I congratulate Coall & Hertwig (C&H) for their thorough and rigorous analysis of grandparental investment. I agree there is a need for a more comprehensive theoretical framework of grandparental investment, one that explains the proximate and ultimate causes of investments, and that takes into account institutional arrangements, such as the presence of a state-financed welfare system. My comments focus on three claims the authors make.
The authors claim that the impact of grandparental investments in contemporary industrialized societies is relatively small when measured by classic indicators of fitness, such as number of grandchildren and grandchild mortality. This point needs some qualification. In post-demographic transition populations where people are living under low-risk conditions, direct grandparental investments will always have relatively little impact on the number and mortality of grandchildren when compared to the impact of direct grandparental investments in societies with high fertility and mortality rates. This is because the probability of grandchildren being born or dying is so much lower in post-demographic transition populations than it is in pre-demographic transition populations. Even if grandparents made very large direct investments in their grandchildren, this will have relatively little impact on the number and mortality of grandchildren in industrial societies.
However, the indirect investments grandparents make in industrialized societies do make a tremendous difference in the likelihood of their children and grandchildren surviving and reproducing. Here, I am referring primarily to the institutions that grandparents, and other taxpayers, support (e.g., Department of Health and Human Services) that create and maintain the relatively low fertility, low mortality conditions under which their children and grandchildren live. Although it is difficult to measure the impact indirect grandparental investments have on classic measures of fitness in contemporary industrialized societies, I would argue that their impact is very substantial.
C&H's position regarding the application of the Trivers–Willard hypothesis to parental and grandparental investments in industrialized societies also merits comment. I agree with them that the evidence of differential resource allocation to sons and daughters by parents in industrialized countries is conflicting, and that there is currently no convincing evidence of sex-biased grandparental investment in humans. However, researchers may have been looking for evidence in the wrong places. In the case of parental investment, Rosemary Hopcroft (Reference Hopcroft2005, p. 1116) notes that researchers generally focus on infants and relatively young children. In her opinion, parental biases in investments for this age group are unlikely in a relatively wealthy society such as the United States.
That is, there are abundant resources available for parents to be able to make these early investments in all children regardless of sex. Evolutionary theory suggests that in such a situation of unusual abundance all children, irrespective of sex, will be highly valued and invested in by their parents… [Status] attainment in American society requires much more than just investments in children as babies and adolescents; it requires ongoing investments in very lengthy periods of education and occupational training… It is provision of support over such long periods of time that is the scarce resource in American society… I suggest, therefore, that the focus of studies testing the resource allocation version of Trivers–Willard should be sex differences in parental provision of access to extensive training and education and encouragement of this process. (Hopcroft Reference Hopcroft2005, p. 1116)
Using years of education as an indirect measure of parental investment, and cumulative U.S. General Social Survey data, Hopcroft found support for the Trivers–Willard hypothesis in the United States. Sons of high-status fathers attain more education than daughters do, daughters of low-status fathers attain more education than sons, and high-status men have more sons among their biological offspring than do lower-status individuals.
In light of this finding, I am optimistic that researchers will eventually find evidence for sex-biased grandparental investments in industrialized societies. Promising avenues of research may include the financial and social investments made by grandparents in their grandchild's education, occupational training, business endeavors, and so forth.
My final comment concerns C&H's view that the poor now out-reproduce the wealthy in post-demographic transition societies. Has the availability of effective contraception really severed the link between status and reproduction in modern human societies? One of the challenges of evaluating this question is having the right kind of data. I again look at Hopcroft's work on reproductive success in the contemporary United States (Hopcroft Reference Hopcroft2006, p. 107) to illustrate my point. She used a representative sample containing data that made a distinction among an individual's biological, step, and adopted children. Previous research on reproductive success in the United States. did not make these distinctions and tended to rely on data that only report female fertility or the number of children in a household. With data that likely includes all, and only, the living biological children of men and women, Hopcroft (Reference Hopcroft2006) found that high-income men have more biological children than do low-income men and high-income women.
There is also reason to believe that the economic status of grandparents in at least one industrialized society is positively correlated with their long-term fitness, as measured by number of grandchildren. Recently, Anna Goodman and Ilona Koupil (2009) reported their research findings on the social and biological determinants of reproductive success in Sweden during the 20th century. They used high quality, multi-generational data on a cohort born from 1915 to 1929, a time when the Swedish demographic transition was largely concluded. Goodman and Koupil had many interesting findings. The main one regarding grandparental socioeconomic status and reproduction follows:
The effect of SEP [family socioeconomic position] is interesting in highlighting the potential…continuity [between pre- and post-demographic transition populations with respect to]… the determinants of reproductive success. The continuity is illustrated by the positive correlation in both sexes [of grandparents] between higher SEP and a greater number of descendents. This is consistently observed in traditional populations (Hill & Kaplan Reference Hill and Kaplan1999; Low Reference Low2000) and, although evidence in modern populations is more variable (Clarke & Low Reference Clarke and Low2001), our study adds to the evidence that the demographic transition does not necessarily erase the relationship between access to resources and reproductive advantage. (Goodman & Koupil Reference Goodman and Koupil2009, p. 339)
I congratulate Coall & Hertwig (C&H) for their thorough and rigorous analysis of grandparental investment. I agree there is a need for a more comprehensive theoretical framework of grandparental investment, one that explains the proximate and ultimate causes of investments, and that takes into account institutional arrangements, such as the presence of a state-financed welfare system. My comments focus on three claims the authors make.
The authors claim that the impact of grandparental investments in contemporary industrialized societies is relatively small when measured by classic indicators of fitness, such as number of grandchildren and grandchild mortality. This point needs some qualification. In post-demographic transition populations where people are living under low-risk conditions, direct grandparental investments will always have relatively little impact on the number and mortality of grandchildren when compared to the impact of direct grandparental investments in societies with high fertility and mortality rates. This is because the probability of grandchildren being born or dying is so much lower in post-demographic transition populations than it is in pre-demographic transition populations. Even if grandparents made very large direct investments in their grandchildren, this will have relatively little impact on the number and mortality of grandchildren in industrial societies.
However, the indirect investments grandparents make in industrialized societies do make a tremendous difference in the likelihood of their children and grandchildren surviving and reproducing. Here, I am referring primarily to the institutions that grandparents, and other taxpayers, support (e.g., Department of Health and Human Services) that create and maintain the relatively low fertility, low mortality conditions under which their children and grandchildren live. Although it is difficult to measure the impact indirect grandparental investments have on classic measures of fitness in contemporary industrialized societies, I would argue that their impact is very substantial.
C&H's position regarding the application of the Trivers–Willard hypothesis to parental and grandparental investments in industrialized societies also merits comment. I agree with them that the evidence of differential resource allocation to sons and daughters by parents in industrialized countries is conflicting, and that there is currently no convincing evidence of sex-biased grandparental investment in humans. However, researchers may have been looking for evidence in the wrong places. In the case of parental investment, Rosemary Hopcroft (Reference Hopcroft2005, p. 1116) notes that researchers generally focus on infants and relatively young children. In her opinion, parental biases in investments for this age group are unlikely in a relatively wealthy society such as the United States.
That is, there are abundant resources available for parents to be able to make these early investments in all children regardless of sex. Evolutionary theory suggests that in such a situation of unusual abundance all children, irrespective of sex, will be highly valued and invested in by their parents… [Status] attainment in American society requires much more than just investments in children as babies and adolescents; it requires ongoing investments in very lengthy periods of education and occupational training… It is provision of support over such long periods of time that is the scarce resource in American society… I suggest, therefore, that the focus of studies testing the resource allocation version of Trivers–Willard should be sex differences in parental provision of access to extensive training and education and encouragement of this process. (Hopcroft Reference Hopcroft2005, p. 1116)
Using years of education as an indirect measure of parental investment, and cumulative U.S. General Social Survey data, Hopcroft found support for the Trivers–Willard hypothesis in the United States. Sons of high-status fathers attain more education than daughters do, daughters of low-status fathers attain more education than sons, and high-status men have more sons among their biological offspring than do lower-status individuals.
In light of this finding, I am optimistic that researchers will eventually find evidence for sex-biased grandparental investments in industrialized societies. Promising avenues of research may include the financial and social investments made by grandparents in their grandchild's education, occupational training, business endeavors, and so forth.
My final comment concerns C&H's view that the poor now out-reproduce the wealthy in post-demographic transition societies. Has the availability of effective contraception really severed the link between status and reproduction in modern human societies? One of the challenges of evaluating this question is having the right kind of data. I again look at Hopcroft's work on reproductive success in the contemporary United States (Hopcroft Reference Hopcroft2006, p. 107) to illustrate my point. She used a representative sample containing data that made a distinction among an individual's biological, step, and adopted children. Previous research on reproductive success in the United States. did not make these distinctions and tended to rely on data that only report female fertility or the number of children in a household. With data that likely includes all, and only, the living biological children of men and women, Hopcroft (Reference Hopcroft2006) found that high-income men have more biological children than do low-income men and high-income women.
There is also reason to believe that the economic status of grandparents in at least one industrialized society is positively correlated with their long-term fitness, as measured by number of grandchildren. Recently, Anna Goodman and Ilona Koupil (2009) reported their research findings on the social and biological determinants of reproductive success in Sweden during the 20th century. They used high quality, multi-generational data on a cohort born from 1915 to 1929, a time when the Swedish demographic transition was largely concluded. Goodman and Koupil had many interesting findings. The main one regarding grandparental socioeconomic status and reproduction follows:
The effect of SEP [family socioeconomic position] is interesting in highlighting the potential…continuity [between pre- and post-demographic transition populations with respect to]… the determinants of reproductive success. The continuity is illustrated by the positive correlation in both sexes [of grandparents] between higher SEP and a greater number of descendents. This is consistently observed in traditional populations (Hill & Kaplan Reference Hill and Kaplan1999; Low Reference Low2000) and, although evidence in modern populations is more variable (Clarke & Low Reference Clarke and Low2001), our study adds to the evidence that the demographic transition does not necessarily erase the relationship between access to resources and reproductive advantage. (Goodman & Koupil Reference Goodman and Koupil2009, p. 339)