THE EFFECT OF SOCIO-ECONOMIC FACTORS ON PARENTAL FINANCIAL SUPPORT FROM THE PERSPECTIVES OF THE GIVERS AND THE RECEIVERS

The transition phases into adulthood have been prolonged, fragmented and diversified in many European countries. Diversification and the postponement of adulthood is, at least partially, attributable to two ongoing changes in the labour market: the role of education has become more important and insecurity has intensified (Järvinen and Vanttaja 2001; Furlong and Cartmel 2007; Swartz and O'Brien 2009). The transition to working life has been postponed by the increases in secondary school attendance (OECD 2009: 302). In practice this means that young adults are rarely economically independent in the early years of adulthood, and rely on public as well as private forms of financial support. Benefits from the welfare state and earnings from the labour market have been primary forms of economic welfare of young adults in the Nordic countries. The student grant, and housing and unemployment allowances have become axiomatic for many during the early years of adulthood in the last few decades, whereas the parents' role is largely complementary (Berthoud and Iacovou 2004; Hämäläinen 2004; Björnberg and Latta 2007).

The level of state support has been diminishing however, and the financial situation of young Finnish adults in particular deteriorated during and after the recession of the 1990s (Honkanen 2006; Moisio 2008; van Gerven 2008). According to previous studies (see e.g., Kohli 1999; Björnberg and Latta 2007) parents in higher social strata tend to support their adult children financially, and young adults in need receive financial transfers from their families. Björnberg and Latta (2007), for example, note that young Swedish adults on low incomes are more likely to receive financial support from their families, and in Germany the parents of unemployed young adults and students support their children financially (Kohli 1999). There is even some evidence that parental support is becoming an established compensatory mechanism for many economically vulnerable young adults (Majamaa 2011).

Even if a good economic position predicts the extent of support by parent to their adult children (see e.g., Fritzell and Lennartsson 2005), one would expect that almost all parents could help at least to some extent. In previous studies, however, small amounts of financial support have not been taken into account in surveys, have been excluded from the final analyses, or have not been separately dealt with as distinct from more extensive support (see e.g., Kohli 1999; Attias-Donfut and Arber 2000; Fritzell and Lennartsson 2005; Björnberg and Latta 2007). This study also takes into account these contributions, which are seen as a positive sign of the existence of a safety net for young adults. If there is an interaction between parental socio-economic characteristics and more modest financial support, there may be a stronger and hidden mechanism behind the socio-economic factors that had not been acknowledged in previous studies. This study has another advantage over previous studies in that we had data sets covering two generations, parents and their adult children, and were thus able to form adult child–parent dyads. In short, the analyses are based on data from two postal surveys and two generations: Finnish baby boomers who were born in 1945–1950 and their adult children. These data sets enabled us to assess the extent of parental financial support from the perspectives of both the givers and the receivers.

The financial position of young adults is weak in many European countries, and especially those who live alone are particularly vulnerable (Aassve et al. 2006; Moisio 2008). Prolonged education, atypical employment and the risk of unemployment are all typical features of the post-industrial society, and people with few skills, dependent children and those who rely on social security for long periods are increasingly likely to face economic difficulties (Taylor-Gooby 2004; Bonoli 2005).

The Finnish welfare state's role in providing economic security has weakened since the 1990s recession, and income differences between low- and median-wage groups have increased. Furthermore, the gulf between basic security benefits and earned income has deepened (Honkanen 2006; Moisio 2009). Poverty rates among the students are particularly high. According to Moisio's calculations (2010: 186), the relative poverty rate among 18–34-year-old Finnish students living in a single-person household in 2008 was 90% as opposed to 77% in 1995. The respective rates among unemployed young adults (basic unemployment benefit or labour market subsidy) were 92 and 62%, and among young working young adults it was 10%, against 7% in 1995.

Despite the high poverty rates among these young adults who are not working on full-time basis, because of individually targeted assistance, they are more likely to live independently from their parents than their counterparts in central and southern European countries. In practice, young Nordic adults are moving out of the parental home at a younger age than their European counterparts (Eurostat 2008): the proportion of women aged 18 to 24 living away from the parental home was highest in Finland at 61%, against 50% in the UK, 44% in France and 10% in Italy. Finnish young men are also more likely to live outside the parental home (44%) than their counterparts in the UK (33%), France (30%) and Italy (6%). The trend is similar among the 25–29-year-olds but the percentages are higher (ibid: 156). This tendency to leave the parental home, and the declining role of the welfare state partly explain Finland's high poverty rates among young adults.

Calculations from four comparable surveys drawn up by Statistics Finland – The Savings and Indebtedness survey in 1987, two wealth surveys in 1994 and 1998, and The Housing and Wealth survey of 2004 (Statistics Finland 2008) – show that household disposable income per average member changed quite dramatically in recent decades (Figure 1), increasing by about 40% between 1988 and 2004.

This growth in disposable income is not equally distributed, however. On an aggregate level it applied to all age groups except those under the age of 25. It reached 27% among those aged 25–34, but was most striking in 2004 among the 55–64-year-olds, in other words the so-called Finnish baby boomers, being about 75% on average per household member. This was more than double the increase the 18–24-year-olds. The trend in disposable income is even more striking if housing costs are taken into account in that it is precisely these costs that are prominent amongst younger age groups (Niemelä 2004). Baby boomers are the wealthiest generation in Finnish history thus far, which partly explains the differences.

In terms of economic survival, young adults dependent mainly on earnings from work, state benefits and/or private money transfers from family members. Many factors determine the potential to earn money as well as the role of different income sources. The phase of life, for example, the length of any studies undertaken, the household structure and educational level are all indirectly related to the earnings potential of the individual. On the aggregate level the trade cycle, the labour-market situation, legislation, cultural norms and welfare provision all have an effect. Furthermore, the role of financial sources varies depending on the historical phase in which the society finds itself. For example, whereas adult children have previously supported their aging parents, the trend nowadays in many European countries is for (grand)parents to support their adult (grand)children thus reshaping family solidarity (Kohli 1999; Attias-Donfut and Arber 2000; Fritzell and Lennartsson 2005; Albertini et al.2007; Björnberg and Latta 2007).

Thus far, the Nordic welfare states have assured the main responsibility for providing financial support during periods of long-term economic difficulties (Björnberg and Latta 2007), but there is evidence that families are playing bigger role in the welfare of young adults (Majamaa 2011), and that private transfers, especially from parents, are an important source of income (see e.g., Björnberg and Latta 2007). Approximately 50% of Finnish and Swedish young adults under the age of 35 receive at least some financial support from their parents (Björnberg and Latta 2007; Haavio-Mannila et al.2009). Thus, parental ability and willingness to give such support after their children have moved away from home plays a major role.

The reasons behind this generational solidarity vary, although altruism amongst family members, and especially of parents towards their children (Becker 1991), social obligations, expectations of reciprocity between generations (Coleman 1990), as well as affection and need (Fingerman et al.2009; Fingerman et al.2010) are all key factors. When children progress to the family-building phase, parents show one type of family solidarity in providing help. This descending familialism (see Dykstra and Fokkema 2010) reflects both the needs of the young adults who are becoming independent and the eagerness of parents to give help. Whatever the motivation is, the best outcome is that private help and financial transfers support and nurture social and emotional relations (see e.g., Björnberg and Latta 2007).

It has been found that married or cohabiting couples are better able to give financial support to offspring (Lennartsson et al. 2010). However, children living alone are more likely to receive such support than those who are living with a partner (Attias-Donfut and Wolff 2000b; Fritzell and Lennartsson 2005). Age also has an effect: the younger the child is, the more likely he or she is to receive support (see e.g., Fritzell and Lennartsson 2005; Haavio-Mannila et al.2009).

From a gender perspective, financial transfers seem to equalise gender differences in wealth in that intergenerational transfers tend to benefit women more than men (Fritzell and Lennartsson 2005). It has been suggested that reciprocity and expectations of future care-giving motivate the giving of support (Rossi and Rossi 1990). Research has also shown that resources are limited, and are often unevenly distributed among family members and across generations. For example, children in larger families receive less support, on average, than children in smaller families (Fingerman et al.2010; see also e.g., Attias-Donfut and Wolff 2000b; Fritzell and Lennartsson 2005).

Previous studies have also shown a strong association between socio-economic status and giving: parents who give financial support tend to belong to the higher social strata (Fritzell and Lennartsson 2005). Having a higher level of education and belonging to a higher income group are key indicators of financial generosity by parents to their own adult children (Kohli 1999; Björnberg and Latta 2007). The association between socio-economic position and the receiving of financial support, however, seems to go two ways, and merits closer attention (Pitrou 1992 as cited in Attias-Donfut and Wolff 2000b: 64). On the one hand, parental support helps to prevent poverty among young adults, and is dispensed to those in need (Fingerman et al.2009). Björnberg and Latta (2007) report that parents tend to give more financial transfers to young adults on a low income, and unemployed young adults and students tend to receive more financial support from their parents (Kohli 1999). Thus private transfers are complementing welfare-state assistance, at least to some extent (Attias-Donfut and Wolff 2000b). On the other hand, some of the research findings indicate that private support is orientated to advantaged young adults who have already established their position in adult life. Receiving such help is common among those in a better economic position (Fritzell and Lennartsson 2005; Lennartsson et al.2010).

Although Björnberg and Latta (2007) note with regard to Sweden that certain risk positions, such as receiving social assistance, unemployment or sickness benefit, do not increase the likelihood of receiving financial support from one's family, Kohli (1999) found that unemployed young adults and students in particular were likely to receive parental support. Tentative research results from Finland also suggest that recipients of benefits, and especially students, are more likely to be supported by their parents than in the 1990s. It seems that parents have stepped in to hold up the crumpling pillars of the market and welfare state, thereby engendering a compensatory mechanism for many economically vulnerable young adults (Majamaa 2011).

The context of our study was the income development mentioned above. People aged between 55 and 64 are likely to be in a good financial position, whereas younger age groups often fare worse, at least on the aggregate level (see Figure 1). The relative difference has grown over time, and one might expect that almost all parents are able to give at least some financial support to their adult children. Even in small amounts this carries the positive message that a somewhat stronger safety net is available. In order to assess the prevalence of given and received support we investigate, first, to whom Finnish baby boomers give financial transfers, and secondly, from whom the adult children of Finnish baby boomers receive transfers. Previous studies related to intergenerational support have not, however, factored in smaller contributions, and the main aim of this study is thus to examine whether there is a connection between the socio-economic characteristics of those who give support and these smaller amounts of support. We also consider support from the receiver's perspective, and specifically investigate whether the higher propensity among adult children who are in need (e.g., receiving social security benefits, or otherwise on a low income) to receive support remains when the characteristics of the child and the parent's socio-economic position are controlled for.

This study is based on information obtained from questionnaires distributed to families across two generations. The first generation comprised the so-called Finnish baby boomers, and their adult children made up the second, younger generation. The first sample consisted of 1998 randomly selected Finns born between 1945 and 1950, and the second sample comprised 3391 of their adult children born between 1962 and 1988. Statistics Finland drew up both data sets in 2007 and, with the permission of the respondents, merged some individual-level administrative register data into the survey data, such us information about employment, level of education and income from 2007 and from previous years. The response rate among the baby boomers was 56% (n=1115), and 42% among their adult children (n=1435).

For the present study we included baby boomers who had at least one adult child living outside the parental home (n=841), as well as adult children who had at least one living parent and who did not live in the parental home (n=1334). We reduced the latter data by a third, because we could make adult child–parent dyads only if both child and parent returned the questionnaire. By means of the family identification number we linked the information on the parents with the data on their adult children. This new dataset included 911 adult children with information about one of their parents.

Because of the low response rate we carried out non-response analyses based on data from the administrative register separately for men and women (Majamaa 2009a). According to the results, both sets of data were fairly representative. However, among the baby boomers there was a higher response rate from women in the higher income deciles (71%) than from those in the lower deciles (48%) although this tendency was not evident among the men. Moreover, those with a higher educational level responded more actively than those with only a basic level (women 69 vs. 55%, men 57 vs. 40%). Non-response bias in the children's data was related to education and socio-economic position: the higher the educational level, the higher was the response rate (women 61 vs. 37%, men 42 vs. 21%) (Majamaa 2009b). Income level did not separate the respondents from the non-respondents among the young adults. According to previous studies, financial transfers are more prevalent and larger sums are given among those with a better socio-economic position. Over-representation of the higher strata may exaggerate the extent of financial support given to young adults (Majamaa 2009a).

The questionnaire items focused on the respondent's social interaction, including social relations, and any financial transfers and practical help received and given over the previous 12 months. Alternative sources of help included the respondent's own and his or her spouse's parents, siblings, cousins, aunts, uncles, grandparents, friends and workmates/neighbours. The emphasis in this article is on the flow of private financial support between parents and their adult children.

According to previous studies (see Attias-Donfut and Wolff 2000b; Fritzell and Lennartsson 2005), substantial private transfers, such as (advance) inheritance, are more typical in higher social strata. Furthermore, large bequests are usually received in the later phases of life. The recipients of support studied here were quite homogenous in terms of age: the average age was approximately 31 years. The probability of receiving any form of inheritance is modest among the younger generation, and was not considered in this study.

The financial support given or received was a dependent variable. The questionnaire for the baby boomers included two items concerning their adult children¹ and other relatives,² friends and workmates, which would yield the relevant information. The adult children's questionnaire also included two separate items covering any financial support received from their own and their spouse's parents³ and other friends and relatives.⁴ Financial support meant the giving of money or covering specific types of costs such as for schooling or certain purchases.

Firstly, in the descriptive part of the study we considered financial support given on a more prevalent level. Figure 2 shows the total amount of financial support a parent gave to his or her adult children, as well as the support the adult child received from both parents. This makes comparison of the different groups more accurate, because the amounts of money given or received from other relatives, friends and workmates were also considered as lump sums and not as individual donations. Secondly, financial transfers were not summed in the explanatory part. In the case of parents (givers), the amount given to adult child was considered. Also for the recipients the amount received from the parent (mother or father) who was included the data was taken into account. The respondents were divided into three categories based on the amounts of given and received: those who did not give or receive any, those who gave or received at the most €500 and those who gave or received more than €500. The cut-off point of €500 reflects, firstly, the student's/unemployed person's monthly allowance in 2007,⁵ and secondly the fact that, according to the data on adult children, the median amount of parental support given was approximately €500 for those who received support.

The choice of control variables was based on previous research. Seven variables covering the socio-demographic characteristics of baby boomers were included in the analysis of the financial support given (Table 2). The first four of these were gender, living with a spouse (living alone or together with a spouse), number of children (one, two, three or more) and average age of all children (under the age of 30, at least 30 years old). The last three represent the baby boomers' socio-economic characteristics: educational level (basic or unknown, secondary and tertiary), received social securitybenefits⁶ (none, some⁷ ) and disposable income (less than €1000, €1000–1999 and €2000 or more). Information about disposable income per month was constructed based on information about taxable income minus paid taxes in 2007, and this sum was divided by 12.

The analyses of financial support received covered seven socio-demographic characteristics of the adult child and two socio-economic indicators from his/her parent (Table 3). The first two determinants were gender and age (under the age of 30, at least 30 years old). The number of siblings was divided into four groups (no siblings, and one, two, three or more). Information about living with a spouse and educational level was also included. In terms of social-security benefits received the information was based on the respondents’ own reports of what they had received in the previous 12 months.⁸ Three categories emerged; none,⁸ some⁹ and the student grant.¹⁰Disposable income was included in the analysis, as well as parental educational level and parental disposable income. Even if parents who still have (numerous) underage children living at the parental home or have an obligation to provide maintenance to underage child(ren) have fewer assets with which to support offspring who have already moved away from home, we could not control for these factors due to the limitations of the data sets.

First, we considered the prevalence of giving and receiving financial support among the baby boomers and their adult children in 2007 (Figure 2). We used multinomial logistic regression in the explanatory analyses. The response variable is dichotomous in binary logistic regression, and comparisons are made between these two categories. In the case of multinomial regression the response variable has more than two categories and comparisons are made between category×and the base category (Liao 1994). Here we made two comparisons between both the giver's and the receivers’ groups: gave at the most €500 vs. gave no financial support (base), and gave more than €500 vs. gave no financial support (base) in the former; and received at the most €500 vs. received no financial support (base), and received more than €500 vs. received no financial support (base) in the latter. The first category in each explanatory variable was the reference group with an odds ratio (OR) of 1. First, each main effect was fitted one at a time into the model (Model 1) in order to find out how each variable was associated with the giving or receiving of financial support. Then, in order to control for all the characteristics of baby boomers (Table 2) and their adult children (Table 3) under consideration, we fitted all the main effects into Model 2. In interpreting the OR one should be aware that comparisons are always made with the base category. For example, as Table 2 shows (Model 1), the odds ratio for men who gave considerable financial support (>€500) to their own adult children compared to those who did not give any (base category) is about 1.35 times as high as the same ratio for women. In other words, it seems that men are more likely than women to give substantial financial support when those who give substantial support are compared to those who do not give any support. Stata's statistical software cluster option was used to compute the standard errors because of the clustered data on adult children.

According to the basic data, Finnish baby boomers gave financial transfers mainly to their own adult children (Figure 2, left-hand side). Almost half of those who had at least one adult child gave some support, but about a third gave more than €500. Among those who gave support, the median transfer was €1000. They also supported their friends, siblings, their own parents and other relatives, but to a relatively lower degree.

An analysis of the sources of financial transfers gave a more varied picture (Figure 2, right-hand side). The most common source was the recipient's own parents: 46% of those with at least one living parent received some financial support from him or her or from both parents, about a quarter received more than €500 and the median amount was €500 among those who received support. Although the most common source of support was their own parents, a fifth of the young adults received contributions from their spouse's parent(s), and about 10% reported that they had received some economic support from their grandparent(s).

Table 1 gives a more accurate picture of the amounts of financial support given to their adult children¹¹ and received by the adult child from both parents. The mean amounts given and received were €1116 and €539, respectively, and the median amount in both groups was zero. The standard deviations among both groups were approximately three times higher than the means, however, indicating large differences in the amounts of support given and received. The amounts of both given and received financial transfers were relatively small: three-quarters of the respondents (baby boomers) gave at most (if any) €1000, and three-quarters of the adult children received less than €500 from their parent(s) (results not shown here).

When we looked more closely at the adult child–parent dyads we observed differences in recollection of the sums of money given or received. About half of the adult children (51%) gave the same information in the questionnaire as his or her parent, whereas a quarter (25%) specified a larger sum and a quarter (24%) a smaller sum. The major difference was in the amounts, however. Among those who mentioned a larger sum than the parent, the average difference was €754, whereas in the case of those who mentioned a smaller sum, it was €1408. These would appear to be the main reasons for the discrepancies in the amounts given and received shown in the Figures 2 and 3. Mandemakers and Dykstra (2008) also found that children have a tendency to under-report the help and support they receive whereas parents tend to over-report the amount given.

Assessing the distributions of financial transfers from a generational perspective gave a more precise picture: such transfers occurred most frequently and were most generous, between family generations, especially when the donor's own adult child(ren) were the recipients (Figure 2). In line with the results of two Swedish studies (Fritzell and Lennartsson 2005; Björnberg and Latta 2007), parents gave financial transfers almost solely to their offspring. For example, only 3% of baby boomers with at least one living parent gave any financial support to their own parents. However, the adult children also received financial help from other relatives and friends – especially from their spouse's parents and their own grandparents – but the amounts were quite small.

Multinomial logistic regression was used in the analysis reported below. Table 2 shows the odds ratios for giving support of at the most €500 (moderate support) or more than €500 (more substantial support), the reference being giving no support according to the chosen variables. Model 1 shows each variable introduced one at a time, and Model 2 presents the estimates adjusted for all the chosen variables.

Gender was not statistically significantly related to moderate amounts of financial support given to adult children whether or not they lived with a spouse (Table 2, Model 1). Other factors (number of children, average age of all children, educational level, social-security benefits and disposable income per month) were related to the giving of moderate support. In the case of more substantial sums (>€500), all the chosen variables except gender and number of children were associated with parental support (Table 2, Model 1). Those who lived with a spouse, whose children were younger, did not receive any social-security benefits, and who had a high socio-economic position (a higher educational level and higher disposable income) were more likely to give more substantial (>€500) support to their offspring.

When all the variables were adjusted for in Model 2, only two (the number of children and their low average age) were significant predictors of giving moderate (≤€500) financial support to adult children. Previous studies have also shown that financial support is more often given to younger children (Attias-Donfut and Wolff 2000b; Hillcoat-Nallétamby and Dharmalingam 2003). Interestingly, when all the variables were fitted into the model, neither educational level nor disposable income was associated with the giving of smaller amounts of support. These results indicate that neither educational level nor financial position affects parent's ability and willingness to give smaller sums of money to their adult children.

As expected, a high level of education and a high disposable income were statistically significant predictors of giving substantial financial support (>€500) to adult children (Table 2, Model 2). These findings are consistent with results reported in Swedish studies (Fritzell and Lennartsson 2005; Björnberg and Latta 2007) that this level of support is more common in higher social strata. Furthermore, even if those whose income consisted partly or wholly of social-security benefits tended to give less financial support to their adult children (Model 1), when all the variables were controlled for this association no longer held among those who made substantial transfers. This result reflects the fact that one's economic position later in life depends on many factors: social-security benefits are usually earnings-related, property is likely to have acquired in the form of one's own home, costs of living are low and the ability to give extensive support is not so strongly related to how current income is formed.

Furthermore, parents also tended to transfer more substantial amounts to children who were relatively young: the odds ratio was more than twice as high among those whose children were, on average, under the age of 30 than when the children were older than 30. In contrast with the results of a Swedish study (see Lennartsson et al.2010), living with a spouse did not predict considerable financial support from the baby boomers. Furthermore, it has been reported in previous studies (see e.g., Attias-Donfut and Wolff 2000b; Fritzell and Lennartsson 2005) that having a small number of children predicts the giving of financial support, whereas in this study there was no statistically significant connection when all the variables were included in the model.

In general, having a good economic position, measured on two different variables, increased the odds of giving more substantial rather than minimal financial support. This reflects the fact that parents have to have some economic resources in order to be able to give money to their adult children, especially when larger amounts are involved. Surprisingly, however, all of the socio-economic factors were associated, at least partly, with giving moderate support (Model 1). In particular, educated parents were more prepared to support their offspring than those with a lower educational level. There also seem to be other factors involved, such as the relative low age of the child.

We then shifted the focus of the analysis to the receivers of support. The adult children in our data received private transfers from several sources, but mainly from their parents (Figure 2). As mentioned earlier, the association between parental educational level and giving financial support held when larger sums were involved after all the control variables had been incorporated into the model (Table 2, Model 2). This encouraged us to add two more variables concerning the parent's socio-economic position to the analysis of adult children in order to test whether or not the children with a more highly educated and better-off parent were more likely to receive financial support from him or her.

In general, almost all the variables we used (age group, number of siblings, living with a spouse, social-security benefits, disposable income, parental educational level and disposable income) were associated with the receipt of both moderate (≤€500) and larger (>€500) amounts of money from the parent (Table 3, Model 1): the only exceptions were gender and educational level, which were not statistically significant with regard to substantial amounts of support (>€500) received from the parent (Table 3, Model 1). The lack of association between educational level and larger received financial transfers may be attributable to the fact that some of those with basic or a secondary education were, in fact, studying at vocational college or university, thus confounding the results. At least according to the data (results not shown here), almost a quarter of those with at least a secondary-level qualification in 2007 were also in a receipt of a student grant. The number of children was associated with giving moderate support, but now the association was negative: a low number of siblings was associated with a greater propensity to receive both smaller and larger amounts of money from one's own parent (cf. the findings for the parents).

Bringing all the main effects into the model together changed the findings somewhat (Model 2). Neither gender nor living with a spouse was now associated with receiving financial support. However, adult children who were younger and had no or only a small number of siblings were more likely to receive either moderate or larger amounts of support. These results are congruent with previous findings (see e.g., Attias-Donfut and Wolff 2000a; Fritzell and Lennartsson 2005).

Moreover, being in receipt of some social-security benefits, usually in the form of the student grant or other allowances was, at least partially, a highly significant predictor of receiving additional private support. This result is consistent with the findings of a study conducted in France (Attias-Donfut and Wolff 2000b), whereas it was found in a Swedish study (Björnberg and Latta 2007) that being the recipient of unemployment benefit, sickness benefit or social assistance had only little relevance to receiving private financial support. Having a lower disposable income was associated with receiving less financial support (Table 3, Model 2). The association between disposable income and receiving more substantial financial support disappeared when social-security benefits were included, which is probably attributable to the fact that a low disposable income mainly comprises such benefits. Finally, consistent with the results concerning their parents, there was a higher propensity among the children to receive financial support from a parent with a high disposable income, even when the child's socio-demographic characteristics were controlled for. This applied only to smaller (≤€500) amounts, however. Parental educational level was not statistically significant when all the variables were included. Even if high level of education also predicted more substantial given and received financial support among parents with a low disposable income (results not shown), it seems that the association between parental educational level and financial support depends, to some extent, on the parent's disposable income, and to some extent that highly educated parents' children were more often students than lower educated parents' children (results not shown).

In sum, our Finnish data gives further evidence that the transfer behaviour of parents is related both to their ability and willingness to give financial support, and to the needs of young adults. Socio-economic factors in particular were associated with the giving as well as the receiving of support. Being a recipient of public income support (student grant) increased the odds of receiving additional, especially larger, financial transfers from parents. Those with a low disposable income, however, were more likely to receive smaller amounts of money from their own parents. Among the parents, however, the negative association between receiving social-security benefits and giving financial support did not hold after all the variables were included in the model. Interestingly, there was also some interplay between socio-economic factors and giving moderate support, which has not been recorded in previous studies. Given that patterns of help among family members are more or less inherited (Rossi and Rossi 1990), and that parents with a higher socio-economic position support their children financially especially when they are studying, the extent of parental support has both direct and indirect effects on the life chances of adult children. In valuing and supporting their children's studies financially, parents also contribute to their educational capital.

This article analyses the flows of private financial transfers to adult children from both the givers’ and the receivers’ points of view. The Finnish baby boomers under study were highly likely to give financial support to their adult children, and these children were highly likely to receive support from their parents, but also from their spouse's parents. Our findings also suggest that the link between parents' socio-economic position and their financial backing is stronger than previous studies have indicated, whereas all the socio-economic characteristics of baby boomers we considered were, at least partly, associated with the support given even when moderate support was categorised as separately (Table 2, Model 1). It seems, however, that the associations between the socio-economic characteristics of the baby boomers and the giving of financial transfers were stronger among those who gave larger sums than among those who gave smaller sums.

Historical time, place and social circumstances are related to the forms and directions of intergenerational support. During the last few decades the main sources of economic welfare among young adults have included the labour market and the welfare state, and parental support has been mostly complementary, at least in the Nordic countries. According to a study on the financial circumstances of Finnish students (Berndtson 2004), the two most important sources of income are public support (study grant and housing supplement) and working during the holidays, whereas financial support from parents is the third most important source. As students have an especially high poverty risk (Moisio 2008: 263), they probably receive more financial support from their parents than working young adults, for example. Berndtson's results (2004) suggest that the role of the family has been mostly complementary. Furthermore, although Hämäläinen (2004): 46) notes that students comprise a very heterogeneous group, the student grant is typically a major source of income for younger students, whereas older students tend to rely more on earned income.

Young adults, in particular are experiencing a more negative trend in disposable income than other age groups (Figure 1): they are entering the labour market later, and there is less job security. Furthermore, the welfare state's role as the provider of financial security has weakened considerably. According to the results of previous research, parental support of their adult children has become more prevalent during the last 10 years. One reason seems to be the diminishing public support and another could be that middle-aged parents nowadays have more assets (see Figure 1). Our results reveal that younger adult children, here those under the age of 30, were more likely to receive private transfers from their own parents than older children. Moreover, those with a low disposable income, especially students, were more likely to be supported by their parents, even if the proportion of people receiving a student grant was less than half (15%) of the proportion of people in receipt of other social-security benefits (36%).

Bourdieu and Passeron (1990) observe that education plays an important role in the process of reproducing the existing social system. In supporting education parents may hope to be making a sensible investment in the future of their children (Becker 1991: 369). Attias-Donfut and Wolff (2000a) argue that in financially supporting their adult children parents are trying to prevent them from slipping down the social ladder. At least in this study, the parent with a higher socio-economic status was still found to support his or her adult children when the analysis was conducted from the child's perspective. Furthermore, young adults who were studying in particular received extra help from their own parents. It was reported in a previous Swedish study (Lennartsson et al.2010) that the needs of children apparently did not play a significant role in the likelihood of receiving financial support. From the two-generation approach taken here, however, it seems that need is the bottom line: a young age, a low disposable income, and being in receipt of social-security benefits was strongly associated with receiving support. It seems that many young adults who do not have sufficient means via the welfare state have parents who can and are willing to help, at least to some extent.

Educational egalitarianism has been highly valued in Finland, and in practice almost all young adults have been able to continue their education if they wished (Antikainen 2006; Kivinen et al.2007). For example, in 2009, 94% of Finnish youth continued on to the intermediate level. Thus far the educational inheritance has not been as evident in the Nordic countries as in many other European countries (Hertz et al. 2007). In fact, inequality of educational opportunity depending on family background has declined in recent decades (Breen et al.2010), in the Nordic countries, too (Jæger 2007; Kivinen et al.2007). Nevertheless, the difference in enrolment in university education is still eight-fold (measured in odds ratios) in favour of children from academic families (Kivinen et al. 2007).

State intervention in the provision of education has been at a high level in Finland as the importance of post-compulsory education has increased in the post-industrial society. If the educational level of parents is a strong determinant of their children's educational level – in other words the children of more highly educated parents tend to be more highly educated than the children of parents with a lower level of education (Kivinen et al. 2007; Myrskylä 2009) – and, as our results show if there is greater tendency among parents with a higher socio-economic status to give financial support to their adult children, and probably larger sums, it could be argued that parental support increases the inequality of opportunity between the higher and lower social strata. In sum, even if parental support fills some of the gap the withdrawal of welfare-state funding has opened up, policymakers should be aware that not all young adults have supportive parents who are both willing and able to help them financially. Furthermore, even if young adults are already on an unequal footing in terms of receiving parental financial support, and given the deterioration in welfare-state provision, the overall situation of young adults would be even weaker without support from their families.

Finland has been a country with relatively even income distribution. However, after the 1990s recession the gap started to widen (OECD 2010). Andersen et al. (2008) suggest that one of the major challenges facing Finland is to maintain a fair distribution of resources and opportunities. Given that one of the targets of public spending in the Nordic welfare states has been investment in human capital, including the provision of child care and education (ibid.), in order to maintain a positive economic and social climate, it is vital – now and in the future – to invest in education and to maintain equality in educational opportunities.

Financial support was provided by the Academy of Finland. This paper is based on GENTRANS data (http://blogs.helsinki.fi/gentrans/). GENTRANS data collection in 2007 was primarily funded by the Academy of Finland (project numbers 29885301, 2006–2009).

Karoliina Majamaa is PhD candidate of Social Policy at University of Helsinki in Finland. Her current research interest includes intergenerational help, in particular helping patterns among parents and their adult children, and young adults' economic position in the contemporary society.