VoxEU Column Frontiers of economic research

Unequal bequests

Bequests have important economic and social consequences. Using a large sample drawn from the Health and Retirement Study, this column documents two results. First, about a third of US parents with wills plan to distribute their estates unequally among their children. This is especially common among families with stepchildren and children with whom the parent has little contact. Second, about 40% of parents die without wills.

The stock of wealth that turns over as people die is enormous. In the US alone, some $41 trillion will pass from the dead to the living in the first half of the 20st century (Friedman 2009). But the social impact of inheritance is more than a matter of money. It is also a matter of what money can buy and bring about and whether or not parents ‘play favourites’ when, at the end of their lives, they distribute their wealth.

Findings from models on end-of-life transfers

Given the magnitude of end-of-life transfers, it is not surprising that theoretical and empirical economists have analysed them. Economic theorists have proposed three models of bequests and their distribution. Each model begins by postulating a motive for bequests: altruism (Becker 1974), exchange (Bernheim et al. 1985, Cox 1987), or – drawing on evolutionary psychology – the ‘genetic child’ hypothesis (Cox 2003). All three models predict that parents will distribute their estates unequally. Contrary to this prediction, empirical economists have found that parents generally distribute bequests equally among their children. This robust finding of equal bequests was summarised 20 years ago by Behrman et al. (1995) who wrote: “In the United States, the limited empirical evidence thus suggests that, regardless of estate size, the typical pattern is equal bequests”. Subsequent research has confirmed that equal bequests are indeed the typical pattern.

New evidence on end-of-life transfers among Americans

In a recent paper, we employ data from the Health and Retirement Study (HRS), a household-level survey of a representative sample of more than 26,000 Americans over the age of 50, to investigate end-of-life transfers. We highlight here two of our main findings.

  • First, we found that unequal bequests are much more common than generally recognised. More specifically, we found that one-third of parents with wills plan to distribute their estates unequally among their children.

These plans for unequal bequests are particularly concentrated in what we call ‘complex families,’ that is, families with stepchildren and families with genetic children with whom the parent has had little or no contact, often, but not always, children from previous marriages.

  • Second, although the literature has focused on the bequest intentions of parents who have made wills, we found that a staggering number of elderly American parents have not made wills.

Figure 1 shows the upward trend in the percentage of parents who intend unequal bequests to their children. Analysing the data, we found that the aging of those in the sample does not explain this trend. We also found that the increase in the percentage of intended unequal bequests is independent of the marital status of the parent. Figure 2 shows that the rise in unequal bequests intentions is greatest in complex families. In simple families, the increase has been less steady and less pronounced, at least until 2008. Because of changes in patterns of marriage, divorce, cohabitation, and non-marital fertility over the past 60 years, simple families are more prevalent among older Americans than they will be in the future. Hence, projection of future patterns of end-of-life transfers should recognise this compositional shift and distinguish between simple and complex families.

Figure 1. Unequal bequest intentions, by parent's gender

Figure 2. Unequal bequest intentions, by family type

Difference in bequests between simple and complex families

We were able to quantify more precisely the differences in bequest intentions of complex and simple families. Controlling for socio-demographic, health, and economic variables, we found that for parents with stepchildren the likelihood of including all children in a will is about 30 percentage points lower than for parents without stepchildren. This negative stepchild effect is even stronger for widowed and divorced parents, with a reduction in the probability of leaving equal bequests of 40 to 60 percentage points, depending on the parent's gender and marital status. We also found that parents who have had limited or no contact with some of their children are much less likely to plan equal bequests. The data indicate that for parents who have less than one contact per year with at least one of their children, the probability that they plan equal bequests is 28 percentage points lower than for parents who are in more regular contact with all of their children.

These results establish that the lack of genetic links (in the case of stepchildren) and proximity (in the case of no-contact children) between parents and children affect parents' plans for end-of-life transfers. Additional results indicate that the existence of interactions between stepparents and stepchildren can offset the stepchild penalty. The more years the stepchild spent living with the stepparent, the higher the likelihood of being included in the will: 9 years of stepchildhood completely eliminates the stepchild penalty. We also found that different types of interactions between parent and child are associated with different likelihoods that the child is included in the will. On average, a stepchild has a 15% lower probability of being included in the stepparent’s will if the stepparent has genetic children. This penalty is entirely eliminated, however, if the stepchild's predicted income is lower than the genetic child's, and is more than outweighed if the predicted income difference exceeds 50%. The stepchild penalty is also outweighed if the stepparent regularly provides care for the stepchild’s children.

Turning our attention to those without wills, we found that the proportion of elderly American parents who do not have a will is surprisingly large. Although the probability of having a will increases with age, a staggering 30% of parents aged 70 and over have no wills. Of the Health and Retirement Study’s respondents who died between 1995 and 2010, 38% died intestate (i.e., without wills). Intestacy, which the economics literature generally ignores, has important implications for the distribution of end-of-life transfers.

Our analysis indicates that the likelihood of intestacy does not differ in simple and complex families. In simple families, in the absence of a will intestacy law distributes the estate equally among the children, consistent with the strong norm of equal distribution in simple families. In complex families, in the absence of a will intestacy law leads to different distributions in stepparent families than in no-contact families. When a stepparent dies without a will, her stepchildren inherit nothing from her. When a no-contact parent dies without a will, the no-contact children receive the same share of the estate as the ‘contact’ children. The extent to which parents in complex families who die intestate are aware of the provisions of intestacy law or would choose these distributions if they were aware of them cannot be assessed with available data.


Families in the US and Europe have changed profoundly in the past 60 years. Divorce, remarriage, cohabitation, and non-marital fertility have influenced the size and the composition of modern families. These changes have influenced the way American parents plan or fail to plan to distribute their wealth and those analysing end-of-life transfers need to recognise these changes. Focusing exclusively on the bequest intentions of parents who have made wills provides an incomplete and misleading picture of end-of-life transfers. Economists should recognise the (perhaps unintended) consequences of intestacy when analysing end-of-life transfers. Similarly, legislators should recognise that legal rules that accommodate the needs of simple families may fail to serve those of complex families.


Becker, G S (1974), "A Theory of Social Interactions", Journal of Political Economy, 82(6), 1063-93.

Behrman, J R, R A Pollak and P Taubman (1995), From Parent to Child, University of Chicago Press Economics Books, University of Chicago Press.

Bernheim, B D, A Shleifer and L H Summers (1985), "The Strategic Bequest Motive", Journal of Political Economy, 93(6), 1045-76.

Cox, D (1987), "Motives for Private Income Transfer", Journal of Political Economy, 95(3), 508-46.

Cox, D (2003), "Private Transfers within the Family: Mothers, Fathers, Sons, and Daughters", in Munnell, A H and A Sunden (eds.), Death and Dollars: The Role of Gifts and Bequests in America, Washington, D.C.: Brookings Institution Press.

Francesconi, M, R A Pollak and D Tabasso (2015), "Unequal Bequests", IZA Discussion Paper 8813, IZA, Bonn.

Friedman, L M (2009), Dead Hands: A Social History of Wills, Trusts, and Inheritance Law, Stanford, CA: Stanford Law Books.

630 Reads