Red candles don’t lie; they just don’t tell you which NAICS code is bleeding.
When the BLS employment report crossed the tape, the headline hit crypto Twitter like a hammer: America’s information industry lost 23,000 jobs in August. Employment in that sector is now at its lowest level since 2015. Instantly, the usual narratives fired off. Tech layoffs are accelerating. AI is eating the knowledge economy. Prepare for a white-collar recession that will drag Bitcoin down with it.
Hold on.
I’ve spent years on a market surveillance desk in Dublin, watching the 8:30 a.m. ET BLS release while crypto positions were still being settled. I’ve learned to distrust every macro headline that uses the word “tech” as if it were a single BLS category. The August information-sector print is worth your attention. But not for the reason the first wave of commentary is screaming about.
The information industry is not “the tech industry.”
That distinction is not academic nerdery. It is the difference between a useful labor-market signal and a self-induced panic.
What the Information Industry Actually Is
The BLS uses the North American Industry Classification System. The sector in question is NAICS 51, officially called Information. That includes a broad and somewhat old-fashioned collection of industries:
- Publishing, including software publishing
- Motion pictures and sound recording
- Broadcasting, including radio and television
- Telecommunications
- Data processing, hosting, and related services
- Internet publishing, broadcasting, and web search portals
Notice what is missing: most crypto startups, most fintech engineering teams, most semiconductor jobs, and a huge share of what ordinary people call “tech.”
Software publishers are in NAICS 51. But a crypto exchange’s trading desk might be classified under finance. A blockchain infrastructure company doing server work might be under data processing. A crypto hedge fund is not in the information sector at all.
So when the media says the information industry lost 23,000 jobs, those jobs are mostly in telecom, traditional publishing, broadcasting, and parts of data services. This is not the same as saying “crypto engineers are getting fired by the thousands.”
Why does this distinction matter so much right now?
Because market sentiment moves on categories, not on underlying data. If traders believe the information sector is collapsing, they sell growth assets. Then rate-cut expectations rise. Then they buy Bitcoin again. That whiplash is not analysis. It is noise trading.
The deeper problem is that many market observers do not actually read the BLS establishment survey. They read Crypto Briefing articles about the BLS survey. By the time the data passes through a niche financial media filter, the nuance is gone. That is how “information sector drops 0.7%” becomes “tech is dead” and then becomes “sell risk assets.”
The Real August Print
Let’s put the actual data in perspective.
Total nonfarm employment in the United States is roughly 159 million workers. The information sector sits at around 3.05 million. That is less than 2% of all payrolls.
A monthly loss of 23,000 jobs in information sounds severe. In percentage terms, that is close to a 0.75% decline in one month. For a single sector, that is a heavy blow. The unemployment rate for the whole economy barely moves from such a number, but the sector itself is feeling real pressure.
Even more striking is the context. The report says employment in the information industry is at its lowest level since 2015. That means a decade of net hiring has been completely erased. The sector did not just dip. It gave back the entire post-2015 expansion.
This is not a rounding error.
But it is also not a sudden AI-driven apocalypse. The information sector has been under structural pressure for years. Traditional media companies have consolidated. Telecom companies have been shedding workers. Broadcast and cable companies have been disrupted by streaming. Publishing has been shrinking. The aggregate number hides a brutal internal story: old information economy jobs are dying faster than new information economy jobs are being created.
The most interesting part of the report is not just August. It is the revision to July. The article notes that July’s information-industry figure was also revised downward by about 20,000 jobs. That means the two-month drag is closer to 40,000 or more. In a sector of 3 million workers, that is a genuine warning light.
Still, I would not take one monthly print as a confirmed trend. The BLS establishment survey is based on a sample of businesses. Sector-level data can be volatile. And in an election year, job numbers attract political heat. The data is not fabricated, but it is noisy. The only responsible approach is to wait for the next two monthly reports.
What Crypto Gets Wrong
Crypto markets love macro narratives because macro narratives create liquidity cycles.
When the labor market cracks, the Federal Reserve is likely to cut rates. Lower rates are, in theory, good for risk assets. Bitcoin is often treated as a duration asset. So crypto traders begin to cheer bad jobs data.
There is a fatal problem with that logic. It works only until the bad jobs data becomes recessionary enough to hurt corporate earnings and consumer spending. At that point, rate cuts stop being “crypto fuel” and start being an emergency response. Risk assets sell off anyway. The exit liquidity is someone else—usually the person who sees one weak jobs report and assumes the Fed will save the market without collateral damage.
That is why I keep coming back to the sector breakdown.
Professional and business services—where many computer systems design and related tech jobs sit—have been more resilient. The people building software for banks, cloud infrastructure, and enterprise clients are still being hired in many cases. The 23,000 loss in information is not, by itself, evidence that white-collar hiring has collapsed.
If professional and business services starts losing jobs at a similar pace, then I will start preparing for a genuine labor market recession. That would matter to crypto far more than a drop in the narrow information sector. The information sector is important. But it is a leading indicator, not the whole picture.
The AI Narrative Is Too Easy
The temptation is to blame artificial intelligence for every job loss in information. That story has emotional power. It feels futuristic and inevitable. But the August data does not prove AI is the primary cause.
We should be honest about what the BLS can and cannot tell us. The employment report does not ask employers why they laid people off. AI is not listed in the survey questionnaire. So any commentator who says “AI caused these 23,000 job losses” is making an inference, not reporting a fact.
That does not mean AI is irrelevant. AI is absolutely reshaping content production, software engineering workflows, and data processing. But AI disruption is showing up across multiple industries, not just in one labeled sector. If AI were the main driver of this month’s loss, we would expect to see weakness in computer systems design as well. That has not fully happened yet.
Wash trading taught crypto to trust no volume number without looking at the order book. BLS payroll prints are no different. A monthly headline that says “information jobs fall to 2015 low” is like a token with suddenly huge volume. It could be real. It could also be the product of seasonal adjustments, business sampling, and a narrow industry definition that the rest of the market does not understand.
The trick is to look under the surface before you trade on it.
A Labor Market Built on Shifting Sand
Here is what does worry me.
The information sector is one of the highest-wage components of the American labor market. When information workers lose jobs, they cut spending. They also stop paying high taxes. The ripple effect spreads to housing, retail, and services.
The reason labor economists watch this sector so closely is that it often leads the broader white-collar cycle. Tech and media companies were among the first to announce big layoffs. Now, months later, the official data is reflecting that reality. If those layoffs spread into finance, legal services, and consulting, the cracks in the labor market will become much more obvious.
Crypto should care because crypto is not isolated from that liquidity cycle.
In a bear market, survival matters more than hope. Stablecoin issuance, DeFi activity, and exchange volumes all depend on the marginal dollar entering risk assets. That marginal dollar comes from workers who still have jobs and still feel confident about spending. When the information sector is shrinking, that confidence is beginning to erode.
Of course, there is a contrarian angle. A weak job market increases the chance of Fed rate cuts. Rate cuts increase the appeal of dollar-neutral assets like Bitcoin. That is why, in the short term, bad news for workers can be good news for crypto.
But that relationship inverts when the labor market falls below a certain threshold. Then the Fed is not cutting rates to stimulate growth; it is cutting rates because the economy is already in trouble. Risk assets rarely perform well in that environment.
That is the line I am watching.
If information employment drops by another 20,000 to 30,000 jobs next month, and if professional and business services starts to weaken, I will treat the macro backdrop as recessionary. That would not be a reason to buy Bitcoin on rate-cut hopes. It would be a reason to respect the possibility of a broader liquidity squeeze.
If, on the other hand, the information sector stabilizes or rebounds next month, the August print will look like an unusually sharp noise event. The market will brush it off, and the macro trade will return to inflation data and Fed speakers.
The Lesson from Old ICO Days
I learned this lesson the hard way.
In 2017, I was chasing ICO stories across Telegram groups. Every project claimed to be building the future. Most had no code. The best way to separate real projects from scams was to check GitHub commits, read the actual smart contracts, and ignore the marketing channels.
The same discipline applies to macro data.
Too many people read one news article and assume the underlying report matches the headline. They do not check whether the data is seasonally adjusted. They do not ask whether the sector definition matches their mental model. They do not look at the historical series to see how much noise appears in ordinary months.
If you want to use employment data in crypto trading, you have to read the actual BLS tables. Look at the information sector’s level. Compare it to the three-month moving average. Watch the diffusion index, which tells you how many industries are adding jobs. And always separate the “information” category from the broader “tech” category.
That is where the real information gain lives.
What to Watch Next
The next two monthly employment reports will tell us whether the information sector is in a systemic decline or just having a bad quarter.
First, watch the level. If information employment falls below the prior month’s revised level for three consecutive months, that is a trend.
Second, watch initial jobless claims. If the four-week moving average begins to climb toward and beyond 260,000, the labor market is cooling faster than the Fed wants.
Third, watch the professional and business services category. That is the larger, more modern slice of the tech economy. If it joins the information sector in negative territory, then the tech-driven labor recession is real.
Fourth, watch Fed language. The Fed does not react to one BLS report. It reacts to a pattern. If two consecutive reports show white-collar weakness, the tone will change. Crypto may rally on the first hint of rate cuts. But the rally will not be sustainable if the labor market is in genuine retreat.
In the meantime, do not let a single August jobs report turn into a fantasy about AI-driven doom or a fantasy about imminent Fed rescue.
The information sector is bleeding. That much is true. Whether the wound is cyclical, structural, or a statistical artifact will only become clear in the weeks ahead.
Red candles don’t lie. But they also don’t explain themselves. Your job is not to react to the first headline. Your job is to read the source data, understand the sector definitions, and decide whether the market is pricing a real shift or chasing an emotional narrative.
The exit liquidity is someone else—usually the person who treats the phrase “information industry” as if it were interchangeable with “crypto and tech.”
Do not be that person.
Watch the trend. Respect the noise. And only when the next two months confirm the direction should you call this a turning point.