Turchin is correct that tempo effects distort period Total Fertility Rate measurements—but the distortion is too small to explain the gap to replacement. When women delay childbearing, period TFR becomes artificially depressed, typically by 0.2-0.4 children per woman. However, tempo-adjusted fertility rates in most developed countries reach only 1.5-1.8, leaving a substantial "quantum deficit" of 0.3-0.6 children below the ~2.1 replacement rate. The demographic evidence shows both timing distortion and genuine fertility decline are occurring simultaneously.
Demographers have understood since Norman Ryder's work in the 1950s that delayed childbearing distorts period fertility measures. The foundational 1998 paper by John Bongaarts and Griffith Feeney formalized this insight with a correction formula: TFR*(t) = TFR(t) / (1 - r), where r is the rate of change in mean age at childbearing. When women postpone births by 0.1-0.2 years annually—common across developed countries since the 1970s—the observed TFR becomes systematically depressed below the underlying fertility level.
The empirical evidence for this distortion is substantial. During the 1980s and 1990s, several countries recorded "lowest-low" fertility rates that proved temporary. East Germany's TFR plummeted to 0.80 in 1994—far below the cohort's eventual completed fertility of ~1.5-1.6. Spain's TFR hit 1.18 in the late 1990s while cohort completed fertility was ~1.45. The Human Fertility Database now provides tempo-adjusted TFR as a standard indicator for multiple countries, reflecting broad acceptance of the methodology.
Quantitative estimates of tempo distortion range from 10-25% of observed TFR during periods of active postponement. The Max Planck Institute found US tempo distortions ranged from +28% in 1948 (when age at childbearing was falling) to -11% in 1975 (when it was rising). In Bangladesh, a 2011 study found the tempo effect depressed observed TFR by 0.51 children (from 2.81 adjusted to 2.30 observed).
The timing shift driving tempo effects is dramatic and well-documented. Across developed countries, mean age at first birth increased from 21-25 years in 1970 to 29-32 years by 2023. The United States saw an increase from 21.4 to 26.9 years. Japan, Switzerland, and Spain now exceed 30 years. Italy leads the EU at 31.8 years.
| Country | Age at First Birth (1970) | Age at First Birth (2023) | Change |
|---|---|---|---|
| United States | 21.4 | 26.9 | +5.5 years |
| Japan | 25.6 | 30+ | +5 years |
| Spain | 25.0 (1980) | 31+ | +6 years |
| Germany | ~24 | 30+ | +6 years |
This shift fundamentally altered fertility timing. In the US 1960 cohort, roughly 50% of first births occurred before age 25, compared to ~75% for the 1935 cohort. The share of women having first children at ages 30-34 doubled or tripled across developed countries. Peak fertility rates shifted from ages 25-29 in 2003 to ages 30-34 by 2023 in the EU.
Completed cohort fertility shows notable variation by country. The 1970 cohort (now aged 55, with completed fertility fully observable) achieved 1.4-1.6 children in Germany, Italy, and Japan, but 1.8-2.0 in France, Nordic countries, and the United States. This variation matters: it suggests that tempo effects interact with underlying fertility decisions differently across policy and cultural contexts.
Here is the critical finding that answers whether tempo effects can "close the gap": even after tempo adjustment, fertility rates in most developed countries reach only 1.5-1.8 children per woman, not the ~2.1 needed for replacement.
| Country | Period TFR | Tempo-Adjusted TFR | Still Below Replacement By |
|---|---|---|---|
| Spain | 1.18 | 1.56 | 0.54 |
| Czech Republic | 1.63 | 1.93 | 0.17 |
| Netherlands | 1.58 | 1.76 | 0.34 |
| Sweden | 1.63 | 1.97 | 0.13 |
| Germany | ~1.4 | ~1.6 | 0.50 |
| EU-15 average | ~1.5 | ~1.8 | 0.30 |
Wolfgang Lutz of the Vienna Institute of Demography summarized: "The measured fertility rate of around 1.5 children per woman would actually be more like 1.8, if the increase in the mean age of childbearing stopped instantly." That 0.3 adjustment is meaningful but leaves a substantial deficit.
Only the United States has shown tempo-adjusted TFR consistently above replacement, with the Human Fertility Database indicating adjusted rates above 2.1 between 1992 and 2015. France's completed cohort fertility for the 1965 cohort reached 2.03—essentially replacement. But these are exceptions. Most European countries, Japan, and South Korea show tempo-adjusted fertility of 1.5-1.8, and former Soviet countries average only 1.45 even after adjustment.
Turchin's framework assumes women who delay will eventually have children at 40. The biological and empirical evidence paints a more complex picture. While roughly 85% of women ages 40-44 have become mothers (and this share has actually increased from 80% in 2006), the remaining 15-20% face permanent childlessness—roughly half of whom wanted children but couldn't have them.
Natural fertility declines sharply with age. Monthly conception probability drops from ~20% at age 30 to less than 5% at age 40. Within a year of trying, 75-87% of 30-year-olds will conceive versus only 44-54% of 40-41-year-olds. Miscarriage rates compound the challenge: 10-16% at age 30 but 40% at age 40 and up to 80% by age 45.
IVF cannot fully compensate for age-related decline. Per-cycle live birth rates drop from 51-56% under age 35 to 25-27% at ages 38-40, then to 12-13% at 41-42, and below 5% by ages 43-44. A Yale/SART study found that for women over 43, the cumulative live birth rate after multiple IVF cycles maxes out at 5-10%. For women aged 40-42, even six complete IVF cycles yield only a 31.5% cumulative success rate.
Childlessness correlates strongly with education and delayed childbearing. Women with PhDs or professional degrees show ~20% childlessness rates (down from 35% in 1994, suggesting some "catch-up"), while those without high school diplomas show only 7%. Japan reports 28% childlessness for women born in the mid-1970s—the highest in the OECD.
Leading demographers agree that tempo effects are real and significant, but disagree about whether fertility will recover. The pessimistic camp, represented by Wolfgang Lutz's "Low Fertility Trap" hypothesis, argues that sub-replacement fertility becomes self-reinforcing through three mechanisms: fewer potential mothers (demographic momentum), declining ideal family size among younger generations (sociological adaptation), and rising aspirations combined with declining relative income for young adults (economic squeeze).
Ron Lesthaeghe, architect of the Second Demographic Transition theory, concluded: "Most cohorts of the world's white (+ Japanese) national populations born after 1960 will not make it to replacement level." His view emphasizes cultural shifts toward individual autonomy and expressive needs that structurally sustain sub-replacement fertility.
The more optimistic position, advanced by Bongaarts and Sobotka, notes that tempo-and-parity-adjusted TFR provides a "remarkable fit with completed cohort fertility" and that the 1998-2008 European fertility recovery was largely explained by declining postponement. However, even they acknowledge that tempo-adjusted rates in most countries remain below replacement.
The methodological critics, including Máire Ní Bhrolcháin, caution that tempo-adjusted TFR is "not a prediction" of future fertility and that "doubts have been expressed as to whether fertility is likely to reach the levels implied by the tempo-adjusted TFR." There are biological constraints on how late postponement can extend—women cannot simply shift all childbearing to age 45.
Turchin's observation about tempo effects is correct in direction but overstated in magnitude. The evidence supports three conclusions:
Tempo effects are real and significant. Period TFR systematically underestimates underlying fertility by 0.2-0.4 children per woman when women are actively delaying childbearing. This explains why extremely low TFRs of 1.1-1.3 observed in some countries proved temporary and why period TFR can diverge substantially from cohort completed fertility.
Tempo effects cannot fully bridge the gap to replacement. Even tempo-adjusted fertility rates reach only 1.5-1.8 in most developed countries, leaving a "quantum deficit" of 0.3-0.6 children below replacement. This deficit reflects genuine behavioral change: women are completing their childbearing with fewer children than previous generations.
Delayed childbearing does not guarantee eventual births. While most women who delay still have children, 15-20% reach their mid-40s childless, and roughly half of these cases are involuntary. IVF success rates at 40+ are far lower than commonly believed, and many women who intend to have children later end up not having them at all.
The balanced expert view holds that tempo effects explain roughly half of the apparent deficit between observed TFR and replacement, while genuine fertility decline explains the other half. Postponement alone cannot stabilize population in the absence of increased quantum fertility—women actually choosing to have more children, not just having them later.