DCP: 10-Year at 5.22% Won't Reverse Until Something Breaks

The bond selloff has outrun the Fed's own signals

DCP, a rates trading veteran who started on the floor in 1983, opened by noting the 10-year closed at 5.22% and that ZB futures are down almost 10 "big boy points" over two and a half months, which he called "truly amazing" given the market had been pricing in cuts before that. He linked the turn to a pop in yields around February that he says can be traced to the period around the Iran war.

He laid out conditions he believes are needed before the bond and STIR market can flip to a favorable long: the Fed signaling hikes are done, AI capex slowing, crack spreads rolling over, material weakness in indices, or a jobs shock. He dismissed a jobs-driven trigger, arguing employers are hoarding workers because the labor pool has shrunk: — DCP

On SOFR, DCP said there is no cut priced in out along the curve, and that the market has swung from pricing three-quarters of cuts to three-quarters of hikes and now roughly four hikes, which he called insane given that shift happened "since March." He said the SOFR curve is pricing in more hikes than the Fed's own recent dots.

Diesel and financing costs are quietly crushing Main Street

DCP argued the pass-through from diesel prices is going to hit "everything" over the next six to nine months, citing diesel at nine dollars a gallon in one location and seven dollars in Chicago. He said his own school catering business is paying double for beef versus a couple of years ago, and that farmers who locked in fertilizer pricing for this year face much higher costs next year.

He contrasted this with corporate credit access: an SBA 7(a) working capital loan is now around 12-13%, while a 504 loan can still be had in the "mid-60s" basis points below prime for larger amounts.

"There's a world of difference between a little guy trying to do an SBA loan and Apple sitting on all their reserves."

— DCP. He said he is watching smaller corporate bankruptcies in the $1-20 million range as a signal that stress is spreading.

Treasury intervention and Fed unity aren't enough to stop it

The host noted Treasury Secretary Besant conducted buybacks the same day, offering to buy $6 billion but only executing $4 billion, which added "fuel into the fire" in bonds. DCP was skeptical Treasury has the firepower:

"I don't think he has the gunpowder to do it... we're going to be at a $50 trillion deficit by the end of the year."

— DCP. He said every uptick in rates eats into issuance capacity and called the intervention comparable to "getting drunk and pissed off the back of the boat into the lake."

He described a near-unanimous vote among Fed governors under Worsh as surprising but welcome, saying Worsh needs unified support to make progress, and referenced a Dario Perkins tweet distinguishing problems the Fed can fix from those it can't, such as a runaway deficit or an energy supply shock.

AI capex, private credit and concentration are the fault lines

DCP said AI capex remains a key driver of what a guest referenced as an "inflationary expansion of all assets," and floated private credit, or a name like Oracle, as the likely trigger for stress, noting Oracle's bonds are "upside down by at least 10 bips" and that the company recently declared a force majeure. The hosts discussed how AI-related capex and markups (citing Anthropic funding rounds) are propping up both equity indices and reported corporate profit and GDP growth.

A list shared from Compound248/Martin Pelletier was read out showing steep declines from all-time highs across many "relatively good companies," including Nike down 84%, PayPal down 83%, Disney down 50%, Boeing down 55%, and Trade Desk down 98%, with the host noting "none of them are AI." A Goldman-sourced chart was cited showing over half of S&P 500 constituents now have negative beta to the index, which the host framed as evidence that a small number of mega-cap names are driving the index while the majority move opposite it.

Geopolitics could still flip the trade, but the panel is split on likelihood

The host raised the possibility that resolution in Russia, Iran or an agreement with China could reverse the bond selloff, especially with meetings in Washington around the taping. DCP was skeptical the Iran situation resolves easily, framing Iran, Russia and China as an "axis of power" working to complicate matters for the US, and said:

"If I were Iran, I would not negotiate with the United States... I would be holding out for a new president to negotiate with."

— DCP

He also cautioned that intelligence visibility into Iran may be weaker than during the Israel-Hamas conflict, meaning outside observers don't know how close the regime might be to cracking, though he noted dictatorships can sustain hardship differently than democracies by rerouting resources to loyalists.

Levels DCP is watching before turning long

DCP said he does not believe this is the "gen carry trade" unwind but rather a structural shift, and floated that par (yields near 6%) on the long end could produce a capitulation event in bonds. He said the economy could "easily handle" 6.25% on the 30-year and 5.75-6% on the 10-year, though it would "crush" refinancing-dependent borrowers and pressure regional banks and CRE; he pointed to KRE sitting on its 200-day moving average as a signal to watch, noting Worsh specifically flagged regional lending risk.

He is watching for an opportunity to fade the roughly four hikes priced into SOFR's further-out contracts, saying he wants to see the contract "overextend" by about 8-12 pips past a quarter-point level, calling 9487 (implied ~1.12 hikes) an attractive entry versus the current level pricing closer to four hikes. He is also watching 103 on ZB futures and the March lows in the ES as key reference points.

Where a short would go if DCP took a bearish equity view

Asked whether he'd rather short "laggard" Main Street names or mega-cap AI leaders, DCP said he'd rather target an AI leader he believes is "just wrong" than short already-beaten-down laggards, comparing AI enthusiasm to leverage-driven excess in 2008 and the dotcom era:

"The only money being made right now in AI is trying to buy the leader and hope you can get out before it's over."

— DCP. He said he expects "mainstream" winners like Google and Meta to ultimately prevail among AI-exposed names, but declined to name which model provider he sees as the weak link, and said he doesn't think markets are at that breaking point yet since none of the AI players have turned a profit.