thesisPrototype

Government data

U-6 Underemployment Rate, July 2026

What will the BLS first-print seasonally adjusted U-6 alternative labor underutilization rate be for July 2026 in Employment Situation Table A-15?

Forecast

7.9%
7.7%–8.1%
80% prediction interval
gpt-5.5Run record ↗

Cumulative probability

Chance that the outcome is at or below each value.

0%25%50%75%100%7.4%7.9%8.4%Forecast value

Shaded band: 80% interval (7.7%–8.1%). Dashed line: point estimate (7.9%).

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

Observed outcomeinside 80% interval
actual
7.9%
forecast
7.9% with 80% interval [7.7%, 8.1%]
error
0.0% · absolute 0.0%
cdf score
CRPS 0.043 · PIT 0.50
source
bls_cps Employment Situation, Table A-15

First print for 2026-07 captured from https://alfred.stlouisfed.org/graph/alfredgraph.csv?id=U6RATE&vintage_date=2026-08-07 on the official release date named by the cell's resolver.

Analysis

§Forecast for BLS July 2026 U-6 first print

Framing and exact resolver: this targets the BLS CPS household-survey U-6 alternative labor underutilization rate, seasonally adjusted, Table A-15, source code LNS13327709 mirrored by FRED/ALFRED as U6RATE. The ledger mechanically resolves through the ALFRED U6RATE first-vintage binding, while BLS Table A-15 is the underlying official release.

▸ Reported tool use: official.lookupmodel report
Checked BLS Schedule of Releases for the Employment Situation for the July 2026 reference month.
↳ BLS lists July 2026 Employment Situation release date as Aug. 07, 2026 at 08:30 AM; the same schedule lists June 2026 on Jul. 02, 2026 and August 2026 on Sep. 04, 2026.
▸ Reported tool use: official.lookupmodel report
Checked BLS CPS release calendar for the July 2026 reference period.
↳ BLS CPS calendar lists Aug. 07, 2026 for July 2026 Employment Situation, with nearby entries Jul. 02, 2026 for June 2026 and Sep. 04, 2026 for August 2026.
▸ Reported tool use: official.lookupmodel report
Fetched the current official BLS Employment Situation Table A-15 for recent seasonally adjusted U-6 values.
↳ BLS Table A-15 reports U-6 seasonally adjusted values of Feb. 2026 7.9, Mar. 2026 8.0, Apr. 2026 8.2, May 2026 8.1, and June 2026 7.9; the same table reports June 2025 seasonally adjusted U-6 at 7.7.
▸ Reported tool use: fred.lookupmodel report
Checked FRED/ALFRED U6RATE as the public history mirror and first-vintage resolution binding for the same BLS series LNS13327709.
↳ FRED/ALFRED U6RATE shows Jun 2026 7.9, May 2026 8.1, Apr 2026 8.2, Mar 2026 8.0, and Feb 2026 7.9, updated Jul 2, 2026 at 8:31 AM CDT, with next release date Aug 7, 2026.

Base rate/reference class: for a monthly labor-underutilization rate already near 8 percent, the strongest reference class is persistence plus small one-month CPS sampling and composition movement. The 2026 official/mirror history gives recent one-month moves of -0.2, +0.1, +0.2, -0.1, and -0.2 percentage points from Jan through Jun, so unchanged or a one-tenth move is the modal case.

Level, momentum, and mechanism: the latest level is 7.9 percent, down from 8.2 in April and 8.1 in May, while June U-3 in the same Table A-15 was 4.2 after 4.3 in May. That argues against a sharp U-6 rise, but U-6 is broader than U-3 and can move with marginal attachment and part-time-for-economic-reasons even when unemployment is steady.

Prior/update/interval: persistence prior = June 2026 U-6 at 7.9 percent; historical sample = Jan-Jun 2026 seasonally adjusted U-6 values 8.1, 7.9, 8.0, 8.2, 8.1, 7.9, with successive changes -0.2, +0.1, +0.2, -0.1, -0.2. Sample sigma = 0.18 percentage point. 80% normal half-width is about 1.28*sigma = 1.28*0.18 = 0.23 percentage point. Adjustment components: level 0.00, recent downward momentum -0.05, mean-reversion from June drop +0.05, policy/mechanism one-off 0.00, leaving point 7.9. Implied continuous 80% bounds are 7.9 +/- 0.23 = 7.67 to 8.13, rounded to 7.7 to 8.1 for the printed one-decimal series. The narrow interval is intentional because it is based on recent realized monthly moves and the one-decimal first-print granularity.

Counter-considerations: upside risk is a July jump in part-time-for-economic-reasons or marginal attachment that lifts U-6 to 8.2 or higher, which would land above the 80% interval. Downside risk is a broad household-survey improvement that pushes U-6 to 7.6 or lower, which would land below the interval. Outside the interval would most likely require a larger-than-recent move in the broader underemployment components rather than just a small U-3 change.

Review disposition: accepted the resolver critique by aligning resolutionSourceUrl and resolutionRule with the canonical ALFRED U6RATE first-vintage binding while keeping BLS Table A-15 as the underlying official release; accepted the calibration note by stating that the interval is based on recent realized monthly moves and one-decimal first-print granularity.

Key drivers

  • June 2026 U-6 first print was 7.9 percent after May 8.1 and April 8.2
  • recent seasonally adjusted U-6 moves are small, with five 2026 monthly changes of -0.2, +0.1, +0.2, -0.1, and -0.2 percentage points
  • June official U-3 unemployment was 4.2 percent, down from 4.3 percent in May, limiting near-term U-6 upside
  • July first print resolves on the August 7, 2026 Employment Situation release through the ALFRED U6RATE first-vintage binding

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 U6RATE first-vintage capture of the U.S. Bureau of Labor Statistics Employment Situation, Table A-15
Resolved
August 8, 2026
Resolution rule
Resolve to the July 2026 first-print U6RATE value captured by the ledger ALFRED binding, corresponding to the BLS-published seasonally adjusted U-6 value in Employment Situation Table A-15, in percent, as released on August 7, 2026. Use the first print rounded to one decimal place; do not apply later revisions or same-day correction exceptions unless the target binding itself records that first public vintage.
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 mostly publication-ready, but the resolver/source fields should be aligned with the canonical ledger binding before publication.

  • warning resolver: Draft JSON uses BLS Table A-15 URL as resolutionSourceUrl, while the ledger sourceBinding resolves mechanically through ALFRED U6RATE first-print capture.
  • info optional_suggestion: The Jan-Jun-only volatility sample is compact but very short; consider stating that the narrow 80% interval is based on recent monthly realized moves and rounded first-print granularity.
  • info optional_suggestion: The tail scenarios are adequate; adding explicit example values such as 8.2+ and 7.6 or lower already makes them concrete.
  • info optional_suggestion: Check that sourceContext does not imply FRED, rather than ALFRED first vintage or BLS Table A-15, is the final resolver.

disposition accepted: Review disposition: accepted the resolver critique by aligning resolutionSourceUrl and resolutionRule with the canonical ALFRED U6RATE first-vintage binding while keeping BLS Table A-15 as the underlying official release; accepted the calibration note by stating that the interval is based on recent realized monthly moves and one-decimal first-print granularity.

disposition not applicable: Review disposition: accepted the resolver critique by aligning resolutionSourceUrl and resolutionRule with the canonical ALFRED U6RATE first-vintage binding while keeping BLS Table A-15 as the underlying official release; accepted the calibration note by stating that the interval is based on recent realized monthly moves and one-decimal first-print granularity.

disposition not applicable: Review disposition: accepted the resolver critique by aligning resolutionSourceUrl and resolutionRule with the canonical ALFRED U6RATE first-vintage binding while keeping BLS Table A-15 as the underlying official release; accepted the calibration note by stating that the interval is based on recent realized monthly moves and one-decimal first-print granularity.

disposition not applicable: Review disposition: accepted the resolver critique by aligning resolutionSourceUrl and resolutionRule with the canonical ALFRED U6RATE first-vintage binding while keeping BLS Table A-15 as the underlying official release; accepted the calibration note by stating that the interval is based on recent realized monthly moves and one-decimal first-print granularity.

Activity artifacts

Complete original trace

§Forecast for BLS July 2026 U-6 first print

Framing and exact resolver: this targets the BLS CPS household-survey U-6 alternative labor underutilization rate, seasonally adjusted, Table A-15, source code LNS13327709 mirrored by FRED/ALFRED as U6RATE. The ledger mechanically resolves through the ALFRED U6RATE first-vintage binding, while BLS Table A-15 is the underlying official release.

▸ Reported tool use: official.lookupmodel report
Checked BLS Schedule of Releases for the Employment Situation for the July 2026 reference month.
↳ BLS lists July 2026 Employment Situation release date as Aug. 07, 2026 at 08:30 AM; the same schedule lists June 2026 on Jul. 02, 2026 and August 2026 on Sep. 04, 2026.
▸ Reported tool use: official.lookupmodel report
Checked BLS CPS release calendar for the July 2026 reference period.
↳ BLS CPS calendar lists Aug. 07, 2026 for July 2026 Employment Situation, with nearby entries Jul. 02, 2026 for June 2026 and Sep. 04, 2026 for August 2026.
▸ Reported tool use: official.lookupmodel report
Fetched the current official BLS Employment Situation Table A-15 for recent seasonally adjusted U-6 values.
↳ BLS Table A-15 reports U-6 seasonally adjusted values of Feb. 2026 7.9, Mar. 2026 8.0, Apr. 2026 8.2, May 2026 8.1, and June 2026 7.9; the same table reports June 2025 seasonally adjusted U-6 at 7.7.
▸ Reported tool use: fred.lookupmodel report
Checked FRED/ALFRED U6RATE as the public history mirror and first-vintage resolution binding for the same BLS series LNS13327709.
↳ FRED/ALFRED U6RATE shows Jun 2026 7.9, May 2026 8.1, Apr 2026 8.2, Mar 2026 8.0, and Feb 2026 7.9, updated Jul 2, 2026 at 8:31 AM CDT, with next release date Aug 7, 2026.

Base rate/reference class: for a monthly labor-underutilization rate already near 8 percent, the strongest reference class is persistence plus small one-month CPS sampling and composition movement. The 2026 official/mirror history gives recent one-month moves of -0.2, +0.1, +0.2, -0.1, and -0.2 percentage points from Jan through Jun, so unchanged or a one-tenth move is the modal case.

Level, momentum, and mechanism: the latest level is 7.9 percent, down from 8.2 in April and 8.1 in May, while June U-3 in the same Table A-15 was 4.2 after 4.3 in May. That argues against a sharp U-6 rise, but U-6 is broader than U-3 and can move with marginal attachment and part-time-for-economic-reasons even when unemployment is steady.

Prior/update/interval: persistence prior = June 2026 U-6 at 7.9 percent; historical sample = Jan-Jun 2026 seasonally adjusted U-6 values 8.1, 7.9, 8.0, 8.2, 8.1, 7.9, with successive changes -0.2, +0.1, +0.2, -0.1, -0.2. Sample sigma = 0.18 percentage point. 80% normal half-width is about 1.28*sigma = 1.28*0.18 = 0.23 percentage point. Adjustment components: level 0.00, recent downward momentum -0.05, mean-reversion from June drop +0.05, policy/mechanism one-off 0.00, leaving point 7.9. Implied continuous 80% bounds are 7.9 +/- 0.23 = 7.67 to 8.13, rounded to 7.7 to 8.1 for the printed one-decimal series. The narrow interval is intentional because it is based on recent realized monthly moves and the one-decimal first-print granularity.

Counter-considerations: upside risk is a July jump in part-time-for-economic-reasons or marginal attachment that lifts U-6 to 8.2 or higher, which would land above the 80% interval. Downside risk is a broad household-survey improvement that pushes U-6 to 7.6 or lower, which would land below the interval. Outside the interval would most likely require a larger-than-recent move in the broader underemployment components rather than just a small U-3 change.

Review disposition: accepted the resolver critique by aligning resolutionSourceUrl and resolutionRule with the canonical ALFRED U6RATE first-vintage binding while keeping BLS Table A-15 as the underlying official release; accepted the calibration note by stating that the interval is based on recent realized monthly moves and one-decimal first-print granularity.

calibrated forecast · 80% CI
7.9%[7.7% · 8.1%]
Target metadata

Data point: bls.cps.u6_underemployment_rate.2026_07.first_print

More government data forecasts