VoxEU Column Frontiers of economic research

On the origin of the family

Why do people form long-lasting marital unions? This column presents new insights on what makes a family stick together. Families dominate more promiscuous pairs, in the sense that they can achieve greater survivorship and enhanced genetic fitness. The column suggests that this might provide an evolutionary explanation for the origin of religion as an institution to protect the family.

Why do humans live in families? The fact that only 3% of avian and mammal species are known to be familial suggests that the emergence of the family cannot be taken for granted, even among humans (Emlen 1995). Divorce is a common feature of modern life and non-traditional family structures are growing more common.

Families matter to economic policy. In many nations, government policy takes explicit account of families in tax policies, welfare policies, housing policies, and pension allocations. In many universities, family incomes affect eligibility for student aid. Research consistently shows that family structure can affect children’s lifetime performance. Recently, the British Conservative Party suggested a tax break for as an effort to stop marriages from faltering.

What is a family?

By family, we mean a “fidelity family”, which is any environment in which all the children share – and are raised by – the same parents. Since families dominate human society over time and across cultures, it seems obvious that there must be something very powerful and very general promoting this particular form of social organisation. One obvious explanation is the advantages of economies of scale. But this explanation encounters problems since families thrive in across very different forms of economic activity. Families are the dominate forms for people engaged in everything from hunting and gathering to investment banking.

Perhaps there is something deeper behind the family, something based on evolution.

Modelling the dominance of the family

In recent research (Francesconi et al. 2010), we address such questions within a simple framework. We build on Hamilton (1964) suggesting males and females only care for the survivorship and propagation of their own genes, and there are no economies of scale associated with the family. Two further assumptions are crucial:

  • Paternity is uncertain and males must resort to guard their opposite-sex mate if they wish to increase confidence in their being the biological parent of the children they feed, and
  • There are overlapping cohorts of dependent children; at each point in time parents need to provide resources to successive generations of offspring.

Our model can be illustrated as follows. Imagine two ancestral villages in which men and women plan to have more than one child and must choose the partner with whom they breed.

In both villages, there is full intra-household labour specialisation, with food being supplied by males and child rearing by females. The two villages, however, are different in the sense that in one village an individual has all his or her children with the same partner in a stable family grouping, while in the other village an individual begets each child with a different partner in a “promiscuous” relationship.

In which of the two villages do men and women achieve greater fitness? Notice that, regardless of where they live, males will have to guard as well as to provide food twice. That is, we impose neither economies of scale in food production nor a more efficient technology of guarding in association with a specific family configuration.

Our analysis indicates that the fidelity-based family dominates the promiscuous pair bonding, that is, the former can attain a higher probability of survival than the latter. This result is driven by the fact that, in the promiscuous village, a male shares the responsibility of supporting his own children with another male, and thus ends up free riding on this other male's food provision. He, instead, devotes more of his time to guarding, increasing his paternity confidence. But since every promiscuous male follows the same strategy and guarding is socially unproductive, children in promiscuous families receive less food and are less likely to survive than children in fidelity families, and therefore the (population) growth in the promiscuous village is bound to be lower.

When the notion of the family is extended to a more general context, in which kinship ties are accounted for, we find that food transfers from older to younger siblings enhance individual and societal fitness, because older males shift time from the unproductive activity of guarding towards greater food provision that is meant to support their younger siblings. Moreover, such a fitness gain can only materialise in the context of the family and not in the context of the promiscuous pair bonding. In an environment in which paternity is uncertain, two individuals who are highly confident that they share the same mother are likely to be genetically closer to each other than to their own offspring.

Essential to our theory is the notion that mate-guarding by males is unavoidable when fatherhood is uncertain, even though from society's viewpoint it is a complete waste. Therefore, any social “institution” that can, in equilibrium, reduce guarding will lead to a fitness gain at the individual level as well as at the group (or society) level. One of such institutions is religious beliefs and norms that put the fidelity family at centre stage. This reasoning will allow us to underpin why virtually all major world religions stress the role of the fidelity family – and, more specifically, marriage – as avoidance of casual sex. Our research illustrates the extent to which family-centred religious beliefs can become a successful adaptation (Wilson 2002) and deliver a more efficient equilibrium.

What are the implications?

One important implication of this analysis is offered by the observation that the promiscuous family in our model shares a number of features with the contemporary “blended” family described in Ginther and Pollak (2004). Like our promiscuous partnership, the blended family is characterised by stepchildren (born from an earlier marital union) living together with their half-siblings who are the joint children from an ongoing union.

A vast empirical literature has documented that experience of life in a blended family is associated with unfavourable child outcomes (e.g., Duncan and Brooks-Gunn 1997). Such poorer outcomes are explained by a variety of mechanisms, including parental stress and family conflict after separation and remarriage, inferior maternal time allocation within blended families, and economic hardship allowing family disruption.

Our model instead suggests an explanation based on the free-riding over resources given to children by the first and the second father figures; males in promiscuous unions undersupply resources to their children, while seeking to free-ride on each others' provisions. Related to this issue, future research may focus on the increasing importance of mothers' contribution to household resources. This will open up new avenues of analysis within the economics of the family.

Another implication of the model is related to the idea that norms and religious beliefs that are centred on the fidelity family can be seen as a successful group-level adaptation. Societies which adopt a belief system that supports the family are likely to outgrow otherwise comparable societies that do not. Our analysis predicts that, in societies with strong family-based religious norms, the family will be the dominant form of family life, whereas in societies in which family-centred norms are weak or not enforced, the fidelity family is expected to coexist with other family types.

It may not be a coincidence that virtually all the surviving world religions emphasise the centrality of the family. It will be interesting to see whether or not the family continues to have evolutionary advantages over other forms of mating when religious beliefs are weak and fatherhood uncertainty can be at least partially resolved through DNA paternity testing. This extension will have consequences for scientific studies of religions.

Moreover, in a world with perfect food storage (as we have assumed in our analysis), parents can easily secure their youngest (still needy) offspring full access to resources even after their death. But if food storage were not perfect, then this security is not automatic and parents will have to resort to other forms of intergenerational transfers. One of such forms is extended kin ties. When storage is impractical or impossible, parents will have to rely on kin transfers for the survivorship of their own non-adult offspring, and the food substitution from parents to kin will be complete. The notion of kinship here should be fairly broad and, besides own children, future extensions could include other related kin, such as uncles and cousins. This extension will lead to a new interesting interaction between economics and anthropology.

Finally, future work may also focus on other systems of parent-child transfers, which rely on goods that either are easier to store than food or can be bequeathed, and may or may not require support from kin other than own offspring. An example is fiat money, i.e., any object with no intrinsic value (such as a collection of pieces of paper or certain types of seashells) that can be used only as a medium of exchange (Kiyotaki and Wright 1989). An extension concerned with fiat money will have to deal with the possibility of saving, as distinct from the (in)ability of storing resources, as well as with strategic interactions between elderly parents and adult children (Lundberg and Pollak 2007). The presence of money may allow us to address new substantive questions by examining, for instance, the extent to which money reduces the need for parents to rely on extended kinship systems to transfer resources to their offspring.


Duncan, Greg J and Jeanne Brooks-Gunn (eds.) (1997), Consequences of Growing Up Poor, New York: Russell Sage Foundation.

Emlen, Stephen T (1995), “An Evolutionary Theory of the Family”, Proceedings of the National Academy of Science of the USA, 92(August): 8092--99.

Francesconi, Marco, Christian Ghiglino, and Motty Perry (2010) “On the Origin of the Family”, CEPR Working Paper DP7629, January.

Ginther, Donna K and Robert A Pollak (2004), “Family Structure and Children's Educational Outcomes: Blended Families, Stylized Facts, and Descriptive Regressions”, Demography, 41(4):671-96.

Hamilton, William D (1964), “The Genetical Evolution of Social Behaviour. I, II.” Journal of Theoretical Biology, 7(1):1-52.

Kiyotaki, Nobuhiro and Randall Wright (1989),“On Money as a Medium of Exchange”, Journal of Political Economy, 97(4):927-54.

Lundberg, Shelly and Robert A Pollak (2007), “The American Family and Family Economics”, Journal of Economic Perspectives, 21(2):3-26.

840 Reads