Bitcoin Brief

Sup, freaks.

The Fed paused yesterday. The bond market kept tightening.

That distinction matters. It tells us where monetary power ends and market discipline begins.

Let's get into it.

LEAD STORY

The Fed Paused. The Bond Market Did Not.

The Federal Reserve held its target rate at 3.50% to 3.75% yesterday. Three voting members wanted a 25-basis-point hike. Chair Kevin Warsh offered almost no forward guidance and tried to reassure the market that inflation would keep moving in the right direction.

The bond market balked.

The curve split. That was the move that mattered.

According to the Treasury's official closing data, the two-year yield fell four basis points to 4.22%. The ten-year rose six basis points to 4.67%. The thirty-year rose eleven basis points to 5.20%.

The spread between the two-year and ten-year widened ten basis points. The spread between the two-year and thirty-year widened fifteen.

The curve sent a more complicated signal than "hike now." The front end accepted the pause and lowered the near-term policy path. The long end charged the government more for inflation, duration, and credibility.

The real-yield curve makes the message clearer. The ten-year real yield was unchanged at 2.41%. The entire six-basis-point increase in the nominal ten-year was accounted for by higher implied inflation compensation. At thirty years, the real yield rose six basis points while implied inflation compensation added roughly another five.

The market is saying that the price of money today and the credibility of money over time are two different things.

Goldman's post-meeting report called this a dovish hold. Warsh noted that rates are higher than they were at the previous meeting, that market rates "didn't pause," and that the markets have done quite a bit of tightening while the Fed has done little over the last 42 days. His argument is that the bond market can substitute for a policy-rate hike.

That may be true mechanically. It makes the situation worse.

If the Fed refuses to raise the overnight rate because it fears what another hike would do to debt markets and asset prices, the long end can raise borrowing costs anyway. Mortgages, corporate debt, infrastructure finance, and government refinancing price off the curve.

The stock market figured this out in real time. J.P. Morgan's desk recorded an initial 0.8% S&P 500 rally between 2 p.m. and 3 p.m. Then long yields turned higher and the rally disappeared. The S&P finished down 1.5%. The Nasdaq 100 fell 2.1%. Semiconductors, cyclicals, and momentum winners were hit hardest.

Warsh had plenty of accomplices. Oil rose roughly 7% as U.S.-Iran tensions escalated. Major technology companies were reporting earnings. Leveraged momentum trades were already being unwound across the United States and Asia. It would be lazy to pin every red screen on him.

It would also be lazy to ignore the timing. The Fed produced the relief rally. The long-end selloff killed it.

This morning's GDP release made the Fed's problem worse. Headline real GDP grew at only a 1.5% annualized rate during the second quarter, down from 2.1% in the first. That looks like an argument for patience until you inspect the composition.

Real final sales to private domestic purchasers grew 3.9%. The gross domestic purchases price index rose 5.7%. Quarterly PCE inflation ran at 5.1%, with core PCE at 3.4%.

Private demand is much stronger than the headline GDP number. Prices are still rising too quickly.

The monthly PCE report gave the doves their counterpoint. Headline PCE fell 0.1% in June and core PCE increased only 0.1%. Year over year, however, headline PCE remained at 3.7% and core PCE at 3.3%.

Both sides have evidence. Monthly inflation softened enough to justify waiting. Private domestic demand and the quarterly price data remain too hot for the Fed to declare victory.

That is how the two-year yield can fall while the thirty-year breaks to another cycle high.

South Korea shows what happens when rising discount rates collide with leverage. The KOSPI closed near 5,594 today, roughly 40% below its all-time high. J.P. Morgan says previous trend-following longs in the Nasdaq, Korea, Taiwan, and Japan have largely unwound. The ratio of leveraged-equity-ETF assets to underlying market capitalization has reversed roughly 85% of its April and May increase across the broader universe.

That is the good news. Much of the mechanical selling may have already happened.

The bad news is that leveraged exposure around memory stocks has retraced only about 55% of its prior increase, and U.S. retail margin leverage was still extremely elevated in June. July's margin data arrive in late August.

Korea offers a warning rather than a forecast. A legitimate investment thesis can become a massacre when momentum, concentrated exposure, and borrowed money pile on top of it.

Bitcoin can get smoked too. Higher real yields, a rising dollar, and forced deleveraging can hit bitcoin in the short run. Anyone claiming that yesterday's relative resilience proves complete decoupling is getting ahead of the evidence.

The structural bitcoin case is getting stronger anyway.

The Fed controls the overnight rate. It does not control the price the market demands for holding long-duration government debt. It cannot decree that inflation compensation, term premium, or fiscal risk remain low.

The Fed now faces three bad paths.

It can hike and defend purchasing power while increasing debt-service costs and pressure on leveraged markets. It can hold and let the long end tighten financial conditions on its behalf. Or it can ease into persistent inflation and sacrifice credibility to protect the debt machine.

Bitcoin sits outside that debtor-creditor chain. Its scarcity stands apart from a committee's ability to convince bondholders that inflation expectations remain anchored. The Treasury's refinancing needs and the Nasdaq's rescue needs cannot alter its monetary policy.

I have no clue what bitcoin does this afternoon. I do know it gives savers an asset that requires no faith in the central bank's ability to serve three masters at once.

The Fed paused. The bond market did not. Bitcoin keeps producing blocks.

SIGNAL

Kentucky Is Choosing Between Building Power and Banning Demand

Western Kentucky is showing two ways to handle power-hungry computation.

The Paducah American Energy Hub proposes pairing a 1.8-gigawatt AI campus with roughly 2 gigawatts of new gas generation, transmission upgrades, and up to 2.6 gigawatts of batteries. The project is proposed, not energized, and still needs regulatory approvals.

In Hopkinsville, an existing curtailable bitcoin mine reportedly supplies revenue that helps a 13,000-customer utility avoid rate pressure. Its proposed 50-megawatt expansion would finance a new substation and add about 15 megawatts of community capacity. Local zoning could choke it off.

New computational loads expose America's power shortage. The shortage came first. Make large loads finance generation, substations, and flexible capacity. Stop treating scarcity as a zoning achievement.

A Million Applications and a Border Government Cannot Control

Look at the footage from Ceuta. The crowds swimming around the breakwaters and pouring onto Spanish territory are overwhelmingly young men. Almost no women. Almost no children.

Ferric William puts the argument bluntly: when fighting-age men force a border in large groups, the receiving country has every reason to treat the event as a security threat. Modern refugee law allows men to seek asylum. Fine. A legal category does not erase what the camera shows or give anyone the right to force entry before being identified and screened.

Spain is simultaneously processing more than 1.17 million regularization applications while losing control of Ceuta. That is a betrayal of sovereignty. My read is simple: a mass forced entry by predominantly fighting-age men looks like an invasion, whatever bureaucratic label Madrid applies afterward.

The state owes Spaniards interception, detention, rapid adjudication and removal of those without a lawful claim. Pedro Sánchez and his government should be removed from power and held politically accountable for allowing this abomination to continue.

Washington Is Still Trying to Turn Code Into a Financial Institution

Treasury rejected new developer-liability language that it said would gut software protections and create liability where none currently exists.

The dispute centers on whether publishing non-custodial software makes a developer responsible for every future user of that code. Criminal operators remain prosecutable.

That line matters. An operator who takes custody, controls transactions, and knowingly participates in crime is a different actor from a developer who publishes tools other people run themselves. Collapse those categories and open-source financial software becomes legally impossible to maintain in the United States.

The rejected proposal has no force of law, and the final negotiating text remains fluid. Congress should target control, custody, knowledge, and intent. Writing code is not running a bank.

Bitcoin Developers Want to Make Stale Blocks Observable

Stale blocks are evidence that part of the network briefly followed a chain that lost the race. They can reveal propagation delays, validation bottlenecks, network partitions, or possible adversarial mining behavior. Today, much of that information disappears from ordinary node visibility.

A new bitcoin-dev proposal would add an opt-in staletip message so nodes can announce recent stale-chain headers and indicate whether they can serve the associated blocks.

The proposal is an unnumbered draft with a proof of concept and remains outside Bitcoin Core. Stale blocks expose symptoms rather than proving a specific cause. Relaying more branch information also creates bandwidth, fingerprinting, privacy, and denial-of-service questions.

Still, the direction is right. Bitcoin's network should be easier to observe without giving a central operator control over it. Better instrumentation lets the network diagnose itself.

The Hormuz Toll Booth Accepts Bitcoin

The Treasury Department alleges that an IRGC-linked maritime-insurance network forced vessels to buy coverage to cross the Strait of Hormuz. One of the designated operations, HormuzSafe, allegedly accepted bitcoin and other digital assets.

The sanctions are administrative allegations, not criminal convictions. Treasury reported neither a bitcoin seizure nor an interruption in oil traffic.

The episode still exposes the physical limit of permissionless money. Bitcoin can route around correspondent banks. It cannot route a tanker around a chokepoint, naval force, insurance demand, or port authority.

Money can be digital and borderless while energy remains stubbornly physical. The entity controlling the water, pipeline, grid, or loading terminal can still demand a toll. Permissionless settlement reduces financial gatekeeping. Geography keeps its veto.

⚡ FREEDOM TECH CORNER

The Bitcoin Moved Owners, Not UTXOs

Lxdev demonstrated a controlled swap of a 10,000-sat Mercury statecoin for one Liquid USDT.

The underlying bitcoin UTXO never moved. Mercury transferred ownership off-chain by replacing the owner's key share. Liquid confidential transactions concealed the stablecoin asset and amount from public observers. The two parties derived a shared viewing key so they could verify the exact USDT contract without obtaining authority to spend it.

That is a useful composition: private bitcoin ownership transfer matched with confidential stablecoin settlement, without changing bitcoin's consensus rules.

The trust has moved. It has not disappeared.

Lxdev controlled both sides of this low-value proof of concept. Mercury still depends on its Latch service and the assumption that previous owners delete their old key shares. The Liquid hashlock script remains visible. Proposed PTLC adaptor signatures, trusted execution environments, and federation improvements remain future work. The demo shipped without those protections.

The execution log, Liquid funding transaction, and Liquid claim transaction make the flow auditable.

The experiment matters because it moves privacy away from one-chain maximalism. Different systems can protect different pieces of a trade. The honest work is identifying which metadata disappears, which trust assumptions remain, and who can still stop the transaction.

DATA SNAPSHOT

As of July 30, 2026, 10:42 a.m. ET

See you tomorrow. This is not investment advice. Do your own research.

YouTube: https://www.youtube.com/@TFTC

podcast: https://www.tftc.io/tag/podcasts/