Silicon Valley's 'Pronatalists' Killed Remote Work. Then the Strait of Hormuz Brought It Back

By early 2026, the return-to-office movement had won. Not gracefully (Amazon could not even find enough desks for the 350,000 corporate employees it ordered back five days a week) but decisively. JPMorgan Chase ended remote work in April 2025. Dell, AT&T, TikTok, Truist, and the Washington Post followed with five-day mandates. Instagram's Adam Mosseri told staff a five-day office presence was needed for a "winning culture." Microsoft began requiring Puget Sound employees three days minimum in February 2026. A KPMG survey found 83 percent of CEOs expected full return to office within three years; a ResumeBuilder survey reported nearly half of all companies demanded four or more in-office days by 2026, with 28 percent phasing out remote work entirely.

Some companies used the mandates as quiet layoffs, hoping workers who valued flexibility would self-select out rather than commute. And the workers complied. By December 2025, only 40 percent of employees said they would quit over a mandatory return-to-office notice, down from 91 percent who said the same in January. The job market had tightened: 2025 produced just 584,000 total job gains, the weakest outside a recession since 2003. The leverage belonged to employers, and employers wanted bodies in chairs. The pandemic experiment was over. Remote work was being consigned to a brief, strange interlude, a concession to extraordinary circumstances that would not recur.

The implicit premise of every return-to-office memo was clear enough: the exogenous global shock that created mass remote work was a one-time event. COVID-19 was the exception. Normalcy was the rule. After all, what were the odds of another global shock that would force millions of workers home overnight?

In March 2026, the Strait of Hormuz closed, and the question answered itself.

We will not rehearse the geopolitics; the reader is more than aware. What bears noting is how governments responded. Thailand ordered all non-frontline public-sector employees home, set air conditioning to 26 degrees, told workers to take the stairs, and replaced travel with online meetings. The Philippines moved to a four-day government workweek. Pakistan's Prime Minister mandated remote work for half of all government employees and shifted universities online. Vietnam urged all employers to allow remote work and encouraged citizens to ride bicycles. Bangladesh closed universities, stationed troops at oil depots, and turned off Eid-al-Fitr light displays. Nepal filled cooking gas cylinders only halfway to stretch supply.

These governments did not mandate remote work because they had read a working paper about fertility. They did it to keep the lights on. But in doing so, they resurrected the very arrangement that corporate America had just buried, and that a growing body of research now links to higher birth rates.

I want to be clear-eyed about what comes next. The commodity shock that forced these mandates will itself devastate birth rates by destroying the employment and economic prospects of young people, especially young men, across the region. No workplace policy can stand against that kind of destruction. But as we argue below, the remote work mandates these governments stumbled into are among the most effective interventions available for raising children per mother among those with the means to form families.

And the loudest pronatalists in American life, the ones who claim declining birth rates are civilization's gravest threat, are the same people who just spent two years dismantling it: Elon Musk, who has fathered at least fourteen children and called declining birth rates "a much bigger risk to civilization than global warming," told tech workers on CNBC to "get off the goddamn moral high horse with the work-from-home bulls***." Marc Andreessen, whose Techno-Optimist Manifesto declares "our planet is dramatically underpopulated," testified before his local town council that he was "immensely against multifamily housing development." The network around them (Thiel, Altman, Armstrong, Buterin) has poured some $800 million into fertility technology while the companies in their orbit dismantle the workplace flexibility that actually raises fertility.

Here is the core of our argument. A growing literature (Lu et al. 2025, using U.S. census data; Wang and Dong 2024, running experiments in Singapore; Bailey et al. 2023, on the COVID baby bump; Chong and Noguchi 2024, on Japanese pregnancy odds) converges on the same finding: remote work raises fertility. The latest and most comprehensive contribution is Davis, Aksoy, Barrero, Bloom, Cranney, Dolls, and Zarate (2026), drawing on original survey data from 38 countries and the U.S. Current Population Survey.

The effect operates on what we call the intensive margin: more children per mother, among people already positioned to have families. It cannot solve the deeper economic preconditions for family formation. But the magnitudes are striking, the evidence is no longer tentative, and the policy it identifies is precisely the one that self-described pronatalists just finished destroying.

The evidence Davis et al. (2026) draw on two original surveys. The Global Survey of Working Arrangements (G-SWA) covers 38 countries; its fourth wave, fielded November 2024 to February 2025, includes a new fertility module. The U.S. Survey of Working Arrangements and Attitudes (SWAA) is a monthly survey yielding 89,886 respondents aged 20–45 from 135,949 total usable observations (December 2022 to December 2025). Both samples are carefully constructed: respondents in the bottom 5 percent of completion times and the 17 percent who fail attention checks are dropped, and the SWAA data are reweighted via raking to match the Current Population Survey on age, sex, education, partner status, region, and employment.

The authors consider three fertility measures: realized fertility from 2023 to early 2025 (including children in gestation), planned future fertility, and lifetime fertility (children to date plus plans). The headline findings are large and consistent. When both partners work from home at least one day per week, estimated lifetime fertility is greater by 0.32 children per woman in the 38-country sample (14.3 percent of mean fertility) and 0.45 in U.S. data (17.5 percent of mean fertility). For realized fertility since 2023, women who work from home had 0.037 more children in the G-SWA (25 percent of sample mean) and 0.091 more in the SWAA (33 percent of sample mean). Planned fertility is also significantly higher. These patterns hold across all specifications, for men and women separately, with controls for age, education, marital status, pre-2023 children, own and partner employment status, and country or state fixed effects.

A complementary analysis using the Current Population Survey (approximately 137,000 respondents ages 30–45, 2023–2025) reinforces the findings by linking fertility not to individual WFH status but to occupation-level WFH shares, the fraction of job postings advertising remote work in one's occupation, as classified by Hansen et al. (2026) using an LLM applied to half a billion postings with 99 percent accuracy. A one-standard-deviation increase in own-occupation WFH share raises one-year fertility by 7.3 percent of mean for women; adding partner's occupation, the total effect reaches 14 percent. The results also hold in 2017–2019 data, before anyone could have chosen occupations with foreknowledge of pandemic-era WFH opportunities. Since nearly nine in ten people aged 30–45 stay in the same occupation year to year, reverse causation is implausible.

Davis et al. also note a "missing intercept" problem: their cross-sectional estimates cannot capture aggregate effects of WFH prevalence on fertility norms, so the true effects are likely larger.

Davis et al. is not the first study to establish this link. It is the most comprehensive. Lu et al. (2025) find that the COVID WFH shock raised fertility, especially through higher-order births, second and third children, not first. Wang and Dong (2024) demonstrate a causal effect of flexible work on fertility intentions in a Singaporean experiment. Chong and Noguchi (2024) report increased pregnancy odds for Japanese women in high-WFH occupations. Bailey et al. (2023) show relative birth-rate gains among college-educated American women in 2021, the group that saw the largest WFH expansion. What Davis et al. adds is scale (38 countries, 135,000+ U.S. respondents), household-level analysis capturing partner effects, and a concrete national-level quantification: WFH accounts for an estimated 291,000 U.S. births per year, 8.1 percent of total fertility, as of 2024.

Since U.S. WFH rates are now three to four times pre-pandemic levels, this implies roughly 200,000 extra births relative to a no-pandemic-WFH counterfactual. By comparison, Olivetti and Petrongolo (2017) find that U.S. government spending on early childhood care and education (0.4 percent of GDP) contributes an estimated 0.08 children per woman. The WFH contribution is 0.13. Remote work is delivering more fertility impact than the entire early childhood spending apparatus, at zero cost to taxpayers.

The debate is no longer about whether the relationship exists. It is about magnitude and mechanisms.

Two margins, two different forces Before we go further, we need to draw a distinction that much of the fertility debate elides. Birth rates reflect two things: how many people become parents (the extensive margin) and how many children those parents have (the intensive margin). These respond to different forces. Entry into parenthood is overwhelmingly driven by economics: stable employment, a partner with stable employment, a reasonable expectation of affording housing and children. When commodity shocks or recessions destroy the prospects of young adults, and especially young men, whose earnings remain disproportionately important to household formation decisions across cultures, fewer people form families. If young men in Bangkok or Karachi cannot find work, they do not marry, and births do not happen. WFH affects the second margin. The Davis et al. results are conditional on age, education, employment status, partner's employment, and marital status. The women and men in their regressions overwhelmingly have jobs and partners. For this population, remote work eases the time and coordination costs of combining employment with child rearing. It shifts the calculus from two children to three, or from "not yet" to "now." The Lu et al. (2025) finding that the COVID WFH shock raised fertility mainly through higher-order births, not first births, reinforces this reading. Remote work is not making new parents. It is giving existing parents room for one more child.

This distinction is essential for everything that follows.

Return-to-office as anti-natalist policy that Elon Musk loves If WFH raises fertility by these magnitudes, the corporate return-to-office wave is, functionally, an anti-fertility policy. Andy Jassy did not set out to reduce birth rates. But eliminating remote work for 350,000 employees in prime childbearing years removes a feature that workers value at roughly 5 percent of earnings, that parents value even more, and that is associated with significantly higher realized and planned fertility. Intentions are beside the point.

The distributional dimension sharpens the concern. WFH concentrates among college-educated, professional workers, precisely the groups with the lowest fertility and the most to gain from flexibility. Bailey et al. (2023) found that college-educated American women saw relative birth-rate gains in 2021, the group with the greatest WFH expansion. Davis et al. reveal a gendered pattern: women's fertility is negatively associated with paid employment in their 38-country data, but this penalty is largely offset when the woman works from home. Return-to-office mandates restore the full penalty. And when companies use those mandates for indirect headcount reduction (as Resume Builder's Stacie Haller notes, leaders are often aware that flexible workers will leave rather than comply) the workers who self-select out are disproportionately parents, the people managing school schedules and sick days and the thousand logistical demands of family life.

A rough estimate: if the U.S. WFH share among women drops from approximately 42 percent to 30 percent, the Davis et al. framework implies about 100,000 fewer births per year. These are births lost on the intensive margin, not fewer parents, but parents who lose the flexibility that made an additional child feasible. That is illustrative, not definitive. But the order of magnitude is sobering: a fertility reduction imposed by corporate fiat that exceeds the effect of many countries' entire pro-natal policy apparatus.

The 2026 energy crisis: a cruel and clarifying test The crisis-driven WFH mandates are analytically interesting for reasons Davis et al. could not have anticipated; their paper predates the conflict. The mandates are exogenous to fertility concerns (no country adopted remote work for fuel conservation to raise birth rates), they are massive in scale (millions of workers moved to remote arrangements within days), and they overlap geographically with the world's lowest-fertility nations. Japan, with a TFR of 1.23, had the lowest WFH share in the Davis et al. sample at 21 percent. South Korea, with the world's lowest TFR at 0.75, stood at 27 percent. Davis et al. estimate that raising WFH to the U.S./UK/Canada average of 45.3 percent would yield 31,800 extra births per year in Japan and 10,500 in South Korea.

This is also not the first time it has happened: after Russia's invasion of Ukraine disrupted European energy supplies in 2022, Germany encouraged remote work to reduce consumption, and a Freshfields analysis found it could cut national gas use by 5 percent. Crisis-driven WFH expansions are becoming a recurring geopolitical feature, not a pandemic anomaly.

But we must be honest about the savage arithmetic of commodity shocks. The same crisis forcing Asian governments to mandate remote work is simultaneously annihilating the conditions under which young people form families. Oil above $100 a barrel in import-dependent economies means surging costs for fuel, food, transport, and electricity; inflation eroding real wages; firms cutting back. South Korea, which imports 70 percent of its oil from the Middle East, has seen its Kospi index post some of the steepest market declines this month. The Economist Intelligence Unit projects elevated energy prices will raise inflation and lower growth across the region. Agriculture faces its own disruption: Gulf fertilizer supplies have been cut, Thai rice exports to the Middle East have stalled, Indian farmers are dumping supply at lower prices. When young men cannot find stable work, household formation collapses and births do not happen, regardless of whether office workers have been told to stay home on Fridays.

The crisis-hit countries will almost certainly see birth rates fall, not rise, in 2026 and 2027. The tragedy is that the crisis simultaneously expands WFH (good for the intensive margin) and contracts the economic foundations of family formation (catastrophic for the extensive margin). The net effect will be negative. But the longer-run story may be different. The oil crisis is temporary. The WFH habits it creates need not be. East Asian countries have spent billions on conventional pronatalist policies (cash bonuses, parental leave, childcare subsidies) with limited results. If crisis-induced remote work practices outlast the crisis, as they did after COVID-19, permanently higher WFH rates and the accompanying fertility gains could follow.

We should not pretend that WFH can solve the fertility crisis in societies where the binding constraint is the economic destruction of young men's prospects. If the shock lingers, it will do more damage to birth rates in one year than a decade of WFH expansion could repair. But on the intensive margin, the gains are real. Even there, WFH is not a panacea. Kim et al. (2024) argue that status-driven competition in child education is a key driver of East Asian fertility collapse: parents locked in a positional arms race over schooling, spending enormous sums and time to secure marginal advantages. WFH makes it easier to drive a child to tutoring. It does not make the arms race less exhausting. Davis et al. acknowledge that these status externalities "could mute the potential impact of expanded WFH opportunities." Fair enough. But status competition suppresses desired family size, while workplace inflexibility suppresses the ability to achieve even a reduced desired family size. WFH addresses only the latter. In East Asia, that may mean a lower ceiling on WFH-induced gains than in North America or Europe. But at TFRs below 1.0 in South Korea, every fraction of a child per woman matters.

Pronatalist paradox I described the pronatalist contradictions above: Musk's simultaneous championing of birth rates and disparagement of remote work, Andreessen's planet-is-underpopulated manifesto paired with opposition to housing construction, $800 million in fertility technology from a network that is dismantling the workplace flexibility that outperforms government early childhood spending. We do not need to redescribe the figures. We still need to diagnose the analytical failure.

The return-to-office case is that in-person collaboration fosters mentorship, company culture benefits from proximity, young workers gain from office exposure, and there are productivity costs to fully remote work. The research is shaky on all those points, and company culture is more affected by layoffs. Let's take them at their word. Hybrid arrangements of one or two days a week capture a lot of the fertility upside while preserving the in-person collaboration that return-to-office advocates prize. Companies are still eliminating hybrid arrangements, let alone remote work.

The deeper problem is the distinction between techno-pronatalism and real world pronatalism. Techno-pronatalism (IVF, genetic screening, embryo selection, artificial wombs) is expensive, selective, and accessible primarily to elites. Real world pronatalism (workplace flexibility, affordable family size housing, child bonuses, etc) is (relatively) cheap, universal, and available to everyone. As the Heritage Foundation's Emma Waters has noted, Silicon Valley pronatalists "tend to promote, in practice if not in speech, a selective pronatalism: more babies of a certain kind." The WFH-fertility evidence is emphatically not selective; it operates across education groups, occupations, and countries. And the pronatalists are not merely failing to support the intensive margin. They are attacking the extensive margin as well. Opposition to housing construction raises the cost of family formation. The use of return-to-office mandates as covert layoffs pushes parents out of employment. The tightening job market of 2025, to which tech layoffs contributed, made it harder for young adults to establish the economic footing that precedes family formation.

They are undermining the workplace flexibility that raises children per mother while doing nothing to strengthen the economic foundations that determine who becomes a parent. That is not a contradiction in emphasis. It is incoherent.

What should be done Davis et al. observe that "for societies faced with undesirably low birth rates, WFH can thus yield societal benefits that go beyond any direct benefits to employees and employers." Governments should stop discouraging remote work, especially if they received poor advice from certain "industry leaders." The public sector is the obvious starting point: because the government is a large employer across many occupations, its workplace practices set private-sector norms.

For crisis-hit Asian economies, the immediate priority is economic stabilization: protecting employment, containing inflation, keeping young men in the labor force. No WFH policy substitutes for those preconditions. But once the crisis recedes, the governments that mandated remote work for fuel conservation should study making some version permanent. Japan and South Korea have low WFH rates, extremely low fertility, and have exhausted most conventional pronatalist options. Hybrid work incentives could complement existing policies at minimal fiscal cost; the Davis et al. estimates imply 31,800 extra births per year in Japan and 10,500 in South Korea if WFH rates reach the U.S./UK/Canada benchmark.

Employers should understand what the evidence implies for talent. Workers value WFH at roughly 5 percent of earnings; parents value it more. Firms offering hybrid arrangements will attract and retain workers with stable families and the lower turnover that follows. Davis et al. caution that "policy interventions that push for a one-size-fits-all approach to working arrangements are likely to yield unhappier workers and lower productivity." The goal is expanding options, not mandating remote work universally. But expanding options is precisely what the return-to-office movement is not doing.

Beating Dead Horses By early 2026, the return-to-office movement had declared victory. Then the Strait of Hormuz closed, and governments from Bangkok to Islamabad ordered workers home, not to build culture, but to save fuel. The arrangement corporate America had just buried was resurrected in days.

I am not romanticize what is happening in Asia. Again, commodity shocks will devastate birth rates by destroying the employment prospects of young people, especially young men. The binding constraint on fertility in crisis-hit Asia is whether young men can find work and young couples can afford a home, not whether they commute. But for those who are still employed, partnered adults who do form families, the question of how many children they have are shaped by the logistics of combining paid work with child rearing. On this margin, the evidence is clear: the ability to work from home even one day a week is associated with 0.32 more children per woman across 38 countries, and 0.45 more in the United States, amounting to 291,000 births per year that would not otherwise occur.

Meanwhile, in Palo Alto, the play pretend pronatalist tech executive who tweets about civilizational collapse from declining birth rates is ordering his employees back five days a week. He has invested millions in artificial wombs and embryo selection. He has not invested a penny in the workplace flexibility that produces, at no cost to anyone, hundreds of thousands of additional births per year, not by creating new parents, but by giving existing parents room for one more child.

The CEOs said the pandemic was a one-time event. They asked what could possibly force remote work back. Now they have their answer, and it arrived not from Silicon Valley but from the Strait of Hormuz. The fertility intervention that works is not a moonshot. It is the ability to work from home at least one or two days a week. Now the question is whether the people who claim to care most about birth rates will read it before the next shock forces their hand again.