Show notes
"Watch who you’re calling childless
Women in America are having as many babies over their lifetimes as they did two decades ago
A set of public signs showing women, pregnancy, baby care, strollers, and children playing
Dec 18th 2025 | 5 min read
**A colony of **sentient lizards, living on a distant planet. It comprises 100 females and 100 males. Each lives for 100 years. There is only one male and one female of any age (eg, one 15-year-old female, one 74-year-old male). The planet is safe and peaceful: no lizards die from disease or violence.
But every year the two 100-year-olds pass away. And every year two babies are born, the fruits of an unusual reproductive cycle. At age 24 every female produces a male baby; at 26, a female. She will never have any more. The population of Planet Lizard is thus fixed at 200. Now imagine an asteroid hits Planet Lizard, altering its atmosphere and climate. At first little seems to have changed. All the lizards miraculously survive, and they rejoice at their good fortune.
But it soon becomes clear something is wrong: the two females, aged 24 and 26, who are due to produce baby lizards do not. The population shrinks to 198. The next year, the same thing happens. The colony dwindles to 196. Concern turns to panic. Prophecies of extinction are splashed across the front pages of lizard newspapers. Councils of senior lizards are convened to devise plans to revive fertility. Nothing works.
Ten years after the asteroid struck, the remaining 180 lizards are resigned to their fate: ageing, death and extinction, two by two. And then: a miracle. The two females who failed to have babies at 24 and 26 are suddenly fertile, at 34 and 36. Soon, two infants are born, to general ecstasy. The next year, the miracle is repeated. And so on. The population stabilises and (a century later) climbs back to 200. Ultimately, the asteroid changed the timing of lizard births but not their number. This fable may seem silly, but it contains a lesson. For decades, across the West, the commonest measure of how many children women have, on average—the total fertility rate (tfr)—has been dropping. It has fallen well below the “replacement rate” (needed to keep a population constant) of 2.1. Between 2007 and 2022 America’s tfr dropped from 2.1 to just 1.6. Warnings of probable population decline, including from this newspaper, abound. Pronatalists are casting around frantically for solutions. A shrinking (human) population could be cause for concern. It means fewer working-age people to support the old. Fiscal crisis could loom. Fewer minds would yield fewer innovative ideas. Less scope for division of labour would mean a less efficient economy.
But tfr has flaws. It measures the number of children a hypothetical woman might have in her life, based on current birth rates for women of different ages. To calculate it, you start by dividing the total number of births to a given group of women (aged, say 15-19 or 35-39) by the number of women in that group. Those group birth rates are then added up and multiplied by the number of years in each group (ie, five).
The result is the number of children a woman will have if, aged 15-19, she has the same number of babies as the average woman of 15-19 is having today, then the same number at 20-24 as the average woman of 20-24 is having today, and so on. It is almost sure to be wrong. That does not make TFR useless. It captures what demographers call “tempo”—the timing of an event, like a birth.
But as a measure of “quantum”—the total number of births—it is likely to mislead. In effect it assumes that a young woman will have babies not only at the same rate as her contemporaries, but also at the same rate as women five or ten years older. That is not necessarily so. A different measure, the completed fertility rate (CFR), captures the average number of births a woman has by the end of her child-bearing years (put at 44).
America’s CFR has not fallen at all over the past two decades. In fact, according to an analysis of census data by Mike Konczal, a former Biden administration economist, it has risen slightly, from 1.91 in 2000 to 1.97 in 2024. Chart: The Economist Back on Planet Lizard, the TFR dropped to zero overnight. After ten years, when females started having babies again, it jumped back to 2. The CFR never budged. The timing of births simply shifted, and eventually the tfr caught up with the cfr. In a doomsday scenario, in which the asteroid did render the lizards infertile, the cfr would have dropped to meet tfr. Which scenario is playing out in the West? Many demographers, including Lyman Stone, of the Institute for Family Studies, a think-tank, point out that tfr is typically a strong predictor of cfr. But there is evidence that women are deferring, not forgoing, childbirth.
In 2000 by age 26 the average American woman had produced one child; the average 32-year-old, 1.6 children; the average 40-year-old, 1.9. In 2024 the average 26-year-old had 0.6 children and the average 32-year-old 1.2. But the average 40-year-old has still had 1.9, having delayed in her 20s and caught up in her 30s. Economies of scales Does this mean, as with the lizards, that tfr will rise to meet cfr? Historical examples are suggestive. As women delayed childbearing in Sweden the tfr dropped sharply in the 1980s, before recovering in the 1990s (though it has fallen since). The cfr, at around 2, barely budged. In America the question is pertinent. The commonest ages are 33 and 34, because of a baby boom in 1990 and 1991. Women aged 34 have had, on average, 1.46 children each. As this group approaches their 40s, many more births may be on the way. As for younger women, those born in 2000 have had fewer children than those born in 1990 had had by 2015.
Perhaps they will catch up—but that won’t be known for at least a decade. To be sure, an average of just under two births per woman is still not “replacement level”. But if that truly is the long-term average, panic is scarcely justified. It would mean a gradual population decline: tricky, but much more manageable than implied by TFR. It is easy to fear the end times. But the problem may be one that humans, like our faraway lizards, never face.
美國女性一生中生育的孩子數量，與二十年前其實一樣多
2025 年 12 月 18 日｜閱讀時間 5 分鐘
想像一下，在一顆遙遠的星球上，有一個由具備感知能力的蜥蜴所組成。這個社會共有 100 隻雌性與 100 隻雄性蜥蜴。每一隻蜥蜴都能活到 100 歲，而且每個年齡層都恰好只有一隻雄性與一隻雌性（例如：只有一隻 15 歲的雌蜥蜴，一隻 74 歲的雄蜥蜴）。這顆星球安全又和平，沒有任何蜥蜴死於疾病或暴力。然而，每年都會有兩隻 100 歲的蜥蜴自然死亡；同時，也會誕生兩隻新生蜥蜴，這來自一種特殊的繁殖週期。每隻雌蜥蜴在 24 歲時會生下一隻雄性幼蜥；在 26 歲時，會生下一隻雌性幼蜥。此後，她將不再生育。於是，「蜥蜴星球」的人口始終固定在 200。
接著，想像一顆小行星撞擊了蜥蜴星球，改變了它的大氣與氣候。一開始，看似毫無異狀。所有蜥蜴奇蹟似地存活下來，並為自己的幸運而歡欣鼓舞。但很快地，問題浮現了：原本應該在 24 與 26 歲生育的兩隻雌蜥蜴，沒有生下幼蜥。人口減少到 198。隔年，同樣的事情再度發生，人口降至 196。憂慮迅速轉為恐慌，關於滅絕的預言登上蜥蜴報紙的頭版。長老議會召開，試圖找出恢復生育力的方法，但一切都徒勞無功。
十年後，剩下的 180 隻蜥蜴只能接受他們的命運：逐漸老去、死亡，然後滅絕——每年少兩隻。
然後，奇蹟出現了。
那兩隻在 24 與 26 歲未能生育的雌蜥蜴，竟然在 34 與 36 歲時重新恢復生育能力。不久後，兩隻幼蜥誕生，整個星球陷入狂喜。隔年，奇蹟再次發生。接著年復一年。人口穩定下來，並在一個世紀後回升至 200。
最終，小行星改變的只是蜥蜴生育的「時間點」，而非「總數」。
這個寓言或許看來荒謬，但其中蘊含了一個重要教訓。數十年來，西方國家最常用來衡量女性平均生育子女數的指標——總和生育率（Total Fertility Rate，簡稱 TFR）——持續下降，且已遠低於維持人口穩定所需的「更替水準」2.1。2007 年至 2022 年，美國的 TFR 從 2.1 降至僅 1.6。各種關於人口即將衰退的警告不絕於耳，其中也包括本刊。鼓吹提高生育率的人士正四處尋找解方。
人口萎縮確實值得憂慮。這意味著勞動年齡人口減少，難以支撐高齡人口；財政危機可能浮現；創新思維的數量下降；分工的空間縮小，經濟效率隨之降低。
然而，TFR 本身存在缺陷。它是根據不同年齡層女性「當前」的生育率，推算出一名假想女性一生可能生育的子女數。計算方式如下：先將某一年齡組（例如 15–19 歲或 35–39 歲）的出生總數，除以該年齡組女性人數，得出該組的生育率；再將各年齡組的生育率加總，並乘以每組的年數（通常為五年）。結果是一個假設：一名女性在 15–19 歲時的生育行為，會與當今該年齡層女性相同；在 20–24 歲時，也會與當今該年齡層女性相同，依此類推。這種假設幾乎必然與現實不符。
這並不代表 TFR 毫無用處。它確實能反映人口學家所稱的「節奏」（tempo）——事件（如生育）發生的時間點；但作為衡量「總量」（quantum，即一生生育數）的指標，它往往具有誤導性。它實際上假設年輕女性不僅會以同齡人的生育率生育，未來也會以年長女性目前的生育率生育，而這並非必然。
另一個指標——完成生育率（Completed Fertility Rate，CFR）——則計算女性在生育年齡結束（通常定為 44 歲）時，平均實際生育的子女數。根據前拜登政府經濟學家 Mike Konczal 對美國人口普查資料的分析，美國的 CFR 在過去二十年間並未下降，反而略有上升，從 2000 年的 1.91 提高至 2024 年的 1.97。
回到蜥蜴星球：TFR 曾在一夜之間降為零；十年後，當雌蜥蜴重新開始生育時，TFR 又回升至 2。但 CFR 始終未曾改變。生育的時間只是延後，最終 TFR 追上了 CFR。若小行星真的使蜥蜴永久不孕，那麼 CFR 才會下滑，與 TFR 接軌。
那麼，西方世界正在經歷的是哪一種情況？
許多人口學家（包括家庭研究所的 Lyman Stone）指出，TFR 通常是 CFR 的良好預測指標。然而，越來越多證據顯示，女性是在延後生育，而非放棄生育。
2000 年時，26 歲的美國女性平均已有 1 名孩子；32 歲者為 1.6；40 歲者為 1.9。到了 2024 年，26 歲女性平均只有 0.6 名孩子，32 歲為 1.2；但 40 歲女性仍是 1.9——她們只是把原本 20 多歲的生育延到 30 多歲完成。
規模經濟
這是否意味著，美國的 TFR 會像蜥蜴一樣，最終回升以追上 CFR？歷史案例提供了一些線索。瑞典在 1980 年代女性延後生育時，TFR 曾急劇下降，並於 1990 年代回升（儘管之後再度下滑）；但 CFR 始終穩定在約 2。
在美國，這個問題尤為關鍵。由於 1990 與 1991 年的嬰兒潮，當前人數最多的女性年齡層為 33 與 34 歲。34 歲女性目前平均已有 1.46 名孩子。隨著她們邁入 40 多歲，更多出生可能仍在路上。至於年輕世代，2000 年出生的女性，到目前為止生育數少於 1990 年出生者在 2015 年時的水準。她們是否會「追上來」，仍需至少十年才能確認。
誠然，每名女性平均不到兩個孩子，仍低於更替水準。但若這真是長期趨勢，恐慌其實並不必要。這代表的是一個緩慢、可控的人口下降，而非 TFR 所暗示的危機規模。人類很容易想像末日降臨。但這個問題，或許終究不會真正發生——正如那些遙遠星球上的蜥蜴一般。 ■
--
Hosting provided by SoundOn