A CONTEmpOrAry issUE iN dEmOGrAphy: ThE risiNG AGE AT firsT birTh, prOs ANd CONs 1

two decades after aleš Hrdlička’s death the introduction of effective contraceptives broke the evolutionary link between sexuality and procreation. since then we decide about having children or not, and if we want them we can also decide about their timing. as a consequence the number of children declined to numbers that fit modern life styles and the age at first birth has risen considerably. However having children remains a chance, every month again. For the majority having children goes without problems worth speaking of. But a minority risks the chance of remaining without children or with only one child as they started to try having children ‘too late’. this contribution overviews the backgrounds and the pros and cons of the rising age at first birth.


introduction
at the time of anthropologist aleš hrdlička  one was probably not very aware of an issue which is essential now in human reproduction: with the introduction of effective contraceptives the evolutionary link between sexuality and procreation completely broke.now people consider whether or not to take children, they do not get them anymore, but carefully plan to have them.so people weight all the pros and cons of having children, and as long as they are undecided they postpone.
This indeed is an intriguing contemporary issue in demography: families are not only much smaller than in the past but couples decide more explicitly on having children much later in life, if at all.This article will show not only the geographical variation in the changing age pattern at first birth within europe over de past decades but also touch on the pros and cons of postponement behaviour, its determinants and consequences, the (theoretical) interrelation between late and low fertility, as well as on issues around gender inequality.at the end some provocative statements are made on policies towards a more optimal life course.in such a life course motherhood and fatherhood may have another meaning, another life fulfilment and therefore be more ideally timed.
postponement behaviour is a difficult concept to grasp as it does not easily fit individual behaviour on the microlevel.a woman (man) may have a specific plan on the number and timing of children but will not perceive this plan with a strict reference point in time from which to deviate wittingly.however at the macro-level women born in a certain year may show a later start with having children than women born earlier.demographers will then notice, again at the macro-level, that a later start normally goes together with higher levels of childlessness, with higher shares of small families, and with changes in birth intervals.at the micro-level women often have their first baby at a higher age than their mothers did.

the semantics of postponement behaviour
one of the main postponement indicators used in demographic research is the mean age of the mother at first birth.over the past decades this indicator shows a rising trend almost everywhere in Western societies.even in developing countries such an upward trend is observed.it not only means that women have their first child later in life but also that they are without children for a longer period of early adulthood than previous generations were: a longer period in which they can concentrate on other important life fulfilling commitments.and also the age at first fatherhood rose.statistics sweden shows evidence (Figure 1) on the fact that the age at first fatherhood almost perfectly follows the rise of the age at first motherhood with a difference of about 2.5 years.so where in this article motherhood is written, the reader can practically always easily exchange that for fatherhood as well: later parenthood thus.do all countries have this indicator on the onset of childbearing, and if so is it comparable across countries?When analysing data on age at first birth one realises that it is far from easy to find comparable data: A CONTEmpOrAry issUE iN dEmOGrAphy: ThE risiNG AGE AT firsT birTh, prOs ANd CONs1 gi js Be et s netherlands interdisciplinary demographic institute (nidi)

AbstrAct
two decades after aleš Hrdlička's death the introduction of effective contraceptives broke the evolutionary link between sexuality and procreation.since then we decide about having children or not, and if we want them we can also decide about their timing.as a consequence the number of children declined to numbers that fit modern life styles and the age at first birth has risen considerably.However having children remains a chance, every month again.For the majority having children goes without problems worth speaking of.But a minority risks the chance of remaining without children or with only one child as they started to try having children 'too late' .this contribution overviews the backgrounds and the pros and cons of the rising age at first birth.
To be quoted as: beets, G. ( 2011): A contemporary issue in demography: The rising age at first birth, pros and cons AUC Geographica, 46, No. 2, pp.5-14 1.The definition of birth order may differ; it turns out that there are two definitions: a first child to the mother, which is the one we want to obtain, versus a first child to the marriage.as some people marry more than once they may have a 'first birth to the marriage' in each of their marriages.countries with such marriage definitions in their statistics are excluded from this research, and unfortunately some important countries belong to this group (Belgium, France, germany, luxembourg, switzerland, united kingdom).most likely this 'marriage definition' in birth order registration stems from times when non-married fertility was 'not appreciated' and more or less negligible.however here we are explicitly interested in the first biological child to the mother.2. small definitional variation may occur as to number of women delivering versus number of children born, i.e. due to multiple births more children are born than women delivering.additionally small variation may exist as to whether to include only children born alive or also those who are not born alive (still-born babies).3.There usually are two different definitions of age: age in completed years (actual ones, i.e. the exact age of the mother at the day she delivers her baby) or age reached during the year (i.e.age as at 31 december of a given year).4. Finally, obtaining absolute or relative figures may make a substantial difference in the analysis.mean ages are usually computed from age-specific fertility rates (births per all women of a specific age) although fertility tables with age-specific data on first children born to childless women would provide a more precise indication.

trends in postponement behaviour
The significant increase over the past few decades in the age at which women became mother for the first time, i.e. made the transition away from childlessness, represents a major change in demographic behaviour.and as said, later motherhood than before is not only observed now in western, but also in many non-western countries.The contraceptive pill, introduced in the 1960s, gave increasingly larger numbers of women and their partners the possibility to prevent becoming pregnant at younger ages and to have their first child when they felt prepared to give it a warm place.
From Figure 2 we can judge that the age at first birth used to be somewhat higher in the first half of the previous century than after the second World War. it topped at about 27 years in the 1930s, but had dropped to 24 around 1970, at least in the netherlands.since then many western countries observed a switch towards having the first child increasingly later in life.ups and downs in the curve over the past century reflect the age at first marriage, economic prosperity, uncertain prospects, and separation of married partners during war periods.since the 1970s modern contraceptives, rising educational levels and new (non-marital) life styles led to an unprecedented rise in the age at first birth.Figure 3 shows some of the geographical variation in the mean age at first (biological) motherhood from 1960 onwards.curves are shown for eu-15, the old eu, and for nms-12, the 21st century new member states, as well as for the czech republic, the netherlands, spain, sweden and the united states of america.nms-12 follows the same direction as eu-15 had gone already earlier but at a different pace.For several years the highest values were observed in the netherlands, but currently the 'world champion' is spain; the lowest eu values stem mainly from Bulgaria and romania, but the early childbearing pattern, at least for european standards, used to be characteristic also for the czech republic.
it is immediately clear that the age at first motherhood has increased substantially since around 1970. at that time it was for women in europe 'normal' to have the first child when they were between 22 to 26 years, with an average of 24 years.currently the range is slightly larger, in between 24 and almost 30, with an average of 28 years.so, the average rose with 4 age years, while the age range widened from 4 to about 6 years.it points to convergence in direction but to divergence in detail.The divergence can partly be 'blamed' to geographical variation: central and eastern europe did not show significant changes before the fall of the Berlin Wall, but only after.
it is mainly time lags in the same trend direction that explain the divergence in detail. in a way it resembles a tsunami that tumbles over us: data on the average mother's age at first birth for the former german democratic republic show immediately after the fall of the Berlin Wall a drop, as only a young and small selection of the female population still had a child in those years, while those who normally also would have had a child started to postpone, the forerunners; it is like the retreat of water before the large tsunami wave strikes. in following years also the younger ones postponed and a giant rise in the age at first birth resulted, as well as a significant decline in fertility rates.
in the usa the age at first motherhood is substantially lower than in most of europe.however the trend is the same in the 1970s and 1980s but the pace of increase is seemingly slower as from the 1990s.The low us age is mainly related to the prominent role teenage pregnancies continue to play; the us teenage rates are larger than those in Bulgaria, romania or the united kingdom, the eu top-3.low teenage pregnancy rates, like in the netherlands, are essential to the explanation of the relatively high mother's age at first birth.
What happened in detail over time with the agespecific fertility rates for first live-born children?The example is again from the netherlands where we follow at which single age becoming a mother for the first time was most prominent.to age 27 in 1986-1989, age 28 in 1990-1992, age 29 in 1993-1996, even age 30 in 1997-2002, and since 2003 age 29 and age 30 alternate with the highest share.The top for one single age nowadays does not reach higher than only 7%. it means that substantially more first babies have a mother who nowadays is over 30 (45% of nl babies) than before (up from 13% around 1970), and 12% of the babies even have a mother over 35 (Figure 4).so far we looked at women delivering a first baby in a specific calendar year.They happened to do so at various ages.however, it actually is more meaningful to analyse fertility rates for women born in the same year, i.e. we follow their life course and compare these with earlier or later born women (Figure 5). in the netherlands around 70% of the women born in 1945 had their first (biological) baby by age 25 and ultimately 88% ever made the transition to motherhood.Women born in 1975 approach the 70% border by age 33 and we expect that only 81% or so will ever deliver a first baby.it means that childlessness increased from 12 to around 19%.
if up until the fall of the Berlin Wall the netherlands (nl) resembled the group of countries with a late first childbearing pattern, then the czech republic resembled the early pattern (Figure 6).since the early 1990s the nl pattern did not change much -the nl pattern changed in the 1970s and 1980s -and obviously the age at first birth is stalling now, but cz changed significantly: early childbearing has disappeared!The variation between cz and nl has become much smaller.however (not shown in the graph) the cz total fertility rate (the mean number of children per woman) is much smaller than in nl: in 2008 it was 1.50 in cz as against 1.77 in nl.how comes? it is related to less children being born in years in which women massively postpone: take the (unlikely) case that all women would postpone to have their child during exactly one calendar year, then a calendar year without any child born would be the result.however if all these women would have their child in the next year together with women born themselves one year later who happen not to postpone then an enormous baby boom would occur, of about twice the normal size.it means that changes in the age pattern may easily give rise to waves in fertility statistics.We just already came upon a tsunami effect, but we can also find weather and climate effects: the yearly numbers and rates of children born resemble the weather, the life course numbers and rates the climate.compare also what happens on roads: it makes an enormous difference for the emergence of traffic jams whether 1000 cars per hour enter a road all at the same time or enter evenly spread over that hour. in the first case there are too many at the same time at the same spot causing a massive jam (boom) while the rest of the road may be empty; in the second case each car may be able to enter and drive smoothly at maximum speed (bust).
Whether a woman has her two children at age 25 and 28 or at age 27 and 30 does likely not make much difference for herself, and certainly not for her ultimate family size.But while this may not be much of an issue at the personal micro level, the societal macro level is confronted with first a retreat of births (baby bust) followed by a baby boom if ultimately all women would have the same number as they otherwise would have had.Birth waves are normal when changes occur in the number and/or timing of having children.if the life time number of children stays equal and only the timing changes there is demographically spoken 'not much of a problem' in the long run.cz is currently still in the midst of the postponement transition making that the actual fertility rates are low.to many the cz rates are even surprisingly low; politicians even prefer to say 'dramatically low' .But they have been low also in nl when that country was in the same stage of the postponement transition as cz is in now.however, since the age at first birth is stalling now nl fertility rates are higher up again, making the yearly number of children almost equal to the ultimate number of children women have per birth cohort, i.e. in their life course.

Pros and cons
postponement of the first child is a logical outcome of better planning, and modern contraceptives can perfectly help people to fulfil their wish to have children at the best possible moment in life.luckily we see nowadays larger shares of wanted children, and obviously the birth of unwanted children has disappeared (although some still occur).people still value to have children very much, but not too many, while specifically among the higher educated voluntary childlessness is rising.teenage motherhood is hopefully soon to disappear, because these children risk growing up in low-income one-parent families which make their future grim.also from that perspective postponement is a blessing, because it is much easier for women to reach a higher level of education when still childless -and education is nowadays typically a period in the life course where having children is rare.higher education advances a better socio-economic equipment to run a family when one feels ready for that.so it leads to more labour participation and higher tax income while still childless, and it makes that people are much more mature and considerate when they start a family.
In short, from a personal and socio-economic perspective: the later the first child comes, the better.
But on the other side, there are several cons.an unexpected consequence of perfect contraceptive behaviour is that partner selection has changed: if there is no pregnancy risk one can invest more in finding the perfect partner.and that goes together with doubts about whether the current one is the best 'prince on the white horse' .increasingly more couples start cohabiting without marriage.not only many cohabiters but also quite a few just married split up before having children. it means that, different from earlier generations, many more women and men -and most of them already had one or more partners -are single at the moment they want to have children.having children without a partner, although possible nowadays and easier to arrange for women than for men, is rare practice.also increasingly more people with all kinds of doubts about having children are observed: will my children be nice, will i be a good parent, can i manage the large responsibility, do i want to give up the freedom i have now, how do i combine all the commitments i already have?in earlier days such dilemmas were never considered as children just 'arrived from god' .
and the longer we wait/postpone, the higher the risk of not realising the preferred family size, the higher the risk of negative health outcomes (for mother and child), and the higher the risk of having to rely on assisted reproductive technology (art), mainly in vitro fertilisation (ivF).many may think that ivF is the ultimate solution and that everyone can easily have a child via this procedure, but unfortunately the population is not very well informed.ivF is a stressful and risky undertaking, and often donor eggs or semen are used.people may think, because taking the pill is such an easy medium to prevent becoming pregnant, that getting pregnant is equally easy as soon as using the pill is stopped.however, starting a pregnancy is a monthly chance depending on various factors, which may easily change and lower with increasing age, also for men: lower chances for getting pregnant, the so called time-to-pregnancy, as well as lower chances for bringing a pregnancy to term and having a healthy baby.ivF will first and for all remain a therapy for those who have a proven reproductive dysfunction, and the earlier such people are assisted the better their chances.offering ivF to young women without reproductive dysfunction, as has been suggested for example for combating population ageing, is 'a waste of money and energy' , as these women have fairly good chances to conceive naturally, although maybe a few months later then (habbema et al. 2009).
In short, from the health perspective: the sooner the first child comes, the better.But there is an effect of education and wealth: those who are higher educated and/or better off have comparatively better fertility chances (as well as higher life expectancy) than those of lower education/the poor (esveldt et al. 2001).

Determinants
What reasons do people mention for postponement behaviour?Besides partner selection (not having a partner) Education is mentioned: the higher the education, the later the first child arrives.educational expansion explains about half the increase of the mean age at motherhood over the past decades, at least in the netherlands.next to that it is labour market participation; not yet ready for children (too much responsibility); doubting about having children or not; financial concerns (like heavy mortgage on two incomes); view on having children does not match with partner's view; pregnancy did not start yet although I am trying; too little child care facilities (statistics netherlands 2009).This all seems related to what is now called the second demographic transition theory (initially proposed by ron lesthaeghe & dirk van de kaa 1986).This theory suggests that processes of modernisation, secularisation and individualisation in Western societies have reduced people's inclination to adhere to normative guidelines, and have increased the value of individual autonomy.as children are seen as impinging on autonomy, individuals will only choose to have children if the responsibilities can be accepted and if having children contributes to their 'self ' .postponement of having children is an easy way out as long as people perceive that otherwise their individual autonomy is seriously diminished.
In short: whether to have children or not is a very difficult decision to make, moreover an irreversible choice with which it is not possible to experiment -an unknown adventure.it will give many the feeling that whatever you choose, you will loose (at least loose something).no wonder that many struggle with how to make their own rational choice.
We know that people not having children have specific reasons for remaining voluntary childless (statistics netherlands 2009): insufficient time for other important issues / commitments (specifically paid work); children keep you from working outside home; having children is a heavy load / too much responsibility; concerns about the future (overpopulation, environment); concerns about not being a good parent / not being able to raise children; do not have a partner / partner already has children and does not want to have more.
In short: this group of people will not easily change their mind.
The basic determinants for having the first child late seem to be: -rise in educational levels -rise in female labour market participation -changing norms and values on family behaviour, also because of secularisation and immigration -drop in having a sustainable socio-economic position, and in feeling ready and secure.some people may regret their own behaviour later on, although it is always difficult to admit your own mistakes.however, according to a nl sample survey 30% of the women who had their child early (before 25) had preferred to have had it later, 22% of the women who had it after age 30 had preferred to have had it earlier, while 90% of those who had their first child while between 25 and 30 years were very much content (esveldt et al. 2001).so obviously the most preferred age range to have the first child lays somewhere in between 25 and 30 years.
But late parenthood also has advantages: people feel physically and emotionally more prepared for parenthood; they feel more quiet and flexible; they have more deliberately chosen for children; early parenthood and family instability are strongly related: fewer children with young parents (than with older parents) live with both natural parents; late parents have a higher educational level, have more financial security; and as long as childless women are employed, their contribution to the labour market and to tax incomes is much larger (Beets & verloove-vanhorick 1992).

consequences
in this short overview on the consequences we should stress that the disadvantages for mothers are basically in health: longer 'waiting-times-to-conception' , more (stressful) assisted reproductive technology, more miscarriages, more multiple pregnancies, more gestational complications, more caesarean deliveries, more breast cancer later in life.and, all in all, that leads to more hospitalizations, or to say it more bluntly: it is much more costly.
The disadvantages for children are more stillborn children, higher perinatal mortality, higher infant mortality, more immaturity/prematurity, more multiple live births, more congenital malformations, and especially for immaturely born children: more physical and mental handicaps.in short, it means that the next generation is slightly less healthy, although focussed prenatal diagnosis may help the seemingly most unhealthy babies not to be born.
But there are also other consequences.For example in demography, where societies are faced with a slightly lower number of children, more childlessness, a lower number of brothers and sisters, a possibly higher number of daughters with fecundity problems and a possibly lower life expectancy than otherwise had been the casethese two reasons are related to possibly still unknown longer-term effects of assisted reproductive technology -, a larger intergenerational interval as not only the age at parenthood increases but also the age at grandparenthood. in total it leads to a lower population growth rate, and to extra population ageing.
several other issues are partly related with the previous: more absence from the labour market due to 'illness' (when treated for fertility problems); more children enrolled in special education; a lower chance for children to share a specific (substantial) number of years together with their grandparents; grandparents cannot easily look after grandchildren; children sometimes refer later in life to having missed the optimal physical capacities of their parents; children may be embarrassed about the 'old looks' of their parentspeers may even think that they are the grandparents -, and children may have more fears about a possible early death of their parents.
In short, there is a whole list of consequences, some are costly, some add to public finances, some are difficult or even impossible to express in financial terms, like for example becoming a grandparent later in life than otherwise could have been the case.economists ever tried to calculate what is, financially seen, the best age to have the first child, but the exercise remained incomplete as too many issues could not be incorporated, because data were lacking or causal relationships were unknown.

Gender inequality
governments strive for health and happiness for all, and emancipation directs towards gender equality.But what exactly does that mean?
if the complete population is well educated and participates on the labour market this may be seen as a society close to ideal.But an ideal society also needs an optimal, next generation.replacement level fertility is seen as ideal -under zero migration -as that ultimately leads to a constant population size and age structure, which is also beneficial for planning purposes.
teenage pregnancies are perceived as far from ideal, and so are broken families and lone mothers.The societal context points towards postponement, but late fertility (a mean age of the mother at first birth at 30 years or over) is beyond the biological optimum.and here we come to the point that gender equality may be a perfect goal from the socio-economic perspective, but can never be a goal from the biological point of view.
our brains are sexually differentiated.male and female brains are not similar, are fixed during prenatal and early development, and will not lend themselves for a completely equal division of tasks between men and women in the family or on the labour market.men and women differ in their feelings and behaviour (see specifically the chapter by swaab in Beets et al. 2011).Women are less power oriented, less risk taking, more often on the safe side, more careful, more empathic, more oriented towards good and long-lasting relationships -not only privately but also in their professional life.Women are also more family oriented than men are.This world is rather masculine, designed by males.many women obviously feel less at home in this world, as they want the best of both worlds: they do not object at all to labour market participation, but they want it compatible with a nice family.our current western society is considered to be too child-unfriendly, men are too little interested in family and household businesses.Women have adapted much more to men's world than men have adapted to women's (see specifically the chapters by hakim, schippers, te velde, van doorne-huiskes and by van hooff in Beets et al. 2011).
although the society is becoming more feminine we may wonder how tomorrow's society will deviate from today's?Will it be more optimal, and will it make a more perfect socio-economic, as well as health and demographic performance?

Policies
no country has found yet the perfect way of offering optimal life courses to all citizens.The nordic countries seem to be closest with turning into a world where everyone is happy, healthy, and wealthy.and that may have been related with the high level of gender equality.
What kind of policy system could bring us further? in a recently published book (Beets et al. 2011), based on an international expert-meeting organised at the netherlands interdisciplinary demographic institute, the idea of a 'cafeteria' system was developed: like people can compose their own meal by picking in the cafeteria whatever ingredient is preferred, citizens should also be able to compose their own life by picking up a preferred coherent set of 'policy measures' .a set that can change over time according to new needs and preferences in their life course.ideally it provides them every month of life with the best possible combination of income and security, housing accommodation, health, social protection and inclusion.it adapts to personal household and family circumstances and preferences of that moment.it directs towards a world that adapts towards individual wishes and preferences, a world away from people being put up with what governments and employers think is best.if we do not make such a revolutionary switch in thinking, a trend towards having the first child (slightly) earlier will turn out to be an illusion.That also has to do with the fact that there are a few other societal trends that lead to later childbearing: as more than ever young adults face the divorce of their parents and such experience does not contribute to stability in their own unions, as age at first motherhood for daughters is positively related to the age her mother had at first birth, as education is still increasing and the higher educated start later with having children than lower educated, as in our ageing society the labour market pressure on women increases, as (luxury) lifestyle wishes and mortgages increase, as also foreign born women postpone -specifically in the second generation.
In short: this all means that without any change the age at first motherhood will increase further. is turning the rising trend of the age at first motherhood a goal within reach then?yes, it may be within reach if labour market contracts are becoming much more dependent on and perfectly compatible with life course choices people make, and not vice versa.
yes, it may be within reach if that also goes together with good information/education on the relation between age and having children.
yes, it may be within reach because a cafeteria-system may save the time spent on discussions with the partner on how to arrange life when children are born.dual earners seem to spend on average 2.5 years on discussions; and that seems to be a waste of time, certainly if the discussers are already up in their 30s.
turning the trend may also become normal practice some day, if children are becoming such a 'scarce item' that mothers will be paid for having and raising children.so, it may be possible to concentrate having the first child much more around the ages of let us say 27-29.

conclusion
We have to accept that evolution provided us, deliberately, with two different sexes.men and women do not have exactly the same roles or tasks in society and we should stop trying to make them do so.and we should not perceive pregnancies and child care as 'problems' but see them as common responsibility for employers, employees and society.long-term thinking should prevail: appreciate the arrival of future employees and stop short-term thinking that only appreciates this year's profit making.
true gender equality must entail that both sexes are equivalent in the sense of having the same value, which is quite different from being the same.true gender equality is accepting and appreciating that both sexes are different in various essential aspects, for example reproduction.in that society both mother-and fatherhood will have another meaning, another life fulfilment, and may be more ideally timed.

Fig. 1 Fig. 2
Fig. 1 Mean age at first birth in sweden, women and men, 1970-2009.source: statistics sweden Fig. 4 Percentage of first births among women of 30, 35 years or over in the netherlands (all first births per calendar year = 100%), 1950-2008.source: statistics netherlands