US Jobless Claims Drop to 194K: What's Driving the Surge in Continuing Claims?

 

The US Department of Labor published weekly unemployment insurance claims at 8:30 a.m. Eastern on October 22. Per the official release, seasonally adjusted initial claims for the week ending October 17 came in at 194,000, a decrease of 6,000 from the revised prior week, while seasonally adjusted continuing claims for the week ending October 10 reached 1,728,000, an increase of 25,000.

 

The 194,000 figure is strong on its own terms. Per press reports it is the lowest in three months, below the 200,000 expected in a Reuters poll of economists, and within striking distance of the 189,000 recorded in mid-July that was the lowest since 1969. Continuing claims at 1.728 million, meanwhile, is the highest in six weeks.

 

The two series moved in opposite directions, which is what this article sets out to explain. But before the explanation, four technical premises need stating, because together they establish that the real size of this divergence is far smaller than the headline suggests.

 

First, the two series cover different reference weeks. Initial claims cover the week ending October 17 while continuing claims cover the week ending October 10, a one-week gap, so strictly speaking they are not comparing the same period.

 

Second, both prior-week revisions moved in the direction that widens the divergence. Prior initial claims were revised up from 199,000 to 200,000, turning a 5,000 decline into a 6,000 decline, while prior continuing claims were revised down from 1,710,000 to 1,703,000, turning an 18,000 increase into a 25,000 increase.

 

Third, on a four-week moving average basis the divergence shrinks substantially. The initial claims average fell 1,000 to 197,750 while the continuing claims average rose only 4,000 to 1,711,000.

 

Fourth, on a year-over-year unadjusted basis the divergence disappears entirely. Unadjusted initial claims were 172,390 against 205,653 in the comparable 2025 week, while unadjusted continuing claims were 1,481,260 against 1,674,202 a year earlier. Both sit well below their levels of a year ago.

 

 

 

Initial claims beat expectations. Seasonally adjusted at 194,000, a decrease of 6,000, the lowest in three months and below the 200,000 expected in a Reuters poll.

 

Continuing claims hit a six-week high. Seasonally adjusted at 1,728,000, an increase of 25,000, while the seasonally adjusted insured unemployment rate held unchanged at 1.1%.

 

The reference weeks differ. Initial claims cover the week ending October 17 and continuing claims the week ending October 10.

 

Both revisions widened the divergence. Prior initial claims were revised up 1,000 and prior continuing claims revised down 7,000, together adding 8,000 to the headline gap.

 

The four-week averages diverge far less. The initial claims average fell 1,000 to 197,750 while the continuing claims average rose 4,000 to 1,711,000.

 

There is no divergence year over year. Unadjusted initial claims are down roughly 16.2% and unadjusted continuing claims down roughly 11.5% against a year earlier.

 

 

Listing every figure alongside its reference week is the precondition for avoiding misreadings.

 

 

The year-over-year comparison needs listing separately, because it points the opposite way from the week-over-week change. Unadjusted initial claims for the week ending October 17 were 172,390 against 205,653 in the comparable 2025 week, a decline of roughly 16.2% by our calculation. Unadjusted continuing claims for the week ending October 10 were 1,481,260 against 1,674,202 a year earlier, a decline of roughly 11.5% by our calculation.

 

The direction of the revisions warrants separate mention. Had prior initial claims stayed at 199,000, this week's decline would have been 5,000 rather than 6,000. Had prior continuing claims stayed at 1,710,000, this week's increase would have been 18,000 rather than 25,000. Together the two revisions added 8,000 to the divergence as presented in headlines. That is not a criticism of the data, since revisions are routine procedure, but it shows how sensitive a single week's change is to them.

 

 

The two measures capture different stages of the labour market, so moving in opposite directions is mechanically normal.

 

Initial claims measure the flow of new unemployment. A person enters the count only when filing a first claim after losing work, so it directly reflects the pace of layoffs. Initial claims running low means firms are shedding employees slowly, and 194,000 sitting near the lowest range since 1969 indicates layoff pressure is mild in historical terms.

 

Continuing claims measure the stock of people still drawing benefits. After a first claim, a person counts only by continuing to qualify and file each week, so it reflects how long the unemployed take to find work again. Rising continuing claims means the unemployed are exiting unemployment more slowly, which is to say re-employment has slowed.

 

Combining the two produces a simple identity: the change in a stock equals inflow minus outflow. Initial claims represent the inflow, while the outflow depends on re-employment or on ceasing to qualify. Falling initial claims alongside rising continuing claims therefore has only one explanation, that the outflow slowed more than the inflow, meaning fewer people are finding new jobs. That is how low layoffs and slow re-employment can coexist.

 

Continuing claims also carry two statistical limitations that cause the measure to systematically understate labour market slack. The first is eligibility, since per press reports continuing claims exclude recent college graduates and others who do not qualify because of limited work histories. The second is benefit duration, since most states cap benefits at 26 weeks, so the long-term unemployed drop out of continuing claims once benefits are exhausted without having found work.

 

Those two limitations mean rising continuing claims is a signal of a weakening labour market while continuing claims ceasing to rise is not necessarily a sign of improvement, since it may instead reflect benefits running out. Any judgement needs unemployment duration and payrolls data alongside it.

 

 

Placing this week's data in a wider picture yields an explanation with external corroboration.

 

Per press reports the labour market sits in a low-hire, low-fire state, with Eric Basmajian of EPB Research saying layoffs remain extremely low. Evidence on the hiring side comes from monthly data, where September nonfarm payrolls rose just 29,000 and the unemployment rate climbed to 4.2%. Per that reporting, the median duration of unemployment was 11.5 weeks in September, near a four-and-a-half-year high.

 

Those three figures form a complete causal chain. Low layoffs keep initial claims near historic lows. Slow hiring means the unemployed need longer to find work. And longer unemployment duration directly lifts the stock of continuing claims. Rising continuing claims alongside falling initial claims is therefore not merely possible but, under this structure, inevitable.

 

On why hiring has slowed, the explanations cited in reporting include uncertainty from the Middle East conflict and import tariffs discouraging firms from hiring. Worth noting is that those are views quoted in press coverage rather than conclusions drawn from official data.

 

The limits of a single week also need stating, and there are three. One is the reference week gap, with the two series this week separated by one week. Another is sensitivity to revisions, with this week's two revisions together adding 8,000 to the divergence. The third is seasonality, since September's weak payrolls were partly attributed in that reporting to a late Labor Day, while seasonal adjustment of weekly claims is likewise affected by holiday placement.

 

This week's data alone is therefore insufficient to conclude the labour market is weakening. The more robust approach reads the four-week averages and the year-over-year basis, both of which currently give far milder signals than the week-over-week change: the four-week divergence amounts to initial claims down 1,000 against continuing claims up 4,000, while on a year-over-year basis both sit well below a year earlier.

 

 

The purpose of cross-checking is to judge whether this week's change is noise or the start of a trend.

 

On four-week averages, initial claims fell from 198,750 to 197,750 while continuing claims rose from 1,707,000 to 1,711,000. The directions match the week-over-week changes while the magnitudes differ by an order of magnitude, indicating most of this week's divergence came from single-week volatility rather than a change in trend. That the continuing claims average is rising deserves recording, since it reflects the direction of change in re-employment speed better than a single week's figure.

 

On monthly data, September payrolls rose 29,000 with unemployment at 4.2%, against 133,000 in August per that reporting. Monthly data is revised in subsequent releases, so prior-month figures move. Either way, decelerating from six figures to under thirty thousand points the same way as rising continuing claims, and the two corroborate each other. For a full reading of the September payrolls report, see our September nonfarm payrolls analysis.

 

The insured unemployment rate offers another reference. It held unchanged at 1.1% seasonally adjusted and 1.0% unadjusted. That rate uses insured employment as its denominator, so holding steady while continuing claims rose 25,000 indicates the increase remains small relative to the size of that denominator.

 

One figure also tends to get overlooked. Total continued weeks claimed across all programs for the week ending October 3 was 1,485,572, a decrease of 40,033. That series covers a week earlier than seasonally adjusted continuing claims and a broader scope, and its direction is downward. The simple fact that three different measures point in different directions is itself the best demonstration that one week settles nothing.

 

Three verification milestones therefore follow: whether the continuing claims four-week average keeps rising; whether October payrolls published in November extend September's deceleration; and whether the median duration of unemployment continues lengthening. Only two of those three moving together would constitute grounds for a judgement about trend.

 

 

Transmission from employment data to rate expectations requires establishing the policy starting point first.

 

The Federal Reserve raised rates 25 basis points to 3.75% to 4.00% at its September 15 and 16 meeting, with the daily effective federal funds rate at 3.88% in October. Per press reports, economists expect no change at next week's meeting followed by another hike in December.

 

Market pricing converged clearly toward a hold this week. Per Kalshi data as of 8:30 p.m. Eastern on October 22, the probability of no change at the October 28 meeting stood at 92%, with a hike at 7%, split between 6% for 25 basis points and 1% for 50, and a cut at 1%, while fed funds futures implied 92% for a hold and 8% for a hike. Worth noting is that the page is internally inconsistent, with another section showing a 93.5% hold and a 5.5% hike. For comparison, the same tracker showed a hold near 87% to 88.5% and a hike near 12% the previous day. Pricing converged toward a hold across that interval, though this article attributes that shift to no single data release.

 

The December probability needs its basis noted. The 75% figure shown on that page is not the probability of a hike at the December meeting alone but the cumulative probability that the target range sits above the current level after that meeting, including a hike as early as October. The page provides no standalone December figure, while its dot-plot ladder implies a median end-2026 projection near 4.00% to 4.10% on thin volume.

 

Employment data cuts both ways within this framework. Initial claims near historic lows indicate the labour market has not deteriorated, which reduces the urgency of easing. Meanwhile unemployment at 4.2% was read by economists in that reporting as indicating no significant inflationary pressure from the labour market, which reduces the urgency of further tightening. Together they support a hold, which is exactly what the current 92% pricing reflects.

 

What can actually move pricing is inflation rather than employment. With the Federal Reserve having restarted hikes, prices dominate the decision inputs, so weekly claims matter mainly through whether they change the assessment of labour market resilience rather than by directly setting the rate path.

 

For digital assets the transmission is indirect. The sequence typically runs from the size of the data surprise into rate expectations, then through the dollar and real rates into risk asset valuations and liquidity expectations, with the mechanism covered in our analysis of how Fed rate hikes affect bitcoin. As of 9:13 p.m. Eastern on October 22, bitcoin traded near $81,397, down 2.36% over twenty four hours, for a market capitalisation around $1.64 trillion.

 

Monitor BTC pricing on MEXC alongside changes in US rate expectations, where the pair trades as BTC against USDT

 

 

For James Mitchell, the most valuable aspect of this data is not the divergence itself but how the size of the divergence changes with the basis used. On a week-over-week seasonally adjusted basis, initial claims fell 6,000 while continuing claims rose 25,000, which looks like a clear signal of disagreement. On four-week averages it becomes initial claims down 1,000 against continuing claims up 4,000, an order of magnitude smaller. On a year-over-year unadjusted basis the divergence disappears entirely, with initial claims down roughly 16.2% and continuing claims down roughly 11.5%, both well below a year earlier. One data set yielding three different strengths of conclusion across three bases indicates the headline disagreement is substantially a product of statistical convention.

 

The direction of the revisions also deserves recording. Prior initial claims were revised up 1,000 and prior continuing claims revised down 7,000, with both revisions happening to move in the direction that widens the divergence, together adding 8,000 to the headline gap. That is no indictment of data quality, since revisions are routine, but it shows the signal-to-noise ratio in judging trend from a single week's change is low. More consequential still is that the two series are separated by one reference week, so strictly speaking they are not comparing the same period.

 

That said, the divergence is not without information. Mechanically, the change in a stock equals inflow minus outflow, and initial claims represent the inflow, so falling initial claims alongside rising continuing claims can only mean the outflow slowed more, which is to say re-employment has slowed. That inference has external corroboration in monthly data, with September payrolls up just 29,000, unemployment at 4.2%, and median unemployment duration at 11.5 weeks near a four-and-a-half-year high. Low layoffs coexisting with low hiring is therefore an evidence-supported judgement, just not one for which this week's data is the strongest evidence, since the monthly data is.

 

The two statistical limitations of continuing claims warrant particular attention, because they will at some point produce a misleading improvement in the measure. Continuing claims exclude recent graduates and those with limited work histories, while most states cap benefits at 26 weeks, so the long-term unemployed drop out once benefits are exhausted. That means if the current slowdown in re-employment persists, continuing claims may at some stage stop rising or even fall because benefits are running out, which would not indicate an improving labour market. At that point the judgement should rest on unemployment duration and payrolls rather than on continuing claims themselves.

 

For rate expectations I regard this data's practical effect as limited but directionally clear. Initial claims near their lowest range since 1969 indicate the labour market has not deteriorated, weakening the case for easing. Unemployment at 4.2% indicates the labour market is not generating inflationary pressure, weakening the case for further tightening. Together they support a hold, and pricing already assigns that 92%. With the Federal Reserve having restarted hikes, what determines the path is the inflation components, while employment data sets the background assumption about labour market resilience. The turning point genuinely worth tracking is whether the continuing claims four-week average keeps rising, and whether that rise begins showing up as negative payroll growth.

 

 

 

For the week ending October 17, seasonally adjusted initial claims were 194,000, a decrease of 6,000 from the revised prior week of 200,000, with the four-week average at 197,750, down 1,000. For the week ending October 10, seasonally adjusted continuing claims were 1,728,000, an increase of 25,000 from the revised prior week of 1,703,000, with the four-week average at 1,711,000, up 4,000. The seasonally adjusted insured unemployment rate held unchanged at 1.1%. The Department of Labor published the data at 8:30 a.m. Eastern on October 22.

 

 

Because they measure different stages. Initial claims measure the flow of new unemployment and reflect the pace of layoffs, while continuing claims measure the stock of people still drawing benefits and reflect how long the unemployed take to find work. The change in a stock equals inflow minus outflow, so falling initial claims alongside rising continuing claims can only mean the outflow slowed more, which is to say fewer people are finding new jobs. That is the mechanism by which low layoffs and slow hiring coexist.

 

 

It warrants a discount for four reasons. The two series are separated by one reference week. Both prior-week revisions moved in the direction that widens the divergence, together adding 8,000 to the gap. On four-week averages the divergence shrinks by an order of magnitude to initial claims down 1,000 against continuing claims up 4,000. And on a year-over-year unadjusted basis the divergence disappears entirely, with initial claims down roughly 16.2% and continuing claims down roughly 11.5%.

 

 

Quite low. Per press reports it is the lowest in three months, below the 200,000 expected in a Reuters poll, and within striking distance of the 189,000 recorded in mid-July that was the lowest since 1969. On an unadjusted basis the week ending October 17 came in at 172,390 against 205,653 in the comparable 2025 week, a decline of roughly 16.2%. Layoff pressure is therefore mild in historical terms.

 

 

It is a signal worth watching but not conclusive on its own. Supporting that reading are September payrolls up just 29,000, unemployment rising to 4.2%, and median unemployment duration at 11.5 weeks near a four-and-a-half-year high. Weakening that reading is a four-week continuing claims average up only 4,000, with continuing claims on a year-over-year basis roughly 11.5% below a year earlier. A trend conclusion requires the four-week average to keep rising and payrolls to corroborate it.

 

 

Two statistical limitations cause it to understate labour market slack. Per press reports, continuing claims exclude recent college graduates and others who do not qualify because of limited work histories. And most states cap benefits at 26 weeks, so the long-term unemployed drop out of continuing claims once benefits are exhausted despite not having found work. That means continuing claims ceasing to rise is not necessarily a sign of improvement, and any judgement needs unemployment duration and payrolls data alongside it.

 

 

The effect is limited. The Federal Reserve already raised rates to 3.75% to 4.00% in September, and per Kalshi data as of 8:30 p.m. Eastern on October 22 the probability of no change on October 28 stood at 92% against 7% for a hike. Initial claims near historic lows indicate the labour market has not deteriorated, weakening the case for easing, while unemployment at 4.2% was read by economists as generating no significant inflationary pressure, weakening the case for further tightening. Together they support a hold. In the current environment the path is determined mainly by inflation rather than employment data.

 

 

The information above is provided for general market information and analysis only and does not constitute investment advice, financial advice, legal advice, tax advice or a recommendation to trade. Macroeconomic data and asset prices can both move sharply, and historical relationships and market-implied probabilities do not guarantee future results. The claims data, revisions, monthly employment data, rate pricing probabilities and asset prices cited here reflect publicly available information at the time of publication, weekly claims data is advance data subject to revision in later releases, monthly employment data is likewise revised, market-implied probabilities change continuously, actual policy decisions are determined by the official statements of the Federal Open Market Committee, and actual data is determined by the official releases of the Department of Labor and the Bureau of Labor Statistics. Year-over-year changes identified as our calculations are derived from official unadjusted data. Descriptions of changes in rate pricing attribute no causation to any single data release. The response of digital assets to macroeconomic data is indirect and unstable, with no reliable fixed relationship. Readers should conduct their own research and make decisions based on their own financial circumstances, investment objectives and risk tolerance, consulting a qualified professional where appropriate. The MEXC Crypto Pulse team accepts no liability for any direct or indirect loss arising from the use of this information.

 

 

James Mitchell specializes in technical analysis, market trends, and trading strategies for both Bitcoin and altcoins. Based in London, he has over 10 years of experience in financial markets. Before joining MEXC Learn, James worked as a senior analyst at a leading European investment firm, where he developed expertise in risk management and quantitative trading. His transition to cryptocurrency markets began in 2017, and he has since become recognized for his data-driven approach. He holds a Master's degree in Financial Economics from the London School of Economics. His analytical approach combines traditional technical analysis with on-chain metrics to provide readers with actionable insights.

 

Areas of Expertise: Technical Analysis, Market Trends and Cycles, Trading Strategies, Bitcoin and Altcoin Analysis, Risk Management.

 

 

 

Want the fastest access to MEXC's latest updates? Join our official Telegram group now!

MEXC Platform: Download App | Zero Fee Trading | Buy USDT Now

Join MEXC Community: X (Twitter) | Telegram | Discord

Get Insights: MEXC Learn | Terms of Service | MEXC Blog | MEXC News

Account Verification: Understand KYC | How to Complete KYC

External Content Platforms: Substack | Medium | Paragraph | LinkedIn | X(News)