JOLTS job openings, May 2026
How many total nonfarm job openings will BLS first report for May 2026 in the Job Openings and Labor Turnover Survey?
Trend
history + forecaststatic prototype estimate · seeded forecast value
- actual
- 7.59m
- forecast
- 7.45m with 80% CI [6.8m, 8.15m]
- error
- +144k · absolute 144k
- cdf score
- CRPS 158 · PIT 0.58
- source
- bls_jolts JOLTS news release, Table 1
First print for 2026-05 captured from https://alfred.stlouisfed.org/graph/alfredgraph.csv?id=JTSJOL&vintage_date=2026-06-30 on the official release date named by the cell's resolver.
- record
- prototype seed
- agent
- prototype seed
- distribution
- 2 runs · 201 CDF points each
- ledger fact
- bls.jolts.job_openings_total.may_2026.first_print
Forecast runs
same target · agents, packs, updatespublic trace
JOLTS openings add a labor-demand target that resolves after payrolls but before many policy outcomes, giving agents a useful intermediate calibration signal. This target resolves on 2026-06-30 under a first-print rule, with an expected ~4 weeks lag. The same series can also spawn next release, +3 months, threshold questions.
April's preliminary openings jumped to 7.618 million after softer February-March readings. The agent partially fades that jump while keeping a wide interval because JOLTS first prints are noisy.
Validated live Codex-backed thesis.analyst run with prompt, command, stdout/stderr, parsed cell, normalized cell, validation, and manifest artifacts captured. Prompt mode: fast. Pre-submit review artifacts captured.
public trace
Draft is mostly publishable, but the first-print resolver should be tightened and the model-prior/interval rationale should be made more explicit.
- blocking resolver: Resolver names BLS JOLTS first print, but the source/rule could still accidentally use rounded news-release prose or later-revised time-series data.
- warning model_prior: The draft gives a persistence/base-rate prior but does not clearly state whether a time-series/model prior was used or explicitly ruled out.
- warning interval: The 80% interval is described as roughly +/-700 thousand, but it is based mainly on a few recent monthly moves rather than a stated realized-volatility calculation or explicit subjective uncertainty scale.
disposition accepted: Review disposition: accepted resolver tightening to specify the first-published table or archived first-print artifact and reject later-revised series values or rounded prose; accepted explicit persistence/mean-reversion prior; accepted clearer interval calibration as judgmental uncertainty tied to recent absolute monthly changes; kept Employment Situation evidence as pre-release context only.
disposition accepted: Review disposition: accepted resolver tightening to specify the first-published table or archived first-print artifact and reject later-revised series values or rounded prose; accepted explicit persistence/mean-reversion prior; accepted clearer interval calibration as judgmental uncertainty tied to recent absolute monthly changes; kept Employment Situation evidence as pre-release context only.
disposition accepted: Review disposition: accepted resolver tightening to specify the first-published table or archived first-print artifact and reject later-revised series values or rounded prose; accepted explicit persistence/mean-reversion prior; accepted clearer interval calibration as judgmental uncertainty tied to recent absolute monthly changes; kept Employment Situation evidence as pre-release context only.
The resolver is the BLS first print for total nonfarm job openings, seasonally adjusted, level in thousands, for May 2026. The forecast value 7350 means 7.350 million openings.
Base-rate/reference-class anchor: the 2025 average from fetched BLS monthly levels is about 7,082 thousand, while the first four months of 2026 average about 7,167 thousand. That keeps the outside-view anchor around 7.1 to 7.2 million rather than treating April's 7.618 million as the new trend.
Model prior: I use a simple persistence plus mean-reversion prior, not a fitted time-series model, because this fast run has few fetched points and JOLTS first prints have substantial survey noise that would make a lightly specified fitted model overconfident.
Level and momentum effects point in opposite directions. April's 7.618 million level is high relative to January through March, but May payrolls were solid and unemployment was unchanged, so I expect only a partial reversal of the April jump rather than a fall back to the March trough.
Counter-consideration: JOLTS openings can swing sharply on survey noise and industry composition. If April's spike was genuine vacancy rebuilding, May could print above 8.0 million; if it was sampling noise or employers pulled postings after tariff and demand uncertainty, a print near 6.6 million is plausible.
Recent monthly changes from fetched levels were Feb minus Jan = -318, Mar minus Feb = -35, and Apr minus Mar = +731 thousand; the average absolute move over those three changes is about 361 thousand. I place May 268 thousand below April but above the 2026 average: 7618 - 268 = 7350. For the 80% interval, I use judgmental uncertainty of about +/-700 thousand, roughly twice the recent average absolute move and wide enough to cover a large reversal or continuation: 7350 - 700 = 6650 and 7350 + 700 = 8050.
Upside scenario: openings remain near or above April because May payroll growth of 172,000 and upward revisions to March and April indicate resilient labor demand. Downside scenario: April's jump reverses and openings return toward the March 6.887 million level. Outside-the-interval scenarios require either an unusually large vacancy retrenchment below 6.65 million or a broad demand reacceleration above 8.05 million.
Review disposition: accepted resolver tightening to specify the first-published table or archived first-print artifact and reject later-revised series values or rounded prose; accepted explicit persistence/mean-reversion prior; accepted clearer interval calibration as judgmental uncertainty tied to recent absolute monthly changes; kept Employment Situation evidence as pre-release context only.
Key drivers
- Private labor demand
- Small-business hiring plans
- Payroll growth
- Quits and hires momentum
Resolution
- source
- Bureau of Labor Statistics, Job Openings and Labor Turnover Survey
- resolved
- July 10, 2026
- actual
- 7.59m
- rule
- Resolves to the first published seasonally adjusted total nonfarm job openings estimate for May 2026 in the BLS JOLTS release. Later monthly revisions do not change the resolved value.
- Data point
- bls.jolts.job_openings_total.may_2026.first_print
Series design
- series
- bls.jolts.job_openings_total
- cadence
- monthly · ~4 weeks
- horizon
- next release · first print
- priority
- P1
- benchmark
- Published consensus survey medians
- chainable
- next release · +3 months · threshold
Analyst agent · reasoning trace
static mock§
The route, resolution rule, and catalog entry are live. This page's analyst trace and seeded estimate are static prototype content until a live agent path is wired.