The bar on Bořivojova had that 2 a.m. yellow light that makes everyone look like they're winning. It was the second winter of the bear market, and my phone was buzzing against the table in a rhythm I'd learned to read — not panic, not euphoria, just the steady fibrillation of a group chat that can't decide whether it's mourning or arguing.
Tomasz used to trade rates at a bank in Frankfurt. Now he sells insurance in Prague and tells himself he's happier. He read a headline out loud from somewhere in the crypto press: Federal Reserve rate hikes are about Wall Street, not inflation, an economist argues.
He laughed. I didn't. Not because I disagreed, and not because I agreed — because I've heard that sentence before, in a different costume, from my own side of the fence. In 2017 someone in a Telegram group told me the rug pull was the market correcting itself. In 2020 someone told me the oracle exploit was a stress test. In 2021 someone told me the failed mint was gas being gas. Every time, the sentence explained nothing and absolved everyone.
That's what the headline does. It sounds like an indictment. It's actually a shrug.
Here's the number that matters more than the headline: across the four tightening cycles I've watched from inside this industry, on-chain borrowing costs on dollar stablecoins have tracked the U.S. risk-free rate with a lag measured in days, not quarters. Not because DeFi is wise. Because DeFi is a currency board with a Discord server. That single mechanical fact tells you more about whether the Fed hikes for Wall Street than any unnamed economist ever will.
What we actually have
Let me lay out the information base, because it's thin and I refuse to pretend otherwise. A crypto outlet, Crypto Briefing, published a short item relaying a claim from an economist who is not named and not linked. The claim: the Federal Reserve's rate hikes serve Wall Street rather than the fight against inflation. Two derivative judgments follow — the hikes affect market dynamics, and they affect the profitability of financial institutions.
That's it. No core PCE print. No dot plot. No terminal rate. No balance sheet discussion. No mention of employment, even though maximum employment is literally half of the Fed's statutory mandate. A serious argument about the Fed's reaction function that never mentions the unemployment rate is like a post-mortem that never mentions the code.
So before I tell you what I think of the claim, let me tell you what it's actually pointing at, because there's a real conversation underneath it and it's worth thirty seconds.
Central banks are said to operate under three kinds of dominance. Fiscal dominance is when a government's borrowing needs dictate monetary policy — the central bank keeps rates low because the sovereign can't afford to refinance. Financial dominance is when the stability of the financial system, and by extension asset prices, constrains the central bank's choices. And the orthodox story is monetary dominance: the central bank sets the price of money according to its mandate and everyone else adapts.
The headline is a maximalist version of financial dominance. It says the third thing isn't real and the second thing is the whole game. That's a spicy position. It's also, as stated, basically unfalsifiable — no data offered, no mechanism specified, no competing explanation engaged with.
But here's why I don't just throw it out. Because I have watched the exact dynamic it describes, in miniature, in code, in my own projects. I have watched a protocol's governance vote swing not on the merits but on what the largest holders' liquidation prices implied. I have watched a DAO rewrite its emissions schedule overnight because a whale's health factor was about to break. I have watched the tail wag the mandate — not in Washington, but on-chain, in front of me, at 4 a.m., in Prague. And if it happens in a $40 million DeFi protocol with eleven people in a Telegram group, I have no trouble believing it happens in a $27 trillion Treasury market with eleven people in a boardroom.
Walls crumble when the party truly begins. The question is who's holding the wall.
What a rate actually is
Strip the ideology and a policy rate is three prices stacked on top of each other.
It's the price of duration — how much you pay today to receive dollars later. It's the price of collateral — what it costs to borrow against your assets. And it's the price of trust in the unit of account itself.
When the Fed moves the federal funds rate, it moves all three at once, and it moves them mostly through the plumbing: bank reserves, the reverse repo facility, dealer balance sheets, the Treasury's issuance calendar. That's the part nobody tweets about, and it's the part where the "for Wall Street" story dies a quiet death — and then gets reborn in a more interesting form.
Think about the 2022–2023 cycle. The Fed took the target range from 0–0.25% to 5.25–5.50% in sixteen months while running down a balance sheet that had peaked near $8.9 trillion. That's a machine designed to do one thing: make it expensive to hold duration and attractive to hold cash. Reverse repo balances, which had peaked above $2 trillion in late 2022 as money funds parked cash at the Fed, drained to near nothing as bills paid better. Reserves redistributed. Dealers absorbed issuance. None of that is a conspiracy. It's arithmetic.
Now ask the naive question: if the Fed hikes for Wall Street, whose earnings go up?
The honest answer is almost nobody's, at least not in the way the headline implies. Higher rates raise banks' nominal net interest margins — sure — but they also mark down every held-to-maturity bond portfolio on the books. Silicon Valley Bank filed the receipt for that trade in March 2023: roughly $15 billion of unrealized losses on its securities book, a deposit run conducted at the speed of a group chat, and a resolution. Then the Fed opened a lending facility to stop the bleeding. Read that sequence twice. The Fed hiked until a bank broke, then built a new window to keep the system standing. That is not Wall Street writing the script. That is the plumbing choosing the Fed's next move.
So the headline's thesis is wrong in its simple form. But it's right in a form the headline can't articulate. Let me try.
Three readings of "for Wall Street," and why two collapse
Reading one: the Fed hikes to protect bank profits. This collapses immediately. High rates hurt bank profitability in the tail, blow up duration exposure, and raise credit costs. The 2023 regional banking episode was not a profit party.
Reading two: the Fed hikes to protect asset prices — to prevent equities and credit inflating into a bubble that would later destroy the system. More defensible, but self-contradictory. Hiking compresses valuations by construction. A central bank protecting asset prices by crushing them is a strange kind of patron.
Reading three: the Fed hikes to protect the credibility of the dollar as the world's collateral, and financial stability is the proximate constraint through which that mandate gets expressed. This one survives. It explains the hiking, it explains the pivot to liquidity support when something broke, and it explains why the Fed's communication is always about anchoring expectations — because the dollar's value is, at bottom, an expectation.
That third reading is the one worth taking seriously, and it's the one the crypto press won't write, because it isn't a story. "Central bank manages a global collateral system under market and political constraints" is not a headline. "Central bank dances for bankers" is.
The reflexive loop, and why the Fed is a price-taker
Here's where my own experience gives me a weird angle.
I spent 2022 hosting weekly drinks in Prague's Jewish Quarter for developers, traders, and skeptics. Bear Market Bar Stories. The point wasn't to be right about the Fed. The point was to watch how fast a room of intelligent people converged on a narrative that made them feel less stupid about their losses. And the narrative that won, almost every week, was: the Fed is against us.
That narrative is emotionally satisfying and analytically lazy. The Fed isn't against you, and it isn't for Wall Street either. It's reading your prices.
Since the 1990s, and especially since 2008, the Fed has increasingly used the market's own forward curve as an input. Fed funds futures price in the likely path; the FOMC then moves roughly in line with what's priced, with occasional deliberate surprises. That isn't leadership. That's a feedback loop with a lag. If the market prices a hike and the Fed delivers a hike, who moved whom?
When expectations are endogenized into the policy rule, the central bank becomes the largest reflexive trader in the world. It holds the position the entire market is trying to front-run, and its edge comes from being able to change what it holds.
Now I have to be careful, because it's easy to slide from "reflexive" into "captured." They're not the same thing. A reflex is a mechanism. Capture is an intent. The headline asserts capture. The mechanism it should have described is reflex — and reflex is exactly what I've watched eat DeFi protocols alive.
Crypto built the same machine, and it's uglier
Now the part my own side doesn't want to hear, and the reason I'm writing this at all.
In 2020 I was a mid-level dev helping a yield aggregator called VaultPrime launch out of Prague. We had 300% APYs, weekly DeFi Dive parties in my apartment, and documentation written on napkins. I was too busy celebrating to look at the oracle. When the manipulation hit, $2 million left the building.
The lesson everyone took was "audit harder." The lesson I took was different: our APY wasn't a yield. It was a transfer. We were paying users in our own token to pretend our TVL was real. The moment emissions stopped, the depositors left. Not the smart ones — all of them. Because there was never a user. There was a subsidy wearing a user's costume.
That's the same disease the headline describes, running inside crypto's own body. And we diagnose it in the Fed while ignoring it in ourselves, because it's easier to be angry at a building in Washington than at a spreadsheet in our own repo.
Look at where the actual carry is this cycle. The most profitable entity in the crypto economy for two straight years has been a stablecoin issuer whose business model is holding short-dated U.S. government debt. That's not a DeFi protocol. That's a money market fund with a Telegram channel and a marketing budget. Its entire margin is the policy rate. It didn't decentralize the dollar. It bought the dollar's yield curve and sold it back to you as a token.
Then look at the synthetics. The delta-neutral basis trade — spot plus short perp, packaged as a yield-bearing stable — pays what it pays because the funding rate pays. Strip out the leverage wrapper and it's a leveraged bet that perpetual funding stays positive. The APY is a beta on the Fed's balance sheet wearing a delta-neutral hat. When the basis compresses, the product doesn't adapt. It shrinks.
And the plainest evidence of all: I can watch Aave's USDC supply rate in real time. When the risk-free rate was near zero, that number was near zero. As the policy rate climbed through 2022 and 2023, that number climbed with it, within days. It didn't climb because DeFi had become more productive. It climbed because DeFi is a currency board that imports the world's price of money.
So when I read "the Fed hikes for Wall Street," my first reaction isn't agreement or disagreement. It's: brother, you are Wall Street. You just don't have a Bloomberg terminal.
Survival is the first layer of value
This is a bear-market piece in disguise, so let's talk about what actually matters to readers with money on the line: which parts of this system bleed when the plumbing moves.
Start with the denomination. Ask what a protocol's revenue is denominated in. If it's denominated in the policy rate — stablecoin reserves, RWA vaults, T-bill wrappers, basis trades — then its income statement is a leveraged expression of the Fed's decision function and nothing else. Those protocols aren't safe havens. They're the most rate-sensitive instruments in your portfolio. You can call it yield. It's duration.
Then the subsidy question. Ask whether the yield survives the emissions. If a lending market pays 12% to suppliers while underlying borrower demand pays 4%, the difference is a token transfer from a foundation wallet. It has a half-life. The subsidy is not the yield; it's the marketing budget. I learned that on a napkin in my own apartment, and I'd like you to learn it without losing $2 million.
Then the freeze question. Which brings me to the part of the stack I've been the most publicly annoying about.
The sequencer is a node with a marketing department
I've said this for two years and I'll keep saying it: "decentralized sequencing" has been a PowerPoint for two years. In practice, a Layer 2 sequencer is a single operator that orders transactions, decides inclusion, and — critically — decides what happens when things go wrong. You can have a validity proof covering the state transition and still have one machine deciding the order in which your transaction enters the block. Those are two different guarantees, and only one of them is sold to you.
Why does that matter in a Fed article? Because the pitch for L2s has always been "we inherit Ethereum's security." What they inherit is settlement assurance. What they don't inherit is liveness and censorship resistance — the exact properties that matter in the scenario where macro stress becomes regulatory stress becomes somebody deciding your transaction shouldn't land.
And there's a rate story inside L2s, and it's brutal. When EIP-4844 shipped blobs in March 2024, the marginal cost of posting L2 data to L1 collapsed. Great for users. Catastrophic for the sequencer business model. Several large L2s watched fee revenue fall by the kind of percentage that makes a CFO stop answering Slack. The response, in every case, was to search for new revenue: new fee markets, new "applications," new token dynamics. The L2s discovered, publicly and painfully, that cheap base-layer data is the same thing as a subsidy ending. That's the DeFi Summer lesson from four years earlier, wearing a rollup costume.
Scale the analogy up one level and you get the whole point of this essay: a layer that rents its scarcity from somewhere else will eventually discover who owns the scarcity.
Cosmos, or the value capture question nobody likes
Same disease, different organism. IBC is the most elegant interoperability design in production. Genuinely beautiful engineering — light clients, no shared validator set, no bridge honeypot in the middle. I've said nice things about it in public and I'll say them again.
And the application layer on top of it is a fragmented archipelago, and ATOM captures almost none of it. Hundreds of chains. A fraction of the aggregate economic activity settling into the base asset. The token that made the party possible doesn't get paid at the door.
Elegance without accrual is a hobby. And it turns out to be a rate story too: in a high-rate world, the opportunity cost of holding a non-yielding base asset inside a fragmented ecosystem rises sharply. Every basis point the Fed adds raises the bar for "hold this because it's beautiful." Belief is a subsidy. Subsidies get priced.
That's not me dunking on Cosmos. That's me noticing that every layer of this industry shares one structural flaw: the entity that produces the guarantee also fails to capture the value of the guarantee. Then everyone acts shocked when the entity with the balance sheet starts making decisions for the entity without one.
Which is, of course, the headline's claim. Just pointed at us.
Where I actually land
The claim that the Fed hikes for Wall Street, not inflation, is true in the way that "the house always wins" is true. It's a mood, not a mechanism. It collapses the moment you ask whose earnings actually rise on the hike — because almost nobody's do, and one March 2023 weekend proved it. It offers no data, names no one, distinguishes nothing. It's a vibe with a byline.
But here's the contrarian part, aimed at my own side. Crypto's version of financial dominance is worse than the Fed's, because crypto's version has no mandate at all. No dual mandate. No statutory obligation toward employment, or price stability, or anything. When a whale's health factor constrains a DAO's emissions schedule, that isn't a bug in the vision of decentralization. That's the vision of decentralization working exactly as designed for anyone with enough capital to matter. The Fed is at least accountable to a Senate committee. Your favorite protocol is accountable to the three addresses that can move its price.
The Fed at least has to pretend. We don't even pretend.
And the second contrarian swing: if you take the headline seriously, you should notice it cuts the wrong way. If the Fed is constrained by financial stability, then the tightening path is less aggressive than the inflation data alone would justify — because the plumbing breaks first. That's a bullish argument for long-duration assets, not a bearish one. The people sharing that headline as evidence the system is rigged are, mechanically, sharing evidence that the system has a floor. The reflex that supposedly enslaves the Fed is the same reflex that puts a bid under your bags.
Uncomfortable thought for a doom account. Sit with it.
What I'm carrying out of the bar
No, I don't think the Fed hikes for Wall Street. I think the Fed hikes for the plumbing, and the plumbing is made of everyone — the Treasury's issuance calendar, the money funds at the repo window, the regional bank with a duration mismatch, and yes, the crypto fund that borrowed in dollars to buy a yield that was itself a dollar yield. We didn't dodge the chaos; we danced through it, and the chaos had a policy rate attached. The network breathes in Prague, pulses in Ethereum, and settles in the same 4 a.m. dollar as everyone else.
Which leaves one question worth carrying out of this bar and into next week, and it isn't about the Fed.
If every yield you hold is a function of somebody else's balance sheet, and every layer of your stack rents its scarcity from a layer above it — how much of what you call "decentralized" is actually a position, and who is on the other side of the trade when it closes?
I'll be at the bar on Thursday. Bring your balance sheet.