and Italian sole parents with a child aged under 6 are employed.20 In general, New Zealand does not stand out from the rest of the OECD in an analysis of women’s participation by educational attainment. International comparisons are difficult in this area, however, because of the differences in educational systems between countries and because of the opacity of the International Standard Classification of Education (ISCED) which is commonly used to make comparisons between OECD countries. It does appear, however, that New Zealand women who have attained a degree have relatively low employment rates compared to women with a degree in other OECD countries, although there is little to separate most countries (Figure 23). This relatively low participation of New Zealand women with a degree may be due to the low participation rates of women with young children in New Zealand. Even mothers of young children who have a post-school qualification have relatively low participation rates in New Zealand (see Table 3). Figure 23 – Employment of women 25-54 with a university degree, 2002 0 10 20 30 40 50 60 70 80 90 100 Iceland Portugal Sweden Finland Austria Norway United Kingdom Netherlands Poland Ireland Denmark Czech Republic Belgium Switzerland Germany Australia Canada France United States Italy New Zealand Greece Spain Japan Employment rate Source: OECD Labour Market Statistics, 2002 data. A university degree is defined here as the attainment of ISCED level 5A or 6. 5.4 Accounting for paid parental leave The previous section has shown that differences between countries can be attributed in part to differences in the participation of, and prevalence of, different types of families. At least some of the difference between countries, however, might simply be due to the definition of “participation”. We illustrate this by comparing participation rates in New Zealand to those in the highest-participating countries in the OECD, the Nordic countries. In official statistics, women on paid parental leave should be counted as employed, even though they are not working.21 Nordic countries have amongst the most generous paid 20 The circumstances under which sole parent families have been formed—for example, whether they have been formed through marriage break-ups later in life, or because of teen parenthood—might have an effect on the labour force participation of sole parents. These factors could influence, to some extent, the participation of sole parents in different countries. 21 The OECD-ILO definition of employment says that “Persons temporarily not at work because of illness or injury, holiday or vacation,…maternity or parental leave [etc]… should be considered as being in paid employment provided they have a formal job attachment” and that “formal job attachment should be determined… according to one or more of the following criteria: (1) the continued receipt of wage or salary; (2) an assurance of return to work following the end of the contingency, or an agreement as to the WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE 32 parental leave provisions in the OECD but New Zealand has one of the least generous provisions (Table 8). Therefore at least some of the difference between the best performers of the OECD and New Zealand could be due to the unproductive “employment” of women on paid parental leave, who might equally be considered as not in the labour force, since they are not at work or available for work. Table 8 – Paid maternity leave provisions in Nordic countries and New Zealand Country Maximum paid maternity leave Sweden 78 weeks Denmark 52 weeks Finland 52 weeks Norway 48 weeks Iceland 26 weeks New Zealand 12 weeks Source: Social Policy Research Unit, University of York ( http://www.york.ac.uk/inst/spru/research/summs/welempfc.htm). The maximum effect of these differences in paid parental leave can be modelled by assuming that women take the maximum leave available for all their children and adjusting the reported participation figures to reflect this. Figure 24 shows that after this adjustment there is a marked ‘dipping’ in participation rates in the Nordic countries. When adjusted, the profile of women’s employment rates in these countries loses its n-shape, and becomes much more like the profile in New Zealand. The difference in participation rates of women aged 25 to 39 years is also markedly reduced after adjustment. Figure 24 – Women’s participation in Nordic countries and in New Zealand, raw and adjusted for maximum time on paid parental leave 0 10 20 30 40 50 60 70 80 90 100 15-19 20-24 25-29 30-34 35-39 40-45 45-49 50-54 55-59 60-64 Age group (years) Participation rate Nordics raw Nordics adjusted NZ raw NZ adjusted Source: OECD Labour Market Statistics, 2003 data; age-specific fertility rates from Eurostat on-line data. Participation rates for the Nordic countries are the mean, for each age group, for Iceland, Norway, Denmark, Finland and Sweden. date of return…”. It is difficult to know exactly how each country operationalises this definition, but a formal job attachment is usually taken as including paid parental leave, but not long periods of unpaid leave (Gruen and Garbutt 2003). Nordic official publications show that a large number of mothers of young children are employed but not actually at work during the week of the relevant survey. WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE 33 This may indicate that some of the difference in participation rates between countries, at least in the peak child-bearing ages, is an artefact arising from the way women on paid parental leave are treated in labour force statistics. WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE 34 6 Conclusion and discussion People who are not participating in the labour force are in a position where the net income to be gained from paid work is not sufficiently attractive to entice them into foregoing any of the other unpaid, but no less valuable, uses of their time. Labour force participants, on the other hand, are either employed or actively seeking work. New Zealand women’s labour force participation rates are lower than men’s at all ages and exhibit a characteristic “dip” in the younger age groups. Women’s participation rates have been rising since the Second World War, and each cohort of women has had a greater level of participation, at all ages, than their predecessors. Women’s participation rates are projected to rise further, at least in the older age groups, over the next decade. Among New Zealand women, the presence and age of children, being a sole or partnered mother and level of qualifications have a strong effect, and each factor has an effect which is independent of the other two. Mothers with different combinations of these characteristics have widely varying participation rates. At one extreme, for example, sole parents with a pre-school child and no school qualification have a participation rate of only 32%. At the other extreme, partnered mothers with a child aged 10-17 and with a postschool qualification have a participation rate of 91%. Clearly, in terms of labour force participation, women, even younger women with children, are a diverse group. Family factors also explain the dip in women’s participation: women aged 25 to 39 years have low participation rates because women in this age group are the most likely to have preschool children at home. Compared to other OECD countries, New Zealand has a relatively high overall female participation rate. Yet, participation rates for women between the ages of 25 and 39 are conspicuously low by international standards. Few other countries show a dip in participation rates in the peak childbearing ages. This point of difference seems to be driven by a combination, in New Zealand, of relatively low participation rates among mothers with young children and sole mothers, together with high fertility rates and high proportions of sole parent families. While New Zealand women tend to leave the labour force when they have children, they also tend to return strongly to paid employment when their children get older. The participation of New Zealand women with older children, and with no children, is relatively high compared to other countries and, as a consequence, New Zealand women aged 45 years and over have strong rates of participation compared to older women in other countries. This pattern of participation over the life-cycle may in part reflect a different life-course approach in New Zealand compared to many other countries: in New Zealand the balance between motherhood and participation might be achieved more by sequential rather than by concurrent patterns of paid and household work. In other words, women in New Zealand do not necessarily wish to combine work and childrearing at the same time, but to wait until their children are older before spending a good period of time in paid employment again. Alternatively, the participation of younger mothers in New Zealand may be low because of difficulties in accessing satisfactory childcare or suitably flexible jobs, or because the market wages that younger mothers can earn are not high enough to entice them into joining the labour force. WP 05/06 | WOMEN’S PARTICIPATION IN THE LABOUR FORCE 35 Differences in participation between countries may also reflect differences in government policies (such as tax and benefit policies) or social norms (such as the attitudes towards, and expectations of, women working compared to looking after their children). OECD countries can be grouped according to their pattern of women’s participation across ages, and these groupings to a considerable extent reflect similarities in the countries’ values, social conventions, institutions and recent histories. Not surprisingly, New Zealand’s profile is most similar to Australia and the United Kingdom: countries with which we share a common heritage. The participation profile of New Zealand men, relative to the OECD, is also similar in many ways to that of New Zealand women, with relatively high participation rates for younger and older people, but relatively low rates for people aged around 25-39. These similarities support the case for the existence of particular “country effects”, which affect both women and men. Finally, as illustrated in Section 5.4, the differences in participation between countries might to some extent be spurious, reflecting the definition of participation, or each country’s interpretation of that definition. It is difficult to know how big this effect is, although the maximum effect is marked. How does all this inform the public policy question of whether, and how, to encourage the participation of New Zealand women? Some initial thoughts are hazarded here. Firstly, since different groups of women, and mothers, have widely differing participation rates, any policies which aim to increase the participation of women would need to be carefully focused. One type of policy is unlikely to work for all women. Also, since some groups of women already have high participation rates, policies which aim to increase this participation even further may incur high deadweight costs. The analysis in this paper also shows that the presence of pre-school children is a key factor in explaining women’s participation in New Zealand, and compared to other countries. This reinforces the fact that the desirability of time spent with young children versus working is central to policy considerations. Such considerations need to go beyond the simple participation/non-participation analyses presented in this paper, and think about the time spent at work and at home by both parents combined. As Figure 1 shows, the hours that New Zealanders work, relative to the size of the working age population, are already among the highest in the OECD. In the short-to-medium term, overall participation rates for women in New Zealand are likely to rise without any change in policy settings, although there may be some fluctuations according to the state of the economy. Policies which aim to increase women’s participation may be pushing a rock downhill. Historically, women’s and men’s participation rates in New Zealand have been gradually converging, as have been women’s participation rates in the OECD. It would be no surprise if these trends continued in future decades. Finally, we need to consider how much to concern ourselves with differences in participation rates between countries. Some of these differences are rather fine and perhaps even spurious. Some are more considerable, such as the difference between the participation of young women in New Zealand and in the Nordic countries. However, although this “gap” clearly shows that an increase in participation is possible, it says nothing about the costs and benefits of increased participation. Also, it may not be straightforward to quickly imitate the patterns of participation which other, very different, countries demonstrate. Women’s participation rates as high as those in the Nordic countries are not likely to be achieved through one or two policy instruments, but may require sustained social change over a longer period of time.