Abstract
Parenting support is a new policy field, directed toward teaching parents how to assume their role. Its foundations are embedded in a child-centered social investment approach, which is becoming dominant in Western European welfare states. This article aims at exploring the extent to which ideas underlying these policies are coherent with individual attitudes toward children, parents, their relationship and their change over time. We analyze how these attitudes changed in five selected countries (France, Germany, the Netherlands, the United Kingdom, and Sweden), using data from all four waves of the European Values Study (1981, 1990, 1999, 2008). We also test what kind of values are behind these attitudes, employing logistic regression as method. Our main finding is that there has been a value shift in public sentiments regarding the primacy of children, which is no longer to be viewed as a traditional type of attitude.
Parenting support is a new policy field that encompasses a variety of interventions—ranging from information and advice to education and training—directed at parents and aimed at facilitating parents and carers in their role (Boddy et al., 2009). By providing parents with support and by influencing their understanding and exercising of the parenting role, these policies aim at ensuring that children and youth develop better cognitive, emotional, and social skills. The principles on which parenting support policies are constructed are embedded in a child-centered social investment approach that is traced by Jenson (2004) to have emerged in the 1980s and 1990s in Canada and subsequently traveled all over the world. The turn toward a child-centered social investment approach applied to family policy, emphasizes three elements: lifelong-learning as a form of security in the knowledge society, an orientation toward the future, and the idea that social investment has positive effects not only on individuals but on the community as a whole (Jenson & Saint-Martin 2006). In this respect, child centeredness does not necessarily imply that the interests, well-being, or happiness of children are at the center of policy concerns. Rather, child-centered social investment policies cut across diverse welfare state institutions and political ideologies. Since early childhood is crucial for the upbringing of future citizens, workers, and consumers, early-stage social interventions are required in order to increase competitiveness thanks to “a future of well trained, flexible and productive workers” (Jenson, 2004, p. 431), maintain social order (Lewis, 2011a; Williams & Roseneil, 2004), or help break the cycle of inequality (Esping-Andersen, 2002). For instance, in the United Kingdom, “parenting programmes are far from new but in the last decade, parenting programmes have been designed to encourage the kind of behaviour on the part of the parent that will both promote and reinforce good behaviour in the child” (Lewis, 2011a, p. 107). New Labour policy has had a strong accent on children since the early 2000s, to the point that they have become “the most legitimate welfare subject[s]” since they deserve protection and support to prevent deviant behavior and to become future citizens-workers (Williams & Roseneil, 2004). Esping-Andersen (2002), when criticizing the configuration of conservative-continental welfare states against the rise of “new social risks,” argued that they should direct more funding toward children, as they represent an investment in the future. International organizations have contributed to this policy direction (Organisation for Economic Co-operation and Development [OECD], 2001; UNICEF, 2000). This focus on children’s interests has been accompanied by an approach based on children’s rights (Eurochild, 2010).
In this paradigm, parenthood is constructed as a sort of occupation with a specific set of skills provided to parents and that are instrumental toward achieving specific outcomes for children. In doing so, parenting policies clearly set children as the outcome of the interventions and parents as their instrument. While Esping-Andersen’s (2002) argument that sometimes families that are in complete control of their children may not act in their best interest and that the state should take part of the responsibilities was initially directed mainly at supporting the claim for extending early childcare services, similar views parallel the development of ideas around parenting support measures—the more so because quantity and quality of time dedicated to children matters for their development and it is not evenly distributed across social classes. It follows that this represents a key issue for social policies (Esping-Andersen, 2007).
However, it is not only social policy concerns that set the agenda for social investments, particularly parenting support. Parents themselves worry about the development of their children and how to prepare them for their lives as adults under conditions that are increasingly out of their control. Social and technical innovations and the loosening of social bonds are part of human history, but the accelerating speed of innovations, the amount of media exposure, the rapid spread of information, and the perceived lack of parental control on children’s lives are rather new. Also new is that more people than ever before live at a distance from relatives who can offer support and advice to parents of both newborns and growing children. In other words, parenting support interventions can also be envisioned as fulfilling a demand from increasingly uncertain parents (see also Knijn & Ostner, 2008; Riesman, Glazer, & Denney, 1961/2001). To summarize, on the one hand, parenting support policies fit into a children-first strategy aimed at societal developments that range from nurturing adequate and effective workers for the future, ensuring social cohesion, order, and inclusion, to increasing social equality. The approach sees parents and substitute carers as instruments toward reaching these goals. On the other hand, the notion that parenting support has been developed in reaction to (constructed?) parental demands should not be excluded a priori. The assumption might hold that, more than in the past, parents (and the population in general) assume that difficulties and problems can be solved by professional support, as faith in expert systems is one of the characteristics of late modernity (Giddens, 1991). Parents might also be aware of the social, economic, and cultural effects of a suboptimal education. Hence, they try to optimize their children’s development by the available methods and instruments. Whatever the drivers of parenting support policy are, its implication is that parents are supposed to be less able to solve parenting issues by themselves and are in need of information, expert supervision, and even professional surveillance.
This article aims at exploring the extent to which the ideas underlying parenting support policies are coherent with individual attitudes toward children, parents, their relationships, and their change over time. In other words: Do these new policy directions come as a state intervention to (increasingly) detached and individualized parents? Or, by contrast, do these policies combine well with parents’ need for support in what they aim crucial—their involvement and investment in children? This article does not aim to seek a causal relationship between policy ideas and individual values on parenting though, but rather to determine whether a parallel exists between values promoted by policies at the macro level—a strong focus on the importance of children and their rights and on parents as “instruments” enabling children to properly develop—and those shared by the majority of the population. After presenting our conceptual and theoretical background on the relationships between social policy ideas and individual attitudes and values with specific reference to child centeredness and parental investment in children, we analyze how these attitudes changed in five selected countries (France, Germany, the Netherlands, the United Kingdom, and Sweden) between the early 1980s and the late 2000s. These countries represent different welfare regimes: liberal (the United Kingdom), conservative (France, Germany, and the Netherlands), and social democratic (Sweden; (Esping-Andersen 1990). Moreover, parenting support policies have been introduced in each of these countries (Daly, Knijn, Martin, & Ostner, 2012; Lundqvist, 2012), albeit in the framework of different family policy traditions and through a variety of approaches and policy instruments. For our purposes, we will use data from all four waves of the European Values Study (EVS; 1981, 1990, 1999, 2008) and employ logistic regression as main method.
“Welfare Culture,” Values, and Attitudes
Culture, defined as shared norms, beliefs, values, and attitudes by most of the people who live within one nation-state (Baldock, 1999) is an important variable in understanding the origins and effects of social policy, more specifically of family policy. On the one hand, welfare regimes are based on different sets of values about crucial aspects of human behavior and the role of state institutions in promoting such behavior. On the other hand, governments express and construct cultural values by promoting social policies that prioritize and support some types of behavior, while sanctioning others (Fraser, 1989; Kremer, 2007; Pfau-Effinger, 2005). Moreover, value systems of welfare regimes are not entirely coherent and might deviate in some respect from values shared by parts of the population. An interesting question therefore is whether the meaning attached to child centeredness could be considered as a revival of traditional family/parenting values or does it stand for a different type of attitude?
Values are conceptions of the desirable that guide the way social actors (e.g., organizational leaders, policy makers, individuals) select actions, evaluate people and events, and explain their actions and evaluations (Rokeach, 1973; Schwartz, 1992). Values do refer to general guiding aims, abstract ideas about what is right, good, and desirable in a group and is explicitly or implicitly shared by its members (Williams, 1970). Attitudes orient behavior in more concrete situations and are influenced by the more general abstract values. Having this distinction in mind, this article considers child centeredness as representing attitudes that are embedded in more general values. We assume that these vary per country, given their various historical embedded social policies, academic discourses on parent–child relationships, and family life. On the other hand, travelling ideas (Béland, 2009) may result in some homogenization of attitudes via the international exchange of academic thoughts, media, and policy. Comparative studies show that cultural differences are to be found among national populations as well as between countries (van Oorschot, Opielka, & Pfau-Effinger, 2008).
Modernization theory claims the ensuing breakdown of traditional social ties (like those related to extended family and community/neighborhood (see e.g., Putnam, 2000) and increased rationalization, secularization, and individualization of lifestyle choices (Beck, 1992, Inkeles & Holsinger 1974). However, not all social scientists find support for this claim. For example, Galland and Lemel (2008) have studied the attachment to traditional values, including those related to the family. They conclude that there is a constant tension between two types of values—tradition-oriented versus individualization—and not an actual convergence toward the modern pole of a singular dimension. Inglehart (1977) shows that in postwar modern affluent societies, material security has been achieved and people’s basic needs have been covered; therefore, parents socializing and children socialized in such times will cherish “postmodern values,” such as self-fulfillment and self-expression more than “modern ideals” of rational economic and technological development. In a refinement of his theory, Inglehart concludes that societies can be grouped according to two value dimensions: (a) modernity as expressed in traditional versus secular/rational values and (b) postmodernity as expressed in survival versus self-expression. The groupings obtained tend to be homogeneous according to religious heritage, level of economic development, and occupational structure (prevalence of the industrial or service sector in the economy). Both cultural and structural factors are therefore important in influencing values. Using this paradigm, we could assume that family values, including those on the relationship between parents and their children, might shift between those two poles; values considered traditional in a modernizing world and values considered self-expressive in a postmodern world.
Child Centeredness and Changing Family Values
The value attached to children, their social meaning, and position in society varies historically, across and within societies. The seminal work of Ariès (1962) put forward the idea that the concept of childhood was virtually absent in premodern and early modern societies. Children in lower class families had economic value (Hammel, Johansson, & Ginsberg, 1983), they were contributing to maintaining the household and were therefore an integral part of family and community life. According to Ariès (1962), the attitude of upper-class parents toward children, in contrast, was predominantly characterized by emotional detachment and neglect (for a different view see: Pollock, 1983). And yet it is exactly among the upper classes that important attitudinal changes took place since the Enlightenment. An increasing interest in the destiny of each individual child accompanies the transition to lower mortality rates as well as birth control and reduced fertility rates. The rise of the “modern” family was linked to the social construction of the “modern” child: “a much-loved, emotionally important being who was extremely expensive to raise in terms of parental time, concern and material-social resources” (Johansson, 1987, p. 352). Only after becoming a norm among upper-class families did these ideas also spread across the middle and lower classes. In her account of the rise of a new conception of childhood, Zelizer (1985) shows that early 20th-century American middle-class reformers fostered a sentimentalized view of childhood and the removal of children from work and the streets among the working classes. As a result, in modern societies, children have become economically worthless but psychologically priceless for their parents: their instrumental value has been replaced by their expressive value (Scheper-Hughes & Sargent, 1998). Interestingly enough, the mainstream literature on value change tends to treat child centeredness, manifested through high investment in children and strong parent–child bonds, as a “traditional” trait. According to Inglehart and Baker (2000),
in traditional societies a main goal in life is to make one’s parents proud—one must always love and respect one’s parents, regardless of how they behave. Conversely, parents must do their best for their children even if their own well-being suffers. (p. 25)
However, this form of attachment to and investment in children echoes more the construction of sentimentalized childhood that diffused and crystallized in Western societies in the early 20th-century rather than that of a generic premodern or traditional society.
The second half of the 20th century has seen considerable social and cultural transformations that might have had an impact on the social value of children and on parenting attitudes, also with respect to the centrality of children. Individualization and rationalization are deemed as important drivers of the weakening of family bonds and of “traditional” values. At the structural level, the extreme consequence of individualization theory is an increasing atomization of social bonds, including those between parents and children. Parents would be less inclined to invest economic and emotional resources in their offspring and would privilege their own individual well-being instead. However, despite increasing divorce and cohabitation rates, the extent to which individualism would create an opposition between family- and individual-based values, as Popenoe (1988, p. 305) argues, is contested (Scott & Braun, 2006). According to Riesman et al. (1961/2001, also Knijn & Ostner, 2008) a consequence of individualization is that “other-directedness” substitutes “inner-directedness,” making parents more sensitive to how they, their parental behavior, and their children are envisioned by others. Following that argument leads, paradoxically, to argue that individualization results in less instead of more parenting autonomy. So far however, there is no evidence available, neither to support the increasing value attributed to parental independence and children’s autonomy nor to support the hypothesis of a decreasing importance of closeness between parents and children. By contrast, research shows an increasing time investment in care for children among both male and female parents and grandparents, as well as strong bonds between parents and their adult children, at least in European countries (Bucx, 2009; Kalmijn & Vries, 2009; Kohli & Künemund, 2003). The fact that this increase is concomitant with the growing employment status of mothers and is higher among the middle and upper classes (Bianchi, 2000) further underpins the increasing importance contemporary Western societies attach to parental investments in (the relationship with their) children. On a more theoretical level, it is reasonable to assume that parenting is becoming more important at the micro level and not only at the policy level, also in relation to the deinstitutionalization of marriage (Bianchi, 2000; Cherlin, 2004).
There are thus reasons to expect child centeredness to become an increasingly widespread attitude that may present different value components. The traditional form of child centeredness and parental investment does not necessarily contradict a “social investment strategy.” A specific type of child centeredness should then not be associated with traditional family structures but instead with parents’ concerns about their relationship with, and the well-being and future life chances of their children. Also, it is unlikely that individualization has, so far, diminished the importance parents attach to children and child centeredness. Consequently, our study asks the following two research questions: Is there an increasing convergence in the dominance of child-centered attitudes in the five European countries selected? Which parental values (traditional/authoritarian vs. late modern/autonomy based) are associated with child-centered attitudes, and has this association changed over time? The second question is especially important for understanding the meaning of the changes observed. Is agreement with the primacy of children to be understood as the expression of traditional (early 20th century) authoritarian parental values, or is it the expression of a new (social) investment approach toward children characteristic of late modern values?
Data and Method
The World/ EVS is the largest cross-national, longitudinal, social survey with values as main research focus. It covers value areas such as general life attitudes, family, work, religion, politics, and society. The program started in 1981, with only 12 countries participating, and is repeated every 9 years. The number of countries collecting data, using a standardized questionnaire that is translated into each national language, has increased, reaching 47 in 2008. 1 For each country, the data are collected for a representative sample at the national level. The four consecutive waves in the five selected countries constitute the data set used in our analysis. In Germany, for the first wave only, West Germany was included in the survey. In the following years, representative samples are available for both West and East Germany, separately. Although differences between the two are quite marked, we have decided to focus our analysis on the national level.
Dependent Variable
Child centeredness, as a value orientation, also manifests itself through an attitude that can be labeled as primacy of children. This is measured in our study by agreement with the statement, Parents’ duty is to do the best for their children, even at the expense of their own well-being (vs. Parents have a life and should not be asked to sacrifice their own well-being for the sake of their children or agreement with neither statement), constructed as a dummy variable. It should be underlined that this type of variable has been previously used as part of scales to measure attachment to the traditional family 2 (Galland & Lemel, 2008; Inglehart & Baker, 2000). This item was assumed to measure the opposition between family values typical of the 1950s and increasingly individualized and hedonistic values emerging after the 1970s. For example, Lesthaeghe and Meekers (1986) label this scale familism, identifying two not entirely separated dimensions to it, tolerance of nonconformism in family formation (deinstitualization) and the meaning attached to parenthood. Their analysis, using EVS data (first wave), shows that familialist values can be predicted by other value scales; they are positively associated with religiosity and nationalism and negatively with postmaterialism and leftism. But other scales developed to measure familism do not include any item of the sort and rather stress strong family ties through hierarchical relations (Lugo Steidel & Contreras, 2003). For the reasons discussed in the theoretical section, we consider that this selected item about the primacy of children expresses an attitude toward parenting that might be compatible with the social investment approach and that might be less related to traditional parental values in recent times. Since the dependent variable is a binary one, logistic regression was used to test if differences between countries are significant in each wave (to assess convergence) and to study the profile of people who agree with the aforementioned statement.
Independent Variables
As building a scale for traditionalism that can be used for all five countries and all four waves proves to be a very challenging task, both at the conceptual and the empirical level, we opt for two variables—parental values and self-positioning on the left/right scale as a proxy for it—in order to test if the increase in support for the primacy of children is due to a recrudescence in this type of values or whether it stands for something else. Parental values refer to one dimension, labeled as autonomy, versus authority orientation in child rearing. The battery was designed by Kohn (1959), building on results from previous open-ended questions, in order to study the impact of social class on child-rearing values. He assumed that these values (autonomy oriented for the middle and upper classes vs. authority oriented for the working class) are part of the causal mechanism explaining social reproduction. His analysis shows indeed a latent dimension called self-direction versus conformity, which correlated as expected with social class. It was included in the World/EVS, in the form of choosing 5 out of initially (in the first wave) 17 values, then 13. Following an analysis by Halman and Moors (2003), we have selected four values for each pole of the dimension 3 and computed the index by adding the four related to autonomy and subtracting the four related to authority. Political orientation is measured with self-positioning on the right–left 10-point scale (where 1 means far right and 10 far left). In Western European countries, it correlates with traditional values, people with more modern/autonomous values having a tendency to place themselves more to the left end of the scale (Galland & Lemel, 2008). For the variable country (which tests the convergence assumed by the first hypothesis), the reference category is France, as its evolution is the most stable one (a constant upward trend, with a very high starting point). In the case of Germany, we include only the West in the analysis, as in the East there is a very high variation in the dependent variable that is probably caused by factors related to the 1989 transition.
Control Variables
Both values and interests (and power resources) of social actors can depend on their location within the social structure (gender, social class, employment status). We control for the impact of these variables in order to assess the real effect of parental values on the primacy of children. Phipps (1999) shows that women are less inclined to agree that parents should sacrifice their own well-being in favor of children. They are more likely to be the ones who have to commit to that. From the perspective of the person who has to make the sacrifice and the extent of it, the position held in the labor market could also be important, as employed parents could have more difficulties in reconciling the two tasks. Parents as well as older and less affluent persons are more likely to agree with this statement. Religion has a very important influence on values, at both the societal and the personal level. Inglehart and Baker (2000) show that there are still important differences between Catholic and Protestant countries, but not so much between Catholics and Protestants living in the same national state. Galland and Lemel (2008) also prove that religion has an important impact, but it is rather the contrast between belonging to a religion—any religion, and not between members of different confessions—that is much more significant. Knijn and van Oorschot (2008) show that Dutch people of the Protestant denomination are significantly less likely to support child care–related social investment instruments than religiously unaffiliated individuals. Inglehart (1972) also showed that elites (defined in terms of education and socioeconomic status) hold more postmaterialistic values. We have used all three determinants of status (education, occupation, and wealth) as controls, in separate models, but none of them had a significant impact and are therefore not included in the final model. Studies about value change take into account both the age and the cohort to which respondents belong to as control or independent variables in explanatory models (e.g., Lesthaeghe & Meekers, 1986; Voicu & Tufiş, 2012), because people become more traditional in their views, attitudes, and values as they grow older, while on the other hand each generation is influenced by specific events that take place in their formative and socialization years as well as later in life. Following this, we measure age as a continuous variable, and the cohort as a dummy one formed by three categories: prewar, baby boom, and X generation. The baby boom generation is defined as those born between 1946 and1964 (Owram, 1997), in times of economic growth and security while X generation is more difficult to identify but following a definition by Ulrich (2003), we consider them born after 1964 in our samples. Most of the sources cite 1981-1982 (e.g., McClendon, 2000) as the end of generation X and the start of a so-called generation Y, but people born after these years are only present in the most recent data set (EVS, 2008). For comparability reasons (between the waves), we include them together with the Gen-Xers, considering the whole group as more affected by insecurity (real and perceived) than the baby boomers, and thus holding a different value orientation.
Results
The first step is to look at how the percentages of respondents who agree with the statement, Parents’ duty is to do the best for their children, even at the expense of their own well-being, rather than Parents have a life and should not be asked to sacrifice their own well-being for the sake of their children (or neither statement) evolve over time in the five countries selected, treating West and East Germany separately for reasons related to the 1989 transition.
Agreeing with this statement indicates a higher appreciation of the primacy of children over that of parents. While autonomy-oriented values are becoming increasingly widespread in family values up to the EVS 1999 wave (Hagenaars et al., 2003), agreement with this item, which at first sight seems to indicate a nonindividualistic/traditional perspective, is on an almost constant ascending trend in all five countries since the turn of the century. When analyzing West and East Germany separately, one can note that the pattern is followed mostly by the West (although a very small decrease can be seen between 1999 and 2008), while in the East there is a strikingly low value registered for the 1999 wave. We can only speculate on the reasons for that low value and assume that it might be occasioned by the struggle of East German parents to survive while facing the lost of childcare facilities, income, and employment security. Under such conditions postmodern values are less prioritized.
The only exception is a decrease between 1990 and 1999 for the Netherlands and the United Kingdom, and between 1981 and 1990 in Sweden. In the last wave (2008), the level of support for this item is very similar in four out of the five countries (Germany being the exception, especially the Western part), at around 80%. There has been a rather small increase in the proportion of people agreeing with the primacy of children in Germany, the United Kingdom, and France (less than 10 percentage points) if we compare data from the most recent survey with the first one available, but the difference for the Netherlands is quite striking (almost 25 percentage points). It should be noted that, by contrast, agreement with the statement, One should always love and respect one’s parents, regardless of their qualities and faults has a dissimilar evolution over time, and the correlation between these two items, which initially measured two faces of the same coin, becomes weaker over time (see Figure 2).
Percentage of people agreeing with the statement, Parents’ duty is to do the best for their children, even at the expense of their own well-being.
Percentage of people agreeing with the statement, One should always love and respect one’s parents, regardless of their qualities and faults.
With few exceptions, agreement with unconditional love and respect for parents is constantly decreasing in all five countries, while the opposite holds true for the variable regarding the primacy of children’s well-being over that of their parents. The data support a positive answer to the first research question: child-centered attitudes, as measured by the primacy of children, are increasingly widespread in the five countries—even in Germany, which displays a lower level of agreement (almost 60%)—and converge toward a similar value in 2008. One can also look at how the coefficient of variance is changing from one year to the other within one country, for the variable of interest, in order to tap the extent to which attitudes are becoming normative. Except in the case of Germany (both East and West), where the coefficients are rather stable, in the other countries they have the lowest value in the 2008 wave (around 0.15 for most countries, and even lower, 0.10, for Sweden).
Survival analysis can be used to estimate if the countries’ curves are significantly different. The pair-wise comparisons based on the Tarone–Ware test—used as alternative to the log-rank test in absence of proportional hazards—show nonsignificant differences between the curves of France, Sweden, and the United Kingdom and between those of the Netherlands and West Germany. Only the curve of East Germany is significantly different from that of all the other countries.
The logistic regression results (see Table 1) show that differences between the countries are significant in the first three survey years (1981, 1990, and 1999, with the exception of the United Kingdom in 1981) and become nonsignificant in 2008. Only the difference with Germany remains significant in 2008 too (as well as with Sweden, although this is not visible in the descriptive data, probably because of different categories of the population supporting this item).
Logistic Regression Results for Dependent Variable Agreement With Parents’ Duty Is To Do The Best For Their Children, Even at the Expense of Their Own Well-Being Versus Parents Have A Life and Shouldn’t be Asked To Sacrifice Their Own Well-Being for the Sake of Their Children (+ Neither Statement).
p ≤ .1. **p ≤ .01. ***p ≤ .001.
Convergence in terms of attitudes regarding primacy of children is therefore to be assumed for at least four of the five countries. Concerning our second research question, we see that autonomy-oriented parental values have a strongly significant negative effect on the probability of agreeing that parents should sacrifice their own well-being when their responsibilities toward their children are at stake, but the effect becomes weaker from one survey wave to the other. Not surprisingly, people who situate themselves rather on the right end of the political scale tend to agree with this statement more, but the effect becomes insignificant and the sign of the coefficient is reversed in the 1999 and 2008 waves. This shows a change in the profile of people supporting this view, as the increase we can see in absolute percentages can be attributed to those individuals holding more modern/individualistic views. Catholic affiliation increases the likelihood of agreeing with the aforementioned statement in the first and third waves, while in the 2008 wave it becomes insignificant. Education, income, and social status (which were introduced separately because of possible issues of multicolinearity) are not significant when it comes to support for the idea that parental duties should have primacy over personal well-being. Being married (as opposed to being single) does not seem to have an impact either (and has been excluded from the final model); it is having children which for all survey years seems to have a significant positive effect on the probability of agreeing that children’s interests should have priority over those of their parents. In the first three survey waves, men are more likely to believe that parents should sacrifice for their children. However, a change took place between 1999 and 2008 and the difference in attitudes between the two genders became insignificant. The same type of change can be observed regarding age. Older people tend to agree to a larger extent with the first item, except in 2008, when age has a negative albeit nonsignificant impact on the likelihood of choosing this statement—meaning that the probability of stressing the primacy of children lowers with age. All these findings show that not only the proportion of people who agree that children’s well-being is more important than that of their parents has changed over time but also their profile. Given the fact that the percentages have increased more among those categories that are more involved in childcare (women and younger people), this may mean that an actual change has taken place in the public’s values regarding children and their relationships with their careers.
By identifying the cohort period, we can conclude that it plays a role, as values formed in the early years are supposedly very difficult to change later on. On the other hand, individuals from a cohort can also be influenced in a similar way by events taking place at a certain time. We can clearly note from Table 1 that it is the baby boomers (born between 1946 and 1964) who are significantly less likely to agree with this statement, compared with the younger cohort. One possible explanation is that there is a general reverse modernization trend for younger generations, at least concerning issues related to the primacy of children. Another is that this type of values is more related to another dimension (that of self-expression), rather than degree of modernization as opposed to traditionalism. Although the fairly modest values of approximate R2s suggest the model does not fit the data extremely well, the focus lies on studying the profile of people who agree with this statement, more than on explaining the dependent variable.
Discussion
Agreement about the primacy of children’s well-being over that of their parents is on the rise in the five European countries investigated, and it converges at values around 80% for France, the Netherlands, Sweden, and the United Kingdom, and about 60% in Germany in 2008. Interestingly, both parts of Germany stay behind the tendency of the other countries; the West (56.2%) even more than the East (69.3%). A parallel seems to exist between the individual level (population attitudes, public sentiments) and the macro, policy level concerning child centeredness and the need for parental investments in children, which can be supported by state-guided professional interventions.
Germany deviating from this general pattern of northwestern European welfare states needs some explanation. Overall, motherhood norms appear to be more conservative in the western part of the country (Pfau-Effinger, 2005), and after the transition, institutions are less supportive of balancing care with other activities in the eastern part. As such, the “sacrifice” asked from parents may be perceived as harsher, which may explain why fewer people are willing to make it. The subjective cost of having children in Germany should be perceived as higher, given its relatively low fertility rates (1.3). In 2007, a new law concerning parental leave was passed, inspired by the Nordic social policy model, which allowed a more generous income replacement for parents. It would be interesting to analyze its effect on the agreement with the primacy of children in the next survey wave, which will take place in 2017, and to see if better reconciliation policies and thus a less burdening sacrifice have an impact on the orientation of parents toward child centeredness.
The profile of people who agree that parents’ duty is to do the best for their children, even at the expense of their own well-being, looks quite different in 2008 compared with previous survey years. In the first waves, the categories that hold this view seem to be rather the conservative ones (for example older people, more religious ones, right-wing political orientation, men), while in 2008 many of these coefficients become insignificant and even show a reverse sign. The influence of authoritarian-oriented parental values on the agreement with primacy of children becomes weaker with each survey wave. These findings, together with the increase (steep in some countries, linear and steady in others) in the actual percentage of people who agree with this statement, demonstrates a value shift in public sentiments regarding the primacy of children, which is no longer to be viewed as a traditional type of attitude. As such, it can be assumed to be compatible with a social investment strategy at the policy level. Parenting support programs, as part of this political strategy, fit in with such an attitude and are probably welcome and even requested by child-centered parents in need of finding information and help about the most appropriate and efficient way of investing in their children.
Further research is needed, as the present study is limited by its use of only one item as a dependent variable. Another limitation, although we do not focus on the causal mechanisms that contribute to forming this type of attitude, is the fact that the regression model presented has a limited explanatory power. The endogeneity that is present in model could be due to important variables missing from the data set that could have an impact on the results. Also, the relationship between values and policies is difficult to assess, especially regarding the sense of causality. However, a parallel between the two levels does seem to exist.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the Open Area Research Program “Governing New Social Risks: The Case of Recent Child Policies in European Welfare States,” financed by NWO, ANR, DFG and ESGC.