Empirica

, Volume 40, Issue 3, pp 483–504

Family background and the decision to provide for old age: a siblings approach

Authors

    • Munich Center for the Economics of AgingMax-Planck-Institute for Social Law and Social Policy
Original Paper

DOI: 10.1007/s10663-013-9212-4

Cite this article as:
Lamla, B. Empirica (2013) 40: 483. doi:10.1007/s10663-013-9212-4

Abstract

The Riester pensions in Germany provide helpful evidence to better understand the determinants of and the barriers to the demand for old-age provision products. The paper argues that families are of key importance in the decision making process to buy such a private pension. Families do not only shape the way we make our financial decisions they can also be a source for cost-effective and reliable information. Depending on certain characteristics some individuals can process this information more easily. Results confirm that individual characteristics, in particular income and education, as well as family characteristics are correlated with Riester ownership. Adding a dynamic element to the analysis I find strong sequential correlations in Riester ownership between siblings. However, these correlations become weaker over time as the number of Riester owners in other social circles grows. Once a critical mass has been reached, positive spillovers can create a social multiplier leading to a higher coverage with private pensions in the future.

Keywords

Old age provisionFamily backgroundInformation sharingTransaction costsRiester pension

JEL Classification

D83D91

1 Introduction

Almost all western countries face the challenges of an aging population. To support the elderly with decent living standards without imposing a crushing burden on the young, far reaching economic reforms have been introduced in the pension systems (as well as the health care system and the labor market). These reforms have in common that they shift income risks from the state to the individual and therefore require individuals to make their own provisions.

In Germany, although saving rates are on average quite high by international standards, there is a rising concern on the adequacy of savings after deep social reforms in 2001 and 2004.1 Compared to a situation without reforms, the public pension level will be lower by 14.4 % in 2030 (Börsch-Supan and Gasche 2010). With the aim of promoting the take-up of supplementary pensions to fill the emerging pension gap Riester pensions were introduced in 2001 (Riester reform). Riester pensions are state-subsidized private saving plans with an (largely) annuitized payout plan.2 Several authors have analyzed which social groups reacted more promptly to the reform, buying one of these private pension plans. This follows research in the US which shows that participation rates in similar plans are very heterogeneous, especially by income classes (see e.g. Venti and Wise 1988, 1990; Feenberg and Skinner 1989; Gale and Scholz 1994). Germany appears to be no exception: In general, individuals with low-income and low levels of educational attainment are less likely to make use of Riester subsidies (e.g. Coppola and Reil-Held 2009; Börsch-Supan et al. 2008; 2012; Pfarr and Schneider 2011).

In 2001 1.4 Mio persons owned a Riester pension. However, after the initial enthusiasm, demand for Riester pensions flattened in 2003 and 2004 (see Fig. 1 in the “Appendix”). These disappointing initial results led to a lively discussion about possible modifications to the law (see e.g. Fehr et al. 2003), some of them were implemented in another reform in 2005.3 Administrative data show that after 2005 the number of new subscriptions increased very steeply. While some socio-demographic groups may have reached saturation levels, there is a continuous increase in the uptake rate of Riester pensions among low income households (see Fig. 2 in the “Appendix”). There are several explanations for this development: In addition to the simplifications in 2005 social learning appears to be one of them (Börsch-Supan et al. 2008, 2012). While other forms of long-term saving plans, such as life insurances, are well-known by the population Riester pensions are relatively new and their financial advantages might be hard to grasp (see chapter 2). Several studies have examined how the design of retirement programs could improve the decision making process by emphasizing the role of information (e.g. Madrian and Shea 2001; Iyengar et al. 2004; Lusardi 1999). As more and more people learn about eligibility criteria and subsidies this might result in a social multiplier effect and lead to a better coverage in the future (Glaeser et al. 2003).
https://static-content.springer.com/image/art%3A10.1007%2Fs10663-013-9212-4/MediaObjects/10663_2013_9212_Fig1_HTML.gif
Fig. 1

Development of Riester pensions over time. Source Bundesministerium für Arbeit und Soziales (2012)

https://static-content.springer.com/image/art%3A10.1007%2Fs10663-013-9212-4/MediaObjects/10663_2013_9212_Fig2_HTML.gif
Fig. 2

Uptake of Riester pensions by quintiles of monthly household disposable income. Source Börsch-Supan et al. (2012)

This paper adds to the discussion on why individuals respond differently to the subsidies by stressing the role of families. Parental erudition as well as experience in financial matters and attitude towards saving are strong determinants of children’s preferences and ability in financial decision making (Lusardi et al. 2010; Ashby et al. 2011). Beyond that families can be a source for cost-effective and reliable information. Given the public debate on the intransparency of the cost structure of Riester contracts (Hagen and Reisch 2010) reliable information from a trusted person might be of particular importance in overcoming possible barriers to entry. More specifically, the contribution exploits information on family background in order to reduce omitted variable bias in two ways: (1) by subtracting a family fixed effect and, (2) through the inclusion of proxy variables. Furthermore, the paper provides circumstantial evidence for information sharing by considering the sequence of Riester market entrance between siblings.

The remainder of the paper is organized as follows. Section 2 presents some key features of Riester pensions and contains an overview of the relevant literature regarding the importance of families in financial decision making. Section 3 describes the identification strategy including the data and sample used as well as the model specifications. Estimation results are reported in Sects. 4, 5 concludes.

2 Related literature and hypotheses

2.1 Riester pensions: the key features

The decision regarding retirement saving is complex: Individuals are confronted with the complicated task to figure out how much to save and which financial instruments to choose. In contrast to the traditional economic perspective this task is not realized easily. Individuals incur cost of information processing (when making the original decision) and self-control (when adhering to it). These cost lead to boundedly rational behavior as described by a range of models such as prospect theory (Kahneman and Tversky 1979), mental accounting and the behavioral life-cycle hypothesis (Thaler 1990; Shefrin and Thaler 1988), and quasi-hyperbolic discounting (Laibson 1997; Laibson et al. 1998). The design of Riester products can add to this complexity.

To start with, subsidies are bound to eligibility criteria. Basically everyone who is affected by the decreasing statuary pensions is eligible for subsidies, yet the concrete eligibility rules are complicated (Börsch-Supan et al. 2012). A distinction is made between direct and indirect eligibility. Directly eligible are employees paying mandatory contributions to social insurance, unemployed and recipients of other wage compensation benefits, self-employed, farmers as well as civil servants. Indirect eligibility is derived from eligibility of the spouse. Coppola and Gasche (2011) demonstrate that especially low-income households are ignorant of their eligibility for subsidies under the Riester scheme. The authors find that low knowledge of the pension system is correlated with a higher probability to misreport households’ eligibility for the Riester-subsidies.

For certified Riester products subsidies exist in the form of a basic benefit matching the own contribution and a tax deduction, depending on the amount contributed to the contract and the marginal tax rate of the owner of the contract. Moreover, there is an additional subsidy for each child. Table 1 provides an overview of the state subsidies for Riester products as applicable for 2008 onwards.
Table 1

State subsidies for Riester products (as of 2008)

Minimum percentage of income required to be saved to obtain full subsidies

4

Minimum own contribution in Euro per year

60

Per capita subsidy in Euro per year

154

Subsidies for children in Euro per year:

 

 Children born before 01.01.2008

185

 Children born 01.01.2008 and after

300

One-time bonus if the subsidized individual is younger than 25 years in Euro

200

Maximum tax deductible amount in Euros per year

2,100

Source Bucher-Koenen (2011)

The financial advantages of Riester pension plans are not immediately obvious to everyone and its quantification requires some mathematical skills (Börsch-Supan et al. 2012). Low income individuals receive a relatively high subsidy due to the matching basic benefit, higher income individuals profit from the tax deductions (Börsch-Supan and Gasche 2010). Overall, the subsidies average about 45 % of contributions, depending on income and number of children (Fig. 3). While subsidies are particularly generous for low income individuals, targeting individuals with high risk of old age poverty, there is widespread misperception of the generosity of the state-subsidy in this group (Coppola and Gasche 2011). At the same time, only those who are aware of incentives can respond to them (Chan and Stevens 2008).
https://static-content.springer.com/image/art%3A10.1007%2Fs10663-013-9212-4/MediaObjects/10663_2013_9212_Fig3_HTML.gif
Fig. 3

Subsidy as percentage of total contribution. Source Deutsche Bundesbank (2002)

Riester pensions might not be advantageous for everyone. As they were only introduced in 2001 some individuals might already own other un-subsidized pension products. Others might be able to derive higher returns from investing in e.g. stocks. Moreover, social welfare in the old age (“Grundsicherung im Alter”) is means-tested with Riester pensions being fully accountable. If an individual expects to rely on such a program he will not benefit from his retirement savings (see Gasche and Lamla 2012). Then again, if an individual chooses to buy a Riester pension he might find it difficult to decide which product to pick. Saving options are diverse with more than 5,000 different certified products. The decision might be further impeded by the ongoing public debate about the intransparency of costs related to the contracts. Hagen and Reisch (2010, p. 5) point out the nature of Riester contracts as “trust-goods”, with consumers being unable to evaluate the value of the contract even after the purchase. Ziegelmeyer and Nick (2012) analyze the reasons behind the termination of Riester-contracts, finding that in about one-third of the cases miscounseling or bad products were the only cause for terminating or stop paying contributions in Riester contracts, thus further stressing the need for a more transparent market and a simplification in the design of Riester contracts.

To sum up, an individual has to manage a series of tasks until he can make the decision to buy a Riester pension: First, one needs to identify the need to provide privately for old age. Second, to make an informed decision the individual needs to gather information on other relevant pension products as well as Riester-specific information on eligibility and subsidies, which can be hard to grasp. Third, the individual has to decide whether a Riester pension suits best to his personal circumstances. Finally, the individual has to pick one of the many available contracts on the market. The next section argues why families might be of key importance in this complex decision making process.

2.2 The role of families in financial decision making

Families shape the way we handle investment decisions to a large extent. Lusardi et al. (2010) examine financial literacy among the young and find that financial literacy is strongly related to socio-demographics and mothers’ education, which is—among other variables- interpreted as a proxy for family financial sophistication. Mothers’ educational attainment also proves to be an important determinant for thinking about retirement (Lusardi et al. 2010; Lusardi 2003). These studies, however, lack sufficient information on fathers’ educational attainment. Indeed, Loehlin (2005) finds that across studies the correlations in different characteristics between mothers and children are on average higher than the correlation between fathers and their children. However, the still prevalent male breadwinner model in Germany might make fathers the financial decision maker of the household and, therefore role models in investment decisions. The fact that men are usually found to be financially more literate than women, might reinforce their position (Lusardi and Mitchell 2011).

Using a sample of identical and fraternal twins Barnea et al. (2010) decompose variance in investment decisions. The authors claim that similarities in investment behavior are to some degree due to genetic predisposition, even after controlling for a wide range of covariates and the frequency of interaction.

Beyond predisposition and preferences social learning appears to be necessary to improve financial decision making. Lusardi (2003) uses a sample restricted to 50–61 year old Americans. She reports that individuals learn to plan for retirement from older siblings and from the experience of old parents. In her descriptive analysis she shows that respondents who do not think about retirement are less likely to have older siblings that could provide advice on preparation for retirement. Moreover, she uses the age difference to the oldest sibling as an instrument for planning in a savings regression. The author claims that this should capture “search and psychological costs of planning” (Lusardi 2003, p. 8). Many studies make peers the main source of financial advice and contributors to financial decisions (e.g. Brown et al. 2008). Hong et al. (2004) develop a theoretical model with two types of investors, non-social and social ones. Non-social and social investors face fixed participation costs when entering a market, but for the latter group these costs decrease when the participation rate among peers is higher. The model predicts a social multiplier effect due to positive externalities. Empirically the authors show that sociable households who interact more with neighbors or attend church are more likely to possess stocks. Similar results are reported in Guiso et al. (2004) who find that households living in high social capital areas, measured as participation rates in elections and blood donation, tend to invest more in stocks. Especially in low social capital areas, narrow subgroups are considered a valuable source for information (see Guiso et al. 2004 and references therein). Given the public debate on the intransparency of the cost structure of Riester contracts reliable information form a trusted person, such as a family member, might be of particular importance in overcoming possible barriers to entry.

The main identification problem in studies on social interaction is that households are not randomly assigned to peer groups: Sorting and matching is endogenous (Li 2009). Manski (1993) formally demonstrates the considerations involved in identifying peer effects: First, group members share a common social environment. Second, people with similar preferences are members of the same group. In order to overcome the endogeneity problem, Duflo and Saez (2003) analyze a randomized experiment. They study the role of peer effects in TDA (Tax Deferred Accounts) plan participation using data on employees at a university. The experiment encouraged randomly selected employees of certain departments to attend an information fair which promised rewards for attendance. The authors show that the effect on enrollment rates in TDA plans was similar between treated and untreated departments a few months after the fair, which they take as evidence for social networks dispersing information. Without a credible instrument or experiment at hand, an alternative solution to overcome the endogeneity problem is proposed by Li (2009). The author looks at the sequence of stock market entrance of family members. He finds that the likelihood of entering the stock market within the next 5 years is higher if the respondents’ parents or children had entered the stock market during the previous 5 years.

Following this approach I expect the likelihood of a family member entering the Riester market to be higher if someone within the family has bought a Riester contract in the previous period, other things held constant. If positive externalities exist that might create a social multiplier once a critical mass has been reached (Becker and Murphy 2000; Glaeser et al. 2003). The family as a source of information should become less important as soon as the group of Riester owners in the population is large enough and information on eligibility and subsidies is widely dispersed.

3 Identification strategy

3.1 Data and sample

The Socio-Economic Panel Study (SOEP) is an annually conducted, representative household panel study starting in 1984.4 The SOEP provides information on all household members who reach the age of 17. Its structure is very similar to the American Panel Study of Income Dynamics (PSID) with an individual questionnaire for all household members containing questions about e.g. education, occupation and earnings and a complementing biography questionnaire which covers information on the life course (e.g. marital history, social background, employment biography etc.). In addition, there is a questionnaire answered by only one person with questions on the situation of the household as a unit. A question on Riester ownership was part of the individual questionnaire in the years 2004, 2006, 2007 and 2010. The wording is as follows: “Did you subscribe to a Riester contract since 31st December 2001?”. Unfortunately the SOEP does not provide information on how much individuals save in their contract and thus, whether they qualify for the full subsidy or, whether they pay contributions at all.5 Moreover, I have only very limited knowledge on the portfolio of the individuals as detailed information on the financial balance sheet of the household is collected in 5-year intervals. Thus the data is not suited to investigate the optimality of the Riester decision and the research question is restricted to asking whether there is a correlation between family background and Riester ownership.

SOEP has wide tracing rules, namely to follow everyone living in an original sample household.6 Individuals leaving original sample households form new households, consisting of grown children and separated spouses. These newly formed households are added to the SOEP population, including all non-original sample household members. Due to this feature I am able to construct a sample consisting of siblings, defined as persons having the same mother and matched using her never-changing identification number. Information on the father is associated with each sibling also using his identification number. In principle, even richer family relationships could be established by matching grandparents and own children. This, however, would result in a rather small sample size. In summary individuals who have at least one sibling in the SOEP and whose mother is already part of the SOEP population are considered in the sample.7 As pointed out in chapter 2, in order to be eligible for subsidies certain criteria have to be fulfilled.8 Because all persons in a household answer the individual questionnaire, I am also able to account for eligibility of the spouse which in turn leads to indirect eligibility for the individual under observation. The analysis is restricted to individuals with German nationality as foreigners might have a limited opportunity to learn from their relatives on public and private old age provision in Germany. For the first part of my analysis wave 2010 is used, that is the latest publically available wave. Later on, when looking at sequential correlations I exploit the longitudinal character of SOEP and use waves 2004, 2006, 2007 and 2010. The final sample size accounts to 1228 siblings in year 2010.

While the special tracing rules in combination with the length of the panel allow the construction of such a complex sample, attrition is a potential weakness.9 If only certain siblings remain part of the SOEP after moving out from their parents, the sample will suffer from sample selection bias. Because the sample requires the observation of at least two siblings as adults this adds to the scope of the attrition problem (Fitzgerald 2011). Schonlau et al. (2010) show that even such wide tracing rules as in the SOEP do not exceed attrition and hence, do not result in an “ever expanding panel”. Spiess et al. (2008) compare sample members and their descendants. Favourably, the author finds no evidence that one group is more volatile in participation behaviour than the other. Fitzgerald et al. (1998) show that intergenerational correlations in earnings, education and welfare participation are slightly stronger for a subsample of children who did not drop out from the PSID and whose parents were already part of the PSID sample. If family ties are especially strong in samples of second generation respondents and the finding can be conferred to the SOEP this would imply that the inter-generational as well as intra-generational correlations found in my results should be understood as an upper bound.

Table 2 compares the coverage rate with Riester pensions in the Siblings Sample with data from the German SAVE study.10 In comparison to the overall eligible population in the SAVE sample a lower share of individuals own a Riester contract in the Siblings Sample. This can be partially ascribed to the low average age (Table 3): Schunk (2007) finds that saving for the old age becomes more important for older age groups as saving motives change over the life cycle. Table 3 reports descriptive statistics for the Siblings Sample as well as for a sample of all individuals in the SOEP who are eligible for subsidies under the Riester scheme. Given that the individuals are rather young the influence of parents on their children should be pronounced. The average network size is 2.4 siblings with a maximum of 7 adult siblings under observation in 2010, reflecting the extent of possible social interaction that can be observed in the data. In comparison to the overall SOEP population the parents of the siblings sample are on the average better educated.11
Table 2

Coverage rate of eligible population with Riester pensions

 

2004

2005

2006

2007

2008

2009

2010

SOEP sibling sample

9.18

n.a

15.57

21.07

n.a

n.a

32.70

SAVE

n.a

n.a

n.a

26.23

31.37

33.33

37.71

Source Own calculations, n.a = not available

Table 3

Descriptive statistics for Siblings sample and the eligible SOEP population

 

Siblings sample

Eligible population

Male

0.53 (0.50)

0.47 (0.50)

Age

29.77 (8.19)

44.00 (12.04)

Lives in east Germany

0.23 (0.42)

0.25 (0.43)

No degree/low secondary education

0.22 (0.42)

0.28 (0.45)

Medium secondary education

0.34 (0.47)

0.37 (0.48)

High secondary education

0.33 (0.47)

0.31 (0.46)

Post-secondary education

0.22 (0.42)

0.28 (0.45)

Married

0.29 (0.45)

0.65 (0.48)

Children in HH

0.32 (0.47)

0.34 (0.47)

Income

3,121.38 (2,141.67)

3,203.62 (1,990.91)

Mother: no degree/low secondary education

0.45 (0.50)

0.68 (0.47)

Mother: medium secondary education

0.38 (0.48)

0.22 (0.41)

Mother: high secondary education

0.11 (0.31)

0.07 (0.26)

Father: no degree/low secondary education

0.50 (0.50)

0.67 (0.47)

Father: medium secondary education

0.28 (0.45)

0.17 (0.38)

Father: high secondary education

0.16 (0.36)

0.13 (0.33)

Mother: post-secondary education

0.78 (0.42)

0.62 (0.48)

Father: post-secondary education

0.87 (0.33)

0.81 (0.39)

Parents own life insurance

0.54 (0.50)

 

Own Riester

0.33 (0.47)

0.29 (0.45)

Sibling(s) own(s) Riester

0.31 (0.46)

 

No. siblings

2.39 (0.74)

 

N

1,228

10,433

Source SOEP 2010. Own calculations. Standard errors in bracket

3.2 Model specifications

Family background should be strongly correlated with the decision to buy a Riester pension and therefore needs to be considered in an analysis of the determinants of Riester ownership. In order to reduce omitted variable bias, I (1) estimate a family-fixed effects model and, (2) include proxy variables for family background. To allow comparison of results across specification- including the family-fixed effects model- the models are estimated using Ordinary Least Squares.12 Both proposed strategies have their advantages and allow for different interpretations:

The family-fixed effects model exploits the idea that—at least part of—the unobserved heterogeneity is common to members of one family. Under this assumption the difference in unobserved characteristics should be lower within than between families. Index s identifies a sibling while f denotes the family (Eq. 1). The error term is split into two components, αf and εsf. Explanatory variables captured under Xsf are assumed to be correlated with the family specific component αf, but not with the idiosyncratic error εsf. The idiosyncratic error needs to be strictly exogenous after taking out αf. This assumption must hold for all regressors included (Wooldridge 2002). The reference period is 2010, the latest available wave of SOEP. The fixed effects transformation will eliminate all effects which do not vary within the same family by subtracting the family averages (Eq. 2).13 By taking differences measurement errors are amplified which might lead to attenuation bias (Grilliches 1977). The bias resulting from comparison across siblings is nevertheless lower than comparing individuals across time. Moreover, standard errors are large due to the large number of parameters that have to be estimated (Schnabel and Schnabel 2002, p.9). Nevertheless, the family-fixed effects model is valuable: It reflects pure individual decisions to buy a Riester contract as opposed to coordinated family decisions by disentangling genetic influences and shared preferences from individual characteristics.
$$ Riester_{sf} = \beta_{0} + X_{sf} \delta + \alpha_{f} + \varepsilon_{sf} $$
(1)
$$ Riester_{sf} - \overline{{Riester_{s} }} = (X_{sf} - \overline{{X_{s} }} )*\delta + \varepsilon_{sf} - \overline{\varepsilon }_{s} $$
(2)

In the model above the family-fixed effect is treated as a nuisance parameter (Durlauf and Ioannides 2010, p. 465). Yet, it is interesting to study the influence of the family in a more direct way. In an attempt to partially capture unobserved components in the error term I start with a baseline model (which includes only a dummy indicating the gender and whether the household is located in East Germany, age as well as age squared, the educational attainment of the respondent, whether he/she is married, has children under 16 years living in the household and household net income divided into quintiles) and nest it in gradually richer models, which are then compared testing the hypothesis that the coefficients on the additional covariates are jointly zero using a Wald-test. More specifically, in the extended models I control for the educational attainment of the mother and the father separately. As pointed out in chapter 2, whose characteristics are more strongly correlated with their children’s financial decisions remains a priori unclear. Therefore, I add a variable which indicates if the parents own a life insurance contract. Specific information on eligibility, subsidies and on the design of a particular Riester product should lower entry costs even further. Therefore, a dummy variable is included which indicates whether at least one sibling owns a Riester contract.14 In addition, I control for the size of the network using the number of siblings observed in the sample.

The added variables are endogenous to the model. In order to partially overcome the identification problem, I focus on sequential correlations. The underlying idea is that in t = 0 no family member owns a Riester contract. For some exogenous reasons the first sibling buys a Riester contract in the period between t = 0 and t = 1.15 During the next period information sharing takes place, lowering the entry barrier for the other siblings. In t = 2 the next sibling has bought a Riester contract until in infinity all siblings own such a private pension plan (Fig. 4). This approach obviously assumes that everyone can benefit from a Riester contract. As pointed out above, this might not be the case but constitutes a necessary assumption for the analysis. A correlation in Riester ownership between siblings, however, could also result from a sub-optimal decision. For instance, to mimic the behavior of siblings could be considered a straight forward form of financial decision making similar to a rule of thumb (see Winter et al. 2012). If sequential correlations are supposed to be only due to information sharing the identification assumption requires that the factors influencing the decision of one sibling between t and t + 1 are not correlated with the decision of the other siblings made after t + 1 (Li 2009, p. 8). As I am not willing to make this assumption shared preferences still play a role.
https://static-content.springer.com/image/art%3A10.1007%2Fs10663-013-9212-4/MediaObjects/10663_2013_9212_Fig4_HTML.gif
Fig. 4

Information sharing within the family over time

The model to be estimated is a piecewise constant hazard function. A discrete-time hazard model consist of (a) a baseline risk profile over time and (b) shift parameters that capture the effect of the covariates on the baseline hazard. The higher the hazard, the greater the risk to buy a Riester contract in a given year. The main variable of interest is the one-period lag of whether a sibling has bought a Riester contract. Variation in the covariates results in a vertical shift of the entire baseline hazard function the entire baseline hazard function. Among the usual assumptions of linearity and homogeneity, a proportionality assumption must hold in a simple discrete-time hazard model: Proportionality assumes that the vertical displacement in the hazard rate per unit difference in the predictor is the same in every year (Singer and Willet 1993). However, I have reason to believe that the family as a source of information has a time-varying effect if a social multiplier exists. Therefore interaction terms with time are included in a further step. The models are estimated by complementary loglog regressions (cloglog).16

4 Results

4.1 The influence of individual and family characteristics

I estimate a series of linear probability models (LPM) explaining the probability to own a Riester contract. Table 4 displays regression results for the baseline model (column 1), the family-fixed effects model (column 2) as well a model including parental education (column 3) and whether the parents own a life insurance as proxies for family background (column 4). Column 5 extends the latter model by controlling for contemporaneous correlations in Riester ownership between siblings. Standard errors have been adjusted for heteroskedasticity, allowing for clustering at the level of the mother.
Table 4

Regression results for LPM

Dep. var: owns Riester contract

(1)

(2)

(3)

(4)

(5)

Individual characteristics

Male

−0.03 [0.03]

−0.04 [0.04]

−0.03 [0.03]

−0.03 [0.03]

−0.03 [0.03]

Age

−0.02 [0.01]

−0.01 [0.02]

−0.02 [0.01]

−0.02 [0.01]

−0.02 [0.01]

Age squared

0.00 [0.00]

0.00 [0.00]

0.00 [0.00]

0.00 [0.00]

0.00 [0.00]

Lives in east Germany

−0.02 [0.03]

0.29 [0.11]***

−0.04 [0.04]

−0.03 [0.04]

−0.01 [0.03]

No degree/low secondary education

Ref.

Ref.

Ref.

Ref.

Ref.

Medium secondary education

0.1 [0.03]***

0.09 [0.05]

0.08 [0.03]**

0.07 [0.03]**

0.07 [0.03]**

High secondary education

0.00 [0.04]

0.1 [0.08]

−0.02 [0.05]

−0.03 [0.05]

−0.02 [0.05]

Post-secondary education

0.1 [0.05]**

0.03 [0.08]

0.1 [0.05]**

0.1 [0.05]**

0.09 [0.05]*

Married

0.08 [0.04]*

0.11 [0.06]*

0.08 [0.04]**

0.09 [0.04]**

0.09 [0.04]**

Children in HH

0.04 [0.04]

0.04 [0.05]

0.05 [0.04]

0.05 [0.04]

0.05 [0.04]

Income: 1st quintile

Ref.

Ref.

Ref.

Ref.

Ref.

Income: 2nd quintile

0.11 [0.05]**

0.11 [0.07]

0.11 [0.05]**

0.11 [0.05]**

0.1 [0.05]**

Income: 3rd quintile

0.09 [0.04]**

0.13 [0.07]*

0.09 [0.04]*

0.08 [0.04]*

0.07 [0.04]*

Income: 4th quintile

0.12 [0.04]***

0.16 [0.08]**

0.11 [0.04]**

0.09 [0.04]**

0.09 [0.04]**

Income: 5th quintile

0.04 [0.04]

−0.01 [0.08]

0.03 [0.04]

0.02 [0.04]

0.01 [0.04]

Family characteristics

     

Mother: no degree/low secondary education

  

Ref.

Ref.

Ref.

Mother: medium secondary education

  

0.06 [0.04]*

0.06 [0.04]*

0.06 [0.03]*

Mother: high secondary education

  

0.05 [0.05]

0.05 [0.05]

0.05 [0.05]

Father: no degree/low secondary education

  

Ref.

Ref.

Ref.

Father: medium secondary education

  

−0.02 [0.04]

−0.03 [0.04]

−0.04 [0.04]

Father: high secondary education

  

−0.07 [0.05]

−0.07 [0.04]

−0.06 [0.04]

Mother: post-secondary education

  

0.02 [0.04]

0.01 [0.04]

0.01 [0.04]

Father: post-secondary education

  

0.11 [0.04]***

0.1 [0.04]**

0.1 [0.04]***

Parents own life insurance

   

0.08 [0.03]**

0.07 [0.03]***

Sibling(s) own(s) Riester

    

0.13 [0.04]***

No. siblings

    

−0.03 [0.02]*

Constant

0.44 [0.21]**

0.2 [0.40]

0.33 [0.22]

0.33 [0.21]

0.41 [0.21]**

R2

0.04

0.06

0.05

0.05

0.07

Wald test

 Prob > F

0.03

0.01

0.00

N

1,228

1,228

1,228

1,228

1,228

Source SOEP 2010. Cluster-robust standard errors in parentheses. * p < 0.1; ** p < 0.05; *** p < 0.01

4.1.1 Individual characteristics

Similar to Coppola and Reil-Held (2009) I do not find significant differences between men and women in the likelihood to subscribe to a Riester contract. In contrast to other findings I do not find a significant age effect. This can be partially ascribed to the lower variation in age in comparison to the overall SOEP population and the low average age (Table 3). While married individuals are significantly more likely to subscribe to a Riester pension, individuals with children are not. Only the family-fixed effect model is in line with existing literature finding that individuals in the East of Germany have a higher probability to buy Riester pensions (Coppola and Reil-Held 2009; Pfarr and Schneider 2011). The difference between models is difficult to explain and might result from the selective sample used for the analysis. One of the reasons named for the positive relationship is the lower coverage with occupational pensions in East Germany (Kriete-Dodds 2008). Moreover, evidence suggests that the precautionary saving motive is of great importance in the East (Fuchs-Schündeln 2008).

Education and income are usually found to be among the most important determinants for Riester ownership (Blank 2011). Especially individuals with a medium degree as well as respondents who have some sort of postsecondary training are more likely to buy a Riester pension. It should be noted that education might not fully capture the ability to handle financial decisions. For instance, Bucher-Koenen (2011) shows that higher levels of financial literacy are correlated with a higher probability to subscribe to a Riester contract a concept that is not measured in the data at hand. Furthermore, assuming that education is a good proxy for permanent income, individuals with low education might act fully rational in not subscribing to a Riester pension if they are certain to rely on means-tested social welfare after retirement (Gasche and Lamla 2012). Similar to the results of Barnea et al. (2010), the effect of education turns insignificant in the family-fixed effect model and decreases when adding controls for parental education. Using a sample of twins the authors find that education is relevant for stock market participation, but it is mainly the genetic factor of education that is important.

Related to this, also the effect of income varies across models. The size of the income effect is more pronounced in the family-fixed effect model with the difference being especially large for the 4th income quintile. That might imply that especially for the upper end of the income distribution much of the observed differences in uptake rates are driven by family background, while for individuals in the lower income quintiles current own income dominates the influence of family characteristics. Individuals from the 5th quintile, however, are not found to be more likely to own a Riester pension. Bucher-Koenen (2011) proposes to use income as a proxy for the size of the subsidy. Contrary to the incentive scheme, however, results show that individuals with low income are not more likely to subscribe to a Riester pension. This common result might be partially artificial as incentives depend on gross earnings with contributions to Riester contracts being deductible from taxes.17 Household net income, however, captures a different aspect, namely whether financial constraints prevent individuals from saving for retirement. Asking for the reason why individuals do not own a Riester pension, Gasche and Lamla (2012) find that a considerable share claims not to have enough money left.

4.1.2 Family characteristics

The probability of owning a Riester pension does not seem to follow the incentive scheme. As proposed by Börsch-Supan et al. (2008, p. 297), “in many cases [enrolment patterns] better fit the patterns predicted by the availability of information about the pension system”. While education and income can be considered as proxies for individual financial education (Börsch-Supan et al. 2008) parental education and whether parents own a life insurance measure family financial education. In line with existing results (e.g. Lusardi et al. 2010) parental education is found to be positively and significantly correlated with the decision to save for retirement in the form of a Riester pension. While the usual finding is that the characteristics of the mother are stronger determinants for the behavior of their children (Loehlin 2005), the fact that fathers are still the main earners in families might make them role models in financial decisions. In addition, a common finding in financial literacy literature is that men have higher financial literacy than women (Lusardi and Mitchell 2008). This might strengthen their position as the person in charge of the households’ finances. Results show that while medium secondary education of the mother is significantly correlated with Riester ownership, it is fathers’ post-secondary education. In terms of magnitude the influence of the father exceeds the influence of the mother.

Life insurances are the most wide-spread financial asset in Germany (Börsch-Supan et al. 2009). On the one hand, this variable accounts for the parents’ preferences for saving, especially as most of these people coming from older cohorts did not have to save privately to sustain adequate living standards after retirement because they could rely on high public pensions. On the other hand, this variable influences the extent to which children could learn from their parents about financial matters and thereby, how easy it is for them to overcome possible barriers to entry in the Riester market. As expected, the variable has a significant and positive effect on Riester ownership of the children.

Intra-generational correlations between siblings might be of even more importance as siblings are usually in the same stage of the life cycle and have therefore similar preferences and needs. Börsch-Supan et al. (2008) show that knowledge about the pension system is correlated with Riester ownership. While other saving plans (e.g. life insurances) are well-known by the population (Corneo et al. 2010) particular knowledge about Riester pensions might be necessary to overcome skepticism and a general lack of knowledge. The added variables in column 5 reveal that Riester ownership between siblings is significantly correlated. The size of the network is significant at the margin with an effect close to zero.

The model fit is rather unsatisfactory: Only around 7 % of the variation in Riester ownership can be explained by the full model. Savings decisions are complex and may involve psychological aspects that can neither be captured by the proxy variables included nor be explained by an effect common to all family members. For instance, Coppola and Reil-Held (2009) show that saving motives are important determinants for Riester ownership. These motives can even change over the life cycle (Schunk 2007). Coppola and Reil-Held (2009) find that magnitude and significance of the predictor variables change over their observation period. A static consideration of the determinants for Riester ownership might therefore not be appropriate.

4.2 Sequential correlations between siblings

Adding dynamic elements to the analysis Table 5 displays results after complementary loglog regressions. The model in column 1 includes only time dummies in order to identify time trends in uptake rates. Column 2 extends the previous model by adding a lagged dummy for whether a sibling has bought a Riester contract in the previous period taking into account sequential correlations. As to account for non-proportionality in the effect of the main variable of interest over time, i.e. the lagged dummy indicating whether a sibling owns a Riester contract, interaction terms with time are included in column 3.
Table 5

Regression results for hazard models

Dep. var: owns Riester contract

(1)

(2)

(3)

2004

0.10 [0.01]***

  

2006

0.17 [0.01]***

0.05 [0.04]***

0.05 [0.04]***

2007

0.24 [0.01]***

0.06 [0.05]***

0.06 [0.05]***

2010

0.40 [0.02]***

0.10 [0.08]***

0.12 [0.11]*

Sibling(s) own(s) Riester at t−1

 

1.74 [0.24]***

 

Sibling(s) own(s) Riester at t−1 * 2006

  

2.35 [0.49]***

Sibling(s) own(s) Riester at t−1 * 2007

  

2.17 [0.36]***

Sibling(s) own(s) Riester at t−1 * 2010

  

1.27 [0.20]

Further controls

No

Yes

Yes

Wald test

 Prob > F

0.00

0.00

N

5,908

3,035

3,035

Source SOEP 2004, 2006, 2007, 2010. Exponentiated coefficients. Cluster-robust standard errors in bracket. * p < 0.1, ** p < 0.05, *** p < 0.01

Further controls See Table 4, column 5

If the risk of event occurrence was independent of time the hazard function would be flat (Singer and Willet 1993). However, positive externalities in the form of information advantages should help to overcome barriers to entry in the Riester market and create a social multiplier (Becker and Murphy 2000; Glaeser et al. 2003) which in turn would result in dynamic demand for Riester contracts. Indeed, the hazard of buying a Riester contract significantly increases with time during the period between 2004 and 2010 (column 1). There is a large increase in the hazard rate between 2004 and 2006 which is probably the result of simplification reforms kicking in. After a period of initial enthusiasm, demand for Riester pensions flattened already shortly after introduction in 2003. The limited growth and public debates about the complex eligibility and subsidy structure led to a simplification of the design in 2005, aiming at an improvement in the acceptance of the new product by the eligible population as well as providers (Börsch-Supan et al.,2012). Official statistics display a dramatic increase in Riester subscriptions after the legislative changes (see Fig. 1 in the “Appendix”), confirming that the complexity of the design did constitute barriers to entry.

The paper argues that narrow sub-groups might be of special importance as long as skepticism towards Riester products is high. Provided that a sibling has bought a Riester contract in the previous period the likelihood of buying a Riester pension is significantly elevated (column 2). Once someone within the family has acquired knowledge on Riester products he will communicate this information to the rest of the family and can give advice to the others in their decision making process. As presented in Ziegelmeyer and Nick (2012) around 14 % of previous Riester-pension owners have terminated their contracts with most of them not signing a new contract after the termination. Evidently, the advice could also be not to buy a Riester contract this, however, is not captured by the analysis.

Considering interaction terms with time I find that sequential correlations in Riester ownership between siblings become weaker over time. While shortly after the introduction of Riester pensions reliable information from a trusted person, such as a sibling, was crucial in order to overcome barriers to entry, especially when taking into account that the product was new and already publically criticized, the family as a source of information becomes less important as the group of Riester owners in the other social circles grows. Once a critical mass has been reached positive spillovers will create a social multiplier (Glaeser et al. 2003). In that case public policy will have a direct effect on individuals and an indirect effect through social interaction.

5 Conclusion

The Riester pensions in Germany provide helpful evidence to better understand the determinants of and the barriers to the demand for old-age provision products. Existing evidence shows that deep subsidies alone do not appear to provide strong enough incentives (Börsch-Supan et al. 2012). This paper argues that families are of key importance in the decision making process to buy such a private pension. Families do not only shape the way we make our financial decisions they can also be a source for cost-effective and reliable information.

Apart from individual characteristics, in particular education and income, family characteristics are correlated with Riester ownership. Individuals with low financial education seem to find it difficult to make use of the subsidies. Moreover, (perceived) financial constraints prevent many people from subscribing a Riester pension (Gasche and Lamla 2012). Parental education as well as whether parents own a life insurance and thus, have saved for retirement themselves, appear to shape their children’s decision to buy a Riester contract. The paper provides circumstantial evidence for information sharing between siblings. The hazard of buying a Riester contract significantly increases with time during the period 2004 and 2010, with a large step-up after 2005 when simplifications to eligibility rules and product design were introduced. Sequential correlations in Riester ownership between siblings become weaker over time indicating that the family as a source of cost-effective reliable information becomes less important as the number of Riester owners in other social circles grows.

Models deviating from the fully rational decision making paradigm are much richer and as a consequence, much harder to identify. Endogeneity of Riester ownership of siblings, omitted variable bias, and measurement error do not allow for a causal interpretation. Moreover, attrition cannot be controlled for and might threaten the validity of my results. Without a cleanly designed experiment the identification strategy cannot isolate the pure effect coming from social interactions and hence, results need to be interpreted with caution. The data does not allow evaluating the quality of the financial decision making as more detailed information on the Riester contract itself as well as the financial balance sheet of the household, including assets owned by the spouse, would be required. The targeting success of Riester subsidies, however, also depends on the question whether subsidized saving contracts represent new savings, or if they are merely displacing other forms of savings. In Germany the issue has been explored by Corneo et al. (2009, 2010) who find that Riester contracts do not increase savings.

The paper cannot identify whether the development of Riester pensions is due to the product design, incentives or the availability of information. Nevertheless, the following lessons can be drawn: Riester pension have a complex design which is not easily understood (Börsch-Supan et al. 2008, p. 297) and –depending on certain individual and family characteristics- some individuals can process this information more easily. Positive externalities help to overcome barriers to entry in the Riester market by dispersing information on eligibility and the generosity of subsidies. An optimistic interpretation of the results is that the more time elapses since the pension reforms, the higher the coverage with private provisions. Combining my findings however draws another picture: If only certain groups, depending on their family background and their financial education, get in touch with information on and, hence engage in voluntary old age provision wealth inequality might rise in the future. While lowering the barriers to entry through further simplifications and more transparency in the Riester market is certainly a promising way, public policy needs to help people identify the need to save for retirement. Raising awareness through better information and educational programs is therefore of utmost priority.

Footnotes
1

See Börsch-Supan and Wilke (2004) and Wilke (2009) on the pension reform process in Germany.

 
2

See Becker (2004) for a discussion on voluntary versus compulsory systems.

 
3

The reform included as simplification of the application procedure for subsidies, a reduced number of certification criteria, a standardized minimum own contribution, improved transparency of products and a different cost structure. For a detailed description see Börsch-Supan et al. 2012.

 
4

See Wagner, Frick and Schupp (2007) and Haisken-DeNew and Frick (2005) for a detailed description of the SOEP.

 
5

See Ziegelmeyer and Nick (2012) for an analysis of contribution free Riester contracts.

 
6

See Schonlau et al. (2010) for a discussion on tracing rules.

 
7

Data was extracted using PanelWhiz (Haisken-DeNew and Hahn 2010).

 
8

Eligible are all individuals where at least one spouse is an employee subject to social security contributions, pays voluntary social security contributions, is a civil servant or unemployed.

 
9

See Kroh (2011) for an analysis of attrition in the SOEP.

 
10

Corneo et al. (2010) also compare both data sets. For a description of the SAVE data set, see Coppola and Lamla (2013).

 
11

Table 2 displays parental education as reported by the observed individual in order to compare the siblings sample with the overall population. The information might be prone to measurement error.

 
12

See Wooldridge (2002) for a discussion on the advantages and drawback of linear probability models. Pfarr and Schneider (2011) use a time-fixed effects logit model to estimate the determinants of Riester ownership. This however would require dropping 806 cases which lack variation in the dependent variable Riester between siblings.

 
13

Notation is analogous to Schnabel and Schnabel (2002).

 
14

I do not consider whether parents own a Riester as this would require accounting for their eligibility. However, some of the parents have never been eligible as they were already retired when the product was introduced. This would reduce the samples size even further.

 
15

Corneo et al. (2009; 2010) consider the introduction of Riester pensions as a natural experiment as by law certain groups are eligible while others are not. Apart from the introduction itself, individuals can become eligible due to e.g. marriage or when entering the labor market after graduation representing such an exogenous reason.

 
16

See Singer and Willet (1993) and Jenkins (2005).

 
17

I thank one anonymous referee for this valuable comment.

 

Acknowledgments

I thank Michela Coppola who has given me advice throughout the project. Moreover, I am grateful to Axel Börsch-Supan, Joachim Winter and Michael Ziegelmeyer for their helpful comments. I have benefited from comments coming from participants at the MEA seminar (Munich), the Annual Meeting of the Austrian Economic Association (Vienna),the German Socio-Economic Panel User Conference (Berlin) as well as the CeRP conference (Turin). Special thanks go to the two anonymous reviewers for their valuable comments and suggestions to improve the paper.

Copyright information

© Springer Science+Business Media New York 2013