Gifts, Bequests, and Social Inequality in West Germany

Leopold, Thomas; Schneider, Thorsten Article Gifts, Bequests, and Social Inequality in West Germany Schmollers Jahrbuch – Journal of Applied Social Science Studies. Zeitschrift für Wirtschafts- und Sozialwissenschaften Provided in Cooperation with: Duncker & Humblot, Berlin Suggested Citation: Leopold, Thomas; Schneider, Thorsten (2011) : Gifts, Bequests, and Social Inequality in West Germany, Schmollers Jahrbuch – Journal of Applied Social Science Studies. Zeitschrift für Wirtschafts- und Sozialwissenschaften, ISSN 1865-5742, Duncker & Humblot, Berlin, Vol. 131, Iss. 2, pp. 263-276, https://doi.org/10.3790/schm.131.2.263 This Version is available at: https://hdl.handle.net/10419/292332 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Intergenerational Transmission Gifts, Bequests, and Social Inequality in West Germany By Thomas Leopold and Thorsten Schneider* Abstract We compare large inter vivos transfers to bequests using retrospectively surveyed event history data from the German Socio-economic Panel Study (SOEP). We find the chances of receiving gifts and bequests to depend strongly on the socio-economic position of the parents and on the number of siblings. The same is true for the amount of bequests, but not for the amount of gifts. For women, both the chances of receiving a gift and the amounts transferred are considerably lower than for men. Bequests, on the other hand, are distributed equally between sons and daughters. JEL Classifications: D10, J10, C41 1. Introduction A large body of sociological literature on social stratification has been produced measuring social fluidity by intergenerational status (im)mobility and the influence of social origin on educational attainment (Breen et al., 2010; Erikson/Goldthorpe, 2002). In the field of economics, the focus has been on intergenerational income mobility (Solon, 2002). There are also a few studies on the relationship between parents’and children’s wealth, revealing strong correlations (Bowles/Gintis, 2002; Beller /Hout, 2006). The high degree of wealth reproduction might arise through educational investments, personality characteristics shared within a family, or direct wealth transfers from one generation to the next. Such direct transfers can be made either while parents and children are still alive (inter vivos transfers) or after the death of one party, usually a parent (bequests). Schmollers Jahrbuch 131 (2011), 263 –276 Duncker & Humblot, Berlin Schmollers Jahrbuch 131 (2011) 2 *The data used in this publication have been made available by the German Socio- Economic Panel Study (SOEP) at the German Institute for Economic Research (DIW Berlin). OPEN ACCESS | Licensed under | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.131.2.263 | Generated on 2023-01-16 13:36:19 There is a growing literature on bequests in Germany (e.g., Szydlik, 2004; Szydlik/Schupp, 2004). Research interest in these transfers has increased as the aging, well-off generation that grew up in the period of peace and sustained economic prosperity following World War II has set in motion a major intergenerational reallocation of wealth. Less is known, about large1inter vivos transfers (Kohli et al., 2005; Künemund et al., 2005). These gifts are of special interest as they differ from bequests in several important ways: First, a gift is always the result of a decision-making process that might be driven by altruism, reciprocity, strategic exchange motives or normative obligations. In contrast, bequests can also occur accidentally or unplanned (e.g., Modigliani, 1988). Second, gifts are highly private and subject to few regulations, whereas the division of bequests is strongly restricted by German inheritance legislation. For example, German law does not allow siblings the right to claim a share of gifts received from parents by a brother or sister.2According to the inheritance law, legal heirs can always claim 50% of their intestate share, even if the share is specified otherwise in the testator’s will. For example, if a widowed father of three daughters dies intestate, every daughter inherits one third of their father’s belongings. If the father states in his last will that everything should be devised to his mistress, every daughter has a legal right to claim one sixth of the whole estate. Gifts and bequests have very similar tax regulations in Germany. There are large tax allowances for sons and daughters (currently of 400,000 EUR per child and donor, ErbStG § 16) that treat gifts and bequests equivalently and can be fully claimed every ten years. In the case of large wealth and few descendants, a long-term combination of gifts and bequests can save taxes. The present study compares gifts and bequests, relating transfer chances and amounts to social background and family characteristics. Specifically, we address the following research questions: How common are large gifts compared to large bequests? How do social background and family structure affect transfer changes and transfer amounts? Do these relationships differ between inter vivos transfers and bequests? 264 Thomas Leopold and Thorsten Schneider Schmollers Jahrbuch 131 (2011) 2 1There is no standard definition of which amount constitutes a “large”transfer. Some surveys use explicit lower thresholds (e.g., 5,000 Dollars, Health and Retirement Study; 5,000 Euros, Survey of Health, Ageing and Retirement in Europe). The present study uses data based on a question that referred to gifts and bequests “of great value”, giving no specific lower threshold. Over 90% of the reported gifts have values above 5,500 Euros, and over 90% of the bequests exceed a value of 3,500 Euros (for details, see section 4). 2The only exception in the former German inheritance legislation concerned gifts that a child received less than 10 years before the parent’s death. These gifts were credited against the statutory share of the bequest. OPEN ACCESS | Licensed under | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.131.2.263 | Generated on 2023-01-16 13:36:19 We start by giving a brief overview on the determinants of financial transfers. Then we describe our data from the German Socio-economic Panel Study (SOEP) (Wagner et al., 2007), the dependent variables, and the methods used. The respondents reported only on transfer receipt. Therefore, we cannot analyze intergenerational transfers from the giver’s perspective, although some important information on the parents is included. In the fourth section, we discuss our empirical findings on the chances of receiving gifts and bequests as well as on transfer amounts. We conclude by summarizing our findings, discussing some limitations of our study and outlining future research perspectives. 2. Determinants of Financial Transfers between Generations A framework that covers different determinants of intergenerational solidarity has been proposed by Szydlik (2000). The functional dimension of solidarity comprises different types of intergenerational transfers, including gifts and bequests. The model distinguishes between four groups of determinants: opportunities, need, family structure, and the cultural context. Opportunities refer to the resources of parents and children, like time and money. Large gifts and bequests can only be passed on if parents have at least some assets. Accordingly, empirical findings have shown that parents with higher levels of education, occupational prestige, income, and wealth give inter vivos transfers and bequests more frequently and in larger amounts (Künemund/Motel, 2000; Kohli et al., 2005). The opportunity structure should also affect inter vivos transfers, as parents with large assets might split their giving to save taxes. For bequests, the death of one or both parents can also be considered an “opportunity”that triggers the transfer. Considering the need for intergenerational support, adult children’s own incomes and wealth should influence the receipt of transfers. From a within-fa- mily perspective, McGarry / Schoeni (1997) found that financial inter vivos transfers were compensatory, that is, targeted to the poorer children. Considering the whole generation of recipients, however, empirical findings point to a “Matthew effect”: Children with higher education and incomes had better chances of receiving gifts and bequests (Szydlik/ Schupp, 2004; Kohli et al., 2005). The family structure is defined by the joint family history as well as the past and current family composition. For example, the literature has shown that the number of siblings is negatively correlated with the chances of receiving bequests (Szydlik, 2004). This effect, however, was only pronounced when the respondent had at least three siblings. Due to statutory shares, the number of siblings diminishes the amounts inherited rather than the chances of receiving at least something. This need not be true for gifts, as these transfers can be targeted at one specific child and siblings cannot claim statutory shares. The num- Gifts, Bequests, and Social Inequality in West Germany 265 Schmollers Jahrbuch 131 (2011) 2 OPEN ACCESS | Licensed under | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.131.2.263 | Generated on 2023-01-16 13:36:19 ber of siblings should therefore be negatively correlated with the chances of receiving a gift (“competitors”), whereas it should affect the amount received to a lesser extent. Finally, the cultural-contextual structure refers to the conditions under which wealth can be accumulated or not. The key determinants are the economic system (market or planned), the tax regulation and welfare regime, as well as norms regarding gender, first-born children, etcetera. With respect to gender differences, we observe a long history of penalizing daughters in bequeathing (e.g., Kosmann, 2003). In today’s Germany, however, these differences have vanished. Recent studies did not find any gender differences with regard to bequests (e.g., Künemund et al., 2005). However, gender differences might appear in the non-regulated sphere of gift-giving. 3. Data and Method In 2001, the eighteenth wave of the SOEP, one page of the questionnaire was devoted to gifts and bequests. Respondents were asked: “Have you yourself ever inherited something or received a gift of great value? We are referring to gifts or inheritance of house or land, securities, investments, other forms of wealth or large amounts of money.” Respondents who answered positively were asked further questions about the year of receiving, the type of transfer (bequest or gift), its value at the time, and the giver(s). Information could be provided on up to three transfers. We focus only on intergenerational gifts and bequests, that is, gifts from parents as well as from grandparents (but not from parents-in-law). As almost 90% of all intergenerational transfers were received from parents, we simplify by always referring to parents as the givers, although the analysis also covers transfers from grandparents to grandchildren. We restrict our sample to West Germans born between 1930 and 1984 to reduce bias by selective mortality and to obtain a sample that is relatively homogeneous with respect to the main direction of transfer streams (downward), parents’chances of accumulating private property (market economy), and legal regulations on gift-giving and bequeathing. To analyze the transfer chances, we construct episodes starting at birth and ending with an event at the age of receiving the first transfer (in the first sample: a gift, in the second sample: a bequest). An episode is right-censored if a person has not received a transfer before the interview date. Further, episodes are censored two years after the last parent has died. We analyze this data with transition rate models (see equation 1). The transition rate r(t) is the intensity of experiencing an event under the condition of not having experienced such an event before (Blossfeld et al., 2007). The major advantage is that this analytical strategy takes into account (young) persons who have not yet received a trans- 266 Thomas Leopold and Thorsten Schneider Schmollers Jahrbuch 131 (2011) 2 OPEN ACCESS | Licensed under | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.131.2.263 | Generated on 2023-01-16 13:36:19 fer but may do so in the future. Another advantage is the possibility to consider time-varying covariates. rðtÞ¼expðXÞðequation 1Þ Transfer amounts are analyzed by ordinary least square (OLS) regression. As the respondents provided retrospective information on transfers received in different years, we adjust the value by the consumer price index. In addition, we take the logarithm of this adjusted value to avoid heteroscedasticity. Two variables have substantial shares of missing data. First, information on the father’s occupation was not sufficient to assign ISEI scores in almost 40% of all cases. Second, information on the number of siblings could not be obtained from respondents that participated in the survey in 2001, but no longer in 2003 (13% of all cases). Listwise deletion of these cases could lead to biased estimates. Therefore, we impute all missing data by chained equations, producing ten stacked sets of imputed data on which we run our analyses. The background model for the imputation includes all variables from the multivariate models and a number of auxiliary variables additional information from the SOEP data, such as sample membership, father’s education, and respondent’s education. In all analyses, we follow the procedures suggested by Rubin (1987). For details on theoretical and methodological aspects of our sample selection, definition of dependent variables, operationalization, and imputation, see Leopold /Schneider (2010). Descriptive statistics on the independent variables used in the models are given in the appendix (Tables A1 und A2). 4. Results Figure 1 presents descriptive results on the chances of receiving transfers divided by four different groups of parents’opportunity structures. We use the father’s score on the International Socio-Economic Index of Occupational Status (ISEI) as an indicator for parental resources. The score on the ISEI scale (ranging from 16 to 90) is derived from information on the father’s occupation when the respondent was 15 years old. We assigned respondents to three nearly equal-sized groups according to the father’s ISEI. This results in a lower (ISEI < 31), a middle (31 ISEI 43), and a higher (ISEI 43) status group. In addition, we define a separate group of respondents who are daughters or sons of farmers. Although farmers score low on the ISEI scale, they often own property, a home, and land. We estimate survivor functions for our event history data. The curves report the proportion of persons in each group who have not received a transfer up to a certain age. Gifts, Bequests, and Social Inequality in West Germany 267 Schmollers Jahrbuch 131 (2011) 2 OPEN ACCESS | Licensed under | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.131.2.263 | Generated on 2023-01-16 13:36:19 268 Thomas Leopold and Thorsten Schneider Source: SOEP, release 2007, own calculations. Estimates are based on 10 imputed data sets. Figure 1: Survivor functions for gifts (upper panel) and bequests (lower panel) by father’s socio-economic status, West Germany Survivor functions for receiving a gift are displayed in the upper panel of Figure 1, survivor functions for receiving a bequest in the lower panel. Up to the age of around 25, people very rarely received larger gifts. Subsequently, all survivor curves start to fall. We observe the strongest decline, and therefore the best chances of receiving a gift, for sons and daughters of farmers. But we also see a strong decline for the group with higher ISEI scores compared to those Schmollers Jahrbuch 131 (2011) 2 OPEN ACCESS | Licensed under | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.131.2.263 | Generated on 2023-01-16 13:36:19 with middle or lower scores. The lower panel for bequests shows the same rank order: The higher the father’s socio-economic position on the labor market, the higher the proportion of sons and daughters who received a bequest. However, children of farmers again seem to have high chances of receiving bequests. There are also some differences between gifts and bequests. First, the decline of the curves for bequests starts later, but the curves fall more steeply. Second, the proportion of persons receiving at least one bequest is much higher than for gifts, irrespective of the father’s occupation. In Models 1, 2, and 3, presented in Table 1, we estimate transition rate models analyzing the impact of different covariates. Model 1 refers to gifts; Models 2 and 3 refer to bequests. In all transition rate models, we allow for a time-dependent hazard rate by including yearly updated variables for age in linear and quadratic form.3We also include variables for gender, birth cohort, agricultural background, father’s ISEI, and the number of siblings.4In addition, we introduce time-varying covariates indicating the death of the first and second parent. Model 3 is expanded by three additional time-varying covariates. First, we include a dummy variable that goes from zero to one when a person receives a gift. The two other variables are interaction terms, which are calculated by this dummy variable and the indicators of opportunity structures. The most striking result from Table 1 is that men had better chances of receiving gifts than women (Model 1). In contrast, no gender differences were found for bequests. In additional analyses, we estimated Model 1 separately for men and women to test for gender differences in the effects of other predictor variables (estimates not shown). The results did not indicate any interaction effects between gender and other predictor variables. Further consideration of life course events showed that the gender effect was robust to controlling for the marital and birth biographies. Finally, analyses on the type of wealth transferred revealed that sons’higher chances of receiving were attributable to gifts of houses or land, whereas we found no gender differences in the transmission of liquid assets such as cash or bank deposits (Leopold / Schneider, 2010, in press). Concerning parental resources, our multivariate findings are consistent with the descriptive results presented in Figure 1: the higher the father’s ISEI, the better the child’s chances of receiving a transfer. These estimates yielded equal Gifts, Bequests, and Social Inequality in West Germany 269 Schmollers Jahrbuch 131 (2011) 2 3Using the first derivative with respect to age, we can calculate the maximum rate. The chance of receiving a gift is highest at age 42 (Model 1); the chance of receiving a bequest is highest at age 60 (Models 2 or 3). 4In 2001, the question on siblings referred only to living siblings. As information on transfers was collected retrospectively, other siblings might have still been alive when the transfer was received. Therefore, we use the information surveyed in the year 2003, which refers to all sister and brothers, even if they are deceased. OPEN ACCESS | Licensed under | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.131.2.263 | Generated on 2023-01-16 13:36:19 270 Thomas Leopold and Thorsten Schneider Schmollers Jahrbuch 131 (2011) 2 Table 1 Transition Rate Models and Linear Regression Models, West Germany Receiving gift Receiving bequest Value of gift Value of bequest Model 1 Model 2 Model 3 Model 4 Model 5 Constant -17.05** (0.50) -13.50** (0.35) -13.59** (0.36) 13.51** (0.36) 12.06** (0.32) Process time tvc Age 0.34** (0.02) 0.12** (0.01) 0.12** (0.01) Age 2 (/10) -0.04** (0.00) -0.01** (0.00) -0.01** (0.00) Male (ref:. female)0.16* (0.08) 0.01 (0.06) 0.01 (0.06) 0.45** (0.13) 0.15 (0.11) Father‘s Occupational Status Farmer 1.24** (0.14) 0.66** (0.13) 0.70** (0.13) 0.38 (0.23) 0.53* (0.25) ISEI 0.02** (0.00) 0.02** (0.00) 0.02** (0.00) -0.00 (0.00) 0.01* (0.00) Gift received tvc -1.67* (0.70) Interaction terms tvc Farmer * gift -0.17 (0.78) ISEI * gift 0.02 (0.01) Siblings (ref:. none) 1-0.35** (0.11) -0.06 (0.11) -0.06 (0.11) -0.17 (0.18) -0.42** (0.15) 2-0.36** (0.11) -0.21* (0.11) -0.21* (0.11) -0.38 + (0.21) -0.66** (0.18) 3-0.79** (0.16) -0.24* (0.12) -0.25* (0.12) -0.25 (0.25) -0.94** (0.19) 4ormore -0.78** (0.16) -0.66** (0.13) -0.67** (0.13) -0.47 + (0.25) -0.99** (0.25) Birth year (–1900) 0.07** (0.00) 0.06** (0.00) 0.06** (0.00) -0.04** (0.00) -0.02** (0.00) Parents deceased (ref.: both alive) tvc One parent -0.01 (0.10) 1.82** (0.11) 1.82** (0.11) Both parents 1.40** (0.22) 4.43** (0.12) 4.43** (0.12) OPEN ACCESS | Licensed under | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.131.2.263 | Generated on 2023-01-16 13:36:19