Human capital growth persists with upcoming population decline

The world's population has grown exponentially over the past few centuries, rising from around 1 billion in 1800 to just above 8 billion today. However, demographers across the globe agree that this remarkable growth will likely grind to a halt before the end of the 21^st century. This anticipated end to world population growth is the culmination of the demographic transition, a process through which nations progress from high fertility and mortality rates to low rates of both over time.

This transition has happened at different times across the regions of the world. Many European and North American countries were the first to complete the transition in the late 19^th and early 20^th centuries, as family planning spread and child mortality plummeted. In recent decades, populous nations like China, Brazil, and Mexico have followed suit, with their total fertility rates dropping below the population replacement level of 2.1 children per woman. Recently, the largest country in the world, India, saw its fertility rate falling below replacement level for the first time. While sub-Saharan Africa is still lagging behind in the demographic transition, the fertility rate has also been declining steadily and is likely to hit replacement level in the future.

At the same time, the world is undergoing a pivotal education transition. As countries develop economically and socially, more children gain access to schooling opportunities. There has been a persistent secular trend, with each new generation achieving higher levels of schooling than previous cohorts. Indications are that this trend will continue into the future. As a result, the younger, more educated cohorts will gradually replace and phase out the older, less formally educated generations over the coming decades.

Education is a powerful force that shapes human potential and drives societal progress on multiple fronts. Higher educational attainment opens doors to better career opportunities, higher lifetime earnings, improved health outcomes, and enhanced critical thinking abilities. But the benefits of education extend far beyond the personal level. More educated societies tend to have lower levels of poverty and crime and experience faster economic growth, higher innovation, stronger institutions, and more robust civic engagement, while being more resilient to climate change and crises in general.

Traditional analyses focusing solely on age structure as a predictor of economic growth do not fully capture the ability of the working-age population to productively support the dependent population, which is strongly influenced by factors such as educational attainment, labor force participation, and productivity. For these reasons, we developed a new Human Capital Weighted Population (HCWP) dataset^[i] in a recent paper published in Scientific Data. This indicator provides a more comprehensive measure of a population's productive capacity and human capital heterogeneity than traditional demographic indicators like total population size, average educational attainment levels, or age structure.

The HCWP adjusts the working-age population using weights based on educational attainment and education system quality, summarizing human capital heterogeneity in a single metric. The HCWP represents the number of people with the world average level of human capital in 2015 who would have the same productive capacity as the actual population. By accounting for both the quantity and quality of education across populations, the HCWP aims to enhance the explanatory power of statistical models used to assess the socioeconomic impact of demographic change, leading to more accurate economic growth forecasts. By applying weights to population estimates by the Wittgenstein Centre for Demography and Globa Human Capital (WIC) by age, sex, and education, we provide HCWP estimates for 185 countries from 1970 to 2020 and forecasts for five Shared Socioeconomic Pathway (SSP) scenarios from 2020-2100.

But, how do these new data help us to understand the interaction between population and productive capacity dynamics? The figure above illustrates the projected trajectories of the total population, the working-age population (aged 20-64), and the newly developed HCWP for the world, spanning the period from 1970 to 2100 under the SSP2 scenario.

While total population and working-age population show similar trends, with a peak in the second half of the 21^st century followed by a slow decline, the HCWP follows a different trajectory. Initially lower than the working-age population, the HCWP rises at a steeper rate than the other two measures. Although the rate of increase will slow down in the last decades of the 21^st century, no decline is expected. This divergence suggests that after mid-century, as the working-age population begins to decline, its human capital will continue to increase, offsetting the decline in raw numbers.

Analyzing the HCWP also provides a different view of population dynamics across world regions. The chart above compares the regional distribution of the world population versus its human capital-weighted population, both historically from 1970 to 2020 and projected from 2020 to 2100 under the SSP2 scenario. A striking difference emerges.

Up to the 2000s, Europe and North America have a significantly higher share of the HCWP compared to their total population share, as these developed regions have an advantage in terms of educational attainment and the quality of their education system. In 1970, Europe and North America together accounted for only 26% of the world population, but 60% of the HCWP. Conversely, Asia represented 60% of the world population but only 33% of the HCWP. Africa had a minimal HCWP share in the 1970s, despite comprising 10% of the world population. By 2020, Asia had significantly improved its educational attainment, aligning its HCWP share (60%) with its population share.

Looking forward, Asia's population is projected to decline, while Africa's population will continue to grow and further educational progress is expected in the continent. By 2100, Africa is expected to account for a third of the world's HCWP, and all continents will have a HWCP matching their population share. However, this reduction in inequalities in human capital among regions would only materialize if progress in education in less developed regions continues and is translated into higher productivity per worker. This analysis highlights how conventional population measures can obscure substantial inequalities in productive capacities driven by human capital differentials across regions.

While the world's total population is projected to peak and then decline later this century, the human capital of the global population is expected to continue rising. By accounting for heterogeneity in educational attainment and quality across populations, the new human capital-weighted population metric provides a more comprehensive measure of a population's overall productive capacity compared to traditional demographic indicators. As developing regions make progress in education, their share of the world's human capital-weighted population will continue to grow, reflecting their rising economic potential.

While demographic trends can seem daunting, this analysis underscores how investments in human capital development through expanded access to quality education can prepare societies to thrive despite population decline. By improving our ability to quantify population dynamics, the human capital-weighted population offers a powerful new lens for assessing economic impacts and informing policies to harness the opportunities of the 21^st century demographic landscape.

Reference:
Marois, G., Crespo Cuaresma, J., Zellmann, J., & Reiter, C. (2024). A dataset of human capital-weighted population estimates for 185 countries from 1970 to 2100. Nature – Scientific Data DOI: 10.1038/s41597-024-03466-y

^[i] The complete dataset is available on Zenodo: Marois, G., Crespo Cuaresma, J., Zellmann, J. G. & Reiter, C. Human Capital-weighted population estimates for 185 countries from 1970 to 2100 (1.0) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.10573016 (2024).

Note: This article gives the views of the author, and not the position of the IIASA blog, nor of the International Institute for Applied Systems Analysis.