Chapter 14 Altruism, exchange or indirect reciprocity: what do the data on family transfers show?

https://doi.org/10.1016/S1574-0714(06)02014-8Get rights and content

Abstract

Most models of family transfers consider only two generations and focus on two motives: altruism and exchange. They also assume perfect substitution between inter vivos financial transfers and bequests to children. On the contrary, this survey of recent developments in the literature emphasizes the strong heterogeneity of downward financial transfers and motives for these transfers over the life-cycle. In face of the empirical failure of standard models in developed countries (these models may perform better in less developed countries or in old Europe), it also advocates “mixed” motivations of transfers, such as strategic altruism, models with endogenous heterogeneous behavioral regimes (Becker, Cigno), and especially indirect reciprocities between three generations, which lead to the replication of the same type of transfer from one generation to the next. Indirect reciprocities appear able to accommodate several empirical puzzles: they are thus compatible (against altruism) with small compensatory effects of transfers both between and within generations, and (against exchange) with the lack of parents' observable counterpart to financial or time support given by their children. They also predict “3rd generation effects”—transfers between parents and children being determined by grandparents' transfers or again grandchildren's characteristics—which appear corroborated by (mainly French or U.S.) available evidence. We thus face the challenge of innovative modelling of indirect reciprocities within the framework of individual forward-looking rationality.

Introduction

The economic literature on intergenerational transfers within the family has considerably developed since the last thirty years, or so. “Standard” models in this field emphasize two main competing motives for inter vivos transfers (i.e. between living members of the family):

  • altruism towards children, especially in a Barro-Becker form: parents care for the well-being of their offspring;

  • self-interested (inter-temporal) exchange between parents and children, meaning that the implicit contract where (e.g.) parents trade prior education, or the promise of future inheritance, for children's support in their old age, is expected to be mutually advantageous—if enforceable.

Moreover, bequests (post-mortem transfers) may also reflect a precautionary motive against lifetime uncertainty in the absence of efficient annuity markets (the ownership of durable, illiquid, indivisible assets, such as homes, is another reason).

There are several theoretical and/or empirical surveys of this literature, which distinguish various forms of altruism and exchange—as well as their recent extensions based upon processes of imitation or cultural transmission (such as children's “preference shaping”)—, and compare the predictions derived with observed transfer behavior in the U.S. and elsewhere. Their general conclusions tend to emphasize the poor empirical performances of altruistic and exchange-motivated models, regarding equally the determinants of inter vivos transfers and those of bequests.1

In comparison with these studies, our own review, which focuses on modern developed countries, does not intend to be exhaustive and may look slightly oriented. Its main objective is threefold. First, we try to grasp better why the predictions of standard models of altruism or exchange fail to apply in previous tests. A key reason is the perfect substitution that these models assume between financial help, various gifts and bequests to children: it does not accord at all with the observed strong heterogeneity of downward financial transfers and motives for these transfers over the life-cycle (see section 3).

Second, we show more precisely where models of altruism and exchange fail, empirically, using new (French) data to design more powerful tests and to underline stylized facts about transfers which are generally overlooked or unexplained (see sections 5 More on inter vivos transfers in developed countries, 7 Indirect reciprocities: preliminary French and U.S. evidence).

Third, to solve the empirical puzzles encountered by standard models, we advocate “mixed” motivations for transfers, including “strategic altruism”, Becker's or Cigno's models with several endogenous behavioral regimes (section 5.4) and, especially, indirect (serial) reciprocities between three generations, which lead to the replication of the same type of transfer from one generation to the next (see sections 6 Indirect reciprocities between generations: theory, 7 Indirect reciprocities: preliminary French and U.S. evidence). In the absence of satisfactory economic models of indirect reciprocities, the important part of the paper devoted to these mechanisms is bound to be more tentative and to have a dominant empirical orientation; one of its objective, however, is to make a link between the models of family intergenerational transfers and the approaches to reciprocity developed, in other contexts, in this handbook.

Standard models of transfers rely indeed on a simplistic view of the intergenerational family: either no apparent family, for the selfish life-cycler who leaves “accidental” bequests owing to random life duration; or the unified Beckerian family, headed by a benevolent patriarch driven by altruism towards his progeny; or even pure self-interested family relations, where intergenerational exchanges act as (imperfect) substitutes for private exchanges or contracts that should exist on ideal markets. Drawing on anthropology, the concept of reciprocity, based upon the gift-return-gift relation, should lead to a more satisfactory view of family linkages, providing new motivations for transfers and more realistic norms of behavior between kin generations.2

More specifically, we shall try to convince the reader that indirect forms of reciprocity between generations may be viewed as appropriate dynamic syntheses of altruism and exchange allowing, with minimal deviations, the introduction of “intermediate” motivations for transfers which better fit the data.

Note first that our approach to reciprocity will be quite specific by comparison with other analyses in this handbook. Instead of being applied to general human sociality, or to explain reciprocity between strangers by a norm of “fairness” leading to retaliation behavior in experimental games,3 or again to foster cooperation within small communities,4 it concerns family and kinship, i.e. blood relationship and asymmetric links between parents and children. Moreover, the concept must be adapted to the succession of generations, taking a particular form which has been introduced by the French anthropologist Mauss (1968): namely, indirect (serial) reciprocity, involving three successive generations at a time and leading to infinite endless chains of descending or ascending transfers between parents and children.

Reciprocity differs from market exchange in that it proceeds from a set of “internal” obligations—to give, to receive, and to give back—, whether driven by norms or collective values, group or social pressure (Kolm, 2000). Moreover, if direct reciprocity looks still like standard quid pro quo exchange between two parties, indirect reciprocity implies either that the beneficiary generation gives back to a third generation (e.g., provides bequests to one's children “in return” for inheritance received from one's parents), or that the giving generation will be paid back by a third generation (e.g., will receive support given by her children in return to past support given to own parents):5 in each case, it leads to the replication of the same type of transfer along the intergenerational chain. For instance, the way to pay back my parents for the education I received is to give myself proper education to my own children, and so on; of course, this process will often work through imitation or transmission of norms.

Our initial motivation for extending exchange or reciprocity to three generations within this encompassing framework came from French evidence on parent-to-child transfers: strong and highly significant retrospective effects, both qualitative and quantitative, have been systematically found for downward transfers and transmission practices on different data sets by different authors (Arrondel et al., 1997). Hence, what is left in bequest (and declared gifts) to children appears commensurate to what has been received from parents, the life propensity to bequeath out of inheritance being much higher than the one out of human resources. Moreover, transmission patterns and behaviors tend to be reproduced from one generation to the next: everything being equal (especially the amount of wealth owned), inheritance received through a will increases the probability to make a will, recipients are more likely to become donors, etc. (sections 7.1 “Retrospective effects” on parent-to-child inter vivos transfers and bequests, 7.2 Discussion: robustness of retrospective effects on parent-to-child transfers).

These results refer to transmission practices and downward transfers influenced by corresponding behaviors of the previous generation. For that reason, they will be interpreted as a backward-looking (i.e. retrospective) and downward indirect reciprocity. Likewise, Barro-Becker dynastic altruism—where parents care about their child's utility and expect their children to exhibit a comparable degree of altruism and to adopt a similar bequeathing behavior towards their own children, and so on—may be seen as a particular variant of forward-looking downward indirect reciprocity, where agents are endowed with an infinite horizon.6 Four types of these reciprocities will then be considered, according to the orientation of time (backward or forward), and the direction of the transfer considered: upward (child-to-parent) or downward. Thus, examples of upward and forward-looking ones are so-called “demonstration effects”, where effective support of elder parents is assumed to favor, by one way or another, later support by own children during one's old age.7,8

Section 2 provides a quick reminder of theoretical models of family transfers, relying on three main explanations: precautionary savings against lifetime uncertainty, leading only to accidental bequests; altruism and exchange, in different forms, that should also account for the determinants of inter vivos transfers, whether from parents to children or, especially in the case of exchange motivations, from children to parents.

A major source of misunderstanding and confusion in the literature comes from the lack of a proper definition and classification of inter vivos transfers. Section 3 claims that this identification problem is of central importance for parent-to-child, financial transfers. Their strong heterogeneity is not enough acknowledged, as if college fees to a 20-year old child and major (official, declared) gifts received some 25 or 30 years later as anticipated inheritance were perfect substitutes. We propose a typology of transfers according to their timing and their objective (whether they add to child's income, consumption or wealth). This typology has an obvious bearing on the lively Kotlikoff–Modigliani debate concerning the share of received wealth in total existing accumulation, as well as on the related issue about the quantitative importance of inter vivos transfers with respect to bequests.

Section 4 sums up existing tests of standard transfer models, with rather negative conclusions for developed countries: in particular, pure altruism cannot account for the observed absence, or limited importance, of compensatory effects for parent-to-child transfers, either between generations, or among siblings.9 Indeed, we emphasize the dominant feeling of disillusion in the profession about the explanatory power of these models.

Section 5 adds to previous empirical studies in several directions. Most models of transfers, including recent ones based on values transmission or preference formation, assume a unique family composition of either two or three generations: data show, however, that the most frequent one is rather with four overlapping generations (with two working ones); also, there is no such thing as a “representative” family but a large diversity of compositions—a result which has implications for policy design and inequality of well-being between families.

Furthermore, while adopting a restrictive definition of time transfers and using richer data sets, with detailed information on services and financial support given to old parents, we show precisely at which steps most forms of exchange fail empirically in modern occidental countries—and incidentally why exchange motivations might be more relevant in old Europe or in less developed countries. First, support of old parents by their children appears the only significant (time or financial) upward transfer in developed countries. Second, the latter transfers cannot be explained—in French or U.S. data—by any observable, past, present or expected counterpart, from parents to children: helpers have not received more than non helpers, and do not expect higher inheritances, but smaller ones. This empirical puzzle is solved by two non-standard forms of exchange and altruism, with heterogeneous behavioral regimes: the self-enforcing family constitution model (Cigno, 1993); and the model of either free or constrained altruism (Becker and Tomes, 1986).10 It can also be explained by indirect reciprocity: in helping children expect then to be paid back not by parents, but by similar help received later from their own children.

Section 6 first emphasizes that reciprocity encompasses mixed or intermediate (“other-oriented”) motivations between (strategic) exchange and pure altruism, while allowing for richer relations within the family, especially through the notion of gift “ambivalence”: a gift is both a positive act of sharing, and a negative one of domination exerted on the receiver. Indirect reciprocity is then introduced as a form of “general” reciprocity, a concept already analyzed at length by Kolm (1984), who has identified its two basic ingredients: the “rebound effect”—one gives to the givers—and the “propagation effect”—the helped help in turn. Applied to family transfers between three generations, this analysis leads to four types of serial reciprocities: backward- or forward-looking ones, for upward or downward transfers. Upward transfers are governed by the rebound effect, downward ones by the propagation effect; whatever the specific motivation at work (imitation, habits, education, social approval, etc.), backward-looking reciprocities typically reflect the obligation to comply to the relevant effect, and forward-looking ones the intentional desire to provoke it.11

Section 7 reports preliminary tests, performed on French or U.S. data, concerning specific predictions of indirect reciprocities, i.e. “3rd generation effects”: transfers between parents and children depend on grandparents' or grandchildren's characteristics/behavior. Parent-to-child transfers appear strongly influenced by the corresponding behaviors or transmission practices of the previous generation. In fact, estimated retrospective effects appear not only very significant and robust but also highly selective: thus, wealth gifts bestowed on children depend specifically upon wealth gifts received, not so much upon other receipts, whether inheritance or financial help. Moreover, in favor of demonstration and related effects, observed (time or financial) support given to old parents could be motivated by the expectation of receiving comparable assistance from one's own children during old days—although the possibility of alternative interpretations of the findings cannot be ruled out.

Section 8 deals briefly with the macroeconomic and policy implications of alternative motivations of transfers, including: Cigno's self-enforcing family “constitutions”, where the agent has the choice to “go-it-alone” (life-cycle saving) or to “comply” to a family norm of extended exchange between three generations; Becker's parental altruism with investment in a child's human capital under two regimes, either free with operative transfers a la Barro that crowd-out public transfers, or constrained by the interdiction of negative bequests; and indirect reciprocities, introducing specific links between three successive generations. We consider in turn the impact of government redistribution policy on growth, saving and education; the interaction effects between public and private transfers; and distributive issues, such as the effect of transfers on income inequality, on wealth concentration and inter-generational mobility. In particular, altruism may paradoxically give complementary roles to the family and the State, since more public transfers towards the elderly—pensions, health—should entail more private transfers to (grand) children—education, gifts, bequests.12

Section 9 sums up the main conclusions drawn in this survey and indicates some directions for future research. Especially welcome would be a more thorough comparison of the features and motives of family upward and downward transfers in modern developed countries and in less developed countries (LDCs thereafter), or in old Europe.

Section snippets

Altruism, exchange, and other motives: a quick reminder

A theorist not familiar with the topic might be scared by the impressive blossoming of miscellaneous models and motivations introduced in the literature in order to explain bequests and other family transfers between generations: capitalist (or entrepreneurial) bequests; precautionary (or accidental) bequests; transfers motivated by parental altruism, using different specifications—pure (Beckerian) altruism, retrospective (or golden-rule) bequests, paternalistic, “joy of giving” or “warm glow”

Heterogeneity of (financial downward) transfers

The shift of interest in the literature from bequests to inter vivos transfers has created a difficulty: as opposed to the former, the latter encounter theoretical and empirical problems of definition and typology which are far from innocuous but have been largely overlooked so far. We shall first give a general idea of these conceptual pitfalls before focusing on the case of parent-to-child financial transfers, which has the most important implications.

Previous tests of transfer models

Keeping in mind this heterogeneity and distribution of financial parent-to-child transfers, we may now give a brief appraisal of existing empirical tests, which have been performed (for a large majority of them) on various U.S. data sets, including panel data. French and other European evidence will also be considered for comparison. These tests lead to rather jaundiced conclusions for the three broad types of transfer models considered in the literature: precautionary savings against lifetime

More on inter vivos transfers in developed countries

In order to pinpoint further specific shortcomings of standard models and to decide which extensions may better work, we shall take a closer look at three characteristics of observed family transfers (section 3.1): their timing—at which stage of the life-cycle are they received or given; their direction—the relative diffusion and quantitative importance of upward and downward transfers; their nature—whether in financial form or time transfers: the ultimate goal would be to get an overall

Indirect reciprocities between generations: theory

Cigno's and Becker's 2-regimes models represent one way to remedy for major empirical failures of standard exchange and altruistic models, while only introducing a few changes or extensions. In the same spirit, indirect reciprocities between generations constitute another, also with minimal deviations from standard models.

At the theoretical level, they can be seen as a reconciliation of exchange and altruism, that blur somewhat the distinction between the two generic motivations for transfers:

Indirect reciprocities: preliminary French and U.S. evidence

Precise predictions of indirect reciprocities are difficult to formulate because of the lack of completely specified and articulated models. Nevertheless, three kinds are easy to state and have been already emphasized: the first two ones give indirect reciprocities an empirical advantage over altruism and exchange; the last ones—“3rd generation effects”—provide specific tests of these reciprocities.

First, contrary to pure altruism, indirect reciprocities should entail at most limited

Economic implications of family transfers (prolegomena)

The theoretical and empirical analysis of the motivations for private intergenerational transfers is not only of interest for family micro-economists. These motivations are even more important for their economic implications concerning the interactions between public and private transfers and the effects of public policies on growth, saving and education, as well as distributive issues, such as the effect of transfers on income inequality or on the concentration and intergenerational immobility

Conclusions

To conclude this empirical survey devoted to the types and motives of family intergenerational transfers in developed countries, five points are worth being (re)emphasized.

References (135)

  • J.D. Adams

    Personal wealth transfers

    Quarterly Journal of Economics

    (1980)
  • J.G. Altonji et al.

    Parental altruism and inter vivos transfers: theory and evidence

    Journal of Political Economy

    (1997)
  • J.G. Altonji et al.

    The effects of income and wealth on time and money transfers between parents and children

  • D. Anderberg et al.

    Self-enforcing intergenerational transfers and the provision of education

    Economica

    (2003)
  • J. Andreoni

    Giving with impure altruism: application to charity and Ricardian equivalence

    Journal of Political Economy

    (1989)
  • L. Arrondel et al.

    Successions et héritiers à travers les données fiscales

    Economie et Prévision

    (1991)
  • L. Arrondel et al.

    Les partages inégaux des successions entre frères et sœurs

    Economie et Statistique

    (1992)
  • L. Arrondel et al.

    Succession capitaliste et succession familiale: un modèle econométrique à deux régimes endogènes

    Annales d'Economie et de Statistique

    (1998)
  • L. Arrondel et al.

    Que nous enseignent les enquêtes sur les transferts patrimoniaux en France?

    Economie et Prévision

    (1991)
  • L. Arrondel et al.

    Family transfers involving three generations

    Scandinavian Journal of Economics

    (2001)
  • L. Arrondel et al.

    Bequest and inheritance: empirical issues and French–U.S. comparison

  • L. Arrondel et al.

    Préférences de l'épargnant et accumulation patrimoniale

    Economie et Statistique

    (2004)
  • L. Arrondel et al.

    Les transferts inter vivos en France: investissements humains, aides financières et transmission du patrimoine

    Economie et Prévision

    (1998)
  • A.B. Atkinson

    Comments on redistribution, inheritance and inequality: an analysis of transitions of J. Davies and P. Kuhn

  • R.J. Barro

    Are government bonds net wealth

    Journal of Political Economy

    (1974)
  • R.J. Barro

    The Ricardian approach to budget deficits

    Journal of Economic Perspectives

    (1989)
  • Barry, C., de Eneau, D., Hourriez, J.M. (1996). “Les aides financières entre ménages”. Insee Première, (441),...
  • G.S. Becker

    A theory of social interactions

    Journal of Political Economy

    (1974)
  • G.S. Becker

    Family economics and macro behavior

    American Economic Review

    (1988)
  • G.S. Becker

    A Treatise on the Family

    (1991)
  • G.S. Becker

    The economic way of looking at behavior

    Journal of Political Economy

    (1993)
  • G.S. Becker

    Accounting for Tastes

    (1996)
  • G.S. Becker et al.

    The family and the state

    Journal of Law and Economics

    (1988)
  • G.S. Becker et al.

    Human capital and the rise and fall of families

    Journal of Labor Economics

    (1986)
  • J. Behrman et al.

    Parental preferences and provision for progeny

    Journal of Political Economy

    (1982)
  • Bennet, S.K. (1990). “Economic and non-economic factors motivating bequest patterns”, mimeo, Trinity...
  • T.C. Bergstrom

    Economics in a family way

    Journal of Economic Literature

    (1996)
  • T.C. Bergstrom et al.

    How altruism can prevail in an evolutionary environment

    American Economic Review

    (1993)
  • B.D. Bernheim

    A neoclassical perspective on budget deficits

    Journal of Economic Perspectives

    (1989)
  • B.D. Bernheim

    How strong are bequest motives? Evidence based on estimates on the demand for life insurance and annuities

    Journal of Political Economy

    (1991)
  • B.D. Bernheim et al.

    Is everything neutral?

    Journal of Political Economy

    (1988)
  • B.D. Bernheim et al.

    Bequests as signals: an explanation for the equal division puzzle

    Journal of Political Economy

    (2003)
  • B.D. Bernheim et al.

    The strategic bequest motive

    Journal of Political Economy

    (1985)
  • D.L. Bevan et al.

    Intergenerational transfers and inequality

    Greek Economic Journal

    (1979)
  • A.S. Blinder

    Comments on Modigliani (chap. 1) and Kotlikoff and Summers (chap. 2)

  • Brezis, L. (2000). “Population, social classes, and economic growth during industrialization”, mimeo, Bar-Ilan...
  • Brown, J.R., Weisbenner, S.J. (2002). “Is a bird in hand worth more than a bird in the bush? Intergenerational...
  • N. Bruce et al.

    The Rotten Kid theorem meets the Samaritan dilemma

    Quarterly Journal of Economics

    (1990)
  • J. Caldwell

    A theory of fertility: from high plateau to destabilization

    Population, Development Review

    (1978)
  • C.D. Caroll

    Why do the rich save so much

  • Cited by (125)

    • Crowding-out or crowding-in? Effects of LEAP 1000 unconditional cash transfer program on household and community support among women in rural Ghana

      2021, World Development
      Citation Excerpt :

      When a government program comes in and provides a similar transfer the support giver will lose the reason for providing support and stops doing so without changing the recipient’s utility level. Alternative and more recent models, such as the exchange model (Arrondel & Masson, 2006; Cox, 1987), repeated games and investment in social capital (Ostrom & Ahn, 2009) or social norms (Cox & Jimenez, 1992; Sunstein, 1996) challenge the assumption of altruism and argue that support is given based on personal or future expectations. As an example to the exchange model, the empirical study by Bernheim et al. (1986) shows that support of children in the form of visits and phone calls to their parents can be seen as a trade for future inheritance.

    • A longitudinal perspective on inter vivos transfers between children and their parents in need of long-term care

      2021, Journal of the Economics of Ageing
      Citation Excerpt :

      This share tends to be even higher when the parent needs LTC. By contrast, the share of adult children that provides financial support to their parents is commonly reported to be around 3–10% in cross-sectional datasets (Altonji et al., 1996; Arrondel and Masson, 2006; Attias-Donfut et al., 2005; Cox and Stark, 2001). In a next step, we analyzed the effect of parents’ need for LTC on inter vivos transfers from children to parents.

    • Intergenerational transfers of time and money over the life course

      2021, Handbook of Aging and the Social Sciences
    View all citing articles on Scopus
    *

    We would like to thank Alessandro Cigno, Michael Hurd, Serge-Christophe Kolm, Jean Mercier-Ythier, Muriel Roger and especially François-Charles Wolff for their very helpful and stimulating remarks on successive versions of this paper. We are also grateful to Jim Ogg for his careful rereading of the paper.

    View full text