Washington Apparently Knows Where the Spare Key Is
Last week had a lot of action - let's recap.
Scotty-B intervenes: the Treasury doubled the cap on long-end buybacks, $2B to at least $4B per operation. The next day Bessent went on CNBC and said it could go bigger than that.
Bitcoin wakes from the dead: +23% on the week, ~$1.9B into spot Bitcoin ETFs (best week of 2026), briefly poking above $79K before settling in the $77Ks.
The Fed says "not so fast.": July minutes showed a surprisingly broad willingness to hike if inflation doesn't cool. The vote itself was 9-3, with three regional presidents wanting a hike right then and there.
Everyone is still side-eyeing oil: Iran ruled out extending the June MOU and told Reuters its people should be ready to escalate. Brent is back above $90.
Jobs aren't collapsing: 206K initial claims, but continuing claims crept up to 1.80M. Still very much a low-firing, low-hiring economy.
Each data point is interesting on its own - geopolitics, Treasury, jobs, Bitcoin (BTC). Stack them and something more intriguing shows up.
Markets have been unsure how far the Fed is willing to go, and lately the honest answer looks like "up". Meanwhile, Treasury is leaning on a completely different lever: long-end buybacks. Nothing new in concept, just bigger. With enormous hyperscaler capex crowding into the same pool of long-duration capital, Treasury's willingness to absorb some duration matters at the margin.
Risk assets heard that liquidity was getting freed up and moved. BTC cleared $70K, ETF flows came back for five straight sessions, and record short liquidations poured gasoline on it. Trump also spent the week publicly pushing Congress on the Clarity Act, which didn't hurt.
Initial claims are in a decent spot right now - four weeks ago they hit 189K, roughly the lowest since 1969. There's no big layoff cycle. But the people who do lose jobs are taking longer to find another one, and continuing claims are drifting toward the top of their recent range.
This sets up the conflict:
Fed: inflation may require rates to go UP.
Treasury: long rates are too high.
Market: that's fine, we'll buy BTC.
The market is testing the waters to see whether "easing" can come from somewhere other than the Fed.
Treasury can put downward pressure on term premium. The White House can encourage the crypto and bond markets. Those decisions can loosen financial conditions while the Fed stays hawkish on rates. (Take a breath. Everything is going to be okay.)
An unusual period of fiscal easing + monetary restraint. Never a boring day with the fiat regime!
The Fed doesn't own the whole liquidity cycle. It controls the overnight rate and influences the curve. Treasury works through issuance composition, buybacks, and the broader plumbing of the Treasury market. Markets don't especially care who loosens financial conditions. They care that conditions are getting looser.
And no, Treasury buybacks aren't QE with makeup and a skirt. The Treasury isn't creating money to buy bonds - it has to fund the repurchases by borrowing elsewhere, effectively swapping long paper for short. TD Securities called it Treasury's own little version of Operation Twist, which is about right. It can improve liquidity and relieve pressure on parts of the curve at the margin.
That's where BTC gets interesting.
The bull case is straightforward: Fed cuts, yields fall, liquidity improves, BTC takes flight.
But what if Warsh doesn't cooperate?
BTC may not need the Fed to cut if Treasury is trying to keep long-term financing conditions manageable, the White House is encouraging crypto adoption, and institutional money keeps flowing through ETFs.
In other words, BTC might be trading Washington more than Warsh.
But Huston...we still have problems (two to be exact)...
The first is oil. Hormuz isn't a risk that might materialize - it's been closed or close to it for six months, transits are running a fraction of pre-war levels (see chart from TankerMap), and the IEA is warning that global stockpiles are draining fast. Any further escalation feeds inflation, keeps the Fed hawkish, and pushes yields higher from an already uncomfortable starting point.
The second is that we already ran the experiment, and it didn't go great. Treasury pulled the lever on Wednesday. The 30-year fell ten basis points. By Thursday it had erased the entire move and closed the week above where it started, near 5.27%, after touching a 19-year high the day before the announcement. The same week U.S. debt crossed $40 trillion.
So the tug-of-war...
On one side: Treasury, ETF inflows, and a crypto-friendly White House.
On the other: a hawkish Fed, an active war in the Gulf, an economy that isn't weak enough to force Warsh's hand, and a long end that just told Bessent it wasn't impressed (oooft).
Maybe the question isn't "when is the Fed going to cut?" - maybe it's "does the market even need the Fed to cut?"
For now, Bitcoin is betting the answer is no. The 30-year is betting the answer is yes.
Warsh can keep the cookie jar locked up. Washington apparently knows where the spare key is. The bond market hasn't decided whether it works.
Dates we will be watching watching:
August 28: Warsh's first Jackson Hole keynote as Chair. He's said he wants to frame big questions rather than give guidance, and 72% of fund managers believe no Fed rate hike will occur before the November midterm elections. Neutral is priced. Anything else isn't.
September 9: the larger buyback sizes take effect, running through November 4.
September 15: cloture vote on the motion to proceed to the Clarity Act. Needs 60. Not a vote on the bill itself, just a vote on whether the Senate gets to debate it.
September 15-16: FOMC. Same week. GG.
