Australia CPI annual rate, July 2026

Framing and exact resolver: the target is the ABS Monthly Consumer Price Index Indicator, All groups CPI annual movement for July 2026, original first print, percent, one-decimal published value. This is the monthly indicator variant, not quarterly CPI, not trimmed mean, and not a later revised vintage.

Resolver-source note: the intended resolver page is the future ABS July 2026 Monthly CPI Indicator release at https://www.abs.gov.au/statistics/economy/price-indexes-and-inflation/monthly-consumer-price-index-indicator/july-2026. That URL is used only as the stable first-print resolver location, not as evidence for the unreleased outcome.

Base rate/reference class: for a one-to-two-month-ahead forecast of a monthly year-over-year inflation rate, the outside-view base rate is recent persistence in the same ABS all-groups annual series. The prior is anchored on the latest May 2026 level of 4.0 percent, with the Feb-May mean of about 4.1 percent used as a cross-check rather than as the mechanical starting point.

Level, momentum, one-off, and policy mechanisms: level is still above target because housing and services remain firm; momentum from May is slightly down; temporary fuel and petrol declines depress headline inflation; a partial rebound or base-effect reversal by July argues against projecting May's 4.0 percent mechanically lower. A richer time-series model was not used because this is a short-horizon first-print forecast and the main uncertainty is ordinary monthly movement around recent persistence.

Prior/update/interval: persistence prior uses the same-series recent historical sample available in the draft: February-May 2026 at 3.7, 4.6, 4.2, 4.0. Adjustment components are +0.1 pp for possible fuel/base-effect rebound by July and 0.0 pp for underlying inflation persistence because trimmed mean at 3.6 is already below headline but still elevated. Successive changes are +0.9, -0.4, -0.2, so sigma = 0.70 using sample standard deviation of those changes; the 80 percent half-width is roughly 1.28*sigma = 0.90. Because this is only a short volatility sample, the interval is judgmental but kept at the direct realized-change width rather than narrowed. Point = 4.0 + 0.1 = 4.1, interval = 4.1 +/- 0.9 = [3.2, 5.0].

Counter-consideration: upside risk is a renewed fuel-price or utility-price jump plus sticky rents, which would land above the interval if July prints over 5.0 percent. Downside risk is a larger petrol reversal or broader demand slowdown, which would land below the interval if headline falls under 3.2 percent. An outside the interval result would most likely require a sharp energy or administered-price shock rather than ordinary month-to-month noise.

Review disposition: accepted the resolver critique by explicitly documenting the ledger sourceBinding mismatch while keeping the forecast tied to the registered dataPointId; accepted the leakage critique by treating the July 2026 URL only as the intended resolver page and removing it from evidence sourceContext; accepted the interval critique by stating the short-sample limitation and judgmental calibration rather than claiming a stronger realized-volatility basis.

Framing and exact resolver: forecast the ABS Monthly Consumer Price Index Indicator, Australia, July 2026, All groups CPI annual movement, first print, not seasonally adjusted headline indicator, in percent rounded to one decimal. This is the monthly CPI indicator variant; every anchor and historical value used here is the same All groups CPI annual movement variant, not quarterly CPI, trimmed mean, seasonally adjusted CPI, or a later revision.

Base rate/reference class: for monthly All groups CPI annual-rate targets two months ahead, a persistence prior using the latest official-source headline rate and recent successive changes is a strong base rate because the target is a year-over-year rate with overlapping 11 of 12 months already mostly determined by recent prices. The reference class here is the latest contiguous observed annual movements from February through May 2026: 3.7, 4.6, 4.2, 4.0. June 2026 was not used because it was not available at the run time.

Level, momentum, one-off, and policy mechanisms: level starts from May at 4.0 percent; momentum from March to May is downward after the fuel-shock peak, with changes of +0.9, -0.4, and -0.2 percentage points; one-off fuel volatility can pull July either way; policy-rebate and fuel-excise timing can distort monthly headline CPI; persistent rents, housing, food, and services keep the center above the RBA target band.

Prior/update/interval: persistence prior = 4.0 from May 2026 All groups CPI annual movement; historical sample = February-May 2026 annual rates 3.7, 4.6, 4.2, 4.0; adjustment components are deliberately coarse rather than precise: +0.2 for sticky inflation pressure indicated by May trimmed mean at 3.6 percent and home building at 5.6 percent, +0.1 for possible fuel rebound after March automotive fuel was up 32.8 percent year over year and then partly reversed, and -0.1 for the observed May-April/March headline deceleration, giving a 4.2 center before ladder rounding. Successive changes are +0.9, -0.4, -0.2, so sample sigma = 0.70 percentage points and the normal 80 percent half-width is roughly 1.28*sigma = 0.90. This three-change shock-window sample is thin, but it is a conservative realized-volatility anchor for July because the current forecast is explicitly exposed to the same fuel, rebate, and monthly-indicator noise that produced those recent changes. The ladder-implied 80 percent interval is 3.2 to 5.1, half-width about 0.95 around the 4.2 median, very close to the 1.28*sigma width.

Ladder: P(X <= 2.8) = 0.03; P(X <= 3.0) = 0.06; P(X <= 3.2) = 0.10; P(X <= 3.4) = 0.16; P(X <= 3.6) = 0.24; P(X <= 3.8) = 0.33; P(X <= 4.0) = 0.42; P(X <= 4.2) = 0.50; P(X <= 4.4) = 0.59; P(X <= 4.6) = 0.68; P(X <= 4.8) = 0.77; P(X <= 5.0) = 0.86; P(X <= 5.1) = 0.90; P(X <= 5.3) = 0.95. Linear interpolation gives p10 = 3.2, p50 = 4.2, and p90 = 5.1, each already at one-decimal print precision.

Counter-considerations: upside risk is a renewed fuel or import-cost spike, stronger rent and dwelling-price pass-through, or rebate expiry that would land above the interval at more than 5.1 percent. Downside risk is a sharper fuel reversal, broader household-demand weakening, or new price subsidies that would land below the interval at less than 3.2 percent. Outside the interval would require a materially larger one-month shock or policy-price adjustment than seen in the February-May reference window.

Resolver discrepancy note: the registered ledger sourceBinding URL points to the ABS Consumer Price Index, Australia, June 2026 page, but the operative target is the Monthly Consumer Price Index Indicator, Australia, July 2026 All groups CPI annual movement first print at dataPointId abs.cpi.all_groups.yoy.2026-07.first_print; I keep the forecast tied to the canonical slug and document the mismatch rather than changing identity.

Review disposition: accepted the interval critique by explicitly flagging the three-change volatility basis as thin but conservative for a July print exposed to the same recent fuel/rebate volatility; accepted the update critique by making the component adjustments coarse and tying them to the cited 4.0 headline, 3.6 trimmed mean, 5.6 home-building, 32.8 fuel, and recent headline changes. Kept the ledger discrepancy note and removed non-official news sources from sourceContext.

Framing and exact resolver: forecast the first ABS print for the July 2026 All groups CPI annual movement, in percent and rounded to one decimal. I keep the catalog slug, dataPointId, resolutionDate, and first_print policy fixed. The ledger sourceBinding discrepancy is material: it points to the June 2026 quarterly Consumer Price Index page, while the resolver text points to the July 2026 monthly CPI indicator page. I do not change target identity.

Base rate/reference class: for the official ABS monthly headline inflation-rate series available on the page, Sep-2021 through Sep-2025 values ranged from 1.9 to 8.4, with the late-2025 run Jul 2.8, Aug 3.0, Sep 3.5 showing renewed upward momentum after the June 2025 trough of 1.9. The reference class therefore supports a persistent monthly annual-rate prior rather than a sharp one-month break.

Prior/update/interval: persistence prior starts from the latest accessible same-variant official headline rate of 3.5, with current-release adjustment +0.4 for the Sep-2025 uptrend and sticky housing/services components, +0.2 for possible ongoing energy/base-effect pressure, and 0.0 net policy/mechanism offset because electricity rebate timing can reverse but rents and services remain firm, giving point 4.1. Historical sample is ABS official monthly all-groups annual movements from Sep-2021 to Sep-2025; using successive monthly changes for this rate series gives sigma = 0.51 percentage points. The normal 80% half-width is roughly 1.28*sigma = 1.28*0.51 = 0.65, rounded and slightly widened to 0.7 because the ledger’s monthly-indicator-versus-complete-monthly-CPI transition raises mapping risk. Final 80% bounds are 4.1-0.7 = 3.4 and 4.1+0.7 = 4.8.

Variant consistency: anchors use the ABS headline All groups monthly CPI annual movement, not seasonally adjusted and not trimmed mean, excluding-volatile, quarterly CPI, or a later revised value. Trimmed mean and components are used only as drivers, not as the resolving series.

Counter-consideration: upside risk is a renewed electricity, fuel, rent, or insurance shock that keeps annual headline inflation near or above 4.8; downside risk is a reversal of energy-base effects plus weaker goods prices that pulls the first print below 3.4. A print outside the interval would most likely require one of those energy or policy-rebate mechanisms to dominate the broad CPI basket.

Framing and exact resolver: forecast the ABS Monthly Consumer Price Index Indicator, Australia, All groups CPI annual movement for July 2026, first print only, rounded to one decimal in percent. This uses the monthly headline indicator variant throughout, not quarterly CPI, trimmed mean, seasonally adjusted components, or later revisions.

Reference class and base rate: for this one-to-two-month-ahead headline annual-rate forecast, the outside view is a random-walk/persistence base rate anchored on the latest same ABS monthly headline series values. The recent path is 3.7% in February, 4.6% in March, 4.2% in April, and 4.0% in May, so the base rate is near 4.0% before July-specific adjustments.

Inside-view update: March's fuel shock has partly faded, pulling the headline rate down from 4.6% to 4.0% by May, but trimmed mean at 3.6% and continued RBA concern argue against a fast return toward the 2-3% band by July. The July 2025 comparison month was already lifted to 2.8% by electricity rebate effects, limiting upside from base effects, but current domestic inflation is still too sticky to forecast a clear drop below May.

Prior/update/interval: persistence prior = May 2026 headline annual CPI 4.0% from the ABS monthly headline reference class; historical sample = Feb-May 2026 monthly headline annual rates 3.7, 4.6, 4.2, 4.0; adjustment components = +0.2 for sticky trimmed-mean/domestic services pressure, -0.1 for partial fuel-shock unwind and high July 2025 electricity base, final point = 4.0 + 0.2 - 0.1 = 4.1. Successive changes are +0.9, -0.4, -0.2 percentage points, so sigma = 0.7; one-month 80% half-width is roughly 1.28*sigma = 0.9. I widen to 1.1 for the two-step May-to-July horizon and energy-policy volatility, giving 4.1 +/- 1.1 = 3.0 to 5.2.

Counter-consideration: upside risk is a renewed fuel or electricity price jump plus continued housing and food pressure, which would land above the interval if the July annual print exceeded 5.2%. Downside risk is a sharper reversal of the March fuel shock or a larger temporary policy subsidy effect, which would land below the interval if the July annual print fell under 3.0%.

Framing and exact resolver: this is the ABS Monthly Consumer Price Index Indicator, Australia, July 2026, All groups CPI annual movement, not the quarterly CPI and not a seasonally adjusted or trimmed-mean variant. The ledger target resolves on the first ABS print, rounded to one decimal, and the ledger discrepancy is that sourceBinding names a June 2026 quarterly CPI page while the target identity and resolver clearly refer to the July 2026 Monthly CPI Indicator page.

Base rate / reference class: for a two-month-ahead forecast of the monthly headline annual rate, the strongest public reference class is the recent same-series path: 3.7, 4.6, 4.2, and 4.0 percent. That says the outside-view anchor should remain near 4 percent rather than reverting quickly to the RBA 2-3 percent target.

Level, momentum, one-off, and policy mechanisms: the level is still elevated at 4.0 percent in May; short-run momentum is down from the March fuel spike; the March one-off fuel shock has partly unwound; the July expiry or reversal of temporary fuel-excise relief is an upside mechanism, while tighter rates and weak demand are downside mechanisms.

Prior/update/interval: persistence prior uses the same-series monthly annual-rate history 2026-02=3.7, 2026-03=4.6, 2026-04=4.2, 2026-05=4.0. Successive changes are +0.9, -0.4, -0.2 percentage points; sample sigma = 0.70. The one-month 80 percent half-width is about 1.28*sigma = 0.90. I widen to 1.10, within 1.75x, because the target is two monthly releases ahead and July has a fuel-policy/base-effect risk. Starting from 4.0, I add +0.2 for underlying inflation near 3.6 and possible fuel-excise reversal, giving point 4.2 and 80 percent bounds 4.2 +/- 1.1 = 3.1 to 5.3.

Counter-consideration: upside risk is a renewed oil-price or excise-driven petrol jump plus freight pass-through, which would land above the interval if July annual CPI prints above 5.3 percent. Downside risk is a fast fuel-price reversal, demand weakness, and goods disinflation, which would land below the interval if July prints below 3.1 percent.

Derived run: the pointwise median of the CDFs of 3 independent thesis.analyst fast rollouts on this target — median prediction sampling per Turtel et al. 2025 (arXiv:2505.17989). No new model call; this run is a deterministic aggregate of the recorded rollouts at 2026-07-10T13:52:30Z, 2026-07-10T13:54:35Z, 2026-07-10T13:55:57Z. Drivers and resolver fields mirror the rollout closest to the median.

Median CDF quantiles: q10 = 3.1, q50 = 4.1, q90 = 5.2. Constituent points [4.1, 4.1, 4.2] with 80% widths [1.4, 2.2, 2.2]; the median interval inherits the central rollout mass rather than averaging tails.

The target is the ABS first-print All groups CPI annual movement for July 2026, rounded to one decimal. All anchors use the same national weighted-average-of-eight-capital-cities All groups annual CPI variant; the ledger sourceBinding points to the June 2026 Consumer Price Index page rather than the July monthly-indicator page, but the forecast remains tied to abs.cpi.all_groups.yoy.2026-07.first_print.

The reference class/base rate is the recent complete-monthly All groups annual-CPI sequence, centred near 3.9% over December-May. July 2025's unusually large 1.3% monthly increase is the dominant mechanical comparison effect, so a lower July 2026 annual rate is more likely even if underlying housing pressure persists.

Prior/update/interval: Persistence prior is the December 2025-May 2026 All groups annual-rate mean of 4.02%; the historical sample is April 2025-May 2026 annual rates, whose successive changes have sample sigma = 0.46 percentage points. The base-effect adjustment for July 2025's +1.3% monthly print is -0.7pp after allowing a positive July 2026 monthly print, while persistent Housing (+6.5%) and trimmed mean (+3.6%) add +0.0pp net versus the recent level, implying 3.3%. The ordinary 80% half-width is 1.28*0.46 = 0.59pp; I use 0.8pp (1.36x) because a large July base month makes the annual-rate mapping unusually sensitive to the unobserved July monthly print, giving 2.5% to 4.1%.

Upside risk is a renewed energy or housing-price jump that produces a July monthly increase well above the assumed positive print; downside risk is another transport-led fall combined with weak discretionary prices. Either would land outside the interval: a July monthly rise near or above the July 2025 1.3% would land above 4.1%, while a material monthly fall would land below 2.5%.

This is the original, not-seasonally-adjusted All groups CPI annual movement for the weighted average of eight capital cities. The matching ABS release table is TABLE 1, CPI: All Groups, Index numbers and Percentage change. The ledger remains tied to abs.cpi.all_groups.yoy.2026-07.first_print: its sourceBinding points to a June 2026 CPI page, while the resolver specifies the July 2026 Monthly CPI Indicator release.

Base rate/reference class: the recent complete-monthly-CPI annual-rate sample from Apr-25 through May-26 rose from 2.4% to 4.0%, with the latest three prints 4.6%, 4.2%, and 4.0%. I use persistence around the current 4% regime rather than the earlier low-inflation 2025 level.

Prior/update/interval: Persistence prior is May's 4.0% annual rate; the 14-print Apr-25–May-26 historical sample is 2.4, 2.1, 1.9, 3.0, 3.2, 3.6, 3.8, 3.4, 3.8, 3.8, 3.7, 4.6, 4.2, 4.0. Its 13 successive changes have sample standard deviation sigma = 0.46 percentage points, so 1.28*sigma = 0.59 points. I add a 0.1-point persistence/energy-and-housing update to 4.1%, then widen the half-width to 0.7 points (1.19x) because the July 2025 monthly comparison was +1.3% and July seasonal price movements make the annual base effect unusually consequential; implied 80% bounds are 3.4% to 4.8%.

Counter-consideration: upside risk is a renewed electricity or fuel increase combined with a stronger-than-July-2025 monthly rise, which could land above the interval. Downside risk is a larger fall in fuel prices or a weak seasonal clothing/travel outcome; a monthly rise far below July 2025's 1.3% would land below the interval.

The target is the ABS All groups CPI, Australia, annual movement headline series, not trimmed mean or a seasonally adjusted analytical series. The resolver is strict original first print; the ledger's June-page sourceBinding discrepancy is documented but does not alter the target identity or first-print policy.

Base rate: the matching ABS All groups annual-movement reference class ran 2.4%, 2.1%, 1.9%, 3.0%, 3.2%, 3.6%, 3.8%, 3.4%, 3.8%, 3.8%, 3.7%, 4.6%, 4.2%, and 4.0% from April 2025 through May 2026. July 2025's 1.3% monthly rise creates a substantial favorable year-ago comparison for July 2026.

Prior/update/interval: persistence prior is May's 4.0% matching headline rate; historical sample is the 14 ABS annual rates from April 2025-May 2026; successive annual-rate changes give sigma = 0.476 percentage points, so 1.28*sigma = 0.609. I apply a -0.4pp July base-effect adjustment, partly offset by persistent housing and underlying inflation, for 3.6%; rounding the 0.609pp half-width to the published 0.1-point precision gives 3.0% to 4.2%.

Counter-consideration: upside risk is a renewed energy or administered-price jump combined with sticky housing, which could land above the interval. Downside risk is a stronger-than-expected July reversal after the 1.3% July 2025 base; a broad goods-price decline would land below the interval.

Derived run: the pointwise median of the CDFs of 3 independent thesis.analyst fast rollouts on this target — median prediction sampling per Turtel et al. 2025 (arXiv:2505.17989). No new model call; this run is a deterministic aggregate of the recorded rollouts at 2026-07-10T15:37:10Z, 2026-07-10T15:40:54Z, 2026-07-10T15:45:17Z. Drivers and resolver fields mirror the rollout closest to the median.

Median CDF quantiles: q10 = 3.0, q50 = 3.6, q90 = 4.2. Constituent points [3.3, 4.1, 3.6] with 80% widths [1.6, 1.4, 1.2]; the median interval inherits the central rollout mass rather than averaging tails.

Framing and exact resolver: the target is the ABS first print for the July 2026 Monthly Consumer Price Index Indicator All groups CPI annual movement, rounded to one decimal, tied to dataPointId abs.cpi.all_groups.yoy.2026-07.first_print. The ledger also documents that the registered sourceBinding URL appears to point to a June 2026 CPI page; I keep the canonical ledger resolver fields unchanged.

Variant consistency: the ledger target names the Monthly CPI Indicator, while ABS transition pages indicate the ongoing complete Monthly CPI publication now carries the monthly All groups CPI annual movement. Forecast anchors use the ABS All groups CPI, Australia, original annual movement series; I do not use seasonally adjusted, trimmed mean, or old indicator values as the target series.

Prior/update/interval: base rate/reference class prior is persistence in the complete Monthly CPI annual rate from Apr-25 to May-26. The latest annual rate is 4.0%; the high Jul-25 base month (+1.3% m/m after Jun-25 +0.1%) mechanically pulls July 2026 annual inflation down unless Jun-Jul 2026 monthly gains repeat the rebate/fuel spike, while sticky trimmed mean at 3.6%, housing at 6.5%, and broad services pressure offset some downside. Successive annual-rate changes from Apr-25..May-26 were -0.3, -0.2, +1.1, +0.2, +0.4, +0.2, -0.4, +0.4, 0.0, -0.1, +0.9, -0.4, -0.2 percentage points; sample sigma = 0.48, so 1.28*sigma = 0.62 percentage points. I widen to a ladder-implied 80% half-width of 0.8 points, about 1.29x the sigma half-width, because June and July are both still unknown and July base/rebate effects are unusually lumpy; final implied bounds are 3.0% to 4.6% around a 3.8% median.

Ladder: P(X <= 2.6) = 0.04; P(X <= 2.9) = 0.08; P(X <= 3.1) = 0.13; P(X <= 3.3) = 0.20; P(X <= 3.5) = 0.30; P(X <= 3.7) = 0.43; P(X <= 3.8) = 0.50; P(X <= 3.9) = 0.58; P(X <= 4.1) = 0.70; P(X <= 4.3) = 0.80; P(X <= 4.5) = 0.88; P(X <= 4.6) = 0.92; P(X <= 4.9) = 0.97. Linear interpolation gives p10 between 2.9 and 3.1 at 2.98, rounded to 3.0; p50 at 3.8; p90 between 4.5 and 4.6 at 4.55, rounded to 4.6.

Counter-considerations: upside risk is a renewed fuel or electricity/rebate shock plus sticky housing that would land above the interval. Downside risk is further fuel reversal and weak discretionary prices pushing the July print toward the low 3s. Outside the interval below 3.0 would likely require very soft June-July monthly CPI despite the known sticky components; outside the interval above 4.6 would likely require another March-like energy or administered-price jump.

Review disposition: accepted the blocking resolver critique by restoring the canonical ledger resolutionSource, resolutionSourceUrl, and resolutionRule exactly, while keeping the ABS naming transition and sourceBinding discrepancy as reasoning context. Kept the ladder, base-rate arithmetic, and 80% interval because the critique did not identify an evidence or calibration error.

Framing and exact resolver: I forecast the first ABS print for July 2026 All groups CPI annual movement, rounded to one decimal percent. The target ledger names the old Monthly Consumer Price Index Indicator URL, but the ABS current public release and future calendar now label the relevant product Consumer Price Index, Australia; I keep the forecast tied to the provided slug, dataPointId, resolutionDate, and first-print rule rather than changing target identity.

Reference class and base rate: the relevant reference class is month-to-month changes in the annual All groups CPI rate from the ABS monthly CPI publication since the complete monthly history begins printing annual rates. The base rate is persistence from the latest 4.0% annual print, adjusted by expected June-July 2026 monthly price growth relative to the 1.4% combined monthly rise in June-July 2025.

Prior/update/interval: persistence prior is May 2026 annual CPI = 4.0 using the ABS Apr-25 through May-26 annual-rate sample. Adjustment components: expected Jun-26 plus Jul-26 headline monthly CPI about 1.2% versus Jun-25 plus Jul-25 base of 1.4%, subtracting about 0.2 percentage points from the annual rate; housing and trimmed-mean persistence add back about 0.0 to 0.1, leaving point 3.8 after one-decimal rounding. Interval method uses realized dispersion of successive annual-rate changes: differences are -0.3, -0.2, 1.1, 0.2, 0.4, 0.2, -0.4, 0.4, 0.0, -0.1, 0.9, -0.4, -0.2; monthly-difference sample sd is about 0.48, two-step sigma = sqrt(2)*0.48 = 0.68, so sigma = 0.68 and 1.28*sigma = 0.87. Rounded 80% interval is 3.8 +/- 0.9, or 2.9 to 4.7.

Momentum assessment: May headline fell on the month but that was after a large March rise; trimmed mean rose to 3.6% and housing inflation remained 6.5%, so a forecast materially below 3.5 would require softer June and July prints than the underlying measures currently imply.

Counter-consideration: upside risk is another July electricity, fuel, or administered-price jump like July 2025, which would land above the interval if combined with broad services persistence. Downside risk is a repeat of May's negative headline monthly print plus falling fuel and travel prices, which would land below the interval if June and July together were near zero or negative.

Framing and exact resolver: this is the ABS Monthly Consumer Price Index Indicator, All groups CPI annual movement for July 2026, first print only, percent rounded to one decimal. The ledger sourceBinding discrepancy is noted: the target resolves from the July 2026 Monthly CPI Indicator page, not the June 2026 quarterly CPI page.

Base rate / reference class: for this same ABS monthly All groups annual-rate series, the short-run reference class is the recent monthly annual-rate sequence itself. A persistence prior from the latest official 4.0 percent is preferred over a long-run 2-3 percent target because the July print is only two monthly observations ahead and the annual rate still reflects earlier fuel, housing, and services shocks.

Prior/update/interval: persistence prior = May 2026 All groups annual CPI at 4.0. Historical sample for realized dispersion = successive changes in Feb-May annual rates: 4.6-3.7 = 0.9, 4.2-4.6 = -0.4, 4.0-4.2 = -0.2. Sample sigma = 0.70 percentage points; 80 percent normal half-width is about 1.28*sigma = 0.90. Adjustment components: +0.25 for possible July fuel-excise and energy rebound, +0.15 for sticky housing/services and trimmed mean at 3.6, net roughly +0.40 from the 4.0 prior. Final point = 4.4; 80 percent interval = 4.4 +/- 0.9 = 3.5 to 5.3.

Variant control: anchors and dispersion use the monthly CPI indicator All groups annual movement, not quarterly CPI, not seasonally adjusted monthly change, and not trimmed mean except as context for underlying pressure.

Counter-consideration: upside risk is a renewed fuel or electricity jump after temporary relief fades, which would land above the interval if July All groups annual inflation prints above 5.3. Downside risk is a faster petrol reversal plus weak discretionary demand and housing-cost cooling, which would land below the interval if the first print is under 3.5.

Framing and exact resolver: this targets the ABS first-print July 2026 All groups CPI annual movement, percent, weighted average of eight capital cities, original annual movement, rounded to one decimal. The ABS current-release path now shows Consumer Price Index, Australia rather than the older Monthly Consumer Price Index Indicator wording, but I keep the ledger target identity and resolver text unchanged and note the sourceBinding discrepancy rather than changing the target.

Reference class and base rate: for this rate series I used the ABS monthly annual All groups CPI history since monthly annual movements begin in April 2025 as the reference class, with May 2026 at 4.0 as the persistence base rate. The recent annual-rate sequence is elevated but volatile: 3.8 in January, 3.7 in February, 4.6 in March, 4.2 in April, and 4.0 in May.

Prior/update/interval: persistence prior/model is May 2026 annual CPI at 4.0, historical sample is ABS Apr 2025-May 2026 monthly annual All groups CPI, adjustment components are high July 2025 base month (+1.3 monthly) lowering the July 2026 annual rate, assumed Jun-Jul 2026 original monthly CPI of about +0.4 and +0.6 partly offsetting that base effect, sticky housing/services/trimmed-mean pressure adding upside, and May fuel weakness adding uncertainty. Mechanics: July annual approx 1.040*(1.004*1.006)/(1.001*1.013)-1 = 0.0359, rounded to 3.6. Successive annual-rate changes are -0.3, -0.2, +1.1, +0.2, +0.4, +0.2, -0.4, +0.4, 0.0, -0.1, +0.9, -0.4, -0.2, so sigma = 0.47 from RMS successive changes and 1.28*sigma = 0.61; final implied 80% bounds are 3.6 +/- 0.61, rounded to 3.0 to 4.2.

Counter-considerations: upside risk is a rebound in fuel/transport after May's -3.9 original monthly transport fall plus still-high electricity or rent inflation, which would land above the interval if June and July monthly CPI together exceed roughly 1.7%. Downside risk is another energy-price fall or faster goods disinflation, which would land below the interval if June and July combined monthly CPI is near flat or negative despite services persistence.

Derived run: the pointwise median of the CDFs of 3 independent thesis.analyst fast rollouts on this target — median prediction sampling per Turtel et al. 2025 (arXiv:2505.17989). No new model call; this run is a deterministic aggregate of the recorded rollouts at 2026-07-10T16:19:40Z, 2026-07-10T16:30:43Z, 2026-07-10T16:42:38Z. Drivers and resolver fields mirror the rollout closest to the median.

Median CDF quantiles: q10 = 3.0, q50 = 3.8, q90 = 4.7. Constituent points [3.8, 4.4, 3.6] with 80% widths [1.8, 1.8, 1.2]; the median interval inherits the central rollout mass rather than averaging tails.

The canonical target is the first July 2026 Monthly Consumer Price Index Indicator All groups CPI annual movement, printed to one decimal, not trimmed mean, a seasonally adjusted monthly change, or a later replacement value. The registered sourceBinding points to a June 2026 Consumer Price Index page; this discrepancy is documented without changing the ledger target or resolver.

The reference class/base rate is the six first-published annual observations from December 2025 through May 2026: 3.8%, 3.8%, 3.7%, 4.6%, 4.2%, and 4.0%. Persistence around the latest 4.0% is the outside-view anchor; the six-observation median is also 4.0%.

Level, momentum, one-off, and policy mechanisms are separated as follows: the level remains near 4%; momentum eased by 0.6 percentage point from March to May; volatile fuel, travel, and food can move individual months; and electricity-rebate exhaustion keeps Housing inflation elevated. These effects support a central value near 4.0% without extrapolating March's spike.

Prior/update/interval: persistence prior = May annual CPI of 4.0%, using the December-May historical sample 3.8, 3.8, 3.7, 4.6, 4.2, 4.0. Successive changes are 0.0, -0.1, +0.9, -0.4, and -0.2 percentage point; their sample sigma = 0.50. Because only five changes were available in the fetched pre-resolution history, 1.28*sigma = 1.28*0.50 = 0.64 point is a short-sample normal-approximation half-width, not an empirical 80% quantile. Adjustments are -0.1 for recent easing, +0.1 for persistent Housing/electricity pressure, and 0.0 net for other one-offs, leaving 4.0%. The ladder implies bounds of 3.3% and 4.8%, total width 1.5 points versus the sigma-based 1.28 points; the 1.17x widening is judgmental allowance for rebate and base-effect volatility.

Ladder: P(X <= 3.0) = 0.04; P(X <= 3.2) = 0.08; P(X <= 3.4) = 0.12; P(X <= 3.6) = 0.20; P(X <= 3.8) = 0.30; P(X <= 3.9) = 0.40; P(X <= 4.0) = 0.50; P(X <= 4.2) = 0.62; P(X <= 4.4) = 0.73; P(X <= 4.6) = 0.82; P(X <= 4.8) = 0.90; P(X <= 5.2) = 0.95. Linear interpolation gives the median at 4.0, the 10th percentile at 3.3, and the 90th percentile at 4.8, each rounded to one decimal.

Upside risk comes from further electricity-rebate unwinding, fuel disruption, or unusually strong rents and services; a combined shock would land above the interval. Downside risk comes from fuel reversal, discounting, or favorable July base effects; synchronized declines across volatile and core components would land below the interval. Outside the interval therefore requires a broader or larger shock than ordinary month-to-month variation.

Review disposition: Accepted the blocking resolver correction by restoring the canonical source, URL, and rule while retaining the sourceBinding discrepancy only as context. Accepted the interval warning by describing 1.28*sigma as a short-sample normal approximation and identifying the 1.17x widening as judgmental; no unavailable longer history was introduced.

The target is the first-print July 2026 ABS All groups CPI annual movement for Australia, rounded to one decimal. All anchors use the same headline, not-seasonally-adjusted All groups variant. The ledger calls the release the Monthly CPI Indicator and supplies a future indicator URL, while the official ABS calendar labels it Consumer Price Index, Australia; the sourceBinding also incorrectly points to the June release, so the forecast retains the registered target and records the discrepancy.

The reference class is the 14 official annual prints from April 2025 through May 2026. Its simple persistence base rate is the latest 4.0% print, but July 2025's unusually large 1.3% monthly increase will leave the twelve-month comparison. Assuming roughly 0.3% monthly inflation in both June and July 2026, June 2025's 0.1% base adds about 0.2 point before July 2025's 1.3% base subtracts about 1.0 point, implying approximately 3.2%.

Level and momentum remain firm because May annual CPI was 4.0% and Housing was 6.5%. The principal one-off is the July 2025 base effect; easing Transport inflation reinforces the decline. Policy works only with a lag over this horizon, so no separate large policy adjustment is imposed.

Prior/update/interval: persistence model prior = 4.0%, using the 14-print April 2025–May 2026 official history. Adjustment components are +0.2 percentage point for replacing June 2025's 0.1% monthly rise with an assumed 0.3%, then -1.0 point for replacing July 2025's 1.3% rise with an assumed 0.3%: 4.0 + 0.2 - 1.0 = 3.2%. Successive annual-rate changes are -0.3, -0.2, +1.1, +0.2, +0.4, +0.2, -0.4, +0.4, 0.0, -0.1, +0.9, -0.4, and -0.2 percentage point; their sample standard deviation is sigma = 0.476. The empirical 80% half-width is 1.28*sigma = 0.610, giving 3.2 ± 0.61 = [2.59, 3.81], rounded to [2.6%, 3.8%].

Upside risk comes from another unusually large July price increase, persistent Housing inflation, or renewed fuel pressure and would land above the interval. Downside risk comes from a flat or negative July monthly print combined with broader goods disinflation and could land below the interval. An annual print outside the interval would require the two-month June–July price change to differ materially from the roughly 0.6% central assumption.

The resolver is the first July 2026 ABS All groups CPI annual movement, national weighted average of eight capital cities, original headline series, printed to one decimal. ABS now labels the release Consumer Price Index, Australia rather than Monthly CPI Indicator. The ledger's June-page sourceBinding is discrepant; target identity and first-print policy are unchanged.

The reference class and base rate are short-horizon monthly annual-CPI persistence. The same-series path from January through May was 3.8%, 3.7%, 4.6%, 4.2%, and 4.0%. Persistence anchors July near 4.0%, while the RBA's 4.8% June-quarter projection and fuel/raw-material pass-through argue for an upward adjustment.

Level, momentum, one-off, and policy mechanisms point in different directions: the May level was 4.0% and recent momentum eased; fuel and electricity-rebate effects lift the level; July rolls over a large 1.3% July 2025 monthly increase, restraining annual inflation; tighter monetary policy mainly weighs on subsequent months.

Prior/update/interval: persistence prior = May annual CPI 4.0%; historical sample = January-May annual rates 3.8, 3.7, 4.6, 4.2, 4.0; adjustments = +0.5 percentage point for the RBA near-term energy/cost path and -0.1 for easing momentum plus July's strong base, giving 4.0 + 0.5 - 0.1 = 4.4%. Successive changes are -0.1, +0.9, -0.4, -0.2 percentage point; their sample standard deviation is sigma = 0.50. The Gaussian 80% half-width is roughly 1.28*sigma = 1.28*0.50 = 0.64 percentage point; rounding bounds to the ABS one-decimal print grid gives 4.4 - 0.6 = 3.8 and 4.4 + 0.6 = 5.0.

Upside risk comes from persistent fuel disruption, second-round transport and materials pass-through, or another electricity-price increase and would land above the interval at 5.1% or more. Downside risk comes from a rapid fuel reversal combined with the large July 2025 base effect and would land below the interval at 3.7% or less.

The target is the first-print ABS All groups CPI annual movement for July 2026, printed to one decimal. All anchors use this same original, not seasonally adjusted, All groups variant. ABS now labels the publication Consumer Price Index, Australia; the ledger's generic sourceBinding incorrectly points to the June page, while the exact July release page is the substantive resolver.

The reference class or base rate is monthly annual-CPI persistence since monthly annual observations began in April 2025: 2.4%, 2.1%, 1.9%, 3.0%, 3.2%, 3.6%, 3.8%, 3.4%, 3.8%, 3.8%, 3.7%, 4.6%, 4.2%, and 4.0%. Persistence alone anchors July near 4.0%, but the unusually large July 2025 monthly increase creates a strong adverse base effect.

Level and momentum effects point in different directions. Headline annual inflation eased from 4.6% to 4.2% to 4.0%, while May trimmed mean rose to 3.6% from 3.4%. Housing inflation was 6.5%, including electricity at 21.1%, but goods inflation slowed to 4.2% and automotive-fuel inflation fell to 7.7% from 18.6%.

Prior/update/interval: The persistence prior is May's 4.0%, using the 14 monthly annual observations from April 2025 through May 2026. Adjustments are approximately -0.5 percentage point for July 2025's 1.3% monthly rise leaving the annual comparison, +0.2 for firm housing/electricity and underlying inflation, and 0.0 net for softer fuel and other momentum, yielding 4.0 - 0.5 + 0.2 = 3.7%. Across the 13 successive annual-rate changes, sample sigma = 0.48 percentage point; 1.28*sigma = 0.61, so the rounded 80% half-width is 0.6 and the implied bounds are 3.1% to 4.3%.

Upside risk comes from another large electricity-tariff step, persistent rents, or renewed fuel inflation and would land above the interval if July prints above 4.3%. Downside risk comes from a weak July monthly print combined with the 1.3% July 2025 base rolling out; a broad goods or fuel decline could put the result outside the interval below 3.1%.

Derived run: the pointwise median of the CDFs of 3 independent thesis.analyst fast rollouts on this target — median prediction sampling per Turtel et al. 2025 (arXiv:2505.17989). No new model call; this run is a deterministic aggregate of the recorded rollouts at 2026-07-10T17:17:18Z, 2026-07-10T17:23:36Z, 2026-07-10T17:29:16Z. Drivers and resolver fields mirror the rollout closest to the median.

Median CDF quantiles: q10 = 3.1, q50 = 3.7, q90 = 4.3. Constituent points [3.2, 4.4, 3.7] with 80% widths [1.2, 1.2, 1.2]; the median interval inherits the central rollout mass rather than averaging tails.

Framing: the resolver is the ABS Monthly Consumer Price Index Indicator, Australia, July 2026 release, All groups CPI annual movement, first print, in percent rounded to one decimal. I use the same monthly-indicator all-groups annual variant for anchors; I do not mix in quarterly CPI, seasonally adjusted variants, trimmed mean, or later revisions.

Reference class and base rate: for near-term Australian monthly CPI indicator all-groups annual forecasts, the strongest public reference class is recent same-series persistence plus one- to three-month shock reversal. The fetched same-variant sequence 3.7%, 4.6%, 4.2%, 4.0% anchors the ladder span: most mass stays in the high-3s to high-4s, with a meaningful right tail if fuel or housing re-accelerates.

Prior/update/interval: persistence prior is the recent same-series All groups CPI annual movement centered around the May 2026 4.0% print, using the February-May 2026 sample of 3.7%, 4.6%, 4.2%, and 4.0%; I apply a judgmental +0.3pp central update, roughly +0.15pp for July fuel-excise/oil pass-through risk, +0.10pp for sticky housing and rents, and +0.05pp for monthly-indicator sampling and momentum risk, limited by likely normalization from the March fuel spike. The interval method is a threshold ladder grounded on the fetched 3.7%-4.6% recent range, widened for two unknown monthly prints before July and one-off fuel-policy effects; the 10th-90th percentile 80% interval is 3.5 to 5.2 with median 4.3.

Ladder: P(X <= 3.0) = 0.03; P(X <= 3.3) = 0.07; P(X <= 3.6) = 0.12; P(X <= 3.8) = 0.20; P(X <= 4.0) = 0.31; P(X <= 4.2) = 0.43; P(X <= 4.4) = 0.54; P(X <= 4.6) = 0.65; P(X <= 4.8) = 0.75; P(X <= 5.0) = 0.83; P(X <= 5.2) = 0.90; P(X <= 5.5) = 0.96; P(X <= 5.8) = 0.98. Linear interpolation gives 10th percentile at 3.48, median at 4.33, and 90th percentile at 5.20; rounded to the ABS print precision this is 10th percentile at 3.5, median at 4.3, and 90th percentile at 5.2.

Counter-considerations: upside risk is a larger July fuel rebound, rent acceleration, or supply shock that would land above the interval near 5.3% or higher. Downside risk is a sharper fuel reversal, energy subsidy effect, or broad goods disinflation that would land outside the interval below 3.5%. The central case keeps annual inflation above target but below the March shock peak.

Review disposition: accepted the warning to make the +0.3pp upward update explicit and judgmental, with approximate component contributions; accepted the confidence-semantics suggestion by naming the ladder-derived 10th-90th percentile 80% interval. I did not add a new oil or policy datapoint because it was not in the public evidence available to the draft trace.

The target is the original, not seasonally adjusted, All groups CPI annual movement for July 2026, printed to one decimal by ABS. It resolves only to the first print. The ledger calls the release the Monthly CPI Indicator and supplies a sourceBinding for the June CPI page, while the current ABS calendar labels the July publication Consumer Price Index, Australia; I preserve the registered dataPointId and resolver and document this discrepancy rather than changing target identity.

Reference class/base rate: the fetched annual sequence from July 2025 through May 2026 was 3.0%, 3.2%, 3.6%, 3.8%, 3.4%, 3.8%, 3.8%, 3.7%, 4.6%, 4.2%, and 4.0%. Persistence near the latest 4.0% is the outside-view anchor, while the observed range of 3.0% to 4.6% anchors the ladder span.

Level, momentum, one-off, and policy mechanisms point in different directions. The 4.0% May level and 3.6% trimmed mean show persistent underlying pressure; headline momentum eased from March. Expiring electricity rebates raised measured housing inflation, while the unusually large 13.5% July 2025 electricity increase and 1.3% headline monthly increase become adverse base effects when they roll out. Higher fuel costs can offset part of that decline.

The ten consecutive changes in the fetched July 2025-May 2026 annual-rate sequence were +0.2, +0.4, +0.2, -0.4, +0.4, 0.0, -0.1, +0.9, -0.4, and -0.2 percentage points, with a sample standard deviation of about 0.4 point. Scaling that volatility across the two releases from May to July gives roughly 1.28 × 0.4 × sqrt(2) = 0.7 point for an 80% persistence interval; allowing additional uncertainty around electricity base effects and fuel supports ladder tails about 0.8 point from the 4.0% anchor.

Ladder: P(X <= 2.8) = 0.04; P(X <= 3.0) = 0.07; P(X <= 3.2) = 0.11; P(X <= 3.4) = 0.18; P(X <= 3.6) = 0.28; P(X <= 3.8) = 0.40; P(X <= 4.0) = 0.52; P(X <= 4.2) = 0.64; P(X <= 4.4) = 0.75; P(X <= 4.6) = 0.84; P(X <= 4.8) = 0.91; P(X <= 5.0) = 0.95; P(X <= 5.2) = 0.97. Linear interpolation gives the 10th percentile at 3.15, median at 3.97, and 90th percentile at 4.77; rounded to the ABS print precision these are 3.2, 4.0, and 4.8.

Prior/update/interval: A latest-value persistence model uses the July 2025-May 2026 reference-class sample (3.0%-4.6%) and starts from May's 4.0%. Updates comprise sticky services/non-tradables and fuel pressure upward, offset by recent headline slowing and the July 2025 electricity/headline base effects. Historical annual-rate change volatility implies about a 0.7-point two-release 80% width around persistence; direct threshold-ladder quantile inversion, widened modestly for one-off uncertainty, yields final implied bounds of 3.2% to 4.8% and a 4.0% median.

Upside risk: a larger fuel shock, renewed electricity-price increases, or broader pass-through into goods and services would land above the interval. Downside risk: a sharp fuel reversal plus a stronger-than-expected electricity base-effect unwind would land below the interval. Historical levels and prospective month-to-month shocks are distinct: the cited history includes a 3.0% level, so it does not imply that every outcome outside the interval requires an unprecedented historical level.

Review disposition: Accepted both critiques. Added an explicit volatility calculation to calibrate the ladder interval and removed the overstated tail claim by distinguishing historical levels from prospective annual-rate shocks.

Target framing: forecast the ABS All groups CPI annual movement for July 2026 in original terms, rounded to one decimal and resolved on the first print. The registered resolver names the ceased Monthly Consumer Price Index Indicator and its sourceBinding points to the June 2026 CPI page; the active ABS production is Consumer Price Index, Australia. I retain the supplied dataPointId, date, and first-print policy rather than changing target identity.

Reference class/base rate: the six fetched monthly annual prints span 3.7% to 4.6%, with a 4.0% latest print and a 3.9% simple average. All anchors are the same national All groups CPI annual-movement variant, not quarterly CPI or a smoothed series.

Level and momentum point mildly lower after the 4.6% March spike eased to 4.2% and then 4.0%; persistent housing inflation and a 3.6% trimmed mean limit the expected decline. June and July observations were not yet published at this forecast date. No separate fitted time-series model was used: the six-month sample is too short and volatile for a reliable fitted model, so the persistence/reference-class prior is used instead.

Ladder: P(X <= 3.2) = 0.01; P(X <= 3.3) = 0.03; P(X <= 3.4) = 0.08; P(X <= 3.6) = 0.16; P(X <= 3.7) = 0.29; P(X <= 3.8) = 0.46; P(X <= 3.9) = 0.59; P(X <= 4.0) = 0.70; P(X <= 4.1) = 0.79; P(X <= 4.2) = 0.86; P(X <= 4.3) = 0.93; P(X <= 4.4) = 0.97; P(X <= 4.6) = 0.99. Linear interpolation gives 10th percentile at 3.45, median at 3.83, and 90th percentile at 4.26; rounded to the ABS one-decimal print precision these are 3.5%, 3.8%, and 4.3%.

Prior/update/interval: persistence/reference-class prior is the fetched Dec-May All groups sequence (3.8, 3.8, 3.7, 4.6, 4.2, 4.0), centered near its 3.9% mean; adjustments are easing headline momentum, persistent 6.5% housing inflation, and uncertainty from volatile monthly components. The 80% interval is the ladder-derived 10th-to-90th percentile range, calibrated with rung placement against the fetched 3.7% trough and 4.6% March high, yielding 3.5% to 4.3% after one-decimal rounding.

Counter-consideration: upside risk is a renewed housing, fuel, or food acceleration that would land above the interval; downside risk is a broad decline in goods and services prices that would land below the interval. A July print above 4.3% or below 3.5% is outside the interval.

Review disposition: accepted the request to state that no separate fitted time-series model was used and to clarify the ladder-derived 10th-to-90th interval; accepted the release-status clarification that June and July observations were not yet published.

This targets the ABS All groups CPI annual movement for July 2026, first print, unadjusted and rounded to one decimal. The ABS release calendar verifies publication on 2026-08-26 at 11:30am AEST. The registered sourceBinding names the June 2026 CPI page, but the target remains the July 2026 first-print dataPointId and July release page specified by the ledger.

The base rate/reference class is the recent complete monthly ABS sequence: annual inflation was 3.0% to 3.8% across July 2025 to January 2026, then 3.7%, 4.6%, 4.2%, and 4.0% from February to May 2026. The central tendency has recently eased, but the level remains above the mid-3% range.

Level and momentum point modestly lower: the March spike of 4.6% has fallen by 0.6 percentage points over two releases. Persistent housing and non-tradables inflation provide an upside offset, while fuel and other volatile components can create a sizeable month-specific base effect. The target is the same gross, unadjusted All groups CPI annual movement throughout; no trimmed-mean or seasonally adjusted variant is substituted.

Prior/update/interval: Starting from a persistence prior centered near the recent May 2026 reading of 4.0%, using the ABS Jul-25-to-May-26 reference class, I apply a modest easing adjustment for the March-to-May decline, retain an upside adjustment for 4.7% non-tradables inflation and housing persistence, and allow volatile fuel/base effects. The ladder-implied 80% central interval is 3.3% to 4.5%, with the rungs anchored by the fetched 3.0%, 3.2%, 3.6%, 3.8%, 4.6%, 4.2%, and 4.0% observations.

Ladder: P(X <= 3.1) = 0.05; P(X <= 3.3) = 0.10; P(X <= 3.5) = 0.20; P(X <= 3.7) = 0.35; P(X <= 3.8) = 0.45; P(X <= 3.9) = 0.55; P(X <= 4.1) = 0.70; P(X <= 4.3) = 0.82; P(X <= 4.5) = 0.90; P(X <= 4.7) = 0.95; P(X <= 4.9) = 0.97. Linear interpolation gives 10th percentile at 3.3, median at 3.85, and 90th percentile at 4.5; rounding to the ABS one-decimal print precision gives ciLow 3.3, pointEstimate 3.9, and ciHigh 4.5.

The main upside risk is a renewed fuel or housing-related acceleration that would land above the interval, especially at 4.6% or higher. The main downside risk is another large favorable volatile-item or base-effect move, with a result at or below 3.2% landing outside the interval.

Review disposition: Accepted the reviewer’s optional clarity suggestion by explicitly labeling 3.3%–4.5% as the 80% central interval. No required critique items were identified, so the resolver, source grounding, base-rate treatment, and forecast values remain unchanged.