The End of Hyperbolic Growth in Human Population and CO 2 Emissions

Victor M. Yakovenko · 2025

Read The End of Hyperbolic Growth in Human Population and CO 2 Emissions by Victor M. Yakovenko (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-11038.txt.

Introduction

Victor Yakovenko presents a 2025 study of human population growth and carbon dioxide in the air. In 1960, researchers found that growth followed a hyperbolic curve, which pointed to a runaway peak in 2026. Yakovenko rechecks that claim with data running from 10,000 BCE to 2023 CE.

Early growth was exponential but very slow, then sped up over the ages. Around 1700, with the Industrial Revolution and large-scale use of fossil fuels, growth entered the hyperbolic regime. In this century, the curve bends away from a runaway peak. Instead, the model points to a population peak around 2030, spread over a width of about 32 years. The rise in carbon dioxide fits a matching curve tied to the same dates.

Catalog record

CREATORVictor M. Yakovenko
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE59 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-11038

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

The End of Hyperbolic Growth in Human Population and CO 2 Emissions Note: Published in Physica A 661 , 130412 (2025), https://doi.org/10.1016/j.physa.2025.130412 Victor M. Yakovenko Abstract Using current empirical data from 10,000 BCE to 2023 CE, we re-examine a hyperbolic pattern of human population growth, which was identified by von Foerster et al. in 1960 with a predicted singularity in 2026. We find that human population initially grew exponentially in time as N ⁡ ( t ) ∝ e t / T N(t)\propto e^{t/T} with T = 2080 T=2080 years. This growth then gradually evolved to be super-exponential with a form similar to the Bose function in statistical physics. Around 1700, population growth further accelerated, entering the hyperbolic regime as N ⁡ ( t ) ∝ ( t s − t ) − 1 N(t)\propto(t_{s}-t)^{-1} with the extrapolated singularity year t s = 2030 t_{s}=2030 , which is close to the prediction by von Foerster et al. We attribute the switch from the super-exponential to the hyperbolic regime to the onset of the Industrial Revolution and the transition to massive use of fossil fuels. This claim is supported by a linear relation that we find between the increase in the atmospheric CO 2 level and population from 1700 to 2000. In the 21st century, we observe that the inverse population curve 1 / N ⁡ ( t ) 1/N(t) deviates from a straight line and follows a pattern of “avoided crossing” described by the square root of the Lorentzian function. Thus, instead of a singularity, we predict a peak in human population at t s = 2030 t_{s}=2030 of the time width τ = 32 \tau=32 years. We also find that the increase in CO 2 level since 1700 is well fitted by arccot ⁡ [ ( t s − t ) / τ F ] {\rm arccot}[(t_{s}-t)/\tau_{F}] with τ F \tau_{F} =40 years, which implies a peak in the annual CO 2 emissions at the same year t s = 2030 t_{s}=2030 . 1 Introduction In 1960, Heinz von Foerster et al. published a paper [ 1 ] with the provocative title ‘‘Doomsday: Friday, 13 November, A.D. 2026.’’ Fitting the data then available, the paper argued that growth of human population is described by a hyperbolic function of time that extrapolates to infinity in the year 2026. The particular day of 13 November was a joke, as it is the birthday of Heinz von Foerster. 1 1 1 https://en.wikipedia.org/wiki/Heinz_von_Foerster A historical and philosophical review of the work by von Foerster et al. is presented by Umpleby [ 2 ] . This predicted “doomsday” seems to be actually arriving now. An article [ 3 ] “Suddenly there aren’t enough babies: The whole world is alarmed” by Greg Ip and Janet Adamy, published in The Wall Street Journal in May 2024, opens with a rather dramatic statement: “The world is at a startling demographic milestone. Sometime soon, the global fertility rate will drop below the point needed to keep population constant. It may have already happened.” The article implies that global human population is about to start decreasing. The question thus arises: what will happen in 2026? Will human population go to infinity or begin to collapse? In this paper, we re-examine human population dynamics using the currently available data from 10,000 BCE to 2023 CE. We present quantitative fits of these empirical data using simple mathematical functions that we describe along the way together with the relevant literature. The aim of these fits is to capture the dominant features of the data on a large scale, while avoiding fits of minor fluctuations and deviations on a small scale. Highly detailed accuracy is not expected from these fits; yet they provide important insights into human population dynamics. In Secs. 2 – 5 , we demonstrate that human population growth has gone through four stages described by distinct mathematical equations and culminating in the ever-increasing pace of growth analyzed in Sec. 6 . In Sec. 7 , we compare historical data for human population and \rm CO_{2} level in the atmosphere and demonstrate a linear relationship between them. Moreover, we present a mathematical fit of the time-dependent level {\rm CO_{2}}(t) and make a near-term forecast. From a unified perspective, we introduce coupled differential equations for human population and \rm CO_{2} emissions in Sec. 8 and further elaborate on them in Sec. 9 using a complex variable. We compare our predictions with the reports by the Population Division of the United Nations (UN) in Sec. 10 . Our paper leaves aside dynamics of economic and financial indicators, which were considered, e.g., in Refs. [ 4 , 5 , 6 , 7 , 8 ] , and spatial dynamics [ 9 ] . More generally, the subject of our paper fits into a broader field of “social physics” [ 10 ] , which applies mathematical methods developed in physics to social phenomena. We conclude that, paradoxically, both claims cited at the beginning are inherently related and represent two sides of the same coin. On the one hand, we confirm the hyperbolic growth pattern identified by von Foerster et al. [ 1 ] , but with a slightly different year t_{s}=2030 of the extrapolated singularity. On the other hand, human population cannot become literally infinite. Thus, the extrapolated singularity is avoided in close proximity and transforms into a peak in human population at t_{s}=2030 . After this peak, the world population is expected to decrease, which is consistent with the conclusions of Ref. [ 3 ] . Moreover, our analysis also predicts a peak in the annual \rm CO_{2} emissions at the same year t_{s}=2030 and a subsequent decline, while the overall atmospheric level of \rm CO_{2} will continue to increase monotonically in time. The primary source of the population data for our paper is the aggregator Our World in Data [ 11 ] . As indicated on their Web site, the historical estimates of population for the time period from 10,000 BCE to 1799 CE come from the History Database of the Global Environment (HYDE, v 3.3) of the Netherlands. For the time period 1800–1949, the historical estimates are from Gapminder (v7). For the time period 1950–2023, Our World in Data uses the population records from the UN World Population Prospects (2024 revision) and updates them biannually. However, instead of the UN data, we use the population data from the World Bank (WB) available from 1960 to the latest year 2023 [ 12 ] . The WB population data are updated annually and represent the middle of the corresponding year. Their single data variable SP.POP.TOTL for the total world population [ 12 ] is easier to handle than a multitude of data entries contained in the UN files. For some reason, the WB numbers for total population seem to be slightly lower in recent years than the UN numbers. Our mathematical fitting applies to the WB data [ 12 ] , but not necessarily to the UN data. 2 Exponential growth from 10,000 to 3000 BCE The growth rate r=b-d of the population N(t) in time t is the difference between birth rate b and death rate d . Taking this rate to be

Read the complete text