pop·u·la·tion ag·ingAmE /ˌpɑpjəˈleɪʃən ˈeɪdʒɪŋ/ · BrE /ˌpɒpjʊˈleɪʃən ˈeɪdʒɪŋ/nounA progressive shift in the age structure of a population toward older groups — driven by declining fertility, rising life expectancy, and the changing balance of births, deaths, and migration.
About
This page accompanies the paper:
Fernandes, Fernando, Cássio M. Turra, and Eduardo L. G. Rios Neto. 2023. “World Population Aging as a Function of Period Demographic Conditions.”Demographic Research 48 (13): 353–72. https://doi.org/10.4054/DemRes.2023.48.13.
Editor's Choice, Volume 48.
Across 236 countries and areas from 1950 to 2100, the Preston–Himes–Eggers period identity (Preston et al. 1989) decomposes annual change in population mean age into contributions from births, deaths, and net migration, drawing on United Nations population estimates and projections.
This companion recomputes the paper’s rejuvenating-effects figure (Fig. 2) with the medium variant of World Population Prospects 2024 (United Nations 2024), reprints its stage classification (Table 1), and provides an interactive table of population aging indicators by geographic level. The paper used World Population Prospects 2022 (United Nations 2022), so the figure and the indicators here are not the published values.
Figures
Figure 1 · Rejuvenating effect of births by rejuvenating effect of deaths
236 countries and areas, 1950–2100.
Source: Prepared by the authors from United Nations (2024).
Notes: Recomputed with World Population Prospects 2024 following Fig. 2 of Fernandes et al. (2023), which used World Population Prospects 2022 (United Nations 2022). Each point is a country or area in one calendar year; the vertical axis is the rejuvenating effect of births (bNa), the horizontal axis that of deaths (dDaNa). The mean age stability line marks where their sum is 1 and exactly offsets natural aging; points above it belong to populations becoming younger, those below to aging populations, net migration aside (ρ, see Data). Not shown: 30 country-years beyond the axis ranges (deaths −0.4 to 0.8, births 0 to 1.4): China, Hong Kong SAR 14, Monaco 8, Puerto Rico 6, Rwanda 1, South Sudan 1; and the 17 years in which Tokelau has no mean age at death, hence no rejuvenating effect of deaths.
Table 1. Stages of population aging according to the rejuvenating effects of births and deaths
Stage
Mean age
Rejuvenating
effect of
births
Rejuvenating
effect of
deaths
Combined
rejuvenating
effect of births
and deaths
Ap
dAp/dt
bNa
dDaNa
bNa + dDaNa
1
Decreasing
Negative
1.0, 1.4
−0.4, 0.0
> 1
1A
Minimum
Zero
1.0
0.0
= 1
2
Increasing
Positive
0.6, 1.0
0.0, 0.2
< 1
3
Increasing
Maximum
0.6
0.2
< 1
4
Increasing
Positive
0.4, 0.6
0.2, 0.6
< 1
4A
Maximum
Zero
0.4
0.6
= 1
5
Decreasing
Negative
0.4
> 0.6
> 1
Source: Fernandes et al. (2023), Table 1. bNa and dDaNa are rejuvenating effects of births and deaths; bNa + dDaNa = 1 on the mean age stability line (Figure 1). 1A and 4A are the turning points at which mean age reaches its minimum and its maximum; 1 to 5 are the stages around them. Stage 5 is projection-only in the paper.
Stage assignment in this companion: the WPP 2024 pipeline gives each unit-year the first stage whose rule it meets, in the order 1: bNa > 1, dDaNa < 0 and sum > 1; 1A: within ±0.02 of (1.0, 0.0); 2: 0.62 ≤ bNa ≤ 0.98, 0.02 ≤ dDaNa ≤ 0.18 and sum < 1; 3: within ±0.02 of (0.6, 0.2); 4: 0.38 ≤ bNa ≤ 0.58, 0.22 ≤ dDaNa ≤ 0.58 and sum < 1; 4A: within ±0.02 of (0.4, 0.6); 5: bNa ≤ 0.5 and sum > 1. Each build checks this rule against every row. It departs from the table above in two places: stage 4 starts at bNa = 0.38, not 0.4, and stage 5 admits bNa up to 0.5. Rows that meet no rule are left Unclassified: 15,075 of 39,109 (38.5%) across the three levels. 4,383 of them are populations past stage 4 that have not reached turning point 4A, for which Table 1 has no stage: the effect of births is below 0.38 while the sum is still below 1 (Japan and Italy in 2023, for example). 17 are the years without a mean age at death (Tokelau); the others lie between or outside the stage rules.
Data
Population mean age Ap changes over a calendar year according to the Preston–Himes–Eggers decomposition (PHE-I):
dAp/dt = 1 − (bNa + dDaNa) − ρ
The 1 is natural aging: a year later, every survivor is one year older. bNa = b·Na is the rejuvenating effect of births, which enter at age 0, Na years below the mean age. dDaNa = d·(Da − Na) is the rejuvenating effect of deaths, positive when people die older than the population’s mean age. ρ is the residual: the effect of net migration plus the error of applying the continuous-time identity to annual data. For the World, where net migration is zero, ρ averages 0.004 in absolute value. Positive values of each term lower the mean age. See eq. (1) in Fernandes et al. (2023).
Filters.Level selects World (one series), Subregion (22 UN subregions), or Country/Area (236 units). Region is the UN SDG continental grouping and narrows the country list (not a separate WPP aggregate). Subregion and Country pick geographic units within the selected level. Year range spans 1950–2100. Estimate vs projection follows UN WPP labels. Stage filters by the five stages and two turning points of Table 1 (Figure 1 maps country-years to stages); Unclassified selects the rows that meet no stage rule (see the note under Table 1).
Variables
Notation
CSV column
Description
Ap
Nat
Population mean age at the start of year t (1 January)
dAp/dt
dNat
Observed change in mean age over year t, Ap(t+1) − Ap(t); natural aging alone would make it 1
Na
Na
Mean age of the population in year t, the reference age of the births and deaths terms
Da
Da
Mean age at death in year t (blank for Tokelau 2001 and 2004–2019)
bNa
bNa
Rejuvenating effect of births, b·Na
dDaNa
dDaNa
Rejuvenating effect of deaths, d·(Da − Na)
bNa + dDaNa
bNa_dDaNa
Combined rejuvenating effect of births and deaths
ρ
rho
Residual, 1 − (bNa + dDaNa) − dAp/dt: net migration plus discrete-time error; positive values lower the mean age (ρio in the paper)
b, d
cbr, cdr
Crude birth and death rates, per person per year
CNMR
cnmr
Crude net migration rate, per person per year (positive: net immigration)
Stage
stage
Aging stage of Table 1: stages 1–5 and turning points 1A, 4A; blank when no stage rule is met (shown as Unclassified in the table)
Geographic unit, year t, level, ISO alpha-3 code, SDG region, UN subregion, and UN estimate or projection
Notation: PHE-I expression (Fernandes et al. 2023), after equation (1) of Preston et al. (1989). Column names are those of the full data file, phe-indicators-interactive.csv (10 MB, every row and column at full precision). Filtered downloads rename Nat, dNat and bNa_dDaNa to Ap, dAp_dt and bNa_plus_dDaNa, keep a blank stage blank, and leave out iso3_code, Na and Da.
Sources
Preston, Samuel H., Christine Himes, and Mitchell Eggers. 1989. “Demographic Conditions Responsible for Population Aging.” Demography 26 (4): 691–704. https://doi.org/10.2307/2061266.
United Nations. 2022. World Population Prospects 2022: Online Edition. United Nations, Department of Economic and Social Affairs, Population Division.
United Nations. 2024. World Population Prospects 2024: Online Edition. United Nations, Department of Economic and Social Affairs, Population Division. https://population.un.org/wpp/.
