MASARYKOVA UNIVERZITA Fakulta sociálních studií Katedra sociologie

Mgr. Lada Železná

Intergenerational solidarity and the influence of informal transfers on fertility behavior

Disertační práce

Školitel: prof. Martin Kreidl, Ph.D.

Brno 2018

I declare that the thesis is the result of my own work and I have cited all my sources.

Brno, 20/04/2018 ……………………………….

TABLE OF CONTENTS

INTRODUCTION ...... 3

1. Dimensions of intergenerational solidarity ...... 7

1.1 Individual factors influencing intergenerational solidarity ...... 8

1.2 Structural factors influencing intergenerational solidarity ...... 14

1.3 Life course approach ...... 17

1.4 Intergenerational solidarity: summary ...... 19

2. Fertility in Europe ...... 20

2.1 Fertility, gender equity and conflicting roles of parents ...... 26

2.2 Fertility and public policies arrangements ...... 38

2.3 The intergenerational solidarity and fertility ...... 41

4. Methodological approach ...... 46

5. The effect of normative intergenerational solidarity on the probability of childbirth ...... 50

5.1 Latent solidarity and the operationalization of the term ...... 52

5.2 Model of the relationship between grandparents’ obligations perception and grandchild’s birth 60

5.3 Summary and discussion: the latent support from parents as an encouraging factor to fertility .. 70

6. The effect of income on childbearing intentions ...... 72

6.1 The association between economic factors and fertility ...... 74

6.2 Education, fertility and family gap ...... 76

6.3 Data and research questions ...... 78

6.4 Measurements of fertility intentions and income ...... 78

6.5 Control variables ...... 84

6.6 Basic models of fertility intentions and subjective income...... 86

6.7 Fertility intentions and alternative indicators of financial situation ...... 90

6.8 Fertility intentions and interaction of subjective income and education ...... 92

6.9 Fertility intentions and interaction of subjective income and country ...... 94

6.10 Fertility intentions and interaction of income and order of the child ...... 98

6.11 Summary and discussion: Fertility intentions and subjective income in interaction with education, country and order of the child ...... 100

7. Financial transfers from the family and its effect on the fertility behavior: the role of child´s order 102

7.1 Financial transfers as a form of intergenerational solidarity ...... 104

1

7.2 The financial resources and fertility ...... 105

7.3 Models for the effect of financial transfers on the fertility behavior ...... 108

7.4 Summary and discussion: the parents as a source of money before and after the childbirth ..... 115

8. The influence of informal financial support and childcare on fertility intentions: gender differences ...... 117

8.1 Fertility intentions and social capital...... 118

8.2 Data and results ...... 123

8. 3 summary and discussion: The effect of persisting traditional gender roles in fertility decision- making process ...... 131

9. Conflicting intergenerational roles: caregiving to grandchildren and elderly parents ...... 132

9.1 Macro level determinants of informal help and care ...... 133

9.2 Individual determinants of informal help and care ...... 135

9.3 The sandwich generation in the four-generation approach ...... 137

9.4 Data and methods ...... 139

9.5 Macro level analysis ...... 142

9.6 Individual-level analysis ...... 145

9.7 Summary and Discussion: family solidarity across generations ...... 152

CONCLUSION AND DISCUSSION ...... 154

REFERENCES ...... 159

LIST OF TABLES AND GRAPHS ...... 176

AUTHOR INDEX ...... 178

2

INTRODUCTION

One of the often-cited characteristics of the modern society is the individualization. The individualization theories often suppose the erosion of the traditional family and a decline of emotional and economic bonds among family members as a part of a general process of detraditionalization.

Individuals are more independent on each other, they do not rely on collective family resources and bear responsibility for their own action. Functions and roles traditionally performed by the family are now carried out by the state and its institutions. As a result, modern individualized societies face decreasing fertility rates because family values lose their relevance and raising the children without the support of the family represents a burden for isolated individuals.

There is however also another view of the modernization process. Recent research indicates that people are not fully independent, and they rely on external sources of support. Generally, modern society provides two channels of support. The traditional role of the family can be to some extent substituted by the state and public institutions. Nevertheless, the state and family are complementary institutions, rather than institutions excluding each other. Social policy in most of the European countries is based on different levels of subsidiarity; the idea that the state should intervene only in case of exhaustion of individual resources and resources of private groups such as the family. This principle was accepted to a various extent in European countries, but its importance is on the rise since welfare states generally struggle with an insufficient financing mainly as a result of ageing populations but also due to legitimizations issues. Furthermore, social institutions are not able to adopt all the roles performed by the family. Even though the economic function of the family has been weakened, the family is the most important source of an emotional support and even its economical function still holds its relevance at least within the relationships of the nuclear family.

Generally, the flow of resources takes place among people of different generations and generational relations are therefore important both for formal institutions and informal support. The

3

formal part of an intergenerational solidarity is primarily represented by the public system of pension funding, which is, in most of the European countries, fully or mostly dependent on a contribution of economically active younger generations. As has been shown by the recent research, the intergenerational solidarity on the private level usually has an opposite direction. Transfers of time and money flow from older generations to younger generations; besides other things it is because pensions funding enables older people to help their children and grand-children in times of need, especially in form of financial transfers, but also by providing a social support since payments of pensions allow healthy older people to dedicate their time to their families instead of having a paid job.

The downward flow of an informal support occurs also due to an insecurity of younger generations.

Young people tend to be more exposed to precarious and unstable job positions on the globalized labor market, especially in times of the economic downturns since they are not protected by seniority and working experiences. Some countries also struggle with a low accessibility of housing. A various support provided by generations of parents and grandparents helps young people to overcome the difficulties connected with the transitions to the adulthood: finishing the education, arranging the independent housing, establishing a stable relationship and starting the family. These transitions have been recently postponed to a higher age and as a result the rate of marriages and more importantly the level of fertility has declined significantly.

The investigation of the fertility process is of a great importance for European countries since the fertility rates are still unstable and low in most of the European countries. The preferred number of children is now lower than in the past, but the main issue lies in a discrepancy between a preferred and actual number of children since a high proportion of people fail to fulfill their fertility plans. One of the indisputable reasons is the increasing age at the first birth. With an increasing average age at the first birth, there is a tendency to a lower total number of children. However, social political arrangements designed to support young people in their childbearing activity do not lead straightforwardly to desired outcomes. Direct financial incentives are not efficient for people of all social classes and the same applies also to other forms of support such as childcare services, flexibilization of a labor market, providing social

4

housing etc. Furthermore, these interventions lead to different results in different countries. If the public social system is complementary to the private solidarity, it is important to investigate the role of intergenerational solidarity in the private sphere across the European countries. The findings will be possibly helpful for developing more efficient family interventions that do not suppress family relationships.

The main question of the presented study is the following: How is fertility affected by different forms of the intergenerational solidarity? The intergenerational solidarity is understood as a broader concept of the relationships between young generation and older generation of parents and grandparents and when relevant, it is also linked to support from other informal resources.

The first chapter introduces the topic of the intergenerational solidarity. What is a general understanding of the term, the relationship between public and private intergenerational solidarity, the measurement and manifestation of solidarity in different societies and individual and structural factors od the intergenerational solidarity. The following chapter (Chapter 2) discusses the fertility on an aggregate level: the development of fertility and its relations to others phenomenon such as the education system and female labor market participation and most importantly it summarizes the previous findings on the relationship between fertility and the informal solidarity. Chapter 3 is than an analysis of the relationship between parents´ normative intergenerational solidarity and fertility behavior of their children. The next chapter (Chapter 4) focuses on the analysis of the relationship between a subjective perception of household´s income and women short-term fertility intentions. It is shown that a higher income increases the probability of fertility intentions only under some circumstances and for some groups. Chapter 5 follows these findings and examine the effect of parents´ financial transfers to their children and their childbearing activity. The attention is given to the role of the child´s order. This topic is further elaborated in Chapter 6, where the effect of financial transfers on fertility intentions is analyzed and beside that also the effect of the childcare is considered. The most important finding is a different effect of these two forms of support for men and women. The final chapter (Chapter 7) deals with the problem of a potential

5

role conflict for people of middle-age generation who care of their grandchildren but also support their ageing parents.

6

1. Dimensions of intergenerational solidarity

As has been already indicated in the introductory section, the intergenerational solidarity is a multidimensional concept. The intergenerational solidarity is usually approached as a behavior. Social actors actively provide support in form of help, financial transfers or shared housing. Intergenerational solidarity can be, however, approached also as an issue of values, norms and attitudes (i.e. Kalmijn and

Saraceno 2008; Petrová Kafková 2010; Silverstein, Gans and Yang 2006). These two forms of solidarity can be identified as a latent and manifest solidarity (Silverstein and Bengtson 1997). Latent solidarity or normative solidarity corresponds to the responsibility to other family members or the emotional closeness. A supposed association between the normative solidarity and manifest solidarity is not always straightforward. For example, acceptance of the filial responsibility to parents is a necessary condition for providing a help and support to older parents by adult children, but it is not a sufficient condition

(Silverstein, Gans and Yang 2006). It is necessary to include the needs and resources of parents and children. For another example, the coresidence of adult children and their parents can be perceived as a form of a strong family support, but in fact it may manifest the inaccessibility of housing and a high unemployment rate among young people (Insegrad and Szydlik 2012; Albertini and Kohli 2012).

Coresident young adults in fact report worse affective relationships with their parents (White and Rogers

1997). The relationship between latent and manifest solidarity is elaborated in Chapter 5.

Manifest solidarity exists in three main forms: the coresidence and geographical distance of family members, the frequency and intensity of contact among family members and transfers among family members (financial, help, property)1. These three forms are interconnected on the individual level.

1 According to another terminology, all three forms can be perceived as transfers. In more general terms it represents material transfers (money or goods), time (care, help, contact) and space (coresidence). It seems, however, more legitimate to use a more general term of solidarity, because the term transfer indicates an action of giving or receiving some kind of support. Solidarity, on the contrary, indicates more static phenomenon, which can be present, even if not currently manifested.

7

Several studies have examined this connection, for example Albertini and Kohli (2012) showed that adult children in the southern countries tend to coresident with their parents, share consumption and help each other, but there is a lower level of financial transfers. On the contrary, the probability of adult children and their parents’ coresidence is low in the northern countries, as well as transfers in form of help, but there is a higher probability of a financial support. Financial transfers can be therefore understood as an alternative form of support, which allows a greater independence of individual family members. In case of coresidence, family members (parents and adult children) are more dependent on each other. Financial transfers serve as a substitute of coresidence primarily for young people who are approximately between 20 and 30 years old. On the other hand, transfers for children who are older than this age are more probably a substitute for bequests (Arrondel and Mason 2001).

The occurrence of the above cited dimensions of the intergenerational solidarity is based on many factors, both individual and structural. The next two chapters summarize these factors affecting intergenerational solidarity with an emphasis to solidarity between parents and their children.

1.1 INDIVIDUAL FACTORS INFLUENCING INTERGENERATIONAL SOLIDARITY

According to a previous research, the intergenerational support is influenced by the structure of opportunities and needs both on the provider´s and the recipient´s side. However, many authors have also tried to identify the motivation beside the private solidarity. These general theories are presented in following subchapters.

1.1.1 Private transfers behavior theories

The private intergenerational solidarity as a form of social action is supposed to be based on some motivation. Researchers, mostly in economics, tried to find an empirical evidence for one of two theories for motives of transfers: altruistic theory and exchange theory. The altruistic theory emphasizes the ties

8

among family members which are primarily emotional, and they are motivated and maintained by a sympathy to each other2. This type of motivation enables existence of one-sided transfers with no reciprocity, but it can be also two-sided (e. g. from a child to parents and from parents to a child). The exchange theory, on the contrary, supposes the existence of a rational choice as an important factor for keeping the individual motivated for exchange with other relatives. Some researchers (Laferrere and

Wolff 2006) add the third model in which is altruism impure and transfers are provided to make the receiver behave in a desired way.

Revealing the motivation for transfers is important not only for the social and economic theory, but also for social political interventions. The main question is, if the private transfers can be “crowded out” by public transfers. If members of extended family are linked together by altruism, the income and public benefits provided to one person will be equally shared by other family members. Hence, it is possible to employ the principle of subsidiarity and expect that person in need will obtain private transfers from own family and providing public benefits to this person is unnecessary and it could potentially violate family bonds. Some form of public help should be provided only if the whole family is in need. The example of this approach is the conservative welfare state (i.e. Germany, Austria), where the male breadwinner model is dominant. On the contrary, if rather exchange theory is plausible, then the principle of subsidiarity can be employed only to a limited extent, because sharing financial means among family members is constrained. The person in need can be still helped by another relative, but only if reciprocity is expected. Social system then should not rely on the family solidarity and it should be based on the individualized help.

2 There were several debates about the term „altruism“, which is usually understood in terms of morality. Nevertheless, regarding economic notion of private transfers, the altruistic behavior toward own relatives is perceived as a way of increasing own utility by increasing utility of close people.

9

Both theories were empirically tested with contradictory results. Cox (1987) considered altruism and exchange as motives for transfers between living people (inter vivos transfers3) and he concluded that the findings are more consistent with the exchange theory. Cox supposes a negative relationship between receiver´s income and the probability of receiving a transfer. This assumption is in consistence with both theories. However, the amount of transfers should be positively related to receiver´s income if exchange motives exist, because the person who receives the support from another family member will pay this back by providing some services to him or her. If the receiver´s income is higher, the donor of the financial transfer must increase the amount to compensate for the service. The data showed a positive relationship between income and value of transfers and for that reason, Cox concludes that his observations are more consistent with the exchange theory. He confirmed his findings later by analyzing another data set4 (Cox,

Rank 1992). However, Cox notes that the existence of the exchange is consistent both with the exchange theory and the theory of altruism. A similar conclusion was made also by Altonji, Hayashi and Kotlikoff

(1989), who compared the division of the total consumption and the division of collective resources inside an extended family5. Because the individual consumption corresponded to individual resources, but not to collective resources, they casted doubts on the altruistic theory. Their results were confirmed later by using the similar method as Cox, leading to a clear rejection of the altruism hypothesis (Altonji, Hayashi,

Kotlikoff 1997).

The same scholars (Altonji, Hayashi and Kotlikoff 2000), however, casted doubts also on the exchange theory. They considered a disproportion of income and amount of transfers among siblings and they found that children with a higher income tend to receive less transfers from their parents and give more in comparison with children with a lower income. These findings are incompatible with the exchange theory, which supposes the direct reciprocity in transfer behavior. Ioannides and Kan (1994)

3 Cox was one of the first scholars, who investigated transfers between living persons. Most of the previous research focused on bequest data (Ishikawa 1975; Becker and Tomes 1979; Adams 1980; Menchik and David 1983; Blinder 1973 and others). 4 Both the first and second analysis was based on US data: President's Commission on Pension Policy survey (PCPP) and National Survey of Families and Households (NSFH). 5 The analysis was based on US Panel Study of Income Dynamic (PSID) data. 10

obtained similar results. They considered both transfers from parents to children and transfers from children to parents in form of money and time. They found some proportion of exchange, but they do not deny the altruistic theory as they show that parent's needs are important determinants for transfers from children.

These contradictory results are probably linked to methods and interpretations of the analysis. As

McGarry (2000) pointed out, we can distinguish between exchange and altruistic behavior only to a limited extend. Furthermore, he suggests that using the dynamic model instead of the static one is convenient to follow the fluctuation of income and transfers. Generally, both the exchange and altruistic theories have been already approved and rejected and probably both the altruistic and reciprocal motives are present. For example, the provider of transfers is motivated by altruism but increases the frequency or intensity of support once he receives something in exchange. Furthermore, the inter vivos periodical transfers can be balanced by a one-time transfer of a high value (house, large amount of money etc.) or by expected bequest. Some scholars also distinguish between direct and indirect reciprocity (Arrondel and

Masson 2001; Arrondel and Masson 2006). For example, parents invest into children who don’t directly pay this support back, but they support their own children, for example, by financing of their education.

For that reason, an investigation of the motivation behind the transfers requires a longitudinal observation and consideration of the extended family network.

Furthermore, Kohli and Künemund (2003) show that transfer´s provider usually does not have single motive for giving. They also show that the motivational mechanism is not the same for different family types and gender. In addition, it can be supposed that the motivation varies regarding social stratification. While people from a lower social class might give all additional income to their family members who are in need and for whom family transfers can be crucial and necessary for their well- being, parents from a higher social class might tend to prefer strategic transfers which motivate their children in their own effort to achieve some goals such as to finish the education and find a well-paid job.

Leopold and Raab (2011) show that exchange motives take place when parents have sufficient financial resources to share with their children and they obtain in return help and care from their children.

11

To sum up, the individual motivation for transfer behavior varies and cannot be universally defined.

Even the same person can be motivated differently regarding a situation. Both altruistic and exchange patterns are usually present at the same time even though the social actor is not necessarily aware or does not acknowledge the existence of multiple sources of motivation.

1.1.2 Individual factors of transfers between parents and their adult children and grandchildren

In general, for most of the European countries, regardless the total amount and intensity of the transfers, holds true that the flow of private transfers among generations is downward. Parents and grandparents tend to be givers and their offspring receivers of the support (Albertini and Kohli 2012). The

“pivot generation” is the main provider of financial support to young generations. (Attias-Donfut, Ogg and

Wolff 2005). According Litwin et al. 2008, the downward flow of transfers continues to the age of 79 years and majority of downward transfers showed the age group between 50 and 59 years. Direct financial transfers are not, however, the only possible form of a material support provided by parents. Adult children may obtain help in form of shared housing. In this case, the amount of financial transfers is lower, because it is balanced by shared consumption (Schenk, Dykstra and Maas 2010). It concerns not only very young adult children, who are still in education or who have graduated recently, but also those who undergo some personal crisis such as a long-term unemployment, divorce or the death of a partner6.

Young people are in need of financial transfers in times of major life transitions such as finishing education, moving out of the parents’ house, finding a job, getting married and having a child. On the basis of a comparison of families in which children received more support than aging parents, Fingerman

6 Even though parents are usually the ones who provide housing, a child can be a provider of housing too. According to Choi (2003), unmarried parents as receivers of help tend to be older and less healthy. Except of health reasons, economic issues often come into play (Cooney 1989). Widowed parents are the most predisposed for coresiding with their children (Cooney 1989; Brody et al. 1995; Roan and Raley 1996). Children who did not experience the coresidence with their parents in their early adulthood are more rarely providers of housing and care to their elderly and dependent parents (Goldscheider and Lawton 1998). 12

et al. (2011) found that in these families were children in education or unmarried children. Hogan,

Eggebeen and Clogg (1993) also found that parents tend to support adult children who already have their own offspring. Financial help is provided also to divorced children (Hamon 1995).

Having a child in need does not make a sufficient condition for a decision of parents to provide a support. Parents need to have sufficient resources to be able to provide help to their children (McGarry

1999; Berry 2008; Albertini, Kohli and Vogel 2007). According to the theories of social stratification and inequality, the poverty can be ‘inherited’ – transmitted between generations. Because parents with a higher income tend to support more their children and increase their living standard, the social inequality is reproduced (Semyonov and Lewin-Epstein 2001).

As has been indicated, the transfers between parents and their children are not limited to a financial and material support. A crucial part of the support within family is a social support or help.

Middle-aged generation is in involved in childcare as a high share of grandparents regularly or occasionally care of their grandchildren (Guzman 2004; OECD 2012). Similarly as in case of a financial and material support, the grandparents’ care is connected with increased needs of children such as being full- time employed (Coall, Hilbrand and Hertwig 2014; Kuhlthau and Mason 1996; Vandell et al. 2003) or being a lonely parent (Hank and Buber 2009). Childcare is provided more by grandparents who are younger and healthy (Silverstein and Marenco 2001) and who live with a partner (Knudsen 2012). Grandparents who are employed provide childcare with a higher probability than retired grandparents (Guzman 2004), but less intensively (Attias-Donfut 2005 et al., Gray 2005). The grandparents’ care is also connected with the age of parents, grandparents and grandchildren. Grandparents take care of younger grandchildren and grandparents-caregivers are also younger than grandparents who are not involved in childcare (Coall,

Hilbrand and Hertwig 2014; Luo et al. 2012; Silverstein and Marenco 2001).

To sum up, private transfers usually take place under three circumstances: a receiver has a need, a provider has resources for giving and a provider has a motivation to provide a transfer.

13

1.2 STRUCTURAL FACTORS INFLUENCING INTERGENERATIONAL SOLIDARITY

A substantial part of studies on intergenerational solidarity focus on relationships within the family

(e.g. Lawton, Silverstein and Bengtson 1994; Van Gaalen and Dykstra 2006; Hank 2007). The family is approached as a self-dependent unit with a social network of supportive relations. Intergenerational relationships within the family are, however, shaped in a wider social and cultural environment. Parent's resources are, for example, supposed to be incentive for early nest leaving (Nilsson and Strandh 1999;

Mandic 2008). Manacorda and Moretti (2006), however, suggested that for Italian parents is coresidence with their adult children a desired living arrangement. Parents with a higher income live with their adult children more often than lower income parents. In addition, children´s labor market participation decreases in a response to rise in parent's income. It is therefore necessary to consider not only individual preferences and characteristics and institutional arrangements, but also cultural circumstances, habits, values and attitudes.

Several internationally comparative studies on intergenerational transfer behavior have been conducted recently (i.e. Schenk, Dykstra and Maas 2010; Albertini, Kohli and Vogel 2007; Albertini and

Kohli 2012; Bonsang 2007). These studies indicate a connection between private transfer behavior and welfare state regimes. There is a divide between the southern and northern European states that have very different social system arrangements and to some extent also between western and eastern part of

Europe (i.e. Herlofson et al. 2011) even though especially in the eastern part of Europe this topic is generally under-investigated. The southern countries (Italy, Spain and Greece) show a lower frequency and a higher intensity of intergenerational solidarity (i.e. financial transfers, help, coresidence of parents and their adult children) in comparison with the norther countries (Swedish, Denmark). This pattern is in accordance with the typology of welfare state regimes (Albertini, Kohli and Vogel 2007; Albertini and Kohli

2012) since social policy is much more stressed in the norther countries. It can therefore substitute a support among family members. On the other hand, it has been recently suggested that crowding-out hypothesis, according which the state is a substitute of the family, has some limitations. Kohli (1999) showed that part of the public transfers to an older generation is given to a younger generation in form of 14

private transfers. Künemund and Rein (1999) made a similar conclusion based on an investigation of data from five countries including Germany, United States and United Kingdom. Even though they found different patterns of state and family support in these five countries, their conclusion also cast doubts on the crowding out hypothesis. In this regard, public transfers can even reinforce solidarity inside the family.

Fokkema, Bekke and Dykstra (2008) suggested a typology of families based on their patterns of private solidarity and they identified four types of families: descending familialism (i.e. Belgium), ascending familialism (primarily represented in southern countries such as Italy, Span or Greece), supportive at distance (northern countries such as Sweden, Denmark and Netherland) and autonomous families (France and Switzerland). This typology is based on a few characteristics of intergenerational solidarity such as distance of living, frequency of contact, perception of family care obligations and the direction of transfers. All the family types exist in all examined countries, but their distribution is different.

The authors have casted doubts on a supposed weakening of family solidarity in the northern countries.

The typical family in the northern countries is family with children, who don't live nearby their parents (in the same household or house) and they declare weak family care obligations (normative solidarity).

However, these families still have a rather frequent contact with each other and there is a downward flow of financial transfers from parents to adult children.

Furthermore, Gibney and McGovern (2011) contributed to the debate about crowding-out effect of welfare state by testing a hypothesis of a causal effect of a social support network type on mental health.

Their findings suggest that weak family relationships are much more common in the southern countries

(Greece, Italy) than in the northern countries. The greatest proportion of people who are fully integrated in their family and community and report a high level of mutual help among family members (parents, children, grandchildren) have been found in Sweden, Denmark and Netherlands. These countries also show the lowest level of loneliness and depression. In consideration of this evidence, a divide between the familialism of the southern countries and de-familialism of the norther countries is increasingly recognized as an outdated notion.

15

As noted above, public benefits and services do not only necessarily lead to weak family relationships but sometimes even reinforce them, due to an increased amount of time and money, which can be shared by relatives. A strong social system can enable a free choice of the most convenient model of a family life and support. People do not live with their families and care about their relatives just because they are constrained by their economic situation, insufficient public services or a lack of housing.

If they still do that, they know that it is a matter of their choice. A voluntary decision is usually related to a higher satisfaction and reduce possible conflicts among relatives. This is in accordance with findings on transfers among family members: a high frequency and low intensity in the northern countries (Sweden,

Denmark and Netherlands) and a low frequency and higher intensity in the southern countries (Spain,

Greece and Italy). For example, a help provided by adult children to their ageing parents is more occasional and distributed among all adult children in the northern countries, on the contrary, a help is provided with a high intensity by one coresiding children (Bonsang 2007) in the southern countries.

Similarly, Brandt, Haberkern and Szydlik (2009) distinguish help and care as two forms of possible support.

Whereas care is based on a physical need and usually is very intense and demanding, help tend to be more spontaneous and occasional. There is a higher proportion of children who provide both help and care in the southern countries. On the contrary, care is secured by professional workers in the northern countries and children tend to provide only occasional help to their parents. As mentioned above, parents in the northern countries tend to be more satisfied with their social integration and quality of their social network. This could suggest that providing care to older parents directly by their adult children does not necessarily lead to better relationships between parents and children.

Some conclusions can be made based on the above cited findings. First, there is some connection and complementarity between institutional arrangements in form of the welfare system and family. The patterns of intergenerational transfers and solidarity behavior in European countries follow the patterns of welfare regimes. This link, however, does not simply correspond to crowding-out theory. The state substitutes for some obligations that would be else secured by the family (crowding-out), but in the same time enables people to spend their time and money with their relatives voluntarily and in a more relaxed

16

way (crowding-in). Some evidence indicates that this arrangement is beneficial to quality of relationships within the family.

1.3 LIFE COURSE APPROACH

Life course perspective is one of the recent approaches to an investigation of individual histories. It focuses on a social change over human lives and it builds on an assumption that prior experiences in an individual's history has an influence on following events, values, decisions and beliefs. Furthermore, according to the life course perspective, it is necessary to consider a broader social surrounding affecting individual´s decisions. These assumptions are crucial for the intergenerational solidarity investigation which is based on the relationships among people in small social groups such as the family. Szydlik (2012) suggests the life course approach as a convenient tool for studying of intergenerational solidarity because of the existence of lifelong solidarity between generations. This dissertation focuses on help that is provided by parents to children in early adulthood, however, the range of support between parents and children is much broader.

The life course approach considers the effect of a structure and events within the family on solidarity and relationships between parents and children. Family disruptions have a long-term impact on children’ willingness to help their parents (Cooney and Uhlenberg 1992; Eggebeen 1992; Pezzin and

Schone 1999) and they also influence a general relationship between parents and adult children (White

1992). A similar effect was found in stepfamilies and single-parent families (Aquilino 2005).

Apart from the investigation of long-term effects of family disruptions, transitions to adulthood have recently occupied attention of researchers within the life course perspective. The transitions to the adulthood can be characterized by several events or role transitions such as leaving home (Goldscheider and Goldscheider 1999), marriage or financial independence but also by an attainment of some personal characteristics, e.g. accepting different kinds of responsibilities and an ability to make independent decisions (Arnett 1997, Arnett 1998). The major life transitions are assumed to occur in a specific order

17

(Elder and Shanahan 2006; Gauthier 2007a; Marini 1984), but it has been shown that there is no clear agreement on the order of events and on the meaning and importance of different transitions (Arnett

2001; Shanahan 2000). A perception of conditions for entering adulthood is culturally defined and it is also changing rapidly during the time. Along with the differences in the welfare state arrangements, labor market, housing market and other macro social factors, a high diversity of pathways into adulthood exists in the European countries (Buchmann and Kriesi 2011).

The changes in life course are explained in the context of the second demographic revolution.

According to that, all European countries go through this demographic process, but they are on different stages of the demographic revolution. If that is true, a convergence of the European countries is to be expected along with an increasing diversity within the countries because of the individualization process.

Recent investigations, however, showed that countries do not converge to each other over time (Billari and Wilson 2001). On the contrary, there are some persisting patterns in most of the countries such as early nest leaving in the Scandinavian countries and late nest leaving in the southern countries (Billari

2004) or smooth transitions from the education to work in Germany, Austria or Denmark and difficult entries on the labor market in the southern countries (Brzinsky-Fay 2007). According to these findings, it is suggested that the second demographic revolution is not a completely identical process across all countries and it is related to other characteristics such as cultural conditions and institutional arrangements.

Besides the effects of mostly cultural changes represented by the second demographic revolution, changing economic aspects play an important role for delaying some steps to an adulthood. Young people are more at risk of having an unstable and uncertain job position in comparison with older generations since they are not protected by seniority (Blossfeld and Hofäcker 2011). In most of the European countries, young people also face higher unemployment rates. This perceived insecurity about future earnings leads to an unwillingness to make long-term and irreversible decisions such as having a child. The childbirth is traditionally the last step of the entry into adulthood. Before starting their family, young people are supposed to leave their parents’ house, finish their education, find a job and enter into stable

18

partnership (marriage or cohabitation). While these transitions can be done even in a different order (e.g. moving out from parents’ house before finishing education or after finishing education), the childbirth is usually postponed until previous steps have been completed. For example, Barbieri et al. (2015) show that in countries with weak welfare states, women tend to postpone their motherhood until they have a stable employment. Similarly, a negative correlation between a proportion of young adults living with their parents and fertility rate has been observed (Billari and Kohler 2004; Del Boca and Locatelli 2003). A postponement of the nest leaving leads also to a postponement of childbearing. Young adults need to gain a certain level of independence before becoming parents. While a postponement of most of the steps into adulthood usually does not seriously harm future opportunities, a postponement in childbearing leads to the decreased level of fertility. That implies that social policy should not be directed only to childbearing behavior, but it should also focus on other parts of the process such as housing and job opportunities or system of education.

To sum up, the life course approach points out the importance of the link between steps in individual´s histories. An investigation of fertility necessarily includes also topics of marriage, education, employment and others. Furthermore, previous life-course research indicates that even though most of the transitions in life are taken by adults, they are not always taken independently and without external sources of support. As has been said, the intergenerational solidarity accompanies these transitions and smooth the process of adulting.

1.4 INTERGENERATIONAL SOLIDARITY: SUMMARY

Despite a supposed substantial increase in individualization and a decline of traditional family forms in current European countries, recent research indicates that an informal solidarity across generations and family members is still of a not negligible importance. It has been shown that various dimensions of solidarity are differently represented in countries across Europe and besides cultural variation, there is also a substantial individual variation. However, an originally accepted crowding-out hypothesis which

19

supposed the rise of welfare state at expense of informal family relationships must be revised and elaborated. The intergenerational solidarity is changing rather than disappearing and losing its significance. Previous research has also pointed out the importance of informal solidarity in life course perspective. It is shown that main channels of informal support are directed from older generations to younger ones and this kind of support is especially important in times of increased needs and insecurity.

From a sociological and demographical point of view, one of the most important and most difficult steps to adulthood is starting the family. The next chapter deals with the fertility in Europe and theoretical explanation of persisting low fertility rates in most of the European countries.

2. Fertility in Europe

Fertility in most of the European countries is low and recently the total fertility rate (TFR) has oscillated around 1.5 child per woman (Eurostat 2017a). Nevertheless, the total fertility rate still varies considerably across Europe, starting from countries with the TFR above or close to the replacement level

(Ireland, France, Sweden) to the lowest-low7 fertility (Spain, Greece, Portugal, Poland) (Eurostat 2017a).

This pattern was different in 1980, when Scandinavian countries showed the lowest level of fertility, while statistics for southern and eastern European countries were still exceeding the replacement level (Castles

2003; see Table 1). For a long time, low levels of fertility were ascribed to consequences of the second demographic revolution, the term proposed by Lesthaeghe and Van de Kaa (1986). The second demographic revolution is characterized by many changes in family and fertility behavior. These changes are interconnected; nevertheless, a postponement of childbearing and a decline in number of children per woman are the most important for the overall decrease of total fertility rate. The mean age of women at childbirth in EU has risen from 29.0 in 2001 to 30.5 in 2015 (Eurostat 2017a) which indicates a

7 The term lowest-low fertility has been proposed by Kohler, Billari and Ortega (2002) for the total fertility rate below 1.3.

20

postponement of childbearing of European women. Delayed births contributed to decreased fertility rates

(Frejka and Sobotka 2008).

21

Table 1: Total fertility rate 1980-2015

1980 1985 1990 1995 2000 2005 2010 2015 Trend Austria 1,65 1,47 1,46 1,42 1,36 1,41 1,44 1,49 Belgium 1,68 1,51 1,62 1,56 1,67 1,76 1,86 1,70 Bulgaria 2,05 1,97 1,82 1,23 1,26 1,37 1,57 1,53 Cyprus 2,35 2,43 2,41 2,03 1,64 1,48 1,44 1,32 Czech Republic 2,08 1,95 1,90 1,28 1,15 1,29 1,51 1,57 Denmark 1,55 1,45 1,67 1,80 1,77 1,80 1,87 1,71 Estonia 2,02 2,13 2,05 1,38 1,36 1,52 1,72 1,58 Finland 1,63 1,64 1,78 1,81 1,73 1,80 1,87 1,65 France 1,85 1,86 1,77 1,74 1,89 1,94 2,03 2,01 Germany 1,44 1,37 1,45 1,25 1,38 1,34 1,39 1,50 Greece 2,23 1,67 1,39 1,28 1,25 1,34 1,48 1,33 Hungary 1,91 1,85 1,87 1,57 1,32 1,31 1,25 1,45 Iceland 2,48 1,93 2,30 2,08 2,08 2,05 2,20 1,80 Ireland 3,21 2,48 2,11 1,84 1,89 1,86 2,05 1,85 Italy 1,64 1,42 1,33 1,19 1,26 1,34 1,46 1,35 Latvia 1,86 2,08 2,02 1,25 1,25 1,39 1,36 1,70 Lithuania 1,99 2,08 2,03 1,55 1,39 1,29 1,50 1,70 Luxembourg 1,50 1,38 1,60 1,70 1,76 1,63 1,63 1,47 Malta 1,99 1,95 2,04 1,77 1,68 1,38 1,36 1,37 Netherlands 1,60 1,51 1,62 1,53 1,72 1,71 1,79 1,66 Norway 1,72 1,68 1,93 1,87 1,85 1,84 1,95 1,72 Poland 2,28 2,33 2,06 1,62 1,37 1,24 1,41 1,32 Portugal 2,25 1,72 1,56 1,41 1,55 1,41 1,39 1,31 Romania 2,43 2,31 1,83 1,33 1,31 1,40 1,59 1,58 Slovakia 2,32 2,26 2,09 1,52 1,30 1,27 1,43 1,40 Slovenia 2,06 1,71 1,46 1,29 1,26 1,26 1,57 1,57 Spain 2,22 1,64 1,36 1,16 1,22 1,33 1,37 1,33 Sweden 1,68 1,74 2,13 1,73 1,54 1,77 1,98 1,85 Switzerland 1,55 1,52 1,58 1,48 1,50 1,42 1,52 1,54 United Kingdom 1,90 1,79 1,83 1,71 1,64 1,76 1,92 1,80

Source: The World Bank

22

The postponement is usually named as the main cause of a lower number of births per woman, but it is accompanied by other factors such as changing family size preferences. Some demographers suppose that an extreme decline in fertility is just temporary until delayed births will be realized (Bongaarts 2001); however, as Sobotka (2004) and others (Lesthaeghe and Willems 1999) show, even after a recovery of fertility, it can be predicted that fertility remains below replacement level. The decline in number of children in families has two explanations. Firstly, people prefer a higher number of children, but they are prevented from a further reproduction by external factors: biological constraints, financial costs or inappropriate institutional arrangements. An increase in fertility rate can be achieved by implementation of pro-natalist policies (Chesnais 1996; 2000). Other explanation is based on changes in values leading to a decrease in family size ideals. People delay and constrain the childbearing because they prefer a smaller number of children or even childlessness. According to Eurobarameter 2001 survey, an average number of children preferred by young people in German speaking countries is below the replacement level

(Goldstein, Lutz and Testa 2003) and preferences for sub-replacement levels have been observed also in other countries (Frejka and Sobotka 2008). A significant share of young people in European countries also declare an intention to remain childless (Sobotka and Testa 2008). Table 2 shows ideal size family size in

European countries in three age groups based on the European Value Study 1990. It is obvious that a percentage of people who don´t want any kids was close to zero in most of the European countries and age groups. There are, however, some exceptions: about 3% of the youngest people in Germany and about 4% of young and middle-age people in Netherlands prefer to stay childless. In addition, a significant share of people (about 15%) between 15 and 49 years in Germany believe that ideal size of family is one child. Surprisingly, one child has been marked also by quite a lot of young people in the eastern European countries: Bulgaria (10%) and Romania (12%). These countries also show a rather low percentage of people thinking that an ideal family size is three and more children. This applies also to the other eastern

European countries: the Czech Republic and Hungary. This could indicate a delay in consequences of the second demographic revolution, but economic constraints are probably involved too. On the contrary, a

23

large share of people preferring a big family has been found in high fertility countries: Iceland and Ireland.

Most of the other countries are characterized by almost equal share of preferences in either 2-children family or a larger family.

This relatively old statistic can be compared with a newer one from Eurobarameter 2011 (see Table

3). The percentage of people who prefer to stay childless seems higher in this case, the highest value show Austria (9%) and the Netherlands (8%). A decreasing tendency to big families is obvious since only in

Ireland a majority (54%) of respondents consider three or more children as an ideal size of family. On the other hand, a tendency to have only one child does not seem to grow considerably. For that reason, most of the people in majority of European countries prefer two children. According to these statistics compared with the figures from 25 years ago, it is possible to predict an increasing tendency to smaller families and a higher proportion of childlessness; however, it is indicated that people who decide to have a child usually prefer more than one.

24

Table 2: Ideal number of children in 1990, age groups, percentages

Ideal number of Ideal number of children children Age 0 1 2 3+ Age 0 1 2 3+

Austria 15 -29 2 5 63 30 Latvia 15 -29 0 4 39 57 30-49 2 6 60 32 30-49 0 3 44 53 50+ 1 4 56 39 50+ 0 0 41 60 Belgium 15-29 0 7 53 40 Lithuania 15-29 0 2 51 47 30-49 0 9 54 38 30-49 0 0 41 59 50+ 0 7 52 42 50+ 0 0 25 74 Bulgaria 15-29 1 10 66 23 Netherlands 15-29 4 5 50 42 30-49 1 6 67 27 30-49 4 4 52 41 50+ 0 1 55 44 50+ 0 0 50 50 Czech R. 15-29 0 3 78 19 Norway 15-29 0 1 55 43 30-49 0 4 74 22 30-49 0 1 53 46 50+ 0 3 70 27 50+ 1 1 37 61 Denmark 15-29 1 3 60 36 Poland 15-29 0 4 63 33 30-49 0 3 60 38 30-49 0 4 56 40 50+ 0 1 45 54 50+ 0 1 46 53 Finland 15-29 0 8 43 50 Portugal 15-29 0 7 66 26 30-49 2 3 47 49 30-49 1 5 61 33 50+ 0 1 48 52 50+ 1 3 53 43 France 15-29 1 2 51 45 Romania 15-29 1 12 73 15 30-49 1 3 46 50 30-49 2 8 64 27 50+ 0 1 42 58 50+ 0 4 52 44 Germany 15-29 3 16 63 19 Slo vakia 15-29 0 4 62 34 30-49 2 14 65 19 30-49 0 4 60 36 50+ 2 6 64 28 50+ 0 2 47 51 Hungary 15-29 0 4 74 23 Slovenia 15-29 2 5 61 33 30-49 0 6 65 29 30-49 2 4 63 32 50+ 0 2 57 41 50+ 1 3 55 42 Iceland 15-29 1 1 32 67 Spain 15-29 2 6 56 36 30-49 0 0 25 75 30-49 1 6 56 37 50+ 0 0 11 89 50+ 1 4 45 50 Ireland 15-29 0 1 29 70 Sweden 15-29 1 2 54 43 30-49 0 1 27 73 30-49 0 2 56 42 50+ 0 0 14 85 50+ 0 1 57 42 Italy 15-29 1 6 56 37 UK 15-29 1 4 63 33 30-49 1 8 58 34 30-49 1 3 68 29 50+ 0 4 52 44 50+ 0 1 52 47

Source: European Values Study 1990

25

Table 3: Ideal number of children in 2011

Ideal number of children 0 1 2 3+

Austria 9 17 54 20 Belgium 6 10 47 37 Bulgaria 0 13 76 11 Czech Republic 2 15 67 17 Denmark 4 3 51 42 Estonia 1 6 50 43 Finland 6 7 46 42 France 3 8 50 39 Germany East 5 13 61 20 Germany West 6 9 59 27 Hungary 3 12 59 26 Ireland 3 4 39 54 Italy 4 14 61 21 Latvia 2 8 54 36 Lithuania 2 8 60 30 Malta 4 13 58 25 Netherlands 8 6 50 37 Poland 2 9 55 34 Portugal 3 15 59 23 Romania 2 14 67 17 Slovakia 2 14 60 24 Slovenia 2 7 60 32 Spain 4 7 60 29 Sweden 4 6 54 37 Great Britain 6 6 58 30 Total 4 9 56 31

Source: Eurobarometer 2011

2.1 FERTILITY, GENDER EQUITY AND CONFLICTING ROLES OF PARENTS

The fertility is substantially connected with gender roles and gender equity. A decreasing fertility in developed countries is among other factors the result of a greater availability of a contraception, an

26

increasing level of women´s education and a higher rate of women´s participation on the labor market.

On the other hand, developed countries with a very low level of fertility need to even increase the gender equity to reach a higher level of fertility (McDonald 2000a; McDonald 2000b). In modern societies women have an equal access to education and they can participate on the labor market, but their opportunities are constrained by childbearing and they can choose to reduce their childbearing activity to achieve success in their working life. Torr and Short (2004) demonstrated the relationship between gender equity and fertility also on the individual level and they suggest that both the most traditional model with a low level of gender equity in the division of housework and the most modern model with a high level of gender equity are positively correlated with fertility. Similar results have been obtained by Gil (2004) who analyzed a relationship between private gender equity and fertility in European countries. Mills et al.

(2008) describe it as a role conflict since they observed that only women who perform a large share of housework and at the same time work a lot of hours in a paid employment or already have one or more children have lower fertility intentions. Cooke (2009) also analyzed a relationship between the gender equity in a paid employment and the likelihood of the second childbirth in Italy and Spain with the results suggesting a negative effect of women’ employment in case that a couple does not dispose of formal or informal sources of care for the first child. A negative effect of mothers’ employment can be partially balanced also by childcare provided by fathers (Cooke 2004). Adsera (2011) furthermore observed a tendency to postpone childbearing in case of being employed with a temporary contract. The role of women’s labor market attachment has been shown also by others (Andersson 2000; Hoem 2000). In case of an insecure employment women usually wait until they achieve a stable employment. In some countries, such as Sweden, this tendency is further supported by arrangements of welfare regimes where parental allowance is based on parents’ income before the parental leave (Andersson 2008).

A possible conflict between job and the family and gender roles are important considerations for social interventions. An effect of financial incentives occurs only for lower income groups and not for groups with a higher income (Cohen, Dehejia and Romanov 2007). On the other hand, the accessibility of childcare services may have an effect only on dual-earners families, while women who prefer to stay at home and provide a full-time care of the children will probably take the availability of the childcare 27

services into consideration only to a limited extent and it will not affect the timing or the quantity of their childbearing. Rindfuss et al. (2007) showed on the Norwegian data that availability of the childcare services influences the timing of first childbirth. On the contrary, the effect of childcare services has not been observed in Western Germany (Hank and Kreyenfeld 2003). Authors suggest that one of the reasons for that may be the general conception of welfare state politics which supposes the male breadwinner model where mothers are rather discouraged from being employed.

The rate of mothers´ employment varies considerably across Europe. Figure 1 illustrates a proportion of employed women with one child under age of six and with three children and youngest child under age of six. The highest rate of employment show Luxembourg, the Netherlands, France,

Austria, Sweden and Denmark, on the other side of a scale are primarily the eastern and southern countries: Hungary, Slovakia, the Czech Republic, Estonia and Greece. Most of the countries show also a high distinction between employment of women with one and with three children. From that point of view, the statistics for Scandinavian countries (Sweden and Denmark) are interesting since there is no difference between one child and three children. Both employment rates of women with one child and women with three children are one of the highest in Europe. This figure indicates the efficiency of a wide scale of childcare services that are provided in these countries. The welfare benefits for parents in the

Scandinavian countries are also apparently arranged in a way encouraging women to work. What is important is the fact that these politics have supposedly a positive impact on the fertility rates since the total fertility rate was 1.73 in Denmark and 1.91 in Sweden in 2013. These rates are still below the replacement level; nevertheless, they are well above the EU average rate (1.58 in 2015).

28

Graph 1: Mothers in the labor market in 2016

Source: Eurostat 2016

29

Recent research has shown great differences across European countries also in terms of attitudes.

There is a consensus about some general trends such as a delay of the second demographic revolution’s attitudes in the eastern European countries or attachment to more conservative familialistic attitudes in the southern countries. Graph 2, 3 and 4 show a comparison of 29 European countries regarding the aggregated statistics of some gender-family attitudes based on the European Value Study 2008. The first question considers the perception of a general importance of children. The differences are very high in this case varying from less than 10% of agreement in Sweden to almost 90% in Hungary. The eastern

European countries but also Greece and surprisingly also Denmark show a high level of agreement. On the contrary, the lowest level of agreement is in the other northern European countries (Sweden, Finland,

Norway), the Netherlands and the United Kingdom. These attitudes are not linked to fertility levels, for example, Greece show a high agreement with a general importance of children and a low level of fertility, on the contrary, respondents in Sweden and UK are not very familialistic in their attitudes but total fertility rate is close to the replacement level in these countries.

30

Graph 2: Percentage of agreement with the statement “Women has to have children in order to be fulfilled”

Source: European Values Study 2008

31

The two following figures consider the attitudes on working mothers. Although they are based on similar questions, the first of them focuses on the working mothers regardless the age of their children, while the second asks on the mothers of pre-school children. The first question is not so controversial and all the countries show over fifty percent of agreement starting from less than 70% in Poland to almost

100% of agreement in Finland and the other northern countries. On the contrary, the attitudes over working mothers of pre-school children vary considerably across countries. The level of agreement is lowest in the southern countries where most of the people (over 70%) agree that children of working mothers suffer while only around one fifth of respondents in the northern countries agree with the statement.

32

Graph 3: Percentage of agreement with the statement “Working mother can establish just as warm and secure relationship with her children

Source: European Values Study 2008

33

Graph 4: Percentage of agreement with the statement "A pre-school child is likely to suffer if his or her mother works"

Source: European Values Study 2008

34

A tendency to liberal attitudes toward family life in the Scandinavian countries is indicated also by the other question from the European Value Study: If someone says a child needs a home with both a father and a mother to grow up happily, would you tend to agree or disagree? As Graph 5 shows, the

Scandinavian countries belong to countries with the lowest proportion of people who tend to agree with this statement. Similarly, most of the people in the Scandinavian countries do not agree that “a marriage or a long-term stable relationship is necessary to be happy” (see Graph 6). These attitudes can be therefore in contrast with the demographic indicators suggesting higher rates of fertility in the northern countries in comparison with the countries with more conservative attitudes toward different aspects of family life. There are several possible explanations of this discrepancy. First, there might be only a weak relationship between attitudes and a real behavior. People express their generalized attitudes, but they can act rather differently in their private life. Secondly, more liberal attitudes over family can indicate more relaxed demands on parenting in a society. While people in more conservative countries believe that they have to accomplish the most convenient conditions before having children (such as ensure that they have a stable relationship and a sufficient income), people in the northern countries do not rely too much on these demands because they believe that a child can grow up happily even when a mother is working, a child is raised by an external child-care provider or when one of the parents is not present.

Andersson (2008) suggests that welfare state politics that are not aimed at setting certain conservative family norms (marriage, children raised by both parents) and that allow for a greater diversity of family settings may lead to a higher level of fertility. The Scandinavian public policy arrangements also support the dual-earner model (Korpi 2000). Crompton and Lyonette (2006) explored that in the Scandinavian countries (Norway and Finland) work-life conflicts occur significantly less often than in other countries

(Britain, France and Portugal) and they suggest that aside from childcare services the gender division of domestic work can affect the work-life balance. It is, however, difficult to observe the direction of the causal link between public policy and attitudes. Will attitudes change after introducing some political arrangements facilitating the gender equity between men and women and balance between work and family life? Or are these arrangements going to fail in other countries because people express in their attitudes disagreement with some values contained in these measures? 35

Graph 5: Percentage of agreement with the statement "Child needs a home with both a farther and a mother to grow up happily"

Source: European Values Study 2008

36

Graph 6: Percentage of agreement with the statement "A marriage or a long -term relationship is necessary to be happy”

Source: European Values Study 2008

37

2.2 FERTILITY AND PUBLIC POLICIES ARRANGEMENTS

While the research on the relationship between private solidarity and fertility is not very rich, a lot of studies focused on the effects of welfare state politics. However, findings of these studies can indicate an expected effect of private support because it is based on a similar structure of needs. Many researchers tried to provide an international comparison of politics arrangements. The expenditure on family/children policies varies considerably across European countries from less than 4% of total expenditure on social benefits in Italy to almost 15.7% in Luxembourg (Eurostat 2017b). However, as

Castles (2003) shows, the social expenditure on families are not significantly correlated with the fertility and other factors such as a proportion of children using formal childcare services and a proportion of women in a flexible employment correlate much stronger.

Billingsley and Ferrarini (2014) tested the effect of family policies in 21 European countries distinguishing two kinds of social policies designed to support fertility and families: traditional family support and earner-carer support. While the first one supports the family as a unit in form of child benefits, tax and marriage subsidies and home-care allowances, the second one is focused on the individual and it allows both parents to participate on the childcare and on the labor market. The allowances are earning-related and childcare for children under age of three is provided. They show that the earner-carer support has a stronger and positive effect on an intention to have (another) child. This kind of analysis, however, casted doubts on a relevance of the observed effect since countries are clustered in groups based on a few characteristics but usually they share other common characteristic such as cultural values and norms or similar economic situation that are not observed. Bonoli (2008) therefore tried to overcome these limitations by analyzing the cantons in Switzerland and he concludes that there is a significant and positive effect of childcare availability but also the generosity of child benefits. The effect of financial incentives has been observed also in Israeli by Cohen, Dehejia and

Romanov (2013) with the strongest impact on households with a lower income. Laroque and Salanié

(2004) observed in France a notable effect of financial incentives on the probability of birth of the first and the third child but not so much for the second child. Fenge and Meier (2008) suggest that the 38

effectivity of family allowances system is the highest when the optimal number of children is set and people having a lower number of children have to pay penalty.

The effect of social policies has been indicated also by others (Castles 2003; Oláh 2003) and it seems to be strong especially in case that it enables a balance between work and family life. The availability of childcare facilitates the return on the labor market and parental leave benefits related to the income before the childbirth decreases the motherhood penalty.

Furthermore, the availability of a preschool formal childcare and parental benefits are related to the length of parental leave. The effect of maternity and parental leave can be twofold. On the one hand, it can encourage young people to have a child because they do not have to rely on external childcare providers and one parent (usually woman) can provide full-time childcare. On the other hand, if the maternity or parental leave is too long and there are no other alternatives such as early childcare services or informal childcare, women can be discouraged from having a child early or ever. An insufficient provision of childcare may lead to a prolonged period that woman spend out of the labor market and restrict their future opportunities.

Most of the countries introduced a maternity or parental leave but the leave is not always paid. For example, in the countries with a relatively high level of fertility, Sweden and Norway, there is up to one- year paid leave for mothers and also fathers (Duvander, Lappegard and Andersson 2010). The

Scandinavian countries make efforts to find a balance between the parental care, enabled by a sufficiently long parental leave, and childcare services. For example, Finland has less childcare services in comparison with other northern countries, but the state provides a long maternity leave. On the contrary, Denmark provides a high coverage of subsidized childcare, but the paid maternity leave is shorter than in Norway,

Finland and Sweden (Gupta, Smith and Verner 2008). Regarding the effect of prolonging the parental leave on the fertility, Luci and Thevenon (2012) noticed that it can be negative if it is not controlled for the provision of childcare services for children aged 3 years and younger. These authors also point out that an unobserved heterogeneity probably plays an important role in an investigation of the impacts of family policies on the fertility in different countries. It is highly probable that a paid maternity (parental) leave in

39

a connection with subsidized, high-quality, affordable and broadly available public and private childcare services helps to increase the fertility level. The positive effect of a prolonged paid parental leave on fertility has been observed also in Austria (Lalive and Zweimüller 2009) – the country with a considerably lower level of fertility in comparison with the Scandinavian countries. On the other hand, a quite generous family policy in Slovenia does not seem to have a noticeable effect on the level of fertility, even though the system seems to be like the Scandinavian model: one year of parental leave, affordable and available childcare and other family financial benefits (Stropnik and Šircelj 2008). This study indicates that a generous family policy has to be accompanied by other factors; authors suggest that the effect of

Slovenian family policy is weakened by a difficult position of young people on the labor market characterized by high unemployment rates and insecure employment and also by a low availability of an affordable housing and persisting of traditional gender roles.

Overall, it seems clear that there is no universal solution which could be applied to family policies of all countries, not even the similar ones because of the ambiguous effects of different political arrangements. Looking for an inspiration in more developed countries with successful family policies and a relatively high level of fertility is useful, but the implementation of the measure must be done with a caution since the level of development of the targeted country may not be the same as the original country. Usually some preconditions must be achieved such as a stable and flexible labor market, satisfactory living standard and a decrease of traditional gender roles importance. Nevertheless, the present state of research in family policies does not indicate that the intervention of the state is unnecessary, and fertility related problems should be solved only privately. However, the role of private resources within the informal social networks is of a great importance and this importance is supposedly not going to diminish in the future since most of the European states cannot afford to extend the coverage of their welfare states. The role of solidarity between children and parents as the main form of family solidarity is discussed in the following section.

40

2.3 THE INTERGENERATIONAL SOLIDARITY AND FERTILITY

A decision about the fertility is not an individual process, but it takes place within a group of close people and a broader social environment. The influence of significant others on fertility behavior can be carried out either directly, e.g., parents provide money, housing or help to their children with young offspring to improve the child’s family living conditions and to facilitate another childbearing activity, or indirectly, e.g., parents share their values and attitudes, and children are socialized within family norms and accept some general beliefs about family life.

Some authors (Balbo and Barban 2014; Balbo, Barban and Mills 2013) investigated if the fertility of friends has an impact on the individual fertility decisions, and they revealed the short-term effect of friend’s childbearing behavior. Similar results have been obtained by Laura Bernardi (2003) for women in

Italy, Khadivzadeh et al. (2013) for women in Iran, Lois and Becker (2014) and Pink, Leopold and

Engelhardt (2014) for young people in Germany, Bergnéhr (2009) and Hensvik and Nilsson (2010) for young Swedish adults, Bühler and Fratczak (2007) for the Poles.

The siblings’ influence has been also observed by some studies (e.g., Lyngstad and Prskawetz

2010; Raab et al. 2014; White and Bernardi 2008), appearing to be similar to the effect of members of the peer group. Unlike the emotional affection, people have only limited resources of time and energy

(Grundy and Henretta 2006), and children can in their decision reflect that the burden put on their parents will probably prevent them from providing other support. Emery (2013) showed that only child has also a much higher probability of receiving a financial transfer from the parents than children with one or more siblings.

While friends and other members of peer groups create a role model, and their influence is mediated by social learning (Pink, Leopold and Engelhardt 2014), the parents are also a potential source of a financial or practical help (Mathews and Sear 2012; 2013). Some authors (Bernardi, Keim and von der

Lippe 2007; Rossier and Bernardi 2009) distinguish three different mechanisms of social network effect on the fertility behavior: social learning, social influence, and social support. Parents can be channels of all

41

these mechanisms. Children are in their fertility decision either positively or negatively affected by their own childhood experiences (social learning). In addition, children having good relationships with their parents are also influenced by their parents’ attitudes and values (social influence), and they are often provided with support (social support). Some authors (Liefbroer and Elzinga 2012; Murphy and Knudsen

2002; Murphy 2013) in their research showed that fertility behavior is transmitted between generations of parents and children. Despite substantial demographic changes in contemporary societies, that are observable even between two subsequent generations, children still follow same behavioral patterns of their parents. Some authors also showed that children tend to decide about their parenthood in compliance with their parents’ expectations (Axinn, Clarkberg and Thornton 1994; Barber 2000).

Other studies indicate the importance of parents’ willingness and opportunities to provide intergenerational transfers for intentions of their children to have a second or subsequent child, especially in terms of care (Aassve, Meroni and Pronzato 2012; Bühler and Philipov 2005; Del Boca 2002; Kaptijn et al. 2010). Hank and Kreyenfeld (2003) also observed increased fertility for women living in the same town as their parents. Other researchers have shown that parents tend to support children, who have already started their own family more than childless children (Hogan, Eggebeen and Clogg 1993). It can be suggested that parents either intentionally or unconsciously increase the probability of another grandchild’s birth.

Besides the formal childcare services, the informal childcare by other members of family or by acquaintances is sometimes provided especially when the formal services are not available, or they are too expensive, or their quality is not sufficient. Overall, the proportion of caring grandparents is still rather large in most of the European countries and the USA (Fuller-Thomson and Minkler 2001; Guzman 2004;

OECD 2012).

Fergusson, Maughan and Golding (2008) showed that grandparents’ care is complementary to the paid care and it is used especially in case that mother works part-time (because she does not earn enough to pay for a formal care) or if she returns to work early after the birth. Kaptijn et al. (2010) suggest that grandparental care is important particularly in the welfare regimes providing a low level of public care

42

services provision since they observed the important role of grandparents’ care for intention to have another child in the Netherlands where the childcare is mostly secured privately. Grandparents’ care can also balance a not sufficient length of the maternity or parental leave. Some studies investigated the effect of grandparental care on the participation of women on the labor market and they conclude that the role of grandparents is crucial at least in some European countries (Aassve, Arpino and Goisis 2012;

Dimova and Wolff 2011).

Aside from the positive effects of the grandparents’ involvement in childcare on the mothers’ employment opportunities and for intentions to have another child, some effects on children’ development has been also observed. There is, however, a lack of studies on the topic. Barnett et al.

(2010) suggest a positive effect of grandmothers’ care on children’ behavior and social competences.

Deleire and Kalil (2002) found that grandparents’ involvement in childcare can serve as a protective factor in case of children living in non-married families with at least one coresiding grandparent. Dunifon and

Kowaleski-Jones (2007) and Monserud and Elder (2011) suggest a similar effect of grandparents’ coresidence on children in single-mothers families. On the other hand, an excessive involvement of grandparents in childcare can have negative effects for grandparents themselves. The impact of grandchildren care on health of the caring grandparents have been studied several times with ambiguous results. While some researchers suggest the negative influence of active grandparenthood on the physical and mental health (i.e. Fuller-Thomson and Minkler 2000; Minkler et al. 1997; Solomon and Marx 1999), others on the contrary indicate that the effect might be rather neutral (Hughes et al. 2007) or sometimes even positive (Waldrop and Weber 2001).8 Grandparents also often face the conflicting roles that they perform within and outside the family. Employed grandparents often tend to give up their own work and retire early to care of their grandchildren (Hochman and Lewin-Epstein 2013; Wheelock and Jones 2002).

8 It should be distinguished between occasional care of grandchildren and becoming a primary carer of grandchildren (custodial grandparents) or having an adult child with a grandchild coresiding in the same household. The custodial care is usually connected with some difficult family situation (Baker and Silverstein

2008) and a custodial care has specific characteristics.

43

The riskiest for caring grandparents is the conflict of multiple roles such as caring for older family members and grandchildren together with an active participation at the labor market. Previous research has shown that particularly middle-aged women are struggling with competing roles and they tend to suffer from mental diseases (Glynn et al. 2009; Opree and Kalmijn 2012). The effect is probably affected by the voluntariness of grandparent obligations and also by the intensity of provided childcare since some countries do not provide a sufficient supply of formal childcare services and in these countries, grandparents have to secure the childcare on a more frequent basis and might tend to perceive their involvement in childcare as involuntary and enforced by external circumstances with negative impacts on their other opportunities. For that reason, the negative effects of these obligations might be more evident in the countries with insufficient grandchildren services supply. Fokkema, Bekke and Dykstra (2008) suggest that northern European countries are characterized by a less intense and more voluntary help among family members. Hank and Buber (2009) found that the frequency of grandparent care is higher in the northern countries than in the southern countries, while the intensity is higher in the southern countries. At the same time, most of the southern European grandparents perceive their involvement in childcare as a duty, while the northern European grandparents tend to disagree that they are obligated to care of their grandchildren regardless their real involvement in childcare. Igel and Szydlik (2011) associates observed differences to the welfare state arrangements in these countries. Overall, caring of grandchildren is associated with a higher quality of life if grandparents accept their caring role as an obligation and they therefore meet what is expected by themselves and by a society (Neuberger and

Haberkern 2014).

Several implications of these findings can be formulated. First, the intergenerational relations within the family are still of a great importance for fertility behavior since public services cannot cover all needs of families with children. Grandparents primarily play role of caregivers and they substitute for formal childcare services. The middle-aged and older generation is, however, important also as other source of support; it provides the housing, it serves as a financial backup and it is also a source of information and emotional support. Therefore, the investigation of fertility behavior of a young generation must be inevitably connected with an investigation of the behavior, expectations and roles of 44

older generation. The supporting role of a middle aged and older generation might be in a conflict with their other roles within and outside the family. For example, the investigation of pension system and transitions into retirement is important since the participation of the older generation on the labor market can limit their opportunities to be involved in the childcare and therefore lead to the shortage of informal care. Moreover, the middle-aged people and younger senior age group might be involved also in multiple caring roles since they are the main providers of care and support to dependent older and handicapped family members.

The focus of this study is on young people of childbearing age and their fertility behavior. However, the fertility decision-making process is approached from the social network perspective since the decision is not made independently on other people related to the potential parents. The study does not aspire to cover the whole concept of social influence on fertility behavior since the goal is to bring the internationally generalizable information using the quantitative methods of analysis. The conceptual framework of this study is limited to one aspect of social network influence which is the intergenerational solidarity. The main actors are parents as the main providers of support and their adult children. Besides these two generations, other two generations should be inevitably considered in the process of the intergenerational solidarity examination. From the perspective of middle ‘pivot’ generation, it is the youngest generation of grandchildren and older generation of ageing parents. These four generations are interacting within each other and the dynamic of relationships and support across all generation is important.

The thesis will cover the following subtopics and questions:

1. The normative attitudes of parents toward their grandparents’ obligations as a predictor of the

fertility behavior of their children: What is the relationship between latent and manifest solidarity

across European countries? Is the probability of childbirth higher for children of parents who

express a strong agreement with the normative obligations connected with the grandparent role?

45

2. The effect of income on fertility intentions: What is the effect of subjective perception of

household´s income on women´ short-term fertility intentions? What role does education play in

this relationship?

3. The influence of financial transfers from parents to their adult children on the probability of

childbirth: What is a general effect of income and financial situation for a decision to have a

child? Do children who have been in the past financially supported by their parents tend more to

have a child? What is the effect of financial transfers on childless children and children who

already have one or more children?

4. Financial transfers and help in form of childcare: Which kind of transfers is more important? Are

these kinds of transfers of the same importance for men and women?

5. The care of grandchildren and ageing parents: Do people who help their ageing parents care less

for their grandchildren?

4. Methodological approach

The goal of the study is to investigate various forms of intergenerational solidarity and to test the relationship between individual fertility behavior and the intergenerational solidarity. Moreover, the findings should be comparable and generalizable for different cultural and political contexts. The analyses will be therefore based on the international representative datasets.

Beside the international and representative character of the data, the causal processes behind the observed relationships should be observed whenever it is possible. For that reason, a longitudinal data will be used for a part of the investigation. Furthermore, data should provide information about four generations: middle aged people, their parents, their adult children and grandchildren. Because of these requirements, the substantial part of the analysis (Chapter 5, 7 and 9) is based on the data from the

Survey of Health, Ageing and Retirement in Europe (SHARE). These multidisciplinary and international data are based on the longitudinal investigation of people aged 50+ in twenty European countries and 46

Israel. The sampling was based on the households with at least one member born in 1954 or earlier (it applies for the first wave in 2004).

The first wave was conducted in 2004 and the data collection of the latest wave has been finished in 2015. So far, five regular waves of the panel survey have been completed (2004, 2006, 2010, 2012 and

2014), plus one wave of the retrospective survey (SHARELIFE) in 2008. The data contain information about respondents but also information about their children, grandchildren and parents. It is possible to follow the transitions in children’ life across waves, including their fertility behavior. The data also contain two modules focusing on the help between respondent and other people. The first of them focuses on the social support, while the second one collects the information on the financial transfers which respondents provided or obtained from relatives, friends and others in last twelve months.

As has been said, the main respondents are people aged 50 and more. For this study, they will be considered as a middle-aged generation. Exclusion of people who are older is necessary. However, the main interest of this study lies in the younger generation and therefore respondents were chosen for the analysis not based on their age but based on their children’ age. A substantial part of the analysis considers children who are older than 16 years and younger than 50 because only minority of people outside this age range shows a childbearing activity. To be able to follow children across waves and to control for the most important characteristics such as age, gender and education, the dyads of the parents and children have been created. The basic unit for a substantial part of the analyses is a pair of a child and his parent. In most of the cases, the questions about children were answered only by one of the parents, the other one is excluded from the analysis. Every parent could provide detailed information about up to four children and all of them were included in the final dataset. The character of the data implies that observations on children are not independent and could cause distorted results. If the clustered data are treated the same way as independent data, the standard errors might be underestimated (McCoach and Adelson 2010). There are two possible solutions, and both will be used in the analyses. Firstly, the multilevel analysis has been conducted with a child as the first level of analysis and the family as the second level. The second option is an adjustment of standard errors by clustering of

47

the children of the same parent. This option will be preferred, if possible, since it is much less computationally demanding and multilevel analysis may be performed only under several assumptions.

Because of the requirements of multilevel analysis, the country effects will not be investigated as a separate level of analysis. The total number of countries in SHARE datasets is twenty. Only twelve of them participated in the first wave of survey: Denmark, Sweden, Austria, Germany, Switzerland, Belgium,

France, Netherlands, Spain, Italy, Greece and Israel. In the second wave three more countries joined the survey: Poland, Czech Republic and Ireland. Four more countries participated in the third regular wave:

Estonia, Hungary, Portugal and Slovenia but Greece, Ireland and Israel dropped from the survey. Finally, the data collection for the last wave completed fifteen countries with one new country (Luxembourg) and

Hungary and Portugal dropped from the survey. The twenty countries therefore participated in total across first four waves, but only a few of them participated in all waves and some of them participated only in one wave so they have not joined the panel research yet. Apparently, a real number of countries which can be used for the panel data analysis is much lower than twenty, varying between twelve and fifteen, depending on the chosen waves. A suggested number of countries for a multilevel analysis is usually much higher, for example, Bryan and Jenkins (2015) suggest that minimal number of countries for linear models is around 25 and 30 for logit model. Most of the analyses in this study employ country dummy effects as a control variable but not as a separate level of analysis and no country-level variables are included. Whenever appropriate, the individual-level analysis is supplemented by a descriptive statistics of country differences.

Since the SHARE is not sufficiently comprehensive to study all suggested research questions, two other international surveys will be used: European Social Survey (EES) in Chapter 6 and Generations and

Gender Programme (GGP) in Chapter 8. European Social Survey is not a panel research; however, its main advantage is a coverage of a quite broad range of European countries. Another advantage is the fact that it covers all adult age groups since SHARE is limited only to people aged 50+ which does not allow to directly study younger people. GGP is a longitudinal survey of people aged 18-79 years, so it also provides data on people in fertile age. Besides that, it also includes several measures of informal solidarity and

48

questions on childbearing planning based on Ajzen´s theory of planned behavior (TPB). This enables to analyze the relationship between the intentions to have a child and a support obtained from significant others. The result can be then compared to the findings based on the analysis of SHARE panel data and real fertility behavior. Fertility intentions and realized fertility are different concepts, as will be explained later, but they are both of a great importance for the study of fertility.

The main general research question is the following: How is fertility behavior affected by various forms of intergenerational solidarity? This question requires a causal explanation. The SHARE data are longitudinal which are perceived as the best kind of nonexperimental data for the casual inference

(Allison 2005). However, neither the theoretical base of the causality between fertility and the intergenerational solidarity, nor the empirical methods for investigations of the causal effects are straightforward. From the theoretical point of view, the causal relationship between fertility behavior and different form of solidarity can be bidirectional. Firstly, a support obtained from parents or grandparents serve as a source of help, finances, housing, emotional support or information and leads to the increased chances of childbirth. The manifestation of the intergenerational solidarity therefore precedes fertility. On the other hand, fertility can also encourage the intergenerational solidarity because grandparents might profit from the births of grandchildren – previous research has shown that grandparents are more satisfied and healthier than older people without grandchildren or grandparents who meet with their grandchildren sporadically (Arpino and Bordone 2014; Di Gessa, Glaser, Tinker 2015; Hughes et al. 2007).

This profit will likely encourage a willingness to support the generation of children and grandchildren.

Furthermore, it has been shown that parents support more their children who already have their own children (Hogan, Eggebeen and Clogg 1993). These findings suggest that there is a reciprocal relationship between intergenerational solidarity and fertility and there is not the one-directional causal chain.

Whenever it is possible, these two directions of causality should be considered. However, it is not possible to draw a completely clear picture of these processes since they are tightly interconnected. The issue is also complicated by the fact that solidarity might be latently present even before the childbirth, but it is manifested only after the birth of grandchildren since it is not required earlier, or it might not be manifested at all, because own resources of children and their offspring are sufficient, and an external 49

support is not necessary. For that reason, an empirical investigation of the intergenerational solidarity will be introduced by an analysis of the normative solidarity. It assumes that solidarity remains latent until two other conditions are met: the support is needed, and resources exist to provide a support. The analysis of the latent solidarity aims to broaden the concept of the intergenerational solidarity and its relation to fertility since most of the previous studies have focused on the realized or manifested kinds of support.

5. The effect of normative intergenerational solidarity on the probability of childbirth

Previous research has shown that young people in the childbearing decision-making process consider many factors (Haas 1974), starting from their own economic situation and position on the labor market, housing and other basic living conditions to a partnership quality. The timing of the parenthood is decided based on expected costs and rewards of childbearing (Liefbroer 2005). Individual opportunities and constraints are, however, not the only factors to consider in an examination of the fertility behavior.

Due to increasing rates of female labor market participation, young parents usually must rely on other sources of help, both informal within a family and formal from external providers. The members of social network constitute the source of social capital and have an impact on the timing of childbearing and the number of children (Bühler and Philipov 2005). Furthermore, significant others play a role not only of a possible source of support, but they also influence the individual’s attitudes and motivation, and exert a social pressure (Balbo and Mills 2011).

The research in family solidarity identified parents as a one of the important sources of support for their adult children (Albertini, Kohli and Vogel 2007; Albertini and Kohli 2012). The decision to have a child can be therefore influenced by the children’ expectations about their parents as a possible source of a financial or social support. The parents can encourage the fertility if they express their willingness to support their offspring in case of need. On the contrary, their expressed reluctance to help can increase

50

the insecurity of their children, and lead to a postponement of childbearing because potential parents must rely on their own resources and they are waiting for more suitable circumstances. Parents can also carry general attitudes toward family and obligations of family members toward their relatives. Children perceiving their parents’ unwillingness to help may be less familialistic than children of supportive parents who believe in the importance of family and reciprocal help between family members. A low level of individual familialism can lead to a voluntary childlessness or a smaller number of children.

The goal of the first part of the research is to analyze the effect of parents’ attitudes toward grandparents’ obligations on the probability that their child will have an own child during the observed period. To examine this effect, the analysis of three waves of panel data from the Survey of Health, Ageing and Retirement in Europe is performed (wave 1, 2 and 4). The goal is to research not an actually provided support but an expected support, which is supposedly perceived by children.

This approach introduces several limitations, but it can also have some advantages in comparison with an investigation of a realized support. First, it allows a consideration of the probability that the first child will be born. If the realized support is considered, it is necessary to research the effect of a grandparents’ involvement in the first child’s care on the probability that the second or another child will be born. The probability of having the first child cannot be observed, because we suppose that childless children do not require a support, or they demand other kind of help. However, a decision to have the first child is substantially different from a decision about a subsequent child and it is taken by people with different characteristics. For that reason, it seems important to distinguish each level of childbearing.

Secondly, an investigation of an expected support does not exclude parents, who are generally willing to help, but they do not provide any support to their children because the children are currently not in a need of help, or parents do not dispose of necessary resources for help. Even these children can be, however, affected in their decision to have a child by their parents’ willingness to help because they know that they can rely on their parents’ support if it is necessary.

There are also obvious drawbacks: first, it is not self-evident that attitudes toward family obligations are in an accordance with a behavior. This relates to the second limitation: it is not clear to

51

which extent children perceive the parents’ attitudes, or if they rather consider a behavior. These limitations will be discussed in a more detailed way in the concluding part of the chapter.

5.1 LATENT SOLIDARITY AND THE OPERATIONALIZATION OF THE TERM

As has been discussed in the first chapter, the intergenerational solidarity exists in two main forms.

Apart from solidarity expressed as a behavior (active support of other family members), it can be approached also as an issue of values, norms and attitudes (e.g. Kalmijn, Saraceno 2008; Silverstein, Gans,

Yang 2006). Latent solidarity or normative solidarity is represented by a perceived responsibility to other people. Previous research has shown that people in the northern countries prefer the support of state, and people in the southern countries usually highlight the importance of the family (Daatland and

Herlofson 2003; Fokkema, Bekke and Dykstra 2008). That corresponds to the welfare state regime arrangements as northern countries with a lower level of normative solidarity usually provide a strong support from the state institutions. However, it is not clear to what extent the weak normative solidarity is a result of a strong welfare state.

On the individual level Dykstra and Fokkema (2011) show that normative solidarity is an important condition for transfers of help (practical assistance) but not so much for financial transfers. Generally, help is perceived as a more demanding form of support. On the contrary, providing of financial transfers can ensure a greater independence and autonomy of each family member, but it more depends on the resources that a supporter has, e.g., parents can strongly perceive their obligations toward their children, but they do not dispose of sufficient financial resources for the realization of a support. The relationship between a normative solidarity and an actual provision of support has been investigated on longitudinal data also by Herlofson et al. (2011) in Norway and the Netherlands and they found a positive relationship between attitudes toward filial responsibility and an actual support. The actual support is, however, highly connected to a structure of need, which is influenced not only by a family but also by institutional arrangements. For that reason, Herlofson et al. (2011) suggest different roles of grandparents in countries

52

with various levels of formal care services. While in the northern countries a higher percentage of grandparents is involved in childcare, but they provide it with a lower intensity, it is the opposite on the south, where grandparents’ care is necessary because of inaccessibility of services. It seems therefore legitimate to observe not only the influence of actual support but also general attitudes toward a potential support provided by grandparents. The presented study analyzes the effect of normative solidarity expressed as attitudes toward family obligations.

The first two waves of Survey of Health, Ageing and Retirement in Europe (2004-2006) contained a drop-off questionnaire with a battery of questions about family obligations attitudes. Respondents were asked how much they agree or disagree with four statements: “Parents' duty is to do their best for their children even at the expense of their own well-being.” “Grandparents' duty is to be there for grandchildren in cases of difficulty (such as divorce of parents or illness).” “Grandparents' duty is to contribute towards the economic security of grandchildren and their families.” “Grandparents' duty is to help grandchildren' parents in looking after young grandchildren.” Respondents could choose from the following answers: strongly agree, agree, neither agree nor disagree, disagree, strongly disagree.

These four items could measure general attitudes toward parents' and grandparents' obligations.

Since one question considers parents-children relationship, and the remaining include grandparents- grandchildren relationship, the factor analysis was conducted9 to verify, if the whole battery of questions includes one or more latent dimensions of family obligation attitudes. The only factor has been gained by employing the factor analysis. The uniqueness of the first item is, however, slightly higher (0.3625) in comparison with other items. If the factor analysis is conducted independently for separate countries, still only one factor is derived. The uniqueness of the first item varies between 0.2255 in Denmark to 0.4942 in

Ireland (Table 4). These considerable differences could be an expression of different roles of parents and grandparents in different countries. While in some countries can be the grandparent role perceived as a comparatively similar obligation as the parent role, in other countries may be the grandparenthood perceived as a more voluntary commitment rather than an obligation.

9 For conducting factor analysis, the principal component factor method of extraction with Varimax rotation was used, with minimal eigenvalue 1. 53

Table 4: Factor analysis and reliability analysis for family obligations attitudes battery

Factor analysis Reliability analysis (without the first item)

Uniqueness (first item) Cronbach's alpha Austria 0.3402 0.9192 Germany 0.3512 0.8505 Sweden 0.4419 0.9178 Netherlands 0.3549 0.9082 Spain 0.4380 0.8861 Italy 0.1929 0.9446 France 0.3994 0.8839 Denmark 0.2255 0.9490 Greece 0.4423 0.8350 Switzerland 0.4348 0.8714 Belgium 0.3794 0.8923 Israel 0.4020 0.8627 Czech Republic 0.3855 0.7905 Poland 0.3876 0.8759 Ireland 0.4942 0.8688

All countries 0.3625 0.8976 N 29430 29430

Note: Principal component factor method, Varimax rotation, Min. eigenvalue = 1 Source: Survey of Health, Ageing and Retirement in Europe 2004 a 2006

Regarding these results, the first item has been dropped from the battery, and this decision has been further supported by conducting reliability analysis using Cronbach's alpha for the three remaining items. The final figure for the aggregated data set of all countries reaches almost 0.9 (Table 4), and it varies between 0.7905 in the Czech Republic and 0.9490 in Denmark. The index of family obligations can be therefore created as a mean of the three items. The final index is rounded as a 5-items scale.

54

The relevance of the final index in terms of measuring individual rate of familialism can be verified by analyzing its relationship to another battery of questions in the SHARE drop-off questionnaire. The level of familialism is usually discussed as a characteristic complementary to the level of benefits and services provided by the state or market. According to Esping-Andersen (1999), we can distinguish between familialistic and de-familializing regimes. It is supposed that people who adopt greater parents' and grandparents' obligations toward their offspring tend to rely less on a public provision of financial allowances and emphasize more the responsibility of the family. On the contrary, people, who tend to stress parents' and grandparents' obligations less, tend to rely more on the state. We can therefore use the battery of the following questions: ‘In our opinion, who - the family or the State - should bear the responsibility for each of the following: (a) Financial support for older persons who are in need? (b) Help with household chores for older persons who are in need such as help with cleaning, washing? (c) Personal care for older persons who are in need such as nursing or help with bathing or dressing?’ Respondents could choose from answers on a five-points scale: totally family, mainly family, both equally, mainly state, totally state10. The relationship between these three items have been tested and an index for family obligations has been created to verify the validity of using the index of grandparents’ obligations as a measure of individual familialism. The rounded index for family obligations was used (five-points scale).

By computing chi-square test and Goodman and Kruskal's gamma for every item and country independently, it can be concluded that there is a clear relationship between index of family obligations and battery of questions considering the state versus family responsibility, even though a strength of the relationship varies considerably across items and countries. The relationship is weaker in the southern

European countries and Israel. Nevertheless, the chi-square test is highly significant (at least at 0.05 level) for most of the items and according to gamma (Table 5) it is possible to conclude that people who tend to score more on family obligations index also tend more to highlight the responsibility of family over the responsibility of the state. In other words, these people are more familialistic and persuaded about their

10 This scale was reversed to be in the same direction as the family obligations index. The highest score (5)

therefore marked the highest tendency to emphasizes the family responsibility.

55

obligations toward other family members. The scale therefore convincingly measures the level of individual familialism. The other part of the scale is less clear. Generally, it might be termed as the de- familialism. However, de-familialistic attitudes might be expressed as a reliance on the state but also as the individualism. People who do not accept the grandparental obligations might perceived that the state should secure the childcare and a financial support for families with small children. On the other hand, it might be also a group of people who think that care of children must be primarily secured by parents themselves and it is obligations of neither the state nor the other family members. That might be the explanation of the limited correlation between the grandparents’ obligations scale and the items of responsibility for help and care of older people since these items distinguish the state and the family.

56

Table 5: Correlation between forms of help and care for older people

Pearson chi-square (Goodman and Kruskal's gamma)

Financial help Household chores Personal care

Austria 46.63*** (0.10) 95.16*** (0.23) 64.55*** (0.20)

Germany 88.68*** (0.17) 143.86*** (0.22) 128.96*** (0.20)

Sweden 124.05*** (0.25) 138.87*** (0.27) 119.24*** (0.25)

Netherlands 70.85*** (0.16) 44.85*** (0.09) 60.79*** (0.11)

Spain 29.54** (0.09) 44.99*** (0.11) 44.93*** (0.11)

Italy 34.94** (-0.03) 30.99* (0.07) 32.78** (0.10)

France 50.73*** (0.11) 44.43*** (0.10) 49.84*** (0.08)

Denmark 64.86*** (0.24) 47.62*** (0.11) 70.10*** (0.15)

Greece 49.69*** (0.05) 53.78*** (0.13) 56.22*** (0.10)

Switzerland 62.56*** (0.21) 77.58*** (0.11) 63.13*** (0.18)

Belgium 111.10*** (0.15) 91.40*** (0.13) 121.71*** (0.16)

Israel 28.26** (0.07) 39.94** (0.08) 23.39 (0.06)

Czech Republic 63.41*** (0.20) 74.79*** (0.23) 123.83*** (0.22)

Poland 57.24*** (0.10) 118.16*** (0.16) 68.93*** (0.16)

Ireland 60.42*** (0.09) 64.52*** (0.20) 45.50*** (0.07)

All countries 647.77*** (0.14) 834.91*** (0.17) 1000*** (0.20)

Significance levels: * 5%, ** 1 %, *** 0.1% Source: Survey of Health, Ageing and Retirement in Europe 2004-2006

Another test of an external validity of the family obligations’ index is the analysis of a correlation between attitudes as a latent form of intergenerational solidarity and realized transfers as a manifest form of solidarity. Table 7 provides the values of Pearson chi-square and Goodman and Kruskal's gamma 57

for the index of family obligations and a scale of grandparents’ care of grandchildren. A frequency of care has been used: almost daily, almost every week, almost every month, less often, never. In this case, only grandparents with at least one grandchild younger than 14 years were considered in the analysis since people who have not became grandparents yet would distort the results. The association exists but it varies more distinctively across countries. However, it is important to note that we do not consider a structure of opportunities and needs. For that reason, people who tend to agree that grandparents should support their grandchildren might not have a chance to really support them by care, either because there are some constraints that prevent them from filling up the obligations (health issues, geographical proximity) or because their care is not needed (parents care about children themselves or other people are involved in childcare). However, generally, there is a positive correlation between normative and manifest solidarity. It can be concluded that family obligations’ index can be used as a measure of normative solidarity toward children in the following analysis.

58

Table 6: Correlation between frequency of childcare and family obligation´s index

Pearson chi-square (Goodman and Kruskal's gamma)

Austria 92.90*** (0.30)

Germany 127.33*** (0.24)

Sweden 50.37*** (0.14)

Netherlands 74.19*** (0.13)

Spain 42.59*** (0.13)

Italy 36.02*** (0.06)

France 39.92*** (0.09)

Denmark 68.14*** (0.14)

Greece 38.02*** (0.14)

Switzerland 22.54 (0.04)

Belgium 63.34*** (0.11)

Israel 40.56*** (0.05)

Czech Republic 44.06*** (0.16)

Poland 33.38*** (0.08)

Ireland 18.26 (0.17)

All countries 403.06*** (0.13)

Source: Survey of Health, Ageing and Retirement in Europe, 2004-2006

59

5.2 MODEL OF THE RELATIONSHIP BETWEEN GRANDPARENTS’ OBLIGATIONS PERCEPTION AND GRANDCHILD’S BIRTH

Based on previous findings, two hypotheses are formulated. First, if parents’ attitudes toward obligations of family members to each other are perceived by their children, it can be supposed that children partially reflect these attitudes in their decision to have a baby since parents can be an important source of support. Furthermore, this process can be supported by transfers of familialistic attitudes between generations of parents and children. For these reasons, it can be hypothesized that children of parents who agree more with attitudes toward grandparents’ obligations tend to have more offspring than children of parents who are less familialistic. Secondly, this relationship between attitudes of parents and fertility behavior of children is causal; the attitudes precede the children’ behavior, which indicates that children are aware of parents’ attitudes and take them into consideration in the fertility decision- making process.

Graph 7 and Graph 8 display aggregate mean values of normative (latent) and manifest solidarity.

The normative solidarity is the mean value of grandparents’ obligations index in 15 European countries in

2006. The manifest solidarity is here expressed as a proportion of grandparents involved in childcare in

2006. It is obvious that there are not striking differences in latent solidarity over the participating

European countries (Graph 7). As expected, the familialistic southern countries show the highest agreement, and, on the contrary, the individualistic countries such as Denmark and the Netherlands reach the lowest average score.

60

Graph 7: Attitudes toward grandparents´ obligations

Source: Survey of health, ageing and retirement in Europe 2004-2006

The result is quite different for the manifest solidarity (Figure 8). The differences are more distinctive in this case as the lowest proportion of caring grandparents is in Spain (about 30%), and the percentage for Ireland and the Netherlands is almost two times higher. A remarkable finding shown on the first and second graph is a discrepancy between manifest and latent solidarity in some countries.

Spain reaches one of the highest score of agreement with normative intergenerational solidarity, but it shows the lowest proportion of a realized support. On the contrary, the Netherlands and Denmark have a high proportion of grandparents involved in childcare (over 50%), but the grandparents do not so much agree with the statements about grandparents’ obligations. It is, however, necessary to note that an intensity of childcare is not considered, and the descriptive analysis also does not control for the needs and opportunities of grandparents, children and grandchildren.

61

Graph 8: Grandparents involved in childcare, percentage

Source: Survey of health, ageing and retirement in Europe 2006

The multivariate analysis has proceeded in two steps. Firstly, a cross-sectional analysis of data from

2004 and 2006 was performed to investigate the effect of respondents’ attitudes toward family obligations on a number of his or her grandchildren11. The parent-child dyads have been created to enable controlling for children’ and parents’ characteristics, including the following variables: parent’s age, education (ISCED-97) and subjective indicator of income; child’s age, education, gender, marital status and employment; geographical distance of the child from parents’ house and country.

The results of multinomial logistic regression (Table 7) suggest a significant relationship of parent’s attitudes toward grandparents’ obligations and the number of grandchildren of a child. This however applies only to the difference between childless children and children with at least one child. A higher number of children is not significantly related to respondents’ attitudes. Children of respondents who don’t agree with normative statements tend to be more often childless in comparison with children of respondents who express a stronger agreement with the statements in grandparents’ obligations battery.

11 Childless respondents were dropped off the analysis.

62

The relationship between attitudes and number of grandchildren exists even after controlling for both parent’s and child’s age and other influential variables. The other factors influence the number of children in an expected way. The number of children is positively related to age and childless children have a higher education, live with their parents in the same household, and they are never married. On the contrary, having a child is connected with a part-time working and being out of the labor market. The effect of parents’ income is also interesting since parents who manage easily on their income tend to have more grandchildren than parents dissatisfied with their income. This can indicate the validity of the assumption that young people do not consider only their own resources but also resources of other family members. This topic will be elaborated in Chapter 7.

Based on the first part of the analysis, it is possible to conclude that there is a clear relationship between respondents’ attitudes toward obligations of grandparents and the probability that respondent has a grandchild. The cross-sectional analysis, however, does not reveal the causal effect of this relationship. Do children of parents with a strong belief in the role of grandparents tend to have children because they know they can rely on their parents’ help in case of need? Or do the parents’ change their attitudes toward their grandparents’ obligations only after they have had their own grandchild?

63

Table 7: Attitudes towards grandparents´ obligations and number of children, models of multinomial logistic regression

Base: 1 Child Childless 2 Children 3 and more RRR RRR RRR Grandparents’ obligations (ref. Strongly disagree) Disagree 0,692* 0,843 1,783* Neither agree nor disagree 0,532*** 0,878 1,891* Agree 0,421*** 0,883 1,784* Strongly agree 0,365*** 0,904 1,713 Age parent (ref. 40-60) 61-70 1,052 1,296*** 1,281*** 71 and more 1,016 1,462*** 1,768*** Education parent (ref. ISCED 0-1) ISCED 2-3 0,987 0,891** 0,753*** ISCED 4-6 1,012 0,862** 0,815** Income parent (With great difficulty) With some difficulty 1,125 0,908 0,825* Fairly easily 1,162* 0,940 0,857 Easily 1,343*** 0,995 0,891 Age child (ref. 16-25) 26-35 0,342*** 1,839*** 2,142*** 36-50 0,186*** 3,777*** 8,196*** Education child (ref. ISCED 0-1) ISCED 2-3 1,169** 0,969 2,127*** ISCED 4-6 1,491*** 0,980 8,230*** Gender (ref. Female) Male 0,889** 1,040 0,891* Geographical proximity (ref. Same household) Same building 0,333*** 1,247 1,023 1-100 km 0,374*** 1,216 1,053 More than 100 km 0,498*** 1,028 0,956 Child's marital status (ref. Married) Separated/divorced/widowed 0,991 0,571*** 0,493*** Never married 13,523*** 0,280*** 0,127***

64

Table 7 continued

Child's employment (ref. Full-time employed)

Part-time employed 0,483*** 1,220** 1,475***

Unemployed 0,776** 0,973 1,364**

In education 2,464*** 0,924 1,326

Out of labor market 0,333*** 1,347*** 2,485***

Country (ref. Austria)

Germany 1,286* 0,885 0,866

Sweden 1,745*** 1,852*** 2,626***

Netherlands 2,448*** 1,256* 1,502**

Spain 1,943*** 0,878 0,470***

Italy 2,133*** 0,711** 0,346***

France 0,901 1,511*** 2,476***

Denmark 0,598*** 1,908*** 2,764***

Greece 2,737*** 1,561*** 0,776

Switzerland 3,161*** 1,444** 1,895***

Belgium 1,621*** 1,102 1,440**

Israel 1,475** 1,461** 4,683***

Czech Republic 0,738* 1,503*** 1,054

Poland 0,742* 1,324* 1,907***

Ireland 1,715** 1,192 2,710***

N 48872

Clusters (households) 15549

AIC 83456

BIC 84511

Significance levels: * 5%; ** 1%; *** 0.1% Note: Parent-child dyads. Standard errors clustered for households. Source: Survey of Health, Ageing and Retirement in Europe, 2004-2006

65

To investigate the causality between parents’ attitudes and their children’ childbearing behavior, the second part of the analysis has been conducted. Even though respondents answered the question about family obligations only in one wave (either in 2004 or 2006), so it is not possible to control for the change in their attitudes after the grandchild’s birth, it is still possible to use the panel data from the

SHARE and analyze the opposite direction of causality: the effect of attitudes on the probability that a child will be born between two waves of the survey. For that reason, another wave of the SHARE was used (2010) and the effect of respondents’ attitudes in 2004 or 2006 on their children’ childbearing activity until 2010 has been investigated12. The results of logistic regression (Table 8) suggest that the effect of parents’ attitudes is similar as in the previous model, but it is not so strong (statistically not significant). This may indicate that the causal effect can be opposite; parents tend to change their attitudes when they have grandchildren. To verify this idea the interaction of attitudes and a dummy variable for respondents with no grandchildren and at least one grandchild has been added to the model.

The interaction is significant, and it indicates that grandparents’ attitudes influence the probability of the grandchild’s birth, but this only applies for the first grandchild in the family. The effect of attitudes decreases for the second and a subsequent grandchild.

The decision about having the first child substantially differs from the decision about other child, since it is a decision about the parenthood itself. For that reason, it can be suggested that parents´ attitudes are more important for childless people who decide to either start or not to start their own family. On the contrary, those who have already decided to have a child have a real idea of the parenthood and of the support they can get, and they can decide about other child on the basis of their previous experience. The attitudes are not always correlated with a real behavior and those who already have a child can consider the real support that might be provided or not provided regardless the attitudes. Another possible explanation is the change of the parents´ attitudes once they have own grandchild. The grandparent role is linked to some benefits and costs, which can be fully considered only

12 Dummy variable indicating if a new child was born to a child.

66

after having the first grandchild. Expressing attitudes about a general role of grandparents is substantially different than expressing attitudes about own grandparenthood.

There is also an alternative explanation which is related to the fact that grandparents have only limited resources of support (time, energy, finances). For that reason, supporting one child and his or her offspring can exhaust available sources. Even though parents accept their role, and they are persuaded about the importance of grandparents’ support, children can be aware of the fact that they will share time and energy of their parents with their siblings’ families, and their willingness to start a family decreases.

67

Table 8: Attitudes toward grandparents´ obligations in 2004/2006 and childbirth in 2010, models of logistic regression

Odds ratio Odds ratio Grandparents’ obligations (ref. Strongly disagree) Disagree 1,178 1,651 Neither agree nor disagree 1,162 1,890* Agree 1,190 2,252** Strongly agree 1,340 2,522** Siblings have children 2,899** Grandparents’ obligations # Siblings have children Disagree 0,406 Neither agree nor disagree 0,312** Agree 0,260** Strongly agree 0,279** Age parent (ref. 40-60) 61-70 0,815** 0,823** 71 and more 0,340*** 0,347*** Education parent (ref. ISCED 0-1) ISCED 2-3 1,071 1,068 ISCED 4-6 1,092 1,088 Income parent (With great difficulty) With some difficulty 1,031 1,034 Fairly easily 1,243 1,241* Easily 1,295* 1,300* Age child (ref. 16-25) 26-35 3,627*** 3,641*** 36-50 1,547 1,583 Education child (ref. ISCED 0-1) ISCED 2-3 1,471*** 1,471*** ISCED 4-6 2,393*** 2,365*** Gender (ref. Female) Male 0,739*** 0,743*** Geographical proximity (ref. Same household) Same building 2,358*** 2,414*** 1-100 km 2,506*** 2,555*** More than 100 km 2,880*** 2,927***

68

Table 8 continued

Child's marital status (ref. Married) Separated/divorced/widowed 0,496*** 0,496*** Never married 0,638*** 0,618*** Child's employment (ref. Full-time employed) Part-time employed 1,544*** 1,540*** Unemployed 0,911 0,912 In education 0,310*** 0,396*** Out of labor market 3,354*** 3,380*** Country (ref. Austria)

Germany 1,146 1,153 Sweden 1,784*** 1,807*** Netherlands 1,472* 1,497** Spain 1,589** 1,594** Italy 1,677** 1,655** France 1,816*** 1,827*** Denmark 1,943*** 1,995*** Switzerland 1,144 1,136 Belgium 1,288 1,304 Czech Republic 1,168 1,185 Poland 1,430* 1,460*

N 13073 13073 Clusters (households) 6353 6353 AIC 10286 10241 BIC 10563 10555

Note: Parent-child dyads. Standard errors clustered for households. Significance levels: * 5% ** 1% *** 0.1% Source: Survey of Health, Ageing and Retirement in Europe 2004/2006 and 2010

69

5.3 SUMMARY AND DISCUSSION: THE LATENT SUPPORT FROM PARENTS AS AN ENCOURAGING FACTOR TO FERTILITY

The decision making about the childbirth is not a strictly individual or couple process. People who decide about parenthood gain information, adopt attitudes and strategies and consider potential sources of support from their social-network´s partners. The parents seem to be one of the most important members of young people’s social network since they play multiple roles in encouraging or discouraging of the fertility behavior. Previous research has shown that parents are role models for their children because children follow parents’ childbearing patterns; parents also transmit family values and attitudes, and they are an important source of information and support in case of need. It has also been shown that children consider previous help provided by their parents when planning to have another child.

The limitations of previous research lie in either not considering of the first childbirth or not using the representative data from a longitudinal research, allowing for a consideration of different characteristics of both parents and children and controlling for causal effects. To overcome these restrictions, the analysis of the attitudes toward grandparents’ obligations has been performed. The results suggest that parents’ attitudes are in a relationship with a childbearing activity of children. The role of attitudes is, however, causal only in case of having the first child. The probability of having the second or a subsequent child is not significantly influenced by the parents’ tendency to agree with grandparents’ obligations toward their grandchildren. This holds true also after taking into consideration a wide range of parents’ and children’ characteristics. The suggested idea that children do not consider only their own resources, when making a childbearing decision but also their parents’ resources, is further supported by a significant effect of parents’ subjective satisfaction with their financial income.

Based on present findings, it can be concluded that parents’ willingness to support their children and grandchildren might be perceived by their children, and they take these attitudes into consideration when making their fertility plans. It is however necessary to note that these findings are not sufficient to clarify, how are parents’ attitudes translated into children’ behavior. The attitudes can serve as a signal

70

for children that their parents are willing to support them. Nevertheless, the attitudes can be also transmitted into parents’ behavior, and children might orient their behavior according to their parents’ behavior. For example, parents might support their children by a shared housing or financial help even before the childbirth. The attitudes toward grandparents’ obligations can also represent a general level of familialism (i.e. the general inclination to family values) which is shared between a generation of parents and children.

Overall, the findings implicate the importance of the perception of security which has to be established before the childbirth. Bringing a child to a family which is familialistic and supportive is more secure than bringing a child to a family which tend to rely on external sources of support or highlight an individual and independent management of major life transitions. It does not necessarily imply that the support in familialistic families will be realized but a willingness to provide a support likely contributes to a feeling of security and acceptance. From that point of view, it seems important to encourage and maintain relationships within the family.

It has also been indicated in this chapter that the latent solidarity is related to the manifest solidarity. It means that people who accept the grandparental role as an obligation are also more likely to realize their obligations. For that reason, the effect of a realized support on the fertility behavior can be expected. As has been indicated in the first section, one of the forms of realized support are financial transfers. Supposedly, children who are financially supported by their parents or other relatives are more willing to have a child. This assumption is based on the perception of the childbirth as costly event of a long-term character. Children can be more affordable and less risky for people with a higher income. The next section deals with the association between income and related financial concepts and fertility intentions. Then, the relationship between financial transfers from parents to adult children and fertility is tested.

71

6. The effect of income on childbearing intentions

In developed countries, having a child is usually the result of a planned and rational behavior. In most cases, childbearing does not occur as a spontaneous act, but potential parents consider the benefits and costs. Regarding high and unpredictable costs of children and the long-term and unrecoverable character of childrearing, a child is perceived as a risky investment (Gete and Porchia 2014). Following this logic, many previous studies worked on an assumption that people who dispose of a higher amount of financial resources can afford to have more children. However, the results of previous research are ambiguous. According to Jones, Schoonbroodt and Tertilt (2008), the prevailing part of studies in economics found a negative correlation between income and fertility. The complexity of the relationship between individual economic factors and fertility have been suggested also by sociological studies. The inconsistency can be demonstrated by ambiguous results of investigation in different countries. A positive effect of women´s income has been observed in Sweden, Denmark or Finland (Andersson 2000;

Andersson, Kreyenfeld and Mika 2009; Vikat 2004), but not in Germany (Andersson, Kreyenfeld and Mika

2009) and Norway (Kravdal 2002).

The following chapter aims to contribute to this debate by analyzing the relationship between a subjective perception of income and short-term childbearing intentions using cross-national data. It builds on the findings from previous research, particularly on a suggestion that the effect of income and labor market position on fertility is not the same for all groups of people. Fahlén (2013) shows that the effect of unemployment is not the same for people with a low and high education and in different European countries. A similar effect can be expected for income and education since these factors are correlated. It is not sufficient to only account for education as a control variable since the direction and strength of the income-fertility relationship may vary strongly across educational group and the overall effect can be therefore misleading. Income has a not equal weight in fertility decision-making process for lowly and highly educated people since they don´t share the same motivation, opportunities and constraints.

72

The focus here is on financial costs of childbearing, but the financial costs are inevitably connected with non-financial costs in form of time or opportunities. These expenses are higher for people (in most cases women) who earn more (see e.g. De La Croix and Doepke 2003), but also for those who are more ambitious about their future activities on the labor market since they lose their human capital while they take care of children. The substitution effect occurs in the fertility decision making process (Sommer

2014). It means that children are generally more affordable for people with higher earnings, but these people face higher opportunity costs if they have to reduce their paid job. This is true especially for women who usually spend more time with children and they are forced to interrupt their career.

However, low education is usually connected with low future working prospects and having a child might be a strategy of reducing the uncertainty (Kohler and Kohler 2002). On the contrary, more educated women might be motivated to postpone the childbirth until they reach a stable position on the labor market. Furthermore, there is a trade-off between quality and quantity of children (Becker 1960; Borg

1989; Becker and Lewis 1973). Since highly educated people demand quality of children more, they are prepared to invest more into their education, leisure time activities or health care and therefore they have higher demands on their income before starting a family. This leads to a hypothesis that a low income reduces the fertility among highly educated women, while it is less discouraging factor for lowly educated women or it has even positive effect for them since having a child might be an alternative strategy of reducing the future uncertainty.

To provide more robust conclusions, various indicators of household´s financial situation are used.

Firstly, the effect of a subjective perception of current household´s income is tested. Secondly, we analyze the effect of financial constraints during the past three years. The third indicator of a financial security considers how difficult would be for respondents to borrow money in case of need. These three indicators are used to account for the risk connected with fertility behavior since children represent “a durable good of irreversible nature” (Sommer 2014). It is necessary to consider the current income as well as the past savings and debts and financial risk in the future.

73

The chapter is organized as follows. The first section provides a brief review of previous research in sociology and economics, focusing on the effect of income on fertility and possible confounding and intervening variables. Then, the key concepts are identified and operationalized based on the fifth wave of the European Social Survey. Finally, the results of logistic regression analyses with interaction effects are presented.

6.1 THE ASSOCIATION BETWEEN ECONOMIC FACTORS AND FERTILITY

The connection of fertility end economic factor is pronounced on both the micro and macro level.

Fertility follows the business cycle and it tends to decline after the period of an economic recession or stagnation causing the rise of unemployment and insecurity on the labor market (Neels, Theunynck and

Wood 2013; Sobotka, Skirbekk and Philipov 2011; Testa and Basten 2012). Most of the studies on the association between economic (un)certainty and fertility focus on the effect of the employment.

However, regarding the labor market position, the earnings are not the only important factors and also the stability of employment affects the timing of fertility for both men and women (Lundström and

Andersson 2012). Fahlén and Oláh (2015) suggest that economic uncertainty related to the job insecurity leads to a decline in fertility. Bratti and Tatsiramos (2010) provided a comparison of the effect of a labor market attachment on a transition to the second birth across European countries. They found a positive effect due to an increased income of people who are more active on the labor market. However, a positive effect of the labor market attachment and income was found only in countries that provide childcare services and part-time opportunities such as Denmark and France. Similar studies by Adsera

(2004; 2005) showed a negative effect of high unemployment rates and unstable employment. Women in southern countries tend to postpone the childbirth if they are unemployed or if they don’t have a stable position on the labor market because of the risk of a lower income in the future. Barbieri et al. (2015) also found a negative effect of unstable employment for women in Italy and Spain but not for women in

74

Germany and the United States. They attribute this effect to weak welfare state in the southern countries and to inconvenient labor market arrangements.

Angrist and Evans (1996) suggest that intentions to have children in fact precede the efforts on the labor market since people with intentions to have third child work and earn less. These family-oriented people have lower incentives to increase their human capital by obtaining more working experience since they focus on their family life and are prepared to interrupt or abandon their job. Income can be therefore the cause of orientation toward family but also the consequence of this value orientation.

The effect of value orientation toward family or toward work is more important for women than for men since women are those who must leave their job at least for a short period of time. A clear evidence about a different impact of income and employment on fertility for men and women exists

(Kreyenfeld 2005; Schmitt 2012; Tölke and Diewald 2003). Generally, authors mostly highlight the role of women´s wage (e.g. Borg 1989; Day and Guest 2016; Heckman and Walker 1990). An increase in female wages can lead to a delay in childbirth (Kalwij and Gustafsson 2006; Heckman and Walker 1990; Mills et al. 2011) due to high opportunity costs but this effect can be reduced by an external childcare services

(Martínez and Iza 2004; Yasuoka and Goto 2011).

Previous research has also suggested that the effect of income and labor market position differs for the first child and children of a higher order. People with a higher income tend to postpone the first childbirth bud they have a higher probability of intentions to have a third child (Bühler and Philipov 2005;

Bühler 2004).

To conclude, the most important factors that influence the effect of economic uncertainty are the following ones: gender, labor market position, country, order of child. Finally, there is an effect of education. The following chapter deals with relationship between income and education and between education and fertility. It is suggested that education is a confounding factor in a relationship between income and fertility.

75

6.2 EDUCATION, FERTILITY AND FAMILY GAP

The education is one of the crucial factors affecting the relationship between income and fertility since there is a clear positive relationship between education and income. However, the relationship between education and fertility is much less clear since the numerous studies on this topic do not reach to unambiguous conclusions. The direction and the strength of the relationship between fertility and education depends on the operationalization of both fertility and education. If, for example, the women’s educational enrolment and the timing of first birth is considered, the effect is clearly negative (Hoem

2000; Kravdal 1994; Ní Bhrolcháin and Beaujouan 2012). However, other studies have showed that this negative effect exists mainly due to the postponement of childbirth and if the whole fertility period is investigated, the negative effect of education disappears or even becomes positive (Kravdal 2008; Kravdal and Rindfuss 2008; Liefbroer and Corijn 1999). Furthermore, several studies have indicated a positive effect of women’s education on higher order births (Gerster et al. 2008; Kreyenfeld 2002; Oláh 2003). It can be concluded that in European countries that enable the work-family reconciliation the effect of women’s education tends to be rather positive.

However, a better education also generally implies a better position on the labor market and higher earnings. It has been repeatedly shown that women who interrupt their career loose not only the income covering the period of maternal or parental leave but they future earnings are reduced due to a loss of working experiences and opportunities. The gap in pay between people (primarily women) with children and childless people is usually called the “family gap” in literature. According to Gupta and Smith

(2002), a loss of human capital is the main reason of the family gap. Budig and England (2001) found that the loss of working experience account for approximately one third of the family gap, but they also found indications of an opposite effect. The family gap does not have to be only the result of loss during the maternity leave, but it can be also the reason of low previous earnings since women with intention to become mothers tend to prefer jobs that are paid less. The family gap is strong particularly in the private sector which is not family-friendly (Nielsen, Simonsen and Verner 2004). Phipps, Burton and Lethbridge

76

(2001) add another explanation of the family gap which is the reduced working time due to the necessity to care for children. The wage gap of parenthood can be reduced by family policies aimed at reducing of the costs of childbearing, such as maternity leave (Waldfogel 1998).

Regarding the association between the wage gap of mothers and their education, for example Todd

(2001) suggests that the motherhood is more harmful for women with a lower education in some countries since they face a higher wage gap than college graduated. According to Amuedo-Dorantes and

Kimmel (2005) the wage gap is even positive for college educated women who postpone the childbirth of the first child. This is in accordance with the observation that highly educated women tend to postpone the motherhood until they have established their position on the labor market and reached the sufficient wage. There are indications that this postponement can beneficial for them. However, even though a proportional wage gap can be higher for lowly educated women, the absolute loss is higher for women with a higher education and higher earnings.

To conclude, regarding the previous research, the association between education and income is positive and also the association between education and fertility is generally positive if the whole fertility period is followed. However, it is not clear to what extent higher earnings, which are associated with high education, are responsible for a positive relationship between fertility and income and to what extent this association exists regardless the confounding variables. The crucial question is, how income impacts the fertility within each education category. The effect of income on the fertility behavior is probably not the same across educational groups so the overall effect can be either negative or positive as an average across all degrees of education. If, for example, the effect of income is highly negative for women with a lower education and slightly positive or none for highly educated women than the overall effect can be still negative. The purpose of the following analysis is to show that a high or low income motivates differently people with low and high education.

77

6.3 DATA AND RESEARCH QUESTIONS

The analysis will be based on the fifth wave of the European Social Survey data which incorporates the question on fertility intentions within three years. The analysis will be performed only for women which is a usual approach chosen by most of previous studies. Regarding the fertility intentions, when couple plans the fertility, the intention of woman predicts the fertility more reliably then the man’s

(Berrington 2004). Women are those who have more control over contraception (e.g. Darroch 2008) and they are also those who are more impacted by conception, childbirth and who are more involved in childcare.

The sample is limited to 8719 women aged 18 to 40 regardless the number of children they already have. The total number of participating countries in this wave is 27, mostly from Europe but also from

Israel. The logistic regression models are employed, first without interaction and followed by more complex models to explore the interaction between key variables and other indicators.

The main research question is whether childbearing short-term intentions are affected by the feeling about the household’s income adequacy and whether is the relationship of income and fertility the same for all educational groups? Furthermore, two other indicators of financial security are used to consider the robustness of findings regarding different operationalization of economic uncertainty.

6.4 MEASUREMENTS OF FERTILITY INTENTIONS AND INCOME

The indicator of fertility intentions is used in the following analysis. Fertility intentions do not measure directly the fertility behavior and there might be some discrepancies between intended and realized births. However, fertility intentions are a strong predictor of fertility behavior (Schoen et al. 1997;

Ajzen 1991). The advantage of the fertility intention indicator lies in fact that it is possible to distinguish between the deliberate and unintended human behavior. If the real fertility is investigated, it is not

78

possible to say if the birth of a child has been planned and wished or if it has been caused by an insufficient birth control. For example, Becker (1960) supposes that people with a lower income has a lower knowledge of contraception methods. Even though, the general knowledge on the birth control is larger nowadays, we can still suppose some differences in the awareness and use of contraception across educational and income groups in a society. By using the intentions rather than the realized births, the preferences and rational behavior can be tested without an unknown rate of confusion with an unintended fertility behavior. The fifth wave of ESS survey contains the following question: Do you plan to have a child within the next three years? Respondents could choose from the four options: definitely not, probably not, probably yes, definitely yes. The responses were recoded into a binary variable.

The main indicator of income is a subjective perception of income based on the following question:

Which of the descriptions on this card comes closest to how you feel about your household’s income nowadays? Respondents could choose from four options: living comfortably on present income; coping on present income; finding it difficult on present income; finding it very difficult on present income. The four items scale was kept for the analysis. The indicator does not measure the objective level of income but a subjective perception of income adequacy. However, there are some benefits of this concept over the objective measures. Firstly, questions on objective income have usually a large share of missing answers since a part of respondents denies replying. Furthermore, the objective income of household has to be adjusted for the number of people in a household since the costs of living does not increase linearly for each additional member. For that reason, it is not easy to make assumptions about adequacy of income for a household. An alternative way is a consideration of an individual income rather than household’s income but people living with a partner consider also the partner’s income. Finally, a subjective adequacy of income might be more important for the fertility decision because people with a lower income can also have lower demands and vice versa.

The second indicator of financial security is based on the debts and savings in last three years.

Respondents were asked whether they had to draw on their savings or get into debt to cover ordinary living expenses. This indicator takes into consideration that even people who have a sufficient current

79

income can feel financial constraints since they spent their savings in last three years or they had to put themselves in debts and therefore they probably don´t have any savings.

The third indicator focuses on the future financial security and it is based on the question how easy would be for respondents to borrow money in case of serious financial difficulties. In other words, the indicator measures how risky it would be for respondent and his or her household to be found in a situation of financial constraints. Respondents can rely on informal help from members of family or friends but also on formal loans provided by a bank or other financial institutions. It can be hypothesized that respondents who can rely on help from other people or who are not already indebted and can take a loan from financial institution are more willing to have a child in the next three years since they can obtain a help in case of financial problems.

Table 9 shows the associations across three financial indicators calculated by Goodman and

Kruskal's gamma. The strongest is the association between the indicator of debts and savings and present income. People who are satisfied with their present income are the least likely to have relied on debts or savings in previous three years. This association is not surprising: income tends to be persistent over time and furthermore, those people who had to rely on debts in previous years are probably paying this money back and their income is low.

There is also an association between indicator of how easy it is for respondent to borrow money and subjective income. People who are not satisfied with their income mostly replied that it would be very difficult for them to borrow money. It might be because they are already paying back their loans from previous years, so they don´t have enough money left. However, the association between borrowing money and debts and savings in last three years is the weakest (gamma = 0.3321). The indicator of money borrowing may be rather an expression of the social capital´s strength than of the current level of debts.

All indicators are interconnected, and they measure similar phenomenon; however, they don´t measure exactly the same concepts and it is therefore useful to compare their effect on the fertility intentions.

80

Table 9: Associations across individual financial indicators, Goodman and Kruskal's gamma

Feeling about household´s Borrow money to make ends income nowadays meet, difficult or easy

To what extent had to draw on gamma = 0.5599 gamma = 0.3321 savings/debt to cover ordinary living expenses last three years

Borrow money to make ends meet, gamma = 0.4760 difficult or easy

Source: European Social Survey, wave 5 (2010-2011)

Furthermore, an external validity of the indicator of subjective income is tested. To assess the link between the subjective measure of income from the ESS and an objective measure, the country level analysis between the subjective indicator and Actual Individual Consumption indicator (AIC) have been conducted. The AIC is constructed to describe the wealth of households over European countries

(Eurostat 2017c). The indicator is based on the GDP per capita but it also considers the differences in price levels and since it calculates only a real individual consumption it also allows to account for the services that are financed by public institutions in some countries (such as health care and education). The comparison of a subjective perception of income is based on the question from the ESS cited above. The proportion of people who said that it is difficult or very difficult to live on the present income of household has been calculated.

Graph 9 shows the association between the AIC indicator for selected countries in Europe and a proportion of people who find it difficult to make ends meet in these countries. Obviously, these two indicators are in a clear relationship and the subjective perception of income represents quite precisely the objective state of households’ wealth in European countries. A few countries are slightly deviating from the pattern: Slovenia, Croatia, Estonia and in part also Poland. These countries do not reach high on the objective measurement but the proportion of people who find it difficult to live on the present income is also not so high. These discrepancies might be caused by different perception of standards of living in various countries due to a diverse level of demands on goods and services. However, the overall 81

tendency indicates there is an acceptable rate of an agreement between the objective measurement and the chosen subjective indicator.

82

Graph 9: Proportion of subjectively poor people and Actual Individual Consumption (AIC)

Source: Eurostat 2017c, European Social Survey, wave 5 (2010-2011)

83

6.5 CONTROL VARIABLES

As indicated in the first part, the association between income and fertility is disturbed by many factors such as gender, labor market position, country or education but we need to control for even more variables. These factors may completely reverse the association between income and fertility. This is because a subjective income and fertility intentions are influenced by common factors (see Graph 10). For example, living with a partner/spouse increases the probability to plan a child but it also increases the satisfaction with household´s income since it is easier to cope on two incomes. Secondly, having a higher education increases the probability to plan a child but it is also associated with a higher income. The position on the labor market is similar as education: having a paid job increases the probability to have a child but it also leads to a higher probability to start a family. Furthermore, income increases with age and regarding the current postponements of fertility to later age, fertility is also increasing with age (to some extent). Finally, people who are happy and satisfied with their life in general are also more satisfied with their income (Ball and Chernova 2008) and previous research has shown that happiness is associated with more childbirths (Billari 2009). Since most of these factors are positively associated with income and they are also positively associated with fertility, the overall effect of income should be positive. People who are more satisfied with their income are also more likely to plan a childbirth in short time.

84

Graph 10: Factors affecting subjective income and fertility intentions

However, it is not clear to what extent this relationship is mediated by confounding variables and to what extent there is really a direct association between a subjective income and fertility intentions. Do people plan to start a family if they perceive their income as satisfactory because they have enough resources, or it is rather because wealthier people are also more educated, older, employed, have a partner and are happier than people who are less satisfied with their income? In the following analysis, the association between income and fertility is considered firstly without controlling for potential confounding variables and then after controlling for them. The control variables are the following ones: age (18-26; 27-35; 36-40), economic status (employed; unemployed; in education; retired or disabled; at home), partner and his status (partner has a paid job; partner has not a paid job; not have a partner); education (low; medium; high13), happiness (10-point scale); number of children; age of the youngest child (0-3; 4-6; 7-12; 13 and older).

13 Categories of education are based on the classification of education ISCED-97. Low education corresponds to level 0 and 1, medium is from 2 to 4 and high is from 5 to 6.

85

6.6 BASIC MODELS OF FERTILITY INTENTIONS AND SUBJECTIVE INCOME

To be able to follow the effect of controlling for confounding variables, the building of a final model will be presented. For a clarity of presentation, the building process will be followed only for the first indicator of subjective income. The effect of other two indicators will be tested in the final version of the model.

Model 1 (Table 10) shows the association between women´ fertility intentions within three years and subjective income. The effect of income is sharply positive. Women who say that they are living comfortably on present income or they are coping on present income are the most likely to plan a child within three years. The probability of fertility intentions for women who are living on present income with a great difficulty is about 40 % lower than for the wealthiest category.

To account for the effect of education, Model 2 controls for three degrees of educational attainment. As expected, education has a strong positive effect. The likelihood of fertility intentions of women with the tertiary education is more than two times higher than the likelihood of people with the primary education. Controlling for education also leads to the drop in the effect of income. Part of the positive effect of income can be explained by the fact that people who are more educated have on average a higher income. However, the effect of income still exists, and it is positive.

The next model (Model 3) incorporates also the country indicator. As has been shown in the previous section, a subjective income varies strongly across European countries and so does also the fertility rates. Once we control for country, the effect of income is even weaker than in the previous model. However, the category of people who are the least satisfied with their income still differs about 15

% in their fertility intentions from people who are the most satisfied with their income. As has been indicated, wealth increases with age and it is higher for people who are active on the labor and who have partner with a paid job.

86

The indicators of age and labor market position are added in Model 4. Furthermore, this model controls for a cohabitation with partner and whether the partner is working or not. The effect of age is nonlinear. The highest probability of childbearing intentions has been found for the middle category of women between 27 and 35 years. The effect of the labor market activity indicates that the theory of the reduction of uncertainty can be valid. Women who are unemployed and looking for a job are the most likely to plan a child within three years. However, the second highest probability exists for women who have a paid job. On the contrary, as expected, being in education reduces the fertility intentions sharply and quite surprisingly, so does also being in household and taking care of children. However, we don´t control for number of children yet and women who are at home could already accomplished their fertility desires. Regarding the partner´s labor market status, it seems that it is not very important if a partner has a paid job, while it is very important if a respondent lives with a partner whether he has a job or not. The effect of income is now very weak, the first and the last categories of income differ between each other only about 7 %.

Model 5 controls for a quite important and potentially confounding factor which is happiness. This factor is important since we use a subjective indicator of income. By controlling for happiness, we can see that part of a subjective perception of income is mediated by a general tendency of a respondent to see her life positively or negatively. According to results of Models 5, happiness significantly increases the likelihood of fertility intentions. After controlling for happiness, the effect of income is not present anymore.

However, the final model (Model 6) controls also for number of children and the age of the youngest child. Those who already have children are less likely to have another one. The intentions to have another child within three years are the most likely to occur in group of people who have a child aged 0-3 years or 4-6 years. However, the most important change in this model is that the effect of income is now negative. If we control for all factors, women who are the most satisfied with their household´s income are those who are the least likely to start a family within three years. The probability

87

of the poorest women to plan a child is about 25 % higher than probability of the wealthiest women. This effect probably occurred because women who are childless are the most likely to plan a childbirth within three years and these women are also the most satisfied with their income because they don´t have to cover the costs of children.

88

Table 10: Models of short-term fertility intentions; the effect of income: logistic regression

Model 1 Model 2 Model 3 Model 4 Model 5 Model 6

Income (Ref. Comfortably on present income) Coping on present income 0,90* 0,98 1,03 1,06 1,08 1,16** Difficult on present income 0,75*** 0,87** 0,96 1,01 1,04 1,21** Very difficult on present income 0,60*** 0,75*** 0,85 0,93 0,98 1,29** Education (ref. ISCED 0-1) ISCED 2-4 1,53*** 1,48*** 1,42*** 1,41*** 1, 14 ISCED 5-6 2,60*** 2,61*** 2,41*** 2,38*** 1,63*** Country (not shown) controlled controlled controlled controlled Age (ref. 18-26 years) 27-35 years 0,91 0,92 1,55*** 36-40 years 0,20*** 0,21*** 0,58***

Partner (Ref. Partner has a paid job) Partner without a paid job 0,94 0,95 0,97 No partner 0,20*** 0,41*** 0,25*** Main activity (Ref. Paid work) In education 0,49*** 0,49*** 0,45*** Unemployed 1,22* 1,23** 1,18 Sick or retired 0,93 0,94 0,88 In household 0,77*** 0,77*** 1,04 Happiness 1,04** 1,05*** Number of children 0,32***

Age of youngest child (ref. 0-3 years) 4-6 years 2,12*** 7-12 years 1,22 13 and older 0,89 Childless 0,57***

N 8719 8719 87 19 8719 8719 8719 BIC 10651 10486 10578 9830 9834 9042 AIC 10622 10444 10352 9547 9544 8716

Note: Models 3 – 6 control for country effects that are not presented here. Significance levels *** 1%, ** 5%, * 1% Source: European Social Survey, wave 5 (2010-2011)

89

To conclude the first part of the analysis, it can be suggested that the effect of income is positive, but it is positive only because women who are satisfied with their income have specific characteristics that are also related to fertility intentions. They are more educated, have a paid job, live with a partner, don´t have any children yet. Furthermore, they are happier and live in countries with better living conditions (and higher overall fertility rates). However, after controlling for these factors, the effect of income is rather negative. In the next section, two other indicators of financial security will be tested and then the analysis of interaction between income and education will be performed.

6.7 FERTILITY INTENTIONS AND ALTERNATIVE INDICATORS OF FINANCIAL SITUATION

Firstly, the effect of relying on debts or savings to cover the basic needs is tested in Model 7 (Table

11). The impact of this variable seems to be like the previous factor. Relying on debts and savings does not decrease the probability of fertility intentions, but it even increases the likelihood of fertility plans within three years. Women who recently had to draw a lot on savings and debts about 22 % higher odds of fertility intentions in comparison with women who didn´t have to rely on debts and savings at all or almost not at all. The last model (Model 8) tests the effect of indicator how easy would be for a respondent to borrow money in case of serious financial difficulties. The effect is generally weaker in this case and it does not reach the level of statistical significance. Furthermore, the direction of the effect is opposite in this case. Women who say that it would be very difficult to borrow money in case of financial difficulties are the least likely to plan a child within three years. This is quite surprising since this factor is correlated with two others (see Table 11). However, it will still be tested in the next section if this factor interacts with education.

90

Table 11: Models of short-term fertility intentions; the effect of alternative indexes of financial situation; logistic regression

Model 7 Model 8 Draw on savings or get into debt (Ref. Not at all) Some deal 1,04 A great deal 1,21** How easy to borrow money (Ref. Very easy) Quite easy 1,00 Neither easy nor difficult 0,96 Quite difficult 1,01 Very difficult 1,89 Education (ref. ISCED 0-1) ISCED 2-4 1,15 1,11 ISCED 5-6 1,59*** 1,56*** Country (not shown) controlled controlled Age (ref. 18-26 years) 27-35 years 1,53*** 1,55*** 36-40 years 0,57*** 0,58*** Partner (Ref. Partner has a paid job) Partner without a paid job 0,94 1,02 No partner 0,25*** 0,26*** Main activity (Ref. Paid work) In education 0,45*** 0,45*** Unemployed 1,19 1,24** Sick or retired 0,9 0,91 In household 1,07 1,07 Happiness 1,05*** 1,04*** Number of children 0,32 0,32*** Age of youngest child (ref. 0-3 years) 4-6 years 2,13*** 2,20*** 7-12 years 1,24 1,25 13 and older 0,89 0,95 Childless 0,56*** 0,61**

N 8427 8363 BIC 8771 8725 AIC 8454 8394

Significance levels *** 1%, ** 5%, * 1% Source: European Social Survey, wave 5 (2010-2011)

91

6.8 FERTILITY INTENTIONS AND INTERACTION OF SUBJECTIVE INCOME AND EDUCATION

Graphs 11, 12 and 13 display the average marginal effects of income across educational groups on three indicators of financial security. The interaction terms have been added to the final version of model for each indicator (see Model 6, Model 7 and Model 8). The models are not presented here.

Graph 11: Marginal effects of income across education

Source: European Social Survey, wave 5 (2010-2011)

Graph 12: Marginal effects of savings and debts across education

Source: European Social Survey, wave 5 (2010-2011)

92

Graph 13: Marginal effects of how easy to borrow money across education

Source: European Social Survey, wave 5 (2010-2011)

The effect of income varies across educational groups. The effects of all three factors are negative for women with primary education. In other words, lowly educated women who feel that it is very difficult to live on present household´s income are more likely to have a child than lowly educated women who are satisfied with their income (Graph 11). Relying on debts and savings in last three years also increases the probability of fertility intentions for women with lower education (Graph 12). Finally, low educated women who say that it would be not easy for them to borrow money in case of financial difficulties have a higher probability of fertility plans than low educated women who could easily borrow money in case of need (Graph 13). The effect of all three indicators is considerably weaker among middle and highly educated women. After controlling for a set of the most important confounding variables, the significant effect of income exists only for lowly educated women.

Several explanations of these findings can be named. Firstly, it can be argued that respondents have a non-equal level of financial literacy. We can suppose that women with a low education might be less aware of the costs of children and generally think less about the future incomes and expenses.

Furthermore, a gender gap in financial literacy exists since men usually score better in a financial literacy

93

and they are more often responsible for household’s finances (Smith, McArdle and Willis 2010), especially if their education is higher than the education of a female partner (Fonseca et al. 2012). However, it still does not fully explain why lowly educated women are willing to bring another child to a household if they perceive their income as unsatisfactory rather than when they are living comfortably on their present household´s income. Another possible explanation can be that they are going to rely on some external source of money, such as family allowances. To sum up, women with a low education and women with a high education can choose a completely different strategy. Low education is connected with current and future low earnings and in this situation having a child might be a strategy of reducing the uncertainty.

6.9 FERTILITY INTENTIONS AND INTERACTION OF SUBJECTIVE INCOME AND COUNTRY

It was indicated in previous section that the early childbirth can be a strategy of uncertainty reduction. However, if the childbirth could reduce the uncertainty, it should be related to getting some additional income. The public support of families by provision of allowances or services varies considerably across European countries. While Scandinavian countries provide a wide range of support to the vast majority of population, southern countries do not provide a lot of support and families have to mostly rely on their own resources. By comparing of two sets of countries with a high level of family support and a low level of family support we can verify if the theory of uncertainty reduction might be valid. It can be hypothesized that there is a negative or none effect of income in countries with a high level of family support, while there is a positive effect of income in countries with a low level of family support.

Two sets of countries have been selected as follows. Based on previous findings, it can be suggested that support of families tend to be strong in Scandinavian countries and weak in southern countries. According to Saraceno and Keck (2010), Scandinavian countries and France create the clear

94

cluster of de-familizated countries that are characterized by a great state support of families with children. On the contrary, southern and eastern countries such as Poland, Italy, Spain, Greece and

Bulgaria are on the opposite side of the familialistic scale. The responsibility over family members is held within family. The support does not include only financial benefits but also subsidized childcare or flexible labor market legislation. Overall, the costs of children are higher in southern countries because most of the expanses bear the family. Since there is no reliable comparison of all European countries regarding their family support and costs of children14, we will choose just four Scandinavian countries which are well known for their extensive support of families and three southern countries which are on the contrary known for their low level of support to families. The first cluster consists of Denmark, Finland, Norway and

Sweden, and the second is composed of Spain, Greece and Cyprus. It is necessary to have more countries in a cluster since the national dataset are too small after choosing of only women of fertile age. However, the number of people in some categories of a subjective income is still not very high (only minority of women in Scandinavian countries said that it is very difficult to live on a present income), so a binary variable has been created for this part of analysis (0 = living comfortably or coping on present income; 1 = difficult or very difficult to live on present income).

Two separated models controlling for the same set of explanatory variables (same as in the previous analyses) have been made (Model 9 and Model 10 in Table 12). It is apparent that the effect of a subjective income is exactly opposite for southern and northern countries. Scandinavian women who see their income as inadequate have about 34% higher odds of fertility intentions in comparison with women

14 It is basically impossible to provide a reliable comparison of countries in terms of family support since the factors are not comparable between each other and interact with other country-specific characteristic such as religion, culture or labor market arrangements. Castles (2003) suggests that on the spectrum of family policies some of them enable women to leave the labor market in order to care of child (financial benefits) and others on the contrary facilitate the possibilities of mothers to have a job (childcare services). Furthermore, the costs of children vary significantly across European and other countries (DiPrete et al. 2003).

95

who see their income positively. On the contrary, the negative perception of household´s income decreases the odds of fertility intentions about 20% in southern countries. Even though, neither effect is significant, the difference is still very clear and very well visible. The effect of labor market activity is also worth noticing. Being a housewife or caring for children increases the odds of fertility intentions about two times in comparison with employed women in Scandinavian countries. On the contrary, in southern countries the effect of being a housewife is negative in comparison with employment. It has been already suggested by previous research: due to a low support to families and low support of family-work reconciliation, women in southern Europe have to stabilize their position on the labor market before childbirth (e.g. Barbieri et al. 2015; Gonzáles 2006). The effect of other variables in models is similar.

96

Table 12: Models of fertility intentions and income for northern and southern countries

Northern countries Southern countries

Income (High income) Low income 1,34 0,80 Education (ref. ISCED 0-1) ISCED 2-4 2,11*** 0,88 ISCED 5-6 2,66*** 1,27 Age (ref. 18-26 years) 27-35 years 2,07*** 1,84*** 36-40 years 0,49*** 0,88 Partner (Ref. Partner has a paid job) Partner without a paid job 0,92 0,66 No partner 0,24*** 0,12*** Main activity (Ref. Paid work) In education 0,80 0,27*** Unemployed 0,74 0,90 Sick or retired 0,46 0,56 In household 2,14** 0,74 Happiness 1,07 0,98 Number of children 0,23*** 0,24*** Age of youngest child (ref. 0-3 years) 4-6 years 4,82*** 2,54** 7-12 years 2,52** 1,56 13 and older 1,68 1,52 Childless 3,07 0,31

N 1042 1133 BIC 1122 1157 AIC 1033 1066

Significance levels *** 1%, ** 5%, * 1% Source: European Social Survey, wave 5 (2010-2011)

97

The analysis presented here is only tentative because it is based on few countries and the results cannon be straightforwardly interpreted since we do not use any country level variables. The alternative approach is the multilevel random effects analysis. However, the requirements for the employment of random effects are not fully met to provide reliable results. First, the number of countries is 27 and logit models are employed. Bryan and Jenkins (2015) suggest that the sample with number of countries smaller than 30 can face substantial problems especially if the number of country level variables is high.

Moreover, the number of country-level variables represents another issue. To provide a reliable conclusion about the effects of macro factors such as labor market arrangements, the family policies or the costs of children a large amount of comparable information is necessary which is currently not possible to obtain for a sufficient number of countries. The limited number of degrees of freedom and unavailable country-level factors can lead to the omitted variable bias (Möhring 2012). However, the future research can take advantage of multilevel analysis methods and add another piece of knowledge to presented findings.

6.10 FERTILITY INTENTIONS AND INTERACTION OF INCOME AND ORDER OF THE CHILD

Finally, previous research has also indicated that the effect of financial incentives depends on the child´s order (e. g. Laroque and Salanie 2004; Vikat 2004). This is in accordance with the uncertainty reduction theory. Uncertainty is reduced after the first childbirth; however, the utility of the second childbirth is substantially lower and different kinds of motivation are involved. For example, people have another child if they feel that their child should have siblings. Because of different motivation, the perception of income can play a different role for childless people and people who already have a child.

The final part of the analysis investigates the effect of income in interaction with the child´s order.

98

Graph 14 shows marginal effects of interaction between income and order of the child. As has been already shown in the previous part of the analysis, the order of child has a strong effect. The highest tendency of fertility intentions is observed in a group of childless women. On the contrary, women who have two or more children have a low probability of planning to have another child. However, the effect of income is different for childless women, women with one child and women with two or more children.

While the effect of income is negative for childless women and also for women with two or more children, it is rather positive for one-child women. It means that childless women with a not satisfactory income have a higher probability of fertility intentions than women who are satisfied with their income. On the contrary, women who have a child are discouraged from other fertility plan if their income is inadequate.

That indicates that the uncertainty reduction exists only for the first child.

Furthermore, the effect of income in a group of women with two or more children is also noteworthy. Despite expectations, it has a negative effect. Even though, financial constraints are usually named as a factor of smaller families, here we can see that low-income families might be more inclined to have three or more children. Nevertheless, it should be reminded that fertility intentions are investigated here so it is not clear if these intentions will be realized.

99

Graph 14: Marginal effects of income across number of child

Source: European Social Survey, wave 5 (2010-2011)

6.11 SUMMARY AND DISCUSSION: FERTILITY INTENTIONS AND SUBJECTIVE INCOME IN INTERACTION WITH EDUCATION, COUNTRY AND ORDER OF THE CHILD

Despite quite a broad range of literature in economics and sociology that have already tried to provide a clear description and explanation of the economic determinants of fertility, the research in this topic is still far from the definite conclusion. The aim of the presented paper was to show that the relationship between individual financial situation and individual fertility intentions varies considerably across groups of people living in different countries and having a different education degree. However, an interpretation of these results is not unambiguous, and it will desire a further research to clarify the observed relationships and connections.

Nevertheless, the mere fact of the variation in effect of income on fertility across educational groups and countries is important and it has several implications. First, it is not possible to come with a 100

unified theory of economic background of fertility. Fertility intentions are not based on a strictly rational calculation of costs and benefits and in developed countries it is not true that children are simply more affordable for wealthier people or more beneficial for less wealthy people as a future resource of money or work. The connection of fertility and economy is much more complex. The crucial role seems to be played by welfare state arrangements. They disturb the straightforward association between finances and fertility intentions because the social inequalities are influenced by the redistribution of financial resources. Furthermore, welfare states also influence the costs of children, for example, by subsidization of childcare.

On the other hand, the results of presented analysis suggest that even though the costs of children are not only financial and some of the effect of income is mediated by other variables such as education, labor market position, number of children or having a partner, there is still a significant effect of financial income itself. However, the effect does not occur in an expected direction. Rather, it has been suggested that having a child might be a way how to get to an acceptable standard of living for people who don´t have good prospects and who are limited in increasing their income by their own effort. This is indicated both by the negative effect of income in a group of lowly educated women and by the negative effect of income in countries with a high level of family support and subsidies of services for parents.

The presented analysis has a several limitations which must be acknowledged. Firstly, it is possible to see how a subjective perception of income affects the statistical probability of fertility intentions, however, the interpretation is based on theoretical assumptions and it is only tentative. Future research should focus on the explanation of individual motivation, opportunities and constraints which relate to economic background of individual decision-making process over fertility behavior.

Secondly, if we accept the idea that the association between income and fertility intentions is influenced by welfare state arrangements, then it is necessary to account for a fact that the state affects also the initial perception of income. Some people obtain financial help from the state even before the childbirth which is probably the case of Scandinavian countries where a relatively low proportion of

101

people say that their household´s income is insufficient (less than 10%, see Graph 9). We can suppose that the group of people who has an insufficient income in north countries has different characteristics than subjectively poor people in southern or eastern countries. While it is relatively common to have a low income in eastern and southern countries even for people who have a paid job, it is much more unusual in northern countries to face financial difficulties since the state support covers most of the population.

Being subjectively poor affects fertility intentions differently in southern and Scandinavian countries.

Finally, the number of women in fertile age within countries is not high enough to test more complicated hypotheses such as three-way interaction of income, education and country because some categories are underrepresented. However, the effect of income does not have to only vary across educational group but also the interaction of income and education can vary across countries. The future research should focus on explanation of cross-national variability.

7. Financial transfers from the family and its effect on the fertility behavior: the role of child´s order

The previous chapter has dealt with a general relationship between a subjective perception of income and fertility intentions. It has been shown that the direction of the association between income and fertility differs for women with low and high education and it is not the same in southern and

Scandinavian countries. Furthermore, there is an effect of child´s order. It is suggested that not only the current income plays a role in the fertility decision-making process but also the prospective and expected income intervenes. In most of the European countries, women can expect an external financial support after the childbirth. However, as indicated by previous research, formal support can be to some extent substituted by informal support from members of the family. The following chapter therefore deals with the financial help from the parents.

102

In terms of financial help, parents can serve as a financial backup for children planning to start the family, especially if the future income of adult children is insecure. Young people can be less anxious about the costs of children if they know they can rely on their parents if necessary. The financial transfers from parents can therefore encourage the children’ willingness to start a family soon and to have more children.

Realized financial transfers might play a role of a direct support of children with young offspring or with an intention to have a child in a foreseeable future. However, the direct financial transfers not necessarily occur and there still might be some effect of parents’ wealth and generosity on the decision of their children to start a family. Firstly, wealthy parents might not provide regular financial transfers of a lower value, but rather they donate a one-time large amount of money or material gift such as a house or apartment. Secondly, we can suppose that children of richer parents are advantaged by obtaining a better education or because they are not forced to accept a poorly paid job because they can afford to stay jobless for a longer period since they parents can secure their living. Finally, as mentioned above, parents with a sufficient and stable income can represent a financial backup, the source, which is not necessarily used, but it encourages the feeling of security and childbirth might be perceived as a less risky investment. For that reason, the investigation of financial help from parents to children should not be limited to the observation of direct financial transfers. The consideration of the overall financial situation of the family is desirable.

In the following chapter the investigation of the intergenerational solidarity and fertility behavior by an analyzing of the effect of direct financial transfers from parents to their adult children and parents’ income on the probability of the childbirth in the subsequent years. Two waves of data (2006 and 2010) from the Survey of Health, Ageing and Retirement in Europe for twelve European countries are used and the probability of the birth between 2006 and 2010 is considered after controlling for the occurrence of financial transfers from parents to children in 2006, parents’ subjective perception of their income and other most important parents’ and children’ characteristics.

103

7.1 FINANCIAL TRANSFERS AS A FORM OF INTERGENERATIONAL SOLIDARITY

Financial transfers represent a more universal form of support in comparison with other kinds of intergenerational solidarity: it can substitute for a provision of housing for not coresiding children since it can be provided on a long-distance15 (Schenk, Dykstra and Maas 2010); it can replace a direct help (Litwin et al. 2008) and inter-vivos transfers also sometimes represent a substitute for bequests (Arrondel and

Mason 2001). On the other hand, financial transfers are limited by the resources which providers have at disposals. The income and wealth of parents therefore belong to the most important factors determining the probability of providing a financial help to adult children (Albertini, Kohli and Vogel 2007; Berry 2008;

McGarry 1999). As, for example, Ho (2013) indicates, the elderly women even tend to increase their labor market participation if one of their adult children has a relatively low income.

Besides the resources of parents, the characteristics and needs of children influence the probability and amount of financial transfers. Previous research has shown that parents financially support divorced children (Hamon 1995), unmarried children and children in education (Fingerman et al. 2011). The presence of young offspring also positively influences the likelihood of obtaining the financial help from parents (Schenk, Dykstra and Maas 2010). Cox and Stark (2005) observed transfers for housing down payments and they found that the probability of transfers is higher if children plan to start a family and in the same time they are concerned about the housing. Therefore, it might be argued that parents are responsive to children’ needs and they act to encourage the birth of grandchildren. The transfers from parents to adult children with offspring can be provided to children but also directly to grandchildren by their grandparents. Hoff (2007) shows that transfers toward grandchildren tend to represent an occasional support rather than regular financial gifts.

15 Albertini and Kohli (2012) showed that the probability of coresidency is higher in lower-income families.

104

Despite a substantial evidence of parents’ responsiveness to the children’ needs, it is substantively limited by the socio-economic position of parents. Some authors therefore suggest that transfers contribute to the transmission of inequalities within families (Lennartsson 2011; Künemund, Motel-

Klingebiel and Kohli 2005). Nevertheless, these inequalities can be at least partially balanced by public transfers. An extensive evidence of the interaction and complementarity between private transfers within the family and the welfare state regimes exist. The differences across countries in terms of family support are usually explained by the distinctive differences across welfare state arrangements. The northern countries with a strong welfare state show a relatively higher probability and less intensity of transfers between parents and children, while the frequency is relatively lower and the intensity higher in the southern countries with a minor role of the welfare state (Albertini, Kohli and Vogel 2007; Bonsang 2007;

Fokkema, Bekke and Dykstra 2008). The public provision of support therefore does not substitute for the private transfers, but it rather influences the way and intensity in which is private support realized. The southern European countries are often dependent on the support from other family members and this support is intended to cover the basic needs. The support in the northern European countries is usually a supplement to public services and benefits rather than the main source of support.

7.2 THE FINANCIAL RESOURCES AND FERTILITY

Relevant members of social networks for the people undergoing the fertility decision making process are parents. As has been mentioned in the previous chapter, parents tend to support children who already have their own children. It has also been shown by a previous research that support from parents and other members of social network increases the probability of the second or higher order births (Aassve, Meroni and Pronzato 2012; Bühler and Philipov 2005; Del Boca 2002; Hank and Kreyenfeld

2003; Kaptijn et al. 2010). Most of these studies focus on care of grandchildren or other kinds of help. The provision of childcare can partially substitute for the financial transfers because it can replace the formal childcare which is costly in some countries and hardly affordable for young people. There is, however, a lack of evidence on the effect of direct transfers of money on fertility. 105

Some findings indicate the existence of the effect of public financial incentives on the fertility behavior; e. g. Cohen, Dehejia and Romanov (2013) in Israeeli; Bonoli (2008) in Switzerland and Laroque and Salanie (2004) in France. However, the financial incentives have an effect only in low income groups

(Cohen, Dehejia and Romanov 2007). Caldwell, Caldwell and McDonald (2002) conclude that large government investments into fertility social policy could be probably efficient, especially if it is aimed at the harmonization of working and family life. The results of studies on the effect of cash benefits are mixed. While the macro level analyses showed that the direct effect of cash benefits is rather small

(Blanchet and Ekert-Jaffé 1994; Gauthier 2007b; Gauthier and Hatzius 1997), the results of micro-level analysis usually show some impact of benefits, but the effect varies across countries and with respect to the birth order (e. g. Laroque and Salanie 2004; Vikat 2004). The order of child plays a crucial role since the low fertility rate in current European societies are primarily caused by a decline in families with a higher number of children. According to Morgan (2003), there is a tendency to small families rather than no families. Children don’t have an economic function anymore and marginal utility of higher order births is usually not sufficient to balance the additional costs16. The social policies aiming to reduce the costs of higher order births might be therefore efficient for the increase of the total fertility rate.

16 The term “marginal utility” has an origin in the economic approach to fertility. In terms of fertility it expresses the idea that the additional gain from every other birth is lower than the previous. At the same time, the additional costs are higher with every other birth. The “gain” and “costs” are understood in a broader sense here, not only in economic terms. The gain includes all the motivation connected with having a child such as affection for children, reduction of uncertainty, coming up to expectations (both internal and external) or fear of the solitude. The costs are as well understood not only as financial costs but also opportunity costs. In addition, parents tend to focus more on the quality rather than quantity of children in developed countries and higher order births can reduce the investment into each child. See the recent studies on economic conception of fertility, e. g., Cordoba and Ripoll (2015); Hirazawa, Kitaura and Yakita (2014); Jones and Bird (2014).

106

It might be suggested that the efficiency of public financial incentives signalizes that private financial transfers can have an impact on fertility decision-making, unless public transfers fully cover all the necessary costs of childbirth, which is not the case in any European country. Based on previous findings on public financial incentives and benefits for parents and prospective parents, it can be assumed that the effect of family support also depends on the order of birth. Harknett et al. (2014) analyzed the relationship between family support and fertility intentions and found a more significant effect for the second and higher order births. For that reason, a distinction between the first and higher order births will be included in the following analysis. Moreover, the income and other factors related to the financial situation will be controlled for.

The following analysis aims to contribute to the debate over the effect of financial transfers between parents and adult children on the fertility behavior. The main research question is the following:

Do transfers of money from parents to children given for non-specific reason increase the probability of childbirth within four years periods? The use of four-year periods follows from the characteristics of chosen data and it introduces several limitations. Generally, what is tested in the presented analysis is not the effect of financial transfers given for the specifics purposes such as birth of a child, but it rather represents a general willingness and capability of parents to financially support their children. It is supposed that parents who once financially contributed to the budget of their children will be ready to support them also later in their life and the birth of a child might be perceived as a less risky. Since that approach to the fertility decision-making process is based on the relationships between parents and children, the models will control for both the parents’ and children’ characteristics.

107

7.3 MODELS FOR THE EFFECT OF FINANCIAL TRANSFERS ON THE FERTILITY BEHAVIOR

The analysis will be performed on the panel data from the two waves of Survey of Health, Ageing and Retirement in Europe. Data collections took place in 2006 and 201017 in twelve European countries:

Austria, Germany, Sweden, Netherlands, Spain, Italy, France, Denmark, Switzerland, Belgium, Czech

Republic and Poland.

Respondents were asked about the financial transfers which they provided to other people during the last 12 months: Now please think of the last twelve months. Not counting any shared housing or shared food, have you or your husband/wife/partner given any financial or material gift or support to any person inside or outside this household amounting to 250 euros or more? In the following question, respondents were asked to whom they provided money. If they named one of their children, the data allows to connect the other information about this child, so it is possible to follow the impact of the transfers on subsequent transitions. Unfortunately, this connection is possible only in 2006 but not in

2010. For that reason, the information about transfers will be based only on the data from 2006.

The dependent variable is dummy: the birth of a child between 2006 and 2010. Since the dependent variable examines the fertility behavior, it is necessary to include only children who are of the childbearing age. Children younger than 16 and older than 50 were excluded from the analysis because the probability of childbirth is minimal for these age groups.

To be able to control for the characteristics of both children and their parents, the parent-child dyads have been created. In most of the families answered questions about children and financial transfers only one parent, who could provide information about up to four children and up to three children on the financial transfers questions. The other children were dropped from the analysis. The following characteristics of parents and children were included into the analysis: parent’s age (60 years

17 The data collection has been conducted also in 2008, but it focused on the retrospective questions (SHARELIFE). For that reason, the use of data from 2010 was necessary.

108

and younger; 61 – 71 years; 71 years and older), parents’ subjective income (able to make ends meet with great difficulty; with some difficulty; fairly easily; easily), child’s education (primary education; secondary education; tertiary education), gender, marital status (married; divorced or separated; never married), geographical proximity (same household; same building; 1 – 100 km; 100 km and more), occupation (full- time employed; part-time employed; unemployed; in education; parental leave or at home) and number of children (no children; at least one child). The dummy for countries were included. These characteristics were mostly adopted from the 2010 data, except the dummy variables of financial transfers and number of children.

The models were estimated using the logistic regression with clustered standard errors for families since the children share the characteristics of their parents and observations are interdependent across clusters of children. The main purpose is to test the effect of financial transfers in 2006 on the probability that a child will be born during the following four-year period.

The probability of financial transfers varies considerably across European countries (see Graph 15)

The highest percentage of parents who provided a financial help in last 12 months has been observed in

Scandinavian and German-speaking countries and the least proportion is in the southern and eastern countries. The differences are rather high, starting from approximately 10% in Spain to almost 45% in

Sweden. Overall, around 26% of children obtained in 2006 a financial or material gift from their parents

(See Table 13). Almost 2000 new childbirth occurred between 2006 and 2010 which represents about 16% of children with a new child. Approximately half of children were childless in 2006.

109

Graph 15: Proportion of parents who provided financial transfers in last twelve months

Source: Survey of Health, Ageing and Retirement in Europe 2006, own calculations

110

Table 13: Descriptive statistics for parent-child dyads

% N

Childbirth between 2006 and 2010 Yes 16 % 1955 No 84 % 10246 Financial transfers from parents in 2006 Yes 26 % 3123 No 74 % 9078 Childless in 2006 Yes 52 % 6353 No 48 % 5848 Geographical proximity in 2010 Same household 17 % 2034 Same building 4 % 438 1 - 100 Km 63 % 7670 100 Km and more 17 % 2059 Marital status of child in 2010 Married 55 % 6701 Divorced/Separated 6 % 736 Never married 39 % 4764

Total 100 % 12201

Source: Survey of Health, Ageing and Retirement in Europe 2006 and 2010, own calculations. Aggregated dataset of all countries: Austria, Germany, Sweden, Netherlands, Spain, Italy, France, Denmark, Switzerland, Belgium, Czech Republic, Poland.

Logistic regression models are presented in Table 15. The first model without interaction terms does not show a significant effect of financial transfers. However, there is a significant effect of parents’ income. The probability that a child will be born is considerably higher for families who say that they are able to make end meets easily. This might suggest that regardless the occurrence of financial transfers, children of wealthy parents can feel more relaxed about the costs of raising a child and might perceive

111

their parents as their back-up source of money. However, it is important to note that we do not check for a child’s income, so the effect might be caused by the fact that children of wealthier parents are also wealthy. Therefore, what is investigated here is the effect of wealth of the wider family rather than specific effect of parents’ income.

Since the previous research has shown that the first childbirth is determined by different factors than higher order birth, the second model includes the interaction of financial transfers and the number of children. This interaction is significant and signalizes that the effect of financial transfers is substantially higher for the second and higher order births. It corresponds to the theoretical assumption that people have different kinds of motivation when deciding about the first and another child. The marginal utility of the first child is considerably higher than marginal utility of the higher order births and people therefore consider more carefully the possible advantages and disadvantages of the second child and available resources. Furthermore, the costs of two or more children are considerably higher and parents might be more dependent on external sources of financial support.

The third model (Table 14) adds also the interaction of financial transfers and subjective income of parents. It shows that the effect of financial transfers is lowest for the families who are the least satisfied with their income. Surprisingly, the second lowest effect has been observed for the highest income group.

The interpretation of the low effect of financial transfers for the low and high- income families must be different. Parents who are barely able to make end meets probably provide money to their children only in case of necessity and this money might be used to cover the basic needs. These transfers are probably of a lower amount and the capacity of lower-income parents to represent a financial back-up for the future needs of children is rather limited. Under these circumstances, the financial help does not contribute to the probability of childbirth since it does not decrease the risk of having a child. On the contrary, the high-income groups might choose a different strategy of support: giving a one-time gift of a high value such as a house or apartment or a large amount of money. Furthermore, we do not check for the income of children (since the original data do not include the question on child’s income), so a

112

possible explanation may lie in the fact that children of high-income parents also belongs to a higher income group and they do not need a financial help from parents. Children of wealthy parents might be advantaged already because they obtain a better education (see for example Torche 2011).

The comparison of three models has been made based on AIC. The third model is the best, but it seems that the second interaction does not contribute strongly to the quality of model. The effect of parents’ income would probably require a more detailed examination than enables the available data.

113

Table 14: Model of probability of childbirth in 2010; logistic regression

Model 1 Model 2 Model 3 Age parent (Ref. 60 or younger) 61 - 70 years 0,825*** 0,877* 0,879* 71 or older 0,343*** 0,520*** 0,521*** Subjective income parent (Ref. Great difficulty) Some difficulty 1,122 1,093 0,992 Fairly easily 1,333** 1,302** 1,200 Easily 1,429*** 1,381*** 1,352** Age child (Ref. 16 - 25 years) 26 - 35 years 3,196*** 1,708*** 1,708*** 36 - 50 years 1,484 0,440*** 0,440*** Education child (Ref. ISCED-97 1-2) ISCED-97 3-4 1,463*** 1,471*** 1,468*** ISCED-97 5-6 2,330*** 2,292*** 2,290*** Gender (Ref. Male) 0,745*** 0,738*** 0,737*** Geographical proximity (Ref. Same household) Same building 2,353*** 2,399*** 2,399*** 1 - 100 Km 2,638*** 2,534*** 2,541*** 100 Km and more 3,030*** 3,019*** 3,024*** Marital status child (Ref. Married) Divorced or separated 0,471*** 0,518*** 0,517*** Never married 0,468*** 0,474*** 0,475*** Occupation child (Ref. Employed full-time) Employed part-time 1,583*** 1,554*** 1,560*** Unemployed 0,891 0,860 0,855 In education 0,348*** 0,335*** 0,335*** Parental leave/at home 3,509*** 3,610*** 3,607*** Financial transfers in 2006 (Ref. No) 1,046 0,833** 0,425** Already has child (Ref. No) 0,608*** 0,589*** 0,589*** Interactions Already has child * FT in 2006 1,580*** 1,567*** Income * FT in 2006 (Ref. Great difficulty) Some difficulty 2,397** Fairly easily 2,119* Easily 1,801

Continued

114

Table 14 continued

Country (Ref. Austria) Germany 1,176 1,157 1,158 Sweden 1,739*** 1,686*** 1,690*** Netherlands 1,376* 1,370* 1,371* Spain 1,473** 1,313 1,316 Italy 1,669*** 1,550** 1,554** France 1,944*** 1,846*** 1,844*** Denmark 2,049*** 2,054*** 2,052*** Switzerland 1,109 1,087 1,084 Belgium 1,250 1,215 1,216 Czech Republic 1,190 1,093 1,102 Poland 1,603*** 1,428** 1,434** Constant 0,0449*** 0,095*** 0,100***

N 12201 12201 12201 Number of families (clusters) 5973 5973 5973 AIC 9469,969 9214,131 9212,625

Note: Logistic regression with clustered standard errors for families. Significance levels *** 1%, ** 5%; * 10% Source: Survey of Health, Ageing and Retirement in Europe 2006 and 2010

7.4 SUMMARY AND DISCUSSION: THE PARENTS AS A SOURCE OF MONEY BEFORE AND AFTER THE CHILDBIRTH

The costs of raising a child may represent a significant obstacle for part of people planning to start a family. The welfare state arrangements are not capable to fully cover for all costs and parents mostly must rely on their own resources and informal sources of support coming from the member of their social network. The previous research has shown that grandparents involved in childcare or providing a cheap housing might increase the probability of a higher order childbirth. The presented analysis aimed to

115

contribute to this debate by an examination of the effect of financial transfers from parents on the probability of the first and higher order births.

Despite several limitations, which are discussed below, the results clearly suggest that the effect of financial help from parents exists. However, the effect is present only for people who already have one or more children. This distinction has been already suggested by a previous research and it can be explained using the economic term “marginal utility”. The first child fulfils most of the needs of prospective parents, for example, the uncertainty reduction (Friedman et al. 1994) or social capital building (e. g. Schoen et al.

1997). The second order birth might be motivated by siblings’ companionship (Bulatao 1981). The desire and motivation to have another child can be weaker and parents carefully consider the additional costs.

Moreover, parents can decide to have a smaller number of children because they want to invest into their quality, rather than quantity. The fact that parents can rely on other resources can therefore increases the chances that parents will be able to manage more children. This finding is also supported by the observation of a positive effect of parents’ income on the fertility behavior. The effect of financial transfers is also mediated by a subjective perception of income. The effect of financial transfers is the weakest both for the low-income group and high-income group.

The findings reflect the important impact of financial resources on fertility behavior. Furthermore, the social network character of fertility behavior is emphasized. People planning to start a family consider all possible resources. This applies regardless the welfare state regime since the analysis includes the strong Scandinavian welfare states as well as familialistic southern regimes providing minimal support to individuals and families. It can indicate that finances obtained from parents are of different quality than finances obtained from formal sources. Firstly, parents can intentionally push their children to produce grandchildren. Secondly, the willingness to provide a financial support can be signal of a willingness to provide also another kind of support and it transfers the general emotional inclination and an environment of mutual support. Altogether, it creates a safe space for the childbirth. On the contrary, the

116

welfare state benefits have a rather technical character. They role is limited to the reducing of the costs of raising a child. These distinctions should be considered when creating social political arrangements.

The presented analysis has several limitations. First, a lack of children’ characteristics is available in the chosen data. The interpretative capacity of results is decreased especially because of the missing indicator of child’s health and income. It is not clear how much of the effect of parents’ income would diminish if controlled also for child’s income. The results presented here represent rather an overall effect of the family financial situation. Furthermore, the clear limitation is the impossibility to connect the data on financial transfers across the two waves. It is not possible to say if the financial transfers are a one- time matter or if they have tendency to persist and how they are affected by the transitions in child’s life.

Also, the four years period might seem too long, and it is not clear if there is a real relationship between financial transfers and childbirth after several years. However, what remains after the consideration of these limitations is another demonstration that planned and intended character of the childbearing and tendency to bring a child to broader surroundings of a security and certainty. Furthermore, what has been also shown here is that the family plays a special role for establishing of these conditions and it should not be neglected.

8. The influence of informal financial support and childcare on fertility intentions: gender differences

The previous chapter has shown that a non-negligible part of adult children receives financial transfers from their parents and this has been shown also by others (Mathews and Sear 2013; Attias-

Donfut, Ogg and Wolff 2005). Previous chapters have also suggested that the birth of child is the result of planned behavior; however, the results are influenced by an unknown proportion of unintended or unwanted births. The following chapter therefore consider financial transfers and their effect on fertility

117

intentions. As Fahlén and Oláh (2015) show, a perceived economic uncertainty has a negative effect on short-term childbearing intentions (see also Fahlén 2013; Fahlén and Oláh 2013). For this reason, it can be hypothesized that obtaining financial transfers can increase the individual´s or couple´s intentions to have a child.

The following study investigates the short-term fertility intentions of people aged 20-45 in ten

European countries (Bulgaria, Russia, Georgia, Germany, France, Romania, Austria, Belgium, Poland and the Czech Republic) who already have at least one child. The main research question is whether fertility intentions are positively influenced by receiving financial transfers from informal resources such as members of the family. The additional question is whether providing a childcare by informal providers increases the fertility intentions. This question is not new in fertility and intergenerational-solidarity research (see e.g. Kaptijn et al. 2010); however, it is usually investigated in form of fertility behavior, so it will be useful to examine if the effects is the same for fertility intentions. Furthermore, this research aims to show possible gender differences in perceiving a social support from informal resources. The chapter is organized as follows. Firstly, the concept of fertility intentions and preferences for family size will be discussed in the context of contemporary European countries. Then, the data and methods are introduced, and models of fertility intentions are presented.

8.1 FERTILITY INTENTIONS AND SOCIAL CAPITAL

Generally, the research in fertility intentions on micro level (for an overview of research on macro level see Philipov 2011) goes in two main directions. Firstly, it is investigated if the fertility can be predicted based on fertility intentions and researchers seek to find factors that influence the realization of fertility intentions (see e.g. Berrington 2004; Morgan and Rackin 2010; Schoen et al. 1997; Spéder and

Kapitány 2009). This part of fertility intentions research also deals with the ideal family size and the

“fertility gap” which is a term used for the difference between desired and actual fertility (e.g. Quesnel-

118

Vallee and Morgan 2003). However, the short-term fertility intentions are a more accurate predictor of the fertility behavior than long-term intentions (see Dommermuth et al. 2009; Philipov 2009) or family size preferences.

The second part of research in fertility intentions focuses on factors that influence the intentions to have a child regardless their realization. Fertility intentions or plans are then approached as an independent phenomenon which is in some respects different from fertility behavior. While part of realized births is unplanned (or even unwanted) or it is a result of a spontaneous act or it is mainly a decision of the other partner, fertility intentions are supposed to be based on a consideration of the costs and benefits of childbearing and individual motivation, opportunities and limitations. For that reason, they might be more suitable for an investigation of the fertility decision-making process since it is important to know what motivates or constraints individuals or couples in their fertility plans. On the other hand, they should not be approached as a direct measure of fertility because part of intended birth is not realized.

The recent research in fertility intentions has often built upon Ajzen´s theory of planned behavior

(Ajzen 1991) to explain the link between intentions and behavior. According to the theory of planned behavior (TPB) intentions precede behavior and explaining intentions is important for explaining the decisions. It is supposed that the factors which influence intentions are similar to those affecting behavior while some factors that influence the childbirth can be specific since part of the births is unintended

(Philipov, Spéder and Billari 2006). TPB is based on three components: attitudes, norms and perceived behavioral control. It concerns personal beliefs, a subjective perception of conditions for behavior. For that reason, Philipov (2011) points out that the TPB does not built on an assumption of a rationality since beliefs are composed also of an intuition and other non-rational factors. From the social-capital perspective it means, for example, that personal social network does not have only instrumental effect of a direct and immediate support, but it also increases the individual´s belief that more support can be obtained in case of need (which does not have to be necessarily the true in reality). The theory has been

119

tested on survey data and it proved to be useful for explaining fertility intentions. For example, Billari,

Philipov and Testa (2009) found that attitudes, norms and behavioral control together influence the fertility intentions in Bulgaria even after controlling for background factors such as gender, age or education.

The original theory of planned behavior explains individual beliefs and behavior. However, Rossier and Bernardi (2009) suggests that the TPB can be extended by incorporating mechanisms of social interactions in personal network. The research in social network has identified three main mechanisms of the social network´s influence on the individual behavior (see e.g. Keim 2011): social learning, social influence and social support. Social learning is responsible for a reproduction of social behavior, in context of fertility it is expressed, for example, by the intergenerational transmission of fertility behavior (see e.g.

Liefbroer and Elzinga 2012; Murphy and Knudsen 2002; Murphy 2013). Social influence can be represented by a pressure of significant others to behave in a certain way such as having a baby or on the contrary to postpone the childbirth (e.g. Barber 2000). This mechanism is also attributed to a social

“contagiousness” of the childbearing behavior among siblings, friends or acquaintances (see Bernardi

2003; Pink, Leopold and Engelhardt 2014). The third mechanism of social networks is social support.

Rossier and Bernardi (2009) suggest that within the theory of planned behavior, social support affects both perceived and actual behavioral control. According to Ajzen and Klobas (2013: 206) behavioral control or control beliefs are “concerned with the perceived presence of factors that influence a person´s ability to have a child”. These factors may include biological factors such as infertility but also economic conditions and instrumental issues such as suitable housing, financial resources or availability of childcare.

Significant other may influence the ability to have a child by providing or not providing resources that are necessary for the childbearing. Social support can therefore reduce the costs of childbearing and for that reason it can influence the fertility intentions and fertility behavior.

A positive effect of social capital on fertility intentions has been shown, for example, by Bühler and

Philipov (2005) in Bulgaria. Women aged 18 to 34 with close people who provide “important and

120

substantive support” have a higher probability of fertility intentions. Three aspects of their findings are of an importance for further research in this field. Firstly, their investigation is gender-specific since they do not consider men. Many studies choose to investigate female fertility intentions because of unequal costs of childbearing for men and women and because the intentions of women predict the fertility behavior more reliably than man´s (Berrington 2004). However, for example, Balbo and Mills (2011) analyzed both sexes and found that economic situation is important for men´s fertility intentions. For that reason, it is necessary to distinguish different kinds of support that could be gender-specific. This is related to the second aspect of the study by Bühler and Philipov (2005). In their Bulgarian study, they consider a general support and the intensity of this support – small help and substantial and important support. It is not specified what kind of support is provided. The support can include financial transfers and other material support, or it could be a help, for example, in form of childcare. As explained above, the effect of support may vary according to different kinds of support and receivers of support. Thirdly, they choose parity- specific approach. It is supposed that having the first child is different than having a second or other child since the decision to have the first child is a decision about the parenthood itself (see also Balbo and Mills

2011; Billari, Philipov and Testa 2009; Philipov et al. 2006; Schoen et al. 1997). This parity-specific effect of social capital has been suggested also in the study of Bühler and Fratczak (2007). They show that fertility- related social capital is important for intentions to have a second child. Some of the existing studies on the effect of social capital on fertility focus only on the childbirth of a higher order since they show the effect of grandparental care of the first child (e.g. Aassve, Meroni and Pronzato 2012).

The presented study is going to build on these previous studies and the theoretical framework is adopted from Rossier and Bernardi (2009) who connect the social network perspective and TPB. They suggest the importance of social support for perceived and actual behavioral control as one of the determinants of fertility intentions and fertility behavior. In their paper, they propose to investigate the effect of the informal childcare. Based on the findings from the previous chapter, this study will also include financial transfers as another form of support.

121

The effect of the childcare and financial transfers will be tested independently since it can be supposed that their mechanism of support is different, and they can be important in different situation.

While childcare helps reducing the time burden and opportunity costs, the financial transfers are directed on economic costs of children. It can be hypothesized that they have a different impact on men´s and women´s fertility intentions because perceived costs of childbearing are different for men and women

(see e.g. (Liefbroer 2005). Women are those who in most cases at least temporarily reduce their activity on the labor market but also cut their leisure time activities. It has been shown that the postponement of the childbearing can be an advantageous strategy for women with a higher education since it leads to a lower maternity gap (see e.g. Joshi 2002; Taniguchi 1999). On the contrary, men are still those who bear the main responsibility for a financial security of the family and the financial constrained might be more important for them (see Liefbroer 2005). Low-educated men, for example, more often stay childless

(Kravdal and Rindfuss 2008). This can be explained, among other factors, by their relatively worse position on the partnership market and the insecurity in terms of their abilities to secure family needs. Childless women can be, on the contrary, more often found among women with higher education.

Accordingly, it can by hypothesized that there is an effect of different forms of social capital on fertility intentions. More specifically, two hypotheses are established. Firstly, it is supposed that the informal childcare reduces the time burden and opportunity costs of childbearing and therefore positively influences the fertility intentions, primarily of women. Secondly, financial transfers from significant others reduce the economic costs of childbearing and therefore they increase the chance of the short-term fertility intentions, primarily of men. Since the childcare is considered, only respondent with at least one child will be analyzed because childless people do not obtain a help in form of childcare. For that reason, the study investigates the effect of social support on fertility intentions of people who already have at least one child.

Several contributions are expected. Firstly, different kinds of support will be considered. While many previous studies analyzed a general effect of an informal support, the following study distinguishes

122

between financial support and childcare support. It is supposed that the effect is not the same. Secondly, both men and women will be included into the analysis. However, the separate datasets have been constructed to test whether the effect of both kinds of support is the same for both sexes. Finally, data from several European countries are analyzed. The existing studies have been focusing mainly on eastern countries (Bulgaria, Poland or Hungary) which could differ from other part of Europe, for example, because of their underdeveloped systems of public support of families or worse economic situation.

8.2 DATA AND RESULTS

The analysis will be based on the first wave of GGP data (Generations and Gender Programme). The data have been collected between 2002 and 2011 in fifteen European countries: Bulgaria, Russia, Georgia,

Germany, France, Hungary, Italy, Netherlands, Romania, Austria, Belgium, Lithuania, Poland, Czech

Republic and Sweden. However, not all these countries collected data on all questions that are used in analysis. Part of them, for example, did not include a general question on financial transfers (Hungary,

Lithuania). Italy was also excluded from the analysis since Italian respondents were asked about transfers obtained during last four weeks, while it is last twelve months in an original questionnaire. Furthermore, respondents from Sweden were not included because of a large share of missing answers.

The indicator of financial transfers is based on the following question: “During the last 12 months, have you or your partner/spouse received for one time, occasionally, or regularly money, assets, or goods of substantive value from a person outside the household? Please think also about land and property or inheritance that was transferred to you or your partner/spouse during this time.” Respondent were then asked who the person who provided money, assets or goods was, and they could name up to six persons.

Mostly, close members of family are named, primarily parents but also friends or acquaintances could be givers. Overall, there is around 8% of people who obtained a gift in last twelve month among people aged

18 to 45 and having at least one child younger than 12 years. However, the proportion substantially varies 123

between countries, see Table 16. The highest proportion of people who obtained any transfer is in

Georgia (18%). The proportion is higher than average also in Russia (13%), France (11%) or Belgium (10%).

On the contrary, a low informal support is observed in Romania (3%), Germany (4%) or the Czech Republic

(4%). Apparently, there is no visible trend of transfers across countries such as a higher support in eastern countries.

The second dependent variable was based on an indicator of informal childcare obtained in last 12 months. Respondents were asked if they get regular help with childcare from relatives or friends or other people for whom caring for children is not a job. This kind of support is much more common as about 42% people obtain this help from others. However, as in case of financial help, the proportion of people who obtain childcare varies substantially across countries, see Table 15. The highest proportion of parents who are supported by their friends or relatives is in Netherlands, where almost all (95%) parents gen regular help with childcare. A high proportion is also observed in Belgium (60%), Hungary (58%) and Austria

(56%). On the contrary, informal childcare is not very common in Romania (20%), Lithuania (25%),

Germany (26%) or Georgia (28%). Nevertheless, it is important to note that the frequency of help is not considered here. The question on frequency was not asked in all countries. As an illustration, the mean score on frequency per week for the main caring person has been computed. The relationship between indicator of frequency of care and proportion of people who are provided by help with childcare across countries is shown in Graph 16. It is obvious, that these two measures are in a rather negative relationship. The frequency is high in Romania and Georgia, where the main caring person provide care on average more than 200 days in a year. The overall proportion of people who can rely on an informal help is, however, rather low as it reaches less than 30%. On the contrary the frequency is more than 50% lower in Austria and Belgium, however, most of the people with young parents are supported by their family members or friends. A similar inconsistency of two indicators of social support has been already shown by previous research on intergenerational solidarity. For example, Albertini, Kohli and Vogel (2007) showed that intergenerational transfers are less frequent and more intense in Southern European countries while

124

it is the opposite in the Scandinavian countries. Since we have more data on the first indicator, the following analysis will primarily use the indicator whether respondent has or does not have someone who provide childcare. However, it will be also than checked if the result change after using the frequency indicator.

Table 15: Proportion of parents who obtained financial transfer or informal childcare across countries

Financial transfers Informal childcare N

Bulgaria 5 % 46 % 2307 Russia 14 % 48 % 1079 Georgia 19 % 27 % 933 Germany 4 % 29 % 1989 France 10 % 44 % 853 Romania 3 % 22 % 1300 Austria 9 % 61 % 896 Belgium 11 % 68 % 635 Poland 8 % 44 % 1506 Czech Republic 4 % 40 % 648

Total 8 % 41 % 12146

Source: Generations and Gender Programme (2002-2011)

125

Graph 16: The relationship between indicator of frequency of care and proportion of people who are provided by help with childcare across countries

Source: Source: Generations and Gender Programme (2002-2011)

As has been said, the analysis will include only those respondents who already have one or two children. Respondents with more than two children were excluded since big families are rare in current

European countries. Furthermore, only respondent who currently have a coresident partner will be analyzed because it is supposed that a partnership is a condition of planning another child. Since the aim is the investigation of the childcare effect, the dataset will be further restricted to respondent with at least one child aged 11 and younger. It is supposed that older children do not require a childcare. One of the controlling variables is the age of the youngest child of the respondent (0-3 years; 4-6 years; 7-12 years) and it is also controlled for the number of children (one; two; three or more).

126

The models also control for the following respondent´s characteristics: age (continuous), sex, education, partnership status (has partner; does not have partner); activity status (employed; unemployed; at home). Furthermore, it is controlled for the household´s subjective income. The indicator is based on the question how difficult is for the household to make ends meet (with great difficulty – very easily, continuous). Finally, it is controlled for characteristics or the respondent´s partner – whether the partner is working and partner´s level of education.

Both indicators of social support have been recoded as dummy variables – respondent either gets or does not get this kind of support. Four models of logistic regression have been made separately for men and women, see Table 16 and Table 17. First model (Table 16) investigates the relationship between financial or material financial transfers and men´s short-term fertility intentions. The association is statistically significant, and it seems to be quite strong, about 66%. The intentions to have another child decrease with age of the respondent and with the age of the youngest child in the family. Unsurprisingly, men with one child plan to have child considerably more often than those with two children. The effect of other variables is rather week. The model shows no significant effect of respondent´s education, partner´s education, activity status and partner´s activity status. There is some indication of the income´s effect.

Men who say that they are easily able to make ends meet are more willing to have another child.

The second model shows the relationship between women´s short-term fertility intentions and financial or material transfers. The association is insignificant this time and it seems to be weaker.

Contrary to the men´s model, women show some effect of high education. Highly educated women tend to plan another child with a higher probability. The effect of other variables is similar as in case of men.

The following two models (see Table 17) show the relationship between informal help with childcare and fertility intentions. The Model 3 shows no association between men´s short-term fertility intentions and informal childcare. The effect of other controlling variables remains mostly the same. The same does not apply to women in Model 4. Although the effect is not very high, it is still apparent that the support in form of informal childcare has some effect on women´s fertility intentions.

127

Overall, the results confirm the expectations as the effect of both kinds of social support exists, but it is to some extent different for men and women. Since the costs of children are differently perceived by men and women, the factors that can reduce these costs are also different. The male breadwinner model is still prevailing in most of the European countries at least when small children are present in the family.

It is suggested that man perceive their obligations of securing the financial and material needs of the family. Having an external source of finances can increase their feeling of insecurity about the future financial situation of the family. The same applies also to women but it is suggested that the importance of this factor is lower for them. On the contrary, women are influenced more by time costs of children and having someone who can care for their children is more important for them.

128

Table 16: Models of fertility intentions - financial transfers; logistic regression

Men Women OR OR

Financial transfers 1.66** 1.24 Country (ref. Bulgaria) Russia 1.34 1.40* Georgia 0.43*** 0.61** Germany 0.68** 0.90 France 0.19*** 0.28*** Romania 0.26*** 0.23*** Austria 0.37*** 0.58** Belgium 0.13*** 0.23*** Poland 0.38*** 0.55*** Czech Republic 0.51** 0.61* Age 0.91*** 0.86*** Education (Ref. Low) Medium 0.93 1.18 High 1.00 1.64** Education partner (Ref. Low) Medium 0.95 0.98 High 1.18 0.95 Number of children (Ref. One child) 0.15*** 0.13*** Income (Ref. Low) Medium 1.06 1.10 High 1.34 1.33 Age of the youngest child (Ref. Age 0 -3) Age 4-6 0.53*** 0.65*** Age 7-11 0.30*** 0.39*** Activity status (Ref. Employed) Unemployed 0.86 0.93 Inactive 1.00 1.19 Partner working 1.04 1.00

N 4871 6575

Significance levels: *** 0.1%, ** 1%, * 5% Source: Generations and Gender Programme (2002-2011)

129

Table 17: Models of fertility intentions - childcare; logistic regression

Men Women OR OR

Help with childcare 1.05 1.21* Country (ref. Bulgaria) Russia 1.39* 1.40* Georgia 0.47*** 0.64*** Germany 0.69** 0.93 France 0.20*** 0.28*** Romania 0.27*** 0.24*** Austria 0.39*** 0.57** Belgium 0.13*** 0.22*** Poland 0.39*** 0.56*** Czech Republic 0.51** 0.61* Age 0.91*** 0.86*** Education (Ref. Low) Medium 0.93 1.19 High 1.01 1.66** Education partner (Ref. Low) Medium 0.97 0.97 High 1.20 0.94 Number of children (Ref. One child) 0.15*** 0.13*** Income (Ref. Low) Medium 1.05 1.09 High 1.34 1.29 Age of the youngest child (Ref. Age 0 -3) Age 4-6 0.52*** 0.65*** Age 7-11 0.30*** 0.40*** Activity status (Ref. Employed) Unemployed 0.87 0.96 Inactive 1.06 1.23* Partner working 1.04 0.98

N 4871 6575

Significance levels: *** 0.1%, ** 1%, * 5% Source: Generations and Gender Programme (2002-2011)

130

8. 3 SUMMARY AND DISCUSSION: THE EFFECT OF PERSISTING TRADITIONAL GENDER ROLES IN FERTILITY DECISION-MAKING PROCESS

According to the increasing possibilities of fertility planning, investigating the fertility intentions is now an important part of fertility research. The conceiving occurs as a spontaneous and unplanned act to a much lesser extent than in the past and parents usually consider the conditions for the childbirth.

Previous research has repeatedly shown that the disposal social capital can play an important role besides the individual resources and the resources of the partner. However, the presented analysis reveals, among other things, that these resources might be variously important for men and women. Although women are those who have more control over a contraception, the intentions of male partners should not be neglected. Men can consider different aspects than women even within the same couple. As shown above, men emphasize the financial side of the childbearing more than women, while women stress the time costs more. For both men and women, the external help received from the close people are of a significant importance in their decision to have the second or other child.

Some conclusions and questions for future research can be generated. Firstly, the social capital is an important aspect of the fertility decision-making process. Supposedly, it is true especially for the decision over the second and other child (see Chapter 6 and Chapter 7) as previous research has shown that insufficient resources threaten especially having more than one or two children. Secondly, the analysis controls for the country and the dataset includes countries from different part of Europe, however, the effect of support can differ across countries. The future research should focus on different roles of informal support regarding the scope of formal support from the state. More countries are necessary to make a reliable comparative analysis. Other limitation of presented study is the insufficient data about the intensity of support. It is now not possible to say if the role of the support is substantial and the amount of resources is important for the decision to have another child or if it rather plays a symbolic role (see the Chapter 5 for the discussion of the latent and manifest solidarity) and influence the decision over fertility even if the amount of support is low.

131

9. Conflicting intergenerational roles: caregiving to grandchildren and elderly parents

Intergenerational relationships in contemporary European societies have been considerably influenced by recent demographic changes. Firstly, European populations are aging in a consequence of an increasing life expectancy and decreasing fertility rate (see e.g. Lutz, Sanderson and Scherbov 2008).

The increasing proportion of elderly people puts more requirements on a public system of pensions and care services, as well as on family members, who provide help and care to ageing members of the family.

Secondly, most European countries face an increasing age at first birth. Due to postponement of childbearing, not only parents, but also grandparents are older than in the past. According to Leopold and

Skopek (2015), increasing life expectancy and the postponement of grandmother-hood lead to the overlapping of grandparental and filial roles. Middle-aged adults are expected to support or care for their own ageing parents and parents-in-law and for their grandchildren. Moreover, people are forced to work longer since the retirement age has increased, and they must balance their roles within the family with their activity on the labour market.

The current research on the middle-aged generation focuses primarily on a conflict between working and caring responsibilities for children and grandchildren, elderly parents or a disabled partner

(see e.g. Bolin, Lindgren and Lundborg 2008; Crespo and Mira 2014; Ettner 1995; Gray 2005; Hochman and Lewin-Epstein 2013; Leopold and Skopek 2014; Wang and Marcotte 2007). The conflict in multiple caring responsibilities has been addressed by research on the 'sandwich generation': mostly women who care for an older family member as well as for their own dependent children. A substantial number of middle-aged people provide support to their elderly parents in the form of time; for example, according to Bonsang (2007), it is about thirty per cent of adult children in European countries. Previous research has shown that the proportion of people who care for both elderly parents and underage children is not very high, however, many middle-aged people experience coresidency with their adult children (Angelini,

132

Laferrère and Pasini 2011; Isengard and Szydlik 2012; Le Blanc and Wolff 2006), support them (Grundy and Henretta 2006) and those who have grandchildren are often involved in their care (Guzman 2004;

OECD 2012). For that reason, the concept of the sandwich generation might be used also for people who simultaneously care or support both their elderly parents and their adult children or young grandchildren.

Following Grundy and Henretta (2006), two competing hypotheses are tested in the present study.

Firstly, it can be hypothesized that people experiencing the responsibilities of assisting their elderly parents are discouraged from caring for their grandchildren because they lack time, energy or other resources. There is however also an alternative hypothesis: due to the stronger emotional and solidarity bonds to other family members, those people who help their elderly parents are also more likely to take care of their grandchildren.

The following study investigates how the probability to look after grandchildren is influenced by the provision of care to elderly parents by the same person. It analyses the panel data from the Survey of

Health, Ageing and Retirement in Europe (SHARE) collected between 2004 and 2012 in 13 European countries and Israel. The macro level analysis of countries in propensity and intensity of care for elderly parents and underage grandchildren is provided to show significant differences and clustering of

European countries. In the second section, an individual level multivariate analysis is conducted to test the influence of taking care of an older family member on the probability of looking after grandchildren on an occasional and regular basis.

9.1 MACRO LEVEL DETERMINANTS OF INFORMAL HELP AND CARE

The frequency and intensity of grandparental childcare vary considerably across European countries. Some authors indicate (Fokkema, ter Bekke and Dykstra 2008; Hank and Buber 2009; Ogg and

Renaut 2006) that grandparents in Scandinavian countries and the Netherlands show a higher propensity

133

of care, but they are less likely to provide care on a regular basis. On the contrary, southern countries

(Italy, Spain and Greece) show the lowest proportion of caring grandparents, but grandparents who provide some care tend to provide it regularly. This variation is partially caused by different family settings, for example the coresidency of parents and adult children and grandchildren. Hank and Buber

(2009) also point out the potential differences in perceiving the intensity of care by respondents.

A similar pattern has been observed also in case of adult children taking care of their elderly parents. Support to parents is provided in two forms: help and care (Brandt, Haberkern and Szydlik 2009;

Igel et al. 2009). Help is characterized as less demanding and it is more prevalent in Northern European countries. On the contrary, care as a more demanding form of support is more common in Southern

European countries. The divide between northern transfer regimes with a high proportion of people providing less intense support and southern transfer regimes with a low proportion of people providing more intense support has been suggested also by other authors (see e.g. Bonsang 2007; Ogg and Renaut

2006).

The variation across countries is associated with cultural aspects. Fokkema, ter Bekke and Dykstra

(2008) found that people in Scandinavian countries demonstrate low responsibility toward their family members, while people in southern countries generally feel much more obligated to support their relatives. Besides culture, the welfare state regime seems to be a crucial factor for macro level differences. Northern countries are characterized by strong welfare states that secure a substantial part of the caring responsibilities for dependent people. On the contrary, southern countries rely on families as key providers of help and care. According to Igel and Szydlik (2011), a strong welfare state motivates

("crowd-in") family members to help each other but discourages them ("crowd-out") to provide demanding care on a regular basis. The crowding-in effect has been shown also by others (Brandt,

Haberkern and Szydlik 2009; Künemund and Rein 1999; Motel-Klingebiel, Tesch-Römer and von

Kondratowitz. 2005). The main idea is that the existence of a strong welfare state does not lead to the

134

weakening of family solidary but it rather changes the structure and characteristics of interpersonal relationships within the family (Daatland and Lowenstein 2005).

On the basis of the findings cited above, it can be hypothesized that the help and care of parents and the care of grandchildren are positively related on the macro level. Countries with a high proportion of people who support their elderly parents have a high proportion of grandparents looking after grandchildren and vice versa. Before testing this hypothesis, the conclusion of findings on individual factors related to help and care is provided in the subsequent section.

9.2 INDIVIDUAL DETERMINANTS OF INFORMAL HELP AND CARE

Gender is one of the most distinct factors influencing the likelihood of grandparents' care. Women look after grandchildren considerably more often than men (Hank and Buber 2009; Danielsbacka et al.

2011), but as mothers they are also more likely to be helped by their parents since maternal grandmothers invest the most, followed by maternal grandfathers, paternal grandmothers and paternal grandfathers (Coal, Hilbrand and Hertwig 2014; Euler and Weitzel 1996; Laham, Gonsalkorale and von

Hippel 2005). There is also clear evidence about the influence of geographical proximity and contact between parents and children (Baydar and Brooks-Gunn 1998; Guzman 2004; Hank 2007; Hank and Buber

2009; Vandel et al. 2003).

Working grandparents are generally willing to care equally as non-working grandparents, but with less intensity (Attias-Donfut, Ogg and Wolff 2005, Gray 2005, Hank and Buber 2009). Lakomý and Kreidl

(2015) suggest that some grandparents tend to reduce their employment in order to provide care for their grandchildren and grandparents tend to retire earlier (Hochman and Lewin-Epstein 2013; Van Bavel and De Winter 2013). Besides the position on the labour market, there is also the effect of education:

135

highly educated grandparents tend to care for their grandchildren more than grandparents with a lower education (Baydar and Brooks-Gunn 1998; Silverstein and Marenco 2001).

The likelihood of the grandparents' care is, however, not formed only by the structure of grandparents' characteristics and opportunities, but also by parents' characteristics. Grandparents tend to care more regularly if the mother is employed (Del Boca 2002; Del Boca, Locatelli and Vuri 2005). Women who are helped by grandparents are more likely employed, they work more hours and earn more

(Brewster and Rindfuss 2000; Del Boca 2002; Gray 2005; Hank and Kreyenfeld 2003; Vandell et al. 2003).

Both grandparents' and grandchildren's age are important factors. Younger grandparents tend to look after their grandchildren more than older ones and younger grandchildren receive more care from their grandparents (Coall, Hilbrand and Hertwig 2014; Luo et al. 2012; Silverstein and Marenco 2001).

Regardless of age, the health limitations of grandparents decrease the propensity of care (Hank and Buber

2009). Some authors also suggest that older grandparents may be less involved in care of their grandchildren because they must deal with not only their own health issues, but a lot of them also care for their dependent elderly parents (Minkler and Fuller-Thomson 2000).

According to the studies on the intergenerational care of elderly people, a significant proportion of the European and the American population provides help to the older generation (Brandt, Haberkern and

Szydlik 2009; Grundy and Henretta 2006). Individual determinants of elderly parents' caregivers are similar as determinants of grandchildren's care. As in the case of grandchildren’ care, significantly more women than men provide some support to parents, assistance is provided by children living nearby, without health limitations, with a higher education and the propensity for giving help decreases with the increasing age of the care provider (Bonsang 2007). Apart from the opportunities of children, the needs of elderly parents strongly influence the propensity for receiving some assistance (Silverstein, Gans and Yang

2006). Children in family-oriented countries tend to be more responsive to the needs of their parents

(Kalmijn and Saraceno 2008).

136

In sum, the characteristics of people who provide care or help are shared both by people looking after grandchildren and assisting their elderly parents. They are usually middle-aged, women, without health limitations, with a rather high education and living near their relatives. Furthermore, normative solidarity and responsibility toward family members influence the support provided to both elderly parents and grandchildren. Several studies have shown the effect of filial responsibility on the provision of support to parents (e.g. Bromley and Blieszner 1997; Silverstein, Gans and Yang 2006). Others show the positive effect of perceived family obligations on grandparents’ engagement in grandchildren’ care (Coall,

Hilbrand and Hertwig 2014). These findings indicate that in some cases, it might be the same person who simultaneously provides help or care to both grandchildren and elderly parents. The term 'sandwich generation' can be extended to describe this group of people who are in a similar position as parents raising underage children and assisting their elderly parents. The next section provides information about previous findings on the sandwich generation.

9.3 THE SANDWICH GENERATION IN THE FOUR-GENERATION APPROACH

The sandwich generation is usually defined as a middle-aged generation (or 'pivot generation') of people who simultaneously care for their elderly parents or parents-in-law and dependent children (Tebes and Irish 2000). This generation is involved not only in roles related to family, but they are usually also at the peak of their careers and for that reason, they must balance competing demands inside and outside the family (Riley and Bowen 2005).

Investigation of the sandwich generation requires a multigenerational approach to analyse how different family roles interact with each other. Some authors (Fingerman et al. 2011; Grundy and Henretta

2006) employ a three-generation approach: a generation of children, a middle-aged generation of parents and a generation of grandparents. However, the contemporary family is characterized as

137

multigenerational (Bengtson 2001) and it might be desirable to include even more generations into the investigation of related concepts. Due to increasing life expectancy and healthy life years, people can spend a significant part of their life with their grandchildren. The following analysis therefore considers four generations: children, parents, grandparents and their elderly parents.

According to the research using the three-generation approach, recent findings suggest that downward care prevails: parents tend to help their children more than their parents, but in case of parental disability they support the parents more (Fingerman et al. 2011). This could indicate that support provided to dependent elderly parents represents a burden, which discourages people of the middle-aged generation from supporting their children.

It is, however, necessary to distinguish different levels of a support, for example, more demanding care of strongly dependent elderly parents and less demanding help such as assistance in the household

(Brandt, Haberkern and Szydlik 2009), as well as various levels of frequency and intensity. Less demanding support does not take much energy and time and therefore does not necessarily constrain other activities and caring duties. In fact, there can even be a positive relationship because providing support can represent the expression of a general willingness to support other family members. On the other hand, the effect of highly intense and demanding care for the elderly can prevent grandparents from looking after their grandchildren, either due to a lack of energy or time. The literature shows extensive evidence on the negative effect of caregiving on the mental and physical health or general well-being of the caregiver (see e.g. Hiel et al. 2015; Marks, Lambert and Choi 2002; Pavalko and Woodbury 2000).

The frequency and intensity of support can play an important role in the investigation of the conflict between caring roles. For that reason, the following analysis distinguishes regular and occasional help or care provided to grandchildren and parents.

138

9.4 DATA AND METHODS

The analysis is based on data from the Survey of Health, Ageing and Retirement in Europe. It collects information about people aged 50+, which is the age at which a substantial number of people already have adult children and also have parents after or close to retirement age. So far, a module on intergenerational support has been included in all four regular waves (except for the retrospective survey

SHARELIFE in 2008) in 2004-2012. The SHARE enables using panel data for respondents participating in at least two waves and its international approach gives access to comparative information.

9.4.1 Dependent measures

Two dependent variables have been constructed. The first is a dummy indicator based on the question asking whether a respondent looked after grandchildren during the last 12 months. The second dependent variable differentiates between regular care (at least once a week) of grandchildren and occasional care (less than once a week). Two separate models with these dummy dependent variables have been constructed. The first model includes all respondents with at least one grandchild younger than

15 years and distinguishes between those who provided any care of grandchildren and those who did not.

The second model covers only those respondents who provided any care and investigates the factors affecting the probability of regular care.

139

9.4.2 Independent measures

The main independent variable considers if a respondent provided any assistance to one of his or her parents, stepparents or parents-in-law in the previous year. Assistance is defined as one of the three kinds of help: personal care, practical household help and help with paperwork. Furthermore, respondents were asked how often they had been providing this help. The main independent variable was created based on a question about the frequency of the provided help to parents: daily; about once a week; about once a month; less often; never.

The models moreover control for many factors that have been already identified as important determinants of caregiving by previous research. First, they control for characteristics of the main respondent (grandparent): level of education (low, medium or high); labour market status (retired; employed or self-employed; unemployed; sick or disabled; homemaker); age (55 and less; 56-60; 61-65;

66-70; 71-80; 81 and more); marital status (married; divorced, separated or widowed; never married) and health status (excellent; very good; good; fair; poor). Secondly, they control for characteristics of children: education; gender; geographical proximity (in the same household; in the same building; 1 to 25 kilometres; 26-100 kilometre; more than 100 kilometres); marital status; employment status. Age of the youngest grandchild has also been considered.

Furthermore, the models control for the perception of responsibility towards family members. The indicator of family responsibility has been constructed on the following battery of questions on normative family obligations: “Parents' duty is to do their best for their children even at the expense of their own well-being.” “Grandparents' duty is to be there for grandchildren in cases of difficulty (such as divorce of parents or illness).” “Grandparents' duty is to contribute towards the economic security of grandchildren and their families.” “Grandparents' duty is to help grandchildren' parents in looking after young grandchildren.” Respondents were asked to what extent they agree with these statements: strongly agree; agree; neither agree not disagree; disagree; strongly disagree. While the first item measures the 140

parents’ obligations, the others focus on the grandparents’ role. For that reason, the first item has been dropped from the analysis and the other three items have been used for a construction of an index of family obligations as a mean of these questions. The Cronbach’s alpha confirmed the reliability of the index since it is almost 0.9 for the aggregated data set and it is between 0.79 and 0.95 in country data sets.

The data have a multilevel structure. The primary respondents are people aged 50 and over who were asked about their children. The first level of analysis therefore consists of children clustered by the family on the second level. The third level is a panel data component; data contain information about each child at least in two years. For that reason, the methods of multilevel data analysis are employed.

Since the dependent variables are coded as binary indicators, the mixed-effects models for binary responses are used. The final dataset is large, and three level of analysis make it computationally demanding. Since the focus of the analysis lies in the fixed-effects estimates, the Laplacian approximation18 has been chosen for estimation of models to increase computational efficiency.

Furthermore, data contain another level of analysis, which is the country. The number of countries is, however, too low (14 countries) for using the country indicator as another level of analysis and therefore dummy indicators for country are used as control variables.19

18 Laplace approximation is equivalent to one integration point and can be used as an alternative to multiple integration points if emphasis is placed on fixed effect estimates (StataCorp 2015). 19 Bryan and Jenkins (2015) suggest that the minimal number for computing multilevel models is 25 countries for linear models and 30 countries for logit models. The authors also indicate that estimations of individual effects are correct if the number of cases within clusters is large. Since the present analysis focuses on individual-level explanation, it is possible to rely on computations even when number of countries is small. 141

9.5 MACRO LEVEL ANALYSIS

As mentioned above, previous research has suggested a general pattern of private caring regimes in different countries. As Graph 1720 shows, a clear south-north gradient exists for the probability of looking after grandchildren and any kind of assistance to elderly parents by people aged 50+. Recent investigations are extended here by adding more countries from Eastern Europe. These countries (Poland and the Czech Republic) are similar to countries in Southern Europe with a rather low proportion of people providing care to their elderly parents and low engagement in grandchildren’ care. The macro level analysis does not suggest a conflict between caring roles because countries with a relatively high proportion of people helping their elderly parents show also a high proportion of people looking regularly or occasionally after their grandchildren.

20 Graph 16 and Graph 17 are based on four waves of SHARE survey; however, a random selection of only one year per respondent has been used for respondents participating in more than one wave. Since some respondents participated only in one wave, their probability of providing help would be lower compared to respondents participating repeatedly. This procedure of respondents’ selection applies only to the first part of the analysis, which investigates the differences between countries on an aggregate level. 142

Graph 16: Proportion of people who regularly or occasionally look after grandchildren and provide help to parents in European countries

Note: data for one wave per respondent; author’s own calculation Source: Survey of Health, Ageing and Retirement in Europe 2004-2012

The first graph, however, does not take into consideration the frequency of provided care. As has been shown, southern countries (and as expected, eastern countries) demonstrate a rather low proportion of people caring for their family members, but with a higher frequency and intensity than people from northern countries. Graph 17 therefore displays an association between regular care of grandchildren (daily or at least once a week) and any assistance to elderly parents. The association is opposite in this case. Southern and eastern countries show a higher proportion of regularly caring grandparents in comparison with northern countries. However, the negative association between regular care of grandchildren and assistance to elderly parents does not necessarily indicate the conflict between

143

caring responsibilities but rather different arrangements of family services in European countries.

Scandinavian countries ensure services and financial benefits for both childcare and the care of elderly people; southern countries, on the contrary, keep most of the caring responsibilities within the family. A higher proportion of people who occasionally look after their grandchildren suggests that due to high accessibility of childcare services, people in Scandinavian countries do not care for grandchildren because it is necessary but because it brings them joy. Since their engagement in childcare is rather sporadic, they can devote their time also to other members of the family.

Graph 17: Proportion of people who regularly look after grandchildren and provide support to parents in European countries

Note: Data for one wave per respondent; author’s own calculation Source: Survey of Health, Ageing and Retirement in Europe 2004-2012

144

To sum up, according to the macro level part of the analysis, there are significant differences across

European countries in terms of caring responsibilities. While any care of grandchildren is positively associated with helping parents, regular care of grandchildren is correlated negatively on the country level. The next part of the analysis investigates if this pattern exists also on the individual level after controlling for country effects.

9.6 INDIVIDUAL-LEVEL ANALYSIS

Roughly two thirds of all grandparents in the SHARE dataset looked after their grandchildren during the previous year (see Table 18). However, only less than one third of respondents provided care regularly. The proportion of people who provided help to their parents is significantly lower since only 12 per cent of respondents assisted their elderly parents.

145

Table 18: Descriptive statistics: Grandparents’ care and help to older parents

Percentage Number

Provided care or help to Grandchildren 38% 474 Parents 11% 136 Both grandchildren and parents 28% 354 Neither 23% 280 Subjective health Very good 33% 406 Good 41% 509 Fair 21% 264 Poor 5% 65 Number of children 1 child 9% 114 2 children 43% 530 3 children and more 49% 600 Country Austria 6% 76 Germany 6% 71 Sweden 14% 183 Netherlands 8% 97 Spain 5% 60 Italy 5% 66 France 10% 120 Denmark 7% 90 Greece 5% 57 Switzerland 4% 54 Belgium 11% 140 Israel 9% 112 Czech Republic 5% 62 Poland 5% 56 Total 100% 1244

Source: Survey of Health, Ageing and Retirement in Europe 2004-2012

146

To investigate the relationship between the help provided to parents and the care provided to grandchildren on the individual level, two mixed-effects models for binary responses have been created

(see Table 19). The final models show that the variation on the individual level cannot be explained only by differences across countries caused by different social policies. Models control for country fixed effects and there is still a rather strong connection between caring responsibilities of grandchildren and helping elderly parents.

147

Table 19: Odds ratios of grandchildren’ care regardless of the frequency and regular care of grandchildren

Any care of Regular care of grandchildren grandchildren Help to parents (ref. Never) Less than once a month 1.37 1.53 Once a month 1.72* 0.97 Once a week 1.55* 1.05 Daily 1.83** 1.51 Age of respondent (ref. 50-59) 60-69 0.74* 0.99 Health status (ref. Very good) Good 0.91 0.88 Fair 0.68* 0.66 Poor 0.42** 0.74 Employment status of respondent (ref. Retired) Employed 0.80 0.43*** Unemployed 0.73 1.29 Sick or disabled 0.92 1.07 Homemaker 0.71 0.72 Education of respondent (ref. ISCED 0-1) ISCED 2-3 1.91*** 2.07** ISCED 4-6 2.37*** 2.82** Gender of respondent 1.79*** 2.28*** Marital status of respondent (ref. married) 1.36* 1.50* Geographical proximity (ref. Same household) Same building 4.81 1.08 1-25 km 1.39 0.29* 26-100 km 0.48* 0.03*** More than 100 km 0.13*** 0.04*** Child's marital status (ref. Married) Separated/divorced/widowed 0.91 0.92 Never married 0.79 1.19 Age of youngest grandchild (0-3 y.) 4-6 years 1 .24 0.97 7-15 years 0.73* 0.89

148

Table 19 continued

Child's employment (ref. Employed) Unemployed 0.53* 0.30**

Out of labor market 0.84 0.61* Year of survey 1.04 1.00 Attitudes toward grandparents' obligations 1.37*** 1.28* Country (ref. Austria) Germany 1.44 0.40 Sweden 1.19 0.21** Netherlands 3.03** 0.65 Spain 0.98 1.97 Italy 1.26 4.29* France 3.73*** 0.32* Denmark 3.79*** 0.14* Greece 2.13 2.19 Switzerland 2.04 0.96 Belgium 3.35*** 1.18 Israel 1.01 1.49 Czech Republic 1.15 0.82 Poland 0.61 0.94 Observations (N - level 1 - year) 3267 1952 N (children - level 2) 1828 1228 N (parents - level 3) 1244 947

Random effects Standard Standard deviation deviation Family RE 1.22 1.49 Individual RE (panel ID) 0.73 0.00

Note: Mixed-effects models for binary responses: odds ratios. Significance levels: *** 0.1%, ** 1%, * 5% Source: Survey of Health, Ageing and Retirement in Europe 2004-2012

149

The direction of the association is the same as for the macro level analysis. People who provide help to their parents also tend to look after their grandchildren more than their counterparts without helping responsibilities. In addition, the tendency to help even increases with the increasing frequency of the provided help. The dependency is still present after controlling for age and subjective health of respondents.

This pattern applies not only to any care (Model 1 in Table 20), but it holds also in the model distinguishing between regular and occasional care (Model 2 in Table 20), even though the effect is somewhat smaller in this case and it reaches the level of statistical significance only in case of daily care.21

Regular carers, who provide help to their parents daily, tend to look after their grandchildren on a regular basis; on the contrary, people who provide occasional help to their parents also tend to look after their grandchildren rather sporadically. This negative effect is particularly strong for people who never provide any help to their parents since these people are the least prone to look after their grandchildren. We can therefore suppose that the assistance provided to one family member generally does not decrease the probability of help to other family members. Furthermore, the individual random effects show a very low variability over time (approaching zero in the first model), indicating a strong stability of the caring tendencies.

Other factors affecting the probability to look after grandchildren are similar for both models and are in accordance with previous findings. The likelihood of providing care of grandchildren is negatively associated with age (both grandparents' and grandchildren's), poor health, lower education and being employed or unemployed (in comparison with retirement). Furthermore, the employment status of children has also a significant effect since children who are employed receive help from their parents more often than children currently unemployed or out of the labour market. The child's marital status does not affect significantly the probability of grandparents' involvement in the grandchildren' care, but it

21 The number of observations is, however, lower in the case of regular care and for that reason, the results are less significant, even though the odds ratios are not so much lower than in the first model.

150

is obvious that never-married children tend to receive regular help from their parents more often than children who are married or divorced. This could be explained by a higher prevalence of single parents among never-married people. Furthermore, the effect of geographical proximity is remarkable. The probability that grandparents look after grandchildren is higher if grandparents share the same building with their children than if they share the same household. However, respondents were asked if they had looked after grandchildren without the presence of parents. Grandparents sharing the same household with their children and grandchildren probably spend a significant amount of time with their grandchildren but usually the parents are also present. Furthermore, those people who share the same household with their grandchildren might not report the time spent with their grandchildren. In some countries, the coresidency of parents and adult children is widely acceptable and common (Albertini, Kohli and Vogel 2007; Albertini and Kohli 2013). People in multigenerational households can feel that they do not look after their grandchildren but simply spend time with them.

One of the most important factors is the indicator of normative attitudes toward grandparents' obligations. Respondents who tend to agree that grandparents are obligated to support their grandchildren tend to look after them considerably more often than grandparents who do not think that it is the grandparents' obligation. This effect is not surprising; however, the normative attitudes obviously do not explain the association between help provided to parents and care provided to grandchildren. The explanation might be therefore related to other factors such as emotional bonds between family members or to the saturation of respondent' social needs rather than normative obligations.

The country effects show persisting significant differences across countries even after controlling for many individual factors. Sweden, the Netherlands, France and Denmark show the highest tendency to any care of grandchildren, but lower relative tendencies toward regular care. On the contrary, Italy and

Greece have the highest probability of being involved in the regular care of grandchildren.

151

9.7 SUMMARY AND DISCUSSION: FAMILY SOLIDARITY ACROSS GENERATIONS

The present analysis follows the findings from research on intergenerational relationships that suggests there are increasing demands on the pivot generation. A substantial number of middle-aged people provide help to either their adult children or elderly parents or both. Providing instrumental help is related to different structures of needs and resources. Unlike the resources of emotional closeness, individual resources of time and energy are limited (Grundy and Henretta 2006). The resources can be therefore exhausted by supporting some family member, while other family members are deprived of help. On the other hand, people who tend to support their relatives may share some characteristics that influence the support of both grandparents and grandchildren.

To investigate these two competing hypotheses, the association between looking after grandchildren and providing help to aging parents has been examined. It is shown here that the positive association between caring responsibilities exists on the country level. The effect is opposite if the regular care of grandchildren is considered. People in northern countries care for their grandchildren rather sporadically but a higher number of them devote their time to elderly parents. Scandinavian countries are known for their high availability and accessibility of both childcare and care services for older people. The state, however, obviously does not crowd out the family but it rather contributes to the voluntary and pleasurable character of family relationships.

Individual results indicate that the hypothesis of family solidarity is closer to reality. The care of grandchildren by people aged 50 or more with at least one grandchild younger than 15 years is much more frequent for people who also provide some help to their parents. The positive effect of helping to elderly parents persists even when regular care has been considered. People who help their parents daily also tend to look after their grandchildren regularly. This association is not caused by the most relevant factors such as health status, employment status or age of respondents, because the positive relationship between two analysed caring responsibilities is still clearly present after controlling for these confounding

152

variables. Furthermore, controlling for attitudes toward grandparents’ obligations does not eliminate the association between help provided to parents and care of grandchildren. Thus, it is suggested that the tendency to care is not strictly a matter of solidarity or at least not the normative one.

The results indicate that instead of competition between different responsibilities, there can be some general tendency to care, which is based on factors not considered in this analysis, such as emotional closeness between family members and the saturation of respondent’s social needs. The caring responsibilities can be therefore perceived from a cumulative perspective. Responsibilities of a specific group of people do not compete but accumulate and further investigation is necessary to analyse potential risks of excessive burden related to multiple caring responsibilities.

Future research should focus on several questions. First, what is this general tendency to care and how is related to other individual characteristics? The present analysis does not control for a full range of respondents' attitudes toward family norms and values that are likely linked to their behaviour and relationships with relatives. Second, how do multiple caring responsibilities interact with other demands outside the family and do they have any consequences for the providers of care? Providing help to other people likely limits other activities of caregivers. Do people who simultaneously support more family members restrict their working activity? Does extensive helping to relatives influence the caregivers’ health? Thirdly, available data do not allow a detailed investigation of the different intensity of care and help since SHARE researched the type of provided help only in the first two waves. However, the more demanding regular personal care of a dependent parent can have negative effect on the likelihood of looking after grandchildren. Future research should consider these distinctions. Finally, only a limited examination of country differences has been provided in this paper. It is not clear how different welfare regimes and normative structures interact with multiple caring responsibilities across European countries.

Including more countries in the analysis and employing a multilevel approach could explain country-level differences in a more detailed manner.

153

CONCLUSION AND DISCUSSION

The main goal of the thesis was to find an answer on the question whether, and if so than how is the fertility connected with various forms of intergenerational solidarity. As has been stated in the introductory part, the question is not new in fertility research as it has been repeatedly shown that people are in their decisions influenced by their network partners. However, there is still a lack of studies that show the effect of different forms of support from family members and other close people. This thesis therefore strives to contribute to this research field by analyzing international data from several resources to answer a part of the emerging questions, even though it cannot aim for the exhaustive coverage of this topic.

The first part of the thesis summarizes the theory and research findings on intergenerational solidarity. A several factors are important for further research. Firstly, the solidarity within family members is still of a great importance even in modern individualized society. Despite earlier assumption, the welfare state does not crowd-out the family and channels of private support. However, the welfare state still plays an important role as the patterns of family solidarity differ across European countries.

Secondly, the intergenerational solidarity is a multidimensional concept. It is not only a matter of a direct support, but it should also deal with a latent or normative solidarity. The solidarity can be important also on a symbolic level and normative solidarity to some extent correspond to a potential support that can be activated in case of need. Thirdly, there is no cohesive theory of individual solidarity since the motives for giving are not strictly altruistic, but the exchange motives are also not universal. It is supposed that both motives are usually to some extent present. The givers react on the needs of the close people, but they are more willing to help if they can expect something in return. Finally, what is important for the fertility topic, the most important channel of support exists between parents and their children. Parents are givers in this relationship to their late age, while children obtain the support even if they are adult and living more or less independently of their parents. This form of support is often connected with a life- course steps such as finishing the education, arranging and independent living, getting married or having 154

a child. In view of these findings a substantial part of the presented study focuses on the relationship between parents and their adult children, even though other social network partners can also play a role.

The next part deals with the fertility in Europe and attitudes towards family issues across countries and time. The low levels of fertility have been associated with the changes in values and norms represented, for example, by the decrease in the preferred number of children and an increasing tendency to childlessness. However, as shown later, people do not only prefer a lower number of children, but they are even unable to achieve their desired family size. This leads to a belief that people are prevented to have more children by external factors. The existing research suggest that especially women face various obstacles connected with the changes in gender roles. Despite prevailing beliefs, there is no linear association between women´s labor market participation and fertility levels. However, family-life conflict that occurs in countries with a high female employment has to be balanced by external sources of support. The comparison of attitudes toward traditional family and gender roles also disproves that more liberal attitudes and relativization of family norms lead to a break-up of the family since more liberal countries tend to show higher fertility rates. What seems now indisputable is the effect of social support in current European countries. It has been also suggested that the public support in form of welfare state services and financial benefits can be to some extent substituted or supplemented by a private form of support. Previous research focused mostly on the childcare by grandparents and a positive effect on fertility and female labor market participation seems to be present. However, the other forms of private support are less examined.

As stated above, the intergenerational solidarity is a multidimensional concept and besides support which is manifested by real transfers of money, help or gifts, it can be approached also as attitudes or emotional support as a latent or normative sort of solidarity. The first analytical chapter deals with the effect of the parents´ normative attitudes towards parental and grandparental obligations on the probability of the childbirth. It is suggested that children with parents who believe in obligations of older generations to support their offspring are more prone to having a baby. However, a significant effect

155

exists only for the first child which might indicate that a stated support does not always correspond to a real support. Parents may be willing to help but they do not dispose of necessary resources for a support.

Overall, the results indicate that the effect of support is not limited to real forms of support and the whole social environment can be important for the childbearing decisions. The investment into children is less risky if parents can rely on other people.

Children represent substantial costs, and this leads to an assumption that people with a higher income and wealth will be more willing to have a child sooner or have more offspring since they do not suffer an uncertainty. However, as shown in chapter 6, it is not always necessarily true. The positive effect of income exists for more educated women. They tend to plan to have a child within three years with a higher probability if their household´s income is high. However, the effect is negative for women with a low education. It is suggested that having a child may be a strategy of reducing the uncertainty if a woman can rely on other resources. A tentative prove of this hypothesis is the negative effect of income in

Scandinavian countries that provide a great deal of support to parents with young children. Finally, the analysis also reveals that the effect of income is negative for the first child but positive for the second and another child. As indicated by previous research, decision about the first child is substantially different than decision about other child since the first child represents the transition to parenthood. It is suggested that the income is more important for having more than one child since a marginal utility of the second and every other child is lower. People with a higher income can afford to have more children.

A similar effect is than observed for the financial transfers provided by parents to adult children.

While it does not show any relevance for a decision about the first child, there is a significant positive effect of financial help for higher order children. Again, it is not tested the support directly aimed on parents with young children, but it is rather assumed that parents who have once supported their adult children are generally more willing and able to support them in the future. Even though the support is not provided now, it still can be renewed in the future.

156

The next chapter verify if the effect exist also for short-term fertility intentions and two forms of support are compared – childcare and financial transfers. Furthermore, the analysis deals with a gender effect. Most of the research in this field focus on women who are supposed to be those who control the reproduction, or the research does not differ men and women. However, the decision about the childbirth usually depends on both partners and at the same time, the costs of the children may differ for men and women. When the child is born, men are still those who secure the financial issues, while women provide a higher proportion of housework and childcare. The financial transfers from the family and other close people are therefore more important for men when deciding about the childbirth, while a help with the childcare is more important for women.

To sum up, the analysis confirms that caring for children is still mostly female task. Women are also women expected to participate on the labor market. Some countries (i.e. Scandinavian or France) deal with that issue by providing a wide range of childcare services from early age of the child. Women in other countries still depend to some extent on informal help. Informal childcare is mostly provided by grandmothers; however, the middle-aged generation face multiple roles conflict as they participate on the labor market to later ages and they are supposed to provide support to their ageing parents or spouse. The final chapter of the thesis therefore examine if those women who provide help or care to their parents are prevented from the childcare of their grandchildren. The results show that those who care of their parents also tend to care more of their grandchildren. Even though these findings are positive from the perspective of childcare and older people care, it can mean an overloading of a certain part of the middle-aged generation or early retirement. The question still is what the effect of multiple caring roles on the intensity and frequency of care is. The presented analysis indicates that those who care regularly of their parents are still more prone to care of their grandchildren; however, the frequency and intensity of grandchildren care is not tested. Further research is necessary in this field.

Overall, the thesis confirms the assumptions based on previous research. From the perspective of the intergenerational solidarity it can be said that the family is still important, even though its role may

157

change as a reaction on the changes in external institutions. The reproduction can be an example of areas where close and extended family relationship intervene in a positive way. This research follows the social network research and it confirms that childbearing decisions are not taken in an isolation. Here, the focus was only on a part of the impact which can social network partners make – the social support – but the influence can be much broader in form of a social contagion, reproduction of behavior patterns among generations or social pressure. It is important to note that it is impossible to separate these impacts as they inevitably work together. The social support can then be an instrument of social pressure since it makes a commitment to the provider of the support. As has been stated in the theoretical part, the exchange motives and altruistic motives can be present simultaneously. The results of the thesis therefore should not be interpreted as a demonstration of strictly altruistic bonds among family members. The focus was here on the impact these bonds have on the fertility behavior and fertility intentions and not on the motives of the givers.

158

REFERENCES

Aassve, A., Arpino, B., Goisis, A. (2012). Grandparenting and Mothers’ Labour Force Participation: a Comparative Analysis Using the Generations and Gender Survey. Demographic Research, 27 (3), 53-84. Aassve, A., Meroni, E., & Pronzato, C. (2012). Grandparenting and Childbearing in the Extended Family. European Journal of Population/Revue européenne de Démographie, 28(4), 499-518. Adams, James D. (1980). Personal Wealth Transfers. The Quarterly Journal of Economics, 95(1), 159-179. Adsera, A. (2004). Changing fertility rates in developed countries. The impact of labor market institutions. Journal of Population Economics, 17(1), 17-43. Adsera, A. (2005). Vanishing children: From high unemployment to low fertility in developed countries. The American economic review, 95(2), 189-193. Adsera, A. (2011). Where are the babies? Labor market conditions and fertility in Europe. European Journal of Population/Revue européenne de Démographie, 27(1), 1-32. Ajzen, I. (1991). The theory of planned behavior. Organizational behavior and human decision processes, 50(2), 179-211. Ajzen, I., & Klobas, J. (2013). Fertility intentions: An approach based on the theory of planned behavior. Demographic Research, 29, 203. Albertini, M., Kohli, M. (2012). The Generational Contract in the Family: The Analysis of Transfer Regimes in Europe. European Sociological Review. Albertini, M., Kohli, M., and Vogel, C. (2007). Intergenerational Transfers of Time and Money in European Families: Common Patterns—Different Regimes? Journal of European social policy, 17(4), 319-334. Aldous, J. (1990). Family Development and the Life Course: Two Perspectives on Family Change. Journal of Marriage and Family, 52(3), 571-583. Allison, P. D. (2005, August). Causal inference with panel data. In Amer. Sociol. Association Annual Meeting. Altonji, Joseph G.,Hayashi, F., Kotlikoff, Laurence J. (1992) Is the Extended Family Altruistically Linked? Direct Tests Using Micro Data. The American Economic Review, 82(5), 1177-1198. Altonji, Joseph G., Hayashi, F., Kotlikoff, Laurence J. (1997). Parental Altruism and Inter Vivos Transfers: Theory and Evidence. Journal of Political Economy, 105(6), 1121-1166. Altonji, Joseph G., Hayashi, F., Kotlikoff, Laurence J. (2000). Sharing the Wealth: Demographic Change and Economic Transfers between Generations. New York: Oxford University Press. Amuedo-Dorantes, C., & Kimmel, J. (2005). The motherhood wage gap for women in the United States: the importance of college and fertility delay. Review of Economics of the Household, 3(1), 17-48. Andersson, G. (2000). The Impact of Labour-Force Participation on Childbearing Behaviour: Pro-cyclical Fertility in Sweden during the 1980s and the 1990s. European Journal of Population/Revue européenne de démographie, 16(4), 293-333.

159

Andersson, G. (2008). A Review of Policies and Practices Related to the 'Highest-Low’ Fertility of Sweden. Vienna Yearbook of Population Research, 89-102. Andersson, G., Kreyenfeld, M., & Mika, T. (2009). Welfare state context, female earnings and childbearing in Denmark and Germany. Stockholm Research Reports in Demography, 13. Angelini, V., Laferrère, A., & Pasini, G. (2011). Nest leaving in Europe. In The individual and the welfare state (pp. 67-80). Springer Berlin Heidelberg. Aquilino, W. S. (2005). Impact of family structure on parental attitudes toward the economic support of adult children over the transition to adulthood. Journal of Family Issues, 26(2), 143-167. Arnett, J. J. (1997). Young People’s Conceptions of the Transition to Adulthood. Youth & Society, 29, 1–23. Arnett, J. J. (1998). Learning to Stand Alone: The Contemporary American Transition to Adulthood in Cultural and Historical Context. Human Development, 41, 295–315. Arnett, J. J. (2001). Conceptions of the Transition to Adulthood: Perspectives from Adolescence through Midlife. Journal of Adult Development, 8(2), 133-143. Arpino, B., & Bordone, V. (2014). Does Grandparenting Pay Off? The Effect of Child Care on Grandparents' Cognitive Functioning. Journal of Marriage and Family, 76(2), 337-351. Arrondel, L., Masson, A. (2001). Family Transfers Involving Three Generations. The Scandinavian Journal of Economics, 103(3), 415-443. Arrondel, L., Masson, A. (2006). Altruism, exchange or indirect reciprocity: what do the data on family transfers show. In Serge-Christophe Kolm, Jean Mercier Ythier (Eds.), Handbook of the Economics of Giving, Altruism and Reciprocity, 2, 887-1588. Attias-Donfut, C., Ogg, J., and Wolff, F.-C. (2005). European Patterns of Intergenerational Financial and Time Transfers. European Journal of Ageing, 2, 161-173. Axinn, William G., Clarkberg, Marin E., Thornton, A. (1994). Family Influences on Family Size Preferences. Demography, 31(1), 65-79. Balbo, N., & Barban, N. (2014). Does fertility behavior spread among friends? American Sociological Review, 79(3), 412-431. Balbo, N., Barban, N., & Mills, M. (2013). Friend and peer effects on entry into marriage and parenthood: A multiprocess approach (No. 056). Balbo, N., Mills, M. (2011). The Effects of Social Capital and Social Pressure on the Intention to Have a Second or Third Child in France, Germany and Bulgaria, 2004-05. Population Studies, 65(3), 335-351. Ball, R., & Chernova, K. (2008). Absolute income, relative income, and happiness. Social Indicators Research, 88(3), 497-529. Barber, Jennifer S. (2000). Intergenerational Influences on the Entry into Parenthood: Mothers’ Preferences for Family and Nonfamily Behavior. Social Forces, 79(1), 319-348. Barbieri, P., Bozzon, R., Scherer, S., Grotti, R., Lugo, M. (2015). The Rise of a Latin Model? Family and Fertility Consequences of Employment Instability in Italy and Spain. European Societies, (ahead-of-print), 1-24. 160

Barnett, M. A., Scaramella, L. V., Neppl, T. K., Ontai, L. L., & Conger, R. D. (2010). Grandmother involvement as a protective factor for early childhood social adjustment. Journal of Family Psychology, 24(5), 635. Becker, G. S. (1960). An economic analysis of fertility. In Demographic and economic change in developed countries (pp. 209-240). Columbia University Press. Becker, Gary S., Tomes, N. (1979). An Equilibrium Theory of the Distribution of Income and Intergenerational Mobility. The Journal of Political Economy, 87(6), 1153-1189. Becker, G. S., & Lewis, H. G. (1973). New economic approaches to fertility. The Journal of Political Economy, 81(2), 279-288. Bergnéhr, D. (2009). Social influence and the timing of parenthood. Interpersona, 3, 61. Bernardi, L. (2003). Channels of social influence on reproduction. Population Research and Policy Review, 22(5- 6), 427-555. Bernardi, L., Keim, S., & von der Lippe, H. (2007). Social influences on fertility a comparative mixed methods study in Eastern and Western Germany. Journal of Mixed Methods Research, 1(1), 23-47. Berrington, A. (2004). Perpetual postponers? Women's, men’s and couple’s fertility intentions and subsequent fertility behaviour. Population trends, 117, 9-19. Berry, B. (2008). Financial Transfers from Living Parents to Adult Children: Who is Helped and Why? American Journal of Economics and Sociology, 67(2), 207-239. Billari, F. C. (2004). Becoming an Adult in Europe: A Macro (/Micro)-Demographic Perspective. Demographic research, 3(2), 15-44. Billari, F. C. (2009). The happiness commonality: Fertility decisions in low-fertility settings. How generations and gender shape demographic change, 7-38. Billari, F. C., Wilson, C. (2001). Convergence towards diversity? Cohort dynamics in the transition to adulthood in contemporary Western Europe. Max Planck Institute for Demographic Research, Working Paper WP2001-039. Billari, F., Kohler, H. P. (2004). Patterns of Low and Lowest-Low Fertility in Europe. Population studies, 58(2), 161-176. Billari, F. C., Philipov, D., & Testa, M. R. (2009). Attitudes, norms and perceived behavioural control: Explaining fertility intentions in Bulgaria. European Journal of Population/Revue européenne de Démographie, 25(4), 439. Billingsley, S., & Ferrarini, T. (2014). Family Policy and Fertility Intentions in 21 European Countries. Journal of Marriage and Family, 76(2), 428-445. Blanchet, D., & Ekert-Jaffe´, O. (1994). The demographic impact of fertility benefits: Evidence from a micro- model and from macro-data. In J. Ermisch & N. Ogawa (Eds.) The family, the market and the state in ageing societies (pp. 79–104). Oxford, England: Clarendon Press. Blinder, Alan S. (1973). A Model of Inherited Wealth. The Quarterly Journal of Economics, 87(4), 608-626.

161

Blossfeld, H.P., Hofäcker, D. (2011). Youth on Globalised Labour Markets : Rising Uncertainty and its Effects on Early Employment and Family Lives in Europe. Opladen: Barbara Budrich Publishers. Bolin, K., Lindgren, B., & Lundborg, P. (2008). Informal and formal care among single‐living elderly in Europe. Health economics, 17(3), 393-409. Bongaarts, J. (2001). Fertility and Reproductive Preferences in Post-Transitional Societies. Population and Development Review, 27, Supplement: Global Fertility Transition, 260-281. Bonoli, G. (2008). The Impact of Social Policy on Fertility: Evidence from Switzerland. Journal of European Social Policy, 18(1), 64-77. Bonsang, E. (2007). How do Middle-aged Children Allocate Time and Money Transfers to Their Older Parents in Europe? Empirica, 34(2), 171-188. Borg, M. O. M. (1989). The income-fertility relationship: Effect of the net price of a child. Demography, 26(2), 301- 310. Brandt, M., Haberkern, K., Szydlik, M. (2009). Intergenerational Help and Care in Europe. European Sociological Review, 25(5), 585-601. Bratti, M., & Tatsiramos, K. (2011). The Effect of Delaying Motherhood on the Second Childbirth in Europe. Journal of Population Economics, 25(1), 291-321. Brody, Elaine M., Litvin, Sandra J., Hoffman, Ch., Kleban, Morton H. (1995). Marital Status of Caregiving Daughters and Co-Residence with Dependent Parents. The Gerontological Society of America, 35 (1), 75-85. Bromley, M. C., & Blieszner, R. (1997). Planning for long-term care: Filial behavior and relationship quality of adult children with independent parents. Family Relations, 155-162. Bryan, M. L., & Jenkins, S. P. (2015). Multilevel modelling of country effects: a cautionary tale. European Sociological Review, 32(1), 3-22. Brzinsky-Fay, Ch. (2007). Lost in Transition? Labour Market Entry Sequences of School Leavers in Europe. European Sociological Review, 23(4), 409-422. Buchmann, Marlis C., Kriesi, I. (2011). Transition to Adulthood in Europe. Annual Review of Sociology, 37, 481-503. Bulatao, R. A. (1981). Values and Disvalues of Children in Successive Childbearing Decisions. Demography, 18(1), 1. Bühler, C. (2004). Additional Work, Family Agriculture, and the Birth of a First or a Second Child in Russia at the Beginning of the 1990s. Population Research and Policy Review, 23(3), 259-289. Bühler, C., & Fratczak, E. (2007). Learning from others and receiving support: The impact of personal networks on fertility intentions in Poland. European Societies, 9(3), 359-382. Bühler, Ch., Philipov, D. (2005). Social Capital Related to Fertility: Theoretical Foundations and Empirical Evidence from Bulgaria. Vienna Yearbook of Population Research, 22(4), 53-81. Calwell, J. C., Caldwell, P., & McDonald, P. (2002). Policy Responses to Low Fertility and Its Consequences: A Global Survey. Journal of population research, 19(1), 1-24. Castles, F. G. (2003). The World Turned Upside Down: Below Replacement Fertility, Changing Preferences and Family-Friendly Public Policy in 21 OECD Countries. Journal of European social policy, 13(3), 209-227. 162

Cherlin, A., & Furstenberg, F. F. (1986). Grandparents and family crisis. Generations: Journal of the American Society on Aging. Chesnais, J.-C. (1996). Fertility, Family and Social Policy. Population and Development Review 22(4): 729–739. Chesnais, J.-C. (2000). Determinants of Below Replacement Fertility, Population Bulletin of the United Nations, 1999(40/41): 126–136. Choi, N. G. (2003). Coresidence between unmarried aging parents and their adult children: Who moved in with whom and why? Research on Aging, 25(4), 384-404. Choi, Namkee G. (2003). Coresidence between Unmarried Aging Parents and their Adult Children Who Moved in with Whom and Why? Research on Aging 25(4), 384-404. Coall, David A., Hilbrand, S. and Hertwig, R. (2014). Predictors of Grandparental Investment Decisions in Contemporary Europe: Biological Relatedness and Beyond. PloS One, 9(1), e84082. Cohen, A., Dehejia, R., & Romanov, D. (2007). Do financial incentives affect fertility? (No. w13700). National Bureau of Economic Research. Cohen, A., Dehejia, R., & Romanov, D. (2013). Financial Incentives and Fertility. Review of Economics and Statistics, 95(1), 1-20. Cooke, Lynn P. (2004). The Gendered Division of Labor and Family Outcomes in Germany. Journal of Marriage and Family, 66(5), 1246-1259. Cooke, L. P. (2009). Gender equity and fertility in Italy and Spain. Journal of Social Policy, 38(1), 123-140. Cooney, Teresa M. (1989). Co-Residence with Adult Children: A Comparison of Divorced and Widowed Women. The Gerontological Society of America, 29(6), 779-784. Cooney, Teresa M., Uhlenberg, P. (1992). Support from Parents over the Life Course: The Adult Child's Perspective. Social Forces, 71(1), 63-84. Cordoba, J. C., & Ripoll, M. (2015). Intergenerational Transfers and the Fertility–Income Relationship. The Economic Journal. Cox, D. (1987). Motives for Private Income Transfers. Journal of Political Economy, 95(3), 508-546. Cox, D., Rank, Mark R. (1992). Inter-Vivos Transfers and Intergenerational Exchange. The Review of Economics and Statistics, 74(2), 305-314. Cox, D., & Stark, O. (2005). On the Demand for Grandchildren: Tied Transfers and the Demonstration Effect. Journal of Public Economics, 89(9), 1665-1697. Crespo, L., & Mira, P. (2014). Caregiving to elderly parents and employment status of European mature women. Review of Economics and Statistics, 96(4), 693-709. Crompton, R., Lyonette, C. (2006). Work-Life ‘Balance’in Europe. Acta Sociologica, 49(4), 379-393. Daatland, S. O., & Lowenstein, A. (2005). Intergenerational solidarity and the family–welfare state balance. European Journal of Ageing, 2(3), 174-182. Daatland, Svein O., Herlofson, K. (2003). Lost Solidarity or Changed Solidarity: a Comparative European View of Normative Family Solidarity. Ageing and Society, 23(5), 537-560. 163

Danielsbacka, M., Tanskanen, A. O., Jokela, M., & Rotkirch, A. (2011). Grandparental child care in Europe: Evidence for preferential investment in more certain kin. Evolutionary Psychology, 9(1), 147470491100900102. Darroch, J. E. (2008). Male fertility control—where are the men? Contraception, 78(4), S7-S17. Day, C., & Guest, R. (2016). Fertility and female wages: A new link via house prices. Economic Modelling, 53, 121- 132. De La Croix, D., & Doepke, M. (2003). Inequality and growth: why differential fertility matters. The American Economic Review, 93(4), 1091-1113. Del Boca, D. (2002). The Effect of Child Care and Part Time Opportunities on Participation and Fertility Decisions in Italy. Journal of Population Economics, 15(3), 549-573. Del Boca, D., Locatelli, M. (2003). Fertility and Women’s Work Status. The Rationale of Motherhood Choices: Influences of Employment Conditions and of Public Policies, European Commission, Belgium, 149-169. DeLeire, T., Kalil, A. (2002). Good Things Come in Threes: Single-Parent Multigenerational Family Structure and Adolescent Adjustment. Demography, 39(2), 393-413. Di Gessa, G., Glaser, K., & Tinker, A. (2015). The Health Impact of Intensive and Nonintensive Grandchild Care in Europe: New Evidence from SHARE. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, gbv055. DiPrete, T. A., Morgan, S. P., Engelhardt, H., & Pacalova, H. (2003). Do Cross-National Differences in the Costs of Children Generate Cross-National Differences in Fertility Rates? Population Research and Policy Review, 22(5- 6), 439-477. Dimova, R., Wolff, F-C. (2011). Do Downward Private Transfers Enhance Maternal Labor Supply? Evidence from Around Europe. Journal of Population Economics, 24 (3), 911-933. Dommermuth L., J.E. Klobas and T. Lappegård 2009. Now or later? The theory of planned behaviour and fertility intentions. Working Papers 020. "Carlo F. Dondena" Centre for Research on Social Dynamics (DONDENA), Università Commerciale Luigi Bocconi. Dunifon, R., Kowaleski-Jones, L. (2007). The Role of Grandparents in Single-Mother Families. Journal of Marriage and Family, 69 (2), 465-481. Duvander, A. Z., Lappegård, T., & Andersson, G. (2010). Family policy and fertility: Fathers’ and mothers’ use of parental leave and continued childbearing in Norway and Sweden. Journal of European Social Policy, 20(1), 45- 57. Dykstra, P. A., & Fokkema, T. (2011). Relationships between parents and their adult children: A West European typology of late-life families. Ageing & Society, 31(4), 545-569. Eggebeen, David J. (1992), Family Structure and Intergenerational Exchanges. Research on Aging, 14(4), 427-447. Elder, G.H. Jr, Shanahan, M.J. (2006). The Life Course and Human Development. In W. Damon, R.M. Lerner (eds), Handbook of Child Psychology, 665–715. New York: Wiley. Emery, T. (2013). Intergenerational transfers and European families: Does the number of siblings matter? Demographic Research, 29, 247-274. 164

Esping-Andersen, G. 1999. Social Foundations of Postindustrial Economics. New York: Oxford University Press. Ettner, S. L. (1995). The impact of “parent care” on female labor supply decisions. Demography, 32(1), 63-80. Eurostat. (2017a). Fertility statistics. Eurostat – statistics explained. http://ec.europa.eu/eurostat/statistics- explained/index.php/Fertility_statistics [18-09-2017]. Eurostat (2017b). Social protection statistics - social benefits by function. Eurostat – statistics explained. http://ec.europa.eu/eurostat/statistics-explained/index.php/Social_protection_statistics_-_social_benefits [27-09-2017]. Eurostat. (2017c). GDP per capita, consumption per capita and price level indices. http://ec.europa.eu/eurostat/statistics-explained/index.php/Glossary:Actual_individual_consumption_(AIC) [01-09-2017]. Fahlén, S. (2013). Capabilities and Childbearing Intentions in Europe: The association between work–family reconciliation policies, economic uncertainties and women's fertility plans. European Societies, 15(5), 639-662. Fahlén, S., & Oláh, L. S. (2013). Work and Childbearing Intentions from a Capability Perspective: Young Adult Women in Sweden. Childbearing, Women’s Employment and Work-Life Balance Policies in Contemporary Europe. Hampshire: Palgrave and Mac Millan, S28-64. Fahlén, S., & Oláh, L. (2015). The impact of economic uncertainty on childbearing intentions in Europe. Families and Societies Working Paper 36. Fenge, R., Meier, V. (2009). Are Family Allowances and Fertility-Related Pensions Perfect Substitutes? International Tax and Public Finance, 16(2), 137-163. Fergusson, E., Maughan, B., Golding, J. (2008). Which Children Receive Grandparental Care And What Effect Does It Have? Journal of Child Psychology and Psychiatry, 49(2), 161-169. Fingerman, Karen L., et al. (2011). Who Gets What and Why? Help Middle-Aged Adults Provide to Parents and Grown Children. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 66B(1), 87– 98. Fokkema, T., ter Bekke, S. and Dykstra, Pearl K. (2008). Solidarity between Parents and Their Adult Children in Europe. Amsterdam University Press. Fonseca, R., Mullen, K. J., Zamarro, G., & Zissimopoulos, J. (2012). What explains the gender gap in financial literacy? The role of household decision making. Journal of Consumer Affairs, 46(1), 90-106. Frejka, T., Sobotka, T. 2008. Fertility in Europe: Diverse, Delayed and Below Replacement. Demographic Research, 19, 15-45. Friedman, D., Hechter, M., & Kanazawa, S. (1994). A Theory of the Value of Children. Demography, 31(3), 375-401. Fuller-Thomson, E., Minkler, M. (2001). American Grandparents Providing Extensive Child Care to Their Grandchildren: Prevalence and Profile. The Gerontologist, 41, 201–209. Gauthier A.H. (2007a). Becoming a Young Adult: An International Perspective on the Transitions to Adulthood. European Journal of Population, 23, 217–223.

165

Gauthier, A. H. (2007b). The Impact of Family Policies on Fertility in Industrialized Countries: A Review of the Literature. Population Research and Policy Review, 26(3), 323-346. Gauthier, A. H., & Hatzius, J. (1997). Family Benefits and Fertility: An Econometric Analysis. Population Studies, 51, 295–306. Gerster, M., Keiding, N., Knudsen, L. B., & Strandberg-Larsen, K. (2008). Education and second birth rates in Denmark 1981-1994. Demographic Research, 17, 181-210. Gete, P., & Porchia, P. (2014). Fertility and consumption when having a child is a risky investment. Available at SSRN 1735042. Gil, F. (2004). The Uneven Distribution of Family Responsibilities between Women and Men, and its Link with Low Fertility: Some Evidence for European Union Countries from Eurobarometer Data. Papers de demografia, (253), 1. Gibney, S., McGovern, M. (2011). Social Networks and Mental Health: Evidence from SHARE. Journal of Epidemiology & Community Health, 65. Goldscheider, Frances K., Lawton, L. (1998). Family Experiences and the Erosion of Support for Intergenerational Coresidence. Journal of Marriage and the Family, 60 (3), 623-632. Goldscheider, F., Goldscheider, C. (1999). The changing transition to adulthood: Leaving and returning home. Thousand Oaks, CA: Sage. Goldstein, J., Lutz, W., Testa, M. R. (2003). The Emergence of Sub-Replacement Family Size Ideals in Europe. Population research and policy review, 22(5-6), 479-496. González, M. J. (2006). Balancing employment and family responsibilities in Southern Europe. Revue française des affaires sociales, (5), 189-214. Glynn, K., Maclean, H., Forte, T., Cohen, M. (2009). The Association between Role Overload and Women's Mental Health. Journal of Women's Health, 18(2), 217-223. Gray, A. (2005). The Changing Availability of Grandparents as Carers and Its Implications for Childcare Policy in the UK. Journal of Social Policy, 34(4), 557-577. Grundy, E., & Henretta, J. C. (2006). Between elderly parents and adult children: a new look at the intergenerational care provided by the ‘sandwich generation’. Ageing & Society, 26(5), 707-722. Gupta, N. D., & Smith, N. (2002). Children and career interruptions: the family gap in Denmark. Economica, 69(276), 609-629. Gupta, N. D., Smith, N., Verner, M. (2008). The Impact of Nordic Countries’ Family Friendly Policies on Employment, Wages, and Children. Review of Economics of the Household, 6(1), 65-89. Guzman, L. (2004). Grandma and Grandpa Taking Care of the Kids: Patterns of Involvement. Child Trends Research Brief. (Child Trends Research Brief 2004-017). Available from http://www.childtrends.org Hamon, Raeann R. (1995). Parents as Resources When Adult Children Divorce. Journal of Divorce and Remarriage, 23(1-2), 171–184.

166

Hank, K. (2007). Proximity and Contacts between Older Parents and Their Children: A European Comparison. Journal of Marriage and Family, 69(1), 157-173. Hank, K., Buber, I. (2009). Grandparents Caring for Their Grandchildren: Findings from the 2004 Survey of Health, Ageing and Retirement in Europe. Journal of Family Issues, 30(1), 53-73. Hank, K., Kreyenfeld, M. (2003). A Multilevel Analysis of Child Care and Women's Fertility Decisions in Western Germany. Journal of Marriage and Family, 65(3), 584-596. Harknett, K., Billari, F. C., & Medalia, C. (2014). Do Family Support Environments Influence Fertility? Evidence from 20 European Countries. European Journal of Population, 30(1), 1-33. Hass, Paula H. (1974). Wanted and Unwanted Pregnancies: A Fertility Decision‐Making Model. Journal of Social Issues, 30(4), 125-165. Heckman, J. J., & Walker, J. R. (1990). The relationship between wages and income and the timing and spacing of births: evidence from Swedish longitudinal data. Econometrica: Journal of the Econometric Society, 1411-1441. Hensvik, L., & Nilsson, P. (2010). Businesses, buddies and babies: social ties and fertility at work (No. 2010: 9). Working paper//IFAU-Institute for Labour Market Policy Evaluation. Herlofson, K., et al. (2011). Intergenerational Family Responsibility and Solidarity in Europe. NOVA, Norway. Hiel, L., Beenackers, M.A., Renders, C.M., Robroek, S.J., Burdorf, A. and Croezen, S. (2015). Providing Personal Informal Care to Older European Adults: Should We Care About the Caregivers' Health? Preventive Medicine, 70, 64–68. Hirazawa, M., Kitaura, K., & Yakita, A. (2014). Fertility, Intra‐Generational Redistribution, and Social Security Sustainability. Canadian Journal of Economics/Revue canadienne d'économique, 47(1), 98-114. Ho, C. (2013). Grandchild Care, Intergenerational Transfers, and Grandparents’ Labor Supply. Review of Economics of the Household, 13(2), 359-384. Hochman, O. and N. Lewin-Epstein. (2013). Determinants of Early Retirement Preferences in Europe: The Role of Grandparenthood. International Journal of Comparative Sociology 54(1): 29-47. Hoem, B. (2000). Entry into Motherhood in Sweden: The Influence of Economic Factors on the Rise and Fall in Fertility, 1986-1997. Demographic research, 2(4), 1-28. Hoff, A. (2007). Patterns of Intergenerational Support in Grandparent-Grandchild and Parent-Child Relationships in Germany. Ageing and Society, 27(05), 643-665. Hogan, Dennis P., Eggebeen, David J., Clogg, Clifford C. (1993). The Structure of Intergenerational Exchanges in American Families. American journal of Sociology, 1428-1458. Hughes, M. E., Waite, L. J., LaPierre, T. A., & Luo, Y. (2007). All in the family: The impact of caring for grandchildren on grandparents' health. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 62(2), S108-S119. Igel, C., Szydlik, M. (2011). Grandchild Care and Welfare State Arrangements in Europe. Journal of European Social Policy, 21(3), 210-224.

167

Igel, C., Brandt, M., Haberkern, K. and Szydlik, M. (2009). Specialization Between Family and State: Intergenerational Time Transfers in Western Europe, Journal of Comparative Family Studies 40(2), 203–26. Ioannides, Yannis M., Kan, K. (1994). The Nature of Two-Direction Intergenerational Transfers of Money and Time: An Empirical Analysis. Université Catholique de Louvain. Center for Operations Research and Econometrics [CORE]. Isengard, B., Szydlik, M. (2012). Living Apart (or) Together? Coresidence of Elderly Parents and Their Adult Children in Europe." Research on Aging, 34(4), 449-474. Ishikawa, T. (1975). Family Structures and Family Values in the Theory of Income Distribution. The Journal of Political Economy, 83(5), 987-1008. Jones, J. H., & Bird, R. B. (2014). The Marginal Valuation of Fertility. Evolution and Human Behavior, 35(1), 65-71. Jones, L. E., Schoonbroodt, A., & Tertilt, M. (2008). Fertility theories: can they explain the negative fertility-income relationship? (No. w14266). National Bureau of Economic Research. Joshi, H. (2002). Production, reproduction, and education: Women, children, and work in a British perspective. Population and Development Review, 28(3), 445-474. Kalwij, A., & Gustafsson, S. (Eds.). (2006). Education and postponement of maternity: economic analyses for industrialized countries (Vol. 15). Springer Science & Business Media. Kalmijn, M., Saraceno, Ch. (2008). A Comparative Perspective on Intergenerational Support. Responsiveness to Parental Needs in Individualistic and Familialistic Countries. European Societies, 10(3), 479-508. Kaptijn, R., Thomese, F., Van Tilburg, T. G., Liefbroer, A. C. (2010). How Grandparents Matter. Human Nature, 21(4), 393-405. Keim, S. (2011). Social Networks and Fertility Decision-making: A Mixed-methods Study on Personal Relations and Social Influence on Family Formation. VS Verlag. Khadivzadeh, T., Roudsari, R. L., Bahrami, M., Taghipour, A., & Shavazi, J. A. (2013). The influence of social network on couples’ intention to have the first child. Iranian journal of reproductive medicine, 11(3), 209. Knudsen, K. (2012). European Grandparents’ Solicitude: Why Older Men Can be Relatively Good Grandfathers. Acta Sociologica, 55(3), 231-250. Kohler, H. P., & Kohler, I. (2002). Fertility Decline in Russia in the Early and Mid 1990s: The Role of Economic Uncertainty and Labour Market Crises. European Journal of Population/Revue européenne de Démographie, 18(3), 233-262. Kohler, H. P., Billari, F. C., Ortega, J. A. (2002). The Emergence of Lowest-Low Fertility in Europe During the 1990s. Population and Development Review, 28(4), 641-680. Kohli, M. (1999). Private and public transfers between generations: linking the family and the state. European societies, 1(1), 81-104. Kohli, M. (2005). Generational Changes and Generational Equity. In Malcom L. Johnson (Ed.), The Cambridge Handbook of Age and Ageing. Vol. 116. Cambridge: Cambridge University Press, 2005.

168

Kohli, M., & Künemund, H. (2003). Intergenerational transfers in the family: What motivates giving. Global aging and challenges to families, 123-142. Korpi, W. (2000). Faces of Inequality: Gender, Class, and Patterns of Inequalities in Different Types of Welfare States. Social Politics: international studies in gender, state & society, 7(2), 127-191. Kravdal, Ø. (1994). The importance of economic activity, economic potential and economic resources for the timing of first births in Norway. Population studies, 48(2), 249-267. Kravdal, Ø. (2002). The impact of individual and aggregate unemployment on fertility in Norway. Demographic Research, 6(10), 263-294. Kravdal, Ø. (2008). Effects of current education on second-and third-birth rates among Norwegian women and men born in 1964: Substantive interpretations and methodological issues. Demographic Research, 17(17), 211. Kravdal, Ø., & Rindfuss, R. R. (2008). Changing relationships between education and fertility: A study of women and men born 1940 to 1964. American Sociological Review, 73(5), 854-873. Kreyenfeld, M. (2005). Economic uncertainty and fertility postponement. Evidence from German panel data. MPIDR WP 34. Kuhlthau, K., & Mason, K. O. (1996). Market child care versus care by relatives: Choices made by employed and nonemployed mothers. Journal of Family Issues, 17(4), 561-578. Künemund, H., & Rein, M. (1999). There is more to receiving than needing: theoretical arguments and empirical explorations of crowding in and crowding out. Ageing & Society, 19(1), 93-121. Künemund, H., Motel-Klingebiel, A., & Kohli, M. (2005). Do Intergenerational Transfers from Elderly Parents Increase Social Inequality among Their Middle-Aged Children? Evidence from the German Aging Survey. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 60(1), S30-S36. Laferrere, A., Wolff, François Ch. (2006). Microenomic Models of Family Transfers. In Serge-Christophe Kolm, Jean Mercier Ythier (Eds.), Handbook of the Economics of Giving, Altruism and Reciprocity, 2, 887-1588. Lalive, R., Zweimüller, J. (2009). How Does Parental Leave Affect Fertility and Return to Work? Evidence from Two Natural Experiments. The Quarterly Journal of Economics, 124(3), 1363-1402. Laroque, G., & Salanié, B. (2004). Fertility and financial incentives in France. CESifo Economic Studies, 50(3), 423- 450. Lawton, L., Silverstein, S., Bentson, V. (1994). Affection, Social Contact, and Geographic Distance between Adult Children and Their Parents. Journal of Marriage and Family, 53(1), 57-68. Lennartsson, C. (2011). Need and Support: Determinants of Intra‐Familial Financial Transfers in Sweden. International Journal of Social Welfare, 20(1), 66-74. Le Blanc, D., & Wolff, F. C. (2006). Leaving home in Europe: The role of parents’ and children’s incomes. Review of Economics of the Household, 4(1), 53-73. Leopold, T., & Raab, M. (2011). Short‐Term Reciprocity in Late Parent‐Child Relationships. Journal of Marriage and Family, 73(1), 105-119.

169

Leopold, T., & Skopek, J. (2014). Gender and the division of labor in older couples: How European grandparents share market work and childcare. Social Forces, 93(1), 63-91. Lesthaeghe, R., Van de Kaa, D. J. (1986). Twee Demografische Transities? (Two Demographic transitions?), 9-24 In D. J. van de Kaa and R. Lesthaeghe (eds.), Bevolking: Groei en Krimp (Population: Growth and Decline), Deventer, Van Loghum Slaterus. Lesthaeghe, R., & Willems, P. (1999). Is low fertility a temporary phenomenon in the European Union? Population and Development Review, 25(2), 211-228. Letablier, M.T., Hantrais (1996). Families and family policies in Europe. Longman, Londres/New York. Liefbroer, Aart C. (2005). The Impact of Perceived Costs and Rewards of Childbearing on Entry into Parenthood: Evidence from a Panel Study. European Journal of Population, 21, 367-391. Liefbroer, A. C., & Corijn, M. (1999). Who, what, where, and when? Specifying the impact of educational attainment and labour force participation on family formation. European Journal of Population/Revue Européenne de Démographie, 15(1), 45-75. Liefbroer, Aart C., Elzinga, Cees H. (2012). Intergenerational Transmission of Behavioural Patterns: How Similar are Parents’ and Children's Demographic Trajectories? Advances in Life Course Research 17(1), 1-10. Litwin, H., et al. (2008). The Balance of Intergenerational Exchange: Correlates of Net Transfers in Germany and Israel. European journal of ageing 5(2), 91-102.

Lois, D., & Becker, O. A. (2014). Is fertility contagious? Using panel data to disentangle mechanisms of social network influences on fertility decisions. Advances in life course research, 21, 123-134. Luci, A., Thevenon, O. (2012). The Impact of Family Policy Packages on Fertility Trends in Developed Countries. INED Working Paper 174. Luo, Y., LaPierre, T. A., Hughes, M. E., & Waite, L. J. (2012). Grandparents providing care to grandchildren: A population-based study of continuity and change. Journal of Family Issues, 33(9), 1143-1167. Lundström, K. E., & Andersson, G. (2012). Labor Market Status, Migrant Status, and First Childbearing in Sweden. Demographic Research, 27(25), 719-742. Lutz, W., Sanderson, W., & Scherbov, S. (2008). The coming acceleration of global population ageing. Nature, 451(7179), 716. Lyngstad, T. H., & Prskawetz, A. (2010). Do siblings’ fertility decisions influence each other? Demography, 47(4), 923-934. Manacorda, Marco, Moretti, E. (2006). Why do Most Italian Youths Live with Their Parents? Intergenerational Transfers and Household Structure. Journal of the European Economic Association, 4(4), 800-829. Mandic, S. (2008). Home-Leaving and Its Structural Determinants in Western and Eastern Europe: An Exploratory Study. Housing Studies, 23(4), 615-637. Marini, Margaret M. (1984). The Order of Events in the Transition to Adulthood. Sociology of Education, 57(2), 63- 84. 170

Marks, Nadine F., Lambert, James D. and Choi, H. (2002). Transitions to Caregiving, Gender, and Psychological Well‐Being: A Prospective US National Study. Journal of Marriage and Family, 64(3), 657-667. Martínez, Dolores F., Iza, A. (2004). Skill premium effects on fertility and female labor force supply. Journal of Population Economics, 17(1), 1-16. Mathews, P., & Sear, R. (2013). Does the kin orientation of a British woman's social network influence her entry into motherhood? Demographic Research, 28, 313-340. McCoach, D. B., & Adelson, J. L. (2010). Dealing with dependence (Part I): Understanding the effects of clustered data. Gifted Child Quarterly, 54(2), 152-155. McDonald, P. (2000a). Gender equity in theories of fertility transition. Population and development review, 26(3), 427-439. McDonald, P. (2000b). Gender equity, social institutions and the future of fertility. Journal of population Research, 17(1), 1-16. McGarry, K. (2000). Testing parental altruism: Implications of a dynamic model. No. w7593. National Bureau of Economic Research. Menchik, Paul L., David, M. (1983). Income Distribution, Lifetime Savings, and Bequests. The American Economic Review, 73(4), 672-690. Millar, J., Warman, A. (1996). Family Obligations in Europe: the Family, the State and Social Policy. Joseph Rowntree Foundation, York. Mills, M., Mencarini, L., Tanturri, M. L., Begall, K. (2008). Gender Equity and Fertility Intentions in Italy and the Netherlands. Demographic Research, 18(1), 1-25. Mills, M., Rindfuss, R. R., McDonald, P., & Te Velde, E. (2011). Why do people postpone parenthood? Reasons and social policy incentives. Human reproduction update, 17(6), 848-860. Minkler, M., Fuller-Thomson, E., Miller, D., Driver, D. (1997). Depression in Grandparents Raising Grandchildren: Results of a National Longitudinal Study. Archives of Family Medicine, 6(5), 445–452. Monserud, M. A., Elder, G. H. (2011). Household Structure and Children's Educational Attainment: A Perspective on Coresidence with Grandparents. Journal of Marriage and Family, 73 (5), 981–1000. Morgan, S. P. (2003). Is Low Fertility a Twenty-First-Century Demographic Crisis? Demography, 40(4), 589–603. Morgan, S. P., & Rackin, H. (2010). The correspondence between fertility intentions and behavior in the United States. Population and Development Review, 36(1), 91–118. Möhring, K. (2012). The fixed effect as an alternative to multilevel analysis for cross-national analyses. Cologne: University of Cologne (GK Soclife working paper 16). Motel-Klingebiel, A., Tesch-Roemer, C., & Von Kondratowitz, H. J. (2005). Welfare states do not crowd out the family: evidence for mixed responsibility from comparative analyses. Ageing & Society, 25(6), 863-882. Murphy, M. (2013). Cross-national Patterns of Intergenerational Continuities in Childbearing in Developed Countries. Biodemography and social biology, 59(2), 101-126.

171

Murphy, M., Knudsen, L. B. (2002). The Intergenerational Transmission of Fertility in Contemporary Denmark: The Effects of Number of Siblings (Full and Half), Birth Order, and Whether Male or Female. Population Studies, 56, 235–248. Neels, K., Theunynck, Z., & Wood, J. (2013). Economic Recession and First Births in Europe: Recession-Induced Postponement and Recuperation of Fertility in 14 European Countries between 1970 and 2005. International Journal of public health, 58(1), 43-55. Neuberger, F. S., Haberkern, K. (2014). Structured Ambivalence in Grandchild Care and the Quality of Life Among European Grandparents. European Journal of Ageing, 11(2), 171-181. Nielsen, H. S., Simonsen, M., & Verner, M. (2004). Does the Gap in Family‐friendly Policies Drive the Family Gap? The Scandinavian Journal of Economics, 106(4), 721-744. Nilsson, K., Strandh, M. (1999). Nest Leaving in Sweden: The Importance of Early Educational and Labor Market Careers. Journal of Marriage and Family, 61(4), 1068-1079. Ní Bhrolcháin, M., & Beaujouan, É. (2012). Fertility postponement is largely due to rising educational enrolment. Population studies, 66(3), 311-327. OECD (2009). Society at a Glance 2009. OECD Social Indicators. OECD (2012). The Future of Families to 2030. Paris: OECD Publishing. Ogg, J., & Renaut, S. (2006). The support of parents in old age by those born during 1945–1954: A European perspective. Ageing & Society, 26(5), 723-743. Oláh, L. S. (2003). Gendering fertility: Second births in Sweden and Hungary. Population research and policy review, 22(2), 171-200. Opree, S. J., Kalmijn, M. (2012). Exploring Causal Effects of Combining Work an Intergenerational Support on Depressive Symptoms Among Middle-Aged Women. Ageing and Society, 32(01), 130-146. Pavalko E. K and Woodbury S. (2000). Social Roles as Process: Caregiving Careers and Women's Health. Journal of Health and Social Behavior, 41(1), 91-105. Pezzin, Liliana E., Schone, Barbara S. (1999). Parental Marital Disruption and Intergenerational Transfers: An Analysis of Lone Elderly Parents and Their Children. Demography, 38(3), 287-297. Philipov, D. 2009. Fertility intentions and outcomes: the role of policies to close the gap. European Journal of Population 25: 355-361. Philipov, D. (2011). Theories on fertility intentions: A demographer's perspective. Vienna Yearbook of Population Research, 9, 37-45. Philipov, D., Spéder, Z., & Billari, F. C. (2006). Soon, later, or ever? The impact of anomie and social capital on fertility intentions in Bulgaria (2002) and Hungary (2001). Population studies, 60(3), 289-308. Phipps, S., Burton, P., & Lethbridge, L. (2001). In and out of the labour market: long‐term income consequences of child‐related interruptions to women's paid work. Canadian Journal of Economics/Revue canadienne d'économique, 34(2), 411-429.

172

Pink, S., Leopold, T., & Engelhardt, H. (2014). Fertility and social interaction at the workplace: Does childbearing spread among colleagues? Advances in life course research, 21, 113-122. Powdthavee, N. (2011). Life satisfaction and grandparenthood: Evidence from a nationwide survey. IZA Discussion Papers 5869, Institute for the Study of Labor (IZA). Quesnel-Vallée, A., & Morgan, S. P. (2003). Missing the target? Correspondence of fertility intentions and behavior in the US. Population Research and Policy Review, 22(5-6), 497-525. Raab, M., Fasang, A. E., Karhula, A., & Erola, J. (2014). Sibling similarity in family formation. Demography, 51(6), 2127-2154. Rindfuss, R. R., Guilkey, D., Morgan, S. P., Kravdal, Ø., Guzzo, K. B. (2007). Child Care Availability and First-Birth Timing in Norway. Demography, 44(2), 345-372. Roan C. L., Raley R. K. 1996. Intergenerational Co-Residence and Contact: A Longitudinal Analysis of Adult Children's Responses to Their Mother's Widowhood. Journal of Marriage and the Family, 58, 708–717. Rossier, C., & Bernardi, L. (2009). Social interaction effects on fertility: Intentions and behaviors. European Journal of Population/Revue européenne de Démographie, 25(4), 467-485. Saraceno, C., & Keck, W. (2010). Can we identify intergenerational policy regimes in Europe? European Societies, 12(5), 675-696. Schmitt, C. (2012). Labour market integration, occupational uncertainty, and fertility choices in Germany and the UK. Demographic Research, 26, 253-292. Silverstein, M, Bengtson Vern L. (1997). Intergenerational Solidarity and the Structure of Adult Child-Parent Relationships in American Families. American journal of Sociology, 103(2), 429-60. Silverstein, M., Gans, D., Yang, Frances M. (2006). Intergenerational Support to Aging Parents the Role of Norms and Needs. Journal of Family Issues, 27(8), 1068-1084. Silverstein, M. and Marenco, A. (2001). How Americans Enact the Grandparent Role Across the Family Life Course. Journal of Family Issues, 22(4), 493-522. Semyonov, M. Lewin-Epstein, N. (2001). The Impact of Parental Transfers on Living Standards of Married Children. Social Indicators Research 54(2), 115-137. Shanahan, Michael J. (2000). Pathways to Adulthood in Changing Societies: Variability and Mechanisms in Life Course Perspective. Annual Review of Sociology, 26, 667-692. Schenk, N., Dykstra, P., Maas, I. (2010). The Role of European Welfare States in Intergenerational Money Transfers: A Micro-Level Perspective. Ageing and Society, 30(8), 1315-1342. Schoen, R., Kim, Y. J., Nathanson, C. A., Fields, J., & Astone, N. M. (1997). Why Do Americans Want Children? Population and Development Review, 333-358. Smith James P, McArdle John J, Willis Robert. (2010). Financial Decision Making and Cognition in a Family Context. The Economic Journal.120: F363–F380. Sobotka, T. (2004). Is Lowest‐Low Fertility in Europe Explained by the Postponement of Childbearing? Population and Development Review, 30(2), 195-220. 173

Sobotka, T., Skirbekk, V., & Philipov, D. (2011). Economic Recession and Fertility in the Developed World. Population and Development Review, 37(2), 267-306. Sobotka, T., Testa, M. R. (2008). Attitudes and Intentions toward Childlessness in Europe. In People, Population Change and Policies (pp. 177-211). Springer Netherlands. Solomon, J., Marx, J. (1999). Who Cares? Grandparent/Grandchild Households. Journal of Women and Aging, 11, 3– 25. Sommer, K. 2014. Fertility Choice in a Life Cycle Model with Idiosyncratic Uninsurable Earnings Risk. Finance and Economics Discussion Series Divisions of Research & Statistics and Monetary Affairs Federal Reserve Board, Washington, D.C. Spéder, Z., & Kapitány, B. (2009). How are time-dependent childbearing intentions realized? Realization, postponement, abandonment, bringing forward. European Journal of Population, 25, 503–523. Stropnik, N., Šircelj, M. (2008). Slovenia: Generous Family Policy without Evidence of Any Fertility Impact. Demographic research, 19(26), 1019-1058. Szydlik, M. (2012). Generations: Connections across the Life Course. Advances in Life Course Research, 17(3), 100- 111. Taniguchi, H. (1999). The timing of childbearing and women's wages. Journal of Marriage and the Family, 1008- 1019. Tebes, J. K., & Irish, J. T. (2000). Promoting resilience among children of sandwiched generation caregiving women through caregiver mutual help. Journal of Prevention & Intervention in the Community, 20(1-2), 139-158. Testa, M. R., & Basten, S. (2012). Have Lifetime Fertility Intentions Declined During The Great Recession? (No. 9/2012). Vienna Institute of Demography Working Papers. Todd, E. L. (2001). Educational attainment and family gaps in women's wages: evidence from five industrialized countries (No. 246). LIS Working Paper Series. Torche, F. (2011). Is a College Degree Still the Great Equalizer? Intergenerational Mobility across Levels of Schooling in the United States. American Journal of Sociology, 117(3), 763-807. Torr, B. M., Short, S. E. (2004). Second Births and the Second Shift: A Research Note on Gender Equity and Fertility. Population and development Review, 30(1), 109-130. Tölke, A., & Diewald, M. (2003). Insecurities in employment and occupational careers and their impact on the transition to fatherhood in Western Germany. Demographic Research, 9, 41-68. Vandell, D. L., McCartney, K., Owen, M. T., Booth, C., & Clarke‐Stewart, A. (2003). Variations in child care by grandparents during the first three years. Journal of Marriage and Family, 65(2), 375-381. Van Gaalen, Ruben I., Dykstra, Pearl A. (2006). Solidarity and Conflict between Adult Children and Parents: A Latent Class Analysis. Journal of Marriage and Family, 68(4), 947-960. Vikat, A. (2004). Women’s Labor Force Attachment and Childbearing in Finland. Demographic Research, Special collection 3, article 8.

174

Wang, Y., & Marcotte, D. E. (2007). Golden years? The labor market effects of caring for grandchildren. Journal of Marriage and Family, 69(5), 1283-1296. Waldfogel, J. (1998). Understanding the" family gap" in pay for women with children. The Journal of Economic Perspectives, 12(1), 137-156. Waldfogel, J. (2001). International policies toward parental leave and child care. The Future of Children, 99-111. Waldrop, D. P., Weber, J. A. (2001). From Grandparent to Caregiver: The Stress and Satisfaction of Raising Grandchildren. Families in Society, 82, 461–472. Wheelock, J., Jones, K. (2002). ‘Grandparents Are the Next Best Thing’: Informal Childcare for Working Parents in Urban Britain. Journal of Social Policy, 31(03), 441-463. White, L. (1992). The Effect of Parental Divorce and Remarriage on Parental Support for Adult Children. Journal of Family Issues, 13(2), 234-250. White, R. G., & Bernardi, L. (2008). Close kin influences on fertility behavior (No. WP-2008-024). Max Planck Institute for Demographic Research, Rostock, Germany. White, Lynn K., Rogers. Stacy J. (1997). Strong Support but Uneasy Relationships: Coresidence and Adult Children's Relationships with Their Parents. Journal of Marriage and the Family, 59(1), 62-76. Yasuoka, M., & Goto, N. (2011). Pension and Child Care Policies with Endogenous Fertility. Economic modelling, 28(6), 2478-2482.

175

LIST OF TABLES AND GRAPHS

Graph 1: Mothers in the labor market in 2016 ...... 29 Graph 2: Percentage of agreement with the statement “Women has to have children in order to be fulfilled” ...... 31 Graph 3: Percentage of agreement with the statement “Working mother can establish just as warm and secure relationship with her children ...... 33 Graph 4: Percentage of agreement with the statement "A pre-school child is likely to suffer if his or her mother works" ...... 34 Graph 5: Percentage of agreement with the statement "Child needs a home with both a farther and a mother to grow up happily" ...... 36 Graph 6: Percentage of agreement with the statement "A marriage or a long-term relationship is necessary to be happy” ...... 37 Graph 7: Attitudes toward grandparents´ obligations ...... 61 Graph 8: Grandparents involved in childcare, percentage ...... 62 Graph 9: Proportion of subjectively poor people and Actual Individual Consumption (AIC) ...... 83 Graph 10: Factors affecting subjective income and fertility intentions ...... 85 Graph 11: Marginal effects of income across education ...... 92 Graph 12: Marginal effects of savings and debts across education ...... 92 Graph 13: Marginal effects of how easy to borrow money across education ...... 93 Graph 14: Marginal effects of income across number of child ...... 100 Graph 15: Proportion of parents who provided financial transfers in last twelve months ...... 110 Graph 16: Proportion of people who regularly or occasionally look after grandchildren and provide help to parents in European countries ...... 143 Graph 17: Proportion of people who regularly look after grandchildren and provide support to parents in European countries ...... 144

Table 1: Total fertility rate 1980-2015 ...... 22 Table 2: Ideal number of children in 1990, age groups, percentages ...... 25 Table 3: Ideal number of children in 2011 ...... 26 Table 4: Factor analysis and reliability analysis for family obligations attitudes battery ...... 54 Table 5: Correlation between forms of help and care for older people ...... 57 Table 6: Correlation between frequency of childcare and family obligation´s index ...... 59 Table 7: Attitudes towards grandparents´ obligations and number of children, models of multinomial logistic regression ...... 64

176

Table 8: Attitudes toward grandparents´ obligations in 2004/2006 and childbirth in 2010, models of logistic regression ...... 68 Table 9: Associations across individual financial indicators, Goodman and Kruskal's gamma ...... 81 Table 10: Models of short-term fertility intentions; the effect of income: logistic regression ...... 89 Table 11: Models of short-term fertility intentions; the effect of alternative indexes of financial situation; logistic regression ...... 91 Table 12: Models of fertility intentions and income for northern and southern countries ...... 97 Table 13: Descriptive statistics for parent-child dyads ...... 111 Table 14: Model of probability of childbirth in 2010; logistic regression ...... 114 Table 15: Proportion of parents who obtained financial transfer or informal childcare across countries . 125 Table 16: Models of fertility intentions - financial transfers; logistic regression ...... 129 Table 17: Models of fertility intentions - childcare; logistic regression ...... 130 Table 18: Descriptive statistics: Grandparents’ care and help to older parents ...... 146 Table 19: Odds ratios of grandchildren’ care regardless of the frequency and regular care of grandchildren ...... 148

177

AUTHOR INDEX

Aassve, 42, 43, 105, 121 Bonoli, 38, 40, 106 Adams, 10 Bonsang, 14, 16, 105, 133, 134, 137 Adelson, 48 Bordone, 50 Adsera, 27, 75 Borg, 74, 76 Ajzen, 49, 79, 119, 120 Bowen, 138 Albertini, 7, 8, 12, 14, 51, 104 Brandt, 16, 134, 135, 136, 138 Allison, 49 Bratti, 75 Altonji, 10 Brewster, 136 Amuedo-Dorantes, 78 Brody, 12 and, 95 Bromley, 137 Andersson, 27, 35, 39, 73 Brooks, 135 Angelini, 133 Brooks-Gunn, 136 Aquilino, 17 Bryan, 48, 98 Arnett, 18 Brzinsky-Fay, 18 Arpino, 43, 50 Buber, 13, 44, 134, 135 Arrondel, 8, 11, 104 Budig, 77 Attias-Donfut, 12, 13, 117, 136 Bühler, 41, 42, 51, 76, 105, 121 Axinn, 42 Buchmann, 18 Bulatao, 116 Baker, 44 Burton, 77 Balbo, 41, 51, 121 Ball, 84 Caldwell, 106 Barban, 41 Castles, 21, 40, 95 Barber, 42, 120 Clarkberg, 42 Barbieri, 19, 75, 96 Clogg, 13, 42, 50 Barnett, 43 Coal, 135 Basten, 75 Coall, 13, 14, 136, 137 Baydar, 135, 136 Cohen, 27, 40, 106 Beaujouan, 77 Cooke, 27 Becker, 10, 41, 74, 79 Cooney, 12, 17 Bekke, 15, 44, 52, 105 Cordoba, 106 Bengtson, 7, 14, 138 Corijn, 77 Bergnéhr, 41 Cox, 10, 104 Bernardi, 41, 42, 120, 121 Crespo, 132 Berrington, 79, 119, 121 Crompton, 35 Berry, 13, 104 Bhrolcháin, 77 Daatland, 52, 135 Billari, 18, 19, 20, 84, 119 Danielsbacka, 135 Billingsley, 38 Darroch, 79 Bird, 106 David, 10 Blanchet, 106 Day, 76 Blieszner, 137 De La Croix, 74 Blinder, 10 De Winter, 136 Blossfeld, 19 Dehejia, 27, 40, 106 Bolin, 132 Del Boca, 42, 105, 136 Bongaarts, 22 178

Deleire, 43 Gustafsson, 76 Di Gessa, 50 Guzman, 13, 43, 133, 135 Diewald, 76 Dimova, 43 Haas, 50 DiPrete, 95 Haberkern, 16, 45, 134, 135, 136, 138 Doepke, 74 Hamon, 13, 104 Dommermuth, 119 Hank, 13, 14, 28, 42, 44, 105, 134, 135, 136 Dunifon, 43 Harknett, 107 Duvander, 39 Hatzius, 106 Dykstra, 12, 14, 15, 44, 52, 104, 105, 134 Hayashi, 10 Heckman, 76 Eggebeen, 13, 17, 42, 50 Henretta, 41, 133, 136, 138 Ekert-Jaffé, 106 Hensvik, 41 Elder, 18, 43 Herlofson, 14, 52, 53 Elzinga, 42, 120 Hertwig, 13, 14, 135, 136, 137 Emery, 41 Hiel, 138 Engelhardt, 41, 42, 120 Hilbrand, 13, 14, 135, 136, 137 England, 77 Hirazawa, 106 Esping-Andersen, 56 Ho, 104 Ettner, 132 Hoem, 27, 77 Euler, 135 Hofäcker, 19 Hoff, 104 Fahlén, 73, 75, 118 Hogan, 13, 42, 50 Fenge, 40 Hochman, 44, 132, 136 Ferrarini, 38 Hughes, 43, 50 Fingerman, 13, 104, 138 Fokkema, 15, 44, 52, 105, 134 Chernova, 84 Fonseca, 94 Chesnais, 23 Fratczak, 41, 121 Choi, 12, 138 Frejka, 21, 23 Friedman, 116 Igel, 44, 135 Fuller-Thomson, 43, 136 Insegrad, 7 Irish, 137 Gans, 7, 52, 137 Isengard, 133 Gauthier, 18, 106 Ishikawa, 10 Gerster, 77 Iza, 76 Gete, 73 Gibney, 15 Jenkins, 48, 98 Gil, 27 Jones, 44, 73, 106 Glaser, 50 Joshi, 122 Glynn, 44 Goisis, 43 Kalil, 43 Golding, 43 Kalmijn, 7, 44, 52, 137 Goldscheider, 13, 18 Kalwij, 76 Goldstein, 23 Kan, 11 Gonsalkorale, 135 Kapitány, 119 Gonzáles, 96 Kaptijn, 42, 43, 105, 118 Goto, 76 Keim, 42, 120 Gray, 13, 132, 136 Khadivzadeh, 41 Grundy, 41, 133, 136, 138 Kimmel, 78 Guest, 76 Kitaura, 106 Gunn, 135 Klobas, 120 Gupta, 39, 77 Knudsen, 13, 42, 120

179

Kohler, 19, 20, 74 Mathews, 42, 117 Kohli, 7, 8, 11, 12, 14, 51, 104 Maughan, 43 Kondratowitz, 135 McArdle, 94 Korpi, 35 McCoach, 48 Kotlikoff, 10 McDonald, 27, 106 Kowaleski-Jones, 43 McGarry, 11, 13, 104 Kravdal, 73, 77, 122 McGovern, 15 Kreidl, 136 Meier, 40 Kreyenfeld, 28, 42, 73, 76, 77, 105, 136 Menchik, 10 Kriesi, 18 Meroni, 42, 105, 121 Kuhlthau, 13 Mika, 73 Künemund, 11, 15, 104, 135 Mills, 27, 41, 51, 76, 121 Minkler, 43, 136 Laferrere, 9 Mira, 132 Laferrère, 133 Möhring, 98 Laham, 135 Monserud, 43 Lakomý, 136 Moretti, 14 Lalive, 39 Morgan, 106, 119 Lambert, 138 Motel-Klingebiel, 104 Lappegard, 39 Murphy, 42, 120 Laroque, 40, 98, 106 Lawton, 13, 14 Neels, 75 Le Blanc, 133 Neuberger, 45 Lennartsson, 104 Ní, 77 Leopold, 12, 41, 42, 120, 132 Nielsen, 77 Lesthaeghe, 21, 22 Nilsson, 14, 41 Lethbridge, 77 Lewin-Epstein, 13, 44, 132, 136 Ogg, 12, 117, 134, 136 Lewis, 74 Oláh, 75, 77, 118 Liefbroer, 42, 50, 77, 120, 122 Opree, 44 Lindgren, 132 Ortega, 20 Litwin, 104 Locatelli, 19, 136 Pasini, 133 Lois, 41 Pavalko, 138 Lowenstein, 135 Petrová Kafková, 7 Luci, 39 Pezzin, 17 Lundborg, 132 Philipov, 42, 51, 75, 76, 105, 118, 119, 121 Luo, 14, 136 Phipps, 77 Lutz, 23, 132 Pink, 41, 42, 120 Lyngstad, 41 Porchia, 73 Lyonette, 35 Pronzato, 42, 105, 121 Prskawetz, 41 Maas, 12, 14, 104 Manacorda, 14 Quesnel-Vallee, 119 Mandic, 14 Marcotte, 132 Raab, 12, 41 Marenco, 13, 14, 136 Rackin, 119 Marini, 18 Raley, 12 Marks, 138 Rank, 10 Martínez, 76 Rein, 15, 135 Marx, 43 Renaut, 134 Mason, 8, 13, 104 Riley, 138 Masson, 11 Rindfuss, 28, 77, 122, 136

180

Ripoll, 106 Thevenon, 39 Roan, 12 Thornton, 42 Rogers, 7 Tinker, 50 Romanov, 27, 40, 106 Todd, 78 Rossier, 42, 120, 121 Tölke, 76 Tomes, 10 Salanie, 98, 106 Torche, 113 Salanié, 40 Torr, 27 Sanderson, 132 Saraceno, 7, 52, 95, 137 Uhlenberg, 17 Sear, 42, 117 Semyonov, 13 Van Bavel, 136 Shanahan, 18 Van de Kaa, 21 Short, 27 Van Gaalen, 14 Schenk, 12, 14, 104 Vandel, 135 Scherbov, 132 Vandell, 13, 136 Schmitt, 76 Verner, 39, 77 Schoen, 79, 116, 119, 121 Vikat, 73, 98, 106 Schone, 17 Vogel, 14, 51, 104 Schoonbroodt, 73 von der Lippe, 42 Silverstein, 7, 13, 14, 44, 52, 136, 137 von Hippel, 135 Simonsen, 77 Vuri, 136 Skirbekk, 75 Skopek, 132 Waldfogel, 78 Smith, 39, 77, 94 Waldrop, 44 Sobotka, 21, 22, 23, 75 Walker, 76 Solomon, 43 Wang, 132 Sommer, 75 Weber, 44 Spéder, 119 Weitzel, 135 Stark, 104 Wheelock, 44 Strandh, 14 White, 7, 17, 41 Stropnik, 39 Willems, 22 Szydlik, 7, 16, 44, 133, 134, 135, 136, 138 Willis, 94 Wilson, 18 Šircelj, 39 Wolff, 9, 12, 43, 117, 133, 136 Wood, 75 Taniguchi, 122 Woodbury, 138 Tatsiramos, 75 Tebes, 137 Yakita, 106 ter Bekke, 134 Yang, 7, 52, 137 Tertilt, 73 Yasuoka, 76 Tesch-Römer, 135 Testa, 23, 75, 120, 121 Zweimüller, 39 Theunynck, 75

181