thesisPrototype

Government data

US nonrevolving consumer credit annual rate, Jun 2026

Federal Reserve G.19 Percent Change of Total Nonrevolving Consumer Credit (NONREVSLAR), seasonally adjusted annual rate, June 2026 first print

Forecast

+1.9%
-0.3%–+4.1%
80% prediction interval
gpt-5.5Run record ↗

Cumulative probability

Chance that the outcome is at or below each value.

0%25%50%75%100%-3.6%+1.9%+7.4%Forecast value

Shaded band: 80% interval (-0.3%–+4.1%). Dashed line: point estimate (+1.9%).

Derived from the point estimate and 80% interval; the agent did not report a full distribution.

Observed outcomeinside 80% interval
actual
+2.3%
forecast
+1.9% with 80% interval [-0.3%, +4.1%]
error
+0.4% · absolute +0.4%
cdf score
CRPS 0.51 · PIT 0.57
source
federal_reserve_g19 G.19 Consumer Credit, outstanding, seasonally adjusted

First print for 2026-06 captured from https://alfred.stlouisfed.org/graph/alfredgraph.csv?id=NONREVSLAR&vintage_date=2026-08-07 on the official release date named by the cell's resolver.

Analysis

§Forecast for June 2026 US nonrevolving consumer credit annual rate

Target is the Federal Reserve G.19 seasonally adjusted nonrevolving consumer credit percent change at annual rate, series NONREVSLAR, for June 2026 first print. The catalog slug, unit, dataPointId, and 2026-08-07 resolution date match the ledger contract; the resolver URL is the ledger-bound ALFRED original-vintage series, while the economic release is the Federal Reserve G.19 table.

▸ Reported tool use: official.lookupmodel report
Opened Federal Reserve August 2026 statistical release calendar for G.19 Consumer Credit.
↳ The official calendar lists G.19 - Consumer Credit at 3:00 p.m. with release date 7 in August 2026, verifying the 2026-08-07 resolution date.
▸ Reported tool use: official.lookupmodel report
Opened Federal Reserve G.19 current release dated July 8, 2026 for the May 2026 print.
↳ The release states May 2026 total consumer credit was 0.0 percent annualized, revolving credit was -4.7 percent, and nonrevolving credit increased 1.6 percent annualized; the table also shows nonrevolving flow of 30.9 billion dollars annual rate and nonrevolving outstanding of 3810.3 billion dollars.
▸ Reported tool use: history.lookupmodel report
Opened FRED mirror for Board series NONREVSLAR to read recent exact monthly observations.
↳ NONREVSLAR observations fetched: May 2026 = 1.61, Apr 2026 = 2.93, Mar 2026 = 3.84, Feb 2026 = 1.94, Jan 2026 = 0.84 percent at annual rate; FRED shows update time Jul 8, 2026 2:01 PM CDT and next release date Aug 7, 2026.
▸ Reported tool use: official.lookupmodel report
Opened Federal Reserve G.19 Data Download Program page to confirm release identity and availability.
↳ The Board DDP page identifies G.19 - Consumer Credit as last released Wednesday, July 8, 2026, and lists Consumer Credit Outstanding (S.A.) as a 72.4 KB preformatted CSV package.

Variant check: all anchors are the same variant as the resolver, namely Consumer Credit Outstanding, seasonally adjusted, nonrevolving credit, percent change at annual rate. I did not use not-seasonally-adjusted flow or level series as substitutes for the forecast target.

Base rate / reference class: the recent 2026 monthly values themselves are 0.84, 1.94, 3.84, 2.93, and 1.61, with mean (0.84+1.94+3.84+2.93+1.61)/5 = 2.23 percent annualized. The 2025 annual nonrevolving rate in the official release is a weak context anchor at 1.8 percent, so the outside-view anchor is roughly 2 percent rather than the higher March-April pace.

Prior/update/interval: model prior is simple recent-mean persistence using the Jan-May 2026 NONREVSLAR mean of 2.23; a formal AR model was ruled out for this fast run because only the fetched same-vintage recent observations were used. Adjust down 0.25 for May's slowdown from Apr 2.93 to May 1.61 and down 0.10 for still-tight auto and personal-loan credit conditions, giving a 1.9 point estimate after rounding. Interval method uses the fetched 2026 monthly values themselves because this target is already a change-rate series: sample sigma = 1.17 percentage points, so 1.28*sigma = 1.50; because five observations are a thin volatility sample for a noisy annualized monthly series, widen the half-width to 2.2, which is 1.47x the mechanical half-width, giving 1.9 - 2.2 = -0.3 and 1.9 + 2.2 = 4.1.

Counter-considerations: upside risk would come from a rebound in auto-loan originations or federal/student-loan components strong enough to put June above 4.1 percent annualized. Downside risk would be a paydown-heavy or weak auto-credit month that lands below -0.3 percent. A technical break or unusually large holder reclassification would be outside the interval mechanism, though the G.19 percent-change method is designed to exclude breaks.

Review disposition: accepted the resolver critique by aligning the resolutionSourceUrl to the ledger-bound ALFRED NONREVSLAR series while preserving the Federal Reserve G.19 release as the substantive source; accepted the interval critique by explicitly flagging the five-month volatility sample and widening within the allowed sigma multiplier range; accepted the model-prior critique by naming the prior as simple recent-mean persistence and noting that no formal AR model was used.

Key drivers

  • May nonrevolving growth slowed to 1.61 percent annualized after stronger March and April prints
  • 2026 year-to-date readings average near 2.23 percent annualized, close to the 2025 annual pace
  • High borrowing rates and softer auto-credit growth cap upside
  • Student-loan and federal-holder flows can add month-to-month noise to nonrevolving credit
  • Thin recent volatility sample argues for a wider interval than the mechanical five-month sigma alone

Tool evidence

This run has no captured tool responses. “Reported tool use” in the analysis is the model’s account, not an archived tool response.

Sources and resolution

Official source
ALFRED/FRED official-vintage mirror for Federal Reserve G.19 Consumer Credit
Resolved
August 8, 2026
Resolution rule
Resolve to the first official Federal Reserve G.19 print for June 2026, table Consumer Credit Outstanding, seasonally adjusted, Nonrevolving percent change at annual rate / NONREVSLAR, in percent at annual rate. Use the original vintage for the June 2026 observation released on 2026-08-07 in the ledger-bound ALFRED source series; do not incorporate later revisions or corrected vintages unless the original release itself is withdrawn before publication.
Run details

The analysis is the model’s written report. Tool-use descriptions in that report are model claims; the activity artifacts contain the execution record.

thesis.analyst · gpt-5.5 · fast · v2.4.0

pre-submit review · completed

Draft is broadly coherent, but the resolver/source fields do not match the canonical ledger source binding and the interval method rests on a very thin volatility sample.

  • blocking resolver: Draft resolutionSourceUrl points to the Federal Reserve current G.19 page, while the ledger contract binds first-print resolution to ALFRED/FRED sourceUrl https://alfred.stlouisfed.org/graph/alfredgraph.csv?id=NONREVSLAR with allowedHosts ["alfred.stlouisfed.org"].
  • warning interval: The 80% interval is based only on five 2026 monthly observations, which is too short to establish realized volatility for a noisy monthly annualized series.
  • warning model_prior: The draft uses a persistence prior but does not clearly say whether a formal time-series/model prior was used or ruled out.
  • info optional_suggestion: Clarify whether the 2025 annual nonrevolving rate is being used as a weak anchor or only as context.
  • info optional_suggestion: Keep the tail scenarios tied to specific G.19 components such as auto credit, student-loan/federal-holder flows, and paydowns.

disposition accepted: Review disposition: accepted the resolver critique by aligning the resolutionSourceUrl to the ledger-bound ALFRED NONREVSLAR series while preserving the Federal Reserve G.19 release as the substantive source; accepted the interval critique by explicitly flagging the five-month volatility sample and widening within the allowed sigma multiplier range; accepted the model-prior critique by naming the prior as simple recent-mean persistence and noting that no formal AR model was used.

disposition accepted: Review disposition: accepted the resolver critique by aligning the resolutionSourceUrl to the ledger-bound ALFRED NONREVSLAR series while preserving the Federal Reserve G.19 release as the substantive source; accepted the interval critique by explicitly flagging the five-month volatility sample and widening within the allowed sigma multiplier range; accepted the model-prior critique by naming the prior as simple recent-mean persistence and noting that no formal AR model was used.

disposition accepted: Review disposition: accepted the resolver critique by aligning the resolutionSourceUrl to the ledger-bound ALFRED NONREVSLAR series while preserving the Federal Reserve G.19 release as the substantive source; accepted the interval critique by explicitly flagging the five-month volatility sample and widening within the allowed sigma multiplier range; accepted the model-prior critique by naming the prior as simple recent-mean persistence and noting that no formal AR model was used.

disposition not applicable: Review disposition: accepted the resolver critique by aligning the resolutionSourceUrl to the ledger-bound ALFRED NONREVSLAR series while preserving the Federal Reserve G.19 release as the substantive source; accepted the interval critique by explicitly flagging the five-month volatility sample and widening within the allowed sigma multiplier range; accepted the model-prior critique by naming the prior as simple recent-mean persistence and noting that no formal AR model was used.

disposition not applicable: Review disposition: accepted the resolver critique by aligning the resolutionSourceUrl to the ledger-bound ALFRED NONREVSLAR series while preserving the Federal Reserve G.19 release as the substantive source; accepted the interval critique by explicitly flagging the five-month volatility sample and widening within the allowed sigma multiplier range; accepted the model-prior critique by naming the prior as simple recent-mean persistence and noting that no formal AR model was used.

Activity artifacts

Complete original trace

§Forecast for June 2026 US nonrevolving consumer credit annual rate

Target is the Federal Reserve G.19 seasonally adjusted nonrevolving consumer credit percent change at annual rate, series NONREVSLAR, for June 2026 first print. The catalog slug, unit, dataPointId, and 2026-08-07 resolution date match the ledger contract; the resolver URL is the ledger-bound ALFRED original-vintage series, while the economic release is the Federal Reserve G.19 table.

▸ Reported tool use: official.lookupmodel report
Opened Federal Reserve August 2026 statistical release calendar for G.19 Consumer Credit.
↳ The official calendar lists G.19 - Consumer Credit at 3:00 p.m. with release date 7 in August 2026, verifying the 2026-08-07 resolution date.
▸ Reported tool use: official.lookupmodel report
Opened Federal Reserve G.19 current release dated July 8, 2026 for the May 2026 print.
↳ The release states May 2026 total consumer credit was 0.0 percent annualized, revolving credit was -4.7 percent, and nonrevolving credit increased 1.6 percent annualized; the table also shows nonrevolving flow of 30.9 billion dollars annual rate and nonrevolving outstanding of 3810.3 billion dollars.
▸ Reported tool use: history.lookupmodel report
Opened FRED mirror for Board series NONREVSLAR to read recent exact monthly observations.
↳ NONREVSLAR observations fetched: May 2026 = 1.61, Apr 2026 = 2.93, Mar 2026 = 3.84, Feb 2026 = 1.94, Jan 2026 = 0.84 percent at annual rate; FRED shows update time Jul 8, 2026 2:01 PM CDT and next release date Aug 7, 2026.
▸ Reported tool use: official.lookupmodel report
Opened Federal Reserve G.19 Data Download Program page to confirm release identity and availability.
↳ The Board DDP page identifies G.19 - Consumer Credit as last released Wednesday, July 8, 2026, and lists Consumer Credit Outstanding (S.A.) as a 72.4 KB preformatted CSV package.

Variant check: all anchors are the same variant as the resolver, namely Consumer Credit Outstanding, seasonally adjusted, nonrevolving credit, percent change at annual rate. I did not use not-seasonally-adjusted flow or level series as substitutes for the forecast target.

Base rate / reference class: the recent 2026 monthly values themselves are 0.84, 1.94, 3.84, 2.93, and 1.61, with mean (0.84+1.94+3.84+2.93+1.61)/5 = 2.23 percent annualized. The 2025 annual nonrevolving rate in the official release is a weak context anchor at 1.8 percent, so the outside-view anchor is roughly 2 percent rather than the higher March-April pace.

Prior/update/interval: model prior is simple recent-mean persistence using the Jan-May 2026 NONREVSLAR mean of 2.23; a formal AR model was ruled out for this fast run because only the fetched same-vintage recent observations were used. Adjust down 0.25 for May's slowdown from Apr 2.93 to May 1.61 and down 0.10 for still-tight auto and personal-loan credit conditions, giving a 1.9 point estimate after rounding. Interval method uses the fetched 2026 monthly values themselves because this target is already a change-rate series: sample sigma = 1.17 percentage points, so 1.28*sigma = 1.50; because five observations are a thin volatility sample for a noisy annualized monthly series, widen the half-width to 2.2, which is 1.47x the mechanical half-width, giving 1.9 - 2.2 = -0.3 and 1.9 + 2.2 = 4.1.

Counter-considerations: upside risk would come from a rebound in auto-loan originations or federal/student-loan components strong enough to put June above 4.1 percent annualized. Downside risk would be a paydown-heavy or weak auto-credit month that lands below -0.3 percent. A technical break or unusually large holder reclassification would be outside the interval mechanism, though the G.19 percent-change method is designed to exclude breaks.

Review disposition: accepted the resolver critique by aligning the resolutionSourceUrl to the ledger-bound ALFRED NONREVSLAR series while preserving the Federal Reserve G.19 release as the substantive source; accepted the interval critique by explicitly flagging the five-month volatility sample and widening within the allowed sigma multiplier range; accepted the model-prior critique by naming the prior as simple recent-mean persistence and noting that no formal AR model was used.

calibrated forecast · 80% CI
+1.9%[-0.3% · +4.1%]
Target metadata

Data point: fed.g19.consumer_credit_nonrevolving_annual_rate.2026_06.first_print

More government data forecasts