CHANGES AND CONTINUITIES IN MATERNITY POLICIES: Comparison of maternity legislation in Estonia and Ireland

This paper draws on path dependency theory to explore continuities and change in maternity protection legislation in relation to time (leave) and money (benefits) in Ireland and Estonia in an attempt to contribute to comparative literature on maternity policy. Ireland and Estonia are two interesting countries on which there are few comparative studies available. There are several reasons for choosing to compare maternity protection legislation in Ireland and Estonia. The choice of countries allows for the consideration of the relationship between religious ideas and political ideologies and the development of social policies. Although both countries are currently experiencing the effects of the worldwide economic downturn both societies have changed unrecognisably in the last 80 years since they achieved independence, Estonia in 1918 and Ireland in 1921. Ireland has changed from an economically depressed, inward-looking protectionist state to an outward-looking open economy and member of the European Union. Estonia experienced a brief period of independence, followed by 50 years of Soviet occupation and re-independence which then enabled the country to become a member of the European Union. Both are small countries on the periphery of powerful colonising eastern neighbours. Both countries inherited the germs of social policy from their former overlords: in Estonia the valid legislation from the 1880s until the 1920s was that of Imperial Russia; Ireland built its welfare state on the foundations laid by the British administration (Poor Relief (Ireland) Act 1838) and in the twentieth century drew on the Beveridge male breadwinner model. Following political independence both countries had to deal with war; widespread poverty; the prevalence of TB; and a high incidence of unemployment. Women in both countries gained suffrage in 1918. Drawing on path dependency theory this article explores critical junctures which have shaped maternity policies for the future (Greer 2008). It shows that in both countries, despite major changes and ‘explosions’ continuities persist.

While maternity policies include complex spheres such as health and childcare (de Vries et al.2001; Kennedy 2002 ,2004) this paper is concerned with women's relationship with the labour market and the extent to which maternity payments in both countries allow for women's de-commodification (ability to exist financially outside the labour market) for varying periods around the time of childbirth. Thus it is concerned with time and money; the sphere of maternity protection legislation directly related to leave and benefits. Throughout the paper ‘maternity’ relates to the period from conception to lactation, the period concerned with the physiological process of conception, pregnancy, birth and nursing. This period is common to all mothers; therefore the state's regulation of this period and the state's investment in it is an interesting subject. Maternity protection policies are embedded in legal frameworks which dictate boundaries between the allowed and the prohibited. Even though these lines are not always followed, they have an impact on how the ideal is perceived and are linked to the principles upon which welfare is organised. In Estonia and Ireland these have been very different, influenced undoubtedly by religious ideas. In the former, there has been the Lutheran ideal of a working individual, strengthened further by the communist ideal of woman as citizen worker and mother, on the other hand, in Ireland one sees the enduring influence of Catholic social teaching and the view of woman as mother first and worker second.

Changes and continuities in social policy have been discussed widely by ‘path dependency’ theorists, emphasizing the historicity of policy making (Room 2008). Present policies have emerged from a historical context whose structures influence present resources and prospects for shaping viable political strategies. The start of a path, taking place in a specific type of situation, often called a ‘critical juncture’ or breakage point, or ‘explosion’, is often crucially important for subsequent policies. In these critical junctures or explosive moments the development curve transforms into a new trajectory (Lotman 2001 [1992]). These moments are likely in periods when the very structure of the state is at issue and undergoing transformation. Particular policies may be advocated or opposed for reasons other than their perceived relevance to socioeconomic needs or cultural ideas (Orloff and Skocpol 1984). Nevertheless, even the most powerful and deep explosions do not touch all the complex richness of social layers, despite the central structure undergoing powerful and catastrophic explosions with far reaching consequences. Certain values of the previous period are maintained, and sometimes shift from the periphery to the centre (Lotman 2001 [1992]). Explosions or critical junctures are not the only pathways to something new, since in actual life most cultural and social areas develop incrementally. It is possible that actors voluntarily take the risk that they may be creating inefficiencies in the present, for they think that such steps are required to create new futures (Pfau-Effinger 2005). These very interesting issues are explored here with reference to maternity protection in Ireland and Estonia.

Independence for both countries could be reasonably viewed as a critical juncture. The very structure of the state in both countries was at issue and opportunities for development of policies in specific directions were possible. The leadership in post-Independent Estonia had strong liberalising inclinations: Estonia's first act in the Manifesto to the Peoples of Estonia (the Declaration of Independence) was to abolish all ethnic, class and gender divisions. This was a definite departure and one which was to have implications for the future. Estonia adopted laws for work protection, and the creation of insurance against accidents and illness (EV kodanikkudele 1919) investing in what were modern social policies of the time, modelled after Bismarckian policies that emphasised the importance of an individual's work history in their entitlement to state provided benefits (Cousins 2005: 80). The greatest problem in the political life of the 1920s was the frequent change of governments: their average life between 1919 and 1933 was 20 days and eight months (Rauch 1987). This became especially problematic when depression hit Estonia, causing the devaluation of the local currency and high unemployment. This was accompanied by political unrest, resulting in a coup by which Estonia changed from a parliamentary republic into a presidential one where the president was empowered to appoint and dismiss the government. All political assemblies and parties were banned. Though Estonia of the late 1930s was economically successful, in political terms it was a ‘silent’ era and conservative nationalism was on the rise, emphasizing the ‘natural roles’ of the sexes and much of the liberal spirit of the 1920s disappeared from the legislation. Thus, we see, despite the explosion of independence, there remained continuity.

Like Estonia in the 1930s, conservative nationalism was on the rise in Ireland. Despite the explosive moment of Independence, the period experienced continuity in terms of policy, political ideology and the role of religion. Kennedy refers to the period between Independence and 1970 as displaying many of the characteristics of a conservative/corporatist regime: strong Church influence, status-differentiated social insurance, underdevelopment of day care and other family services and the principle of subsidiarity was strong (2002: 35). Kahl explains the principle of subsidiarity: ‘helping the poor was a responsibility for the local Christian community and should arise from compassion rather than legal force through the state’ (2005: 101). Social policies not only reflected Catholic social teaching but were shaped by other dominant patriarchal institutions such as the Government, the civil service (from which women had to retire on marriage between 1929 and 1973), the medical and legal professions and the trade unions (Valulius 1997). Lee indicates that post-independence Ireland was an underdeveloped, traditional and mostly rural society; 53 percent of the workforce was employed in agriculture (1989). In 1926, 60 percent of women at work were either in farming or domestic labour, with 10 percent in industry (Daly 1978: 71). One commentator argues that failure to succeed economically was due to the fact that the new leaders were primarily literary men who depended for economic advice on conservative and cautious advisors (O'Hagan 1975: 7). Throughout the 1920s a series of legislative measures particularly relevant to women's lives were introduced (Valulius 1997). The conservativism is clear in such reports as the 1927 Report of the Commission on the Relief of the Sick and Destitute Poor including the Insane Poor (Saorstát Eireann 1927). Presented within is a differential motherhood: mothers who are married; widowed; unmarried, the latter group further divided into those ‘who may be considered amenable to reform’ and those ‘regarded as less hopeful cases’ (1927: 68). The latter category included married mothers who had children ‘not the offspring of their husbands’. Throughout the 1920s, restrictive employment practices were enshrined in law. These included the ‘marriage bar’ of 1929 which ensured women had to retire from employment in the public sector/civil service on marriage.

Estonia has been a protestant (principally Lutheran) country since the Reformation, whereas traditionally Irish society was regarded as synonymous with Catholicism, whose values were incorporated into the Irish Constitution of 1937 (Bunreacht na hÉireann). Looking at the relationship between religion and policy Kahl (2005) suggests that since the roots of modern social policy are in poor relief, attitudes to begging and alms-giving in different Christian denominations have strongly shaped policies in European countries. In Catholicism poverty had a positive connotation and poor relief belonged to the sphere of the Church. Hence in Ireland the Catholic Church provided social services long before the foundation of the Irish state and was particularly powerful in the spheres of education, healthcare and social services (Powell 2002; Kahl 2005: 114). However, the principle of subsidiarity was strong with the effect that motherhood and reproduction were very much a private issue; while the state interfered it was in relation to preventing the delivery of services and policy rather than their development. For example, the Catholic hierarchy intervened in 1951 when there was a political move towards introducing universal medical care for mothers and children in 1951 which had long-term implications (Kennedy 2004). In Ireland from the 1930s onwards the Government began to take an interest in social education with the introduction of policies geared towards controlling young people and promoting very traditional values (Kennedy 1984 ,1999). For decades, State and Church worked together to reinforce a gendered ideology that believed a woman's place was in the home and her primary role was that of wife and mother, dependent on a male breadwinner. The 1937 Constitution was seen as a perfect opportunity to promote Catholic social teaching and to control the citizens of Ireland, and in particular women. In the 1937 Constitution the ‘family’ was given a special status in Irish society and undoubtedly echoed Papal teaching ‘… let women be subject to their husbands as to the Lord, because the husband is the head of the wife, as Christ is the head of the Church’ (Castii Conubii 1930: 13). The Constitution delineated the role of family and woman in Irish society in Article 41: the State recognised the family as the natural primary unit group of society, and as ‘a moral institution possessing inalienable and imprescriptible rights, antecedent and superior to all positive law’. This status of the family hinged to a large extent on ‘woman’ or ‘mother’ who was also endowed with a particularly ‘favoured’ status, which she was expected to live out within the home (Article 41.2). One sees the enduring influence of this ideology in Ireland where there is unpaid parental or paternity leave, and childcare is perceived as a family, in fact women's, issue.

Lutheran ideology, on the other hand, places great emphasis on the value of labour: ‘work became an intrinsically positive activity that was pleasing to God’ (Kahl 2005: 102) and was also a means to overcoming poverty. Parallels are found in Soviet thinking. Lutheran countries were also the pioneers in introducing welfare legislation, starting with social insurance in Germany which ‘was motivated by Bismarck's explicitly Lutheran notion of state activity’ (Kahl 2005: 105; Cousins 2005: 80). Estonia, even when a part of the Russian Empire, had been under strong German influence since the upper class was Baltic German. In twentieth-century Estonia, the Church as an institution has had only indirect impact on policy making since the Church was separated from the state in 1918.

The period from Independence to the end of the 1950s in Ireland was a dark era, marked by poverty, economic stagnation and emigration (Ó Gráda 1997). However from 1958 there was a fresh focus on economic expansion and a new era of economic planning began. The 1960s saw the introduction of free secondary education, the growth of the civil rights movement and the emergence of the women's liberation movement which organised around reproductive rights. Ireland joined the European Economic Community (EEC) in 1973. This has been of paramount importance in the area of social policy and it is in this context, in a climate of change and development that policies from the 1970s onwards began to develop in a more gender conscious fashion. This could be viewed as a ‘critical juncture’ leading to change, but simultaneously continuities persisted, perhaps as a result of a lack of political participation by women and a women's movement which remained very much on the periphery of Irish society. While there was change in the public sphere, there was continuity in the private sphere, where the strong male breadwinner type economy persisted, with the assumption that woman as wife and mother would fulfil the role of primary caregiver – a view also supported by the Catholic Church. As Ireland has moved towards a dual worker economy, this situation still persists with childcare continuing to remain a private issue and women having to cope with the dual burden of work within and outside the home (UNICEF 2008). In contrast Estonia's accession to the EU, while significant, was less so in the context of other major changes which had occurred.

A significant change in Estonian life was Soviet occupation in 1940, followed by Nazi occupation in 1941–1944, to be replaced by Soviet occupation again, lasting until 1994.¹ Estonia had to adopt the legislation of the Soviet Union which in terms of social protection was strictly embedded in an individual's labour history (and thus close to the Bismarckian model of social policy).² In terms of mothers’ rights it was making a step backwards (see the discussion on rural women's maternity protection below). It would be correct, however, to divide the Soviet period into two: the Stalinist era (ending in 1955/1956) of constant tension and suppression, and the post-Stalin period until the late 1980s. The latter 30-year period was an expansion of the legal reforms made in 1955–1956, and social policy of that period remained unchallenged until Gorbachev launched perestroika and glasnost, opening a door to the next explosive period. The collapse of the Soviet Union in 1991 was a cataclysmic event, touching the majority of social structures and causing the total remaking of the existing legislation.³ Compared to this, the EU approximation and later accession in 2004 were not as prominent, since these formed a part of the state's priorities from the very beginning. Thus, membership of the EU would appear to have greater significance for Ireland than for Estonia in terms of acquiring more generous maternity protection for women. In Estonia maternity protection has been one remaining ‘continuity’, despite explosions in political life and in major policy fields.

The late nineteenth century ‘discovered’ poverty and became concerned about high levels of infant and child mortality, poor working and living conditions. Throughout history, adult female mortality has largely been associated with childbirth (Shorter 1984 [1982]). Protecting the health of mother and child, and guaranteeing the job security of the pregnant worker were important considerations in state regulations concerning the right to take time off from work to recover from giving birth (de Vries et al. 2001). In most Western countries the first legislation relating to maternity leave was adopted between 1900 and 1915, and it was mostly unpaid, designed as a means of avoiding severe effects on children's health (Gauthier 1996: 50). From the very beginning this leave has often been divided into two parts: before and after delivery, made available on a national level at first to women working in industry. Civil servants were included later, and women working in agricultural sector were the last ones to receive the benefit (Gauthier 1996). The latter pattern was also true for Estonia.

In Estonia maternity leave has meant both the right to a period off work in connection with parturition, as well as a compensation for this period. However, until the re-independent period women were divided into different classes and therefore gained different support for maternity. From 1913, women working in industry had a right to maternity leave, and employers had to cover women's parturition assistance and hospital stay (Johanson 1920). In 1920 maternity leave was also granted to civil servants (four weeks before expected parturition and six weeks after) with full income replacement, supported by a birth grant (which was paid out even if the child was stillborn, though at a smaller amount) (Riismann 1924). As mentioned above, labour protection for rural workers, including that of maternity, was legislated later, thus maternity leave to rural Estonian women became available only in 1939 (Kaelas 1939). At the same time women's childbearing capacity was protected, still only in industry, by prohibiting women from working in underground mines, and performing certain tasks, such as works connected with transmission (Riismann 1924). Labour legislation of 1937 extended protective measures for women by prohibiting them from lifting heavy items (this has remained unchanged until the present day). Since 1939, legislation required the agricultural sphere to take pregnancy into consideration when allocating tasks to women, and paid maternity leave was also instituted (Kaelas 1939).

As mentioned before, Estonia modelled its social security according to the Bismarckian type that valued an individual's labour history. However, at the time of Soviet occupation (both in 1940 and 1944), Estonia had more extensive maternity protection than the Soviet Union in which women working in agriculture were not legally protected at all, and the length of maternity leave for workers and officials was a week shorter (being nine weeks). Though in 1944 maternity leave was extended to 11 weeks with full compensation, it did not apply to rural women.⁴ It was only after 1949, when Estonia was forcefully collectivised, that maternity leave became available to women working in kolkhozes or collective farms,⁵ by granting the right to 50 percent of the former pay for a month before and a month (ca. 60 days) after parturition (Kovrigina 1947: 22). In the Soviet Union all working-age citizens were expected to work: the slogan being ‘who doesn't work, shouldn't eat’ (compare with Lutheran slogan ‘No one should live idle on the work of others’ based on 2 Thessalonians 3:10; see Kahl 2005: 104); however, the state was very interested in population growth, hence there were several legislative measures taken to protect working women's childbearing capacity. Pregnant and nursing women were under the protection of the law and could not be dismissed. After the fourth month of pregnancy and while nursing, women could not work overtime or in night shifts. (Note: night work for all women had been outlawed both in the Imperial Russia and in the inter-war Estonia even though it existed in reality.) Neither could women be sent on business trips outside their permanent place of residency without their agreement (Kovrigina 1947: 22). The state also made efforts to better reconcile maternity and labour market by ordering enterprises to organise nursing rooms for breastfeeding mothers, and time taken for nursing was to be considered working time (Sepp 1946). Workplaces had to allow nursing mothers additional breaks for feeding children, for both nursing and artificial feeding (Kovrigina 1947; Pergament and Stavtseva 1964). In addition, several workplaces organised crèches and kindergartens for the children of their workers. From the age of one month, children could be taken to crèches; and in addition to ordinary day-care centres there were ‘long-day groups’ in which children were cared for five days a week, going home to their parents only for weekends.

As mentioned above, the major breakage point in the Soviet society was marked by the death of Stalin in 1953, followed by considerable modification of the legislation that remained basically unchanged until the collapse of the system. In 1955 paid maternity leave was extended from 11 to 18 weeks (the compensation level of which depended on former work history). In 1956 women could request three months additional leave without pay. Another shift on the legislative level happened in 1980, possibly caused by the chronic lack of ‘working hands’ – a common problem in the low work efficiency Soviet Union, and also the demographic concerns with a decrease in fertility (Kon 1995). Steps were taken to facilitate part-time work⁶ for mothers with young children on condition that their working week lasted at least 20–24 hours, without any restrictions on their annual holidays or other labour rights (Nirk 1989: 37, 39); four years later flexible work-time was introduced. The Labour Committee had to list professions and jobs in which it would be most efficient and rational to employ mothers with small children, and find ways to organise working at home for non-working mothers with children (Nirk 1989). As previously mentioned, in the Soviet Union women were workers first and mothers second. The right to a job was guaranteed in the Constitution and motherhood was no excuse to refrain from paid labour.

Despite the major ‘explosion’ – the collapse of the Soviet Union and the socialist outlook in general – major continuities in maternity protection have prevailed. Labour legislation in post-Soviet Estonia was moulded after the Scandinavian social-democratic and labourer protective model, and so far maternity has been sacrosanct. One of the reasons for this may be the low birth rate, but it would also be very unpopular in the light of mother-supporting political rhetoric of all political parties. Therefore most of the Soviet labour protection for women has been left intact or slightly modified in Estonia. In order to guarantee safe work for pregnant and nursing women the employer has to facilitate temporary work conditions; change the order of work, including shortening of the working day and facilitating periods of rest; and to transfer the woman temporarily to lighter work or another position. If the latter is not possible the employer has to give the woman temporary leave, the duration of which is determined by the physician. Night work is not allowed for pregnant women and if night work is likely to damage the health of a woman or her breastfed child, the employer is obliged to transfer the woman temporarily to day or evening work. If this is not possible the woman must get temporary leave (Töötervishoiu1999). Enterprises and workplaces still have to guarantee a safe work environment for pregnant and nursing women; the requirements foreseen by the decree are compulsory to the employer in cases when the female worker presents a medical certificate. The employer has to give free time from work and continue to pay, if the pregnant woman's ante-natal appointment has to take place during working hours. Employers are required to guarantee the use of a room for resting for pregnant women or for female workers who have given birth within the previous three months but who are not breastfeeding their children (Töötervishoiu ja tööohutuse nõuded 2001, § 2, 5–7). The law requires that additional breaks for feeding a child should be granted to a person who is raising a child under 18 months old. In addition to the general break for ‘resting and eating’ they should be granted a break at least every three hours which should last no less than 30 minutes. If the woman has more than one child younger than 18 months, the breaks should last at least an hour. If the woman chooses, these breaks are added to the ‘resting and eating’ breaks, or her working time is shortened correspondingly. These breaks are considered worked time, and they are paid by the average salary out of the national social security budget (Töö-ja puhkeaja 2001). In summer 2009 the former work and leisure time legislation was replaced by a Work Contract Act which made labour more ‘flexible’, that is easier to hire. Nevertheless, provisions for pregnant women and mothers with children under three years old are protected in the same way as before (Töölepingu seadus 2008).

As already mentioned, labour legislation and especially the provisions for maternity have not undergone radical changes. Healthcare, on the other hand, has been totally reorganised and is connected to one's employment status; nevertheless, pregnant women from the 12th week of pregnancy are covered, whether working or not⁷ (Ravikindlustuse2002). The maternity leave has been lengthened by 14 days, thus it is 140 days in toto, of which at least 30 days must be given before expected parturition. The replacement level is 100 percent of the former average income (Ravikindlustuse2002), whereas in case of illness it is less. In 2003 mothers were granted additional fully paid six months of leave to top off the maternity leave, and either parent can then take fully paid leave for five months. Even the economic depression has not affected maternity and child-care leave provisions, though it hit the pension system and various other benefits connected with child rearing. Those who were not employed before parturition are paid a monthly benefit at the amount of official minimum wage. In addition to this there are birth grants and each local authority is free to make additional provisions to the inhabitants within its community. This means that even though the healthcare system as a whole has taken a new path, the sections dealing with provisions for maternity have followed the old one. Both Soviet and the present maternity policies are of strongly pro-natalist nature, encouraging women to have children by offering benefits to take time off in order to have a baby.

Maternity payments were introduced in Ireland for the first time in 1911 (1911 National Insurance Act) when maternity benefit of 30 shillings was paid to an insured worker or to the wife of an insured male worker as a one-off payment to meet the extra costs associated with childbirth (the 2009 value is €184.70). Barrington states that ‘this relatively large sum was to pay for the extra expenses of childbirth and to encourage working mothers to remain at home with the child in the early months’ (1987: 76). However, there was no right to leave to accompany the payment. The grant was payable either on the husband's or the wife's insurance records and the contribution requirements were easy to satisfy. This payment, renamed ‘maternity grant’ under the 1952 Health Act, was abolished in 1983. In 1953, a contributory ‘maternity allowance’ was introduced, payable for six weeks before and six weeks after confinement, to an insured woman who met the contribution conditions ‘to relieve her of the necessity of working immediately before and after the confinement’ (Farley 1964: 100). This is interesting because it separates maternity payments from men's insurance contributions for the first time. It was payable on the woman's insurance only. A married woman had to have at least 26 contributions paid after her marriage before she could qualify. Thus, it distinguished between different groups of mothers and is interesting when noted that from 1929 to 1973 the majority of women in Ireland had to retire from employment upon marriage. However, there was still no entitlement to leave. There was an assumption that leave was not an issue as the woman was not expected to be in the labour market but in the home. In October 1973, the rule which disqualified women, upon marriage, from receiving short-term social welfare benefits until a further 26 contributions had been made was removed. Women who were outside the labour market could claim this payment as it was based on past contributions and not on current employment status. The Social Welfare (Pay Related Benefit) Act 1973 provided for the introduction of a pay-related benefit scheme which was an earnings-related supplement to the flat rate maternity allowance. It was introduced with effect from April 1974.

Any developments in maternity payments in Ireland in recent decades have been as a direct result of EU Directives on maternity protection. In Ireland maternity leave (time) and maternity payments (money) were separate up until 1981 when the Maternity Protection Act and the Social Welfare Act addressed the same issue: time off for women prior to, during, and after childbirth and financial support during the period. This is a change in Ireland because up until then the focus was primarily on payment. It is undoubtedly linked to Ireland joining the EEC in 1973 and the subsequent introduction of employment equality legislation. This was only eight years after the removal of the marriage bar. While the introduction of maternity leave was a critical juncture for Irish society, being the first time women had a right to leave from paid employment and a right to return after childbirth, there was also continuity in that the wage replacement was linked to social insurance as has been the pattern with maternity payments since they were first introduced. In 1981 the Social Welfare Amendment Act introduced a ‘maternity benefit’ for women in employment. This was a new pay-related maternity allowance scheme and applied to women covered by the provisions of the 1981 Maternity Protection of Employees Act. The allowance was paid for 14 weeks, initially at a rate of 80 percent of earnings, later reduced to 75 percent and then 70 percent. For low earners there was a minimum weekly payment based on average earnings of female workers in employment insurable for pay-related benefit purposes. In 1991 under the Worker Protection (part-time regulations) Act maternity benefit was extended to include part-time workers earning over a certain threshold. This again was a result of an EU Directive and so was a development imposed from outside of Ireland which leads us to believe that joining the EEC in 1973 was an explosive event in terms of maternity legislation for Ireland, and more generally for gender equality in Ireland. This coincided with another major change when, under the Social Welfare Act, the maternity allowance for those outside the labour market was abolished. This was a critical juncture as it moved maternity policies in the direction of being linked to labour market involvement, and a shift from mother to worker rather than dependent. There was an end to payments for women outside the labour market, marking a major change in the recognition of pregnant women only when workers. This pattern continued until 1986 when the Social Welfare Act legislated for the rationalisation of the maternity scheme for women in employment. Pay-related and flat rate elements were amalgamated and accounted for under the ‘maternity allowance’ heading. The ‘maternity allowance’ scheme continued to apply to women who did not qualify for maternity leave under the Maternity Protection of Employees Act 1981. In 1988, eligibility criteria became more onerous and new claims under the general maternity allowance scheme were required to have 13 weeks paid contributions during the governing contribution year to ensure that women had recently been at work. Once again, this demonstrates a shift in ideology towards the dual role of woman as worker and mother. This was an obvious policy departure which militated against women outside the labour market and led to a huge reduction in the numbers eligible for maternity payments. Entitlement to pay-related benefit was abolished for all new claims to general maternity allowance in 1988, and from 4th April 1988, three ‘waiting days’ were introduced for women transferring to disability benefit from maternity allowance.

It is essential to emphasize that in the 1990s Ireland experienced an unprecedented economic boom, which earned it the title of ‘Celtic Tiger’ (analogous to the Asian Tiger economies). By that time Irish women were better prepared to enter the labour market in terms of education than in earlier decades and they were further encouraged by the social partners (a feature of corporatist policy making) who were concerned with the prospect of labour shortages. In addition, increasing house prices and smaller family size have ensured that more mothers have entered the labour market. Irish women's entry into the labour market accelerated. Ireland, with a figure of 58 percent female employment has already passed the Lisbon interim target of 57 percent. However, Irish women in employment are working some of the longest hours in Europe (NESC 2005: 93). In 2000, 30 percent of women in employment exceeded 48 hours a week as against an EU average of 24.3 hours. NESC (2005: 45) indicates that ‘the growth in women's employment, particularly by mothers with young children, is placing substantial strain on this family based model of care’. While there has been a move towards a dual earner model, an ideology persists which postulates that childcare provision is a private issue, ensuring that Ireland has one of the lowest levels of public childcare in the EU (UNICEF 2008). Childcare facilities are underdeveloped and there is no tax relief on childcare. Unlike in Estonia, Irish mothers do not enjoy extra breaks to rest during the working day. Maternity leave was further extended to 22 weeks in 2006 with an optional eight weeks unpaid leave. These were further extended in 2007 to 26 weeks and 16 weeks, respectively. Thus, once again we see a pattern of differential access to leave for women depending on their financial resources with those in a position to do so able to take additional unpaid leave.

Looking at the idea of path dependency in relation to maternity legislation in Ireland, it seems that changes have come about primarily since the 1970s when Ireland joined the EEC. Since then any changes seem to have been as a direct result of EU Directives and so imposed from outside the country. Continuities have persisted in terms of Ireland remaining a country where the principle of subsidiarity still prevails; childcare is still perceived as a private and in fact a woman's responsibility. The 1937 Constitution remains unchanged and the role of woman in the family is still delineated a particular place. So, while women are no longer expected to remain in the home as dependents of male breadwinners, they are nevertheless not supported in their dual role as mother-worker. This is a continuity in Ireland and has been further reinforced in recent years as Ireland has become increasingly neo-liberal in its political structures. Table 1 presents a summary of the maternal labour legislation in Ireland and Estonia.

This paper looks at changes and continuities in maternity policies in Ireland and Estonia. Both countries appeared onto the world map at almost the same time, but since the religious, cultural and political context of their emergence was very different, their social policies, especially those concerning women, have been contrastive. Estonia arose into the political arena via Russia's 1917 revolution, had strong liberal and social concerns, including the promotion of social protection for its population. With the Lutheran work ethos and the official abolition of all class and gender differences, working hands were needed both in industry and the civil service: there were no strong incentives to push women out of paid labour. Hence maternity protection, already existing at the time of the Republic's birth, was gradually broadened. Soviet occupation, even if socialist in name, was initially a step backwards, especially for rural women, but the death of Stalin and the subsequent re-modelling of legislation corrected the situation. In the Soviet Union, adopted lines of policy were resistant to change; however, maternity policies have proven to be robust even at the times of transition. The collapse of the Soviet system and strong neo-liberal political winds have not caused deviation from the already well-trodden path of maternity protection: in the workplace the rights for (paid) breaks during pregnancy and nursing have been maintained, although women also have a right to stay extensively at home and not work. All women are entitled to maternity benefits, irrespective of their former work status, even though in present day Estonia most social rights have a direct connection to an individual's work history. The structures of past policies influence strongly what is feasible and desirable in the present. Maternity protection has adhered strongly to former regulations even if the surrounding policies (such as healthcare and child policy) have undergone dramatic changes. It seems that if work is highly valued in a society and women have a long history of being treated as both workers and child bearers, maternity protection is perceived not only as a social right but an inalienable human right.

In Ireland, the architects of the Irish Republic were Catholic conservatives. The close allegiance to Catholic social teaching was reflected in the 1937 constitution and in legislation of the period. There was a pattern of restrictive legislation in relation to women's participation in the paid labour market. The male breadwinner model was propounded with women's role was in the home both as wife and mother. Childbirth was favoured only within marriage. Change came about slowly and was linked primarily to Ireland's entry into the EEC in 1973: a critical juncture. In the following decade maternity leave and income replacement was established as a right and in subsequent years leave and accompanying payments have become more generous but at the same time clearly linked to women's position in the labour market. There is a stratification of women with some, i.e. those with more financial resources, able to take more leave (unpaid) than those constrained by their financial situation. Ireland's economic boom from the mid-1990s was accompanied by an increased demand for female labour market participation. However, there remained strong features of the traditions inherent in Irish social policies, with the strong characteristics retained; for example, differentiated social security, underdeveloped childcare services and poor maternity entitlements. The assumption that childcare is a woman's responsibility still exists as evidenced by the absence of paid parental and paternity leave.

Employing ‘path dependency’ theory to explore maternity protection policies in Ireland and Estonia is a useful exercise. It demonstrates the existence of changes and continuities in social policy as discussed by ‘path dependency’ theorists that emphasise the historicity of policy making. It shows how present policy makers are in a sense constrained by the environment designed by past policies – both in Estonia and Ireland the maternity legislation has been rooted in their historical frames. We have examples of how the start of a path, taking place in a specific type of situation is often crucially important for subsequent policies – in Estonia maternity protection has been ‘there’ from the very beginning whereas in Ireland the lack of maternity protection has been challenged only by relatively recent changes connected with EU accession. In these critical junctures or explosive moments the development curve transforms into a new way from that of hitherto. These moments are likely in periods when the very structure of the state is at issue, undergoing transformation, and therefore particular policies may be advocated or opposed for reasons other than their perceived relevance to socioeconomic needs or cultural ideas. It also demonstrates that even the most powerful and deep explosions do not touch all the complicated richness of social layers, despite the central structure experiencing powerful and catastrophic explosions with far reaching repercussions. Certain strongly rooted values of previous periods, like religious ones, will be maintained: in Estonia it has been the value of working mother, whereas in Ireland it is the value of woman as a homemaker and principal child carer.

The research has been supported by the Graduate School on Cultural Interaction and Integration and the Baltic Sea Region, Finland and the Department of Social Policy of the University of Turku. Further support has been obtained from Open Estonian Fund. The authors would like to thank the referees for their very useful comments on earlier drafts of this paper.

Patricia Kennedy is a Senior Lecturer in Social Policy in the School of Applied Social Science, UCD.

Heili Einasto was a research fellow at Turku University, Finland, between 2003–2005, and is currently a lecturer at Tallinn University, Estonia.