Where the Cost-of-Living Pressure Actually Sits

The household type usually cited as Australia’s bottom-of-market cost-of-living benchmark is the Pensioner and Beneficiary Living Cost Index (PBLCI) — the ABS’s own indicator for pension and welfare-recipient households. But PBLCI isn’t a population of its own — it’s a combined reading of two groups the ABS already tracks separately, Age Pensioner and Other Government Transfer Recipient households. Break that combination apart, and new data for the June 2026 quarter tells a sharper story: Age Pensioner households alone are running hotter than the PBLCI blend suggests, hotter in fact than any of the ABS’s four household types. That gap has been widening for two years, not appearing overnight.

Three populations elevated, two below — and PBLCI blends two of them

The ABS Selected Living Cost Indexes (cat. 6467.0) publish index numbers for four independent household populations — Employee, Age Pensioner, Other Government Transfer Recipient, and Self-Funded Retiree — plus PBLCI, a fifth published number that is not an independent population but a separately-weighted blend of the Age Pensioner and Other Government Transfer Recipient groups specifically. All five are shown below because all five are genuinely published, distinct index readings; only the underlying populations overlap.

Of the four independent populations, Age Pensioner households are running the hottest — 0.78 percentage points above CPI, ahead of Other Government Transfer Recipient households (+0.69pp). PBLCI, the blended measure covering both of those populations together, sits between them at +0.71pp — arithmetically consistent with being a weighted combination of the other two, not a separate data point in its own right. Self-Funded Retiree and Employee households are tracking at or below CPI. The real split is between government-payment-dependent households and everyone else — not a four- or five-way race between unrelated groups.

A two-year trend, not a one-quarter blip

The PBLCI-CPI spread has been positive for eight consecutive quarters. It is not a stable 0.7 percentage points either — it moves quarter to quarter with underlying price shocks, but it has not returned to zero since the run began.

The spread peaked at 1.08 percentage points in September 2024, narrowed to 0.42 by December 2024, then widened again to 0.91 in March 2026 before easing slightly to 0.71 in the latest quarter. That volatility matters for how the figure should be read: it is not evidence of a single ongoing shock, but of a persistent structural gap that different quarterly price movements push wider or narrower without ever closing it.

Age Pensioner households run hotter than the pension/benefit population as a whole

PBLCI’s own methodology explains the ordering rather than contradicting it. PBLCI is constructed from Age Pensioner and Other Government Transfer Recipient households combined, using its own expenditure weights rather than a simple average of the two. That Age Pensioner households alone (+0.78pp) run hotter than the PBLCI blend (+0.71pp) means Other Government Transfer households are pulling the blended figure down slightly — consistent with their own independently-published reading of +0.69pp. This is exactly the arithmetic you’d expect from a blended index sitting between its two components, not a coincidence. The more useful question is why Age Pensioner households, specifically, are running hottest of the four independent populations.

That cost is Housing. Age Pensioner households recorded the largest Housing sub-index rise of any household type in the June 2026 quarter — 10.83% year-on-year, ahead of even PBLCI’s 9.03%. A companion piece, “It’s Not Interest Rates,” sets out why that Housing figure is not a mortgage-rate story, and what it says about where the real cost pressure sits for retirement-age Australians regardless of pension status.

Housing isn’t the only place these populations diverge

Excise-sensitive spending shows a similarly uneven pattern, and one that cuts across the retirement/working-age line rather than following it. Other Government Transfer Recipient households — a working-age cohort receiving benefits other than the Age Pension — recorded the fastest Alcohol and Tobacco cost growth of any household type, more than double the Self-Funded Retiree rate.

This is most plausibly a consumption-weighting effect rather than a price effect — the underlying tobacco excise and alcohol price movements are national, not household-type-specific, so the divergence reflects how much of each cohort’s total spending sits in this category, not that Other Government Transfer Recipient households are paying a different price. It is still a real distributional finding: a national excise-driven price rise falls most heavily on the cohort with the largest budget share in that category, and that cohort is not the one most commentary defaults to.

What this means for reading “cost of living” as a single number

A single national CPI figure obscures both of the findings above. It cannot show that Age Pensioner households are running hotter than the PBLCI blend meant to represent the pension/benefit population, because CPI does not track cohorts at all. And it cannot show that the cohorts under the most pressure are not uniformly the ones assumed — Other Government Transfer Recipient households, not Age Pensioners, show the steepest Alcohol and Tobacco growth, even as Age Pensioners show the steepest Housing growth. Treating “cost of living” as one number, or even as one “vulnerable cohort,” misses both.

Two caveats worth stating plainly

First, on population overlap: PBLCI’s reading is not independent evidence alongside Age Pensioner and Other Government Transfer Recipient households — it’s a blend of exactly those two, so treat the five figures in the first chart as four populations plus one combined view, not five separate data points. Second, on basket comparability: the Living Cost Indexes and CPI are not measuring identical baskets — LCI includes mortgage interest and consumer credit charges directly, which CPI’s acquisitions-based approach to housing deliberately excludes. Some of the gap between any household-type LCI and CPI reflects this measurement difference, not pure distributional inequality. For the three populations running above CPI here (Age Pensioner, Other Government Transfer Recipient, and the PBLCI blend of both), that caveat matters less than it might: as the companion piece shows, the driver is concentrated in non-financing housing costs, not the mortgage-interest component the two indices treat differently.

Findings are presented on the basis of data and evidence alone.