Prolepsia forecastYes / no

Will the US unemployment rate be above 4.5% in December 2027?

As of 15 November 2026, Prolepsia puts the chance of yes at 27%.

27%chance of yes

Yes 27%No 73%
Prolepsia forecast made

How Prolepsia reasons

The report Prolepsia wrote with this forecast.

How forecasts are made · Track record

TL;DR

The forecast is 27.3% Yes and 72.7% No. Stable unemployment and low layoffs favor a December 2027 rate of 4.5% or lower, but weak job-finding leaves a meaningful risk of deterioration. A published rate of exactly 4.5% resolves No; Yes requires 4.6% or higher.

Key evidence

  • Historical comparisons favor persistence. Among modern historical months with unemployment between 3.8% and 4.4%, a stable preceding six-month trend, and no recession at the starting point, 16 of 69 observations reached at least 4.6% fourteen months later. These observations overlap and represent far fewer independent economic episodes (BLS unemployment series via FRED).
  • The starting point is reassuring. October 2026 unemployment was 4.1%, down from 4.2% in September, after remaining between 4.1% and 4.3% since March. October payrolls increased by 165,000, although the preceding twelve-month average gain was only 48,000 per month (BLS Employment Situation).
  • Layoffs are not surging. Initial unemployment-insurance claims were 200,000 for the week ending November 7, with a four-week average of 195,250. These benefit-system measures support near-term stability, not certainty about late 2027 (Labor Department).
  • The central outlook is below the threshold, but errors are large. The Federal Reserve projected median fourth-quarter 2027 unemployment of 4.1%. Its historical next-year forecast-error benchmark was 1.3 percentage points—large relative to the distance from the threshold (Federal Reserve projections).
  • Low hiring creates fragility. Kansas City Fed research found a job-finding rate near 34%, versus 39% in 2019, while job loss remained near its earlier level. A rise in layoffs would therefore meet a labor market that absorbs displaced workers slowly (Kansas City Fed).

Uncertainties

The main risk is a demand, financial, or policy shock before December 2027. Elevated inflation could limit the Fed’s ability to ease promptly if employment weakens. Conversely, improving hiring and continued low layoffs would reinforce the No outlook.

Historical comparisons use revised data, while resolution uses December’s initial publication, subject to the specified seven-day reporting-error correction. The Fed’s projection concerns a quarterly average, not December alone. Neither historical frequencies nor a below-threshold central projection eliminates the risk of a late downturn.

Read the full report

TL;DR

The forecast is 27.3% Yes and 72.7% No. The evidence favors December 2027 unemployment of 4.5% or lower: unemployment is stable, layoffs are low, and comparable historical starting points usually stayed below the threshold. The main threat is a downturn hitting a labor market where displaced workers already take longer to find jobs.

Context

The question concerns the national, seasonally adjusted U-3 unemployment rate for December 2027. U-3 measures unemployed people as a share of the civilian labor force; it is not the broader U-6 underutilization measure (BLS definitions). Yes requires an initially published rate of at least 4.6%. Exactly 4.5% resolves No. The first December report, expected in January 2028, determines the outcome, subject to the question’s seven-day exception for an acknowledged reporting or transcription error.

As of November 15, 2026, the latest observation is October 2026. That leaves a fourteen-month observation horizon. Unemployment is not already moving toward the threshold: October’s rate was 4.1%, and the March–October range was 4.1%–4.3% (BLS Employment Situation). The task is to weigh that stable starting point against the risk of a change in economic conditions before the target month.

Evidence

Historical experience favors No. A modern reference class starts with unemployment between 3.8% and 4.4%, a preceding six-month change no greater than 0.3 percentage point in either direction, and no recession in the starting month. From 1984 onward, 16 of 69 qualifying months had unemployment of at least 4.6% fourteen months later. A broader modern comparison without the stability restriction produced 20 successes in 86 observations (historical unemployment data). These are descriptive frequencies, not independent trials or direct probabilities for this event.

The central historical lesson survives changes in the comparison group. Narrower starting bands generally show fewer threshold crossings. A longer record includes more sharp deteriorations, but older labor-market conditions are less comparable to today. The evidence therefore supports persistence without making the recent narrow range a hard constraint (historical series).

The successes also cluster around downturns. In the modern stable comparison, successful starting months were concentrated in the 2000–2001, 2006–2007, and 2019 periods. Several nearby months can capture the same subsequent recession. Counting each month separately makes the record look deeper than it is. The useful conclusion is that a downturn is the main route to Yes, not that dozens of independent precedents establish a precise probability.

Current data support the persistence case. The supplied unemployment observations show the following path; May’s individual reading is not specified. The broader March–October range was 4.1%–4.3% (BLS release; FRED series).

Month in 2026 Unemployment rate
April 4.3%
June 4.2%
July 4.1%
August 4.1%
September 4.2%
October 4.1%

October payroll employment rose by 165,000, and revisions added 84,000 to August and September combined. The preceding twelve-month average payroll gain was much weaker, at 48,000 per month. This is a mixed picture: recent improvement against a slow longer-run hiring pace. Payroll employment comes from a separate survey and does not translate mechanically into U-3 (BLS).

The fall in October unemployment was not simply labor-force withdrawal. Household employment increased by 463,000 while the labor force grew by 421,000. Participation rose from 61.8% to 61.9%, and the employment-to-population ratio rose from 59.2% to 59.3% (BLS household-survey results). That makes the latest improvement more reassuring than a decline caused only by people stopping their job search.

Administrative claims provide a more recent check. Initial claims were 200,000 for the week ending November 7, with a four-week average of 195,250. Continuing insured unemployment was 1,697,000 for October 31, down 20,000; its four-week average fell 1,500 to 1,707,750. These measures cover unemployment-insurance beneficiaries rather than everyone counted in U-3. They nevertheless show no immediate broad layoff surge (Labor Department release).

Labor flows point in the same near-term direction. The available comparable observations show stronger hiring and quits, alongside fewer layoffs. September 2026 estimates are preliminary, and not every change is statistically significant (BLS JOLTS).

Observation Hires, millions Quits, millions Layoffs and discharges, millions Openings, millions
September 2025 5.244 2.990 1.816 7.169
August 2026, revised 5.271 3.044 1.768 7.190
September 2026, preliminary 5.402 3.365 1.641 7.523

The forward baseline also favors No. The Federal Reserve’s September projections put median fourth-quarter 2027 unemployment at 4.1%. But a central projection is not a bound. The same publication reports a 1.3-percentage-point historical next-year unemployment forecast error, measured as root-mean-square error using fall forecasts for 2006–2025. That uncertainty is large compared with the distance to the resolution threshold (Federal Reserve projections). The Philadelphia Fed’s third-quarter survey likewise placed its 2027 annual-average and third-quarter outlook at 4.3%; neither measure answers the December question directly (Survey of Professional Forecasters).

The strongest counterweight is slow absorption of jobseekers. Kansas City Fed research reports job-finding near 34%, compared with 39% in 2019, while job loss remained around 1.8%, near its 2019 average. Its worker-flow estimates use twelve-month moving averages. The mechanism matters: low job loss can keep unemployment stable even while weak job-finding makes the system more sensitive to a future layoff shock (Kansas City Fed research).

Duration data reinforce that concern, though they do not show a smooth deterioration. Here is the full June–October sequence supplied in the evidence (BLS household data).

Month in 2026 Mean duration, weeks Median duration, weeks Long-term unemployed, millions Long-term share
June 25.5 11.0 1.937 27.3%
July 24.9 10.5 1.771 25.5%
August 26.3 11.4 1.930 27.0%
September 24.8 11.5 1.944 27.1%
October 27.1 11.9 1.963 27.6%

Credit and policy add plausible transmission channels rather than proof of an approaching recession. The October lending survey reported tighter commercial-real-estate and nonbank-financial lending standards, plus weaker business-loan demand. Large-firm commercial-and-industrial standards were essentially unchanged. This is a modest adverse signal, not a generalized credit crunch (Federal Reserve lending survey).

Inflation also limits the assumption of automatic policy support. Headline CPI rose 0.6% in September and 0.4% in October; core CPI rose 0.3% in each of August, September, and October (BLS CPI). The Fed maintained a 3.75%–4.00% target range on October 28, with two dissenters preferring an increase. I read this as a constraint on rapid preventive easing if employment weakens, not as evidence that unemployment is already rising (FOMC statement).

What's non-obvious

Low layoffs and a fragile labor market can coexist. The headline rate describes the stock of unemployment. Hiring and job loss describe the flows that sustain it. Today’s reassuring claims data say that large numbers of workers are not suddenly entering unemployment. Weak job-finding says that, if this changes, workers may leave unemployment slowly. Those facts support a No forecast with a meaningful downside-economy risk, rather than either complacency or an immediate recession call.

Weak payroll growth is also not enough on its own to establish rising unemployment. Labor-force growth changes the number of new jobs needed to hold the rate steady, and research estimates of that breakeven pace differ substantially (Federal Reserve research; Atlanta Fed analysis). Likewise, evidence of AI-related weakness among young workers does not establish a national U-3 effect. The cited Census research reports an association in highly exposed early-career employment, not an identified increment to December 2027 unemployment (Census working paper).

Uncertainties

The largest unknown is whether a material demand, financial, or policy shock develops before the target month. Current claims are useful for the near term. They cannot rule out a downturn during 2027. Sustained increases in claims, falling hires, broader credit tightening, and worsening unemployment duration would supply stronger evidence of that transition. Continued hiring improvement and low layoffs would support persistence instead.

Historical evidence has two limits. First, overlapping observations contain few independent cycles. Second, the historical series uses revised unemployment rates, whereas this question resolves on the initial December release. The size and direction of that first-release mismatch have not been quantified. Archived initial releases would help close the gap.

The official forecast anchor also concerns a fourth-quarter average, not December alone. December can exceed the threshold even when the quarter’s average does not, and an earlier breach during 2027 does not settle the question. The decisive observation remains the published December rate. These uncertainties leave the saved forecast at 27.3% Yes and 72.7% No.

Forecast history

How Prolepsia’s forecast has movedOne forecast so far: 27% on 15 Nov 2026.
DateForecast
27% chance of yes

Sources Prolepsia read

  1. Concepts and Definitions (CPS) : U.S. Bureau of Labor Statisticswww.bls.gov
  2. Unemployment Rate (UNRATE) | FRED | St. Louis Fedfred.stlouisfed.org
  3. News Releasewww.dol.gov
  4. The Fed - September 16, 2026: FOMC Projections materials, accessible versionwww.federalreserve.gov
  5. Survey of Professional Forecasterswww.philadelphiafed.org
  6. The Low-Hire, Low-Fire Labor Market Is Calm but Fragile - Federal Reserve Bank of Kansas Citywww.kansascityfed.org
  7. Atlanta Fed analysiswww.atlantafed.org
  8. You’re (not) Hired: Artificial Intelligence and Early Career Hiring in the Quarterly Workforce Indicatorswww.alpha.cms.adsd.census.gov
  9. The Labor Market for Recent College Graduates - FEDERAL RESERVE BANK of NEW YORKwww.newyorkfed.org

The question

This question asks whether the United States' official, seasonally adjusted unemployment rate (U-3), as reported by the U.S. Bureau of Labor Statistics (BLS), will exceed 4.5% for December 2027. The U-3 unemployment rate measures the percentage of the civilian labor force that is unemployed, based on the Current Population Survey (CPS). It is the headline unemployment rate published in the BLS monthly Employment Situation report. As of November 15, 2026, the most recent available unemployment data covers October 2026, when the unemployment rate was 4.1%, down from 4.2% in September. The unemployment rate has remained relatively stable in recent months, fluctuating between 4.1% and 4.3% since March 2026. The question therefore concerns whether unemployment will increase sufficiently over the following 14 months to exceed the 4.5% threshold. The December 2027 unemployment rate is expected to be published in January 2028 as part of the BLS Employment Situation report for December 2027. The question concerns the unemployment rate specifically for December 2027, not the annual average unemployment rate for 2027 or the highest unemployment rate recorded during that year. Background sources: [BLS October 2026 Employment Situation](https://www.bls.gov/news.release/empsit.htm) and [BLS Unemployment Concepts and Definitions](https://www.bls.gov/cps/definitions.htm).

How it resolves

The question resolves YES if the U.S. Bureau of Labor Statistics reports that the seasonally adjusted U-3 unemployment rate for December 2027 is strictly greater than 4.5%, using the rate displayed to one decimal place in the first official Employment Situation report covering December 2027. A reported unemployment rate of 4.6% or higher resolves YES. The question resolves NO if the reported December 2027 unemployment rate is 4.5% or lower. A rate of exactly 4.5% resolves NO because the question specifies 'above 4.5%' rather than 'at least 4.5%.' The primary resolution source is the U.S. Bureau of Labor Statistics' Employment Situation report for December 2027, expected to be released in January 2028, available through https://www.bls.gov/news.release/empsit.htm. The relevant figure is the seasonally adjusted unemployment rate in Table A-1, 'Employment status of the civilian population by sex and age,' or its equivalent if the BLS changes its report format. The initially published December 2027 figure will be authoritative for resolution; subsequent revisions, seasonal adjustment updates, or methodological changes will not affect the outcome. If the initial report is delayed, resolution will be postponed until the BLS publishes the December 2027 unemployment rate. If the BLS permanently discontinues publication of the U-3 unemployment rate before December 2027 but publishes an explicitly designated successor measure with substantially equivalent coverage and methodology, that successor measure will be used. If no official December 2027 unemployment rate or substantially equivalent successor measure becomes available by December 31, 2028, the question will be annulled rather than resolved YES or NO.

Fine print

The unemployment rate refers to the national civilian unemployment rate for the United States, covering the 50 states and the District of Columbia, rather than individual states or territories. It is the standard U-3 measure, not the broader U-6 measure of labor underutilization. The seasonally adjusted rate will be used, rather than the unadjusted rate. The outcome will be determined from the published rate rounded to one decimal place, consistent with standard BLS reporting, rather than from an independently calculated rate using unrounded underlying data. The question does not require unemployment to remain above 4.5% for any minimum period. Unemployment rates in other months, including November 2027 and January 2028, will not affect resolution. The question will resolve based on the December 2027 reference month even though the corresponding data will be published in January 2028. If the first official publication is corrected by the BLS because of an acknowledged reporting or transcription error within seven calendar days, the corrected figure will be used; otherwise, the initially published figure remains authoritative.