The Demographic Index: Construction, Histories and the Migration Offset
The APN Aggregate Demographic Index™ is composed on the platform’s composition chain from its four components at the ratified weights. It holds no data of its own, and each component is standardised against its own 2011–2025 baseline with the sample standard deviation.
| Component | Weight | Z-score | Contribution |
|---|---|---|---|
| Ageing Population & Housing Needs (21420) | 0.30 | −0.1180σ | −0.0354 |
| Demographic Trend Analysis (21440) | 0.30 | +0.3915σ | +0.1175 |
| Population Growth & Distribution (21410) | 0.25 | −0.1883σ | −0.0471 |
| Household Formation Trends (21430) | 0.15 | −0.1217σ | −0.0183 |
| Index | 1.00 | +0.0167 |
Household Economic Cohort Analysis (21450) is certified within the series at weight zero: it measures an economic position rather than a demographic quantity.
| Node | Rate | Latest (Q1 2026) | Average, 2011–2025 | Highest since 2011 | Lowest since 2011 | Position |
|---|---|---|---|---|---|---|
| 21410 | Population growth over the year | 1.43% | 1.52% | 2.53%, September 2023 | 0.10%, March 2021 | 49.4 of 175 |
| 21420 | Change in the dependency ratio | +0.43 | +0.45 | +0.92, March 2021 | −0.03, June 2023 | 75.9 of 175 |
| 21430 | Households formed over the year | 157,272 | 162,289 | 244,713, September 2023 | 54,406, March 2021 | 36.2 of 59 |
| 21440 | Change in people 65+ per 100 households | +0.69 | +0.61 | +1.07, March 2021 | +0.24, June 2023 | 75.9 of 59 |
| 21400 | Index | +0.0167σ | — | +0.71σ, September 2020 | −0.55σ, June 2023 | 62.1 of 59 |
Every high and low of the four rates falls in one of two episodes: the border closure (2020–21) and the migration surge (2023). The index’s own high and low fall in the same two episodes.
Households are modelled, not measured, between Censuses: the population multiplied by the share living in private dwellings and divided by average household size. Both ratios come from ABS Household and Family Projections, Series II. Each edition’s base-year estimate is used as an anchor (average household size 2.6019 in June 2011, 2.5748 in 2016, 2.5227 in 2021), the anchors are interpolated by quarter, and the 2021-based projection is followed from June 2021. Editions are not joined directly, because each drifts from the next edition’s base (by 2.5% at 2021). The method reproduces the editions’ base-year household totals to within ten households.
Two consequences follow. Between anchors the count moves with population, so formation follows population growth closely and cannot show households consolidating or fragmenting faster than the projections assume. And recent quarters rest on the 2021-based projection’s assumptions until the 2026 Census re-anchors them. On this count the number of households rises in every quarter of the record.
Over 2011–2025 the components’ Z-scores correlate as two opposing pairs:
| Pair | Correlation |
|---|---|
| Population growth (21410) and household formation (21430) | +0.91 |
| Rate of ageing (21420) and older residents per household (21440) | +0.92 |
| Population growth (21410) and rate of ageing (21420) | −0.90 |
| Household formation (21430) and older residents per household (21440) | −0.89 |
| Household formation (21430) and rate of ageing (21420) | −0.80 |
Net overseas migration went negative in the June quarter 2020, reached −42,600 in September 2020 and recovered to a single-quarter record of 165,500 in March 2023; it was 298,800 across 2025 and 101,000 in the March quarter 2026. Natural increase was 100,300 across 2025 and reached its lowest quarter since 2011, 22,000, in December 2025. Through the composite weights, the ageing pair (0.60) outweighs the migration pair (0.40), so the index moves opposite to migration.
Z-score of the composite in each quarter in which all four components are available.
Population growth and the rate of ageing rank against 175 quarters from September 1982, because the ABS population series run that far back; household formation, older residents per household and the index rank against the 59 quarters from September 2011, where the household count begins. The longer ranking explains an apparent mismatch at 21420: its Z-score is slightly negative (−0.1180σ) against the 2011–2025 baseline, yet its position is 75.9, because the population aged more slowly through most of the 1980s, 1990s and 2000s than in the baseline years.
The levels are at records: population, the dependency ratio and older residents per household. The rates are ordinary: three components sit within 0.2σ below their averages and the fourth, older residents per household, at +0.3915σ. Both opposing pairs are close to their centres, so the offset that dominated 2020 and 2023 is small, and the index reads +0.0167σ. The companion living-cost reading is the series’ clearest signal, at +1.2242σ for the June quarter 2026.
