The Iron Triangle of Family Policy

Population aging is proceeding apace across the world, but there is little agreement on what to do about it or even whether it is a bad thing at all. On the upside, older, smaller societies will have a correspondingly smaller climate impact, will be able to invest more per child, and will suffer less congestion from overcrowding. Further, many of the causes of population aging are for reasons worth celebrating: improved life expectancies, greater opportunities particularly for women, and re­duced old-age poverty. What tends to give people pause, however, is that sustained low birth rates can threaten the viability of social security systems and lead to lower, slower growth rates.¹ Similar to the risks from climate change, an aging, shrinking world threatens to be a place with lower living standards overall.

Encouraging people to have more children is not the only solution. Immigration from young to old countries can surely help. Yet many of the traditional origin countries for immigrants to the United States, Canada, the UK, France, and Germany are already at or below the replacement-level total fertility rate (TFR) of 2.1 children per woman. India (TFR 2.05), Mexico (1.90), China (1.28), Tunisia (2.11), and Tur­key (1.92) will all soon have fewer potential emigrants, and even those countries that are still above replacement have seen their own birth rates trend steadily downwards since the 1960s.² This means that immi­gration can help shore up many developed-world labor forces for the medium term (though, as seen in recent years, large-scale immigration can lead to political unrest and other trade-offs). But, since no country appears immune to eventually settling into sub-replacement fertility, it is inevi­table that societies will have to grapple with the underlying problem.³

Getting to the root of this latter problem may be a more important challenge than anglophone or western European readers anticipate. Many other aging societies face the same growth and social security financing challenges as the United States but are much less willing to use immigration to bolster their economies. Countries like Japan (1.34) and South Korea (0.84!) have evinced little support for immigration, even as they have begun the process of population decline.⁴ Unless these coun­tries either reverse course on immigration or become more successful in increasing their birth rates, then they and their ultra-low fertility peers with high public debt burdens like Italy (TFR 1.24) and Greece (1.34) may trigger a bond crisis that could be financially destabilizing. Another cause for concern is that these U.S. allies may struggle to fill their armies’ ranks, increasing their dependence on U.S. security commitments. For example, the best estimates are that South Korea has about twice the population of North Korea (51.7 versus 26.0 million), but the United Nations estimates that North Korea had nearly 38 percent more births in 2021 than South Korea. Fast forward this dynamic by about twenty years, and South Korea’s birth dearth may very well be a pressing problem for U.S. national security.

Yet, given the scope of the problem, governments have good reasons to avoid doing anything too radical to boost birth rates. Stronger measures may work, but the crux of the problem is that governments want women to do it all: to have more children and to keep working. Any policy that actively encourages one while discouraging the other will be self-defeating. All OECD countries have female employment-to-population ratios well above 0.5 (and several above 0.8). Modern states have thus become completely dependent on the labor and taxes of their female citizens while largely retaining the tax and education structures that were set up back when boys were the default pupil and men were the default taxpayers.

Governments are thus boxed in by three different considerations: fiscal, political, and demographic. Together, these form an “iron tri­angle” of family policy that effectively limits governments’ scope of action. On the fiscal side, governments tempted to throw money at families are confronted with the realities of basic tax accounting. The average cost of a child appears to be quite high: a 2017 U.S. Department of Agriculture estimate for the lifetime cost of raising a child born in 2015 put the figure at $233,610 ($295,357 in 2023 dollars).⁵ By contrast, as of 2021, the federal government can expect to collect $339,173 in federal income taxes over the lifetime of the average citizen. Additionally, the federal government collects $272,000 in FICA taxes from the average person, for a combined sum of $611,173. This means that the U.S. government would have to spend approximately half of its expected lifetime tax receipts if it wanted to fully offset families’ costs of having a child.

Politically, pro-natalist politics are likely to become more difficult as societies age. Older populations in societies with high female labor force participation have a plethora of reasons to resist spending more money on families. For older people, government spending may be viewed increasingly as zero-sum in an age of ballooning old-age support pay­ments.⁶ Further, as stated above, pro-natalist policies that threaten to set back women’s gains in the workplace are sure to be vociferously op­posed, in addition to being fiscally self-defeating.

Lastly, governments will struggle in the face of these realities to target the demographics that appear to have the highest gaps between their ideal and realized number of children: the working and middle classes. A U-shaped relationship between fertility rates and income has emerged in the United States and several other countries, whereby the poorest and richest families appear to be achieving or exceeding their desired fertility targets, but the shortfall has grown in the wide middle. Better access to childcare, increased subsidies, and steps to lower the cost of housing could probably all help the middle and working classes achieve their desired family sizes. All of these will cost money, however, which is something that many governments will find increasingly scarce. Even in Israel, the one wealthy country with above-replacement TFR, demo­graphics are causing significant political and fiscal problems. Much of their population growth is concentrated among the Haredim, or Ultra-Orthodox Jews, who do have very large families, but whose men have lower labor force participation rates than the national average. From a purely fiscal point of view, increased child cash transfers would not necessarily yield the desired return, unless they could be targeted at just the lower-fertility parts of the population, a blatantly discriminatory policy.⁷ Breaking through the iron triangle thus means reexamining old assumptions about how societies arrived at this juncture and considering fresh approaches to help people achieve their ideal family sizes.

The Origins of the Baby Bust: Cui Bono?

A brief review of the history of falling birth rates shows why smaller family sizes have become a hallmark of modern development, and why there are few effective, off-the-shelf policies to change them. In their 2013 book, Why Nations Fail,⁸ Daron Acemoglu and James Robinson relate the story of the Englishman William Lee, who invented an early mechanized knitter that would allow people to make clothes in a frac­tion of the normal time. In 1589, he finally secured an audience with Queen Elizabeth I with the hopes of being issued a royal patent. Sadly, his hopes were crushed, because Elizabeth feared that the machine would destabilize the realm by throwing too many people out of work. Attempts to acquire a patent in France and from Elizabeth’s successor James I were also refused on the same grounds. Acemoglu and Robinson draw on these examples and many others to argue that growth and change can only occur when elites open their political systems to upwardly mobile elite aspirants, reward innovation, and allow for broad‑based economic participation. While their argument that “inclu­sive” economic institutions are at the center of economic growth is not entirely convincing, they do establish a strong case that states and societies thrive when elites are convinced that permitting change is more advantageous than holding on to a stagnant status quo.

Prior to the twentieth century, demographic realities made rulers especially cautious about the unintended consequences of change. Mal­thusian pressures coupled with high mortality rates ensured that most populations were quite young. For example, the median age reported by the 1820 U.S. Census was a mere 16.7. There is no reason to think that median ages in most societies prior to 1800 would have been any higher than the early twenties. Young populations tend to be more prone to rise up in revolt, so a certain economic and political conservativism by ruling elites was a logical, if suboptimal, choice. Now that the majority of humanity lives in rapidly-aging countries with sub-replacement fer­tility, it is worth revisiting and expanding Acemoglu and Robinson’s framework for insights into how most of humanity arrived in this position and where we might go from here.

There are a wide range of reasons why birth rates have declined since the eighteenth century. A non-exhaustive list of culprits includes capital­ism,⁹ secularization,¹⁰ literacy,¹¹ and declining sperm counts,¹² among count­less other reasons. Each of these may (or may not) be contributing, but it is also useful to ask why states tended to adopt policies or encourage trends that would depress birth rates. After all, from the government’s perspective, fewer births mean fewer potential taxpayers and soldiers. So why have changes that tend to lower birth rates like industrialization, child labor laws, and mass literacy been embraced?

Essentially, the Acemoglu and Robinson framework would argue that innovations occur due to greater education, and greater educational attainment leads to higher “quality” taxpayers who are more innovative even if it means fewer taxpayers altogether. These same technological innovations are often socially disruptive, and societies and their gate­keepers must decide whether to allow them to take root, and then how to adapt to them. As an organizational principle, one can think of a social, economic, or technical innovation as a “shock,” a term economists use to describe unanticipated changes. Shocks are not necessarily expected to act uniformly, since how susceptible a society, region, or subgroup is to the shock’s effects is going to depend on preexisting characteristics. For example, a now-canonical example is China’s acces­sion to the World Trade Organization, which prompted the United States to lower its tariff rates on many goods. Since manufacturers facing this new import competition were more concentrated in some communities and not others, job losses were much more heavily concentrated in some areas than others.¹³ Once the gatekeepers agree to let the shock proceed, they then must adjust and adapt as the consequences become appar­ent.

Unfortunately for policymakers, cultural shocks seem to play a role at least as important as economic shocks in driving the long-run decline in birth rates. For example, a convincing body of evidence has shown that fertility began declining in France even before Malthusian conditions began to wane. This may come as a surprise, because mass literacy precipitates secularization in many societies, and literacy and education are popularly linked to fertility decline. Excellent papers by Enrico Spolaore, Romain Wacziarg, and Guillaume Blanc,¹⁴ however, show fertility decline in the modern era began in France as a symptom of dechristianization, and then diffused throughout Europe along lines determined by cultural and linguistic distance from France.¹⁵

Secularization would come later to Protestant countries for reasons not yet well established empirically, but its effects on birth rates seem to be measurable by about the late 1870s. A paper by Brian Beach and W. Walker Hanlon makes the case that the probable direct catalyst was the Bradlaugh-Besant Trial in 1877 in the UK, which publicized effective contraceptive methods for perhaps the first time.¹⁶ Defendants Charles Bradlaugh and Annie Besant were tried for disseminating Fruits of Phi­losophy, a tract with detailed information on contraceptive use. They incidentally received support from Britain’s National Secular Society, incorporated in 1866.¹⁷ Their qualified guilty verdict was overturned on a technicality and Fruits of Philosophy sold widely, particularly in those areas where press coverage of the trial was most intense. Beach and Hanlon’s paper shows that fertility decline diffused along cultural lines toward other anglophone regions in subsequent years.

While secularization appears to have been the first mover, almost all scholars agree that the Industrial Revolution and its attendant economic changes played a central role in both modernization and (ultimately) shrinking family sizes. Industrialization eventually encourages both urbanization and greater educational attainment, both of which tend to increase the cost of the marginal child. Yet once the Industrial Revolution took off and enriched Britain at the expense of its competitor nations, the logic of state competition ensured that other countries would eventually seek to modernize so as not to get left behind. States pursued development, even if it did initially cause unemployment and social dislocation, because those with prosperous economies could col­lect more taxes per capita, which in turn facilitated financing wars, internal improvements, and the development of cutting-edge technologies.

Promoting mass literacy and numeracy are natural extensions of this process. Pace James Scott,¹⁸ literate citizens who can sign forms make the population more legible to the state, and thus are more easily taxed and conscripted. Education allows workers to better use labor-enhanc­ing technologies,¹⁹ but also helps states to make the most of their conscripts. The eventual shift during World War II and the Cold War into explicitly technological forms of state competition—such as atom bombs, spy planes, etc.—again only reinforced the premium the state placed on having an educated citizenry, whose brightest minds could be coaxed (or pressed) into service.

While the benefits of education to the state and to the individual are powerful, they impose their own demands that would revolutionize the life course of the average person and the relationship between the individual and the state. The syllogism runs as follows: children are more educable than adults. Higher targeted levels of education require more time spent in school. As part of the adaptation and adjustment process by societies, childhood was increasingly set off as a special time where parents and the state would share the burden of educating and raising children.²⁰ This meant, however, that the pecuniary payoff to parents from having children would be increasingly delayed. Accordingly, most studies have found that restricting child labor probably de­creased fertility more directly than the additional years of education in and of themselves.²¹

On the other side of the life cycle, modernization demanded a revolution in the relationship between adults and their extended fami­lies. Scientific advances helped tame mortality at every age, but also created a problem where increasingly valuable prime-age, taxpaying adults were having to take care of elderly parents.²² Longer lifespans also made the sharp increase in poverty among the elderly during the Great Depression a more pressing social problem. Formalizing old age as a time for rest from labor by granting a state pension helped to reduce both old-age poverty and prime-age men’s obligation to provide for their elderly parents. One study showed that men whose fathers were early recipients of Social Security had higher mobility and lifetime earnings, although women did not experience similar benefits.²³ This is likely because Social Security obviated sons’ social obligation to provide materially for elderly parents, but not daughters’ social obligation to provide care. Nonetheless, from the state’s point of view in the mid-twentieth century, this was a good bargain.

The demands of state competition were so powerful over this time period that this basic trajectory was followed regardless of whether a country was a liberal democracy or an autocracy, and transcended even the capitalist-communist divide. For example, democratic politicians could run on a win-win platform of making the individual and the nation richer via investments in education.²⁴ Communism had a special affinity for education and industrial progress, and correspondingly the Soviet Union and Cuba both boasted universal literacy rates by the 1950s and ’60s. Yet even non-communist autocracies embraced the case for mass education, out of fear of falling behind in economic, scientific, and military terms, even if they had to compensate for depriving an educated citizenry of their political rights by engaging in ever more elaborate forms of state propaganda.²⁵ Literacy and numeracy have thus advanced across all corners of the globe since the start of the twentieth century.²⁶

Even at the start of the twentieth century, however, there was growing evidence that this state-charted life course was undermining the economic case for having children. Old-age insurance schemes alone probably played a major role in TFR reductions observed since 1920, with one estimate finding that Social Security generosity is responsible for at least 50 percent of the reduction in U.S. TFR from 3 to 2.²⁷ But more generally, the expectation that children should be educated for eventual employment in the market economy had undermined the idea that children were an economic benefit to their parents even before the advent of Social Security. Phillip Longman’s The Empty Cradle quotes the response given by one anonymous father to Theodore Roosevelt’s call for upper-class families to have more children:

It used to be said that it did not cost anything to raise a baby, but on the contrary it was an investment. Such a proposition can no longer be maintained. . . . Each baby cramps a little all that have come before it, with the result that the mother and father are soon obliged to sacrifice themselves almost entirely on the family altar.²⁸

Nonetheless, the upsides of this development pattern were strong enough that even though many European and Anglosphere states saw fertility rates decline overall across the nineteenth century, movements to encourage more births only started to emerge at the start of the twentieth century.²⁹ This amounted to mostly rhetoric, but history itself would shortly intervene.

Whence The Baby Boom

While birth rates declined to about replacement rate in the world’s richest countries during the Great Depression,³⁰ the postwar period did see a baby boom in many countries under conditions that, if anything, formalized the industrial life cycle model even more stringently. There appears to be no scholarly consensus on what exactly touched off the baby boom, and several leading hypotheses lack strong empirical sup­port.³¹ One possibility is that culture again trumped economics, or at least interacted with it in ways still not entirely understood. The war elevated the status of veterans, which thanks to the draft encompassed about ten million men primarily in their late teens and twenties. The baby boom itself occurred not just in the United States, but among almost all of the victorious Allies.³² Popular sentiment in the United States was that veterans were owed a debt.³³ Earl Warren, then governor of California, expressed this attitude well in his second inaugural address in January 1947: “We are ready and able to do our part in placing California Veterans in the kinds of homes for which they longed while overseas and which they now desperately need.”

Warren matched word to deed. Six million housing units (3.5 million of those single-family houses) were constructed in California during the thirty-year period following World War II.³⁴ This prodigious building boom would be unimaginable in California today, where annual housing unit production has recently averaged only about eighty thousand per annum.³⁵ In much the same spirit, the federal government passed the GI Bill in 1944 with its housing and educational benefits for veterans. Notably, the baby boom was really the result of a marriage boom,³⁶ as the out-of-wedlock birth rate at the time was quite low. Veterans’ ele­vated status was probably a contributing factor to the surge in marriages that occurred after the war, a cultural shock unlikely to be neatly replicated any time soon.

All of these changes were conducive to family formation. Other factors, such as reductions in maternal mortality rates and the rise in real incomes (particularly men’s), probably played a role as well, but do not seem to be the primary drivers. The origins of the baby boom demand more scholarly attention, but one reasonable interpretation of the facts is that when policymakers are motivated to accommodate people’s pre­existing desires for family formation, high levels of family formation are likely to follow suit.

The Rise of the Other Half

The baby boom officially ended in the United States in 1964 when birth rates returned to their pre-boom levels. The mid-1960s popularly mark the beginning of a new era in U.S. history that profoundly changed American society. The civil rights movement, second-wave feminism, Roe v. Wade, and stagflation all followed in quick succession. Beneath these headlines, profound changes in family formation were taking place. The figure below shows that the age at first marriage for women has risen steadily from its low of 20.5 in 1950 to 27 in the 2010 census. Birth rates in the United States and other allied countries began to decline and have since settled in at rates lower than those observed prior to the baby boom.

Further, the decline in birth rates after mid-century in the United States seems to have occurred pretty much in lockstep regardless of region or urbanicity. Studies on local population decline³⁷ show that birth rates have largely converged regardless of one’s location or com­munity type in the United States.³⁸

These changes have all of the hallmarks of a major, national economic and social shock. The 1960s are notable because they appear to mark the time when the same logic that had pushed governments to promote mass literacy, taxable work in the formal sector, and higher educational attainment among men eventually led them toward extending the same value proposition to the female half of the population. Like the spread of education and literacy itself, the logic of welcoming women into the paid labor force transcended the capitalist-communist or the democratic‑authoritarian divides. If anything, communist states led the way: the 1970 Soviet census showed that 80 percent of working-age women (15–54) were employed.³⁹ The comparable rate for U.S. women in 1977 was just 54 percent.⁴⁰

Why this occurred in the middle of the 1960s is of course debatable. The role of rising female educational attainment, however, seems too important to ignore. Already at the start of the 1960s, the average twenty-year-old woman in the United States was a high school graduate, and the average educational attainment gap between men and women was quite small.⁴¹ This small gap between the sexes had actually been a persistent feature of American life since at least the 1870s.⁴² This is less surprising in light of arguments made by Emmanuel Todd’s Lineages of Modernity,⁴³ Doepke and Tertilt,⁴⁴ and others that the rising returns to education increase the relative status of literate women in preindustrial and early industrial societies where they are seen as the first educators of children. Look no further than the Taliban for proof: a recent move to ban female education has encountered strong resistance from both men and women.⁴⁵

Female educational attainment continued to rise beyond a high school diploma with every successive birth cohort for a variety of reasons. The baby boom and the GI Bill spurred the creation of many new public universities, which were eager to fill seats with warm bodies, male or female. Coeducation in the United States had al­ready been rising as part of a slow but steady process among both private and public colleges, so that by the 1960s most of the holdouts were Catholic universities and elite northeastern private schools.⁴⁶ Pressure nonetheless steadily increased on the holdouts, partly by rising demand for coeducation among the baby boomers themselves.⁴⁷ From the perspective of the government, every well-educated person not earning money in the paid labor force is one less salary that could be taxed, making it no coincidence that the 1964 Civil Rights Act also included Title VII banning employment discrimination on the basis of sex.

The problem came largely after college, when women newly armed with degrees then tried to secure employment on equal terms. The battles fought by the first women to break into the male-dominated professions are well-known. Some of the toll can be seen in the rise of women who remained childless by ages 40–44, corresponding to the cohorts of educated women who entered the workforce in the 1960s through the early 1980s. Figure 3a, inspired by Gretchen Livingston’s work at Pew Research,⁴⁸ shows the rise and fall in childlessness among women aged forty to forty-four using June Current Population Survey (CPS) data. These data somewhat overstate the share who will have lifetime childlessness, because birth rates among women in that age bracket are much higher today than they were at the start of the sample period (1976). CPS data are unfortunately sparser on childlessness among women aged forty-five to forty-nine, who better reflect lifetime childlessness. In figure 3b, I use three years where this age group was sampled to show the share of childlessness by educational attainment in 1995, 2012, and 2020. It shows that falling childlessness is chiefly driven by falling childless rates among women with a bachelor’s degree or more.

The across-the-board fall in childlessness, which has been particularly sharp among women with professional degrees, indicates that firms, families, and governments have slowly adapted to this major shock with at least some success. This adaptation has taken several forms. Starting in the 1980s, childcare became increasingly marketized, which helped many working women convert the opportunity-time cost of childrearing into a pecuniary cost that could be benchmarked against their wage. Numerous theoretical and empirical studies show that fully marketized, high-quality childcare has been an important contributor to the decline in childlessness in women with higher education.⁴⁹ Assisted reproductive technologies have also played an important role in easing career-child tradeoffs, shifting outward the age at first birth, increasing total completed fertility, and flattening the education gradient in childlessness.⁵⁰ Parental leave policies also seem to increase both total number of children born and women’s labor force participation,⁵¹ although the effect seems to drop off around one year of paid leave,⁵² presumably because the cost of forgone working experience becomes too high.⁵³

In addition to these practical measures, the emergence of what is often called the neoliberal revolution also offered an intellectual frame­work for accommodating women’s aspirations. In economics, this is related to the idea of the “human capital” model of personal and societal development.⁵⁴ In essence, the argument was that development stages after industrialization would focus more on nonroutine cognitive tasks and involve a lifelong education and training paradigm. Mass college education would allow people from all parts of society to develop their skills to the full extent of their individual abilities. Human capital, in this telling, was as important if not more important than physical capital, because the industries of the future were increasingly seen as tied less to physically-intensive industries like manufacturing and construction and more to the digital sphere and services, where sex differences matter very little or where women may even have an advantage.⁵⁵ The upshot for women and families in the emergence of this paradigm was that as women’s share of the educational pipeline steadily grew, their leverage in white-collar workplaces steadily grew as well. Women first overtook men among college graduates around 1980,⁵⁶ and as of 2021 make up about 60 percent of college students.⁵⁷

Turning the Ship of State Around in a New Era

For some time, then, these accommodations appeared to work, partic­ularly in the United States. TFR hovered around replacement in the United States between the late 1980s and the mid-2000s. As recently as 2007, U.S. TFR was 2.12—just a notch above replacement. Even low-fertility countries like Germany saw a rebound in childbearing in the 1990s and 2000s from lows observed in the mid-1980s. This phase appears to have come to an abrupt end, however, with the Great Recession.

What is strange about the current situation, though, is not just that sub-replacement fertility has steadily taken hold in both the developed and non-developed world; many countries have also jointly experienced a sharp post-2015 TFR decrease for reasons that are cur­rently unclear. Prior to 2015, some countries like China, the United States, Brazil, South Korea, and to a lesser extent the UK seem to have experienced a TFR rise that peaked around the Great Recession, followed by a dip between roughly 2010 to 2013, a recovery that peaked around 2015, and then steady decline since 2015. For China, this decline seems like it took hold closer to 2017, as a onetime boost in 2016 from the repeal of the one-child policy gave way to the bust seen elsewhere. Others, like Russia, Japan, Iran, and Israel had a trajectory that looks more like a steady rise in TFR from 2005 to 2015 or so, followed by a sharp decline. Other countries, like Mexico, saw steady decreases in TFR throughout the entire time period, but the pace of decline picked up around 2015.⁵⁸

Generally speaking, when various countries seem to be on different trajectories and then all snap into a shared pattern at around the same time, that is strongly suggestive evidence of a shared shock. Yet I have no strong hypotheses about what that shock might have been. Making this change all the more puzzling is that a recent story from the Wall Street Journal reported that the latest survey data show that Americans’ ideal family size remains at approximately 2.5 children.⁵⁹ Nonetheless, that this pattern emerged years before Covid suggests that we have entered into some new paradigm, and the need to be innovative to help people close these gaps appears all the more pressing.

Whatever is causing the most recent decline in fertility, the iron triangle paradigm suggests that this accelerated population aging is going to make additional direct spending on families even harder than it is at present. Worse, future dependency ratios will be even higher than initially forecast,⁶⁰ making it more important for mothers (and fathers) to stay in the workforce when they have children. The challenge liberal democracies, in particular, face is to try to address the problem mostly via economic incentives even though the most recent declines (like those in the past) are likely cultural in nature. Lastly, pace Acemoglu and Robinson, societies want to continue to reap the benefits of an educated and innovative workforce, and so will want to keep the quality of children’s human capital the same (or better), while somehow also enabling families to increase in size.

There are actually three possible directions for states to go that do not directly run afoul of either the fiscal, political, or demographic planks of the iron triangle, nor conflict with liberal democratic values. One possibility is to formally recognize the importance of grandparents in the process of helping with a newborn by piloting some sort of grandparental leave, perhaps financed by Social Security. Several papers have shown that close proximity to grandparents (particularly retired grandparents) both increases a mother’s labor force participation⁶¹ and, in some settings, her completed fertility.⁶² It also seems to be the case that women whose mothers are able to provide support after childbirth are less likely to develop postpartum depression.⁶³ Yet this can come at the cost of reducing grandmothers’ labor supply in particular, especially in places where childcare options are poor,⁶⁴ possibly because there are few formal mechanisms for grandparents to take a temporary leave of absence the way there are for parents. One approach would be to allow grandparents to take leave from work for up to one quarter to help the new parents, while receiving some fraction of their expected Social Security wage. The yearly expenditure would not necessarily be very high. The average Social Security monthly payment is $1,782, or about $5,345 per quarter. There are about four million babies born each year, so even under the assumption that at least one grandparent per grandchild would (a) opt into the leave and (b) receive their full expected Social Security payment, the estimated cost per year would be about $21.4 billion, not accounting for forgone FICA taxes from the grandparent’s wages, against the Social Security Administration’s current budget of $1.42 trillion. This approach is unlikely to run afoul of the political side of the triangle, because it benefits voters across generations and is likely to be seen (on net) as helping women maintain their career achievements.

Allowing greater labor force attachment for new parents, and possibly even enabling families to have an additional child if so desired thanks to the expectation of extra help, would probably go some way toward offsetting these costs in the long run as well. Unlike parental leave which is tied to work, this program would benefit all parents, and the health benefits to new moms alone may help it significantly offset the costs, given that a 2020 NIH study estimated that untreated maternal depression costs about $14.2 billion per year. Governments can also double-down on other policies that allow for greater flexibility in conjunction with grandparental leave, including allowing paid-family leave to be used in part-time work arrangements, permitting parents to gradually return to work after parental leave, and encouraging businesses to permit parents to take their leaves within a flexible one-year window postpartum.

Another advantage may be for the grandparents themselves. For example, several programs that have combined nursing homes with day cares have shown that these sorts of intergenerational interactions lower the sense of isolation and increase health and well-being among the older residents.⁶⁵ Using relatively cost-effective measures like these may go a long way toward boosting health for all generations and maintaining prime-age labor supply, with possible beneficial side effects for childbearing as well.

The Future of Family Formation

The next approach would be to take a page out of the baby boom and tackle affordable family formation directly. A good place to start would be to facilitate the building of single-family homes. In the wake of the housing bust, single family homebuilding has lagged household for­mation for almost fifteen years.⁶⁶ Several papers have found that homeownership facilitates both marriage and having children.⁶⁷ Some national-level policymakers have already embraced the challenge, such as British prime minister Rishi Sunak’s recent promise to build one million new homes.

Thinking on this issue needs to be broader than just building new housing, however. As pointed out by Allan Mallach in his book Smaller Cities in a Shrinking World,⁶⁸ many shrinking countries are caught in a vicious cycle in which declining economic activity prompts young peo­ple to move to just one or two thriving cities, usually the capital. This in turn pushes up housing prices in these selected areas, depressing birth rates further, while exacerbating population decline. Even in growing countries like the United States, economic activity has become ever-more concentrated in a handful of so-called “superstar cities”⁶⁹ or “hub cities.”⁷⁰ Place-based and industrial policies should be reassessed not just on their ability to assist distressed regions, but as a way to create places young families can settle in that are more conducive to affordable family formation than the superstar cities. A variety of studies have shown that well-targeted place-based policies like manufacturing exten­sion, custom­ized job training, and infrastructure investments can achieve positive results at relatively low costs.⁷¹ Inasmuch as these policies do stimulate additional local economic activity without raising local hous­ing prices, then their knock-on effects on family formation mean they offer quite a good bargain.

In a similar vein, a fiscally inexpensive but potentially effective measure is to have governments encourage the working from home (WFH) revolution spurred by the pandemic. This obviously would help mostly knowledge workers, but increased WFH seems to have led to a positive bump in the birth rate of college-educated women.⁷² Corporations have reined in WFH significantly since the height of the pandemic, but there may be some actions the federal government can take, such as encouraging remote work by allowing corporations to write down at least some of the costs associated with permitting WFH. It seems like WFH encouraged more people to have children, because it allows for more flexible work arrangements. Programs like Tulsa Remote and As­cend WV that recruit remote workers to cheap locales could be more formally encouraged, possibly with federal grants, because they both can help boost distressed economies and likely yield higher completed fer­tility for the enrollees.

A third direction would be for policymakers to take more seriously the mounting evidence that young men are lagging behind their female peers in educational attainment and career progression. Richard Reeves’s book Of Boys and Men⁷³ is a good compendium of the ways that young men have fallen behind women and how policy interventions often fail to aid boys. One important way this is manifesting itself is the declining share of men and women aged forty and younger who are married. For those born in the 1980s, the projected share married by age forty is expected to be about 60 percent. On current trajectories, for those born in the 1990s this figure could very well be closer to 40 percent. Such a dramatic difference between birth cohorts relatively close in age is hard to ascribe solely to a major shift in tastes and preferences for marriage.⁷⁴ Nor is this is a problem unique to the United States. The disconnect between men and women takes other forms in other countries, but one robust finding is that men’s attitudes toward having a working wife while the children are young increasingly explain a large share of the variation in fertility rates across countries. The more accepting men are of being married to a working mother, the higher the country’s fertility rate and the lower the wife’s child penalty in earnings.⁷⁵ Governments in countries with high child penalties can embark on education campaigns to encourage men to value working wives, even if it means splitting childcare, although the efficacy of this kind of messaging is an open question. In the United States, where working mothers are relatively accepted, we may hope that the spike in manufacturing investment occasioned by the Inflation Reduction and chips Acts may be especially likely to boost male employment. This would raise their economic status and likely allow them to be seen as better marriage partners, but it is too soon to tell.⁷⁶

Lastly, if governments want to spend money directly on the problem, probably the most promising avenue is to target those people who already want a child but are concerned with the up-front pecuniary costs. There is a clear positive relationship between fertility and how much a country spends on childcare and early childhood education, although the results are modest. For example, in spite of Denmark spending up to 2 percent of GDP on childcare and early childhood education (among the highest for developed countries) in 2014, it had a TFR of around 1.69.⁷⁷ In fact, while all OECD countries have increased their spending on families since the mid-twentieth century, only Israel had a TFR above replacement as of 2020 (2.90), in large part due to the aforementioned Ultra-Orthodox Jewish population. In addition to having limited effects, these programs tend to be pricey. Hungary spent around 6.2 percent of its national GDP on various pro-natal measures in 2022, but only achieved a TFR of 1.52.⁷⁸

Of these, spending on improving childcare access seems like the best possible program. It directly targets the parenthood-work tradeoff keenly experienced by many parents while fostering both parental em­ployment and increased family size. Subsidized, high-quality programs that look to foster cognitive development can even enable better parent­ing.⁷⁹ Smartly designed programs can help get directly to the heart of the child quantity-quality trade-off that inhibit time- or money-stressed parents from having another child.

I should note that the quality-quantity tradeoff is a controversial idea in the social sciences. Bryan Caplan amassed a mountain of evidence in his book Selfish Reasons to Have More Kids,⁸⁰ arguing that the quantity-quality tradeoff is mostly bunk and that children’s future outcomes are much less dependent on parenting than is commonly believed. While that is not necessarily my own read of the evidence, I think there’s a good reason that most people have not acted on his advice. The idea of a quantity-quality tradeoff is very much in our heads, and I suspect implanted there quite firmly by Mother Nature. Real or not, people are reacting to perceived scarcity in their environment. This is true whether that scarce resource is financially stable, committed family men for lower middle-class or working-class women,⁸¹ the number of slots at elite schools for upper-middle-class families, or (more generally) time for both work, children, and leisure. Even if Caplan is right on the merits, asking people to go against their instincts and preferences is probably a fool’s errand.

The other benefit from spending more on childcare is that subsidized access to childcare will help precisely those families in the middle of the income distribution where fertility rates have sagged the most in recent years. In 2000, mothers without a college degree were at parity with college-educated women in time spent on care, but by 2019, they were spending an additional ten hours a week.⁸² This divergence suggests that these women have had trouble accessing the childcare market and have reduced their fertility accordingly. Targeting this group would require some finesse. The ranks of childcare workers will be disproportionately drawn from their ranks, creating a classic quis custodiet ipsos custodes problem. This sector may be especially suited, then, for generous wage subsidies. Letting childcare workers earn essentially a full week’s wage for half a week’s worth of work would allow them to have the time and resources to provide care to other people’s families as well as their own.

Humanity appears to be passing through a renewed phase of sharp fertility decline on a near-global basis, but it is too early to say whether this is a blip or whether we are transitioning into a new, lower-fertility equilibrium. This puts pressure on governments in developed countries to address the gap between ideal and realized family sizes perhaps soon­er than many would like, both because the dearth of births will acceler­ate population decline and because many immigrant-sending countries are also aging. Since parental earnings will be ever-more valuable as a source of taxable income, governments will need to be creative in their efforts to stabilize population sizes and shore up their welfare states without discouraging women from remaining in the workforce.

This article originally appeared in American Affairs Volume VII, Number 4 (Winter 2023): 105–30.

Notes