Menu
farexio
Tech

Will US tech layoffs be higher in 2026 than in 2025?

Resolution: Updated:
90%

market consensus

chance the market gives this event โ€” not your chance of being right

Yes โ€” The event happens
90%
No โ€” The event does not happen
10%

Trade this contract

Open Kalshi siteYes 0.90
  • No external wallet needed
  • gas covered
Buy the opposite sideNo 0.10

In short

The market treats this as very likely, and has done so from the moment prices were first recorded. The reason is arithmetic as much as sentiment: more than half of 2026 has already elapsed, the monthly JOLTS layoff readings for the information sector that have been published so far are on the table, and participants evidently judge that the running pace is at or above the 2025 rate. A sharp, sustained slowdown in announced restructuring across the second half of the year โ€” enough to pull the annual average below roughly 37,000 layoffs and discharges a month โ€” is what it would take to move this materially.

How the contract works

A contract on this outcome settles at $1 if the condition is met and at nothing if it is not. The price in between is simply the level at which buyers and sellers currently agree on the chance: a contract trading at 0.30 would mean the market thinks the event happens about three times in ten. Here, settlement depends on the calendar-2026 total for layoffs and discharges in the US information sector in the JOLTS series on FRED, using the figures available at settlement, which is expected by 1 March 2027. Revisions published after that date do not change the result. A position does not have to be held to settlement โ€” it can normally be sold beforehand at whatever the market price is at that moment.
What the market thinks happens
$100
Yes90%

The event happens

Costs now
$0.90
If you put in $100
$111
No10%

The event does not happen

Costs now
$0.10
If you put in $100
$1,000

Probability

0%25%50%75%100%12:0017:3623:1204:4810:2416:00
ConsensusKalshi

How the price has moved

There is very little movement to explain, and that is the point. The market was first recorded on 29 July 2026 at 89% and now sits at a consensus of 88%. Across 122 recorded price observations the entire range has been 88% to 89% โ€” a one-point band. On more than $31 million of volume, that flatness is a market treating the question as close to settled rather than one waiting for news. No single publicly reported trigger accounts for the one-point drift, and at this width it is indistinguishable from ordinary trading noise. The signal to take from the history is confidence, not direction: participants are pricing the residual chance that the remaining monthly JOLTS prints come in low enough to matter, and they judge that chance small.

Context

The question is settled by one specific official series: layoffs and discharges in the US information sector, as measured by the Bureau of Labor Statistics in the Job Openings and Labor Turnover Survey (JOLTS) and republished on FRED. That series covers software publishers, data processing and hosting, web search and other information services, telecommunications, broadcasting, publishing, and motion picture and sound recording. In 2025 it totalled 447,000. For this market to resolve Yes, the 2026 calendar-year total has to come in above that number.

Analysis

Start with the threshold, because it does most of the work. The 2025 total was 447,000 layoffs and discharges in the information sector. Spread across twelve months, that is an average of roughly 37,250 a month. Exceeding it in 2026 does not require a crisis; it requires the monthly pace to average marginally higher than last year's. Markets price questions of that shape confidently once most of the measurement period is behind them, and by late July 2026 the JOLTS releases covering the first several months of the year are already published and unambiguous. The market's high level is best read as a statement that those published months are not running short of the 2025 pace. The price history reinforces that. The consensus across venues sits at 88%. When the market was first recorded, on 29 July 2026, it was 89%, and across 122 recorded observations the whole range has been 88% to 89%. A one-point band is not a market weighing competing scenarios; it is a market that considers the question close to resolved and is trading the residual risk that the second half of 2026 collapses in the specific, narrow way the series measures. On liquidity, the picture is unusual. More than $31 million in volume has traded, and all of it on a single venue, Kalshi, which settles by FRED. That means there is no cross-venue spread to read here โ€” no second price to tell you whether different settlement sources or different user bases disagree. What the size does tell you is that the contract is not thinly traded: a price this stable on this much volume reflects genuine agreement rather than the absence of participants. The honest uncertainty is definitional, not directional. The information sector is a narrow proxy for what most people mean by "tech layoffs". Amazon's warehouse and retail workforce is classified under trade and transportation. Intel's manufacturing headcount sits in durable goods. Microsoft, Meta and Alphabet job cuts largely do land in information, but a year of loud AI-driven restructuring headlines can coexist with a soft JOLTS information print if the cuts fall outside the sector's boundaries or are spread thinly enough month to month. JOLTS is also a sample survey with monthly volatility large enough that a single outlier month can swing an annual total by tens of thousands. Finally, the mechanics of the release calendar matter. JOLTS is published with roughly a six-week lag, so the December 2026 reading โ€” the last piece needed โ€” arrives in early February 2027, comfortably ahead of the 1 March 2027 settlement window. There is no realistic scenario in which the market resolves on incomplete data, but there is a real scenario in which the final answer is decided by one or two borderline months published in the closing weeks.

What moves the probability

  • Half the year is already measured

    By late July 2026, JOLTS readings covering the first several months of the calendar year are published and fixed. That mechanically shrinks the range of possible annual totals and is the single largest reason the probability sits high. Each additional monthly release from here narrows it further, in whichever direction the data points.

  • The 447,000 bar is a pace, not a spike

    Beating the 2025 total means averaging above roughly 37,250 information-sector layoffs and discharges a month. That is a continuation threshold rather than a deterioration threshold, which pushes the probability up. It also means a modest cooling in the second half would not be enough on its own to flip the outcome.

  • Sector definition cuts both ways

    The information sector captures software publishing, data processing, web search, telecoms and media, but not Amazon's logistics workforce or Intel's fabs. Headline-grabbing cuts outside those boundaries do not register, which is the main channel by which a Yes could fail despite a bad-looking year for tech employment. This is the largest source of genuine uncertainty in the contract.

  • JOLTS survey volatility

    JOLTS is a sample-based survey and its monthly layoff figures move around sharply. One unusually low or unusually high month can shift an annual total by tens of thousands, which is meaningful against a threshold this close to the recent trend. It adds noise in both directions rather than a consistent bias.

  • AI restructuring cycle

    A fourth consecutive year of AI-driven reorganisation at the large US software and platform companies keeps the underlying separation rate elevated. If announcements continue at the pace seen since 2023, the pace requirement is met without difficulty. A pause in announcements would take several months to show up in the data, limiting how fast it could change the outcome.

The case for

  • More than half of 2026 is already measured, and the market's stability at a high level implies the published months are not running behind the 2025 pace.
  • Clearing the bar requires averaging only about 37,250 information-sector layoffs and discharges a month, a continuation of recent conditions rather than a deterioration.
  • Large US software and platform employers โ€” the core of the information sector โ€” have been restructuring continuously since 2023, and separations at that pace feed directly into this series.
  • Even a noticeably softer second half can still leave the annual total above 447,000 if the first half ran at or above last year's rate.

The case against

  • The information sector excludes much of what the public calls tech, so a year of loud layoff headlines at retail-classified or manufacturing-classified employers would not lift this series.
  • JOLTS is a survey with substantial monthly volatility, and two or three unusually low prints in the closing months could pull the annual total back below the threshold.
  • If AI restructuring at the major software firms has largely completed its headcount phase, the monthly separation rate could step down for the rest of the year.
  • The 2025 total of 447,000 is itself elevated, making it a harder comparison base than a normal year would be.

Trade this contract

Venues (1)

Open Kalshi siteYes 0.90
  • No external wallet needed
  • gas covered

Venues (1)

Resolution rules

Determined by
FRED (US Bureau of Labor Statistics JOLTS data on layoffs and discharges in the information sector)
Resolution date

The outcome is determined by the Bureau of Labor Statistics JOLTS series for layoffs and discharges in the US information sector, as published on FRED. If the calendar-2026 total exceeds 447,000 โ€” the 2025 total โ€” the market resolves Yes; otherwise it resolves No. The full year requires the December 2026 reading, published in early February 2027, and settlement is expected by 1 March 2027. Figures available at settlement are final for this purpose, and later revisions to the series do not alter the result. Only one venue currently lists the contract, Kalshi, and it settles by FRED, so there is no competing settlement source to create a price gap.

Calculation methodology โ†’

Local context

For readers in the US and UK working in or around technology, the JOLTS information-sector series is the closest thing to an official scoreboard for a restructuring cycle now in its fourth year. Company announcements are self-reported, inconsistently defined and often netted against hiring; JOLTS counts actual separations initiated by the employer, monthly, using the same definition every time. Whether 2026 comes in above or below 447,000 is therefore the difference between a cycle that is still intensifying and one that has plateaued โ€” a distinction that matters for wage bargaining, for how quickly laid-off engineers find new roles, and for how the Federal Reserve reads labour market slack in a sector that has driven a large share of recent US employment income growth. The reach extends beyond the United States. UK, Canadian, Australian and Indian technology employment is heavily tied to US-headquartered employers and their offshore and nearshore centres, and cost programmes decided in Seattle or Menlo Park are typically executed globally. US information-sector separations are a leading indicator of what follows in Bangalore, Hyderabad, Toronto, Dublin and London, even though those jobs never appear in the JOLTS count itself.

What to watch

The BLS publishes JOLTS monthly, roughly six weeks after the reference month, and each release updates the information-sector layoffs and discharges line on FRED. The prints covering June through November 2026 arrive between August 2026 and January 2027 and are the only inputs that can still change the answer; the December 2026 figure, published in early February 2027, completes the calendar year. Watch the cumulative year-to-date total against the roughly 37,250-a-month pace implied by the 447,000 threshold โ€” once the running total passes 447,000, the question is arithmetically decided regardless of later months. Also watch whether major restructuring announcements at large software and platform employers fall inside the information sector or outside it, because only the former reach this series. Settlement is expected by 1 March 2027.

Common questions

What exactly settles this market, and when?
The calendar-2026 total for layoffs and discharges in the US information sector, from the JOLTS series published on FRED. It resolves Yes if that total exceeds 447,000, the 2025 figure, and No otherwise. Settlement uses the data available at the time and is expected by 1 March 2027.
What does the current price actually mean?
A contract settles at $1 if the condition is met and at nothing if it is not, so the price is the market's estimate of the chance expressed in cents on the dollar. A price of 0.30 would mean roughly a three-in-ten chance. The number displayed above this page updates as trading moves it.
What happens if the JOLTS data is revised after settlement?
It does not change the outcome. The rules specify that the figures available at settlement are the ones used, and subsequent BLS revisions are disregarded. This matters because JOLTS is routinely revised, and without that rule the result could flip months after payout.
Does this cover all tech layoffs, or only some?
Only the information sector as the BLS defines it โ€” software publishing, data processing and hosting, web search and other information services, telecoms, broadcasting, publishing, and film and sound recording. Cuts at employers classified elsewhere, such as Amazon's logistics workforce in trade and transportation or Intel's manufacturing headcount in durable goods, are not counted here even when they are widely reported as tech layoffs.
Why is the market so confident this early?
Because most of the measurement period is already in the data. By late July 2026, JOLTS prints covering the first several months of the year are published and fixed, so the plausible range of annual totals is already narrow. The remaining uncertainty is about the second half and the survey's month-to-month noise.
Can a position be closed before March 2027?
Yes. Contracts trade continuously, so a position can normally be sold before settlement at whatever the price is at that moment. Holding to settlement is one option, not a requirement.

Related events

90%/ 10%
Yes / No
Open Kalshi site