Lifehistory Trade-Offs Influence Women’s Reproductive Strategies

doi:10.21203/rs.3.rs-3866853/v1

Download PDF

Research Article

Lifehistory Trade-Offs Influence Women’s Reproductive Strategies

https://doi.org/10.21203/rs.3.rs-3866853/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 26 Mar, 2024

Read the published version in Adaptive Human Behavior and Physiology →

You are reading this latest preprint version

We use a modeling approach parameterized with current UK lifehistory data to show that, if they are to match the reproductive performance of women in the higher socio-economic classes, women in lower socio-economic classes must opt for a significantly earlier onset of reproduction in order to offset the higher class-specific mortality and infertility rates that they face. Women from low socio-economic classes cannot afford to postpone reproduction in order to enter a career unless that career pathway facilitates upward mobility into a higher socioeconomic class.

lifehistory

reproductive decisions

socioeconomic class

women

contingent decisions

Evolutionary theory predicts that the expected duration of the reproductive life span should influence age at first reproduction: individuals who expect to die early should begin to reproduce earlier (Charnov 1991; Stearns 1992). There is some empirical evidence to support this prediction in humans, although much of this evidence is correlational. Low et al. (2008) showed that, across countries, female life expectancy is associated with age at first birth, with earlier onset of reproduction where mortality rates are high (see also Walker et al. 2006). Similarly, Lycett & Dunbar (2000) found that expected future reproductive lifespan was a significant factor influencing single women’s decisions to opt for an abortion. Guegan et al. (2001) reported that disease burden (interpreted as an index of prevailing mortality risk) predicts total fertility rates across human societies, while Quinlan (2007) found that, in societies where death rates are high (due to disease, famine or warfare), weaning is earlier and maternal investment in offspring is reduced compared to societies where conditions are more favourable. In sum, poor quality environments where the future is unpredictable favour a ‘fast’ lifehistory strategy that involves a suite of behaviours that involve an early switch from growth to reproduction, high fertility and low parental investment (Bielby et al. 2007; Nettle 2010).

Essentially similar effects have been noted at a within-society level. In a study of US urban environments, reproduction was shifted earlier in neighbourhoods which had higher mortality and morbidity rates (Geronimus 1996; Geronimus et al. 1999; Wilson & Daly 1997). Similarly, Nettle & Cockerill (2010; see also Nettle 2011) reported that, in the UK, women’s age at first reproduction was, on average, 8 years earlier in a low socio-economic neighbourhood than in a high SES one. Indices of maternal investment (such as birthweight and the duration of breastfeeding) are well known to vary with socioeconomic status (Dubois & Girard 2006; Kohlhuber et al. 2008; Mortensen et al. 2008). Similarly, many studies have noted that women begin reproducing earlier, reproduce more frequently, and invest less in each offspring in neighbourhoods where social and economic deprivation resulted in a shortened expectation of life (Nettle 2010, 2011; Brooks-Gunn et al. 1993; McCulloch 2001; Smith & Elander 2006).

The lifehistory consequences of reproducing may be exacerbated in a knowledge-based economy such as that prevailing in most of the industrialised world. The need to ensure that offspring are competitive in terms of education, wealth and/or social/economic opportunities favours a reduction in fertility and a corresponding investment in offspring quality (Becker & Lewis 1973; Rogers 1990; Mace 1998). In such contexts, future earning potential may be important and may favour the postponement of reproduction in order to further investment in social or career prospects that offer enhanced mate choice opportunities or the acquisition of resources that can be invested in offspring. In such contexts, women who can afford to do so (and hence, especially, women from higher socioeconomic [SES] classes) should be more willing to delay the onset of childbearing in order to further their educational and career opportunities.

Although many of these contingent effects are well known, they have typically been examined piecemeal, in a descriptive rather than an explanatory framework and almost never in a lifehistory context (Nettle 2010, 2011). Here, we use a simulation model to investigate the lifetime fertility consequences of postponing reproduction in the interests of furthering career opportunities when these are likely to have significant socioeconomic consequences. We assume that, in a socio-economic environment where the costs of producing children who will be able to function effectively in the adult economy are high (due to the high costs of education and/or of placing children in a socio-economically advantageous position), parents will reduce family size to that in which they can realistically afford to invest (Mace 1998). Nettle (2008) suggested that women from lower socioeconomic classes might be content to match the reproductive outputs of higher SES women rather than try to out-compete them, in part because paternal investment (of which father absence is a common, though not necessarily the only, component) exhibits a marked socio-economic class gradient. This may add to the economic costs of rearing for mothers in lower SES classes, since the burden of rearing will fall disproportionately on their shoulders and may encourage them to favour a satisficing rather than a maximizing reproductive strategy.

If so, then the higher costs of reproduction experienced by individuals in the lower socio-economic classes is likely to mean that, in order to achieve their reproductive goal, women in these classes will have to begin reproduction earlier, and/or continue reproducing later, in order to arrive at the same final family size as women in the higher social classes. We thus hypothesise that the observed differences in age at first reproduction between socioeconomic classes may become part of a strategy designed to achieve the same desired family size under different constraints.

This could only be the case, however, if it is true that early reproduction does not increase the risks of unsuccessful reproductive events, when socioeconomic status is controlled for. Data for first pregnancies from the British Cohort Study confirm that maternal age currently has no predictive power for a successful outcome for first pregnancy when social class is a covariate (N = 9475: β=-0.001, Wald = 0.010, df = 1, p = 0.920). This is consistent with much recent work showing that the negative effects of young maternal age have previously been exaggerated by a failure to consider the confounding effects of parity and social class (Arif et al. 1998; Malik et al. 1997; Reichman & Pagnini 1997).

In order to test the prediction that the higher mortality and unsuccessful pregnancy rates experienced by women in the lower social classes would result in lower lifetime reproductive success compared to women in the higher social classes if they delayed age at first reproduction, we developed a simulation model using class-specific mortality and fertility data to calculate the probability of women surviving and reproducing in each year between the ages of 15 and 45 years. Our aim is to ask whether social class differences in age at first reproduction could be due mainly to class differences in fertility and mortality, and individual women’s attempts to optimise completed family size (i.e. lifetime reproductive output) under different constraints. We test the model against the observed class differentials in age at first reproduction.

Our model is a Monte Carlo simulation of women’s reproductive life-histories for women who choose to commence reproduction in each of the odd-numbered years between ages 15 and 45, for each of four socioeconomic classes (I + II, IIIN, IIIM and IV + V) as defined by the UK Registrar General. We ran 1000 simulations for each cohort (a total of 64,000 simulated individual lifehistories in all). The equations that set the event probabilities at each life stage are given in Table 1. These equations were directly obtained from contemporary UK national statistics databases (ONS 2001). The Health Survey for England and the British Cohort Study were supplied by the UK Data Archives held at the University of Essex. Mortality Statistics for the UK 1998 and the Health Inequalities: Supplementary Dataset 1987–1991 were supplied by the Office of National Statistics.

Table 1

**Parameter values used in the model.**
Probability of:	Social class I&II	Social class IIIN	Social class IIIM	Social class IV&V
Infertility ⁱ	0.1466	0.1691	0.1705	0.1806
Conceiving within year (women aged ≤ 25) ⁱⁱ	0.6728	0.6683		0.6658
Conceiving within year (women aged 26–30) ⁱⁱ	0.6827	0.6778		0.6753
Conceiving within year (women aged 31–35) ⁱⁱ	0.5628	0.5587		0.5567
Conceiving within year (women aged ≥ 36) ⁱⁱ	0.4982	0.4946		0.4928
Conception not leading to birth (x = pregnancy number) ⁱⁱⁱ	-0.008x² + 0.089x + 0.037	-0.011 x² + 0.082x + 0.061	-0.003x² + 0.041x + 0.068	-0.006x² + 0.050x + 0.046
Conception not leading to birth (x = interbirth interval, years) ⁱⁱⁱ	0.0194x + 0.779
Offspring dying by age 1 ^iv	0.0119	0.0139	0.0143	0.018
Offspring dying by age 24 ^v	0.0094	0.0092	0.0125	0.010
Female dying within year (x = age, years) ^v	0.0000058x³ -0.001 x² + 0.015x-0.071	0.0000057x³ – 0.001x² + 0.015x-0.070	0.0000054x³ – 0.001x² +0.013x-0.060	0.0000053x³– 0.001x² + 0.013x-0.061
ⁱ Based on a population infertility rate of 16.67% weighted according to the effect of smoking¹⁰ and the proportion of people who smoke in the given social class.
ⁱⁱ Based on the probability of conception according to age ¹⁷ weighted according to the effect of smoking¹⁰ and the proportion of people who smoke in the given social class.
ⁱⁱⁱ Data from The British Cohort Study
^iv Mortality statistics for the UK 1998 (ONS).
^v Health Inequalities: Supplementary Dataset 1987–1991

For each reproductive history, the class-specific probability of infertility is first used to determine whether or not the woman is infertile (with completed family size being 0 if she is). For fertile women, a random number generator was used in conjunction with the probability density functions provided by the class-, age- and birth-order-specific probabilities to determine whether or not (a) the woman conceives that year, (b) conception leads to birth (given natural abortion rates only), (c) the infant survives to become a (potentially) reproductive adult (taken to be age 24 years) and (d) the mother survives to the next age interval, for each successive year from the selected age at first reproduction. Completed family size was taken to be the total number of children that survived to age 24 years accumulated over the woman’s entire reproductive history from age 15 years (in effect, menarche) to either age at death or the onset of menopause (at age 45 years), whichever is the earlier. The flow chart for the model is given in the Supplementary Information (Fig. S1).

Figure 1 plots the net completed family size (lifetime reproductive output) for women in the four SES classes as a function of age at first reproduction. These show that the different rates of mortality, subfertility, and unsuccessful pregnancy between the social classes are great enough to result in significantly different patterns of lifetime reproductive success between women in different social classes who begin reproduction at the same age. There were significant effects of both age at first reproduction and social class on lifetime reproductive output (F_3,3996=92.72, p < 0.001; age F_1,3996=247.75, p < 0.001, class F_1,3996=30.33, p < 0.001). Thus delaying reproduction reduces reproductive output for women in every social class but, crucially, within age categories completed surviving family size is always lower for the women in social class IV/V than it is for women in social classes I/II.

Across the full range of age classes represented in Fig. 1, the mean age difference at which women in social class IV/V achieve net reproductive parity with women in social class I/II is 5.65 years (N = 15 age classes). This corresponds well to the observed pattern: in England and Wales, the mean age at first reproduction between 1970 and 2000 was 28.2 years amongst social classes I & II and 23.6 years amongst social classes IV & V (Fig. 2). The difference between the classes (mean 4.6 years) is highly significant (comparison of annual means: F_1,60=74.8, p < 0.001). The difference in mean age at first reproduction between the classes predicted by the model is not significantly different from the observed value (z = 1.270, p = 0.204). In other words, women in social class IV/V have to start reproduction up to half a decade earlier than their counterparts in social class I/II if they are to gain parity with them in completed family size. More importantly, in order to out-compete women in the higher social classes, these women would either need to begin reproduction earlier still (i.e. as late teenagers: Fig. 2) or continue reproducing much later.

We have shown that, within the contemporary UK population, women in lower socio-economic classes must begin to reproduce significantly earlier than women in the higher socio-economic classes if they are to achieve the same completed lifetime family sizes (i.e. fitness) as these women. This is because the cumulative effects of higher rates of infertility, reproductive failure and infant mortality suffered by women in the lower socio-economic classes, combined with a higher risk of dying, mean that fewer of their conceptions result in adult offspring (i.e. offspring in a position to contribute to the parents’ fitness by producing grandchildren). The crucial finding is that these women adjust the onset of reproduction by exactly the amount required to match the lifetime reproductive outputs of better off women. Failure to do so results in significantly lower fitness. Essentially similar results were obtained from an analysis of Finnish historical demography records: if age at first reproduction is held constant, women from higher socio-economic classes out-reproduce less well-off women (Liu et al. 2012), just as is the case in our UK data.

The conventional view in demography and social policy has tended to be that early reproduction and pronatalist attitudes are largely a consequence of differences in education or attempts to access social resources (e.g. housing) (Peckham 1993; Cleland 1995; Franklin & Corcoran 2000; Geronimus 2003; Duncan 2007). However, while education and educational opportunities may be a precipitating factor, our results point to the importance of a strategic lifehistory perspective in understanding human reproductive behavior. They suggest that the higher birth rates and earlier onset of reproduction characteristic of some women may not simply be a consequence of pronatalist attitudes. Instead, earlier reproduction might actually reflect a sensitive strategic response to class-specific risks of reproductive failure in a context where the great majority of women of all classes are in fact aiming at much the same target completed family size. In other words, early reproduction is not a mistake, but an adaptive strategy (Nettle 2010). Indeed, Nettle (2010) offers a compelling argument for the fact the causal logic being that economic circumstances affect lifehistory traits and these in turn affect age at first reproduction rather than the other way around (as has been commonly assumed).

Since the decision about when to start reproduction is made well in advance of any decision about when to cease reproduction, we interpret it as reflecting decisions that women make about whether they can afford to delay the onset of reproduction in order to invest in careers in the light of the effect that career-dependent movement from one social class to another may have on the lifehistory parameters with which they will subsequently have to cope once they start reproducing. In effect, women with limited expectations of future career opportunities (mainly those in lower socieconomic classes whose educational opportunities are limited) should prefer to opt out of career-based life trajectories in favour of early reproduction. Doing so provides them with significantly higher fitness than the opportunity costs offered by trying to compete in an economic market for which they are socially or educationally ill-equipped. Unless they are able to use career opportunities to move into a higher socio-economic class so as to benefit from the reproductive and health advantages of doing so, they will incur a significant cost in terms of lost reproductive opportunities that they will find difficult to recoup. Since success in this market is largely education-driven, poor educational opportunities will invariably constrain these individuals to a best-of-a-bad-job solution (namely, early reproduction).

Belski et al. (2010) found, in a US sample, that degree of maternal harshness when the child was aged 54 months correlated strongly with earlier age at menarche and greater sexual risk-taking during teenage years in daughters. Though they did not consider parental socio-economic status, it is plausible to suggest that this was at least partially correlated with rearing behaviour (Conger & Donnellan 2007); in any case such harshness may well be predictive of maternal support during the daughter’s future reproductive life.

There have been a number of attempts in the social sciences to link late onset of reproduction and delayed marriage to increasing difficulty in finding suitable partners (assortative mating) once women embark on career-oriented life strategies, and in particular the effect that the prolonged education necessary for careers has on the time at which women finally enter the workplace (Becker 1981; Oppenheimer 1988). Nettle & Pollet (2008), for example, found in a UK sample that completed family size (lifetime number of children) declined with income in women (but increased in men), mainly due to increased rates of childlessness in high-income career women. Lawson & Mace (2011), however, caution that there may be benefits to having fewer children for wealthier families in terms of parental investment in future social and economic opportunities. One of those benefits, as we suggest here, may be reduced childhood morbidity and mortality.

However, whereas delayed marriage (and hence delayed reproduction) is often viewed as an unfortunate by-product of education and career opportunities (with potentially adverse consequences at both individual and societal levels), it may, in fact, involve strategic decisions that individuals make in the light of their circumstances in a deliberate attempt to optimise fitness under the particular constraints of their economic and social circumstances. As in many such cases, the choice may be between a low variance, low risk strategy and a high variance, high risk strategy (e.g. peace versus war chiefs among the Cheyenne: Dunbar 1991). Those who can afford to do take the risk opt for the reproductively high risk strategy (where the main risk is failure to reproduce), those who cannot opt for the safer low risk strategy. In an evolutionarily stable strategy set, the mean fitness payoffs should be equal – as they seem to be in the present case.

These results suggest that women’s reproductive strategies may be more subtle and have a much longer time perspective than is often assumed. Implicitly, these decisions are in effect based on fitness-matching (i.e. women trying to match an optimal family size produced by those in the better-off social groups) in order to avoid being disadvantaged in fitness terms. Producing the same numbers of offspring, who in turn will in due course have the same socio-economic choices (depending on their own educational achievements), means that they are not evolutionarily disadvantaged against women in better-off social groups. This suggests that educational opportunities and achievement, and the effect that these have on career opportunities, play a particularly crucial role in women’s strategic decisions in the contemporary world.

Author Contribution

SG conceived the study and collated the data; RD constructed the model; SG and RD analsed the data and wrote the paper.

Acknowledgments.

SG was supported by an ESRC Postgraduate Research Studentship. RD was supported by a European Research Council Advanced Grant (#295663).

Arif, M., Qureshi, A., Jafarey, S., Alam, S., & Arif, K. (1998). Maternal sociocultural status: A novel assessment of risk for the birth of small for gestational age, low birth weight infants. Journal of Obstetrical and Gynaecolical Research, 24, 215–222.
Becker, G. S. (1981). A Treatise on the Family. Harvard University Press.
Becker, G., & Lewis, H. (1973). Interaction between quantity and quality of children. In T. Schultz (Ed.), Economics of the Family: Marriage, Children and Human Capital. Chicago University Press.
Bielby, J., Mace, G. M., Bininda-Emonds, O. R. P., Cardillo, M., Gittleman, J. L., Jones, K. E., Orme, C. D. L., & Purvis, A. (2007). The fast-slow continuum in mammalian life history: an empirical reevaluation. American Naturalist, 169, 748–757.
Brooks-Gunn, J., Duncan, G. J., Klebanov, P. K., & Sealand, N. (1993). Do neighbourhoods influence child and adolescent development? American Journal of Sociology, 99, 353–395.
Charnov, E. L. (1991). Evolution of life history variation among female mammals. Proceedings of the National Academy of Sciences, USA, 88, 1134–1137.
Cleland, J. (1995). Obstacles to fertility decline in developing countries. In R. I. M. Dunbar (Ed.), Reproductive Decisions: Biological and Social Perspectives (pp. 207–229). Macmillan.
Conger, R. D., & Donnellan, M. B. (2007). An interactionist perspective on the socioeconomic context of human development. Annual Review of Psychology, 58, 175–199.
Dubois, L., & Girard, M. (2006). Determinants of birthweight inequalities: population-based study. Pediatrics International, 48, 470–478.
Dunbar, R. I. M. (1991). Sociobiological theory and the Cheyenne case. Current Anthropology, 32, 169–173.
Duncan, S. (2007). What’s the problem with teenage parents? And what’s the problem with policy? Critical Social Policy, 27, 307–334.
Franklin, C., & Corcoran, J. (2000). Preventing adolescent pregnancy: a review of programs and practices. Social Work, 45, 40–52.
Geronimus, A. T. (1996). What teen mothers know. Human Nature, 7, 323–352.
Geronimus, A. T. (2003). Damned if you do: culture, identity, privilege, and teenage childbearing in the United States. Social Science Medicine, 57, 881–893.
Geronimus, A. T., Bound, J., & Waidmann, T. A. (1999). Health inequality and population variation in fertility-timing. Social Science Medicine, 49, 1623–1636.
Guegan, J. F., Thomas, F., Hochberg, M. E., de Meeus, T., & Renaud, F. (2001). Disease diversity and human fertility. Evolution, 55, 1308–1314.
Kohlhuber, M., Rebhan, B., Schwegler, U., Koletzko, B., & Fromme, H. (2008). Breastfeeding rates and duration in Germany: a Bavarian cohort study. British Journal of Nutrition, 99, 1127–1132.
Lawson, D. W., & Mace, R. A. (2011). Parental investment and the optimization of human family size. Philosophical Transactions of the Royal Society London, 366B, 333–343.
Liu, J., Rotkirch, A., & Lummaa, V. (2012). Maternal risk of breeding failure remained low throughout the demographic transitions in fertility and age at first reproduction in Finland. Plos One, 7, e34898.
Low, B. S., Hazel, A., Parker, N., & Welch, K. B. (2008). Influences of women’s reproductive lives: unexpected ecological underpinnings. Cross-Cultural Research, 42, 201–219.
Lycett, J., & Dunbar, R. I. M. (2000). Abortion rates reflect the optimization of parental investment strategies. Proceedings of the Royal Society, London, 266B, 2355–2358.
Mace, R. (1998). The coevolution of human fertility and wealth inheritance strategies. Philosophical Transactions of the Royal Society London, 353B, 389–397.
McCulloch, A. (2001). Teenage childbearing in Great Britain and the spatial concentration of poverty households. Journal of Epidemiological and Community Health, 50, 16–23.
Malik, S., Ghidiyal, R., Udani, R., & Wainganker, P. (1997). Maternal biosocial factors affecting low birth weight. Indian Journal of Pediatrics, 64, 373–377.
Mortensen, L. H., Diderichsen, F., Arntzen, A., Gissler, M., Cnattingius, S., Schnor, O., Davey-Smith, G., & Andersen, A. M. N. (2008). Social inequality in fetal growth: a comparative study of Denmark, Finland, Norway and Sweden in the period 1981–2000. Journal of Epidemiological and Community Health, 62, 325–331.
Nettle, D. (2008). Why do some dads get more involved than others? Evidence from a large British cohort. Evolution and Human Behavior, 29, 416–423.
Nettle, D. (2010). Dying young and living fast: variation in life history across English neighborhoods. Behavioral Ecology, 10, 387–395.
Nettle, D. (2011). Flexibility in reproductive timing of human females: Integrating ultimate and proximate explanations. Philosophical Transactions of the Royal Society London, 366B, 357–365.
Nettle, D., & Cockerill, M. (2010). Development of social variation in reproductive schedules: a study from an English urban area. Plos One, 5, e12690.
Nettle, D., & Pollet, T. V. (2008). Natural selection on male wealth in humans. American Naturalist, 172, 658–666.
ONS. (2001). Births: 1970–2000, Mean ages of women at live births within marriage (according to social class of husband) and birth order. Office of National Statistics.
Oppenheimer, V. K. (1988). A theory of marriage timing. American Journal of Sociology, 94, 563–591.
Quinlan, R. J. (2007). Human parental effort and environmental risk. Proceedings of the Royal Society, London, 274B,121–125.
Peckham, S. (1993). Preventing unintended teenage pregnancies. Public Health, 107, 125–133.
Reichman, N. E., & Pagnini, D. L. (1997). Maternal age and birth outcomes: Data from New Jersey. Family Planning Perspectives, 29, 268–272.
Rogers, A. R. (1990). Evolutionary economics of human reproduction. Ethology and Sociobiology, 11, 479–495.
Smith, D. M., & Elander, J. (2006). The effects of area and individual disadvantage on behavioural risk factors for teenage pregnancy. Psychology Health and Medicine, 11, 399–410.
Stearns, S. C. (1992). The evolution of life histories. Oxford University Press.
Walker, R., Gurven, M., Hill, K., Migliano, H., Chagnon, N., De Souza, R., Djurovic, G., Hames, R., Hurtado, A. M., Kaplan, H., et al. (2006). Growth rates and life histories in twenty-two small-scale societies. American Journal of Human Biology, 18, 295–311.
Wilson, M., & Daly, M. (1997). Life expectancy, economic inequality, homicide, and reproductive timing in Chicago neighbourhoods. British Medical Journal, 314, 1271–1274.

No competing interests reported.

GraingerDunbarSI.pdf

Download PDF

Journal Publication

published 26 Mar, 2024

Read the published version in Adaptive Human Behavior and Physiology →

Editorial decision: Revision requested
20 Feb, 2024
Reviews received at journal
05 Feb, 2024
Reviewers agreed at journal
24 Jan, 2024
Reviewers agreed at journal
17 Jan, 2024
Reviewers invited by journal
17 Jan, 2024
Editor assigned by journal
17 Jan, 2024
Submission checks completed at journal
17 Jan, 2024
First submitted to journal
15 Jan, 2024

You are reading this latest preprint version

Lifehistory Trade-Offs Influence Women’s Reproductive Strategies

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Methods

Results

Discussion

Declarations

Author Contribution

Acknowledgments.

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1