The Island's Last Song: Even the Dragon Year Couldn't Save 0.695

30-Second Overview: 2024 was the Year of the Dragon, when Taiwanese were supposed to rush to have children for good luck. Yet that year saw 715 fewer babies than the previous one¹, marking the first time in 48 years that the 'magic' of the Dragon Year failed. The following year, the fertility rate collapsed to 0.695, the lowest among sovereign nations². The government has spent nearly 50 billion New Taiwan dollars over ten years trying to boost birth rates³, but it has consistently misunderstood one thing: Taiwanese people are not choosing not to have children (over 70% of married couples have two or more children)⁴; they are choosing not to marry. The money was sent to married families, but the gap lies with those who never married. And looking globally (South Korea, Hungary, Singapore), no country has ever bought back its fertility rate⁵. So the real question has shifted from "how do we get people to have more children" to "how do we live with dignity on an island destined to shrink?"

In September 2024, in Liugui, Kaohsiung, it was the first day of school at Xinwei Elementary School.

Founded in 1921, this school has a 104-year history and is the only Hakka-language living school in Liugui⁶. During the flag-raising ceremony that day, the entire school's five classes and 29 students stood on the playground. Among them, zero first-grade freshmen⁷. Not a single one. It was the first time in the school's history.

In August the following year, Xinwei Elementary School was merged into Guangxing Elementary School in Meinong⁸. Acting Principal Chen Yi-ting said in the final semester that the future would focus on Hakka-characteristic teaching, "hoping to attract local students to study here and continue to pass on Hakka culture"⁹. City Councilor Lin Fu-bao was more direct: "Having one child per class is not a good thing either" "If we should merge schools, we should just merge them"¹⁰.

Xinwei Elementary School is not an exception; it is a letter. A letter sent from Taiwan's countryside forty years ago, which urban areas are only now receiving.

Even the Dragon Year Is No Longer Effective

Taiwanese people have memories of the zodiac. Rushing to have children in the Year of the Dragon for good luck, avoiding the "lone phoenix" year of the Tiger—this folk calendar has left fingerprints on birth numbers for nearly fifty years. In the 1976 Dragon Year, 425,125 babies were born nationwide, a surge of 55,776 from the previous year¹¹. In 1988, +28,007; in 2000, +21,651; in 2012, +32,854¹². Four Dragon Years, four jumps upward, without exception.

Then came 2024.

This year was the Year of the Dragon, theoretically a rare peak in birth rates in recent years. Yet the number of births was 134,856, 715 fewer than the 135,571 in 2023¹³. For the first time in 48 years, the Dragon Year did not lift the numbers; it pushed them down. The zodiac's magic officially failed that year.

If the Dragon Year failing was just a warning signal, 2025 was an earthquake. In the Year of the Snake, births collapsed to 107,812, another 27,000 fewer than the Dragon Year, a drop of nearly 20% in a single year¹⁴. The Total Fertility Rate (TFR) plummeted directly from 0.89 in 2024 to 0.695¹⁵, overtaking South Korea to become the place with the lowest fertility rate among sovereign nations¹⁶. In the same year, deaths reached 200,000, with a natural decrease gap of −92,456, the largest natural decrease in Taiwan's history¹⁷.

💡 Did You Know: This is how you read "0.695." It is called the "Period Total Fertility Rate," meaning "if a woman were to have children at the age-specific fertility rates of 2025 throughout her life, she would have 0.695 children." It does not equal "the average woman today will only have 0.7 children in her lifetime." Because Taiwanese people are having children later and later, pushing births that should have been spread over several years into the future, this "period" indicator is underestimated and flattened. Zheng Yan-hsin from the Academia Sinica calculated: looking only at the period fertility rate, Taiwan dropped below 2 in 1985; but when calculated truly by "cohort," it wasn't until the women born in 1965 that lifetime fertility fell below 2, a gap of twenty years¹⁸. So 0.695 is tragic, but the way it is tragic is more complex than it appears.

The numbers are shocking, but what is more worth stopping to think about is that −715 in 2024. It proves one thing: even the lowest threshold of motivation—"seeking good luck, having one more child"—Taiwanese people are no longer willing. The problem hasn't even reached whether people can afford a second child; they can't even have a first. And if you trace back one step from the inability to have a first child, the problem isn't in the delivery room; it's in the wedding.

Not Not Having, But Not Marrying

This is the most counter-intuitive part of the whole story, and the part most often misstated.

When people talk about low birth rates, their brains automatically run to "young people can't afford to raise kids," "housing prices are too high, so they dare not have children." It sounds self-evident. But Zheng Yan-hsin from the Institute of Sociology, Academia Sinica, who has studied Taiwan's fertility data for decades, concludes: "The key code to low fertility is not not having children, but more likely: not yet married."¹⁹

Her evidence is solid. In Taiwan, most couples have children after marriage, "with over 70% having two or more children"²⁰. In other words, as long as they marry, Taiwanese people's willingness to have children is actually not that low. What collapsed is the denominator: fewer people are getting married. In 2024, there were 123,061 marriages; in 2025, it dropped to 104,376 pairs, a historic low²¹. The lifetime unmarried rate for those aged 45 to 49 is approaching 20%, projected to exceed 30% for men and 25% for women by 2050²².

What about "not marrying but having children"? In many European countries, non-marital births account for the majority of births. But Taiwan is not. Zheng Yan-hsin's numbers: "In Taiwan, non-marital births are only 2–4%, unchanged for decades."²³

Stack these three numbers together, and a completely different causal map emerges: Taiwan's fertility is almost entirely bound to marriage (non-marital births are only 2–4%), and the number of people getting married is collapsing. So the drop from the replacement level of 2.1 to below 1.1 is almost entirely driven by the decline in marriage rates²⁴. Rather than saying the fertility rate was crushed by "can't afford to raise," it is more accurate to say it was drained by "not marrying."

📝 Curator's Note: The common narrative gets the causality reversed. The mainstream narrative is "economic pressure → dare not have → low birth rate," so the policy logic naturally becomes "give money to reduce pressure → people will dare to have." But if the real break point is "not marrying," this causal chain is wrong from the first step. Zheng Yan-hsin speaks calmly: if the problem is "cannot afford," subsidies might be effective; but if most are not marrying, "economic policies are of no help." Research from the Taiwan Institute of Economic Research points to the same place: low housing prices and low wages are not the main cause of non-marriage; more importantly, "most unmarried people have the desire to marry, but after entering the workforce, it is difficult to meet suitable partners"²⁵. In other words, this problem cannot be solved by giving money. The money was sent to the wrong link.

And be careful of a deeper trap: framing not having children as "young people being selfish." When women first gain the ability to say no to the "expectation of childbirth," labeling this choice as selfish is unfair and misses the target. After all, the main cause has never been that married people are having fewer children, but that they never entered marriage in the first place. We will return to this later.

Giving Money to the Wrong People

So what did the government do over these ten years?

The answer is: it gave a lot of money. The Total Expenditure for the Strategic Plan for Responding to a Low Birth Rate was approximately 485.1 billion²⁶ from 2007 to 2024. Investments related to ages zero to six increased from 15 billion in 2016 to 140 billion in 2026, breaking 120 billion in a single year²⁷. Childcare allowances are now 5,000 NTD per month for the first child, 6,000 for the second, and 7,000 for the third and above, with means-testing canceled²⁸. In May 2026, Lai Ching-te added an "18-Item New Strategic Plan for Taiwan's Population," calling for a monthly growth allowance of 5,000 NTD for ages 0 to 18 and a unified central birth payment of 100,000 NTD²⁹.

The money was truly scattered. The problem is, it was almost all sent to families who are already married and already have children. To the part that is collapsing (those who are not married), these subsidies do not connect at all. This is what "giving money to the wrong people" means: the problem was never that there was too little money, but that the target was chosen wrong from the start.

More awkwardly, even when aimed at "married but hesitating to have another child" families, the effect is negligible. Yang Tzu-ting from the Institute of Economics, Academia Sinica, conducted a clever study using Taiwan's lottery wins as a "natural experiment": for families that suddenly won over one million in prizes, it was like a "birth subsidy" falling from the sky; would they have more children?

Yes. But only by 0.07 children.

Converted, out of every 100 families winning big, only about seven more children were born due to this unexpected windfall within six years³⁰. Yang Tzu-ting's inference is direct: unless the subsidy scale is large enough to cover the entire cost of raising a child, it will not shake the fertility decision³¹. Why is the money so useless? Because he found that what truly discourages people is time, not money: "Raising children is not just about increasing expenses; more importantly, it impacts personal time arrangements."³² In Taiwan, raising a child to 18 costs about 10 million NTD, but what is more daunting than 10 million is the life parents must hand over during those 18 years.

Wen Tzai-hung from the Center for Population and Gender Studies at National Taiwan University speaks even more sharply. He has long studied population issues and delivered a verdict on the logic of "giving money to boost births":

"The logic of giving you money to tell you to give birth does not work; pursuing a more equal, kinder living environment is the fundamental solution."³³

He added a sentence explaining why this matter is so hard to take seriously: "Population problems are always like boiling a frog in warm water; you never feel it is serious."³⁴ When an earthquake comes, you react within seconds; but the population crisis has had a warning period of ten to twenty years, yet because it is slow, it is always pushed to tomorrow. What Wen Tzai-hung truly cares about is the underlying setting of Taiwan's society. He says, "If we continue to emphasize individual competition, this situation of uneven resource distribution and extreme concentration cannot change."³⁵ In a society that dumps all pressure on individuals to bear, having children becomes a high-risk personal gamble. No matter how much money is given, it cannot change the structure of that gamble.

⚠️ Controversial Viewpoint: Jiang Min-qin from Shih Chien University cuts in from another angle, saying Taiwan "privatizes the risks of child-rearing and socializes the demographic dividend": children grow up to pay taxes, serve in the military, and support the whole society; the benefits are everyone's; but the risks, costs, and sacrifices of raising them are swallowed by parents themselves. He therefore argues: "Low fertility is not a moral defect, but a price signal."³⁶ Young people not having children does not mean they are broken; it is more likely that they have calculated the bill clearly. In this framework, subsidies are "like painkillers, not surgery"³⁷; they can suppress temporary pain, but cannot cure the lesion.

No Country Has Bought It Back

At this point, a natural counter-question arises: what about other countries? Someone must have succeeded, right? Should Taiwan copy their homework?

This is the sentence in the whole article that should be remembered most: To date, no developed country has successfully bought its fertility rate back to the replacement level of 2.1 through policy. Not one.

This sounds pessimistic, but it is a conclusion repeatedly verified by international organizations. Research from IZA World of Labor says that pro-natalist policies can indeed cause a small rise in fertility rates, but "it is unlikely to push it back to the replacement level"³⁸. Economist Gietel-Basten gave a more colloquial footnote to Asia's low fertility: "Throwing money cannot solve this problem."³⁹ The evidence laid out:

• Hungary is the model student for conservative pro-natalism, with the Orbán government's family expenditures accounting for around 5% of GDP for years, ranking in the top tier in Europe, but the 2024 fertility rate was still only 1.39. AEI even estimated that if the world wanted to raise fertility rates by 0.2 through policy, it would need to spend an additional 250 billion USD annually⁴⁰.
• South Korea threw over 270 billion USD, yet it remains one of the lowest in the world⁴¹.
• Even benchmarks like France and the Nordic countries, known as "family-friendly," cannot hold the line: France's 2025 fertility rate was 1.56, the lowest since World War I; Finland's was 1.25, a record low⁴². All model countries are falling.

South Korea's story is especially worth unpacking because it is most easily misread as a case of "policy working." South Korea's fertility rate did rise for two consecutive years: 0.72 in 2023, 0.75 in 2024, returning to 0.80 in 2025⁴³. It sounds like there is hope. But Statistics Korea itself said this rise was "likely driven by an increase in marriages since August 2022"[^44]: a post-pandemic wave of delayed marriages being concentratedly made up, plus the demographic dividend of more women in their early 30s reaching marriageable age, not the credit of any pro-natalist policy.

Source: Statistics Korea, Department of Household Registration, Ministry of the Interior, 2025

(Note: Taiwan here is the Period Total Fertility Rate (period TFR); under the World Bank/UN model methodology, Taiwan is approximately 1.1, belonging to a different system, and the two cannot be mixed⁴⁴.)

And there is a crueler fact hidden here. South Korea's fertility rate rose for two years, but the population is still shrinking⁴⁵. Why? Because of "population momentum." A society's population structure is like a giant ship; even if the fertility rate miraculously returns to 2.1 overnight, the number of women of childbearing age who can give birth has already shrunk significantly, and the total population will still continue to fall for decades⁴⁶. Taiwan's total population peaked at 23,603,121 at the end of 2019⁴⁷, and has been going down ever since. This downhill slope, no policy can turn it around.

📝 Curator's Note: This is why the framework of the question "can pro-natalism reverse low birth rates" is itself wrong. Reversing implies pulling the total population back, and population momentum shows this is physically impossible. The most policy can do is "slow the speed of decline," it cannot "turn around and go up." South Korea is the living evidence of this sentence: fertility rate rising and population continuing to decrease, the two things happening simultaneously, not contradictory at all. Setting the goal as "saving the population" is注定 a losing battle; changing the goal to "falling less, living better in the process of shrinking" is the option still in hand.

Clarifying an old version often cited: many reports say South Korea established a dedicated "Population Strategy Plan Department," and policy thus took effect. In fact, Yoon Suk-yeol did announce in May 2024 that he would establish this department and declared the nation in a population emergency, but the relevant bill got stuck in the legislature and died in the womb with Yoon Suk-yeol's impeachment. This department never truly came into existence⁴⁸. Using a non-existent department as a success case is a mistake.

The Countryside Has Already Walked This Path

Back to that playground at Xinwei Elementary School with no first-grade freshmen.

Urbanites hearing "low birth rate" often feel it is abstract; the MRT is still crowded, kindergartens still have queues. But Taiwan's countryside began receiving this letter forty years ago; now it is the city's turn to open it. Zooming out to the whole island: there are 512 primary schools with fewer than 50 students nationwide; 1,025 with fewer than 100, accounting for 38.4% of all primary schools in the country⁴⁹. The total number of primary school students dropped from 1.2 million in the 113th academic year to a projected 689,000 in the 129th, evaporating over half in fifteen years⁵⁰.

In 2026, in Xingtian, Jiaxian Xiaolin, and Taoyuan Jianshan in Kaohsiung, first-grade freshmen were each only one person⁵¹; in all 79 schools in Kaohsiung City, first grade only had one class. In Danei Toushe and Hede Liuxi branches in Tainan, first grade was also one person each⁵². Schools with only one class for freshmen nationwide have reached 110.

This time machine does not stop at primary schools. Low birth rate is a wave climbing up by age: first hollowing out the delivery room, then the primary school, then the university, the military barracks, the workplace, long-term care, the pension system, arriving on time station by station.

Universities are the first to be hit. The number of first-year university students dropped below 200,000 in the 114th academic year, projected to drop to 146,000 by the 129th⁵³. The Private University Union estimates that when freshmen are only 50,000 to 60,000, it will only be enough to sustain about 50 private universities, meaning about 40 schools will face closure⁵⁴. (To be honest, the number of university vacancies in 2025 hit a five-year low, with only 18 schools having 1,220 vacancies⁵⁵. The long-term "student source cliff" is real, but short-term numbers fluctuate due to closures and reduced enrollment, not linearly.)

Military barracks also received the letter. The number of male conscripts dropped below 100,000 for the first time in 2023 (97,828), projected to drop to 74,000 by 2031⁵⁶. The bankruptcy year for Labor Insurance has been delayed from the originally estimated 2028 to 2031, propped up by an unprecedented appropriation of 387 billion NTD⁵⁷. The talent gap in the workplace is estimated by the National Development Council to expand from 400,000 to 480,000 by 2030⁵⁸. And in 2025, Taiwan officially stepped into a "super-aged society," with the population aged 65 and over exceeding 20.06%, 4.67 million people⁵⁹.

💡 Did You Know: A crossover quietly happened in 2020: that year, the number of dogs and cats in Taiwan exceeded the number of children aged 0 to 14 for the first time. In recent years, there are about 2.79 million dogs and cats, still more than the 0 to 14 population (about 2.68 million)⁶⁰. More pets than children; this sounds like a sigh, but it is actually a concrete measure of a society slowly shifting the object of emotion, companionship, and care from the next generation.

So where do the missing people come from? Part of it has quietly filled the gap. In April 2025, the number of migrant workers in Taiwan reached 830,000, a historic high, of which 215,000 were foreign caregivers caring for Taiwan's elderly⁶¹. There are also about 600,000 new residents, who will soon become Taiwan's fifth largest demographic⁶². In other words, as Taiwan's own birth rate shrinks, it is the labor force from other countries silently holding up the island's factories and long-term care beds. This is an invisible pressure valve, and also an ethical issue we have not yet properly faced: we rely on their prime years, but rarely seriously discuss their situation and future.

Whose Crisis?

At this point, we should turn back and ask a sharper question: whose crisis is low birth rate?

For the past ten-plus years, low birth rate has almost always been packaged into the framework of "national security crisis": insufficient military recruits, labor shortage, economic recession, pension black holes. It sounds urgent. But more and more people are asking back: is this framework turning women's bodies into tools for the state?

Li Meng-yun from the Li Hsin Foundation speaks directly: in the pro-natalist national security narrative, "women will be viewed by the state as resources/tools for population reproduction"⁶³. Lin Lü-hong from the Taiwan Women's Alliance points out a truth often ignored: "Taiwanese women are not unwilling to give birth, but do not wish to be bound by marriage culture and mother-in-law/daughter-in-law kinship relations"⁶⁴. Many women want children, but do not want to be tied to the entire marriage-family structure behind the child that they cannot accept.

This connects back to the thread of "not marrying" mentioned earlier. BBC Chinese interviewed several Taiwanese people who chose not to follow the traditional path. Guo Pei-yu, a 35-year-old independent publisher editor, spoke without embellishment: "Marriage has no benefits for women at all; my mother married and had children according to the script society gave her, and lived in pain within that framework."⁶⁵ Mr. Chen, 33, working in a local public institution, said, "Living alone is also fine."⁶⁶ In 2023, the unmarried rate for Taiwanese aged 35 to 39 reached 47.6%⁶⁷, nearly half. When "not marrying and not having children" shifts from a marginal choice to the norm for nearly half the people, continuing to define it as a "crisis," "selfish," or "outlier" is itself worth questioning.

⚠️ Controversial Viewpoint: Zheng Yan-hsin's proposed solution points the矛头 at the whole society's assumption of "who should be responsible for raising children": "Child-rearing needs to become the business of a whole village, society, and state, not women's natural duty."⁶⁸ This flips the subtext of the entire pro-natalist policy. As long as raising children is still tacitly assumed to be women's natural duty, then subsidies, allowances, and slogans are just adding money to an unfair division of labor, without touching the division itself.

Even the most fundamental assumption that "total population decrease = economic recession" is challenged by some. Commentators point out that one must distinguish between "total GDP" and "per capita GDP": if the population shrinks, but each person creates more value, per capita income may actually rise, and it is estimated that lower fertility rates might increase per capita income by about 10% before 2055⁶⁹. From another angle, the ones truly bearing the crisis might be specific structures like the tax base, the military, the pension system, and domestic industries needing large numbers of people, rather than every specific Taiwanese person. For the individual, fewer people is not necessarily the end of the world.

So if pro-natalism is注定 ineffective, and labor shortage is reality, does Taiwan have another path? Business Today gave an answer blunt enough to sting: "Immigration is the only path to save the population crisis."⁷⁰ But Taiwan has almost stood still on this path: only about 40,000 foreigners have obtained permanent residency, accounting for less than 0.2% of the total population; in contrast, Canada imports about 500,000 immigrants a year, and Japan's specific skilled visa has taken in 300,000 in six years⁷¹. Taiwan cries out for people while keeping the door tightly closed. Pro-natalism or immigration, for this problem, Taiwan has not truly started answering.

📝 Curator's Note: There is an unspoken option standing next to "pro-natalism," and that is adaptation. The National Development Council itself has softened its stance, saying AI and automation "are expected to become the solution"⁷². This is already the language of "reorganizing production in a shrinking population," no longer the language of "reversing population." The subtext of the whole thing is quietly switching tracks: from "trying to get more Taiwanese to have children" to "accepting that Taiwan will shrink, and then figuring out how to live well." The former is a battle that cannot be won; the latter is just beginning.

Living Well on an Island Destined to Shrink

One of the cruelest and most honest things about demography is: it has almost no miracles.

Ten years of pro-natalism, throwing nearly 50 billion, couldn't even buy back the Dragon Year. The component of the fertility rate being crushed by "can't afford to raise" is far less than the component drained by "not marrying"; and for ten years, the government has been throwing money at "fear of having children," sending it to those already married, failing to connect with the part truly collapsing. Even if the money were precisely stuffed into the hands of every hesitating family, Yang Tzu-ting's lottery experiment says, it is only worth 0.07 children. Looking globally, Hungary, South Korea, France, the Nordics, no country has bought back its fertility rate to the replacement level. Taiwan's total population peaked in 2019; since then, this downhill slope, no policy can turn it around.

So what this article wants to leave you with, rather than another empty phrase of "Taiwan needs a better society," is more of a perspective adjustment: we may have been asking the wrong question. The answer to "how do we get more Taiwanese to have children" for ten years is 0.695, the empty playground at Xinwei Elementary School. The real questions to ask are the other two: how do we let those who actually want to marry and have children not be bound by a whole set of structures they cannot accept; and, how do we live with dignity, warmth, and mutual support on an island destined to shrink and age?

On that day in the Dragon Year of 2024, on the playground of Xinwei Elementary School, there was not a single first-grade freshman in a new uniform carrying a too-big backpack.

That is a very quiet scene. Quiet enough that you can't help but wonder: how much longer will this island spend, willing to stop shouting at the delivery room, turn around, and properly look at the people already here on the playground?

The song will get quieter, but as long as someone is still singing, it is not yet the last song.

Further Reading:

• Taiwan's Rural Education — When the countryside enters low birth rate forty years earlier than the city, rural primary schools are where this letter first arrives.
• Education System and Advancement Culture — University closures, student source cliffs, are inseparable from how the whole society views "advancement."
• Development of Taiwan's Long-Term Care System — When over 65 exceeds 20% and foreign caregivers hold up 215,000 beds, long-term care is the other end of low birth rate.
• Taiwan's Robotics Industry — If fewer people is the settled outcome, automation will be one of the key answers to "reorganizing production in a shrinking population."
• Expansion and Closure of Taiwan's Higher Education — The wall of low birth rate hits universities: after 58 upgraded to 148, how do schools close, and who bears the cost?

References